Research Article Interleaved-MIMO DAS for Indoor Radio...

Research ArticleInterleaved-MIMO DAS for Indoor Radio CoverageGuidelines for Planning

E M Vitucci1 F Fuschini1 V Degli-Esposti1 and L Tarlazzi2

1 Dipartimento dellrsquoIngegneria Elettrica e dellrsquoInformazione (DEI) AlmaMater Studiorum-Universita di Bologna 40136 Bologna Italy2 CommScope Italy Srl Via Mengolina 20 48018 Faenza Italy

Correspondence should be addressed to E M Vitucci enricomariavitucciuniboit

Received 25 May 2014 Revised 8 August 2014 Accepted 27 August 2014 Published 15 September 2014

Academic Editor Christoph F Mecklenbrauker

Copyright copy 2014 E M Vitucci et al This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

The combination of distributed antenna systems (DAS) and multiple input multiple output (MIMO) schemes opens the way to avariety of coverage solutions for indoor environment In this paper interleaved-MIMO (i-MIMO) DAS indoor coverage extensionstrategies are studied Their performance in high-order MIMO cases is investigated in realistic conditions through LTE-A link-level simulations based on statistical data extracted from radio channel measurements the impact of the deployment strategyon performance is then evaluated and useful planning guidelines are derived to determine the optimum deployment for a givenpropagation environment

1 Introduction

MIMO transmission techniques are necessary for present andfuture indoor mobile radio systems to achieve a spectral effi-ciency of 30 bitsHz as required by the long term evolution-advanced (LTE-A) standard [1] or even higher figures infuture-generation systems

Differently from what happens outdoors a large numberof easily accessible fixed-terminal locations are availableindoors due to the intrinsically 3-dimensional topology andto the presence of cable raceways double ceilings and soforth Therefore distributed antenna systems (DAS) indoorcoverage-extension schemes can be implemented where thesame signal (of a single base station) is distributed to anumber of remote antenna units (RAUs) using copper cablesor optical fibers (F-DAS) to achieve a better radio coverage[2 3]

The traditional MIMO implementation over a DASscheme requires providing each RAU with all 119899 MIMObranches signals with related cables and antenna elementsthis solution may be called ldquocolocated MIMO DASrdquo Theupgrade from SISO DAS to ldquocolocated MIMO DASrdquo canbe expensive as it requires the installation of new cables

and antennas elements at each RAU location and often themodification of the whole deployment

Recent studies have shown the advantages of adoptingldquodistributed MIMOrdquo schemes where the 119899 antenna elementsrelative to the 119899MIMO branch-signals are further spaced toexploit lower fading correlation among the different branchesand therefore better performance with respect to a colocatedMIMO scheme [4 5] Since antenna displacement can beactually realized through aDAS scheme by connecting differ-ent branches to different RAUs such solutions are often calledldquoMIMO DASrdquo and allow an easy and cheap upgrade fromSISO to MIMO DAS as the antenna and cable deploymentdoes not need to be modified at all A drawback of thissolution is represented by the uneven power distribution ofdifferent MIMO branch-signals at the 119877119909 due to the differentspatial placement of the corresponding antenna elementssuch phenomenon is called ldquopower imbalancerdquo (PI) whoseintensity can be for instance evaluated through the metricsintroduced in [6]

Of course in large indoor environments with many RAUlocations multiple repetitions of the 119899MIMO branch-signalsmust be deployed over the service area the MIMO branches

Hindawi Publishing CorporationInternational Journal of Antennas and PropagationVolume 2014 Article ID 690573 10 pageshttpdxdoiorg1011552014690573

2 International Journal of Antennas and Propagation

must be therefore properly interleaved over the environmentin order to simultaneously guarantee good radio coverageand low PI therefore optimizing MIMO capacity Suchdeployment strategy is called ldquointerleaved-MIMO DASrdquo (i-MIMODAS) and has been introduced and illustrated for thefirst time in [6 7] with particular reference to 2 times 2 MIMOsystems

Interleaved-MIMO DAS schemes are being consideredwith great interest by operators and installers because theymight yield a performance level comparable to that of acolocated MIMO DAS (119888-MIMO DAS) at a much lowercost [6] However high PI values might degrade MIMOperformance to an extent that still needs to be evaluatedin relation to the DAS coverage advantages especially forhigh-order MIMO schemes As already anticipated in [6]the actual PI value distribution depends on the way the 119899branch antennas are interleaved in space and therefore onthe i-MIMO planning strategy Of course PI also depends onthe mobile users spatial distribution but the latter cannot becontrolled and has been assumed uniform in the followingstudy

For these reasons the i-MIMO DAS concept is furtherinvestigated and discussed for higher-order MIMO schemesin the present paper where LTE-advanced link-level simu-lations (Section 2) together with a simple but meaningfulpath-loss formula are used to derive some general planningguidelines (Section 3) for high-order MIMO and for 1D(corridor) 2D and 3D (multifloor) coverage scenarios Suchplanning guidelines can be useful for the system designer toproperly choose the MIMO order and i-MIMO deploymentto optimize the performance to cost ratio It is worth specify-ing that our study is limited to the planning of a DAS systemand not of thewholemobile radio network and therefore doesnot consider channel allocation and interference issues

In order to provide planning guidelines a single bench-mark system standard (LTE-advanced) has been consideredand the problem has been simplified by taking into accountonlymajor parameters such as the signal-to-noise ratio (SNR)and the PI that depend on the particular i-MIMO DASdeployment Other factors such as the multipath richnessand other propagation characteristics can have an importantimpact on performance but they cannot be engineered asthey primarily depend on the propagation environment andtherefore their role has not been deeply analyzed here In thefollowing we will simply assume the multipath richness to belarge enough not to represent a limit to MIMO performance

2 Interleaved-MIMO DAS Evaluation

21 The Concept The application of the MIMO concept todistributed antenna systems is analyzed here with particularreference to the spatial multiplexing capability althoughother transmission techniques such as 119879119909 diversity are ofcourse possibleTherefore in the following each RAU (whichcarries a single ldquoMIMO branchrdquo) is supposed to transmita MIMO signal different from those radiated by the otherRAUs belonging to the same (distributed) transmitting arrayFurthermore since each RAU defines a ldquocellrdquo that is thearea where its signal strength is dominant (usually square or

rectangular cells in indoor scenarios instead of hexagonalcells due to the shape of rooms and buildings) each cell canbe therefore associated with a corresponding MIMO branch

The present work considers the implementation of i-MIMO DAS in large indoor environments with 119872 RAUlocations (with 119872 ≫ 119899 where 119899 is the MIMO order)where multiple repetitions of the 119899MIMO branch-signals aredeployed over the service area The acronym ldquo119899 times 119899 i-MIMO119872-DASrdquo can be used to indicate an 119899th-order i-MIMO DASsolution with M RAUs where each RAU is equipped with asingle antenna element and the mobile terminal is equippedwith an array of colocated antenna elements [6]

In case a linear layout is required andor low-order i-MIMO schemes are considered the best spatial deploymentfor the differentMIMO branches can be often handled on thebase of somehow evident intuitive considerations [6]

In a 2D3D coverage case and for high-order MIMOhowever the best deployment scheme may be no longertrivial as somedifferent interleaved solutionsmight exist andshould be therefore evaluated

In order to limit the spatial PI it is intuitive that thedifferent MIMO branches should be somehow arranged insquarerectangular ldquoclustersrdquo (Figure 1) Of course differentclusters can be laid over the service area in different waysfor instance in the 8-branch case assuming a 2 by 4 clusterscheme the three solutions represented in Figure 1 can beconsidered the clusters can be simply aligned to each other(Figure 1mdashleft) or a 1-2-cell shift can be introduced betweenadjacent columns (or rows) of clusters (Figure 1 center andright resp)

Intuitively the 2-shift solution should lead to a moreuniform branch-signal distribution however actual perfor-mance must be assessed by simulation as shown in Section 3

In conclusion irrespective of the i-MIMO order theconcept of ldquoclusterrdquo widely used in cellular radio planningapplies here However here the mechanisms are somewhatreversed as in traditional cellular planning the cluster-sizeshould be large enough to separate cochannel cells andthereby minimize interference and in i-MIMO DAS thecluster dimension and therefore the MIMO order 119899 shouldbe small enough to minimize power imbalance In fact high-order MIMO deployments (such as the 8-branch case ofFigure 1) will necessarily yield greater spatial spreading ofbranches and therefore higher PI especially in propagationenvironments where path-loss increases rapidly with dis-tance Further i-MIMO DAS planning considerations will beprovided in Section 3

22 Measurement Data Overview In order to effectively setup the system level simulations described in the next sub-section some major parameters of the i-MIMO DAS channelhave been preliminarily investigated through an extensivemeasurement campaign carried out in a typical modernindoor office where up to 4 RAUs have been considered andseveral measurement routes have been deployed in differentrooms and along a corridor All measurements have beenperformed in static conditions at the frequency of 858MHzfor a 2 times 2 i-MIMO arrangement In order to get rid ofthe local multipath effects the measured data have been

International Journal of Antennas and Propagation 3

E F

C D

A B

G H

E

C

A

G

C D

A B

C

A

F

D

B

H

D

B

E F

C D

A B

G H

C D

A B

E F

C D

A B

G H

G H

A B

E F

C D

A B

G H

C D

A B E F

G H

E F

G H

C D

A B

Figure 1 Uniform 2D 8times8 i-MIMODAS coverage Left no-shift interleaved coverage solution Center 1-shift interleaved coverage solutionRight 2-shift interleaved coverage solution The 2 by 4 branch ldquoclusterrdquo is highlighted in black Each colorletter corresponds to a differentMIMO branch-signal

0 005 01 015 02 025 03 035 040

02

04

06

08

x

Prob

|120588|ltx

|120588T|

|120588R|

|120588diag1|

|120588diag2|

Figure 2 CDF of the magnitude of the correlation coefficientsextracted from the measurements

collected several times moving the receiving array over aspatial grid centered on each measurement position Furtherdetails about the measurement campaign can be found in[6 7]

The overall amount of gathered experimental data hasbeen postprocessed to achieve the statistics of both thechannel correlations and the power imbalance In particularFigure 2 shows the cumulative distribution function (CDF)of the magnitude of the complex correlation coefficients (iethe normalized complex covariances between the elementsof the channel matrix) The correlation coefficients at thetransmitterreceiver side and the diagonal correlations havebeen computed according to the following expressions

120588119879119894=

cov (ℎ1198941 ℎlowast

1198942)

radiccov (ℎ1198941 ℎlowast

1198941) cov (ℎ

1198942 ℎlowast

1198942)

1003816100381610038161003816100381610038161003816100381610038161003816100381610038161003816119894=12

120588119877119895=


2119895)


1119895) cov (ℎ

2119895 ℎlowast

2119895)

1003816100381610038161003816100381610038161003816100381610038161003816100381610038161003816119895=12

0 5 10 15 20 25 300

01

02

03

04

05

06

07

08

09

1

x (dB)

Corridor scenario (LOSOLOS)Room scenario (OLOSNLOS)

Prob

PIlt

x

Figure 3 Experimental CDF of the PI in the considered scenarios

120588Diag1 =cov (ℎ

11 ℎlowast

22)


11) cov (ℎ

22 ℎlowast

22)


12 ℎlowast

21)


12) cov (ℎ

21 ℎlowast

21)

(1)

All the CDFs appear rather similar with median valuesalways between 02 and 025 and correlation values seldomlarger than 04 Such results clearly support the assumptionmdashmade at the end of the introductionmdashthat multipath richnessshould not represent a limit to the performance of an i-MIMOsystem which is therefore mainly affected by path-loss andshadowing which practically determine the value of both thePI and the SNR

The experimental CDF of the power imbalance is shownin Figure 3 for two different scenarios in the ldquocorridorrdquo casepropagation occurred in line of sight (LOS) or obstructed-LOS (OLOS) conditions (depending on the RAUs position)whereas OLOS or non-LOS conditions have been observedin the ldquoroomrdquo case [6] The PI values have been computedaccording to the expression introduced in [6] for a 2 times 2


MIMO system The PI values range from 0 to 30 dB with amedian value equal to about 10 dB in both cases though aslightly larger value spread can be noticed in the ldquocorridorscenariordquo as the CDF is less steep around the median value

As described in the next subsection in order to effectivelyassess the performance of i-MIMODAS systems the statisticsshown in Figures 2 and 3 are used to generate realistic randomchannel realizations to be input into the LTE-A link-levelsimulator A more detailed analysis of the MIMO channelcharacteristics in the considered environment can be foundin [6]

23 Performance Evaluation of High-Order i-MIMO DAS Awireless system must be able to cope with high downlinkPI in order to achieve good performance in i-MIMO DASconfiguration For MIMO systems of order higher than 2times 2a unique PI value cannot be defined since different branchpairs can undergo different imbalance values Therefore inorder to estimate the average imbalance level experiencedon a single 119877119909 (mobile) location the following standarddeviation parameter can be adopted [6]

120590119875= radic

1

119899

119899

sum

119894=1

(119875119894minus 120583119875)2 where 120583

119875=

1

119899

119899

sum

119894=1

119875119894 (2)

where 119875119894is the total power (in dB units) received from all

the RAUs radiating the same 119894th MIMO branch-signal 119875119894

is of course the local average of the received signal powerfrom the 119894th 119879119909 branch made over all 119877119909 antenna elementsClearly the parameter120590

119875represents the standard deviation of

the values 1198751 1198752 119875

119899 whereas 120583

119875is the average received

power from all the 119899 119879119909 MIMO branches Assuming aconstant noise power level119873

0for all theMIMObranches the

120590119875parameter can also be expressed as

120590119875= radic

1

119899

119899

sum

119894=1

(SNR119894minus SNR)

2

(3)

where SNR119894= 119875119894(dBm) minus 119873

0(dBm) and SNR = 120583

119875(dBm) minus

1198730(dBm) are the signal-to-noise ratio for the 119894th branch

and the signal-to-noise ratio averaged over all the branchesrespectively

As a benchmark to study i-MIMOperformance and plan-ning strategies in real-life cases the LTE-advanced standardis considered in the present work [1] An LTE-A link-levelsimulator developed at Vienna University of Technology [8]is run in single-userMIMOmode in the 2times2 4times4 and 8times8MIMO cases adopting standard ITU-R power-delay profiles(PDPs) for indoor environment (indoor office channel A) [9]Every simulation run consists of 15 times 100 = 1500 snapshotscorresponding to 15 different SNR values from 0 to 70 dB and100 MIMO channel matrix realizations for each SNR value

For each realization the channel matrices have beenrandomly generated according to the full-correlation model(see [10 page 72]) H = R12H

119908 where R is the 1198992 times 1198992

full-correlation matrix [10] whose elements are randomlygenerated according to the statistics shown in Figure 2 andH119908is a random iid channel matrix

Then random imbalances between the MIMO brancheshave been also introduced according to the PI statisticsshown in Figure 3 In particular for each realization ran-dom power imbalances varying between 0 and 30 dB areintroduced between the columns of the channel matrix bytaking the first column as a reference and properly scaling theothers Finally the generated MIMO channel matrices havebeen normalizedwith respect to the expectation of the squareroot of the average power gain of each channel realizationaccording to

H =

H120573

with 120573 = 119864[

[

[

[

radicsum119894119895

10038161003816100381610038161003816ℎ119894119895

10038161003816100381610038161003816

2

1198992

]

]

]

]

=

119864 [H119865]119899

(4)

where H119865is the Frobenius norm of the channel matrix

(see also [10 page 74]) H is the normalized channel matrixand 119899 is the MIMO order The main input data of the LTE-A simulator are the normalized channel matrices and theaverage signal-to-noise ratio SNR Other simulator parame-ter settings are the same as in [6 11]

The final outcome of the simulations is the downlinkchannel-throughput as a function of 120590

119875and of the average

signal-to-noise ratio as shown in Figures 4 and 5Figures 4 and 5 depict the simulated throughput results

obtained for the 2times2 and 8times8 i-MIMO cases respectively Asimilar result has been found for the 4-branch arrangementnot represented here for the sake of brevity

It is worth noticing that the throughput plot versus SNRand 120590

119901for the 2 times 2 case (119899 = 2) in Figure 4 is very similar

to the corresponding one shown in [6] where the channelmatrices given in input to the LTE simulator have beencomputed through ray tracing simulations [12] in a specificrepresentative indoor environment and not randomly gener-ated on the base of some preassigned statistical distributionsthe satisfactory agreement supports the effectiveness of thegeneral approach here adopted not aiming at investigatingsingle specific cases but rather at identifying some generalbut meaningful planning rules

The results for high-order i-MIMO DAS arrangementsprovide manifold confirmations of the outcomes of theanalysis already carried out in [6] in the 2 times 2 MIMO case

(i) different ldquothroughput zonesrdquo that is portions of the(SNR 120590

119875) plane associated with approximately the

same throughput value are clearly highlighted(ii) the throughput values seem to increase with SNR

and on the contrary decrease for increasing powerimbalance

(iii) the full-throughput achievable within the 8 times 8 case(asymp600Mbps) is 4 times the one achievable in the 2 times2 case (asymp150Mbps) as it should be according to theknown theory that MIMO capacity is proportional toMIMO order if multipath is sufficiently rich

(iv) since the simulated throughput values correspondingto different channel realizations having the same120590119875are very similar 120590

119875together with the SNR can

be considered a reliable performance-determiningparameter


No M

IMO

gain

(le80

Mbp

s)

Full-

thro

ughp

ut(gt140

Mbp

s)

Intermediatethroughput zones

160

140

120

100

80

60

40

20

0

120590P

(dB)

SNR (dB)

12

10

8

6

4

2

10 20 30 40 50 60 70

Figure 4 Downlink throughput values provided by the LTE-A link-level simulator for the 2 times 2 i-MIMO case as a function of average SNRand 120590

119875

600

525

450

375

300

225

150

75

120590P

(dB)

SNR (dB)

12

10

8

6

4

2

100 20 30 40 50 60 70


Full-

thro

ughp

ut

No M

IMO ga

in


119875

With regard to the planning guidelines of Section 3 itshould be highlighted that both 120590

119875and SNR have been

defined as ldquoincoherentrdquo large-scale parameters that is theirvalues are independent of the phase relations between the sig-nals at the119877119909 antennas i-MIMODAS systemplanning can betherefore based on path-loss and shadowing (obstructions)considerations only neglecting small-scale fading effectsThe throughput plots shown above and in particular theinformation regarding the throughput zones are then usedin the present work in combination with a proper path-loss formula to map performance for different i-MIMO DASdeployments as shown in the following section

It is worth noticing that alsomultiuserMIMO is includedin the LTE-A standard where different branches in an i-MIMO scheme would be used to serve different usersAlthough the i-MIMO concept and the methodological

approach of this work are still valid the PI and thus the120590119901will

play a different role in that case In fact interuser interferenceminimization considerations might suggest pursuing high PIvalues which is the opposite of what should be pursued insingle-user cases

3 Planning Considerations

In this section considerations and guidelines for the planningand the deployment of i-MIMO DAS systems in referenceregular 1D (corridor) 2D (single floor) and 3D (multifloor)layouts are provided on the base of both the throughputplots of Section 22 and the following path-loss formulaintroduced and parameterized in [13]

PLdB (119889) = PL (1198890) + 10 sdot 120572 sdot log( 119889

1198890

) + 120573 sdot 119889 (5)


where 120572 is the path-loss exponent mainly accounting forthe wavefront divergence and 120573 is the specific-attenuationaccounting for obstructions Equation (5) has been shown towell reproduce path-loss as a function of link distance com-pared to measurements in reference indoor environments

In [13] it was also shown that the values of the specific-attenuation lie between 05 and 15 in most indoor officescenarios depending on the electromagnetic characteristicsand thickness of the walls while the most appropriate valuefor alpha is equal to 2 with the exception of large indoorscenarios such as shoppingmalls or airports For this reasonin the followingwewill assume120572 = 2 and 0 le 120573 le 15 (120573= 05is the reference value for themeasurement scenario describedin Section 22 as shown in [13])

Given a propagation environment that is specific-atten-uation 120573 in (5) throughput performance mainly depends onPI (120590119901) and SNR as shown in Section 22Therefore i-MIMO

DAS planning can be based on a two-step procedure

(1) SNR-Based Planning RAU antenna spacing and 119879119909power should be chosen so as to guarantee a suf-ficiently high SNR all over the service area and inparticular at cell border

(2) PI-Based Planning The different MIMO branch-signals should be distributed over the RAU antennasin a proper ldquointerleavedrdquo fashion to minimize 120590

119875 as

shown in Section 22 In addition since high-orderi-MIMO DAS deployments andor high 120573 values in(5) lead to high PI between the different branch-signals then MIMO order should be matched to thepropagation characteristics in order for 120590

119875to remain

low enough In other words an increase in theMIMOorder could yield a less-than-proportional increase inthroughput if PL increases too rapidly with distanceso as to produce a too-high power imbalance

There is therefore affinity between i-MIMO DAS plan-ning and traditional cellular planning where radio coveragemust be provided in the first step and then the cluster-sizemust be chosen so as to satisfy signal-to-interference (SIR)requirements Here the MIMO order n plays a similar role ascluster-size while PI replaces SIR

Since planning step (1) does not depend on the planningstrategy and higher attenuation conditions or greater RAUspacing can be easily compensated with a proportional 119879119909power increase step (1)will not be addressed in the followingA typical RAU antenna spacing of 20m will be consideredwhereas realistic119879119909 power and120573 values will be assumed fromcase to case

On the contrary planning step (2) will be investigated insome detail In particular (5) can easily provide the SNR andthe 120590119875values for every possible position of the receiver on

the service area then the corresponding system performancecan be evaluated by means of the throughput plots obtainedwith the LTE-A simulator that is by identifying the corre-sponding throughput zone for each pair (SNR 120590

119875)

It is worth noticing that hereafter the performance of thei-MIMO solutions will not be evaluated in terms of absolutethroughput but rather in terms of ldquoaverage (relative) gain

with respect to the SISO caserdquo (119899119866) which is of course defined

as the ratio between the actual system throughput (evaluatedthrough the previously described procedure) and the SISOmaximum throughput

119899119866=

Average MIMO throughputMax SISO throughput

(0 le 119899119866le 119899) (6)

According to this definition 119899119866ranges from 0 to 119899 119899

being theMIMO order (2 4 or 8 have been here considered)The 119899

119866value can be interpreted as the ldquoequivalentrdquo number

of branches which contribute to achieving the full MIMOcapacity gain (useful branches) For instance if 119899

119866is equal

to 119899 it means that all branches are successfully exploitedand therefore the full MIMO throughput is achieved In thefollowing text 119899

119866

def= 119899 minus 119899

119866instead of 119899

119866is often used since

the corresponding plots usually appearmore readable 119899119866can

be interpreted as the number of branches ldquounder-thresholdrdquowhich do not contribute to the MIMO capacity gain

In general if 119899119866≪ 119899 (or equivalently 119899

119866≫ 0) then the

considered interleaved solution is seriously ineffective Thisis due to the above-mentioned problem when the MIMOorder 119899 is overdimensioned then the i-MIMO ldquoclusterrdquo is toobig and power imbalance is too high to effectively exploit allMIMO branches

It is therefore very important to choose the appropriateMIMO order 119899 for the considered propagation environmentto avoid unprofitable efforts (high-order systems are muchmore complex and expensive than low-order systems) for noor little performance gain

If 119899119866≪ 119899 adequate countermeasures should be adopted

such as a different distribution of the MIMO branches inthe space an increase of the transmitted RAU power or ifthe above-mentioned parameters are already optimized areduction of the MIMO order

In summary the present section provides the followinguseful results

(i) different i-MIMO DAS deployment solutions areevaluated to identify the best one especially in 2Dand 3D cases where different apparently equivalentsolutions are possible as highlighted in Section 21

(ii) for each reference case the equivalent number ofuseful branches is determined as a function of RAU119879119909 power and specific-attenuation 120573 this is a veryimportant result to optimize planning and avoidwaste of power or MIMO order overdimensioning

Three reference regular 1D (corridor) 2D (single floor) and3D (multifloor) layouts are considered in Sections 31 32and 33 respectively In all cases the considered service areathat is the target area where mobile terminals are locatedis assumed separated in space (15m apart along a verticalaxis) from the plane where RAUs are located (eg the ceiling)to avoid power singularities This means that the mobileterminal is always at a distance of at least 15m away fromthe closest RAU antenna A uniformmobile user distributionis assumed in all cases


0

5

10

15

20

25

30

35

40

45

50

55

60

B C D A

SNR

120590P

120590P (120573 = 0)

SNR (120573 = 0)120590P (120573 = 05)

SNR (120573 = 05)120590P (120573 = 15)

SNR (120573 = 15)

Figure 6 Behaviour of 120590119875

and SNR in a 4-branch i-MIMOlinear deployment for different values of the propagation modelparameters (RAU antenna distance 20m)

31 Linear Deployment (RAUs Placed along a Corridor) If alinear deployment is considered (eg RAUs positioned alonga corridor) and assuming that the receiver moves along aroute parallel to the corridor where the RAUs are locatedboth SNR and 120590

119875values peak in proximity to each RAU and

on the contrary reach their minimum close to the middlepoint between adjacent RAUs

This behavior is clearly confirmed by the example inFigure 6 which refers to a 4-branch arrangement for different120573 values (120573 = 0 of course corresponds to an ideal LOScondition)

It is evident that the better the SNR the worse the PI andvice versa therefore the system design (ie the RAUsrsquo spatialdeployment and the choice of theMIMOorder) should aim atproperly balancing coverage (SNR) and imbalance (120590

119875) over

the service areaFigure 7 represents 119899

119866= 119899minus119899

119866as a function of the power

transmitted by each RAU antenna for differentMIMO ordersand different values of the propagation parameter

It is worth noticing that the larger the MIMO orderthe higher the performance sensitivity to the propagationconditions In fact 119899

119866seems to be weakly dependent on 120573 in

the 2 times 2 case whereas performance is dramatically differentfor different 120573 values in the 8 times 8 case

Low 119899119866

values in Figure 7 should be pursued whenplanning a linear i-MIMO DAS system In practice the 2-branch case seems to be almost always a feasible solutionwhereas the adoption of the 4-branch deployment could bequestionable with 120573 = 15 since a quite high power level(at least 25 dBm) would be required to achieve a good

0

1

2

3

4

5

6

7

8

RAU transmitted power (dBm)minus40 minus20 0 20 40

nG

2 times 2 i-MIMO (LOS case)2 times 2 i-MIMO (120573 = 05)2 times 2 i-MIMO (120573 = 15)4 times 4 i-MIMO (LOS case)4 times 4 i-MIMO (120573 = 05)

4 times 4 i-MIMO (120573 = 15)8 times 8 i-MIMO (LOS case)8 times 8 i-MIMO (120573 = 05)8 times 8 i-MIMO (120573 = 15)

Figure 7 Comparison of 2- 4- and 8-branch i-MIMO DASperformance in the linear deployment (corridor) case Averagenumber of branches under threshold (119899

119866) is shown for different

values of the propagationmodel parameters (RAU antenna distance20m)

10 20 30 40

5

10

15

20

x (m)

y (m

)

0

05

1

15

2

A B

TX power minus14dBm 120573 = 05

nG

Figure 8 Performance of 2 times 2 i-MIMO DAS in a 2D deploymentcase Relative gain with respect to SISO is shown (RAU antennapower minus14 dBm 120573 = 05 RAU antenna distance 20m)

performance For the same reason the 8-branch solution isclearly unfit except for the LOS case

32 2D Deployments The combined effects of SNR and 120590119901

on the performance of an i-MIMO DAS system are furtherexplained in Figures 8 and 9 which represent two examples


10 20 30 40

10

20

30

40

0

1

2

3

4

C

B

D

A

TX power 0dBm 120573 = 15

x (m)

y (m

)

nG

Figure 9 Performance of 4 times 4 i-MIMO DAS in a 2D deploymentcase Relative gain with respect to SISO is shown (RAU antennapower 0 dBm 120573 = 15 RAU antenna distance 20m)

of 2D spatial distribution of the 119899119866values over the cluster area

in the 2- and the 4-branch arrangement respectivelyIn both cases the best performance (ie the highest 119899

119866

values) is achieved not too close to the RAUs where the highimbalance prevails over the large SNR values or too far fromthem where the received signal (and therefore the SNR) isweak

Two different values of 120573 are considered in Figures 8 and9 to better highlight the different performance zones

The effectiveness of the interleaved approach is evaluatedin Figure 10 for different shift value The analysis is limited tothe 8-branch arrangement since only a negligible differencehas been appreciated between the shift and the no-shiftsolutions in the 4-branch case

The benefit provided by a shifted deployment is slight if 120573= 05 whereas it becomesmore evident for 120573 = 15 a plausiblereason is that in presence of more favorable propagationconditions (120573 = 05) the power imbalance is likely undercontrol (on the average) even without any shift on thecontrary worse propagation conditions (120573 = 15) produce anincrease of the imbalance which can be then partly reducedintroducing the shift which restores a more uniform branch-power spatial distribution

A performance comparison for different MIMO ordervalues is presented in Figures 11 and 12 in terms of 119899

119866and

percentage of area where users achieve ldquogood throughputrdquothat is a throughput greater than 75 of the maximumachievable given the MIMO order

Similarly to the 1D case Figure 11 highlights the fact thata MIMO solution with order 2 can easily provide satisfactoryperformance irrespective of the propagation condition in the2D case On the contrary the theoretical potential of the8-branch arrangement cannot be fully exploited in hostilepropagation environment if 120573 = 15 and assuming 25 dBmas the upper limit for the RAU transmitted power then 119899

119866

cannot exceed 5

0

2

4

6

8


nG

No-shift 8 times 8 i-MIMO (120573 = 05)

1-shift 8 times 8 i-MIMO (120573 = 05)





Figure 10 Performance comparison of different 8-branch i-MIMO2D deployments (no-shift 1-shift and 2-shift) Average number ofbranches under threshold (119899

119866) is shown for different values of the

propagation model parameters (RAU antenna distance 20m)

The 4-branch case represents an intermediate case butmore specifically it shows a more marked similarity to the 2-branch solution since the RAU transmitted power requiredto reach the full MIMO throughput (119899

119866= 0) is almost

the same for the two deployments (Figure 11) This is alsoconfirmed in Figure 12 where the curves related to the 2-and 4-branch cases are nearly overlapped From Figure 12the minimum 119879119909 power to achieve good performance can beinferred in the different cases For example in a 4times4 case with120573 = 15 a 119879119909 power of at least a couple of dBm is necessaryto achieve ldquogood throughputrdquo (ie a throughput greater than75 of the maximum) over 90 of the coverage area

33 3D Deployments (Outline) For the sake of brevity theanalyses of the 3D case are here limited to the solutionrepresented in Figure 13 where a 4th order i-MIMO DASsystem is deployed over 3 different floors

The 2 times 2 square cluster is repeated over each flooraccording to the no-shift coverage solution shown in Figure 3(center) whereas a possible ldquodiagonal shiftrdquo (ie achievedthrough two subsequent shifts in the 119909- and in the 119910-direction resp) is considered between adjacent floors

Moreover an additional floor penetration loss of 10 dB isconsidered for the evaluation of the intensity of the signalsreceived from RAUs placed on different floors with respect tothe receiver


0

2

4

6

8


nG


2 times 2 i-MIMO (120573 = 05)2 times 2 i-MIMO (120573 = 15)4 times 4 i-MIMO (120573 = 05)4 times 4 i-MIMO (120573 = 15)


Figure 11 Comparison of 2- 4- 8-branch i-MIMO DAS averageperformance in the 2D deployment case Average number ofbranches under threshold (119899



System performance is represented in Figure 14 andclearly shows that no significant performance improvementis provided by the shifted solution for 120573 = 05 In this casethe achieved result is essentially the same as in the 2D case(Figure 11) propagation over the same floor is dominant withrespect to ldquointerfloorrdquo propagation

On the contrary if 120573 = 15 ldquointrafloorrdquo propaga-tion becomes more difficult and therefore signals comingfrom different floors are no longer negligible According toFigure 14 the shift solution yields in this case better per-formance as it providesmdashon the averagemdasha lower powerimbalance

4 Conclusion

Planning strategies for interleaved-MIMO DAS indoorcoverage-extension solutions are studied in the present paperAlthough LTE-A standard is adopted as a benchmark herethe proposed methodology which is based on large-scalepropagation parameters such as SNR and power imbalance(PImeasured through120590

119875) and on a simple path-loss formula

is valid also for interleaved-MIMO DAS solutions imple-mented on different systems

Results show that proper interleaved deployment solu-tions based on the concept of ldquoclusterrdquo should be adopted tokeep PI low enough In some 2D and 3D deployment cases acluster-shifting can help further performance optimization

PI is shown to increase with both MIMO order 119899 andspecific-attenuation 120573 and to decrease with coverage-layout

0

20

40

60

80

100

Are

a with

mor

e tha

n 75

o

f ful

l thr

ough

put (

)


2 times 2 i-MIMO (120573 = 05)2 times 2 i-MIMO (120573 = 15)4 times 4 i-MIMO (120573 = 05)


Figure 12 Comparison of 2- 4- and 8-branch i-MIMO DAS per-formance ( of area where more than 75 of maximum throughputis achieved) in the 2D deployment case for different values of thepropagation model parameters (RAU antenna distance 20m)

A

A

A

B

B

B

C

C

C D

D

D

A

D

D

B

C

C

B

C

B A

D

A

x

y

Figure 13 3D 4 times 4 i-MIMO DAS coverage Left noninterleavedcoverage solution Right coverage solution with branch interleavingbetween different floors

dimensions as 2D or 3D solutions allow a more uniformhearability of a larger number of MIMO branches at thegeneric 119877119909 position than 1D solutions

It is wrong to adopt say a costly 8times8 i-MIMODAS solu-tion in a 1D-deployment with a large specific-attenuation asmost MIMO branches would be useless and full-throughputcould never be achieved

Therefore useful planning guidelines are proposed andgraphs are derived to help the i-MIMOsystemdesigner deter-mine the maximum MIMO order for which performance isnot yet limited by power imbalance and therefore the best i-MIMO solution for given traffic and propagation conditions


0

1

2

3

4


nG

4 times 4 i-MIMO-3D deployment-no shift (120573 = 05)

4 times 4 i-MIMO-3D deployment-shifted (120573 = 05)



Figure 14 Performance comparison of 4 times 4 i-MIMO no-shiftand diagonal-shifted 3D deployments Average number of branchesunder threshold (119899

119866) is shown (RAU antenna horizontal distance

20m RAU antenna vertical distance 3m floor attenuation factor10 dB)

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgments

This work has been carried out in the framework of andpartly funded by the EU Project Architectures for FlexiblePhotonic Home and Access Networks (ALPHA) ICT CP-IP 212 352 This work has also been partly funded by theEuropean Network of Excellence NEWCOM++The authorsthank their graduate student Stefano Fiaschi for the great helpin carrying out system simulations

References

[1] IEEE 3GPP TS 36211 V1110 ldquoPhysical channels and modula-tion (Release 11)rdquo 2012

[2] A A M Saleh A J Rustako Jr and R S Roman ldquoDistributeAntennas for Indoor Radio Communicationsrdquo IEEE Transac-tions on Communications vol 35 no 12 pp 1245ndash1251 1988

[3] M Fabbri and P Faccin ldquoRadio over fiber technologies and sys-tems new opportunitiesrdquo in Proceeding of the 9th InternationalConference on Transparent Optical Networks (ICTON 07) pp230ndash233 Rome Italy July 2007

[4] R Ibernon-Fernandez J-M Molina-Garcia-Pardo and LJuan-Llacer ldquoComparison between measurements and simu-lations of conventional and distributed MIMO systemrdquo IEEEAntennas and Wireless Propagation Letters vol 7 pp 546ndash5492008

[5] M Alatossava A Taparugssanagorn V-M Holappa and JYlitalo ldquoMeasurement based capacity of distributed MIMOantenna system in Urban microcellular environment at 525GHzrdquo in Proceedings of the IEEE 67th Vehicular TechnologyConference-Spring (VTC rsquo08) pp 430ndash434 May 2008

[6] E M Vitucci L Tarlazzi F Fuschini P Faccin and V Degli-Esposti ldquoInterleaved-MIMO DAS for Indoor radio coverageconcept and performance assessmentrdquo IEEE Transactions onAntennas and Propagation vol 62 no 6 pp 3299ndash3309 2014

[7] L Tarlazzi P Faccin E M Vitucci F Fuschini and V Degli-Esposti ldquoCharacterization of an interleaved F-DAS MIMOindoor propagation channelrdquo in Proceedings of the 6th Lough-borough Antennas and Propagation Conference (LAPC rsquo10) pp505ndash508 Loughborough UK November 2010

[8] C Mehlfuhrer M Wrulich J C Ikuno D Bosanska and MRupp ldquoSimulating the long term evolution physical layerrdquo inProceedings of the 17th European Signal Processing Conference(EUSIPCO rsquo09) Glasgow UK 2009

[9] ITU-R Recommendation M1225 ldquoGuidelines for evaluation ofradio transmission technologies for IMT-2000rdquo 1997

[10] B Clerckx and C OestgesMIMOWireless Networks AcademicPress (Elsevier) Oxford UK 2nd edition 2013

[11] E M Vitucci L Tarlazzi P Faccin and V Degli-EspostildquoAnalysis of the performance of LTE systems in an InterleavedF-DAS MIMO indoor environmentrdquo in Proceedings of the 5thEuropean Conference on Antennas and Propagation (EuCAPrsquo07) pp 2184ndash2186 Rome Italy April 2007

[12] EMVitucci VDegli-Esposti andF Fuschini ldquoMIMOchannelcharacterization through ray tracing simulationrdquo in Proceedingsof the 1st European Conference on Antennas and Propagation(EuCAP rsquo06) Nice France November 2006

[13] V Degli-Esposti G Falciasecca F Fuschini and E M VituccildquoA meaningful indoor path-loss formulardquo IEEE Antennas andWireless Propagation Letters vol 12 pp 872ndash875 2013

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014


Active and Passive Electronic Components

Control Scienceand Engineering

Journal of



RotatingMachinery


Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design



Shock and Vibration


Civil EngineeringAdvances in

Acoustics and VibrationAdvances in



Electrical and Computer Engineering

Journal of

Advances inOptoElectronics


Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of


Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014


Chemical EngineeringInternational Journal of Antennas and

Propagation




Navigation and Observation



DistributedSensor Networks



must be therefore properly interleaved over the environmentin order to simultaneously guarantee good radio coverageand low PI therefore optimizing MIMO capacity Suchdeployment strategy is called ldquointerleaved-MIMO DASrdquo (i-MIMODAS) and has been introduced and illustrated for thefirst time in [6 7] with particular reference to 2 times 2 MIMOsystems

Interleaved-MIMO DAS schemes are being consideredwith great interest by operators and installers because theymight yield a performance level comparable to that of acolocated MIMO DAS (119888-MIMO DAS) at a much lowercost [6] However high PI values might degrade MIMOperformance to an extent that still needs to be evaluatedin relation to the DAS coverage advantages especially forhigh-order MIMO schemes As already anticipated in [6]the actual PI value distribution depends on the way the 119899branch antennas are interleaved in space and therefore onthe i-MIMO planning strategy Of course PI also depends onthe mobile users spatial distribution but the latter cannot becontrolled and has been assumed uniform in the followingstudy

For these reasons the i-MIMO DAS concept is furtherinvestigated and discussed for higher-order MIMO schemesin the present paper where LTE-advanced link-level simu-lations (Section 2) together with a simple but meaningfulpath-loss formula are used to derive some general planningguidelines (Section 3) for high-order MIMO and for 1D(corridor) 2D and 3D (multifloor) coverage scenarios Suchplanning guidelines can be useful for the system designer toproperly choose the MIMO order and i-MIMO deploymentto optimize the performance to cost ratio It is worth specify-ing that our study is limited to the planning of a DAS systemand not of thewholemobile radio network and therefore doesnot consider channel allocation and interference issues

In order to provide planning guidelines a single bench-mark system standard (LTE-advanced) has been consideredand the problem has been simplified by taking into accountonlymajor parameters such as the signal-to-noise ratio (SNR)and the PI that depend on the particular i-MIMO DASdeployment Other factors such as the multipath richnessand other propagation characteristics can have an importantimpact on performance but they cannot be engineered asthey primarily depend on the propagation environment andtherefore their role has not been deeply analyzed here In thefollowing we will simply assume the multipath richness to belarge enough not to represent a limit to MIMO performance

2 Interleaved-MIMO DAS Evaluation

21 The Concept The application of the MIMO concept todistributed antenna systems is analyzed here with particularreference to the spatial multiplexing capability althoughother transmission techniques such as 119879119909 diversity are ofcourse possibleTherefore in the following each RAU (whichcarries a single ldquoMIMO branchrdquo) is supposed to transmita MIMO signal different from those radiated by the otherRAUs belonging to the same (distributed) transmitting arrayFurthermore since each RAU defines a ldquocellrdquo that is thearea where its signal strength is dominant (usually square or

rectangular cells in indoor scenarios instead of hexagonalcells due to the shape of rooms and buildings) each cell canbe therefore associated with a corresponding MIMO branch

The present work considers the implementation of i-MIMO DAS in large indoor environments with 119872 RAUlocations (with 119872 ≫ 119899 where 119899 is the MIMO order)where multiple repetitions of the 119899MIMO branch-signals aredeployed over the service area The acronym ldquo119899 times 119899 i-MIMO119872-DASrdquo can be used to indicate an 119899th-order i-MIMO DASsolution with M RAUs where each RAU is equipped with asingle antenna element and the mobile terminal is equippedwith an array of colocated antenna elements [6]

In case a linear layout is required andor low-order i-MIMO schemes are considered the best spatial deploymentfor the differentMIMO branches can be often handled on thebase of somehow evident intuitive considerations [6]

In a 2D3D coverage case and for high-order MIMOhowever the best deployment scheme may be no longertrivial as somedifferent interleaved solutionsmight exist andshould be therefore evaluated

In order to limit the spatial PI it is intuitive that thedifferent MIMO branches should be somehow arranged insquarerectangular ldquoclustersrdquo (Figure 1) Of course differentclusters can be laid over the service area in different waysfor instance in the 8-branch case assuming a 2 by 4 clusterscheme the three solutions represented in Figure 1 can beconsidered the clusters can be simply aligned to each other(Figure 1mdashleft) or a 1-2-cell shift can be introduced betweenadjacent columns (or rows) of clusters (Figure 1 center andright resp)

Intuitively the 2-shift solution should lead to a moreuniform branch-signal distribution however actual perfor-mance must be assessed by simulation as shown in Section 3

In conclusion irrespective of the i-MIMO order theconcept of ldquoclusterrdquo widely used in cellular radio planningapplies here However here the mechanisms are somewhatreversed as in traditional cellular planning the cluster-sizeshould be large enough to separate cochannel cells andthereby minimize interference and in i-MIMO DAS thecluster dimension and therefore the MIMO order 119899 shouldbe small enough to minimize power imbalance In fact high-order MIMO deployments (such as the 8-branch case ofFigure 1) will necessarily yield greater spatial spreading ofbranches and therefore higher PI especially in propagationenvironments where path-loss increases rapidly with dis-tance Further i-MIMO DAS planning considerations will beprovided in Section 3

22 Measurement Data Overview In order to effectively setup the system level simulations described in the next sub-section some major parameters of the i-MIMO DAS channelhave been preliminarily investigated through an extensivemeasurement campaign carried out in a typical modernindoor office where up to 4 RAUs have been considered andseveral measurement routes have been deployed in differentrooms and along a corridor All measurements have beenperformed in static conditions at the frequency of 858MHzfor a 2 times 2 i-MIMO arrangement In order to get rid ofthe local multipath effects the measured data have been


E F

C D

A B

G H

E

C

A

G

C D

A B

C

A

F

D

B

H

D

B

E F

C D

A B

G H

C D

A B

E F

C D

A B

G H

G H

A B

E F

C D

A B

G H

C D

A B E F

G H

E F

G H

C D

A B


0 005 01 015 02 025 03 035 040

02

04

06

08

x

Prob

|120588|ltx

|120588T|

|120588R|

|120588diag1|

|120588diag2|




120588119879119894=


1198942)


1198941) cov (ℎ

1198942 ℎlowast

1198942)

1003816100381610038161003816100381610038161003816100381610038161003816100381610038161003816119894=12

120588119877119895=


2119895)


1119895) cov (ℎ

2119895 ℎlowast

2119895)

1003816100381610038161003816100381610038161003816100381610038161003816100381610038161003816119895=12

0 5 10 15 20 25 300

01

02

03

04

05

06

07

08

09

1

x (dB)


Prob

PIlt

x



11 ℎlowast

22)


11) cov (ℎ

22 ℎlowast

22)


12 ℎlowast

21)


12) cov (ℎ

21 ℎlowast

21)

(1)







120590119875= radic

1

119899

119899

sum

119894=1

(119875119894minus 120583119875)2 where 120583

119875=

1

119899

119899

sum

119894=1

119875119894 (2)





the values 1198751 1198752 119875

119899 whereas 120583





120590119875= radic

1

119899

119899

sum

119894=1


2

(3)


0(dBm) and SNR = 120583

119875(dBm) minus








H =

H120573

with 120573 = 119864[

[

[

[


10038161003816100381610038161003816ℎ119894119895

10038161003816100381610038161003816

2

1198992

]

]

]

]

=

119864 [H119865]119899

(4)




















No M

IMO

gain

(le80

Mbp

s)

Full-

thro

ughp

ut(gt140

Mbp

s)


160

140

120

100

80

60

40

20

0

120590P

(dB)

SNR (dB)

12

10

8

6

4

2

10 20 30 40 50 60 70


119875

600

525

450

375

300

225

150

75

120590P

(dB)

SNR (dB)

12

10

8

6

4

2

100 20 30 40 50 60 70


Full-

thro

ughp

ut

No M

IMO ga

in


119875










1198890

) + 120573 sdot 119889 (5)








119875 as


119875to remain






119875)




119899119866=


(0 le 119899119866le 119899) (6)





119866is equal


119866

def= 119899 minus 119899

119866instead of 119899





119866≫ 0) then the










0

5

10

15

20

25

30

35

40

45

50

55

60

B C D A

SNR

120590P

120590P (120573 = 0)

SNR (120573 = 0)120590P (120573 = 05)

SNR (120573 = 05)120590P (120573 = 15)

SNR (120573 = 15)








119875) over


119866= 119899minus119899






Low 119899119866


0

1

2

3

4

5

6

7

8


nG






10 20 30 40

5

10

15

20

x (m)

y (m

)

0

05

1

15

2

A B


nG






10 20 30 40

10

20

30

40

0

1

2

3

4

C

B

D

A


x (m)

y (m

)

nG




119866






119866and



119866

cannot exceed 5

0

2

4

6

8


nG

















0

2

4

6

8


nG









4 Conclusion






0

20

40

60

80

100

Are

a with

mor

e tha

n 75

o

f ful

l thr

ough

put (

)





A

A

A

B

B

B

C

C

C D

D

D

A

D

D

B

C

C

B

C

B A

D

A

x

y






0

1

2

3

4


nG










Acknowledgments


References
















RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










E F

C D

A B

G H

E

C

A

G

C D

A B

C

A

F

D

B

H

D

B

E F

C D

A B

G H

C D

A B

E F

C D

A B

G H

G H

A B

E F

C D

A B

G H

C D

A B E F

G H

E F

G H

C D

A B


0 005 01 015 02 025 03 035 040

02

04

06

08

x

Prob

|120588|ltx

|120588T|

|120588R|

|120588diag1|

|120588diag2|




120588119879119894=


1198942)


1198941) cov (ℎ

1198942 ℎlowast

1198942)

1003816100381610038161003816100381610038161003816100381610038161003816100381610038161003816119894=12

120588119877119895=


2119895)


1119895) cov (ℎ

2119895 ℎlowast

2119895)

1003816100381610038161003816100381610038161003816100381610038161003816100381610038161003816119895=12

0 5 10 15 20 25 300

01

02

03

04

05

06

07

08

09

1

x (dB)


Prob

PIlt

x



11 ℎlowast

22)


11) cov (ℎ

22 ℎlowast

22)


12 ℎlowast

21)


12) cov (ℎ

21 ℎlowast

21)

(1)







120590119875= radic

1

119899

119899

sum

119894=1

(119875119894minus 120583119875)2 where 120583

119875=

1

119899

119899

sum

119894=1

119875119894 (2)





the values 1198751 1198752 119875

119899 whereas 120583





120590119875= radic

1

119899

119899

sum

119894=1


2

(3)


0(dBm) and SNR = 120583

119875(dBm) minus








H =

H120573

with 120573 = 119864[

[

[

[


10038161003816100381610038161003816ℎ119894119895

10038161003816100381610038161003816

2

1198992

]

]

]

]

=

119864 [H119865]119899

(4)




















No M

IMO

gain

(le80

Mbp

s)

Full-

thro

ughp

ut(gt140

Mbp

s)


160

140

120

100

80

60

40

20

0

120590P

(dB)

SNR (dB)

12

10

8

6

4

2

10 20 30 40 50 60 70


119875

600

525

450

375

300

225

150

75

120590P

(dB)

SNR (dB)

12

10

8

6

4

2

100 20 30 40 50 60 70


Full-

thro

ughp

ut

No M

IMO ga

in


119875










1198890

) + 120573 sdot 119889 (5)








119875 as


119875to remain






119875)




119899119866=


(0 le 119899119866le 119899) (6)





119866is equal


119866

def= 119899 minus 119899

119866instead of 119899





119866≫ 0) then the










0

5

10

15

20

25

30

35

40

45

50

55

60

B C D A

SNR

120590P

120590P (120573 = 0)

SNR (120573 = 0)120590P (120573 = 05)

SNR (120573 = 05)120590P (120573 = 15)

SNR (120573 = 15)








119875) over


119866= 119899minus119899






Low 119899119866


0

1

2

3

4

5

6

7

8


nG






10 20 30 40

5

10

15

20

x (m)

y (m

)

0

05

1

15

2

A B


nG






10 20 30 40

10

20

30

40

0

1

2

3

4

C

B

D

A


x (m)

y (m

)

nG




119866






119866and



119866

cannot exceed 5

0

2

4

6

8


nG

















0

2

4

6

8


nG









4 Conclusion






0

20

40

60

80

100

Are

a with

mor

e tha

n 75

o

f ful

l thr

ough

put (

)





A

A

A

B

B

B

C

C

C D

D

D

A

D

D

B

C

C

B

C

B A

D

A

x

y






0

1

2

3

4


nG










Acknowledgments


References
















RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation













120590119875= radic

1

119899

119899

sum

119894=1

(119875119894minus 120583119875)2 where 120583

119875=

1

119899

119899

sum

119894=1

119875119894 (2)





the values 1198751 1198752 119875

119899 whereas 120583





120590119875= radic

1

119899

119899

sum

119894=1


2

(3)


0(dBm) and SNR = 120583

119875(dBm) minus








H =

H120573

with 120573 = 119864[

[

[

[


10038161003816100381610038161003816ℎ119894119895

10038161003816100381610038161003816

2

1198992

]

]

]

]

=

119864 [H119865]119899

(4)




















No M

IMO

gain

(le80

Mbp

s)

Full-

thro

ughp

ut(gt140

Mbp

s)


160

140

120

100

80

60

40

20

0

120590P

(dB)

SNR (dB)

12

10

8

6

4

2

10 20 30 40 50 60 70


119875

600

525

450

375

300

225

150

75

120590P

(dB)

SNR (dB)

12

10

8

6

4

2

100 20 30 40 50 60 70


Full-

thro

ughp

ut

No M

IMO ga

in


119875










1198890

) + 120573 sdot 119889 (5)








119875 as


119875to remain






119875)




119899119866=


(0 le 119899119866le 119899) (6)





119866is equal


119866

def= 119899 minus 119899

119866instead of 119899





119866≫ 0) then the










0

5

10

15

20

25

30

35

40

45

50

55

60

B C D A

SNR

120590P

120590P (120573 = 0)

SNR (120573 = 0)120590P (120573 = 05)

SNR (120573 = 05)120590P (120573 = 15)

SNR (120573 = 15)








119875) over


119866= 119899minus119899






Low 119899119866


0

1

2

3

4

5

6

7

8


nG






10 20 30 40

5

10

15

20

x (m)

y (m

)

0

05

1

15

2

A B


nG






10 20 30 40

10

20

30

40

0

1

2

3

4

C

B

D

A


x (m)

y (m

)

nG




119866






119866and



119866

cannot exceed 5

0

2

4

6

8


nG

















0

2

4

6

8


nG









4 Conclusion






0

20

40

60

80

100

Are

a with

mor

e tha

n 75

o

f ful

l thr

ough

put (

)





A

A

A

B

B

B

C

C

C D

D

D

A

D

D

B

C

C

B

C

B A

D

A

x

y






0

1

2

3

4


nG










Acknowledgments


References
















RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










No M

IMO

gain

(le80

Mbp

s)

Full-

thro

ughp

ut(gt140

Mbp

s)


160

140

120

100

80

60

40

20

0

120590P

(dB)

SNR (dB)

12

10

8

6

4

2

10 20 30 40 50 60 70


119875

600

525

450

375

300

225

150

75

120590P

(dB)

SNR (dB)

12

10

8

6

4

2

100 20 30 40 50 60 70


Full-

thro

ughp

ut

No M

IMO ga

in


119875










1198890

) + 120573 sdot 119889 (5)








119875 as


119875to remain






119875)




119899119866=


(0 le 119899119866le 119899) (6)





119866is equal


119866

def= 119899 minus 119899

119866instead of 119899





119866≫ 0) then the










0

5

10

15

20

25

30

35

40

45

50

55

60

B C D A

SNR

120590P

120590P (120573 = 0)

SNR (120573 = 0)120590P (120573 = 05)

SNR (120573 = 05)120590P (120573 = 15)

SNR (120573 = 15)








119875) over


119866= 119899minus119899






Low 119899119866


0

1

2

3

4

5

6

7

8


nG






10 20 30 40

5

10

15

20

x (m)

y (m

)

0

05

1

15

2

A B


nG






10 20 30 40

10

20

30

40

0

1

2

3

4

C

B

D

A


x (m)

y (m

)

nG




119866






119866and



119866

cannot exceed 5

0

2

4

6

8


nG

















0

2

4

6

8


nG









4 Conclusion






0

20

40

60

80

100

Are

a with

mor

e tha

n 75

o

f ful

l thr

ough

put (

)





A

A

A

B

B

B

C

C

C D

D

D

A

D

D

B

C

C

B

C

B A

D

A

x

y






0

1

2

3

4


nG










Acknowledgments


References
















RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation
















119875 as


119875to remain






119875)




119899119866=


(0 le 119899119866le 119899) (6)





119866is equal


119866

def= 119899 minus 119899

119866instead of 119899





119866≫ 0) then the










0

5

10

15

20

25

30

35

40

45

50

55

60

B C D A

SNR

120590P

120590P (120573 = 0)

SNR (120573 = 0)120590P (120573 = 05)

SNR (120573 = 05)120590P (120573 = 15)

SNR (120573 = 15)








119875) over


119866= 119899minus119899






Low 119899119866


0

1

2

3

4

5

6

7

8


nG






10 20 30 40

5

10

15

20

x (m)

y (m

)

0

05

1

15

2

A B


nG






10 20 30 40

10

20

30

40

0

1

2

3

4

C

B

D

A


x (m)

y (m

)

nG




119866






119866and



119866

cannot exceed 5

0

2

4

6

8


nG

















0

2

4

6

8


nG









4 Conclusion






0

20

40

60

80

100

Are

a with

mor

e tha

n 75

o

f ful

l thr

ough

put (

)





A

A

A

B

B

B

C

C

C D

D

D

A

D

D

B

C

C

B

C

B A

D

A

x

y






0

1

2

3

4


nG










Acknowledgments


References
















RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










0

5

10

15

20

25

30

35

40

45

50

55

60

B C D A

SNR

120590P

120590P (120573 = 0)

SNR (120573 = 0)120590P (120573 = 05)

SNR (120573 = 05)120590P (120573 = 15)

SNR (120573 = 15)








119875) over


119866= 119899minus119899






Low 119899119866


0

1

2

3

4

5

6

7

8


nG






10 20 30 40

5

10

15

20

x (m)

y (m

)

0

05

1

15

2

A B


nG






10 20 30 40

10

20

30

40

0

1

2

3

4

C

B

D

A


x (m)

y (m

)

nG




119866






119866and



119866

cannot exceed 5

0

2

4

6

8


nG

















0

2

4

6

8


nG









4 Conclusion






0

20

40

60

80

100

Are

a with

mor

e tha

n 75

o

f ful

l thr

ough

put (

)





A

A

A

B

B

B

C

C

C D

D

D

A

D

D

B

C

C

B

C

B A

D

A

x

y






0

1

2

3

4


nG










Acknowledgments


References
















RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










10 20 30 40

10

20

30

40

0

1

2

3

4

C

B

D

A


x (m)

y (m

)

nG




119866






119866and



119866

cannot exceed 5

0

2

4

6

8


nG

















0

2

4

6

8


nG









4 Conclusion






0

20

40

60

80

100

Are

a with

mor

e tha

n 75

o

f ful

l thr

ough

put (

)





A

A

A

B

B

B

C

C

C D

D

D

A

D

D

B

C

C

B

C

B A

D

A

x

y






0

1

2

3

4


nG










Acknowledgments


References
















RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










0

2

4

6

8


nG









4 Conclusion






0

20

40

60

80

100

Are

a with

mor

e tha

n 75

o

f ful

l thr

ough

put (

)





A

A

A

B

B

B

C

C

C D

D

D

A

D

D

B

C

C

B

C

B A

D

A

x

y






0

1

2

3

4


nG










Acknowledgments


References
















RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










0

1

2

3

4


nG










Acknowledgments


References
















RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation











RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation









Research Article Interleaved-MIMO DAS for Indoor Radio...

Documents

Transcript of Research Article Interleaved-MIMO DAS for Indoor Radio...