Advanced wireless systems for ranging and communications:measurement and modelling

Advanced wireless systems forranging and communications:measurement and modelling

Ottavio Gremigni

A thesis submitted for the Degree of Master of Research in TelecommunicationsDepartment of E&E EngineeringUniversity College of LondonMay 2008

Abstract

This thesis contains the results obtained in my two-year work spell in WirelessGroup at Philips Research Laboratories (PRL), Redhill (U.K.). In the thesis twoareas of research are presented.

The first topic of research is concerned with UltraWide Band (UWB), which is re-puted as key technology in many applications such as high data rate short-rangelinks and accurate positioning for asset/personal tracking. UWB has been chosenas the air interface for standards 802.15.3a and 802.15.4a. During the six-monthproject period , I contributed to an extensive measurement campaign to assess theranging capability of a prototype developed by several industrial and academic part-ners within the umbrella of European project PULSERS. The task assigned to PRLwas testing the prototype’s performance in typical indoor environments and un-derstand whether is capable of sufficiently precise distance estimation. The resultsobtained show that the platform can reach excellent accuracy even in tough radioenvironments.

The second research area focuses on Peak-to-Average-Power-Ratio (PAPR) reduc-tion for WCDMA mobile terminals. In mobile networks scenario the increasingdemand for higher data rates led to the introduction of new air interface (e.g.WCDMA, OFDMA) whose modulated signals present high values of PAPR bothin downlink and uplink. Signals with high PAPR seriously hamper the design ofefficient amplifiers. Amplifier’s efficiency, especially in mobile terminals, is funda-mental to avoid frequent battery recharging, therefore many solutions have beenconceived to rule out large amplitude peaks and constrain PAPR to tolerable levels.The aim of the project at PRL was developing simple and effective schemes thatcould tame the negative effects of PAPR and be easily integrated in commercialdevices. A standard-compliant model of a WCDMA mobile transmitter has beendeveloped and an exhaustive simulation work helped to indicating PAPR reductiontechniques which allow to accomplish the goal.The negative effects of PAPR have also been evaluated for OFDM-based systems,specifically for WiMAX. A model for WiMAX physical layer has been developedand the difficulties in designing the radio for WiMAX devices have been highlightedthrough simulation results.

Acknowledgement

I would like to thank all the people who supported me during my stay at PhilipsResearch Laboratories in Redhill and actively contributed to this thesis. In partic-ular, my gratitude goes to my industrial and academic supervisors, Mr DomenicoPorcino, Dr Tim Moulsley and Dr Izzat Darwazeh. I would also like to thank mymother and sisters for being so patient listening to my constant moaning aboutBritish weather.

i

Contents

1 Introduction 11.1 Work at Philips Research Laboratories . . . . . . . . . . . . . . . . 11.2 Structure of the thesis . . . . . . . . . . . . . . . . . . . . . . . . . 41.3 Contributions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51.4 Pubblications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

2 Design of UWB Ranging platform and Performance Tests in Indoorenvironments 72.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

2.1.1 UWB: regulation, key characteristics and applications . . . . 72.1.2 Brief history of UWB standardisation . . . . . . . . . . . . . 122.1.3 PULSERS project . . . . . . . . . . . . . . . . . . . . . . . 132.1.4 UWB Ranging in indoor environments . . . . . . . . . . . . 15

2.2 Design of LDR LT platform and results . . . . . . . . . . . . . . . . 162.2.1 System Description . . . . . . . . . . . . . . . . . . . . . . . 162.2.2 Field Trials: Environments and Measurement setup . . . . . 222.2.3 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232.2.4 Improving performance using a post-processing filter . . . . 31

2.3 Discussion and Conclusions . . . . . . . . . . . . . . . . . . . . . . 35

3 Peak-to-Average-Power Reduction Techniques for WCDMA mo-bile terminals 393.1 Introduction: WCDMA overview, PAPR definition and PA modeling 393.2 WCDMA uplink air interface: Physical channels, Spreading and Mod-

ulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 423.3 Distribution of PAPR and its effect on PA . . . . . . . . . . . . . . 453.4 Radio Specifics for WCDMA transmitter . . . . . . . . . . . . . . . 473.5 PAPR reduction techniques . . . . . . . . . . . . . . . . . . . . . . 493.6 WCDMA Uplink Transmitter Simulator . . . . . . . . . . . . . . . 52

3.6.1 WCDMA uplink transmitter . . . . . . . . . . . . . . . . . . 533.6.2 WCDMA downlink receiver . . . . . . . . . . . . . . . . . . 56

iii

3.6.3 ACLR, EVM and PCDE blocks . . . . . . . . . . . . . . . . 573.7 Simulations Results: a comparison . . . . . . . . . . . . . . . . . . . 583.8 Discussion and Conclusions . . . . . . . . . . . . . . . . . . . . . . 62

4 WiMAX modelling 674.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 674.2 WiMAX: an overview . . . . . . . . . . . . . . . . . . . . . . . . . . 68

4.2.1 WiMAX salient features . . . . . . . . . . . . . . . . . . . . 694.3 Mobile WiMAX system model . . . . . . . . . . . . . . . . . . . . . 74

4.3.1 Transmitter . . . . . . . . . . . . . . . . . . . . . . . . . . . 744.3.2 Receiver . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 794.3.3 Addressing high PAPR in WiMAX systems . . . . . . . . . 81

4.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86

5 Conclusions 89

A IEEE 802.16e PHY layer model 93

References 93

iv

List of Figures

2.1 UWB technology shares frequency whit other existing services with-out causing significant interference [1]. . . . . . . . . . . . . . . . . 9

2.2 Different devices can be connected through a wireless UWB interface(source: www.intel.com). . . . . . . . . . . . . . . . . . . . . . . . . 10

2.3 Examples of possible users scenarios [2].(Reprinted with permissionof the authors). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13

2.4 PULSERS LDR-LT platform. . . . . . . . . . . . . . . . . . . . . . 17

2.5 Temporal shape and spectrum of the pulse transmitted by LDR-LTplatform. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

2.6 Block diagram of the RF Front-end of the PULSERS LDR-LT Plato-form. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

2.7 2 DJ-PPM constellation (signals depicted are different from the onesactually transmitted). . . . . . . . . . . . . . . . . . . . . . . . . . . 20

2.8 Two-way ranging algorithm. . . . . . . . . . . . . . . . . . . . . . . 22

2.9 Clockwise (a) anechoic chamber room (b) wooden partition dividingon one side a large conference room and (c) on the other side a smallconference room (d) office room (approximately 290 × 480 cm) (e)Medium-size conference room (approximately 570× 570 cm . . . . . 23

2.10 Ranging mean error for anechoic chamber test. . . . . . . . . . . . . 25

2.11 CDF of errors - 1σ and 2σ error points at different distances for theanechoic chamber test. . . . . . . . . . . . . . . . . . . . . . . . . . 26

2.12 Mean error and standard deviation, office room test. . . . . . . . . . 27

2.13 CDF of errors - 1σ and 2σ error points at different distances for theoffice room test. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28

2.14 Mean error and standard deviation, conference room test. . . . . . . 29

2.15 CDF of errors - 1σ and 2σ error points at different distances for theconference room test. . . . . . . . . . . . . . . . . . . . . . . . . . . 30

2.16 Mean error and standard deviation, office room test. . . . . . . . . . 32

2.17 CDF of errors - 1σ and 2σ error points at different distances for thepartition test. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

2.18 Block diagram of the filter used to improve measurement accuracy. 33

v

2.19 Effect of filtering on mean error and its standard deviation. . . . . . 34



3.1 Rapp’s model AM/AM characteristic as obtained implementing equa-tion 3.2 with a Matlab c© function. . . . . . . . . . . . . . . . . . . 41

3.2 Frame structure for uplink DPDCH/DPCCH [3]. . . . . . . . . . . 42

3.3 DPDCHs and DPCH uplink configuration [4]. . . . . . . . . . . . . 43

3.4 Uplink DPDCH available bit/symbol rates along with correspon-dent SF [3]. The number of bits per DPDCH is determined bythe parameter k (slot format) according to the following equation:Nbits = 10 · 2k, k = 0, . . . , 6. . . . . . . . . . . . . . . . . . . . . . . . 44

3.5 Signal trajectory for an RRC filtered single data channel when usingcorrect HPSK (a) and employing a wrong choice of spreading codes (b). 45

3.6 Cumulative distribution functions of PAPR for a filtered WCDMAsignal with a single data channel (blue line) and 6 data channels (redline). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46

3.7 Effects of an increased PAPR due to PA nonlinearities. . . . . . . . 46

3.8 Error Vector Magnitude (EVM) and related quantities [5]. . . . . . 48

3.9 Power leakage effect on adjacent channels of a 6 DPDCHs signalthrough a nonlinear amplifier: ACLR1 vs IBO (a). ACLR2 vs IBO(b). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49

3.10 Truncated sinc function used in the peak cancellation algorithm (a)and its spectral properties (b). . . . . . . . . . . . . . . . . . . . . . 51

3.11 Hann window with 49 samples. This is the function applied to ourpeak windowing scheme. . . . . . . . . . . . . . . . . . . . . . . . . 52

3.12 WCDMA-compliant transmitter Simulink c© model developed at PRL. 53

3.13 Constituent elements of the WCDMA model’s block that providedata generation and spreading operation as specified in [3]. . . . . . 54

3.14 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55

3.15 Implementation through Simulink c© blocks of the long scramblingcode generator as used in the WCDMA transmitter model we developed. 55

3.16 WCDMA simulator: despreading stage and bit demodulation. . . . 56

3.17 Details of the block that calculates the out-of-band power leakage(ACLR) of the WCDMA transmitter. . . . . . . . . . . . . . . . . 57

3.18 Constituent blocks of the simulator’s module that estimates EVM. . 59

3.19 Constituent blocks of the simulator’s module that estimates PCDE. 59

3.20 BER performance over AWGN channel for Clipping and Filtering (greendashed-dotted line), Unprocessed signal (blue dashed line) and Un-clipped signal (red line). . . . . . . . . . . . . . . . . . . . . . . . . 60

vi

3.21 BER performance over AWGN channel for Peak Cancellation (greendashed-dotted line), Unprocessed signal (blue dashed line) and Un-clipped signal (red line). . . . . . . . . . . . . . . . . . . . . . . . . 61

3.22 BER performance over AWGN channel for Peak Windowing (greendashed-dotted line), Unprocessed signal (blue dashed line) and Un-clipped signal (red line). . . . . . . . . . . . . . . . . . . . . . . . . 62

4.1 OFDMA subcarriers structure [6]. . . . . . . . . . . . . . . . . . . . 724.2 Functional blocks diagram of the mobile WiMAX transmitter model. 754.3 Insertion of cyclic prefix in OFDM symbols. . . . . . . . . . . . . . 784.4 Functional blocks diagram of the mobile WiMAX receiver model. . 794.5 CCDF of PAPR for mobile WiMAX system. 1024 subcarriers, 16 QAM,

R = 3/4. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 824.6 PA clipping effect: input signal (blue line) and amplified output sig-

nal (red line). Amplitude peaks have been constrained to saturationlevel. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83

4.7 16 QAM constellation diagram immediately pinpoints the presence ofnoise due to PA nonlinearity (all the other possible source of interfer-ence have been excluded). Input power 10 dBm, output saturationpower 43 dBm, linear gain 30 dB. . . . . . . . . . . . . . . . . . . . 84

4.8 Signal spectrum distorted by PA nonlinearity (black line) comparedto the spectrum calculated from the input signal. Input power 10 dBm,output saturation power 43 dBm, linear gain 30 dB. . . . . . . . . 85

4.9 Simulated EVM values as a function of the PA output power. Inputback-off not applied. . . . . . . . . . . . . . . . . . . . . . . . . . . 86

A.1 Mobile WiMAX PHY model . . . . . . . . . . . . . . . . . . . . . . 94

vii

List of Tables

2.1 European limits on emitted power density for UWB transmissionsacross the whole radio spectrum [7]. . . . . . . . . . . . . . . . . . . 9

2.2 Results for the anechoic chamber test. . . . . . . . . . . . . . . . . 242.3 Results for the office room test. . . . . . . . . . . . . . . . . . . . . 252.4 Results for the conference room test. . . . . . . . . . . . . . . . . . 282.5 Results for the partition test. . . . . . . . . . . . . . . . . . . . . . 31

3.1 Simulated values of ACLR1, BER, EVM and PCDE of the clippingstrategies tested. CR= 3 dB. . . . . . . . . . . . . . . . . . . . . . . 63

3.2 Simulated values of ACLR1, BER, EVM and PCDE of the clippingstrategies tested. CR= 1 dB. . . . . . . . . . . . . . . . . . . . . . . 63

4.1 Fixed and Mobile WiMAX Initial Certification Profiles. . . . . . . . 694.2 OFDM parameters used in mobile WiMAX. Boldfaced values corre-

spond to those used in our simulator. . . . . . . . . . . . . . . . . . 73

ix

Chapter 1

Introduction

In 2005 the author joined the Wireless Group at Philips Research Laboratories(PRL) in Redhill (U.K.) as part of project co-funded by Philips U.K. and EPSRC.The Wireless Group at PRL has expertise in many areas of radio technology, span-ning from positioning systems to next generation mobile networks specification andit has represented for several years an advanced research centre in the area of wire-less systems. The thesis illustrates the results of a two-year research activity atPRL, where the author has been involved in projects addressing various aspects oftoday’s wireless communications.

1.1 Work at Philips Research Laboratories

During my stay I had the opportunity of taking part to two projects, which covereddifferent aspects of the wireless communications panorama, namely:

• Ultrawide Band (UWB) ranging measurement campaign

• Solutions for future mobile systems

The first six months of my research work have been dedicated to carrying out anextensive measurement campaign meant to assess the performance of a UWB-basedpositioning prototype. The demonstrator under test was developed within the Eu-ropean project PULSERS. The main objective of this project was demonstratingthe feasibility of accurate ranging (i.e. distance measurement through time-of-flightestimation) exploiting the unique UWB capabilities.Since the regulatory authority of the United States of America (US) in 2001, FCC,liberalised UWB transmission in the frequency band between 3.1 and 10.6 GHz,vast research work has been done to develop UWB devices in the area of short-rangecommunications. In fact, given the regulatory limitations on power emission, UWB

1

application scenario has been restricted to mainly indoor short-range systems. Onthe other hand, the availability of such a wide bandwidth made UWB the suitabletechnology for those applications where high spatial resolution is a necessary re-quirement.The future perspective of wireless communications is providing ubiquitous connec-tivity (any-where, any-time) between many devices as diverse as laptops, mobilephones, etc. In such scenario UWB technology has attracted particular interestfrom standardisation groups (IEEE 802.15) that work in the area of Wireless Per-sonal Area Networks (WPAN) and Wireless Body Area Networks (WBAN). AnUWB-based air interface can ensure either very high data rates or excellent posi-tioning accuracy coupled with lower data rates.Thanks to the FCC prompt decision on UWB regulatory laws (in Europe UWBcommunications have been liberalised in 2007 only), some US manufactures havealready put on the market UWB devices, especially in the area of precision assetlocation and tracking. Nonetheless, the role of PULSERS has proved to be funda-mental in speeding up the European regulatory process as well as in filling the gapwith the US by stimulating research on UWB.PRL was amongst the members in the first phase of the project and its specifictask included carrying out indoor field trials to test the ranging performance ofthe prototype developed. It is also worth to stress out the point that PULSERSdemonstration platform has been designed to deliver good results resorting to alow-complexity/low-cost architecture, which might make easier a future commercialimplementation.Along with Mr Domenico Porcino, I have been in charge of setting up the wholetest activity, which included choosing the premises, collecting data and its relativepost-processing. Trials lasted for five months as we aimed at gathering as muchuseful information as possible about the ranging functionality of the platform. Insuch way we could effectively demonstrate that accurate location is achievable evenemploying a low-cost UWB device. Furthermore, the amount of data available al-lowed to carry out a detailed statistical analysis that pinpointed possible causes oferrors affecting the systems’ accuracy, hence we have been able to propose solutionswhich might notably improve the performance. Results were presented at the IETSeminar on UWB technology held in London in April 2006.

Once completed the measurement campaign, I have been involved in a projectwhich is meant to provide innovative solutions to those issues arising from the newmobile systems scenario. The continued increasing demand for higher data rates isforcing all the players in the market to devise architectures that can deliver betterperformance in terms of speed, quality of service (QoS), flexibility and so on.

2

The introduction of Universal Mobile Telecommunications System1 (UMTS) wasthe first step towards such direction. Well-known Global System Mobile (GSM)cannot easily support many applications that require more flexibility, such as dataservices, as it was mainly designed for voice services. From 2G system the horizonof mobile communications has rapidly widened now including applications like gam-ing, web browsing, video streaming, etc. and the standardisation groups are nowworking on future mobile generation, namely 4G or Long Term Evolution (LTE).Compared to GSM, UMTS and its evolution use a completely different air interface.In particular, WCDMA is the standard physical layer for UMTS, whilst 4G systemswill be based on Orthogonal Frequency Division Multiple Access (OFDMA). Bothof them suffer from the so-called ’high Peak-to-Average-Power Ratio (PAPR)’ prob-lem, which means that modulated signals present large peaks well over their rootmean square value. The main drawback posed by signals with large PAP ratios be-comes evident when feeding those signals into the final stage power amplifier (PA):large amplitudes force the PA to operate into its saturation region which is non-linear. As a consequence of PA’s nonlinearities, signals are affected by in-band andout-of-band distortion. Such detrimental effects must be avoided because standards’regulation sets tight limits on spectrum splatter and signal quality.The simplest way to overcome PAPR effects would be ’backing-off’ the amplifier, butthis would decrease its efficiency. Being amplifiers the most power consuming deviceof a mobile terminal, a lower efficiency would result in a bad user experience, i.e.batteries would need to be recharged frequently. That is the reason why researchwork has been done looking for alternatives that might reduce the PAPR rathershifting the PA’s operating point into its linear region. In my activity at PRL, Ideveloped a Simulink c© model for a WCDMA mobile terminal. The scheme is fullycompliant with the standard in terms of modulation requirements so the modulatedsignal is exactly as it would be in a real device. In order to find simple but effectivesolutions to tame the PAPR negative effects, I have carried out a comprehensivereview of the PAPR reduction techniques available in scientific literature. Three ofthem have particularly drawn our attention because we reckoned that, in a possiblefuture implementation, they would fulfil the related requisites. We then decidedto assess the benefits these techniques could potentially bring through simulation.The results obtained have been presented to LCS in September 2007.

During summer 2007, I spent a spell of two months at UCL Electronic & Electri-cal Engineering department where I have worked on modelling the WiMAX (World-wide Interoperability for Microwave Acces) physical layer. Being based on OFDMA,modulated signals in WiMAX systems are charaterised by high values of PAPR. As

1Often referred as third generation (3G) system in contrast to second generation (2G) systemslike GSM.

3

consequence, designers must cope with the same problems as mentioned earlier forWCDMA-based systems. In particular, in WiMAX those issues related to highPAPR are even more crucial than in WCDMA as the limits imposed by the stan-dard on signal’s quality are more stringent.WiMAX communications systems have been conceived to deliver high data ratebroadband services in both fixed and mobile scenarios (called respectively fixedWiMAX and mobile WiMAX ). In 2001 a broad industry consortium, namely World-wide Interoperability for Microwave Access (WiMAX), decided to promote andaccelerate the the introduction of cost-effective broadband wireless access services(from http://www.wimaxforum.org/about/). The solution devised is based on thestandards developed by the IEEE802.16 group. In particular, fixed and mobileWiMAX comply with the 802.16− 2004 and 802.16e− 2005 standards, which formthe technical foundations of this technology. The WiMAX consortium has the roleof providing certification profiles that ensure interoperability between devices of dif-ferent manufacturers.WiMAX solutions can deliver data rates higher than UMTS and its evolution(HSPA) coupled with an unprecedented flexibility in managing the radio resources.For these reasons WiMAX is deemed a suitable technology for several scenarios.Currently the main application field is DSL cables replacement in remote areas,where it would be unduly costly deploy DSL lines. Nonetheless WiMAX potential-ities can be exploited in wider scenarios like national cellular networks as recentlydone in South Korea, where a version of mobile WiMAX, called WiBro, has beenrolled out. WiBro supports broadband internet service with a peak throughput peruser of 3 Mbps in downlink.The model of WiMAX physical layer I developed during my stay at UCL lacks someessential features, nonetheless it could be fruitfully used to analyse the main draw-backs caused from high PAPR. Unfortunately, a more detailed analysis on PAPRor further development of the model could not be performed due to time constraints.

1.2 Structure of the thesis

The thesis is structured as follows:

• Chapter 2: after an exhaustive introduction to UWB technology, its appli-cation fields and essential regulatory information, we will provide a thoroughdescription of the prototype used during the measurement campaign. We willthen illustrate in detail the analysis carried out on the data collected. Inthis section we also present a data filtering algorithm which, in many occa-sions, boosts the accuracy of the ranging estimations. General discussion onpositioning with UWB and conclusions will follow.

4

• Chapter 3: firstly we will introduce the definition of PAPR and go throughthe details of a popular model for solid state amplifiers. The following sectionscontain a clear explanation of the PAPR issues stemming out from multicodetransmission in WCDMA, particularly focusing on mobile terminals. We willthen illustrate the strategies applied to reduce PAPR and a comparison of theresults obtained through simulations. In the last section we draw our con-clusions on which techniques offer the best compromise between performanceand suitability for implementation.

• Chapter 4after a brief overview on the generic characteristics of WiMAX,we will highlight its salient features with particular interest to the ones in-troduced at physical layer. The bulk of the chapter will be dedicated to adetailed description of the WiMAX physical layer Simulink c© model devel-oped. Based on simulation results, the final sections include a brief analysisof the drawbacks caused by high PAPR when designing the radio of WiMAXdevices.

1.3 Contributions

The author’s main contributions to the work reported in this thesis are:

• Setting up the UWB ranging measurement campaign (section 2.2.2).

• Data collection and statistical analysis (section 2.2.3).

• Design and implementation of data filtering algorithm (section 2.2.4).

• Modelling and implementation of a standard-compliant WCDMA system us-ing Simulink c© (section 3.6).

• Analysis of the effects of high PAPR in presence of nonlinear amplifiers forWCDMA-based systems (section 3.3). Performance comparison (through sim-ulations) between three different PAPR reduction schemes (sections 3.5, 3.7)

• Modelling and implementation of a mobile WiMAX system using Simulink c© (sec-tion 4.3).

1.4 Pubblications

The activity on UWB ranging has been summarised in a paper accepted at theIET symposium on Ultra Wideband Systems, Technologies and Applications held inLondon in April 2006.

5

Main results of the investigation on PAPR reduction techniques for WCDMA sys-tems can be found in the paper presented at London Communication Symposiumin September 2006. Both papers are available in electronic form in the CD-romattached to this thesis.

6

Chapter 2

Design of UWB Ranging platformand Performance Tests in Indoorenvironments

2.1 Introduction

In this chapter we present a general overview of ultrawide band (UWB) radio tech-nology and try to explain the reasons why it is considered as a key technology inthe scenario of future wireless communications.

2.1.1 UWB: regulation, key characteristics and applications

Historically, UWB systems have been mainly developed for military applicationsas, due to their large bandwidth, they could ’see through’ trees and beneath theground surface [8]. Recently, UWB technology has been applied to consumer elec-tronics and wireless communications. It is worth to point out that UWB is a verygeneral term to indicate a specific technology: the first UWB devices have beendesigned to transmit sequences of information using pulses of extremely short du-ration (e.g. 0.1− 2 ns) that are widely spaced such that the waveform’s duty cycleis well below the unity (e.g. 1/10 − 1/1000) [9]. However, in 2002 the FederalCommunications Commission (FCC)1in its First Report and Order [10] allowed acompletely flexible approach in generating UWB signals, the only limits imposedwere in the frequency range and spectral energy. According to FCC, UWB radio

1Regulatory authority of the United States of America.

7

device is defined as any device with a fractional bandwidth2 greater than 0.20 orwith a bandwidth occupation of 500 MHz or more. Given FCC’s broad definition ofUWB transmission, today are available many solutions relying on more conventionalmodulation schemes, such as OFDM or CDMA [9]. The spectrum range spans from3.1 to 10.6 GHz with an average spectral energy limited to −41.3 dBm/MHz, thisconstrains the maximum average EIRP (Equivalent Isotropic Radiated Power) to0.56 mW 3.In Europe UWB transmissions have been approved on 21st of February 2007 by theEuropean Commission’s Radio Spectrum Committee [7]. The definition of ultraw-ide band given by the European Commission is slightly different from the one ofFCC, in particular European regulatory authority states that [7]:

’equipment using ultra-wideband technology’ means equipment incorpo-rating, as an integral part or as an accessory, technology for short-rangeradiocommunication, involving the intentional generation and transmis-sion of radio-frequency energy that spreads over a frequency range widerthan 50 MHz, which may overlap several frequency bands allocated toradiocommunication services.

The power emissions levels allowed across the radio spectrum are reported in Ta-ble 2.1. We can notice that the European Committee sets more stringent powerconstraints for UWB than its counterpart in the US. Actually, UWB systems areforced to work in the frequency range between 6 GHz and up to 8.5 GHz (wherethe maximum EIRP density is at an acceptable level of −41.3 dBm/Hz), unlessinterference mitigation techniques are applied; in that case, a maximum EIRP den-sity limit of −41.3 dBm/Hz is also allowed in the bandwidth spanning from 3.4 to4.8 GHz.The Members States had six months, since the entry into force of the commissiondecision, to allow the use of the radio spectrum as specified in [7]. The U.K. regu-latory body, Ofcom, approved UWB communications on 13th August 2007.

UWB capability of smearing the signal’s power across a wide bandwidth makesit a particularly appealing technology in the present scenario of wireless commu-nications, as it allows the coexistence with several other systems. This aspect isfundamental in the modern world of wireless communications, where spectrum is

2Fractional bandwidth is calculated through the following formula:

BF =B

fc= 2

(fH − fL)(fH + fL)

> 0.2 (2.1)

where fH is the upper frequency of the −10 dB emission point and fL is the lower frequency ofthe −10 dB emission point.

3This limit can be reached only if the whole available bandwidth (7.5 GHz) is exploited.

8

Figure 2.1: UWB technology shares frequency whit other existing services withoutcausing significant interference [1].

Frequency range (GHz) Maximum mean e.i.r.p. density (dBm/Mhz) Maximum mean e.i.r.p. density (dBm/50 Mhz)below 1.6 -90 -501.6 to 3.4 -85 -453.4 to 3.8 -85 -453.8 to 4.2 -70 -304.2 to 4.8 -41.3 4 0.0 5

4.8 to 6.0 -70 -306.0 to 8.5 -41.3 0.08.5 to 10.6 -65 -25above 10.6 -85 -45

Table 2.1: European limits on emitted power density for UWB transmissions acrossthe whole radio spectrum [7].

a scarce resource and the number of wireless devices (and the related services) isgrowing fast. Figure 2.1 clarifies this situation by illustrating the spectral mask ofUWB radio transmission (as defined in [10]) along with the ones of existing services,such as GSM and UMTS. UWB power emission levels have been carefully set inorder to minimize the interference impact on other services.Yet, as suggested in [2], the nature of communication is becoming pervasive andthe future perspective of wireless communications is providing ubiquitous connec-tivity between many devices as diverse as laptops, mobile phones and entertainmentequipment. This ambitious view of communications compels the integration of many

9

wireless systems, including cellular networks (2G and beyond), WLAN, WirelessPersonal Area and Body Area Networks (respectively, WPAN and WBAN). Fig-ure 2.2 shows a possible scenario of UWB application, where several wireless de-vices are linked through an ultrawide band connection. The envisaged scenario forpotential UWB applications comprises a wide range of possibilities, for example,FCC has stated in its First Report and Order [10]:

Based on our review of the record, we continue to believe that UWBtechnology offers significant benefits for public safety, businesses andconsumers. We anticipate that the authorization of UWB technologywill create new business opportunities for manufacturers, distributorsand vendors that will enhance competition and the economy. We alsofind that the use of this technology would promote spectrum efficiencyby sharing frequencies with other services without causing interference.

Figure 2.2: Different devices can be connected through a wireless UWB interface(source: www.intel.com).

Due to the extremely low emission levels allowed by the regulatory bodies, UWBsystems tend to be short-range and indoors. UWB can ensure very high ratesand robust performance under multipath conditions [11] as well as high-resolutionposition location and tracking. The transmission of short pulses is ideal for rangingapplication because it allows to reach a high accuracy (see section 2.1.4, that iswhy tracking and location systems exploit this capability of UWB. Given suchcharacteristics of ultrawide band technology, the following user scenarios have beenenvisaged [8] [2]:

Intelligent Wireless Area Networks (IWAN)

10

UWB positioning devices in a master-slave topology could be used in an IWANto enable context-aware services, such as asset tracking, send control signals tocheck the status of the sensors around the home, etc. Deployment of IWANs ischaracterised by high density of devices (at least 5 per room), which can coverdistances up to 80 meters. Such devices must ensure low power consumptionand have to be very low cost.

Sensors, Positioning and Identification Networks (SPIN)

This scenario is suitable for factories and warehouses environments. A greatnumber of devices (e.g., hundreds per floor) is required to adapt to a contin-uously changing radio propagation environment, thus ensuring high level ofreliability. Network topology might be either master-slave or ad-hoc. SPINswill provide low data rate combined with accurate position location and track-ing (e.g., data rate > 10 kbps and accuracy within 1 meter).

Many potential applications have been conceived within the aforementioned scenar-ios, such as personal location, machine remote control or smart homes and offices.With regard to personal location, in [12] an UWB-based system has been devised foruse in emergencies. The system is meant to provide reliable and accurate positioningfor moving users in situations when GPS is not available, i.e. indoor environments.Requirements for rescue operations also include good radio penetration throughstructures, rapid set-up of stand alone system, tolerance to high levels of reflectionand high accuracy (better than 1 meter). All these requisites can be fulfilled onlyby employing ultrawide band short pulses as radio interface.Currently, some commercial systems that offer asset location and tracking are avail-able on the market. As an example, Ubisense (www.ubisense.net) has conceived areal-time tracking and location system that enables a precision of 15 cm. It is alsoworth to mention that UWB systems are particularly suitable for location in sensi-tive environments, such as hospitals, as their transmitted power is limited to very lowlevels. USA based Parco Merged Media Corporation (www.parcomergedmedia.com)was the first systems developer to deploy a commercial version of this system in aWashington, DC hospital.Recently, another application field that exploits the UWB capabilities is cable re-placement. Wireless USB (see www.usb.org) is meant to provide very high datarates over few meters (480 Mbps at 3 m and 110 Mbps at 10 m) in order to replacestandard USB cables. In conclusion, we can state that UWB technology aims atthe following targets:

• low power

• high data rates

11

• low cost

• precise positioning capability

• extremely low interference

2.1.2 Brief history of UWB standardisation

In many cases the success of a technology might be driven by the existence ofcommon standard, which ensure interoperability with other devices. The essen-tial prerequisite for a potentially successful deployment of UWB is represented bythe availability of suitable physical layer (PHY) and medium access control (MAC)standards [2]. As we mentioned in the previous section 2.1.1, ultra wideband sys-tems will be employed for short-range communications and accurate positioning,that is the reason that lead the IEEE 802.15 6 working group to include UWBtechnology as a viable solution to the PHY standardisation for WPANs.The task group 3a (TG 3a) was an attempt to provide a high-speed UWB-basedPHY enhancement to IEEE 802.15.3 for applications related to imaging and mul-timedia. The group started working in 2002. The initial proposals for UWBPHY specification were 23, eventually the group managed to reduce the numberto just 2: Multi-Band OFDM (MB-OFDM) UWB, backed by the WiMedia Al-liance (www.wimedia.org), and Direct Sequence - UWB, backed by the UWB Forum(www.uwbforum.org). Unfortunately, the two industry alliances failed to agree on acommon proposal and, on January 19th 2006, the members voted to withdraw theproject authorization, thus stopping the standardisation process. The decision onwhich will be the ’winning’ standard is left to the market and, once the technologyhas proven to be commercially viable, then the IEEE can come back and revisitwhether it makes sense to create a standard. Undoubtedly, the lack of a standardmight seriously hamper the uptake of high data rate UWB devices as, from theconsumer’s viewpoint, it is not worth taking the risk of buying a system which isnot universally compatible.A completely different outcome has been achieved by the IEEE 802.15 TG 4a, inwhich the members agreed on UWB PHY specification. The group main targetis providing communications and high precision ranging/location capability (1 me-ter accuracy and better), high aggregate throughput and ultra low power; as well asadding scalability to data rates, longer range, and lower power consumption andcost [13]. In March 2005, the group selected a baseline specification. The baselineconsisted of two optional PHYs consisting of a UWB Impulse Radio (operating inunlicensed UWB spectrum) and a Chirp Spread Spectrum (operating in unlicensed

6The IEEE 802.15 is the 15th working group of the IEEE 802 which specializes in the WPANstandards and it includes five task groups (TGs), numbered from 1 to 5 (www.ieee802.org/15/ ).

12

2.4 GHz spectrum) [13]. The final decision was taken in March 2007, the standardis now complete and about to be published.

2.1.3 PULSERS project

The PULSERS (Pervasive Ultra-wideband Low Spectral Energy Radio Systems) is anIntegrated Project (IP) within the 6th European Commission’s Framework Project(FP6) (www.pulsers.eu). The project started in 2004 with the aim of filling the tech-nological and intellectual property gap in Europe in the area of UWB compared tothe US [14]. Supported by a number of around 32 partners from the academic andindustrial world, the project had the ultimate goal of contributing to developmentand deployment of UWB technology in Europe [15].The project started defining potential user scenarios and applications based on shortrange communications approaches, which leverage the unique capabilities offered byultrawide band. According to the requirements of each application, specificationsfor PHY/MAC layers and higher OSI layers were defined in order to develop com-mercially viable devices. All the potential applications envisaged within the projectscenarios were classified into two main operation modes:

• Very High Data Rate (VHDR) + High Data Rate (HDR)

• Low Data Rate - Localisation and Tracking (LDR-LT)

Some examples of possible user scenarios which fall into these two categories areshown in Figure 2.3.

Figure 2.3: Examples of possible users scenarios [2].(Reprinted with permission ofthe authors).

13

As an example, scenarios like WPAN can be effectively deployed only if a highdata rate is available, i.e. 100−500 Mbps, whilst SPIN main requirement is accuratepositioning (within 1 meter). Thanks to its characteristics, UWB has the ability tosupport the possible applications of both these scenarios. Although the scenariosindicated in Figure 2.3 are completely different in characteristics, functionalities,requirements, quality of service, etc. UWB flexibility allows to merge them intojust two categories.The outcome of this preliminary activity has been an essential input to define thesystems’ technical parameters. Such requirements fueled a wide research work toassess the UWB technology state-of-the-art and investigate possible further devel-opment. PULSERS partners looked at all potential technologies that could enableproper exploitation of UWB capabilities. As the project’s aim was implementingfully operational systems based on ultrawide band physical interface, all the aspectsrelated to pursue this final goal had to be taken into account: PHY/MAC design,regulatory issues, coexistence with incumbent systems and services, etc. The con-sequent technical work performed in the scope of PULSERS has been distributedover dedicated work packages (WP). Philips Research Laboratories (PRL) in Red-hill (U.K.) were active in the so-called work package WP2b, which worked on thedefinition, implementation and testing of two platforms for enhanced UWB analysisand demonstration.The first of the two platforms (Communication Platform 1) was dedicated to LDR-LT (data rates below 10 Mbps) and the goal was showing that accurate indoorranging is achievable with a low-cost architecture. The work on CommunicationPlatform 2 included a feasibility study and practical field testing of performancein order to demonstrate the possibility of building high quality Digital Visual In-terface (DVI) [15] wireless systems, with equivalent throughput over 1.6 Gbps andover-the-air-payloads of over 150 Mbps.In particular PRL have been involved in testing the performance of the rangingfunctionality of LDR-LT system. LDR-LT prototype was shown publicly duringthe PULSERS workshop at the IST summit 2005 [16]. Since then, it has been usedfor trial tests in typical office environments within the premises of PRL. In thefollowing sections the LDR-LT platform will be described in detail along with theresults obtained during an extensive campaign of measurements, which took placein autumn and winter 2005.

Another important objective of the project was contributing to the ongoingEuropean regulatory activity, efforts made by PULSERS in this direction lead to theactual European regulation [7]. A specific work package in PULSERS addressed thecomplex issues related to spectrum regulation. Regulatory authorities and operatorsconsidered UWB a ’disruptive’ technology because of the reuse of frequency bandsalready assigned to incumbent services, thus expressing their concerns over possibleharmful interference. PULSERS strove to rule out these legitimate concerns through

14

technical studies [15]. Besides the results obtained in the European regulatoryprocess, PULSERS has also actively contributed to world-wide standardisation worksubmitting many proposals both to 802.15.3 and 802.15.4a groups.

Now the project PULSERS has entered in its second phase, the new objectiveshave been conceived following the work done in first phase. The future researchwork will aim at consolidating and expand the PHY and MAC schemes, imple-menting a fully operational UWB-based system according to new European deploy-ment rules [17]. Effectively, Phase II will focus on system integration rather thantechnology research. With respect to Phase I, the project will have more emphasison higher OSI layers as a necessary step to complete system implementation andsophisticated system verification. In parallel to the main tasks, PULSERS mem-bers will continue to pursue advanced research topics in the field of distributed orco-located multiple antenna systems (MAS). Even though a preliminary study onMAS was carried out in the previous phase, much work still needs to be done inthis area as one of the objectives of this second phase is delivering a MAS testbed.

2.1.4 UWB Ranging in indoor environments

UWB proved to have unique advantages for precision location applications. Theuse of short pulses (typically hundreds of picoseconds in time duration) provides in-herent accuracy for time-of-flight measurements as well as robustness to multipatheffects in indoor environments [18]. Ranging in today indoor UWB positioning sys-tems is often performed through time-of-arrival(TOA)7 estimation and its accuracydepends on the the signal’s bandwidth: the larger is the bandwidth the better isthe accuracy in TOA estimation. The resolution of the measurement is related tothe bandwidth as follows:

d =c

BW[m] (2.2)

where d is the absolute resolution and BW denotes the bandwidth of the signal (cis the speed of light). From Equation 2.2 it is clear that resolutions of the order offew centimeters are achievable employing UWB technology. For example, a signalwith 2 GHz bandwidth allows a resolution of 15 cm. However, when multipathpropagation occurs UWB systems cannot maintain the theoretical accuracy givenby Equation 2.2, nonetheless noticeable precision can still be reached as shown [18],where a FCC-compliant asset location system obtains a ranging accuracy betterthan 30 cm, with standard deviations as little as 15 cm in an indoor scenario, thusconfirming UWB robustness to multipath fading. This due to the large bandwidthused in such systems: actually transmitting very short pulses reduces the InterSymbol Interference (ISI) and makes possible resolving multiple paths, thereforecontrasting fading and interference [8] [11].

7More details on ranging calculation strategies can be found in [8].

15

The reason why we focusing our discussion on indoor scenarios is that UWB appli-cations will be deployed mainly in this field as regulatory limits on power emissionscompel UWB to use very low energy devices, which are not suitable for outdooruse (see Section 2.1.1). Many field trials have been carried out with testbeds trans-mitting UWB short pulses and all of them have been shown that accurate rangingcan be performed by exploiting UWB characteristics even in dense multipath in-door environments, both in Line-of-Sight (LOS) and Non-Line-of-Sight (NLOS)conditions [19] [20]. Obviously, the performances of a UWB-based location systemalso depend upon the hardware design, processing algorithms,etc. Nonetheless, itis worth to stress out that, in principle, ultrawide band technology is able to ensurevery accurate positioning (below 1 m) even in tough radio propagation conditions.This was unthinkable with previous narrow band systems.In the following section, the LDR-LT prototype developed by PULSERS partnerswill be described in detail, with particular attention to the ranging functionality.The aim of the measurement campaign held at PRL was to demonstrate that achiev-ing accuracy of few centimeters in indoor scenarios is a feasible objective even withlow-cost/low-complexity radio architectures.

2.2 Design of LDR LT platform and results

The PULSERS LDR-LT demonstrator was built to provide a resolution of 30 cmcoupled with a useful data rate of 12.5 Mbps. The emphasis was on a low-cost ar-chitecture, which could allow future developments of intelligent commercial sensors.At the time when the measurement campaign took place the full data functionalitywas not implemented, so we could not be able to test it. Nonetheless, several testshave been carried out at PRL to assess the platform’s ranging capability in indoorenvironments, such as office rooms, conference rooms, corridors, etc. The resultsobtained from the measurement campaign are illustrated in section 2.2.3.The mechanism chosen for ranging calculation is a simple two-way time of arrival(TW-TOA) detection based on non-coherent energy collection, where the transmit-ter and the receiver are both capable of exchanging data to ease synchronizationand clock error removal. The following section contains a detailed description of thesystem’s architecture and main features.

2.2.1 System Description

Each of the two LDR-LT hardware platforms used at PRL for ranging tests iscomprised of the following devices (see Figure 2.4):

• 1 UWB transmitter

16

• 1 UWB receiver

• 1 FPGA (for baseband processing)

• 2 Analogue-to-Digital converters (ADC)

• 2 Wideband antennas

The whole system is divide into two main parts: RF section and Basebandprocessing. The RF section includes the transmitter, the receiver and the antennasand it deals with analogue UWB pulses. The baseband contains the FPGA processorcard that is mounted on a commercial personal computer (PC) and it is in chargeof processing digital data coming from the ADCs and enabling data demodulationand ranging.

Figure 2.4: PULSERS LDR-LT platform.

Transmitter - the pulse generator

The transmitter generates short (time duration around 500 picoseconds) low power(peak-to-peak voltage around 1 V ) pulses that produce a noise-like spectrum whosebandwidth spans from 3 to 5 GHz. The pulse shape and its spectrum at theRF output of the transmitter are shown in Figure 2.5. The transmitter has beendesigned to be compliant with the FCC’s regulation, as a European policy on UWBcommunications was not available yet.

17

(a) temporal pulse shape

(b) Spectrum of the transmitted pulse

Figure 2.5: Temporal shape and spectrum of the pulse transmitted by LDR-LTplatform.

Receiver - RF Front-end

The RF front-end part of the receiver is based on a non-coherent energy detectionscheme. The block diagram in Figure 4 illustrates the main components of thisdevice. The received pulse is filtered to lower out-of-band interferers and then am-plified with a 30 dB Low Noise Amplifier (LNA). The output of the second stagebandpass filter is then split into two separate branches for processing at differentresolutions: the data demodulation and ranging branches.

The demodulation branch comprises a Power Detection circuit (PoD) and anintegrator, whereas the ranging one has PoD and an amplifier with fixed gain of20 dB. The PoD is based on a Schottky diode followed by a capacitor and a resistor(to avoid leakage). This circuit acts like an envelope detector and the signal comingout from this device has a lower bandwidth with respect to the input UWB signal,

18

this was necessary to match the ADC bandwidth of 900 MHz.

The integrator in the demodulation branch is needed to achieve signal synchro-nization; it has an integration window of 20 ns, which can be shifted in steps of4 or 8 ns. In the integration strategy followed by this structure the best startingintegration time is the one that leads to the higher recovered signal energy. In theranging branch there is no integrator as the signal is amplified and sent to the ADCwithout any further processing. The integrator is replaced by the sampling win-dow approximately 1 ns wide. More details on the demodulation mechanism arereported in following sections.

Figure 2.6: Block diagram of the RF Front-end of the PULSERS LDR-LT Plato-form.

Antennas

The LDR-LT demonstrator employs four omnidirectional antennas (two for eachplatform, one for the receive and one for transmit path). For the experimentalcampaign some UWB printed antennas manufactured by TDK have been employed8.

Baseband processing

The baseband part of the receiver includes an FPGA card mounted on a PC. Thebaseband processor is responsible for all the features as synchronization, demod-ulation, ranging and framing. Data from RF modules (both transmit and receivepaths) are conveyed to the FPGA via ribbon cables. The user can set the valuesfor some basic parameters (e.g. type of modulation) of the platform through appli-cation software developed by PULSERS partners. The software allows writing suchvalues into the FPGA registers through a custom interface allowing exchange of

8Details of the antennas cannot be disclosed due to a non-disclosure agreement between PRLand TDK.

19

data with baseband. In the following sections we will focus on the synchronizationalgorithm for ranging and demodulation.

Modulation

The modulation used in the LDR-LT platform is a 2 disjointed pulse position mod-ulation (2 DJ-PPM), where ’disjointed’ means that time slots do not overlap. Ac-cording to the largest data rate available of 12.5 Mbps, the smallest pulse repetitionperiod (PRP) is 80 ns, and this leads to the choice of a 40 ns time slot per symbol.The constellation is illustrated in Figure 2.7.

Figure 2.7: 2 DJ-PPM constellation (signals depicted are different from the onesactually transmitted).

Demodulation and Ranging synchronization

The synchronization algorithm has to find out the best value for the integrationstart time. A scan of the preamble sequence is carried out by shifting the integra-tion window (20 ns wide) at steps of 4 or 8 ns9. The time shift whose integration

9The shorter step leads to a more sharp synchronization, whilst the larger leads to faster butless precise results.

20

value is higher corresponds to the best start time for signal integration. This valueis fed to the demodulator block.After the synchronization process has been successfully completed, the ranging al-gorithm starts. The aim of the ranging process is to seek for the first path arrivedwithin a given demodulation window (i.e., within a symbol time slot). The signalat the output of the PoD circuit of the ranging branch is fed to the ADC. Thesampling time of the ADC is approximately 1 ns and it can be shifted by 1 ns stepsthanks to custom designed delay lines. Considering radio waves travelling at speedof light, this leads to a maximum resolution of 30 cm. The achievable resolution ofthe system is 1 ns even if the clock period of the LDR-LT demonstrator is 40 ns,this is made possible by carrying out the ranging process over a set of consecutivesymbols, assuming the channel unchanged during this period. Samples are takenstarting from the demodulation window’s position at steps of 1 ns apart. The peakvalue with the shortest delay and amplitude above a certain threshold is consideredthe first path arrived.

Two-way ranging algorithm

The algorithm used to calculate the distance between the platforms is a simple twoway time estimate process. As shown in Figure 2.8, the T1 and T2 time intervalsare evaluated by means of four timestamps: Tx1reg, Tx2reg, Rx1reg, Rx2reg. Thefirst two timestamps are 25 bits long and have a resolution of 40 ns (they recordthe symbol time of frame transmission), whereas the last ones are 32 bits long andhave a resolution of 1 ns (5 bits are used to represent the delay of the estimatedfirst path, 1 bit indicates the path position in the neighbouring timeslot).A complete description of the ranging algorithm follows:

• Start of ranging: device I sends a ranging frame to device II and a timestampTX1reg is stored in a register;

• Device II receives the ranging frame, a timestamp RX2reg is stored;

• Device II answers the ranging frame, including the receiving timestamp RX2regas well as the transmission timestamp TX2 in the data field of the frame;

• Device I receives the answer frame, estimates a receiving timestamp RX1regand extracts the timestamps of device II to be stored in registers;

• After this cycle the four register values are passed to the application software,which calculates the distance with the formula in 2.5.

T1 = Rx1reg(31 : 7)−Rx1reg(6)− Tx1reg(31 : 7) ∗ 40 ns + Rx1reg(5 : 0) ∗ 1 ns(2.3)

21

T2 = Tx2reg(31 : 7) + Rx2reg(6)−Rx2reg(31 : 7) ∗ 40 ns−Rx2reg(5 : 0) ∗ 1 ns(2.4)

D = 0.5 ∗ (T1− T2− offset) ∗ 30 [cm] (2.5)

The first seven bits (0 . . . 6) of Tx1reg and Tx2reg are set to zero. The parameter’offset’ takes into account all the delays due frame processing. The ranging cycleis performed once per second and it lasts 1/2 ms. The accuracy for the doubleddistance is 30 cm, hence for the single distance results in a maximum resolution of15 cm.

Figure 2.8: Two-way ranging algorithm.

2.2.2 Field Trials: Environments and Measurement setup

In order to assess the performance of the platform in indoor environments with dif-ferent radio propagations characteristics, measurements were carried out at variouspremises in PRL building. The first step was testing the LDR-LT demonstrator inideal conditions within an anechoic chamber. Assessing platform’s performance in amultipath-free environment helps to understand whether the prototype is workingproperly, which means showing results close to the ones expected from theoreticalanalysis. Then we checked the effectiveness of the platform’s ranging functionalityin scenarios where multipath was present. Tests were carried out both in LOS andNLOS conditions. Measurements in LOS took place in two typical office scenarios:an office room of small dimensions (approximately 290×480 cm) and a medium-sizeconference room (approximately 570 × 570 cm). In the NLOS trial we placed thedevices into two adjacent rooms divided by a wooden partition wall. Pictures inFigure 2.9 illustrate the environments were tests took place.During measurements one platform was moved along a marked track with markers

placed every 30 cm, whilst the other platform remained at a fixed position. Theequipment was placed onto and moved with two trolleys whose heights are 90 and105 cm. A recording time of a fixed length (typically 15 minutes) had been set foreach measured distance to get several estimates per measurement point10. In this

10Ranging cycles are carried out once per second.

22

(a) Anechoic Chamber (b) Partition test:large conference roomside

(c) Partition test:small conference roomside

(d) Office room

(e) Medium-scale con-ference room

Figure 2.9: Clockwise (a) anechoic chamber room (b) wooden partition dividing onone side a large conference room and (c) on the other side a small conference room(d) office room (approximately 290 × 480 cm) (e) Medium-size conference room(approximately 570× 570 cm

set of experiments no coding was employed to protect data from channel errors,therefore some measures could be wrong. Furthermore, due to occasional synchro-nization errors, a ranging cycle could fail and in this particular case no distancevalue was available at the output. As a consequence it is impossible to achieve afixed number of valid distance values per each measurement. This is why it has beendecided to set a constant recording time rather than deal with the same number ofsamples per each measured distance. The amount of valid (i.e., fully synchronized)measures will be showed while presenting the results.

2.2.3 Results

In the following sections we will discuss the results obtained for each of the fourenvironments under investigation.

Anechoic Chamber

The anechoic chamber test is extremely useful as it represents the upper bound forthe LDR-LT demonstrator performance. The absence of multipath scattering leadsto the lowest possible errors, as the system is supposed to be sensitive to multipath

23

effects due to its non-coherent energy detection scheme. The results obtained mustbe regarded as the upper bound of the ranging capability of LDR-LT.Table 3.1 shows the results obtained for the anechoic chamber environment. Thevalues contained in the column Measured Distance represent the average of the mea-surements data obtained at that particular distance (the number of samples takeninto account in the calculation is presented in the rightmost column). In the thirdand fourth columns the mean error between the real and measured distance and itsstandard deviation are displayed.

True Distance [cm] Measured Distance [cm] Mean Error [cm] Std Dev of Error [cm] No. of valid measures

90 87.8 4.8 7.6 202120 124.1 4.7 9.2 192150 157.0 9.9 10.2 207180 178.9 7.7 8.1 188210 216.8 10.4 10.4 171240 240.2 6.8 8.5 205270 269.3 10.1 8.9 188300 292.2 10.3 8.7 179330 326.5 9.0 8.0 195360 362.2 7.6 7.5 160390 379.6 12.1 11.8 194420 420.7 11.7 10.4 174450 446.0 9.5 10.0 173480 480.9 9.9 10.5 173510 513.2 9.1 8.4 192540 539.3 7.3 11.3 64600 597.9 4.3 6.9 28

Table 2.2: Results for the anechoic chamber test.

The mean error has a peak value of 12 cm at a distance of 390 cm (see Fig-ure 2.10). For all other measurements the error is lower, showing clearly that aUWB system can achieve an excellent ranging accuracy even with a low complexityreceiver. Same indications can be drawn from the plot of the Cumulative Distribu-tion Function (CDF) of errors in Figure 2.11, which confirms errors below 30 cm upto 6 m tx-rx distance.

Checking the performance of the LDR-LT demonstrator in the anechoic chamberhas been extremely useful to verify the potential platform’s capabilities in terms ofranging accuracy, on the other hand, this is not enough to claim that high-precisionranging can be achieved by such platform. In fact, we had to gauge the effect ofmultipath on the ranging functionality by carrying out tests in environments wheremultipath comes into play.

24

Figure 2.10: Ranging mean error for anechoic chamber test.

Office Room

This environment has been chosen because it embodies the features of a typicaloffice and can help doing a Line-of-Sight (LOS) analysis in a cluttered environment.Measurements were carried out moving one platform along a straight track fromthe fixed station to the outside. Being the room small in size (approximately 290×480 cm) the maximum distance measured was 3.6 m only.The results obtained are summarized in Table 3.2.


90 89.4 5.5 8.0 396120 115.0 9.8 7.5 434150 144.9 10.5 7.8 464180 179.1 12.6 30.2 410210 233.2 27.8 68.8 327240 243.9 17.2 48.0 458270 318.1 48.1 32.5 421300 318.5 21.8 45.6 383330 394.5 65.4 109.4 392360 363.2 11.1 10.2 443

Table 2.3: Results for the office room test.

Data available show that the errors in ranging estimates are generally low, apartthree cases: 210, 270 and 330 cm, which present remarkably higher average errors(see Figure 2.12(a)). Both 210 and 330 cm also have a higher standard deviationvalue with respect to all the other measurements (see Figure 2.12(b)), this couldmean that there is a small amount of data completely wrong, i.e. some measuresare totally out of range due to demodulation failures or peak misdetections. Thesame argument can not be applied to 270 cm case because its standard deviation is

25

Figure 2.11: CDF of errors - 1σ and 2σ error points at different distances for theanechoic chamber test.

reasonably low.

The analysis of error CDFs gives us a clearer view about errors distribution.From the bar plot in Figure 2.13, we can notice that taking into account ’best’ 67%of the measurements done at distance 210 cm the error is less than 10 cm, whilstif the whole set of data is considered the error rises up to 70 cm because of out ofrange estimation values. Distances at 270 and 330 cm show a different behaviour:the error increase from 67% to 95% is just 20 cm, even though we must remindthat the error’s standard deviation at 330 cm is noticeably higher than at 270 cm.A possible solution to clarify such situations might consist in ruling out evidentlyodd measurements, this would also bring to an increase of the overall platformperformance. In section 3.10(b) a simple technique to reject wrong values will bepresented.

In conclusion, we noted that the measurements carried out at 210 and 330 cmhave higher ranging error due to possible channels errors or peak misdetections,which caused some measurements to fall well beyond a reasonable error range. Thisis confirmed by their high values of standard deviation. On the contrary, the 270 cmcase should be interpreted as failed measure because of its low value of deviation forthe error. Such errors are probably due to multipath effect as they are not presentin the anechoic chamber test.

26

(a) Ranging mean error, office room test.

(b) Standard deviation of error, office room test.

Figure 2.12: Mean error and standard deviation, office room test.

Conference Room

In order to gain a more comprehensive understanding of the platform’s performance,we decided to repeat a LOS in a room with different characteristics with respectto the office space of the previous section. We opted for a medium-size conferenceroom that is common to many other office buildings. Its dimensions are notablylarger than the office room ones.

Even if there is still LOS between the two platforms, radio propagation condi-tions are different as they are determined by the position of furniture, the distanceof the antennas from walls and ceiling, etc. During the test the platform close tothe wide window has been held in a fixed position and the other one moved along

27

Figure 2.13: CDF of errors - 1σ and 2σ error points at different distances for theoffice room test.

a straight track as shown in Figure 2.9. The results obtained are included in Ta-ble 2.4.


90 87.3 7.6 8.0 452120 112.9 10.8 7.3 440150 144.2 10.9 7.8 447180 180 13.9 39.8 454210 223.3 21.8 75.8 419240 232.8 11.8 9.9 419270 327.1 57.4 57.6 358300 307.6 13.6 12.0 430330 422.3 92.7 135.8 405360 386.4 27.0 50.8 349390 405.8 21.2 61.3 320420 437.9 22.0 52.3 288450 464.5 18.8 51.2 429480 507.1 32.7 148.5 316510 565.1 61.5 153.1 174

Table 2.4: Results for the conference room test.

From Figure 2.14 we can notice that measurements at 270, 330 and 510 cmpresent higher error values with respect to the rest. Moreover, the standard devia-tion at 330 and 510 cm reaches large values, whilst the error’s variance at 270 cmis within a reasonable range.

The analysis of the errors CDFs clarifies such situations, see Figure 2.15. It isevident that there are some measures whose relative error is well above the 100%,as the CDFs values at 95% for 330 cm and 510 cm are larger than the distance

28

itself. As in the office room case, there are totally out of range values that makethe error and its standard deviation grow remarkably. This is confirmed by notingthat, if the only values within the 1σ point of the CDF distribution are considered,the error is well below 1 meter for all the distances, which indicates that LDR-LTplatform is able to achieve a good accuracy (actually excluding the aforementionedcases with large errors, the average error is below 40 cm for the remaining distances)and performance are likely to be improved by ruling out those odd values.

(a) Ranging mean error, conference room test.

(b) Standard deviation of error, conference room test.

Figure 2.14: Mean error and standard deviation, conference room test.

Concluding we can state that the overall performances have the same trend asin the office room test: a notable ranging accuracy both in terms of mean error andrelative standard deviation, although with few measurements showing high errors.As already mentioned in the previous paragraph, data post-processing could be the

29

solution to further improve the distance estimation’s precision, i.e. reducing theerror range in those cases affected by large error values.

Figure 2.15: CDF of errors - 1σ and 2σ error points at different distances for theconference room test.

Intra-room partition test

The partition test has been useful to check the performance of the demonstratorin Non-Line-of-Sight (NLOS) radio propagation environment. The partition sep-arating the two conference rooms (see Figure 2.9) is wooden. It has a width of8 cm with two panels joined to create a sound-proof and robust separation of twoadjacent environments. During the tests the partition was closed and each platformwas positioned in one of the two separate rooms. The results obtained are shownin Table 2.5.

Figure 2.16 contain the mean error and the error standard deviation. Measure-ments at 138, 378, 528 and 618 cm have a standard deviation that is above 80 cm,this means that large errors might have occurred causing some values to fall out ofrange, as we already noticed in the office and conference room tests. Nonetheless,we can conclude that the system performs quite well with respect to the harsh radiopropagation conditions in which was forced to operate ranging: the average erroris, in any case, lower than 45 cm. Results are not far from the ones seen in theprevious LOS test.

Going back to measurements with high standard deviation of error, we triedto go through the CDFs analysis in order to better understand the distribution oferrors. Figure 2.17 shows that, in this case, CDFs are not very helpful to clarify

30


108 91.8 17.4 7.9 200138 159.8 39.3 105.4 207168 153.9 18.1 10.5 212198 225.7 31.0 19.9 181228 272.6 44.6 49.1 169258 266.5 22.9 13.7 188288 318.5 32.4 23.6 124318 313.8 14.4 43.0 193348 335.7 15.3 11.4 188378 389.5 24.2 84.3 194408 436.6 30.6 58.3 119438 444.9 17.3 11.6 70468 470.2 13.6 52.3 136498 494.2 18.0 34.8 114528 565.6 40.4 101.7 144558 575.3 20.5 49.5 159588 598.7 14.2 47.6 155618 611.3 28.5 104.0 157

Table 2.5: Results for the partition test.

error distributions; as an example, the measurement at 618 cm presents an errorbelow 10 cm for the 95% of the measures, but the average error and its varianceare well above such value (see Figure 2.16). It could be that few large errors makethe mean error grow noticeably, this is probably due the lack of a proper amountof data to carry out such statistical analysis.

2.2.4 Improving performance using a post-processing filter

While the system ranging accuracy was generally quite good as shown in the pre-vious sections, it was apparent that the lack of any coding and the use of straightunfiltered raw data could cause peak misdetections and the consequent appearanceof few spurious measurements which would influence the system reliability. To im-prove the performance we then decided to implement a post-precessing basic filter.Such solution seems to be the easiest (and cheaper) way to achieve the purpose ofruling out erroneous values from the set of data and reach better performances.The approach used is similar to Kalman filtering technique, which has been widelyused to increase positioning accuracy in GPS [21] [22] and now has been also appliedto UWB-based location systems [23]. This strategy appeared as the most suitable toprocess the raw collected from our tests given the nature of errors occurred, i.e. fewout-of-range values among groups of good estimations. Basically, the filter consistsin the moving average of five consecutive measures - which provides an estimate ofthe distance - the incoming value is taken into account if its relative error is withinthe 50% threshold. It is important to note that rejected measures are considered inthe distance estimation in order to overcome a bad initialization of the algorithm.This filtering technique is quite simple but effective to reduce the number of odd

31

(a) Ranging mean error, partition test.

(b) Standard deviation of error, partition test.

Figure 2.16: Mean error and standard deviation, office room test.

measurements, unfortunately it fails whenever a group of consecutive errors occur.A block diagram of the filter employed is depicted in Figure 2.18.

Applying filtering to the results of the anechoic chamber test - as expected - doesnot bring any improvements, as there were no out of range values. Different is thesituation in the case of the ranging measurements in LOS and NLOS conditions.

Office Room Figure 2.19 show the effect of filtering on mean error and its stan-dard deviation. The average error is remarkably lower for measurements at 210 cmand 330 cm as well as its variance, therefore filtering achieved the goal of ruling outodd values. As expected, there is no improvement by filtering for the 270 cm case:the mean error remains unchanged and its variance is only slightly lower, thereforethis measurement must be considered definitively wrong.

32

Figure 2.17: CDF of errors - 1σ and 2σ error points at different distances for thepartition test.

Figure 2.18: Block diagram of the filter used to improve measurement accuracy.

However, if we consider the whole picture rather than focusing on single cases wemust conclude that generally LDR-LT provides an adequate accuracy. The meanerror never exceeds 50 cm and often it is as low as 10 cm or less (e.g., measurementsat 90, 120, 150, 180, 360 cm).

Conference room Filtering data obtained from the conference room trial hasa beneficial effect on average error and its variance as shown in Figure 2.20. Inparticular, distances at 330 and 510 cm present a significant reduction in bothmean error and standard deviation. The average error fell by 30 cm (from 90 to60 cm) for the measurement at 330 cm and 40 cm (from 60 to 20 cm) at 510 cm.

33

(a) Comparison between the mean error before and after filtering,office room.

(b) Comparison between the error standard deviation before andafter filtering, office room.

Figure 2.19: Effect of filtering on mean error and its standard deviation.

On the contrary, the measurement at 270 cm still has a high error value and theerror distribution deviation remains unchanged, therefore we must interpret it as afailed measurement.For all the other measured distances the average error is no grater than 20 cm,which confirms that in LOS condition precision ranging is undoubtedly feasible.

Partition test The filter has been also applied to the results of the partition test.Figure 2.21 shows that a noticeable improvement is achieved for the measurementswith high error and variance, i.e. 138, 378, 528 cm. Again the filter proves its effec-tiveness in removing random wrong distance estimates, even though the distancemeasure at 618 cm is almost unaffected by filtering and this confirms that the filter

34

(a) Comparison between the mean error before and after filtering,conference room test.

(b) Comparison between the error standard deviation before andafter filtering, conference room test.


is unable to remove clusters of close-by wrong values. Conversely, the standarddeviation at 228 cm is greatly reduced but the error is just slightly lower than priorto filtering, therefore we must conclude that the measurement is definitively wrong.

2.3 Discussion and Conclusions

LDR-LT platform has shown that accurate ranging is achievable even with low-complexity and low-cost devices. PULSERS put particular emphasis on this aspectthroughout the design process. The approach followed in developing this proto-

35

(a) Comparison between the mean error before and after filtering,partition test.

(b) Comparison between the error standard deviation before andafter filtering, partition test.


type was - in some extent - different with respect to other commercial systems. Asmentioned in section 2.1.1, Ubisense offers to the market a 2D − 3D positioningsolution based on a combination of Angle Of Arrival (AOA)/Time Difference OfArrival (TDOA) [24] with tags being provided with an extra control channel inthe ISM bandwidth. LDR-LT operation relies on a more simple location technique,TOA. A drawback of such scheme is that the system needs slightly more ’symmet-rical tags’ (which means that tags must be able to perform synchronisation andclock error removal) than other ranging devices, such as the PAL system describedin [18]. Furthermore, the system envisage by PULSERS is supposed to result inbeing user-friendly in the set-up not requiring any site calibration or fingerprinting

36

of any type.The results obtained from field trials encourage to keep on further developing theLDR-LT platform into a fully-operational system. The PULSERS LDR-LT demon-strator was able to achieve an excellent ranging accuracy (with a relative error below5% up to 6 meters) in a multipath free environment such as the anechoic chamber.Tests held in more realistic premises where multipath was present have shown that’raw’ measurements can present large errors, but still lead to accuracies far betterthan any narrowband system [25]. We have also noticed that data filtering helpsimproving even further such accuracy, as it removes spurious measurements thatmay have occurred. Even in tough radio conditions, such as NLOS, the averageerror in distance estimation does not exceed 45 cm, which is a remarkable result.Also, an in-depth analysis of those cases affected by large error values might leadto the conclusion that they are actually due to the absence of coding, which causessynchronisation failures. In fact, the total lack of data protection mechanism cou-pled with the low complexity of the receiver make the system sensitive to multipath,even though the final results can be still deemed good enough to enable intelligentsensors applications and asset tracking. Obviously, the introduction of a codingalong with other more advanced filtering techniques could further improve the suc-cess rate of synchronisation and the general performance of the ranging detection;however, our results show that short distance ranging is already feasible with lowcomplexity equipment today.In conclusion, we can state that the measurement campaign held at PRL contributedto clearly demonstrate that UWB-technology is capable of accurate distance mea-surement, even with a very simple energy collection and two-way time-of-flight de-vice. Starting from these promising results, PULSERS II new goal is thoroughlyimproving LDR-LT platform into a completely functional system for asset trackingand intelligent sensor networks applications.

37

Chapter 3

Peak-to-Average-Power ReductionTechniques for WCDMA mobileterminals

3.1 Introduction: WCDMA overview, PAPR def-

inition and PA modeling

The Universal Mobile Telecommunications System (UMTS) was introduced to addnew features that could pen up the market to new applications, which the GlobalMobile System (GSM) was not able to support. The best known among these fea-tures of UMTS is higher user bit rate [26], which can allow bit rates up to 2 Mbps onpacket-switched connections. Compared to GSM, UMTS offers more flexibility inmanaging the radio resources and this characteristic allows to support a wide rangeof service with different requirements in terms of Quality of Service (QoS). UMTSsystems relies on Wideband Code Division Multiple Access (WCDMA) air interface,its specification has been created in 3GPP (3rd Generation Partnership Project)which is the joint standardisation project including companies from Europe, Asiaand America. The first full specification for WCDMA was completed in 1999 (theso-called Release 99), since then four other major versions of the standard have beenpublished, each release adds new functionalities to keep pace with the increasingnumber of services. In the last two releases (respectively Release 5 and 6), manyefforts have been done to enhance the maximum bit rate per user both in uplink [27]and downlink [28]. As a result, a slightly different WCDMA physical interface hasbeen introduced in order to implement such enhancement [29] and reach data ratesup to around 4 Mbps in uplink and 14 Mbps in downlink.

39

WCDMA standard1 [3] allows the use of multiple data channels in order toachieve higher data rates. The sum of parallel channels increases the so-called Peak-to-Average Power Ratio (PAPR), i.e. the modulated signal can have high peaks wellabove its average amplitude. According to [30], the PAPR can be defined as:

PAPR = 20 log10

(‖x(t)‖∞‖x(t)‖2

)dB (3.1)

where ‖x(t)‖∞ is simply the peak of the signal and ‖x(t)‖2 is its root mean squarevalue (rms).In mobile terminals the final amplifier is the most power consuming device so itmust work as close as possible to its saturation point, otherwise it would have alow efficiency resulting in frequent battery recharging [31]. As a consequence, sig-nals with high PAPR saturate the power amplifier (PA) causing out-of-band (OOB)emissions and in-band distortion. In an attempt to avoid such effects, the signalat the input of the PA is attenuated applying a back-off (BO) factor. The BOconstrains the signal’s amplitude range, thus the PA can work in its linear region.Unfortunately this yields a drawback: the amplifier efficiency is reduced. As a re-sult, a compromise must be reached between PA efficiency and allowed amount ofspectral splatter/signal distortion.

Throughout this work the PA model is the well-known Rapp’s model for solidstate amplifiers [32], its amplification characteristic (or AM/AM characteristic) is:

A[r] =νr

[1 + ( νrA0

)2p]12p

(3.2)

where A0 = νAs is the saturating , ν is the small signal gain and p is an integer.Note that, as p grows larger, the curve approaches the ideal soft limiter. We set thefollowing values for the amplifier’s parameters:

• ν = 1

• p = 3.

Figure 3.1 shows the AM/AM curve resulting from equation 3.2, where ν and passume the aforementioned values.

The back-off parameter can be defined as input back-off (IBO) or output back-off(OBO):

IBO = 10log10A2

s

〈PIN〉 dB (3.3)

1In the present work we will always refer to Release 99 (R99) of the standard, unless differentlystated.

40

Figure 3.1: Rapp’s model AM/AM characteristic as obtained implementing equa-tion 3.2 with a Matlab c© function.

OBO = 10log10A2

0

〈POUT 〉 dB (3.4)

In this work we will always use the definition in 3.3.

WCDMA standard imposes tight limits on both OOB emissions and signal qual-ity [33]; therefore solutions have to be found to improve amplifier efficiency withinthe limits of the regulation. This is the reason why many efforts have been donein studying PAPR reduction techniques [30], which should guarantee PA linearitywithout resorting to large BO values. In this work, we focus our attention on a par-ticular category of PAPR reduction strategies: signal distortion techniques, whichreduce the peak amplitudes by distorting the signal at or around peaks. Amongstthe available options for such category, we applied three different techniques to aWCDMA transmitter and checked their effectiveness in terms of the parameters im-posed by the standard. Simulation results obtained in each case have been comparedin order to find out which strategy is more suitable for implementation.

41

3.2 WCDMA uplink air interface: Physical chan-

nels, Spreading and Modulation

The 3GPP uplink modulation scheme [4] allows up to six Dedicated Physical DataChannels (DPDCHs) and one Dedicated Physical Control Channel(DPCCH). Theuplink DPDCH transports dedicated data generated at OSI Layer 2 and above,i.e. the dedicated transport channel (DCH)2, whilst the uplink DPCCH is used tocarry control information, such as pilot bits to support channel estimation, trans-mit power-control (TPC) commands, feedback information (FBI) and an optionaltransport-format combination indicator (TFCI) [3]. Physical channels are struc-tured in frames of 10 ms duration, each frame is split into 15 slots of fixed length(equal to 2560 chips). A super-frame corresponds to 72 consecutive frames, i.e. itslength is 720 ms. The frame structure for the uplink dedicated physical channel isshown in Figure 3.2, where the parameter k indicates the number of bits per slotand it is directly related to the SF of the channel as SF = 256/2k.The channels’ configuration is shown in Figure 3.3. WCDMA uses a two level code

Figure 3.2: Frame structure for uplink DPDCH/DPCCH [3].

system: orthogonal spreading codes (cd,n) and pseudo random scrambling codes(Sdpch,n). In order to support variable data rates, the air interface allows to pick

2There might be more than one DPDCH per active connection but always only one DPCCH.

42

Figure 3.3: DPDCHs and DPCH uplink configuration [4].

up spreading codes with different Spreading Factor (SF) and this family of codesis called Orthogonal Variable Spreading Factor (OVSF). In the uplink OVSF codesare used to separate data and control channels from a specific user. After spread-ing control and data channels are I/Q multiplexed. Scrambling codes identify aspecific mobile terminal and it is applied on top of spreading. A detailed descrip-tion of WCDMA uplink physical layer is beyond the scope of the this work, furtherdetails can be found in [26] [34] other than in the related 3GPP specifications [3] [4].

For uplink data rate below 450 kbps, a single DPDCH is always used, in whichcase the in-phase component is made of the data channel and the in-quadraturecomponent is obtained from the control channel (DPCCH). The data rate of theDPDCH can be variable and different values of SF can be used during the trans-mission, whereas the control channel rate is fixed and it uses a spreading code withSF= 256. Table 3.4 contains the bit and symbol rates available (before spreading)for a data channel according to the SF used. When employing multiple data chan-nels the SF is set to 4 for each DPDCH, which corresponds to highest available bitrate (around 2 Mbps), whilst the SF for the control channel remains unchanged at

43

Figure 3.4: Uplink DPDCH available bit/symbol rates along with correspondentSF [3]. The number of bits per DPDCH is determined by the parameter k (slotformat) according to the following equation: Nbits = 10 · 2k, k = 0, . . . , 6.

256. It is worth to point out that only three codes are available at the SF of four,consequently the codes are reused on the in-phase and quadrature branches.βc and βd are adjustable weighting parameters which help to maintain constant thesignal-to-noise ratio of both data and control channels 3.The combined data and control channels are multiplied by a complex scramblingcode (Sdpch,n), which has been designed to avoid 180 degrees transitions from onechip to another, hence leading to a reduction in PAPR. This modulation schemetakes the name of Hybrid Phase Shift Keying (HPSK) and it relies on a proper com-bination of spreading and scrambling codes, a comprehensive explanation of HPSKcan be found in [35]. The PAPR for a single DPDCH is lower than for standardQPSK thanks to HPSK modulation scheme (see section 3.3).The bandwidth of the modulated signal is defined by filtering the I and Q com-ponents with Root Raised Cosine (RRC) filter with a roll-off factor of 0.22 and abandwidth of 1.92 MHz (half the chip rate). The filtered signal trajectory for asingle DPDCH with βc = βd = 1 is depicted in Figure 3.5 (a): it is worth to noticethe low number of 180 degrees phase shifts hence proving the effectiveness of HPSKmodulation. As a comparison, in Figure 3.5 (b) the vector diagram obtained from awrong set of spreading codes is shown: the number of zero crossings has noticeablyincreased.

3Throughout this work both these values are always set to 1.

44

(a) (b)

Figure 3.5: Signal trajectory for an RRC filtered single data channel when usingcorrect HPSK (a) and employing a wrong choice of spreading codes (b).

3.3 Distribution of PAPR and its effect on PA

In Figure 3.6 the CDFs for the PAPR of single data channel and multiple datachannels WCDMA signals are shown. With regard to the 1 DPDCH case, it is easyto notice that HPSK modulation effectively keeps the PAPR within 3.5 dB, which isa remarkably low value. On the other hand, when employing six data channels, thePAPR rises up to a maximum value of around 8 dB. Due to amplifier’s nonlinear-ities, the effects of a higher PAPR are the increased level of out-of-band emissionsand decreased performance because of bad modulation accuracy (or in-band dis-tortion) [36]. As mentioned in section 3.1, power amplifiers have to operate neartheir saturation point to reach high power efficiency, unfortunately this region ofthe AM/AM characteristic is non-linear, therefore -if the signal’s PAPR increases-the PA is forced to work in its saturation region and this causes the negative effectsdescribed above. As an example, Figure 3.7 illustrates the spectrum of WCDMAsignal at the output of the PA when 1 and 6 data channels are employed: thelevel of OOB radiation is clearly higher in the latter case than in the former. Asolution might be setting the amplifier operation point back to the linear regionby introducing a back-off to the input, but this would result in a lower efficiency.Alternatively, PA with a wider linear characteristic could be used to overcome theunwanted effects, unfortunately very linear amplifiers are both very expensive andscarcely efficient. Furthermore, WCDMA standard sets strict limits on signal’s qual-ity (i.e. allowed level of in-band distortion) and adjacent channel power leakage (i.e.OOB emissions), in the next section ( 3.4) such requirements will be discussed in

45

3 3.5 4 4.5 5 5.5 6 6.5 7 7.5 8 8.5 90

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

PAPR [dB]

PAPR − CDFs

Figure 3.6: Cumulative distribution functions of PAPR for a filtered WCDMA signalwith a single data channel (blue line) and 6 data channels (red line).

(a) Spectrum at the output of the PA. 1DPDCH.

(b) Spectrum at the output of the PA. 6DPDCH.

Figure 3.7: Effects of an increased PAPR due to PA nonlinearities.

detail.In this scenario clearly appears the necessity of finding a trade-off between all theseclashing requisites. Summarising, we can conclude that the challenging goal aheadconsists in maximising the PA efficiency by reducing signal’s PAPR without undulycompromising the performance and complying with the restrictions imposed by the

46

standard [37].It is now evident that PAPR reduction techniques represent essential features toachieve the objective because they allow to reduce the PAP ratio and consequentlyto reach a higher efficiency. The main drawback of this approach is given by thedistortions introduced on the signal, which is going to inevitably have a lower qual-ity, hence worse performances (e.g. BER). However had such techniques not beenused a large back-off value would be required thus drastically affecting the systemperformance.In section 3.5 we will go through the details of clipping schemes we applied to limitthe PAPR.

3.4 Radio Specifics for WCDMA transmitter

3GPP standard for WCDMA specifies the following signal-quality requirements forthe waveform at the output of a 3G transmitter [33]:

• Adjacent Channel Leakage Ratio (ACLR) which determines how much oftransmitted power can leak into the first (ACLR1) or second (ACLR2) neigh-bouring carrier. ACLR is defined as the ratio of the RRC filtered mean powercentered on the assigned channel frequency to the RRC filtered mean powercentered on the adjacent channel frequency. ACLR1 should be larger than33 dB and ACLR2 than 43 dB 4.

• Error Vector Magnitude (EVM) which is a measure of the difference betweenthe reference waveform and the measured waveform. This difference is calledthe error vector. The EVM result is defined as the root of the ratio of themean vector power to the mean reference power expressed in %. EVM valueshould be below 17.5 %.In fact, EVM is a measure of the signal’s modulation quality, it helps to imme-diately highlight possible causes of degradation, such as signal compression,local oscillators timing errors, I/Q impairment, etc. In digital modulationthe signal’s amplitude and phase can be measured at any time. Such valuesdefine the actual or ’measured’ phasor. Similarly, a corresponding ideal or’reference’ phasor can be calculated, given the knowledge of data transmit-ted, clock timing and so on. The differences between these phasors form thebasis for EVM measurement [5]. A graphical explanation of this concept isin Figure 3.12, actually EVM is the scalar distance between the two phasors(i.e. the magnitude of the difference vector) and, by convention, is usually re-ported as percentage of the peak signal level. In other words, EVM measures

4The nominal channel spacing for WCDMA is 5 MHz, therefore the first adjacent channel is at5 MHz and the second at 10 MHz.

47

the difference between the transmitted signal and the ideally-modulated one.

Figure 3.8: Error Vector Magnitude (EVM) and related quantities [5].

• Peak Code Domain Error (PCDE) which is computed by projecting the powerof the error vector (as defined for EVM) onto the code domain at a specificspreading factor. The Code Domain Error for every code in the domain isdefined as the ratio of the mean power of the projection onto that code, tothe mean power of the composite reference waveform. This ratio is expressedin dB. The Peak Code Domain Error is defined as the maximum value for theCode Domain Error for all codes. PCDE value shall not exceed −15 dB forspreading factor 4.This figure of merit of the signal’s quality has been specifically introducedin WCDMA standard to address the possibility of uneven error distribution.This test is required only for multicode transmission.

In the present work we considered ACLR1, EVM and PCDE as limiting factorsthat considerably affect the design of power amplifiers for UMTS terminals [38] [39].ACLR2 has not been included in the analysis because the allowed level of emissionsis, in any case, below the threshold. In Figure 3.9 (a) the ACLR1 value for anunclipped 6 data channels signal is plotted as a function of the IBO of the amplifier,the ACLR1 limit is met for back-off values greater than 5 dB only, whereas theACLR2 value is above the threshold of 43 dB regardless of the IBO applied to thePA (Figure 3.9 (b)). In the following of this work we will benchmark the performanceof different PAPR reduction techniques which guarantee a level of spectral splatterabove the minimum limit imposed by ACLR1.

48

0 1 2 3 4 5 6 720

22

24

26

28

30

32

34

36

38

IBO [dB]

AC

LR1

[dB

]

ACLR1 − 6 DPDCHs

ACLR1 − 6 DPDCHsACLR1 limit 33 dB

(a)

0 1 2 3 4 5 6 742

44

46

48

50

52

54

56

58

60

62

IBO [dB]

AC

LR2

[dB

]

ACLR2 − 6 DPDCHs

ACLR2 − 6 DPDCHsACLR2 limit 43 dB

(b)

Figure 3.9: Power leakage effect on adjacent channels of a 6 DPDCHs signal througha nonlinear amplifier: ACLR1 vs IBO (a). ACLR2 vs IBO (b).

3.5 PAPR reduction techniques

Several approaches have been proposed to overcome the problems posed by signalswith high PAPR (a comprehensive summary can be found in [40]). Among thesolutions devised, we decided to chose three techniques that belong to the categorynamed signal distortion techniques. As the name suggests, these schemes aim atreducing PAPR by directly modifying the modulated signal waveform. The reasonfor this choice essentially lays in the simplicity of the approach proposed, which fitswell with the standard boundaries and the requirements of a future implementation.The following sections include a detailed description of the techniques we appliedto the model of a WCDMA mobile terminal..

Clipping and Filtering

The simplest way to reduce the PAPR is clipping the signal, such that peak ampli-tude becomes limited to a desired level, A. We define the Clipping Ratio (CR) asthe clip level A over the rms of the signal:

CR = 20 log10

(A

σ

)dB (3.5)

The main drawback related to this approach is the introduction of in-band and OOBinterference, as clipping is a non-linear operation it causes distortion by giving birth

49

to spurious harmonics (exactly the same effect as real amplifier). Self interference(or in-band distortion) degrades the signal quality which has to stay within theallowed values of EVM and PCDE; on the other hand, OOB emissions must be keptwithin the limits imposed by the ACLR. Filtering after the clipping stage reducesthe spectral splatter, but causes peak regrowth [41]. Despite these side-effects,this simple approach proved to be effective in reducing the amplitude signal rangeand suppressing unwanted OOB emissions. Obviously, the performance will mainlydepend on the characteristics of the filter used. In our case, the filter employed inthis work is a simple equiripple FIR filter with stopband attenuation around 40 dB.

Peak Cancellation

As pointed out in the previous section the main disadvantage of clipping is repre-sented by OOB radiations due to the sharp corners present in the clipped versionof the signal. This undesirable effect can be avoided by doing a linear peak cancel-lation: a time-shifted and scaled reference function is subtracted from the originalsignal in order to reduce the peak power. This technique could not cause anyspectral splatter if a proper function (with the same bandwidth as the transmittedsignal) is selected. One example of suitable function is the sinc function, unfortu-nately it is of no practical use as it has an infinite support. Hence, for practical use,time-limited functions with adequate spectral properties are created.Figure 3.10 (a) shows part of the truncated sinc function which we used as correctionfunction (other types of reference functions can be found in [42] [43]). The spectralproperties of such function assure low OOB emissions as its attenuation outside thesignal’s bandwidth is around 70 dB (Figure 3.10 (b)), conversely no attempts aremade to reduce the in-band distortion.The peak cancellation scheme suggested in [43] employs a correction function whichsuppress both in-band and out-of-band clipping noise. Due to the strict require-ments posed by the filter’s mask specifics in that particular case, the outcome is acorrection function with not enough 5 OOB suppression capability, that is the mainreason why we preferred to let down in-band attenuation and focus on out-of-bandone.

Peak Windowing

Like clipping and filtering and peak cancellation, peak windowing is a signal distor-tion technique (sec. 3.1). We have already pointed out that conventional clippingcauses sharps corners in a clipped signal. This leads to an increased spectral splatterand to a higher ACLR. Peak windowing is meant to reduce the level of unwantedemissions by smoothing the sharp corners related to the hard clipping process. The

5According to the 3GPP specifics [33].

50

500 550 600 650 700 750

−0.2

0

0.2

0.4

0.6

0.8

1

samples

ampl

itude

Reference Sinc function for Peak Cancellation

(a)

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9

−70

−60

−50

−40

−30

−20

−10

0

10

20

Normalized Frequency (´π rad/sample)

Mag

nitu

de (

dB)

Magnitude Response (dB)

(b)

Figure 3.10: Truncated sinc function used in the peak cancellation algorithm (a)and its spectral properties (b).

goal is achieved multiplying the signal with a window function. The spectral prop-erties of this function determine the amount of out-band emissions. Conventionalclipping can be expressed as a multiplication [37]:

xclip(n) = c(n)x(n) (3.6)

where

c(n) =

{1, |x(n)|≤A

A|x(n)| , |x(n)|>A

(3.7)

where A is the clipping level. The idea of this method is to replace the functionc(n) with the following function:

b(n) = 1−+∞∑

k=−∞akw(n− k) (3.8)

w(n) is the window function and ak is a weighting coefficient. To achieve thedesired clipping level the funciton b(n) must satisfy the inequality

1−+∞∑

k=−∞akw(n− k) ≤ c(n) (3.9)

for all n. The difference between c(n) and b(n) depends on the window length andthe coefficients ak. Once the window is chosen, the weighting coefficients ak must

51

be optimised. Usually it is assumed that the window length is so small (accordingto the clipping probability) that the windows do not overlap and the easiest wayto calculate the optimal ak is convolving 1 − c(n) with the window function w(n),yielding to the following expression for b(n):

b(n) = 1−+∞∑

k=−∞[1− c(k)]w(n− k) (3.10)

Unfortunately in a real case windows overlap and, as a result, the signal is clippedmore than needed (overclipping). Many solutions have been proposed to avoid thisproblem [44] [37].In this work we used an approach similar to [44] to overcome overclipping. Thewindow employed is a Hann with 49 samples. This number of samples sets a goodcompromise between the overclipping problem and spectral performance. In Fig-ure 3.11 the window as used in our simulations is depicted.

0 10 20 30 40 500

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

samples

ampl

itude

Hann Window for Peak Windowing

Figure 3.11: Hann window with 49 samples. This is the function applied to ourpeak windowing scheme.

3.6 WCDMA Uplink Transmitter Simulator

The performances of the clipping techniques described in the above sections havebeen evaluated through a Simulink c© model of a standard-compliant WCDMA

52

Figure 3.12: WCDMA-compliant transmitter Simulink c© model developed at PRL.

transmitter.The block scheme of the implemented simulator is shown in Figure 3.12. The modelcomprises a standard transmitter and a simplified WCDMA receiver. The trans-mitter’s modulation and spreading strategy completely fulfils the specifics in [3].

3.6.1 WCDMA uplink transmitter

Figure 3.13 shows the constituent elements of the model’s Data generation andspreading block (see Figure 3.12). Data are generated according to the frame andslot structure specified by the standard and summarised in section 3.2. Each framehas a length of 10 ms and it is divided into 15 time slots. The number of chipsper slot is set to 2560 regardless of the spreading factor, whilst the number of bitsper slot depends on the SF employed. The model we devised can support each ofthe slot format allowed by the standard specifics (see Figure 3.4), although whenusing multicode transmission the only SF value available is 4, which yields to 640bits per slot of a data channel (DPDCH). As already mentioned in section 3.2, theuplink control channel (DPCCH) has a fixed rate and uses a constant spreadingfactor set to 256, which corresponds to 10 bits per slot. On each channel, bits aregenerated independently according to uniform random distributions with differentinitial statistical seeds.

53

After this stage, data are passed on to the spreading block in which data andcontrol bits are multiplied by the OVSF codes, that are allocated according to therule defined in [3]. At the output of the spreading block, each time slot is a sequenceof 2560 chips. The model then provides I/Q multiplexing of the physical channelsfollowing the scheme in Figure 3.3. At this stage the signal must be multiplied by

Figure 3.13: Constituent elements of the WCDMA model’s block that provide datageneration and spreading operation as specified in [3].

the long scrambling code: the result of such operation is a signal with a noise-likespectrum. It is also worth to remind that scrambling codes identify specific usersin a network, hence they have to be uniquely assigned to each mobile terminal thataccess the network. This means that the uplink scrambling sequence has to belong enough to provide a huge number of different codes, in WCDMA standard alength of 38400 chip has been chosen (corresponding to one frame, i.e. 10 ms). Thegeneration of such code is quite complicated and a comprehensive analysis would gobeyond the scope of the work, nonetheless the standard indicates how generatingthe long scrambling code and the relative details are available in 3.2. Figure 3.15shows the Simulink c© implementation of the scrambling sequence generator used inour model. After the scrambling stage, the signal is fed into the chip shaping block.The Chip Shaping block provides in-phase and in-quadrature modulation with aninterpolation factor I = 6, clipping is performed after this stage for a more effectivepeak capturing [30]. The RRC filter roll-off factor α has been set to 0.22, whichcorresponds to a signal bandwidth of 4.68 MHz, given the chiprate= 3.84 Mcps.The simulated signal’s spectrum at the output of the chip shaping filter block isdepicted in Figure 3.14.

Once the root raise cosine filtering has been applied to both in-phase and in-

54

Figure 3.14:

quadrature components of the signal, the uplink transmitter model that we con-ceived allows either processing the modulated signal with clipping techniques orfeeding it directly into the PA without any further manipulation (see Figure 3.12).Furthermore the block named Parameters settings gives the user the possibility of

Figure 3.15: Implementation through Simulink c© blocks of the long scrambling codegenerator as used in the WCDMA transmitter model we developed.

55

easily varying some key-parameters of the transmitter,i.e. choosing between thethree clipping strategies described in section 3.5, setting the IBO of the amplifieras well as the clipping ratio level (CR).PA model is the one described in section 3.1. The amplified WCDMA signal is sentto the receiver through a gaussian channel; the noise power introduced by the chan-nel can be adjusted simply typing the desired value into a tab of the ParametersSettings block.

3.6.2 WCDMA downlink receiver

The receiving end comprises RRC receiving filters and de-scrambling/de-spreadingblocks (see Figure 3.12).De-scrambling operation consists in multiplying the received signal by a complexconjugated copy of the scrambling code used at the transmitter. Similarly, informa-tion bits are demodulated by multiplying each DPPCH/DPCCH by a synchronisedcopy of the OVSF codes followed by an integration over the symbol period, whichcompletes de-spreading delivering a voltage value per symbol to the hard-decisionthreshold block. Figure 3.16 illustrates the blocks employed to build up the sim-ulator’s de-spreading stage. Overall the receiver structure is quite basic becauseof the absence of any kind of channel impairment, i.e. multipath propagation, orsynchronisation issues. In fact, PAPR analysis does not need that level of detailtherefore such additional features have not been included in the simulator design,although - thanks to its flexility - they may be added without significant effort.

Figure 3.16: WCDMA simulator: despreading stage and bit demodulation.

56

3.6.3 ACLR, EVM and PCDE blocks

ACLR is calculated by means of a bank of RRC filters with the same chiprateas the shaping filters and centered on the two adjacent channels at 5 MHz and10 MHz [33]. Figure 3.17 shows the Simulink c© implementation of the block usedin our simulation that measures ACLR values.

Figure 3.17: Details of the block that calculates the out-of-band power leakage(ACLR) of the WCDMA transmitter.

EVM and PCDE have been evaluated6 following the approach suggested in [45]and [39]. A brief, although exhaustive, description of the methods used to calculatethese parameters is presented in this section.

EVM and PCDE are two quality metrics that estimate the amount of distortionon the transmitted signal introduced by non-linear devices, such as real power am-plifiers, mixers, possible clipping stages, etc. According to the definitions given insection 3.2, EVM represents the difference -in the time domain- between the ’ide-ally modulated’ signal and the one actually transmitted. In real testing equipment,EVM can be calculated by substituting to the ’ideally modulated’ signal a referencesignal derived from the real waveform at the transmitter’s output. Such signal isconstructed by the measuring equipment demodulating the signal under test (i.e.at the output of the transmitter) and remodulating it ’perfectly’ according to thestandard specifications. This error free version of the signal is then compared the

6WCDMA standard does not indicate a specific algorithm to calculate these quality metrics.Only their definition as reported in 3.4 is given.

57

actual output of the transmitter under test. Defining the reference signal as a vec-tor, R, of N = n×m complex samples, where n is the number the of symbols in themeasurement interval and m is the number of chips per symbol, and Z the vectorcontaining N samples collected from the signal under test, then the error vector Ein the time domain is:

E = Z −R. (3.11)

Given E, EVM is calculated according to the following equation:

EV M =RMS(E)

RMS(R)× 100%, (3.12)

where the RMS (Root Mean Square) is calculated over the entire measurement in-terval, which is usually a time slot.PCDE is calculated by projecting the time domain error signal onto the channeliza-tion codes domain and selecting, amongst the obtained values, the maximum error.In order to achieve correct PCDE results, the error vector E must be divided into ntime-sequential vectors, e, with m complex samples comprising one symbol interval.Given the matrix C, containing the vectors of the orthogonal channelization codesbelonging to one spreading factor7, the inner product between C and e is calculatefor all the symbols in the measurement interval, the result is a k×n matrix (k is thenumber of codes associated with the spreading factor) and its values represent anerror voltage corresponding to a specific symbol and code. The root mean squarevalue is then performed over each row of the matrix, thus giving a vector, ek, of kelements. Amongst these k values the peak, ek, is selected, finally PCDE can beestimated through the following equation:

PCDE = 10 log10

((ek)

2

RMS(R)2

)[dB]. (3.13)

Figures 3.18, 3.19 illustrate the solution devised in our model to implement theEVM and PCDE calculation algorithms described above.

3.7 Simulations Results: a comparison

The goal of the simulation work done was establishing which is the best clippingscheme for a WCDMA transmitter. Other than compliancy with the standard’srequirements in terms of signal’s quality (EVM, PCDE) and power leakage in neigh-bouring channels, two additional parameters steered our analysis: BER performanceand simplicity, i.e. suitability for implementation.

7In uplink multicode transmission SF= 4 is the only allowed, see section 3.4.

58

Figure 3.18: Constituent blocks of the simulator’s module that estimates EVM.

Figure 3.19: Constituent blocks of the simulator’s module that estimates PCDE.

The first step in evaluating the BER performance in each single case was takinginto account the sole ACLR1 as limiting factor. The back-off of the PA has been setto the minimum value that gives an ACLR1 with a margin of at least 1 dB over thelimit (33 dB). Given this value, the performance in terms of BER has been assessed.Figure 3.20 shows the BER curve for clipping and filtering with CR= 3 dB com-pared to the performances of the unprocessed signal (no IBO applied, hence therequirements on adjacent channel power leakage are not met) and the unclippedcase, where a back-off of 5.5 dB is used to comply with the ACLR restrictions. Itis worth to notice that clipping is clearly effective in increasing PA efficiency as theunclipped signal has worse performance, the amount of distortion introduced by theclipping process is small enough to achieve a gain around 1 dB in signal-to-noiseratio. Now the back-off needed is decreased to 4.1 dB, which confirms that the PAcould be used more efficiently.In Figure 3.21 and Figure 3.22 the results obtained for peak cancellation and peak

windowing are presented. Peak cancellation outperforms both clipping and filtering

59

22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 4610

−6

10−5

10−4

10−3

10−2

10−1

100

AWGN power (dBm)

BE

R

BER performance in AWGN channel

BO=0 dB unprocessed signalBO=5.5 dB unprocessed signalClip&Filt BO=4.1 dB CR= 3 dB

Figure 3.20: BER performance over AWGN channel for Clipping and Filter-ing (green dashed-dotted line), Unprocessed signal (blue dashed line) and Unclippedsignal (red line).

and peak windowing, which has the worst BER performance. IBO values requiredto work above the ACLR1 limit were respectively decreased to 3.2 and 3.1 dB,suggesting that these techniques reduce the signal peaks sharply around the clip-ping level, whilst clipping and filtering suffers from peak regrowth which forces thedesigner to resort to a larger back-off. However, the improvement given by PeakCancellation is just slightly better than the one reached by clipping and filteringand we may claim that they have same roughly performances.

In order to clarify and complete the analysis simulation results, including EVMand PCDE, have been summarised in Table 3.1. For the sake of simplicity, we pre-sented BER at fixed noise power level (30 dBm). The column indicated as ’Back-off’contains the results of the unclipped case, i.e. where the only back-off has been ap-plied. Table 3.1 effectively helps to understand which is the PAPR reduction schemethat sets the best compromise between BER performance and fulfilment of the stan-dard’s requisites. We have already pointed out that BER performances are quiteclose for all the three techniques examined, none of them has a stark advantage overthe others. So it now interesting to note that clipping and filtering introduces lesssignal distortion having lower EVM and PCDE values. At this stage, we can statethat, among the techniques tested, there is not a clear ’winner’, all of them seem

60

22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 4610

−6

10−5

10−4

10−3

10−2

10−1

100

AWGN power (dBm)

BE

R


BO=0 dB unprocessed signalBO=5.5 dB unprocessed signalBO=3.2 dB; CR=3 dB Peak Cancellation

Figure 3.21: BER performance over AWGN channel for Peak Cancellation (greendashed-dotted line), Unprocessed signal (blue dashed line) and Unclipped signal (redline).

to be suitable for implementation in a real system. Nonetheless clipping and filter-ing could represent the most appealing solution due to its straightforward simplicity.

In general, simulation results have clearly shown that employing PAPR schemesis not only essential but also advantageous for WCDMA terminals. Comparing theBER values of the unprocessed signal case to the ones obtained by applying PAPRreduction strategies, we can observe a notable gain in performance: passing from6.0 10−2 to a minum of 3.86 10− 2 for peak cancellation (best values which duly re-spects the imposed limitations). Yet, a significant reduction in back-off requirementhas to be added to the aforementioned improvement. Decreasing the back-off valuesallows a more effective use of the PA amplification range, which is a key-aspect in thedesign of final stage amplifiers. In order to further improve the PA efficiency (thusimproving BER performance), we set the CR to 1 dB. In this way, the signal suf-fers from more distortion and its quality will deteriorate, on the other hand PAPRis now greatly reduced, consequently PA’s efficiency should be enhanced leadingto better BER performance. Table 3.2 illustrates the simulation results obtained.The IBO values which guarantee the required ACLR1 margin are now respectively:2.6 (clipping an filtering), 1.3 (peak cancellation) and 0.8 dB (peak windowing).It is worth to stress out that clipping and filtering and peak cancellation improve

61

22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 4610

−6

10−5

10−4

10−3

10−2

10−1

100

AWGN power (dBm)

BE

R


BO=0 dB unprocessed signalBO=5.5 dB unprocessed signalBO=3.1 dB; CR=3 dB Peak Windowing

Figure 3.22: BER performance over AWGN channel for Peak Windowing (greendashed-dotted line), Unprocessed signal (blue dashed line) and Unclipped signal (redline).

their BER performance compared to the previous case, whilst peak windowing isless effective and has a higher error rate value. This is probably due to excessivelevel of distortion introduced by the clipping stage, which outbalance the benefitsof a lower PAPR. As a confirm, the quality parameters assume values that do notcomply with the standard. Again peak cancellation takes the most advantage outof the lower clipping ratio in terms of BER and ACLR1, unfortunately the qualitymetrics requirements are not met as EVM is greater than 17%.Actually, the only clipping scheme that would allow to operate within the regulationis clipping and filtering, even though its modulation accuracy is dangerously closeto admittable limits (EVM= 14.3%).

3.8 Discussion and Conclusions

The multicode transmission option in uplink WCDMA standard is a crucial fea-ture to support those services that require higher bit rate, e.g streaming applica-tions. The superimposition of numerous channels (up to six) inevitably increasesthe PAPR, i.e. large amplitude peaks are present in the modulated signal. HighPAPR puts strain on the mobile’s amplifier forcing it to work beyond its saturationpoint. The consequences stemming out from this unwanted effect are essentially

62

Table 3.1: Simulated values of ACLR1, BER, EVM and PCDE of the clippingstrategies tested. CR= 3 dB.

CR=3 (dB) Back-Off Clip & Filt. Peak Canc. Peak Wind.

ACLR1 (dB) 34.9 34.5 34.3 34.2

BER @ 30 dBm Noise Pwr 6.0 10−2 4.44 10−2 3.86 10−2 4.33 10−2

EV M (%) 3 8.5 12.7 12.5

PCDE (dB) -38.9 -29.2 -25.7 -25.9

Table 3.2: Simulated values of ACLR1, BER, EVM and PCDE of the clippingstrategies tested. CR= 1 dB.

CR=1 (dB) Back-Off Clip & Filt. Peak Canc. Peak Wind.

ACLR1 dB 34.9 34.4 34.5 34.2

BER @ 30 dBm Noise Pwr 6.0 10−2 3.73 10−2 3.35 10−2 4.61 10−2

EV M (%) 3 14.3 20.5 20.5

PCDE (dB) -38.9 -24.6 -21.5 -21.5

63

two:

• in-band and out-of-band interference/distortion

• decreased amplifier efficiency

WCDMA sets the limits of tolerable modulation distortion (EVM and PCDE) andPA efficiency cannot be too low otherwise terminals would need to recharge theirbatteries frequently. The only viable solution to overcome this such detrimentaleffects consists in reducing the PAPR and many approached have been devised toaccomplish the task. In this work we developed a WCDMA-compliant model of atransmitter and simulated its performance applying a particular category of PAPRreduction techniques (see section 3.5). Among the possible option available in suchcategory, we picked up three schemes, namely: clipping and filtering, peak cancella-tion and peak windowing. System performance in terms of BER, spectral splatter(ACLR1) and modulation quality (EVM, PCDE) have been evaluated in each caseexamined. It is worth reminding that our goal was finding a suitable clipping schemewhich could reach the best trade-off between performance e implementation flexi-bility.From the results we have shown in the previous section we are able to draw someconclusions about the effectiveness of the clipping scheme considered:

• Clipping the WCDMA multiple data channels signal is useful to reduce itsPAPR and restore part of the PA efficiency which would otherwise lost inlarge back-off values needed to meet the requirements of power leakage.

• Among the technique we took into account peak cancellation has the bestperformance in terms of BER. At CR= 3 dB both peak cancellation andclipping after filtering outperform peak windowing, although they show similarresults. Clipping and filtering adds less distortion with respect to the othertechniques showing better EVM and PCDE results. As expected IBO cannow assume lower values thus improving the PA efficiency.

• When the clipping ratio is decreased to 1 dB peak cancellation again setsitself as the more effective solution in terms of BER performance, but it doesnot meet the modulation accuracy requisites being EVM greater than the17% threshold. Peak windowing presents BER values even worse than theprevious case (with CR set to 3 dB), EVM is also out of range as for peakcancellation. Clipping and filtering is the only scheme that allows the systemto operate within the standard limitations, even if an EVM of 14.3% is veryclose to the permitted level. It seems evident that decreasing the clippingratio introduced additional distortion that outbalanced the benefits derivingfrom a reduced signal’s PAPR.

64

• Concluding we can claim that simple signal distortion techniques like the onewe analysed are effective in limiting the signal’s amplitude range accordingto the limits imposed both by the standard and the PA nonlinearities. Inparticular, we noticed that applying PAPR strategies the system is able toachieve better BER performance with respect to the unprocessed signal case.This due to a more efficient use of the PA. On the other hand, the signalquality is deteriorated and the CR cannot be too low otherwise the qualityrequirements are not met (see Table 3.2).

• With regard to suitability for possible implementation, we must conclude thatnone of the three schemes has shown performances remarkably higher than theothers, therefore we would indicate clipping and filtering as the most appealingbecause it merely based on a digital FIR filter.

65

Chapter 4

WiMAX modelling

4.1 Introduction

In chapter 3 we highlighted the issues related to signals with large values of Peak-to-Average-Power Ratio (PAPR) in WCDMA systems. Basically, a large dynamicrange makes the signal very sensitive to nonlinearities due to the transmitter PowerAmplifier (PA) or other nonlinear devices. We pointed out that this result in un-wanted out-of-band and in-band distortion, which degrades the signal quality andaffect the performance of the whole system. Unfortunately, the same problems areencountered when employing another modulation scheme, namely Orthogonal Di-vision Frequency Multiplexing (OFDM) 1. An OFDM signal consists of a numberof independently modulated subcarriers, which can lead to high PAPR when addedup coherently. PAPR values increase with the number of subcarriers employed, astatistical analysis has been carried out in [40] for QPSK symbols and generalisedto QAM constellations in [46]. Simulation results show that an OFDM signal canreach a PAPR greater than 12 dB for 1024 QPSK modulated subcarriers, even ifthe probability of such event is extremely low (0.001%).A physical layer (PHY) based on OFDM technology has been adopted in many mod-ern digital communications systems, i.e. the IEEE 802.11 Wireless LAN (WLAN),Digital Video Broadcast (DVB), as well as the IEEE 802.16 Broadband WirelessAccess (BWA). In particular, a broad industry consortium, the Worldwide Interop-erability for Microwave Access (WiMAX) Forum has started certifying productsthat comply with the standard (and its subsequent amendments) developed by the802.16 group. In the following sections of this chapter we will focus our attention onthis recent technology providing a brief overview of WiMAX and its salient features,many of which are included in the model we built up in order to assess the effects of

1In this chapter we will not go through the basics of OFDM. Detailed information can be foundin many published books and websites, e.g. [40] [46], http://en.wikipedia.org/wiki/OFDM

67

nonlinearities on the system. The remaining part of this chapter will be dedicatedto a detailed description of the simulator, after which we will address the issuesarising from large peak values against the strict quality requirements imposed bythe standard.

4.2 WiMAX: an overview

Originally the IEEE 802.16 group focused on developing air interface standard forLOS point-to-multipoint wireless broadband applications in the 10−66 GHz band.The group subsequently produced a new standard, 802.16a, targeting NLOS com-munications in the 2 − 11 GHz spectrum, using an OFDM-based physical inter-face. Further revisions in the following years resulted in a new standard in 2004,802.16 − 2004 [47], which forms the basis for WiMAX. Being a collection of stan-dards rather than a single interoperable one, 802.16 − 2004 allows several designoptions, such as a variety of different choices in PHY and MAC layers features [46].For obvious practical reasons of interoperability, the scope of the standard had tobe restricted, hence the WiMAX Forum was created to define a limited set of sys-tem profiles, i.e. mandatory and optional features selected from the wide range ofoptions given by the standard.The first WiMAX solution based on 802.16 − 2004, known as fixed WiMAX, ad-dressed fixed applications that could represent a cost-effective alternative to DigitalSubscriber Lines (DSL) and cables. In December 2005, the IEEE group approvedan amendment to 802.16d, namely 802.16e − 2005 (or simply 802.16e) [48], whichadded mobility support. 802.16e standard paved the way for mobile broadbandservices and it is usually referred as mobile WiMAX. The air interface for mobileWiMAX is based on Orthogonal Frequency Division Multiple Access (OFDMA),which introduces the possibility of dividing the available OFDM subcarriers intosubsets (or subchannels) to be allocated to different users. OFDMA will be pre-sented in greater detail in section 4.2.1. Similarly to the previous fixed version, theWiMAX Forum agreed a number of certified profiles that could ensure backwardcompatibility with fixed WiMAX and interoperability between the new mobile ver-sions. Currently the WiMAX Forum specified two different system profiles: onebased on IEEE 802.16 − 2004, OFDM PHY, called fixed system profile; the otherone based on 802.16e−2005 scalable OFDMA PHY, called the mobility system pro-file [46]. Besides, the WiMAX Forum, in its role as guarantor for interoperability,also defined certification profiles, which specify the operating frequency, the chan-nel bandwidth and duplexing mode. WiMAX compliant equipment are certified forinteroperability against a particular certification profile.Today there are five certification profiles for fixed WiMAX and fourteen for themobile version (see Table 4.1). All the initial profiles for mobile WiMAX use Time

68

Division Duplexing (TDD), although Frequency Division Duplexing (FDD) might beconsidered to address specific market opportunities where the spectrum regulatoryrequirements either prohibit TDD or are more suitable for FDD deployment [6]. It

Table 4.1: Fixed and Mobile WiMAX Initial Certification Profiles.

is also worth to point out that that the standard only covers PHY and MAC layers,this forced WiMAX Forum to provide guidelines for an end-to-end implementationof the system [6].In the reminder of this introduction we will focus on WiMAX salient features, es-pecially those introduced at physical layer level.

4.2.1 WiMAX salient features

WiMAX is a wireless broadband solution that supports a wide range of features witha lot of flexibility in terms of deployment options and potential service offerings.The main features offered by WiMAX can be summarised as follows [6] [46]:

• OFDM-based physical layer: As already mentioned, the WiMAX physicallayer is based on OFDM, which is a scheme that offers robustness towardsmultipath and allows WiMAX operating in NLOS conditions. FurthermoreOFDM is easily adaptable to different channelisation bandwidths.

• High data rates: The inclusion of MIMO antenna techniques coupled withflexible sub-channelisation schemes and Adaptive Modulation and Coding(AMC) enable Mobile WiMAX to achieve DL peak data rates per sector upto 46 Mbps, assuming a TDD system with 10 MHz channel and DL/UL ratio3:1, and UL rates up to 14 Mbps, given a DL/UL ratio 1:1.

69

• Scalability: In order to comply with different spectrum requirements, Mobile(and Fixed) WiMAX is able to scale to work in various channelisations from1.25 to 20 MHz. This scalability is supported in the OFDMA mode, where theFast Fourier Transform (FFT) size may be varied according to the availablechannel bandwidth.

• Support for TDD and FDD: A strength of WiMAX is that it can operateeither in TDD and FDD modes. TDD is the preferred option in most partof the system implementations 2 because it offers the possibility to adjust thedownlink/uplink ratio, thus asymmetric traffic can be efficiently supported.Other advantages consist in a less complex transceiver design and the abilityto exploit channel reciprocity. The main drawback related to TDD access isthe need for a synchronised network to counter interference.

• Quality of service: 802.16e MAC layer ha been designed to offer an end-to-end IP based QoS for various classes of services, such as VoIP (Voice overInternet Protocol), streaming audio/video, etc. The goal is reached througha flexible mechanism of resources allocation (scheduling) over the air interfaceon a frame-by-frame basis.

• Mobility: Mobile WiMAx allows optimised handover strategies with latencyless to 50 ms to ensure real-time applications such as VoIP to perform withoutservice degradation.

• IP-based architecture: The WiMAX forum has defined an all-IP net-work architecture. This facilitates integration with other networks and offersthe adavantage of a reduced total cost of ownership during the lifecycle of aWiMAX network deployment [6].

The features listed above make WiMAX a suitable solution for two deploymentscenarios:

• WiMAX as an access solution: as DSL complement to support Internetaccess and VoIP telephony where the cabled option is unfeasible or too ex-pensive, e.g. rural and remote subscribers. WiMAX can also be deployed todeliver wide area BWA, this is an attractive solution for customer who desirea high-speed internet access but do not wish to be tied down with the PSTNline. A WiMAX network might even complement other wide area networks,such as 3G, or be used in conjunction with other wireless technologies, likeWiFi. To date, South Korea is first country that developed and rolled out

2All the initial WiMAX profiles are based on TDD, except for two fixed WiMAX profiles in3.5 GHz

70

a version of mobile WiMAX, called WiBro (www.http://www.wibro.or.kr).WiBro delivers broadband internet service with a peak throughput (per user)of 3 Mbps in DL and supports mobility up to 60 Km/h.

• WiMAX as a transport/backhaul solution: can be used as base for de-velopment of point-to-point or point-to-multipoint wireless IP/Ethernet trans-port connections, e.g. for mobile networks or WiFi hotspots.

In the following section we will focus on the key-features of the WiMAX physicallayer.

PHY layer main characteristics

WiMAX radio interface uses OFDM-based technology, like WiFi and many otherstransmission systems. OFDM is an efficient scheme for high data rate transmissionin NLOS or multipath radio environment. OFDM modulation relies on the idea ofdividing a given high rate data stream into several parallel lower rate streams andnodulating each stream on separate carriers, called subcarriers. OFDM providesprotection against Inter Symbol Interference (ISI) by making the symbol time largeso that channel-induced delay are only a small fraction of the total duration. This isusually achieved by repeating part of the last OFDM data symbols at the beginningof the symbol, the technique is called cyclic prefix and the time duration of thecopied symbols is referred as guard time. Yet, OFDM is spectrally efficient as thesubcarriers are orthogonal one to each other over the symbol time. Another ad-vantage of this modulation scheme resides in its easiness of implementation in bothtransmitter and receiver: OFDM signals can be digitally generated using IFFT al-gorithm and received performing FFT.On the other hand, OFDM is very sensitive to frequency offset and synchronisation,this aspect limits the support of mobility to speeds up to 120 Km/h [6]. Anotherdisadvantage is associated with the signals presenting high PAPR, which causesdegradation in signal quality and forces designers to use very linear devices (e.g.power amplifiers, digital-to-analog converters) to avoid unwanted distortion effects.

Mobile WiMAX multiple access is based on OFDMA (Mobile WiMAX OFDMA-PHY), which is multiplexing scheme relying on subchannelisation 3 of both UL andDL to accommodate users’data streams. OFDMA symbol structure consists of threetypes of sub-carriers:

• data used for carrying data symbols.

3Subchannelisation, as well as frame and slot structure of WiMAX, are quite complex topicsand a comprehensive discussion would be beyond the scope of this work. A detail explanation canbe found in [46], besides the published standards [48] [47].

71

• pilot used for carrying pilot symbols, which are known a priori and are funda-mental for channel estiamtion/tracking. Power on pilot subcarriers is boostedby 2.5 dB to increase channel estimation reliability even at low signal-to-noiseratios.

• null have no power allocated to them. They are located at DC frequency andtowards the edges of the spectrum, guard subcarriers.

Figure 4.1 shows a frequency domain representation of a mobile WiMAX OFDMsymbol. Active sub-carriers (data and pilot) are grouped in sub-sets of sub-carriers

Figure 4.1: OFDMA subcarriers structure [6].

called sub-channel (48 data sub-carriers +6 pilot tones). In principle, the networkassigns a sub-channel for each active user, thus avoiding intra-cell interference. Infixed WiMAX (fixed WiMAX OFDM-PHY), subchannelisation is limited tothe UL only.The FFT size of mobile WiMAX PHY is scalable from 128 to 2048. When theavailable bandwidth increases, the FFT size is increased accordingly so that thesubcarrier spacing remains fixed at 10.94 kHz. As a consequence, the OFDMsymbol duration is constant regardless of the number of subcarriers, hence theimpact on higher levels in minimal. On the contrary, in fixed WiMAX the numberof subcarriers (the FFT size) is fixed at 256, therefore the subcarrier spacing varieswith the channel bandwidth.Other relevant characteristics common to both fixed and mobile WiMAX PHY arelisted below:

• MIMO: use of multiple antenna techniques are integral part of the standard802.16e. Beamforming, Spatial Multiplexing (SM) and Space-Time Codes(STC) are supported. MIMO schemes are essential in order to increase through-put and coverage.

• Hybrid ARQ: Chase Combining HARQ is a mandatory feature of WiMAX,which provides fast response to packet errors and improves cell edge coverage.

72

• Adaptive Modulation and Coding (AMC): AMC is an efficient mecha-nism to maximise throughput in time-varying channels: the highest modula-tion and coding scheme is selected according to the signal-to-noise/interferenceratio measured on the radio link. Mandatory modulations in DL are QPSKand 16 QAM, whilst in 64 QAM is optional both in UL and DL.

• CQICH (fast channel feedback): this channel is used by the MS to sendrelevant channel-state information to the BS. It is essential to support MIMOmode.

• Forward Error Correction (FEC): Convolutional Codes (CC), Convolu-tional Turbo Codes (CTC) and Repetition Codes are supported.

So far, we outlined the main features of WiMAX with particular attention to thephysical layer, highlighting the the slight differences between fixed and mobile ver-sions. From now on, we will take refer to mobile WiMAX as it is subject of our work,all the attributes include in the simulator we developed comply with the 802.16eamendment. Figure 4.2 shows the OFDM parameters values as used in the modelof mobile WiMAX PHY that will be described in the following sections. The pro-files in Table 4.2 are only a set of the possible implementations and they have beenchosen because they are likely to be deployed in the initial roll-outs. In particular,our simulations are based on a channel bandwidth of 10 MHz, the FFT size hasbeen set to 1024 and the frame duration is 5 ms.

Table 4.2: OFDM parameters used in mobile WiMAX. Boldfaced values correspondto those used in our simulator.

73

4.3 Mobile WiMAX system model

In this section we provide an in-depth description of the Simulink c© model formobile WiMAX physical layer that was developed in few months in summer 2007.The simulator in based on IEEE 802.16e specifications and includes only part of thewide set of functionalities offered by the standard. In fact, due to time constraints,many key-features have not been implemented, such as subchannelisation or channelestimation. Nonetheless the options available are sufficient to highlight the criticalissues arising from high PAPR, which was the principal driver of our research work.In our analysis we restricted the scope to DL transmission assuming that the basestation (BS) is transmitting continuously and using all the data and pilot subcarriersavailable (full load), this assumption represent the so-called worst case scenario butdoes not affect the PAPR analysis in any way. However, the system model weconceived can be easily upgraded adding those attributes not included: so far, itcan be regarded as a starting point for further development of mobile WiMAXphysical layer.

4.3.1 Transmitter

Figure 4.2 depicts the functional blocks4 that constitutes the transmitter. The ba-sic time-unit of the simulator is an OFDM symbol; a stream of randomised bitsis generated by a source block (which is not shown in the figure), then bits areencoded and mapped into QAM symbols. Each OFDM symbol comprises 720 datasubcarriers, 120 pilots and 184 null/guardband subcarriers. Tones are assembled inthe assembler block according to the rules in [48] and its structure in the frequencydomain is illustrated in 4.1. Once symbols have been properly ordered, IFFT oper-ation transforms the array of symbols into a time domain signal. Cyclic Prefix (CP)is then added and the signal is fed into the block that performs digital interpolation.Similarly to WCDMA (see chapter 3), the upsampling stage has been included for amore effective signal’s envelope reconstruction, which leads to a more realistic PAPRestimation. After this process, the signal is passed through the power amplifier (RFHPA).

Source

Information bits are generated in random sequences of sequences of length:

Nbits = Ndata ·R · log2(M), (4.1)

where Ndata is the number of data subcarriers, R represents the overall coding rateand M defines the modulation order. As stated in the 802.16e specification [48], the

4A complete Simulink c© block diagram of the simulator is given in Appendix A.

74

Figure 4.2: Functional blocks diagram of the mobile WiMAX transmitter model.

informations bits must be randomised before encoding and the initialisation vectorfor a shift register randomiser is specified. In our simulator, we preferred using thein-built option of the binary source block given by Simulink c© instead of performingthe standard-compliant bits randomisation, but this does not substantially affect theresults obtained.

Encoder

Mobile WiMAX standard supports both convolutional and turbo coding (abbrevi-ated as CC and CTC respectively), although the mandatory option is CC only.In our case, the sole CC has been implemented according to the parameters givenby [48]. The encoder has native code rate (R) of 1/2 and a constraint length of 7.The generator polynomials, expressed in octal alphabet, are:

G1 = 171OCT for X

G2 = 133OCT for Y (4.2)

where X and Y represent the encoder outputs.In order to achieve code rates greater than 1/2, the encoder output is puncturedfollowing the puncturing vectors in [47]. In the DL of fixed WiMAX, where subchan-nelisation is not used, an outer Reed Solomon code is added before the convolutionalencoder. The permitted code rates are: 1/2, 2/3, 3/4, 5/65. In Simulink c© convolu-tional coding and puncturing are supported in a single block, which was used forimplementation.

Interleaving

After the coding stage, bits are interleaved using a two steps mechanism. The firststep ensures that adjacent coded bits are mapped onto non-adjacent subcarriers,

5This value is used in conjuction with Reed Solomon code only.

75

whilst in the second step adjacent bits are alternately mapped to less and moresignificant bits of the modulation constellation. This strategy has been designed toexploit frequency diversity, thus improving the performance of the decoder.The interleaver implemented in our simulator comprises two blocks that performthe aforementioned operations:

Block 1 provides block interleaving by filling the matrix row by row, then valuesare read by columns. The matrix has sixteen rows (Nrows = 16) and a num-ber of columns which varies according to size of the encoded bits array, i.e.Ncolumns = Ntcb/Nrows, where Ntcb is the total number of coded bits;

Block 2 rearranges the elements of the input vector according to an index vectorgiven by the following equation:

mk = s

⌊jk

s

⌋+ mod

(jk + Ntcb −

⌊jkNrows

Ntcb

⌋, s

)+ 1, (4.3)

where mk and jk represent the indices of the bits after and before the inter-leaving operation, whereas s is defined as the half of log2(M) (number of bitsper subcarrier, Ncpc in the standard), i.e. s = (log2(M)/2).

Modulation Mapper

The modulation mapper (or symbol mapper) converts a sequence of binary datainto a sequence of complex-valued symbols. In mobile WiMAX the mandatoryconstellations are QPSK and 16 QAM, with an optional 64 QAM also defined inthe standard which is likely to be implemented - at least for the downlink - in mostof the systems [46].The allocation of bits to a symbol follows the Gray coding method, which ensurethat adjacent constellation points only differ by a single bit. In order to haveequal average symbol power, each constellation is normalised to unit average powerapplying an appropriate scaling factor. In our model, we used the symbol mapperavailable in the Simulink c© library.

Assembler

The mobile WiMAX profile we chose to implement (see Figure 4.2) supports 720data subcarriers, 120 pilots tones and the remaining are null subcarriers. TheAssembler block rearranges data and pilots symbols according to the OFDM sym-bol structure of Figure 4.1. The exact distribution of subcarriers depends on thepermutation mode selected. The PHY model we developed does not include sub-channelisation, therefore pilots tones have been evenly spread across the spectrum,resulting in a pattern of one pilot tone every six data subcarriers. Pilots subcarriers

76

are modulated by pseudo-random sequences of BPSK symbols and their power levelis boosted by 2.5 dB with respect to data subcarriers. Furthermore, the absence of achannel estimation functionality makes unnecessary the implementation of trainingsequences.The null subcarriers are inserted at the edges of the OFDM symbol (guard bands)as this leads to decreasing the emissions in the neighbouring bands; the DC carrieris also suppressed.

IFFT and Cyclic Prefix

As already mentioned in section 4.2.1, OFDM modulation strategy splits a highdata rate sequence of symbols into multiple parallel low data rate streams, each ofwhich separately modulates an orthogonal subcarrier. The output of the assemblerblock is vector of symbols and each element of this array can be regarded as a sampleof one of these multiple parallel symbol sequences. The IFFT operation convertsthe array of symbols into a discrete-time version of the OFDM signal; the output ofthe IFFT block is a vector containing time samples whose values are given by thefollowing equation:

x[i] =1

Nc

Nc−1∑

k=0

s[k]e−2πj(kBc)iTuNc , i = 0, . . . , Nc − 1, (4.4)

where Nc stands for the total number of tones (which, in our case, is 1024), s[k] isthe symbol modulating the kth subcarrier, Bc represents the frequency separationbetween two adjacent subcarriers and Tu is the useful OFDM symbol time (i.e.excluding the cyclic prefix). Actually, the IFFT algorithm returns time-samples ofthe ideal continuous OFDM signal taken at instants multiple of Tu/Nc, which yieldsto a sampling frequency of 11.2 MHz given the set of parameters chosen.After IFFT, the cyclic prefix is added to the signal. The guard interval is createdby copying part of the last samples and appending them to the beginning of theOFDM symbol. The resulting time duration of this portion of samples is namedguard interval (Tg). As a consequence, the total OFDM symbol length (Tsym) isnow increased (see Figure 4.3):

Tsym = Tu + Tg. (4.5)

Guard time Tg values are usually indicated as fraction of the useful symbol time,G

G =Tg

Tu

. (4.6)

In our case, we kept the cyclic prefix length fixed at 1/8 of the useful symbol du-ration, i.e. G = 1/8 or 12.5%. Yet, in mobile WiMAX OFDMA PHY the length

77

of the useful symbol time is constant, hence the guard time has fixed duration too(see Figure 4.2).

Figure 4.3: Insertion of cyclic prefix in OFDM symbols.

The simulator uses the standard IFFT Simulink c© block and CP is appendedusing a simple matrix concatenation block available in the library.

Interpolation

During this stage the signal is upsampled and filtered by a direct-form FIR filter.Upsampling operation inserts L− 1 zeros between signal’s samples, L is called theupsampling factor. The subsequent filtering stage ’reconstructs’ (interpolates) thesignal envelope between the original input signal’s samples. This block has beenintroduced in the attempt of simulating the digital-to-analogue conversion that takesplace in any digital communications device.The signal’s shape at the output of the interpolator presents a greater number ofpeaks with respect to its original version. This is due to the fact that the samplingtime has been ’artificially’ increased (the new sampling time is L times smaller) byadding non-informative samples and the signal’s shape has been ’predicted’ throughlinear filtering. Equation 4.4 indicates that the IFFT algorithm outputs a discrete-time version of the ideally modulated OFDM signal (i.e. the one we would obtainfrom modulating continuous-time tones) and its sampling time is equal to Tu/Nc.As a result, it may happen that possible high signal peaks are not present becauseof a slow sampling rate [30].Real digital-to-analogue converter transforms numerical values into a voltage signalthrough interpolation as described here.The interpolation factor, L, used throughout our simulation work has been set to3.

78

Power Amplifier Model

The AM/AM characteristic6 of the amplifier has been modeled with Rapp’s curveas described in section 3.1. The smoothness factor value, p, has been set to 3 as inWCDMA model (section 3.1), whilst the output saturation power has been raisedto 43 dBm. The linear gain, G, is 30 dB, which is a typical value for DL poweramplifiers for mobile WiMAX applications.The PA model now contain a new feature as a source of white noise has been addedand its power level can be tuned modifying the Noise Figure (NF) parameter. NFis defined as:

NF = log10

(1 +

Noise Temperature

290

), (4.7)

the value 290 corresponds to the standard thermal noise temperature in kelvindegrees and Noise Temperature indicates the equivalent thermal noise temperatureof the PA. The NF range of values can be quite wide depending on the type of PAemployed, in our simulation such value has been set to 3.5 dB, which is usual ingood quality amplifiers for base stations.

4.3.2 Receiver

In Figure 4.4 the functional block scheme of the mobile WiMAX receiver is illus-trated. Basically, the receiver performs the inverse operations as the transmitter.At first, the received OFDM symbols are downsampled to rule out the redundantsamples introduced by interpolation, then the cyclic prefix is removed and the signalis sent to the FFT block, which demodulates the symbols on each subacarrier. TheDisassembler block is in charge of separating the different types of data, i.e. pilot,data or null tones. Actually, this block removes guardband and pilot tones andgroups together data symbols, which are successively sent to the demapper. Oncedata symbols have been revealed, they enter the decoding stage.

Figure 4.4: Functional blocks diagram of the mobile WiMAX receiver model.

6Rapp’s model for PAs does not introduce phase noise, so the only input-output amplitudefunction is available.

79

FFT

In section 4.3.1 we pointed out that the IFFT algorithm converts a set of QAMsymbols into a discrete-time version of an OFDM signal, in which each of the inputsymbols modulates an orthogonal subcarrier. The FFT alogrithm performs theinverse operation of demodulating each tone that compose the OFDM symbols,thus the output of the FFT block is an array of QAM symbols whose size equals thetotal number of subcarriers. Resorting to Equation 4.4, we can derive the followingrelation between the received samples, x[i], and demodulated symbols outputted byFFT algorithm, s[k]:

s[k] =Nc−1∑i=0

x[i]e2πj(kBc)iTuNc , k = 0, . . . , Nc − 1, (4.8)

Disassembler

This block is needed to separate data, pilot and null subcarriers. In our simulator,null subcarriers, i.e. guardabands and DC tones, are discarded along with pilots. Aschannel estimation has not been implemented pilot symbols are useless, otherwisethey would sent to the channel estimator block.

Demapper

The stream of data symbols at the output of the Disassembler is fed into the Demap-per, which converts complex constellation points into a bit sequence. In our model,demapping is based on hard decision metric, which means that the Euclidean dis-tance between the received symbol and the whole set of allowed points of the con-stellation is calculated. The received symbol is then associated to the constellationpoint with the smallest distance. The simulator uses a standard hard decisiondemapper available in Simulink c© communications library.

Decoder

The stream of bits at the output of the demapping stage is passed on to the decoderthat comprises two steps: deinterleaving and Viterbi decoding.Deinterleaving rearranges the bits in the correct order as before the interleavingstage. Similarly to the interleaver block in the transmitter, deinterleaving is madeup of two blocks, which perform exactly the same operations as described in sec-tion 4.3.1.The decoding strategy is based on the well-known Viterbi algorithm, which offera low-complexity solution leveraging an approximated maximum likelihood metric.

80

An exhaustive description of Viterbi decoding is beyond the scope of this disser-tation, more details can be found in [49]. Simulink c© provides a specific Viterbidecoding block with a set of tunable parameters, which conveys the flexibility toadapt to many cases other than WiMAX.

4.3.3 Addressing high PAPR in WiMAX systems

As we anticipated in the introduction to this chapter, OFDM signals present largePAPR values, which leads to deleterious effects in presence of nonlinear devices, suchpower amplifiers or digital-to-analogue converters (DAC). Being based on an OFDMPHY, WiMAX waveforms are characterised by high peaks well above its averagelevel, as a consequence very linear PA and DAC must be employed, otherwise signalswould be unduly distorted affecting the system performance.In WiMAX, problems related to high PAPR are particularly severe resulting indifficult challenges to face when designing the radio. In section 4.1, we pointedout that transmitting high PAPR signals through a nonlinear devices causes in-band and out-of-band distortion. In order to overcome these unwanted effects, twooptions are available:

1. choosing devices that can accommodate signals with wide dynamic range intheir linear region,

2. reducing the PAPR by means of appropriate techniques.

In chapter 3 we discussed about the drawbacks stemming from the first approach:high-linearity PA are expensive and characterised by low efficiency, yet the alter-native solution of applying a back-off to the input signal, IBO (see section 3.1),equally leads to a decreased efficiency that limits battery life and PA gain is pe-nalised by IBO (which means a restriction of coverage range in mobile applications).Conversely, PAPR reduction techniques try lowering the peaks either modifying thewaveform or coding the source data in a way that avoids high PAPR combinations.As shown in section 3.7, approaches based on digital waveform manipulation effec-tively reduce the PAPR, although they introduce a certain amount of distortion.The amount of signal’s degradation allowed is a key aspect associated to radio designas its limit is usually imposed by the standard, e.g. in WCDMA systems, ACLR,EVM and PCDE are quality metrics that specify the permitted distortion (see sec-tion 3.6.3). Similarly to WCDMA, mobile WiMAX specification [48] imposes tightlimits on each QAM constellation quality, again the parameter used is EVM, evenif its definition has been adapted to OFDM signals.

Evaluating PAPR The mobile WiMAX PHY model has been provide with ablock that calculates PAPR for each OFDM symbol transmitted. PAPR definition

81

is given in Equation 3.1. In order to describe the statistical distribution of PAPR,the SImulink c© automatically creates an array containing several PAPR values cor-responding to 1 second record of the interpolated7 WiMAX signal (roughly 10.000OFDM symbols). This set of data is then used to build the Complementary Cumu-lative Distribution Function (CCDF) of the PAPR. The OFDM parameters usedthroughout our simulations are shown in Figure 4.2, the constellation is 16 QAMand the overall code rate, R, is 3/4.The CCDF curve is depicted in Figure 4.5, where it appears evident that OFDMsymbols affected by extremely high peaks seldom occur, nonetheless they must betaken into account in the radio design.Theoretically, an IBO value greater or equal to the PAPR would prevent the input

7 8 9 10 11 12 1310

−4

10−3

10−2

10−1

100

CCDF of PAPR

log1

0(1−

CD

F)

PAPR(dB)

Figure 4.5: CCDF of PAPR for mobile WiMAX system. 1024 subcarriers, 16 QAM,R = 3/4.

signal driving the the PA into its nonlinear region. In our case, Figure 4.5 showsthat an IBO value of -at least- 12 dB would be necessary, although such a conser-vative back-off does not guarantee that a given OFDM symbol will have a PAPRexceeding the IBO. In practice, such large back-off values would decrease the PAperformance to intolerable level.Due to its nonlinear AM/AM characteristic, the PA will clip the high peaks intro-

7It is worth to notice that PAPR calculation is performed after the interpolation stage as thisoperation approximates a DAC of a real transmitter.

82

ducing distortion. As an example, Figure 4.6 shows the signal at the PA input alongwith its amplified8 version at the output, clearly peaks with amplitudes beyond thesaturation level have been cut. As a result, the signal quality is now deteriorated.

0 50 100 150 200 2500

5

10

15

20

25

30

35

40

45

50

samples

Am

plitu

de [V

olt]

Amplifier’s clipping effect on high PAPR input signal

Figure 4.6: PA clipping effect: input signal (blue line) and amplified output signal(red line). Amplitude peaks have been constrained to saturation level.

We have already mentioned that the distortion caused by a nonlinear operationlike amplifying can be divided into two categories: in-band and out-of-band dis-tortion. In-band interference can be identified using a I/Q constellation diagram(symbols have been acquired from the demapper input, see Appendix A) as in Fig-ure 4.7, where the scattered distribution of points around the theoretical 16 QAMconstellation symbols reveals the presence of noise induced by the amplifier clip-ping effect. With regard to out-of-band distortion, it can be easily observed inthe frequency domain through a spectrum analyser (which, in our case, is an FFToperation) as done in section 3.3. Figure 4.8 depicts the signal’s spectrum bothat the PA input and output. As we noticed for WCDMA systems, the growth ofspurious spectral components outside the ideal signal band is a direct consequenceof PA nonlinearity and the effect increases as the signal’s input power approachesthe amplifier’s saturation point.

8For the sake of a neat visualisation the PA gain has been set to 0 dB.

83

Figure 4.7: 16 QAM constellation diagram immediately pinpoints the presence ofnoise due to PA nonlinearity (all the other possible source of interference have beenexcluded). Input power 10 dBm, output saturation power 43 dBm, linear gain30 dB.

The amount of degradation introduced on the signal by the various system com-ponents, such as PA, mixers, local oscillators, etc. must be kept within specifiedlimits, otherwise it would negatively affect the overall performance. In the 802.16especification, EVM is used quality metric to estimate the modulation accuracy.EVM has been comprehensively discussed in section 3.4 and, being a generic pa-rameter commonly used for measuring digital modulations errors, the same generalconcepts can be applied to both WCDMA and WiMAX. Obviously, the definitionhad to be slightly adapted to OFDM signals, yielding to the following equation [48]:

EV M =

√∑k∈S (∆I2

k + ∆Q2k)∑

k∈S

(I20,k + Q2

0,k

) , (4.9)

where I0 and Q0 represent the ideal symbol point on the k-th subcarrier, whilst∆Ik, ∆Qk denote the difference between the actual observed point and the refer-ence one. S is the group of modulated data subcarriers where the measurement isperformed, which, in our case, corresponds to 720 tones. Maximum allowed limit ofEVM are given as function of the data rate, i.e. constellation order and code rate,and they are expressed in decibels9. For 16 QAM-R = 3/4, EVM must not exceed

9In WCDMA standard EVM is given as percentage, see section 3.4.

84

Figure 4.8: Signal spectrum distorted by PA nonlinearity (black line) comparedto the spectrum calculated from the input signal. Input power 10 dBm, outputsaturation power 43 dBm, linear gain 30 dB.

−24 dB [48], which is a notably strict limit.In order to obtain a more detailed estimate of possible EVM values in a real WiMAXtransmitter, we introduced a baseband model of a mixer both in uplink and down-link. In UL, the mixer has been placed immediately after the interpolation stageand before the radio amplifier model, whereas in DL it is positioned at the top ofthe receiving chain (see figure in Appendix A). In Figure 4.9 simulated values ofEVM have been plotted against the power emitted by the PA; results show that theEVM rapidly increases as the output power level approaches the saturation limit.At 40 dBm the EVM is scarcely below the imposed limit of −24 dB, at maximumoutput power EVM has a value greater than −10 dB, which denotes an unacceptableamount of distortion. As a consequence, the limitation imposed by EVM heavilypenalises the PA power range, which would be limited to 40 dBm instead of thenominal 43 dBm. Resorting to a conservative IBO would ulteriorly reduce the PArange and efficiency, thus the only viable solution seems to be represented by PAPRreduction techniques.In section 3.5 we pointed out that clipping a signal reduces the PAPR at the ex-pense of adding distortion, in particular, the level of induced ’noise’ increases as theclipping amplitude is lowered. As explained in section 3.8, designers are forced toseek for a compromise between clashing requirements: improving the PA efficiency-hence lowering the PAPR- and respecting the quality requisites imposed by the

85

10 15 20 25 30 35 40 43 45−40

−35

−30

−25

−20

−15

−10

−5

Pout [dBm]

EV

M [d

B]

EVM vs PA output power

16 QAM R=3/4

Figure 4.9: Simulated EVM values as a function of the PA output power. Inputback-off not applied.

standard. Furthermore, WiMAX designers have to face an even harder task be-cause OFDM signals are characterised by very high values of PAPR and the marginfor error is quite tight.Unfortunately, we could not go further with the analysis due to time constraints.Nonetheless, the main difficulties that may be encountered in addressing PAPR forWiMAX systems have been emphasized. The Simulink c© model has been includedin the CD-rom attached to this dissertation. So far, the simulator can be regardedas a good starting point for further development.

4.4 Conclusions

In this chapter we provided a brief overview of WiMAX focusing on those innova-tive features that make it a promising wireless broadband system in the next future.Both fixed and mobile WiMAX are able to deliver high data rates coupled with anunprecedented flexibility. There are no doubts that the success of WiMAX will bemainly determined by the local spectrum regulatory policies, nonetheless from atechnical viewpoint WiMAX must be recognised as a milestone towards next gen-eration wireless networks and services. As an example, the WiBro network recentlyrolled out in South Korea is the first large-scale system that employs multiple an-

86

tennas schemes.The physical layer of mobile WiMAX, based on the IEEE 802.16e− 2005 has beendescribed in more detail as it represents the core of the work brought forward inthis chapter. In fact, sections 4.3.1 and 4.3.2 contain an exhaustive description ofthe Simulink c© model for mobile WiMAX PHY that we developed in summer 2007at Electric & Electronic Engineering department of UCL. Although many essen-tial features are missing, the simulator allowed to carry out an accurate analysisabout the potential issues related to high PAPR in WiMAX systems. Being basedon OFDM technology, WiMAX signals’ waveform present particularly large peakscompared to their average amplitude level. We have seen in the previous chapterdedicated to WCDMA that high PAPR leads to detrimental effects in terms of con-stellation accuracy and out-of-band spectral splatter. In WiMAX this problem isalso more severe because of the strict limits by PHY specifics, i.e. allowed valuesof constellation error (EVM) are meant to ensure a high signal quality, thus rulingout every possibility of employing aggressive PAPR reduction techniques. In simpleterms, the modulation accuracy required by [48] only allows very little distortion,whilst PAPR scheme may introduce significant distortion. As a result, the marginfor operation is quite tight.The work-plan we had in mind could not be brought to the desired end; an in-depthanalysis of possible solution for decreasing the PAPR (as done for WCDMA trans-mitter) has not been performed, nonetheless the work done so far forms a solid basefor further development.

87

Chapter 5

Conclusions

In the vast panorama of wireless communications, Ultrawide Band systems havegained a crucial position in the area of short-range applications. In chapter 2 wenoted that UWB systems that employ radio pulses with extremely short time dura-tion as physical interface are able to perform very accurate ranging, which is requiredin several applications requiring such high precision, e.g. human/asset tracking andpositioning. Nonetheless the UWB specified bandwidth can be exploited using dif-ferent radio interfaces, such as OFDM. In this case the priority is given to deliveringhigh data rates rather than positioning, as an example, the so-called wireless USBapplications follow this approach as they are meant to replace standard USB cablesdelivering up to 480 Mbps over 3 meters range.The LDR-LT prototype developed by the PULSERS consortium, and tested at PRL,transmits pulses of few nanosecond duration and was designed to provide a resolu-tion of 30 cm in distance measurement coupled with a useful data rate of 12.5 Mbps.PULSERS put particular emphasis on the low-complexity architecture of the plat-form for two main reasons. First the LDR-LT demonstrator should evolve into acommercially viable product, hence the overall cost of the system has been keptintentionally low. The second point is that PULSERS wanted to show that accu-rate ranging is feasible even with devices based on low-complexity architecture andalgorithm, conversely systems already available on the market are based on moresophisticated (and more expensive) architectures. In fact, the results obtained bythe extensive field trials campaign conceived by the author and held at PRL clearlyshow that the prototype is able to perform ranging with an accuracy far better thanany narrow-band system, which is promising outcome considering that the marginfor improving the platform with new features is still broad. It suffice to remindthat the system had no data protection codes or advanced data post-processing,whilst the whole range of commercial positioning devices today available have. Inharsh radio conditions, such Non-Line-of-Sight environments, the LDR-LT platformhas an average distance estimation error that does not exceed 45 cm, which is an

89

excellent result given the simplicity of the system.The analysis of collected data indicates that many errors in ranging estimation areimputable to synchronisation failures or misdetections due to multipath. Such errorsmight be avoided by introducing error correction codes and fixing the synchronisa-tion mechanism. For such reasons, we decided to remove -as much as possible- thosespurious measurements through a simplified Kalman filtering process. As a result,most part of the unwanted errors have been ruled out and distance estimations canrely on a more accurate set of data. Most importantly this allowed us to get closerto the ’actual’ system’s performance, hence we have been able to draw more reliableand realistic conclusions on its effective capabilities.In conclusion, we can state that the measurement campaign carried out at PRL con-tributed to demonstrate that UWB-technology is capable of high precision ranging,even with a low-complexity platform like LDR-LT. The results drawn from fieldtrials form the base for the future development of the prototype into completelyfunctional system for asset tracking and intelligent sensor network applications.The PULSERS project has now entered into its second phase and the partners areworking on those essential improvements needed to make LDR-LT a fully opera-tional system.

In chapters 3 and 4 we addressed the issues related to high PAPR signals fortwo different modulation schemes, namely WCDMA and OFDM.WCDMA has been chosen as the physical layer interface for UMTS, which is a cel-lular system adopted in many countries worldwide. In recent years dedicated stan-dardisation groups have worked to adapt WCDMA to the increasingly demandingrequirements for higher data rates. Recently a revised version of WCDMA has beenreleased and UMTS can now support higher bit rates both in downlink and uplink(respectively HSDPA and HSUPA). Chapter 3 includes a detailed description of theSimulink c© model for WCDMA physical layer (based on the original version com-monly knows as Release 99 ) along with a comprehensive study on possible PAPRreduction techniques that may be applied to mobile devices. The model developedis fully compliant with the standard, this confers a solid reliability to simulationresults. As widely discussed in the introduction to the chapter, PAPR has seri-ous negative effects on modulation quality and system’s performance in presence ofnonlinear devices, such as power amplifiers or digital-to-analog converters. Signalswith high PAPR drive amplifiers (that are the most powering consuming devicesin mobile equipment) into saturation. As a consequence, if not properly addressed,high PAPR can shorten battery life, which is a particularly significant problem inmobile devices. Up to date, several approaches have been devised to tame the detri-mental effects of PAPR and some of them are especially attractive because of theirsuitability to digital implementation. Such techniques are usually know as signaldistortion techniques, as they directly modify the signal’s waveform in order to lowerthe highest peaks. Being in the form of digital FIR filters, their implementation

90

impact on WCDMA systems is minimal. That is the reason why we decided toanalyse the performance of three different strategies belonging to this category.Simulation results show that all the techniques employed are effective in reducingthe signal’s PAPR, although each of them introduces a different amount of dis-tortion. In fact, limiting the signal’s amplitude range to a desired level causes adegradation in modulation quality. The WCDMA standard imposes tight limitson the tolerable amount of signal distortion, this compels the designers to strike acompromise between PAPR reduction and modulation deterioration. Amongst theclipping strategies considered, the one that offers better performance against thestandard’s requirements is clipping and filtering. This technique merely consists inhard clipping the signal and removing the out-of-band distortion due to nonlinear-ities by digital filtering. With the regard to the ease of implementation, clippingand filtering is surely the most suitable. The others strategies show performanceclose to clipping and filtering, though they rely on slightly more complex algorithms.

Like in WCDMA, OFDM-based systems generate signals with large PAPR val-ues. In these systems the problem is even more significant than in WCDMA becausethe PAPR value increases with the number of modulated subcarriers. Amongst thewireless systems employing an air-interface based on OFDM modulation scheme,WiMAX is -at the moment- the most promising technology for delivering broadbandservices. In chapter 4 we highlighted the principal features of WiMAX, especiallythose characteristics that make WiMAX unique in the present scenario of wirelesscommunications.Similarly to WCDMA, we developed a Simulink c© model for mobile WiMAX phys-ical layer. A detailed description of the model is given in chapter 4. Although themodel lacks some key-functionalities, like subchannelisation or channel estimation,the options available are enough to pinpoint those critical points arising from highPAPR. In the case of WiMAX, the standard specifications on modulation accuracyare more stringent than in WCDMA, this narrow the margin left for possible useof PAPR reduction techniques. Our preliminary study on the effect of PAPR inthe presence of a nonlinear amplifier clearly indicates that the power emitted by aWiMAX base station should be unduly reduced to satisfy the signal’s quality req-uisites imposed by the standard. In real scenarios, this would tremendously shrinkthe coverage area, therefore solutions must be found that ensure an efficient useof amplifiers. Undoubtedly, WiMAX designers have to face a hard task given theextremely high values of PAPR and stringent limitations on modulation quality.

All the cited Simulink c© models and relative Matlab c© codes have been includein the CD-rom attached to this dissertation.

91

Appendix A

IEEE 802.16e PHY layer model

Figure A.1 shows the Simulink c© block model for the IEEE 802.16e physical layeras developed at the Electrical and Electronic Engineering department of UCL. Thefunctionalities of each constituent block have been explained in detail in chapter 4.Main simulation paramenters, such as IBO, total emitted power, modulation or-der, etc. can be changed simply typing the desired values into the dialog boxnamed ’paramenters settings ’. As mentioned in chapter 4, the model does notinclude transmit diversity, hence it is assumed that there is a single transmittingantenna. The block named ’EVM’ calculates the modulation distortion according tothe equation 4.9 specified by the standard and reported in chapter 4, section 4.3.3.PAPR is evaluated for each transmitted WiMAX frame and values are displayed ina Simulink c© ’display box’ placed at bottom right side of the screen.

93

Figure A.1: Mobile WiMAX PHY model

94

Bibliography

[1] A.F. Molisch I. Oppermann C. Politano D. Porcino M.G. Di Benedetto,T. Kaiser. UWB communications systems: A Comprehensive Overview. Hin-dawi Publishing Corporation, 2006.

[2] W.Hirt D.Porcino. Ultra-wideband radio technology: Potential and Challangesahead. In IEEE Communications Magazine, volume 41, pages 66–74, July 2003.

[3] ETSI. Physical channels and mapping of transport channels onto physicalchannels (FDD). Technical Report TS 25.211 v2.5.0 (1999-03), 3GPP, 1999.

[4] ETSI. Universal Mobile Transmission System (UMTS); Spreading and Modu-lation. Technical Report TS 125 213 v6.5.0 (2006-03), 3GPP, 2006.

[5] Agilent. Agilent PN 89400-14; Using Error Vector Magnitude Measurementsto Analyze and Troubleshoot Vector-Modulated Signals. Product note.

[6] http://www.wimaxforum.org/technology/downloads/. WiMAX White PaperPart1: A Technical Overview and Performance Evaluation. Technical report,WiMAX Forum, August 2006.

[7] The Commission of European Communities. on allowing the use of radio spec-trum for equipment using ultra-wideband technology in a harmonised mannerin the Community. 2007/131/EC, February 2007.

[8] R.Khono M.Ghavami, L.B.Michael. ultra wideband signals and systems in com-munication engineering. John Wiley & Sons, Ltd, 2004.

[9] R.J.Fontana. Recent System Applications of Short-Pulse Ultra-Wideband(UWB) Technology. In IEEE Transactions on Microwave Theory and Tech-niques, volume 52, Sept. 2004.

[10] Federal Communications Commisssion (FCC). Revision of Part 15 of the Com-mission’s Rules Regarding Ultra-Wideband Transmission Systems. First Reportand Order. ET Docket 98− 153, FCC 02− 48, April 2002.

95

[11] R.A.Scholtz M.Z.Win. Ultra-Wide Bandwidth Signal Propagation for IndoorWireless Communications. In ’Towards the knowledge Millenium’ InternationalConference - ICC 97, volume 1, pages 56–60, June 1997.

[12] M.Quinlan S.J.Ingram, D.Harmer. UltraWideBand Indoor Positioning Systemsand their Use in Emergencies. In Position Location and Navigation Symposium2004 - PLANS 2004, pages 706–715, April 2004.

[13] IEEE 802.15 TG 4a. http://www.ieee802.org/15/pub/tg4a.html.

[14] W.Hirt D.Porcino. Pervasive Ultra-wideband Low Spectral Energy Radio Sys-tems (PULSERS). In Contribution to the 7th WWRF Meeting, Dec. 2002.

[15] S.Zeisberg. Publishable Final Activity Report. PULSERS, Feb. 2006.

[16] 1st PULSERS Workshop (hosted at the IST Summit 2005). Dresden (Ger-many), June 2005.

[17] S.Zeisberg V.Bramkovic. Public project presentation. Integrated PULSERSProject Phase 2 - Deliverable number: D1.3, Feb. 2006.

[18] J.Barney R.J.Fontana, E.Richley. Commercialization of an Ultra WidebandPrecision Asset Location System. In IEEE Ultra Wideband Systems and Tech-nologies Conference, Nov. 2003.

[19] G.J.M.Janssen H.Nikookar C.Tiberius Z.Irahhauten, G.Bellusci. Investigationof UWB Ranging in Dense Indoor Multipath Environments. In 10th IEEESingapore International Conference on Communication Systems ICCS 2006,pages 1–5, Oct. 2006.

[20] C.J.Law W.T.Ng Y.J.Lee Z.N.Low, J.H.Cheong. Pulse detection algorithmfor line-of-sight (LOS) UWB ranging applications. In Antennas and WirelessPropagation Letters, volume 4, pages 63–67. IEEE, 2005.

[21] S.Yamaguchi T.Tanaka. GPS Standard Positioning using Kalman filter. InSICE-ICASE - International Joint Conference, pages 1351–1354, Oct. 2006.

[22] L.N.de Rivera J.C.S.Garcia V.I.Ponomaryov, O.B.Pogrebnyak. Increasing theaccuracy of differential global positioning system by means of use the Kalmanfiltering technique. In IEEE International Symposium on Industrial Electronics- ISIE, volume 2, pages 637–642. IEEE, Dec. 2000.

[23] M.Najar E.Medina. High Resolution Location in Ultra Wideband Communi-cations Systems. In Networking with Ultra Wide Band and Workshop on UltraWide Band for Sensor Networks, pages 40–44. IEEE, July 2005.

96

[24] Ubisense (www.ubisense.net). The Ubisense Precise Real-time Location Sys-tem, 2007.

[25] D.Porcino O.Gremigni. Field Trials of an Accurate UWB Ranging System.In Seminar on Ultra Wideband Systems, Technology and Applications, pages169–175. IET, April 2006.

[26] H. Holma and A. Toskala. WCDMA for UMTS. John Wiley & Sons, 2000.

[27] N.Whinnett R.Ratasuk X.Weimin R.Kuchibhotla A.Ghosh, R.Love. Overviewof enhanced uplink for 3GPP W-CDMA. In IEEE 59th Vehicular TechnologyConference. VTC 2004-Spring, volume 4, pages 2261–2265, May 2004.

[28] N.Whinnett R.Ratasuk X.Weimin R.Ratasuk A.Ghosh, R.Love. Overview ofenhanced uplink for 3GPP W-CDMA. In IEEE 60th Vehicular TechnologyConference. VTC 2004-Fall, volume 5, pages 3359–3363, Sept. 2004.

[29] E.Malkamaki M.Bertinelli. WCDMA ENHANCED UPLINK: HSUPA LINKLEVEL PERFORMANCE STUDY. In IEEE 16th International Symposium onPersonal, Indoor and Mobile Radio Communciations. PIMRC-2005, volume 2,pages 895–899, Sept. 2005.

[30] R. O’Neill and L.B. Lopes. Envelope Variations and Spectral Splatter inClipped Multicarrier Signals. In Proc. of IEEE PIMRC’95, pages 71–75, 1995.

[31] M. Honkanen and S.G. Hggman. New Aspects on Nonlinear Power AmplifierModeling in Radio Communication System Simulations. In Proc. of PIMRC’97, volume 3, pages 844–848, 1997.

[32] E. Costa, M. Midrio, and S. Pupolin. Impact of Amplifier Nonlinearities onOFDM Transmission System Performance. IEEE Communications Letters,3(2), February 1999.

[33] ETSI. Universal Mobile Transmission System (UMTS); UE Radio Transmissionand Reception (FDD). Technical Report TS 25.101 v4.5.0 (2002-06), 3GPP,2006.

[34] R. Tanner and J. Woodard. WCDMA Requirements and Practical Design. JohnWiley & Sons, 2004.

[35] Agilent. Agilent AN 1335; HPSK Spreading for 3G. Application note.

[36] T.Palenius T.Ottosson. THE IMPACT OF USING MULTICODE TRANS-MISSION IN THE WCDMA SYSTEM. In 49th Vehicular Technolgy Confer-ence. VTC 1999., volume 2, pages 1550–1554. IEEE, May 1999.

97

[37] O. Vaananen, J. Vankka, and K. Halonen. Simple algorithm for peak windowingand its application in GSM, EDGE and WCDMA systems. IEE ProceedingsCommunications, 152(3):357–362, June 2005.

[38] S. Marsili. Algorithm for Peak to Average Power Ratio Reduction operatingat Symbol Rate. In Proc. of IEEE Int. Symp. ISCAS 2005, volume 6, pages5497–5500, May 2005.

[39] A. Springer, T. Frauscher, B. Adler, D. Pimingsdorfer, and R. Weigel. Impactof Nonlinear Amplfiers on the UMTS System. In 6th Int. Symp. on Spread-Spectrum Tech & Appli., 2000.

[40] R. Van Nee and R. Prasad. OFDM for Wireless Multimedia Communications.Artech House Publishers, 2000.

[41] X. Li and L.J. Cimini. Effects of Clipping and Filtering on the Performance ofOFDM. IEEE Communications Letters, 2, May 1998.

[42] T. May and H. Rohling. Reducing the Peak-to-Average Power Ratio in OFDMRadio Transmission Systems. In Proc. of Vehicular Technology Conference(VTC’98), volume 3, pages 2474–2478, May 1998.

[43] J. Akhtman, B.Z. Bobrovsky, and L. Hanzo. Peak-to-Average Power RatioReduction for OFDM Modems. In Proc. of IEEE Vehicular Technology Con-ference (VTC 2003), volume 2, pages 1188–1192, April 2003.

[44] H. Kuo, S.W. Cheung, and Simon S.F. Hau. A Novel Peak-Windowing Tech-nique for WCDMA Systems. In Vehicular Technology Conference VTC 2006Spring, volume 4, pages 313–316, 2006.

[45] K. Halonen O.Vaananen, J. Vankka. Effect of Baseband Clipping in Wide-band CDMA System. In 7th International Symposium on Spread-SpectrumTech.&Appl. IEEE, Sept. 2002.

[46] R. Muhamed J.G. Andrews, A. Ghosh. Fundamentals of WiMAX: understand-ing broadband wireless networking. Prentice Hall, 2007.

[47] IEEE. Standard 802.16-2004. part16: Air interface for fixed broadband wirelessaccess systems. Technical report, IEEE, October 2004.

[48] IEEE. Standard 802.16e-2004. part16: Air interface for fixed and mobile broad-band wireless access systems - amendment for physical and medium access con-trol layers for combined fixed and mobile operation in licensed band. Technicalreport, IEEE, December 2005.

98

[49] J.G. Proakis. Digital Communications. McGraw-Hill, 4th edition, 2000.

[50] R. van Nee and Arnout de Wild. Reducing the Peak-to-Average Power Ra-tio of OFDM. In Proc. of IEEE Vehicular Technology Conference (VTC’98),volume 3, pages 2072–2076, May 1998.

[51] N.J. Fliege. Multirate Digital Signal Processing. John Wiley & Sons, 1994.

[52] Agilent Application Note 1536. Designing and Testing 3GPP W-CDMA UserEquipment.

99

Advanced wireless systems for ranging and communications:measurement and modelling

Documents

Transcript of Advanced wireless systems for ranging and communications:measurement and modelling