IJCSIS Paper 30041153

Report as inappropriate Your report has been received

Description

Radio Access Technology (RAT) selection, Joint Radio Resource Management (JRRM), Heterogeneous Wireless Networks.

Shared by: ijcsiseditor

Tags

J. Preethi, Assistant Professor, Department of Computer Science and Engineering, Anna University of Technology, Coimbatore, India

Stats

views:: 79
posted:: 11/26/2011
language:: English
pages:: 6

Public Domain

Document Sample

(IJCSIS) International Journal of Computer Science and Information Security, Vol. 9, No. 5, May 2011

An Overview on Radio Access Technology (RAT)
Selection Algorithms for Heterogeneous Wireless
Networks
J.Preethi Dr.S.Palaniswami
Assistant Professor, Professor,
Department of Computer Science and Engineering Department of Electrical and Electronics Engineering
Anna University of Technology, Coimbatore Government College of Technology
India India
Abstract—Next generation wireless networks (NGWN) will accurately as possible into its original wave form by the base
be heterogeneous in nature where different radio access station.
technology coexist in the same coverage area. The The second generation was implemented to improve
coexistence of different RATs require a need for Joint transmission quantity system capacity and network coverage.
Radio Resource Management (JRRM) to support the In second generation systems (e.g., Personal communication
provision of quality of service and efficient utilization of systems (PCS), Global System for Mobile Communication
radio resources. The Joint Radio Resource Management (GSM), Code Division Multiple Access (CDMA), Time
(JRRM) manages dynamically the allocation and Division Multiple Access (TDMA) and Orthogonal Frequency
deallocation of radio resources between different Radio Division Multiplexing (OFDM)) are based on Digital
Access Technology (RAT). The homogenous Call Transmission.
Admission Control (CAC) algorithms do not provide a In digital systems, more efficient use of the available
single solution to address the heterogeneous architectures spectrum is achieved by digital encoding of the speech data.
which characterize next generation wireless networks. This Due to the transition from 2G to 3G, a number of standards
limitation of homogeneous CAC algorithms necessitates have been developed, which are categorized as 2.5G. These
the development of RAT selection algorithms for are add-ons to the 2G standards and mainly focus on
heterogeneous wireless network. The goal is to select the deployment of efficient IP connectivity within the mobile
most suitable RAT for each user. This paper investigates networks.
ten different approaches for selecting the most appropriate 2.5G is a stepping stone between 2G and 3G cellular
Radio Access Technology (RAT) for incoming calls among wireless technologies, invented for marketing purposes only.
the Heterogeneous Wireless Networks. This RAT selection 2.5G implements a packet switched domain which includes
works in two steps; the first step is to select a suitable GPRS (General Packet Radio Service), EDGE (Enhanced Data
combination of cells among the different RATs. The rates for GSM).
second step chooses the most appropriate RAT to which The objective of the third generation (3G) is to provide
the users can be attached and to choose the suitable fairly high speed wireless communications to support
bandwidth to allocate for the users. multimedia, data and video in addition to voice. 3G includes
Universal Mobile Telecommunications systems (UMTS) [1],
CDMA2000 based on W-CDMA technologies which provides
Keywords- Radio Access Technology (RAT) selection, Joint Radio services like wireless access to the Internet and high data rate
Resource Management (JRRM), Heterogeneous Wireless Networks. applications like real time video transmission. To cope with
these, high bandwidth services and the enormous increase in
I. INTRODUCTION the number of users, a more efficient use of radio spectrum is
In the recent years, mobile communication systems have required [2].
experienced a significant change leading to the introduction of In turn, the perspective of beyond 3G systems is that of
the heterogeneous wireless networks. The first generation heterogeneous networks, which provides wireless services
mobile communications systems (e.g. Nordic Mobile independently of its location in a completely transparent way
Telephony (NMT) and Advanced Mobile Phone System [3]. The user terminal should be able to pick the best access
(AMPS)) are based on analog transmission techniques. Analog technology such as Wireless Local Area Network (WLAN),
signals are radio transmissions sent in a wave-like form. A the Universal Mobile Telecommunication Systems (UMTS)
mobile device sends the waves to a base station where they are and the Global System for Mobile Telecommunication
processed to determine the signals next destination (i.e. (GSM)/Enhanced Data rate for GSM Evolution (EDGE) Radio
another base station, mobile phone, land line phone etc.,) Once Access Network (GERAN) at its current location and use the
the destination is determined, the signal is reconstructed as access technology seamlessly for the provision of desired
service. This leads to the introduction of new Radio Resource

1
(IJCSIS) International Journal of Computer Science and Information Security, Vol. 9, No. 5, May 2011

Management (RRM) techniques referred to as (JRRM) Joint III. BENEFITS OF HETEROGENEOUS JOINT RADIO
Radio Resource Management algorithms which manages RESOURCE MANAGEMENT ALGORITHMS
dynamically the allocation and deallocation of radio resources Each Radio Access Network (RAN) differs from the others
between different Radio Access Technology (RAT). That is, by the air interface technology, cell size, services supported,
instead of performing the management of radio resources bit rate capabilities, coverage, mobility support etc., therefore
independently for each RAT, some form of overall and global the heterogeneous characteristics offered by the network is
management of the pool of radio resources can be envisaged. considered. As a result, RAT provide further flexibility in the
The coexistence of different RATs require a need for Joint way how radio resources can be managed. This lead to the
Radio Resource Management (JRRM) to support the provision introduction of RRM. The basic function of Call Admission
of quality of service and efficient utilization of radio Control (CAC) algorithm is to decide whether a new handoff
resources. In heterogeneous wireless networks, different RATs call can be accepted into a RAT without violating service
coexist in the same coverage area. The goal is to select the commitments [5]. CAC has been used in wired networks and
most suitable RAT for each user. In this paper, a in homogenous wireless networks such as GSM, UMTS,
comprehensive survey of different RAT selection algorithms WLAN, Satellite network etc., However, homogenous CAC
for a heterogeneous wireless network is proposed. Section II algorithms do not provide a single solution to address the
explains about the architecture of heterogeneous wireless heterogeneous architecture. This limitation of homogenous
network. Section III presents the benefits of Joint Radio CAC algorithm necessitates the development of RAT selection
Resource Management algorithms and Section IV presents the algorithm for heterogeneous wireless network.
comprehensive survey of RAT selection algorithms Section V Joint Call Admission Control algorithm is one of the JRRM
gives the proposed work and lastly the conclusions are algorithms. Within the JRRM, the initial RAT selection, i.e the
presented in Section VI. allocation of connections to specific RANs at session initiation
II. HETEROGENEOUS WIRELESS NETWORK BEYOND 3G and the Vertical Handover (VHO) i.e the capability to switch
ongoing connections from one RAN to another. These are the
Next generation networks will be heterogeneous where key enablers to properly manage the heterogeneous radio
different radio access networks such as WLAN, UMTS, access network and become the key for the JRRM functions.
WiMax and satellite networks which is illustrated in the The benefits of Heterogeneous Joint Radio Resource
Figure 1. In order to provide the mobile users with the Management Algorithms are
requested multimedia services and corresponding quality of  Efficient utilization of radio resources,
service (QoS) requirements[4], these radio access technologies  Consistent provisioning of QoS across different
will be integrated to form a heterogeneous wireless access RATs,
network. Such a network will consist of a number of wireless
 Overall stability of network,
network and will form the fourth generation (4G) or next
 Increase in Operator‟s revenue and
generation of wireless networks. However, each access
network provides different levels of QoS, in terms of  Enhancement of users satisfaction.
bandwidth, mobility, coverage area and cost to the mobile
users.

Internet

RRM

WLAN
Node
AP Node
WLAN
WIMAX

Cellular / GSM/ GPRS/
UMTS/3G/B3G

User Equipment /Mobile
Terminal

Fig.1. Integration of Heterogeneous Wireless Access Network

2
(IJCSIS) International Journal of Computer Science and Information Security, Vol. 9, No. 5, May 2011

IV. AN OVERVIEW ON RAT SELECTION ALGORITHMS namely WLAN and WWAN. Four Fuzzy logic controllers are
The section describes different RAT selection algorithms for used separately for selecting the best access network and
initial RAT selection and Vertical Handover. References Particle swarm optimization is used to optimize the fuzzy
[6],[7] presents the good review of RAT selection algorithms. logic controllers and Genetic algorithm is used to optimize the
In Random based RAT selection, when a new call or vertical weight coefficients.
handover arrives, any of the available RATs will be selected
randomly. In Load balancing based RAT selection, the main D) Soft Computing techniques based RAT Selection
objective is to uniformly distribute the load among all the
available RATs in heterogeneous wireless network Soft computing techniques based RAT selection admits an
[8],[9],[10]. In Policy based RAT selection, it allocates users incoming call based on applying soft computing techniques for
to the RAT based on some specific rules specified by the the RAT selection. The soft computing techniques applied for
network. A simple policy has been proposed in [11], which RAT selection are discussed
includes Voice GERAN (VG) policy, Voice UTRAN (UV)  Fuzzy logic [17], focuses on the issues related to
policy and Indoor (IN) policy. Service-class based RAT mobility management in future mobile
admits calls into a particular RAT based on class of service communication scenario where a multi segment
such as voice, video streaming, real-time video, web browsing access network is integrated into an IP core network.
etc., [12]. Service cost based RAT selection admits incoming [17] Proposes a new approach to handover
call into the least expensive RAT in order to reduce the service management by applying a fuzzy logic concept to a
cost incurred by the users. Path loss based RAT selection heterogeneous environment. For handover initiation,
algorithm makes call admission algorithm based on path-loss parameters considered are network coverage,
measurements taken in the cells of each RAT[13]. perceived QoS and Signal Strength (SS).
 The framework for JRRM algorithm [18-19] based on
Some of the other RAT selection algorithms are discussed in fuzzy neural mechanism as explained in Fig.2
this section. consists of three main blocks namely fuzzy based
decision, reinforcement learning and multiple
A) Network layer based RAT selection algorithm objective decision making. The inputs for the fuzzy
based decision block are signal strength of each RAT,
Network layer based RAT selection algorithm admits calls resource availability of each RAT and mobile speed
into a particular layer. If the layer cannot accommodate the of the user. The Fuzzy based decision consists of
call, this algorithm tries to admit the call into the next layer. three parts namely fuzzifier, inference engine and
These algorithms are very simple but can lead to highly defuzzifier. The fuzzifier allocates a value from 0 to 1
unbalanced network load. Network layer based RAT selection for each input. In the inference engine, for each of the
algorithm is explained in [14]. The objective of this algorithm fuzzy subset defined in the fuzzifer, fuzzy rules are
is to minimize new call blocking probability while associated to indicate if it is suitable to be selected.
guaranteeing a hard constraint on handoff call dropping The output of the inference engine is a value that
probability. varies between Y(yes), N (no), PY(probably yes) and
PN (probably no). The defuzzifer will convert the
B) Utility/cost function based RAT selection output of the inference engine into fuzzy selected
decision (FSD). The subjective and techno-economic
Utility/Cost function based RAT selection admits the criteria in the form of user preferences (UP) and
incoming calls into a particular RAT based on some utility or operator preferences (OP) are inputs to the multiple
cost function derived from a number of criteria. It considers objective decision making. The outputs of the fuzzy
user time constraints, and estimates the complete file delivery neural algorithm are cell/RAT selection and amount
time for each available access network and then selects the of bit rate allocated for the selected RAT.
most promising access network. These algorithms are very  Fuzzy MADM (Multiple Attribute Decision Making)
efficient but are very complex and incur high computational method [20-22] operates in two steps. The first step is
overhead. [15] present the utility based RAT selection to convert the imprecise fuzzy variables to crisp
algorithms for selecting the RAT. numbers. The second step is to use classical MADM
technique to determine the ranking order of the
C) Mobile based RAT selection algorithm candidate networks. The highest ranking RAT is then
selected for the call.
This algorithm uses mobile terminal measurements from  Using Fuzzy logic controllers, genetic algorithms and
different radio access technology for the initial RAT selection particle swarm optimization for decision making of
[16]. The inputs to this algorithm are speed of the mobile user, radio access technology selection for the next
signal strength, quality of service and service cost. This generation wireless networks under given input
algorithm uses fuzzy logic controllers, genetic algorithms and criteria on user velocity, type of service and service
particle swarm optimization for decision making under given parameters, Quality of service and service costs of the
input criteria. In this, two access networks are considered mobile user [23].
3
(IJCSIS) International Journal of Computer Science and Information Security, Vol. 9, No. 5, May 2011

REINFORCEMENT LEARNING

Bandwidth

Signal strength MULTIPLE Allocated
INFERENCE Fuzzy
Resource FUZZIFIER DEFUZZIFIER OBJECTIVE Bandwidth
Selected
availability ENGINE DECISION
RAT
MAKING
Mobile speed RAT
User Selected
Fuzzy based Decision Preferences

Operators
Preferences
Fig.2. Block Diagram of the Fuzzy Neural System
(Figure taken from [18])

This algorithm uses mobile terminal measurements G) Network Controlled Cell Breathing RAT selection
from different radio access technologies within a
given time interval, with aim to obtain information The Network controlled cell breathing RAT
for multi criteria decision making between different selection algorithm is applied for the Initial RAT
access networks available to the terminal. selection and Vertical handover between
CDMA/TDMA based systems [26]. In
E) Hopfield Neural network RAT selection FDMA/TDMA systems like GSM/GPRS, there is no
Intra cell interference, but there will be inter cell
Here the RAT selection mechanism is based on the interference for a single user exchanging the
Hopfield neural networks model [24]. The input information between the cells. In case of CDMA
parameters to the HRM model are gathered from based systems like UMTS, there are both Intra cell
different layers of the protocol stack. This approach and Inter cell interferences. CDMA systems are more
allows improvement in terms of service retainability, sensitive for multi user interference than
decreased packet loss, jitter. The main advantage of FDMA/TDMA systems. By effectively controlling
HRM is smaller complexity. The smaller complexity the cell radius of the CDMA RAT, the capacity of the
results in smaller CPU load. CDMA systems is increased and target coverage area
is assured.
F) Operator RAT selection based on the Fittingness
Factor H) Performance Analysis of Radio Access Technology
Selection Algorithms using 2-Dimensional Markov
The initial RAT selection and vertical handover is Model
done based on measuring the fittingness factor
depending on network and terminal capabilities for The performance analysis of Radio Access
each RAT [25]. The fittingness factor is calculated by Technology selection is made using Markov model
three factors namely the capabilities of the terminal, [27]. To evaluate the performances metrics, Mobility
User-centric suitability and network centric suitability based algorithm and Always WLAN algorithm are
of the overall RAT perspective. For Initial RAT considered. Authors Porjazoski and Borislav
selection, fittingness factor is calculated for each Popovski considered the RAT selection algorithms
RAT. The RAT with the highest fittingness factor is performances, like user mobility, call arrival rate, call
selected first and if admission is not possible, select duration, size of RAT cells. The results explained that
the next RAT in the decreasing order with respect to Always WLAN algorithm have higher rate of vertical
fittingness factor. For Vertical handover, the handover than Mobility based algorithm. Mobility
fittingness factor of the serving RAT and the new based RAT selection algorithm shows better
RAT are measured and the RAT with the highest performances in the means of blocking probability for
fittingness factor is given the call. higher percentage of vehicular users. For smaller
percentage of vehicular users and for very high traffic
loads, Always WLAN algorithm performs better then
mobility based algorithm.

4
(IJCSIS) International Journal of Computer Science and Information Security, Vol. 9, No. 5, May 2011

V PROPOSED WORK [12] W.Zhang, “Performan ce of Real-time and Data traffic in Heterogeneous
Overlay Wireless Networks”, 19th International Teletraffic Congress
Applying optimizing methodologies like Fuzzy logic, (ITC 19), Beijing, China, August 29- September 2, 2005.
Genetic algorithm, Neural Networks, Particle Swarm [13] J.Perez-Romero, O.Sallent, R.Agusti, “Enhanced Radio Access
optimization and Simulated Annealing for the RAT selection Technology Selection exploiting Path loss information”, 17th annual
IEEE Symposium on Personal, Indoor and Mobile Radio
of the heterogeneous wireless network and comparing these Communication (PIMRC’06).
algorithms and finding out the best access network. [14] R.B.Ali, S.Pierre, “An Efficient predictive admission control policy for
heterogeneous wireless bandwidth allocation in next generation mobile
networks”, International Conference on Communications and Mobile
VI CONCLUSION computing (IWCMC’06), Vancouver, Canada, July3-6, 2006.
[15] O.Ormond, J.Murphy, G.Muntean, “Utility –based Intelligent network
In heterogeneous wireless networks, different RATs coexist in selection in beyond 3G systems”, IEEE International Conference on
the same coverage area. The goal is to select the most suitable communications (ICC 2006), Istanbul, Turkey, June 11-15.
[16] Aleksandar Tudzarov and Toni Janevski, “Efficient Radio Access
RAT for each user. The coexistence of different RATs
Technology Selection for the Next Generation Wireless Networks” in
requires a need for Joint Radio Resource Management International Journal of Research and Reviews in Next Generation
(JRRM) to support the provision of quality of service and Networks, Vol. 1, No. 1, March 2011
efficient utilization of radio resources. Hence this paper [17] P.M.L. Chan, R.E.Sheriff, Y.F.Hu, P.Conforto, C.Tocci, “Mobility
presents the architecture for heterogeneous wireless networks management incorporating fuzzy logic for a heterogeneous IP
environement”, IEEE Communications Magazine 39(12) (2001) 42-51.
and benefits for Joint radio resource management algorithms. [18] L. Giupponi, R. Agustí, J. Pérez-Romero, and O. Sallent, “A novel
Then an overview of Radio Access Technology selection for approach for joint radio resource management based on fuzzy neural
the Heterogeneous wireless networks is discussed. The future methodology,” in IEEE Transactions on Vehicular Technology, vol.57,
work in this area is to determine the best access technology No.3, May 2008.
[19] R. Agustí, O. Sallent, J. Pérez-Romero, and L. Giupponi, “A
among the available RATs by giving priority levels among the fuzzyneural based approach for joint radio resource management in a
different classes of calls namely new calls, horizontal handoff beyond 3G framework,” in Proc. 1st Int. Conf. Quality Service
calls and vertical handoff calls in heterogeneous wireless Heterogeneous Wired/Wireless Netw., Dallas, TX, Oct. 2004, pp.
networks. 216–224.
[20] W.Zhang, „Handover decision using fuzzy MADM in heterogeneous
networks”, Proceedings of IEEE WCNC’04, Atlanta, GA, March 2004.
REFERENCES [21] A.L.Wilson, A.Lenghan, R.Malyan, “Optimizing wireless network
selection to maintain QoS in heterogeneous wireless environments”,
[1] Radio Resource Management Strategies in UMTS, J. Pérez-Romero, O.
Proceeding of World Communication Forum, Denmark, September,
Sallent, R. Agustí, andM. A. Díaz-Guerra, Eds. Hoboken, NJ:Wiley,
2005.
2005.
[22] O.E.Falowo and H.a.Anthony, “Fuzzy logic based call admission control
[2] V.K.Garg, Wireless Communications and Networking, Morgan
for next generation wireless networks”, Proceedings of 3rd International
Kaufmann Publishers, Sans Francisco, CA, USA 2007.
symposium on wireless Communication Systems, Valencia, Spain,
[3] Marques, R. L. Aguiar, C. Garcia, J. I. Moreno, C. Beaujean, E. Melin,
September 5-8, 2006.
and M. Liebsch, “An IP-based QoS architecture for 4G operator
[23] Aleksandar Tudzarov, Toni Janevski, “Efficient Radio Access
scenarios,” IEEE Trans. Wireless Commun., vol. 10, no. 3, pp. 54–62,
Technology Selection for the Next Generation Wireless Networks”,
Jun. 2003
International Journal for Research and Reviews in Next Generation
[4] J. Perez-Romero, O.Sallent,R.Agusti and M.A.diax-Guerra, Radio
Networks, Vol.1, No.1, March 2011.
Resource Management strategies in UMTS, John Wiley & sons Ltd.
[24] Valentin Rakovic, Liljana Gavrilovska, “ Novel RAT selection
Chichester, UK 2005.
mechanism based on Hopfield Neural Networks” , Internaltional
[5] D.Karabudakm C.Hung,B.Bing, “Call Admission control scheme using
congress on Ultra Modern Telecommunications and Control Systems
genetic algorithms”, Symposium on applied computing, SAC‟04,
and Workshops (ICUMT), 2010.
Nicosia, Cyprus, March 14-17, 2004.
[25] J Perez-Romero, R.Ljung, P.Karlsson, A.Barbaresi, “ Operator‟ s RAT
[6] Abdallah AL Sabbagh, Robin Braun, Mehran Abolhasan, “A
selection policies based on the Fittingness Factor Concept”, 16th IST
Comprehensive Survey on RAT Selection Algorithms for
Mobile and Wireless Communications Summit, Budapest, Hungary, July
Heterogeneous Neworks”, World Academy of Science, Engineering and
1-5,2007, pp.1-5.
Technology 73 2011.
[26] J.Perez-Romero, O.Sallent, R.Agusti, L.Wang, H.Aghavmi, “ Network
[7] O.E.Falowo and H.A.Chana, “Joint Call Admission control algorithms:
Controlled cell-breathing for capacity improvement in heterogeneous
Requirements, approaches and design considerations”, Computer
CDMA/TDMA scenarios”- IEEE Wireless Communications and
Communications, vol.31, no.6, April 2008, pp.1200-1217.
Networking conference (WCNC’06), April 3-6, Las Vegas, NV, USA.
[8] A.Tolli and P.Hakalin, “Adaptive Load Balancing between Multiple Cell
[27] Marko Porjazoski1 and Borislav Popovski1 “Performance Analysis of
layers”, 2002 IEEE 56th Vehicular Technology Conference (VTC 2002)
Radio Access Technology Selection Algorithms in Heterogeneous
vol3., Vancouver, Canada, September 24-28, 2002, pp.1691-1695.
Wireless Networks using 2-Dimensional Markov Model” in the
[9] K.H.Suleiman, H.A.Chan, and M.E.Dlodlo, “Load Balancing in the Call
International Journal of Research and Reviews in Wireless
Admission Control of Heterogeneous Wireless Networks”, International
Communications, Vol. 1, No. 1, March 2011
Conference on Communications and Mobile computing (IWCMC 2006),
Vancouver, British Columbia, Canada, July 3-6, 2006, pp.245-250.
[10] A.Umbert, L.Budzisz, N.Vucevic and F.Bernardo, “An all-IP
heterogeneous wireless testbed for RAT selection and e2e QoS
evaluation”, The 2007 International Conference on Next Generation
Mobile Applications, Services and Technologies (NGMAST 2007),
Cardiff, Wales, UK, September 12-14, 2007, pp.310-315.
[11] J.Perez-Romero, O.Sallent and R.Agusti, “Policy-based Initial RAT
selection algorithms in HeterogeneousNetworks”, The 7th IFIP
International Conference on Mobile and Wireless Communications
Networks (MWCN 2005), Marrakech, Morocco, September 19-21, 2005.

5
(IJCSIS) International Journal of Computer Science and Information Security, Vol. 9, No. 5, May 2011

AUTHORS PROFILE

S. Palaniswami received the B.E. degree in
electrical and electronics engineering from
the Govt., college of Technology,
Coimbatore, University of Madras, Madras,
India, in 1981, the M.E. degree in electronics
and communication engineering (Applied
Electronics) from the Govt., college of
Technology, Bharathiar University,
Coimbatore, India, in 1986 and the Ph.D.
degree in electrical engineering from the
PSG Technology, Bharathiar University,
Coimbatore, India, in 2003.
Coimbatore, India, in 2003. He was the Registrar of Anna University
Coimbatore, Coimbatore, India, from May 2007 to May 2010. Currently he is
heading the Department of Electrical and Electronics Engineering, His
research interests include Control systems, Communication and Networks,
Fuzzy logic and Networks, AI, Sensor Networks. . He has about 25 years of
teaching experience, since 1982. He has served as lecturer, Associate
Professor, Professor, Registrar and the life Member of ISTE, India.

J.Preethi received the B.E. degree in Computer Science and Engineering
from Sri Ramakrishna Engineering College, Coimbatore, Anna University,
Chennai, India, in 2003, the M.E. degree in Computer Science and
Engineering from the Govt. college of Technology, Anna University, Chennai,
India, in 2007 and she is currently pursuing the part time Ph.D. degree in the
Department of Computer Science and Engineering from the Anna University
Coimbatore, Coimbatore, India. Currently, she works as a Assistant Professor
in the Department of Computer Science and Engineering, Anna University
Coimbatore. Her research interests include Mobile adhoc networks, Mobile
Communication systems especially in Radio Access Technology selection,
Fuzzy logic and Neural Networks, Genetic Algorithms and AI.