IERI Procedia 4 ( 2013 ) 59 – 67

Available online at www.sciencedirect.com

2212-6678 © 2013 The Authors. Published by Elsevier B.V.Selection and peer review under responsibility of Information Engineering Research Institutedoi: 10.1016/j.ieri.2013.11.010

ScienceDirect

2013 International Conference on Electronic Engineering and Computer Science

Tightly-Coupled Integrated LR-WPAN/WiMAX Network: Architecture and Performance Modelling

Ayyoub Akbari-Moghanjoughia*, Aduwati Salia, José Roberto Amazonasb, Burhanuddin Mohd Alia, Shamala Subramaniamc, Sabira Khatund

aDepartment. of Computer & Communications Systems Engineerin, Faculty of Engineering, Universiti Putra Malaysia. bDepartment of Telecommunications and Control Engineering of Escola Politécnica of the University of São Paulo, Brazil. cDepartment of Communication Technology & Networks, Faculty of Computer Science and IT, Universiti Putra Malaysia.

dSchool of Computer and Communication Engineering, Universiti Malaysia Perlis, Malaysia

Abstract

In this paper a general architecture of tightly-coupled integrated LR-WPAN/WiMAX network has been introduced by proposing and utilising a novel access router named Wireless Integrated Access Router (WIAR). The suggested access router effectively combines both WiMAX and LR-WPAN technologies in a same device. The main motivation to propose this integration is to deploy Internet of Things (IoT) applications and services in wide area networks. Using WiMAX as backhaul service, the Internet connection can be provided for dispersed mobile LR-WPAN users over a wide area. Moreover, it enables to access all collected data in the local LR-WPAN coverage area – which is named as Senspot in this paper – by third-party clients through the Internet. Furthermore, an analytic model has been developed for the integrated network to evaluate the system’s performance. Theoretical results on blocking probability of new and handover calls from WiMAX subscribers, as well as new connections from LR-WPAN Senspots in a cell of WiMAX have been investigated in terms of call/connection rates and number of allocated channels. © 2012 Published by Elsevier B.V. Selection and peer review under responsibility of Information Engineering Research Institute Keywords: Backhaul Support; Heterogeneous Wireless Network; IoT; Networks Integration; WiMAX; LR-WPAN.

* Corresponding author. Tel.: +6-017-271-4682. E-mail address: ayyoub@ieee.org.

Available online at www.sciencedirect.com

© 2013 The Authors. Published by Elsevier B.V.Selection and peer review under responsibility of Information Engineering Research Institute

60 Ayyoub Akbari-Moghanjoughi et al. / IERI Procedia 4 ( 2013 ) 59 – 67

1. Introduction

Heterogeneous wireless networks have received significant attention during the past few years because of the surge of a variety of wireless communications technologies as well as high-level of interest to provide seamless and continuous access to Internet resources. Wireless and mobile communications systems have become increasingly popular as an inexpensive and promising means of ubiquitous communications. Besides the traditional wide area cellular access technology, recent wireless communications access technologies have expanded to include WiMAX [1] and LTE technology [2] in the metropolitan area, as well as WiFi (IEEE 802.11) [3], WPAN and LR-WPAN (IEEE 802.15.4) [4] in the local and personal areas respectively.

In the past few years, much effort has also been devoted to the integration of networks, as for example, WiFi and WiMAX or cellular networks. Shensheng Tang reported a performance modelling for WiFi/WiMAX integrated networks [5]. The horizontal and vertical handoff rates as well as some interesting performance measures have been derived and calculated by introducing an iterative algorithm. Both the analysis and numerical results show that the performance of the 3G network is significantly increased when the integrated 3G/WLAN network is employed, especially when the WLANs are in the hot-spot areas. The method developed has been shown to be useful for the modelling and performance evaluation of other kind of networks, such as the 3G networks with the infrastructure-less-mode WLAN or ad hoc networks[6].

As already mentioned several works have proposed architectures, integration performance and modelling of integrated wireless/cellular networks [7-10]. However, there is no reported research for the integration of LR-WPAN and WiMAX yet.

Low cost communications and inexpensive equipment are, in general, highly preferred but in some cases, depending on the applications, both reliability and accuracy of communication as well as accessibility are extremely important. For instance, a communications system for mobile patient monitoring, regardless of communication costs, requires full accessibility of patient's information. Nowadays, there is no doubt that WPAN is playing key role in health for monitoring and collecting all possible vital information of patients [11]. The short range and low rate transmission of LR-WPAN makes this technology significantly inexpensive to develop monitoring system not only for health monitoring purposes but for environment monitoring as well. Besides all advantages of the LR-WPAN technology, its limitation to be deployed reliably to cover large distance prevents its use to other mobile objects monitoring. Although applying mesh topology using IEEE 802.15.5 might support the ideas concerning large scale real-time communications as in urban areas, the risk of unreliable communications is considerably high and the cost associated to the deployment of such network in the mentioned scale is prohibitive.

Accordingly, in this work a model is proposed for LR-WPAN and WiMAX networks integration to provide seamless accessibility for LR-WPAN users through WiMAX networks as backhaul support. If the LR-WPAN's cluster head is enabled to move from its local gateway along with all connected Reduced and Full Function devices (RFDs/FFDs), the WiMAX network maybe a promising solution to provide backhaul support for LR-WPAN users. To handle the mentioned traffic through WiMAX networks other intermediate nodes have to play a gateway role and transfer all local traffic to WiMAX and vice versa. However, such intermediate device does not exist yet. So, to solve this problem, a mobile Wireless Integrated Access Router (WIAR) is proposed in this research. Therefore, hand-held and wearable sensor devices do not have to include both WiMAX and LR-WPAN interfaces as the WIAR takes care of protocol conversion. All cluster heads are potentially able to forward their collected information using either the WIAR access router proposed in this work or normal WPAN gateways deployed at regular intervals, or to reply to receive queries.

After this brief introduction, the integrated heterogeneous network modelling is described Section 2, including the system description, its modelling and overall layout. The model's performance metrics and

61 Ayyoub Akbari-Moghanjoughi et al. / IERI Procedia 4 ( 2013 ) 59 – 67

evaluation are carried out and discussed in Section 3. Numerical results are presented and explained in Section 4. Conclusions are summarized in Section 5.

2. Integrated Heterogeneous LR-WPAN/WiMAX Network

Depending on the interdependence between the LR-WPAN and the WiMAX cellular network, the coupling between them can be designated as either tight or loose. In the proposed network integration the WIAR plays the key role to provide the connection among these two different wireless technologies. It establishes the connectivity of the networks without using any other intermediate networks and/or the Internet. The tightly-coupled integration used in this work creates a reliable integrated network given that these heterogeneous networks do not depend on the functionality and availability of third party networks and/or services.

2.1. Network architecture and system description

Smart sensor nodes are low power devices equipped with one or more sensors, a processor, memory, a power supply, radio, and, possibly an actuator [12]. The reduced power available prevents sensors of providing long distance transmission. As illustrated in Figure 1, in a simple scenario of a network made of only low-rate sensor nodes, the Packet Success Rate (PSR) becomes close to zero when the distance reaches 100 meters [13]. This result comes from the reduced available power at the transmitter, associated to the power attenuation over distance and, by consequence, an increase of packet losses in not well organised networks. Furthermore, for those WSN applications with large scale mobility, such as health monitoring, there are severe limitations to access the gateway/backhaul to transmit the collected information to IP network station/server or users. In this paper an integrated LR-WPAN/WiMAX heterogeneous network is proposed to solve the mentioned problem by using WiMAX networks as backhaul for relaying mobile sensors data traffic, without deploying a large number of sinks and gateways in the scale of, for example, a city.

As mentioned earlier, the WIAR supports both 802.15.4 and 802.16 radio interfaces. Moreover, protocol

conversion and encapsulation procedures, especially in the MAC layer, are the main functions of the WIAR. The challenging task is the bidirectional MAC header conversion to make the transferred frames supported by

Fig. 1. Packet Success Rate (PSR) vs. Distance in pure LR-WPAN (IEEE 802.15.4)

62 Ayyoub Akbari-Moghanjoughi et al. / IERI Procedia 4 ( 2013 ) 59 – 67

newM

hoW

( )*

H hoW

VW

newW

OWIAR

TBWIAR

hoM

ChWN Ch

WN

ChWN

ChMN

ChMN

a destination MAC addressed device. To overcome the mentioned challenge in header compression and a header-like payload [14] have been considered in this research. As all traffic is transferred via the aforementioned WIAR instead of using the Internet or any other type of cloud resource, this networks integration shall be considered under the tightly-coupled integration category. As it will be seen, proposed LR-WPAN/WiMAX integrated network involves three main parts that are required to be considered in the performance modelling to evaluate the performance of the networks.

2.2. Proposed system modelling

In the performance modelling of the LR-WPAN/WiMAX integration, in addition to considering vertical handover of users from the LR-WPAN to the WiMAX network, it is also possible to have horizontal handover inside the same radio network, of users traversing different cells/Senspots. In the proposed system modelling, we have mainly focused on both the mobile Wireless Integrated Access Routers (WIARs) which are designed using dual radio interfaces to support Wireless Personal Network's nodes and using WiMAX radio for back-haul support purpose (see Fig. 2).

TrafficClassifierScheduling

Local Traffic

Internet Traffic

Protocol

Encapsulation

IEEE 802.16/802.15.4 Bridge(WIAR)

IEEE 802.16Interface

IEEE 802.15.4Interface

Fig. 2. The Single-hop Proposed General Protocol Converter /Bridge between IEEE 802.16 and IEEE 802.15.4 Fig. 2. Tightly-coupled integrated WPAN and WiMAX traffic model

According to Fig. 3, in the considered scenario there are ChMN channels in a WiMAX cell. On the other hand,

ChWN channels are allocated to new calls in a WiMAX cell and Ch

MN represents all connection requests from LR-WPAN networks. v

W designates the vertical handover rate of a LR-WPAN node to a WiMAX base station through a WIAR. Also TB

WIAR represents the rate associated to those users that handover back to their

63 Ayyoub Akbari-Moghanjoughi et al. / IERI Procedia 4 ( 2013 ) 59 – 67

home network once channel is ready to be rejoined. The important notations and symbols used throughout the paper are listed in Table 1.

Table 1. Notations and symbols

Symbol Comments Symbol Comments Ch

MN Number of total channels in a WiMAX cell ho

W Horizontal handover rate among WPAN Senspots

Ch

SNMaximum number of channels allowed for new calls in a WiMAX cell

( )

*

H ho

W

Horizontal handover rate among WPAN Senspots located in different WiMAX cells

Ch

WNMaximum number of channels for all overflowed, H. and V. handover in a WPAN Senspot

new

M New session requests arrival rate of WiMAX clients in a WiMAX cell

TB

WIARTake-back session rate thru a WIAR from a WiMAX cell new

W New session requests arrival rate from a WPAN to a WiMAX cell

O

WIAROverflow requests rate thru a WIAR to a WiMAX cell. v

W Vertical handover rate from WPAN to a WiMAX cell

ho

M

New handover session requests rate of WiMAX clients in a WiMAX cell

The performance modelling has been developed based on a two dimensional Markov process [5]. Based on

the considered parameters the detailed balance equation could be simplified by the global balance equation [15]. For instance, if WiMAX system placed at state (i, j) the steady state probability of a WiMAX system to go from state i to state j is defined as

1( , )i j and given by Eq. (1).

(1)

3. Performance Metrics

In this section, the performance metrics to evaluate the proposed model of LR-WPAN/WiMAX integration are defined and justified. Two performance metrics have been chosen to assess the rate of successful total traffic carried via WIARs and the handover rate to an alternative networks.

3.1. Total carried traffic

In integrated systems the WiMAX cell is responsible for implementing the handover of traffic arriving from a LR-WPAN Senspot. The total carried traffic depends on the total arrival rate ( k

M ) and on the handover connections ( ho

M ) arrival rate. All arrival rates have been calculated using the iterative algorithm introduced in [6]. The total carried traffic denoted by Total

IntegrationCT and can be evaluated by Eq. (3).

(2)

1 11

1( )

1 11

(0,0), 0 ,0 min( , 1)! !

( , )

(0,0), ,0 min( , 1)! !

Ch ChA A

i jk kn mk lM WIAR Ch Ch Ch

A S Mi j

N jho i Nk kn mk lM M WIAR Ch Ch Ch Ch

A M S Mi j

i N j N Ni j

i j

N i N j N Ni j

1

1

1( )

1

1

(0), 0!

( , )

(0),!

ChChAA

i

Mk ChAi

i

Nn ho i NkM Mk Ch Ch

A Mi

i

n k

i Ni

i j

N i Ni

64 Ayyoub Akbari-Moghanjoughi et al. / IERI Procedia 4 ( 2013 ) 59 – 67

(3)

3.2. Session/Call blocking probability

There is a possibility that a new or a handover call or session request be rejected due to the lack of free channels on a base station or access point. One of the main metrics for evaluation of an integrated network is the call blocking probability. Based upon the outcomes of this metrics, the right number of channels can be allocated to different types of users such as new, handover and etc. To compute the call blocking probability of the new and handover users of a WiMAX cell, Eqs. (4) and (5) have been respectively applied.

(4)

(5)

Moreover, session requests blocking probability of LR-WPAN users of a LR-WPAN Senspot, such as

vertical, horizontal, handover and overflowed can be determined by Eq. (6):

(6)

4. Numerical Results and Discussion

In order to investigate the performance of the proposed integrated heterogeneous wireless network, different numbers of channels have been considered for local and handover call requests. The behaviour of WiMAX call arrival rates and LR-WPAN arrival rates variations have been investigated in terms of both total carried traffic and call blocking probability as well.

The initial network architecture configuration has 20 channels as the total number of WiMAX cell channels. The maximum number of channels allowed for new WiMAX calls and all types of connections in a LR-WPAN Senspot are defined as 15 and 10 respectively. The total carried traffic in an integrated architecture is depicted in Fig. 4. By increasing the new connection arrival rate from LR-WPAN clients and the WiMAX call arrival rate, the result shows an increasing carried traffic for the integrated system. The total carried traffic is almost the same for 6 and 8 channels allocated for handover connection requests from the LR-WPAN, while the lowest carried traffic occurs with only two allocated channels for handover connections from a LR-WPAN to a WiMAX base station.

integrated

intgrated 11 1

min , ( , )ChCh

M NNCh Ch Ch Total

S M Integratedi j

N N N i CT i i j

available S

available

1

1 10 0

min , 1

( , ) ( , )Ch ChCh

M

ChA

Ch Ch ChS M

N NNnew Ch

WiMAX Mj ji N

N N N i

P i j N j j

S

10

( , )ChN

handover ChWiMAX M

jP N j j

M S S 1

1 1 10 0

( , ) ( , )Ch Ch ChN N N

all LR WPAN Ch Ch ChLR WPAN M S M

i jP N i N N j j

65 Ayyoub Akbari-Moghanjoughi et al. / IERI Procedia 4 ( 2013 ) 59 – 67

Figs. 5 and 6 are presented the numerical results for call/connection blocking probability for new and

handover calls. We can see that allocating only two channels out of 20 channels of a WiMAX base station for LR-WPAN users, regardless of new or handover calls, the call blocking probability is significantly low in comparison to the scenarios with 6 and 8 allocated channels. This happens due to the high probability of getting a channel by WiMAX local subscribers. Furthermore, in both figures the obtained results via the numerical model are almost the same.

Fig. 7 shows the advantages of allocating more channels for LR-WPAN users in the proposed integrated network. However, it is worth noting that it causes an increase in the call blocking probability for all types of a WiMAX cell’s subscribers both new and/or handover calls as well. There are several techniques for channel allocation, and the trade-off between channel allocation for LR-WPAN connections and WiMAX calls are considered for further research to achieve an optimised solution.

Fig. 5. New WiMAX calls blocking probability as function of new WiMAX call arrival rates

Fig. 3. Total carried traffic in presence of new WiMAX calls and as function of the LR-WPAN arrival rate

66 Ayyoub Akbari-Moghanjoughi et al. / IERI Procedia 4 ( 2013 ) 59 – 67

Fig. 6. The handover call blocking probability as function of different handover call arrival rates

Fig. 7. Blocking probability of LR-WPAN connection in the proposed integrated architecture for different number of allocated channels for LR-WPAN connections

5. Conclusions

In this paper a solution has been proposed for long-distance transmission of LR-WPAN data using the benefits provided by WiMAX networks. Hence, a two-radio interfaces-based intermediate device, named WIAR, has been introduced to play a key role in this integration. Moreover, the presence of WiMAX as a backhaul support, and using the WIAR as a mobile wireless access router make all local information of WPAN accessible two third party users through the Internet. In this architecture all sensors in an identical Senspot are connected to WIARs and during data transmission the WIAR is responsible to perform the

67 Ayyoub Akbari-Moghanjoughi et al. / IERI Procedia 4 ( 2013 ) 59 – 67

necessary modifications on the IEEE 802.15.4 frame and make it compatible to the IEEE 802.16 MAC header. Calls blocking probability and total carried traffic have been considered as two performance metrics in this paper. The mentioned metrics have been investigated through a numerical model and the results have shown to provide further understanding about the proposed LR-WPAN/WiMAX integrated network performance.

References

[1] "IEEE Std 802.16m-2011, IEEE Standard for Local and metropolitan area networks Part 16: Air Interface for Broadband Wireless Access Systems Amendment 3: Advanced Air Interface," ed, 2011, pp. 1-1112.

[2] "3GPP, Overview of 3GPP release 12 v.0.0.5, Tech. Rep., September 2012.," ed, 2012. [3] "IEEE P802.11-REVmb/D12-2011, IEEE Draft Standard for Information Technology - Telecommunications and Information

Exchange Between Systems - Local and Metropolitan Area Networks - Specific Requirements - Part 11: Wireless LAN Medium Access Control (MAC) and Physical Layer (PHY) Specifications," ed, 2011, pp. 1-2910.

[4] "IEEE Std 802.15.4g-2012, IEEE Standard for Local and metropolitan area networks--Part 15.4: Low-Rate Wireless Personal Area Networks (LR-WPANs) Amendment 3: Physical Layer (PHY) Specifications for Low-Data-Rate, Wireless, Smart Metering Utility Networks," ed, 2012, pp. 1-252.

[5] S. Tang, "Performance Modeling of an Integrated Wireless Network Using WiMAX as Backhaul Support for WiFi Traffic," International Journal of Wireless Information Networks, vol. 19, pp. 73-83, 2012/03/01 2012.

[6] S. Tang and W. Li, "Modelling and evaluation of the 3G mobile networks with hot\&\#45;spot WLANs," Int. J. Wire. Mob. Comput., vol. 2, pp. 303-313, 2007.

[7] L. Hui-Tang, L. Ying-You, C. Wang-Rong, and C. Rung-Shiang, "An Integrated WiMAX/WiFi Architecture with QoS Consistency over Broadband Wireless Networks," in Consumer Communications and Networking Conference, 2009. CCNC 2009. 6th IEEE, 2009, pp. 1-7.

[8] D. Niyato and E. Hossain, "WIRELESS BROADBAND ACCESS: WIMAX AND BEYOND - Integration of WiMAX and WiFi: Optimal Pricing for Bandwidth Sharing," Communications Magazine, IEEE, vol. 45, pp. 140-146, 2007.

[9] D. Cavalcanti, D. Agrawal, C. Cordeiro, X. Bin, and A. Kumar, "Issues in integrating cellular networks WLANs, AND MANETs: a futuristic heterogeneous wireless network," Wireless Communications, IEEE, vol. 12, pp. 30-41, 2005.

[10] P. Neves, S. Sargento, and R. L. Aguiar, "Support of Real-Time Services over Integrated 802.16 Metropolitan and Local Area Networks," in Computers and Communications, 2006. ISCC '06. Proceedings. 11th IEEE Symposium on, 2006, pp. 15-22.

[11] H. Alemdar and C. Ersoy, "Wireless sensor networks for healthcare: A survey," Computer Networks, vol. 54, pp. 2688-2710, 2010. [12] J. Yick, B. Mukherjee, and D. Ghosal, "Wireless sensor network survey," Computer Networks, vol. 52, pp. 2292-2330, 2008. [13] "NS-3 Project, ns-3 Model Library Release ns-3-dev," 31 January 2012. [14] A. A. Moghanjoughi, S. Khatun, B. M. Ali, and R. S. A. R. Abdullah, "QoS based Fair Load-Balancing: Paradigm to IANRA

Routing Algorithm for Wireless Networks (WNs)," in Computer and Information Technology, 2008. ICCIT 2008. 11th International Conference on, 2008, pp. 104-109.

[15] P. G. Harrison and N. M. Patel, Performance Modelling of Communication Networks and Computer Architectures (International Computer S: Addison-Wesley Longman Publishing Co., Inc., 1992.