International Review on Computers and Software (I.RE.CO.S.), Vol. 7, n. 6

Handoff Management in Cognitive Radio Networks: Concepts, protocols, metrics and challenges Samad Nejatian1, Sharifah Kamillah Syed-Yusof 1, Nurul Muazzah Abdul Latiff 1, Nursheila Fisal1, Vahid Asadpour 2

Abstract – In Cognitive Radio (CR) networks, Secondary Users (SUs) use the unused spectrum bands. Once the band of the spectrum, which is occupied by the SU, is claimed by a Primary user (PU), the SU transfers its data transmission into another empty band of the spectrum. Changing the operation frequency is called spectrum handoff. During the spectrum handoff, the SUs transfer their ongoing calls to an unused spectrum band upon the presence of the PU. In this paper, we investigate a brief overview of spectrum mobility and handoff management in cognitive radio. The main concepts, protocols, and tools are also proposed in this literature. We also introduce the concept of integrated mobility and handoff management. In addition, we illustrate the most challenging and open issues of spectrum mobility and handoff management in CR networks.

Keywords: cognitive radio, mobility function, Spectrum Handoff, spectrum Mobility

TABLE I LIST OF ALL THE USED SYMBOLS AND THEIR MEANING

Analytical model variable

Meaning Primary User Secondary User node transmission range total number of available channel number of possible channels at each node channel type total number of channel type number of possible channels of each type at each node probability of a particular channel availability at each node poisson density of nodes’ spatial distribution in the network transmission range of a channel of type l

PU SU RT C c l L cl p

 Rl

I.

Introduction

During the previous decades, wireless communications have been at the forehand of the research areas. Hence, major changes and significant improvements are distinct in the wireless communication systems. These major changes create a heterogeneous and intensively varying radio environment, which forces an accurate and intelligent management system for allocating the scarce radio spectrum resources. The most important solution, which promises to handle this requirement, is cognitive radio.

Manuscript received April2012, revised October 2012

Cognitive radio is an adaptive and intelligent radio that is aware of its operation environment [1]. With reference to Federal Communication Commission (FCC), the current poor spectrum efficiency is because of the fixed spectrum assignment. The fixed spectrum assignment policy leads to inefficient spectrum utilization in which, most of the time, the assigned spectrum is underutilized by the licensed users. Hence, the dynamic spectrum access is proposed to increase the spectrum efficiency by using the spectrum opportunistically [2]. Therefore, the cognitive radio users can capture these unused spectrum bands in an opportunistic manner. CR users are like visitors to the spectrum; hence, in a case that the spectrum mobility occurs, and the specific licensed band occupied by the SU is reclaimed by a PU, the SU’s ongoing call needs to continue on another unused band of the spectrum. Mobility function guarantees a fast evacuation with minimum performance degradation in CR network. It must be equipped with algorithms in order to determine the candidate spectrum based on the channel characteristics and spectrum availability. Spectrum mobility leads to a new handoff, which is known as spectrum handoff. In this paper, we survey the proposals and works found in several literatures concerning spectrum mobility and handoff management in CR networks. In addition, we emphasize the integrated mobility and handoff management while highlighting those factors and types of spectrum mobility, which necessitate comprehensive mobility and handoff management in CR networks. The main challenges concerning the integrated mobility and handoff management with their solutions are also outlined. Copyright © 2007 Praise Worthy Prize S.r.l. - All rights reserved

S. Nejatian, et. Al. The paper is organized as follows. In section II and section III, we have the general overview on cognitive radios, and basic functionalities of mobility function as well as, spectrum handoff in these networks, respectively. In section IV and section V, the mobility procedure and its main functionalities based on the network architecture, infrastructure-based and ad hoc networks are described. The handoff decision phase, different handoff schemes, handoff execution phase and the various steps involved with previously done related work are also introduced. In Section VI, the notion of integrated mobility and handoff management is investigated. Spectrum handoff characterization and its metrics are introduced in section VII. Section VIII investigates some challenges and open research issues for the development of efficient spectrum handoff in cognitive radio networks. Finally, Section IX concludes the paper.

II.

A review on cognitive radio

According to Mitolla, cognitive radio is a radio, which can react to the new conditions and learns from its activities [3]. The limitations on the spectrum as well as, deficient usage of recourses have resulted to implementation of dynamic spectrum access (DSA) or Next generation (xG) networks as a new communication pattern [4]. This new pattern has emerged for utilizing the idle bands and solving the spectrum scarcity challenge. As mentioned above, the first step in the cognitive cycle is spectrum sensing. Spectrum sensing is the most important function of a cognitive radio. Once, an unused spectrum band has been detected, cognitive radio can use this spectrum band in an opportunistic manner. It can capture the best available channel between detected channels. Cognitive radio can also share these spectrum bands between its users and users of other networks. The following described the main functions of cognitive radio: Spectrum Sensing: detecting the spectrum holes (out-ofband sensing), and monitoring the spectrum bands during transmission (in- band sensing). Spectrum management: capturing the best available channel. Spectrum Sharing: enabling spectrum access between different users in a coordination manner. Spectrum mobility: evacuating the licensed channel by the entrance of a primary user and supporting the requirement during this evacuation. Since almost the entire spectrum bands are already licensed, sharing these bands without any interference is the most challenging issue. Meanwhile, spectrum sharing and spectrum management cooperate with each other in order to increase the spectrum efficiency. There are also some cooperative interactions such as routing, medium access, and adaptation between spectrum management and spectrum mobility. Once the occupied band is reclaimed by the licensed user, CR user transforms its current communication to another unoccupied band or stay in the current spectrum by changing its modulation type or power level in order to avoid interference with licensed user.

Copyright © 2007 Praise Worthy Prize S.r.l. - All rights reserved

III. Mobility function and spectrum handoff in CR networks Mobility is one of the most critical functions of cognitive radio because of its effect on network properties such as channel capacity, routing, connectivity, and coverage [5]. The existence of available spectrum holes is random because of randomness in PU’s appearance and accidental user mobility. Hence, it can be stated that the spectrum holes are constantly shifting over time. This is because of the PU’s appearance as well as over space due to the user mobility. This shifting of spectrum holes and changing of spectrum band is known as spectrum mobility. As mentioned before, the spectrum mobility leads to a new handoff, which is known as spectrum handoff. The spectrum handoff occurs when the status of the current channel does not satisfy the quality of service or a licensed user appears. As illustrated in Fig. 1, the function consists of two parts, which are spectrum mobility management and handoff management. Handoff decision making information such as current occupied spectrum bands, suitable spectrum band, and special handoff scheme are obtained from spectrum mobility management and other cognitive functions. In order to perform various aspects of mobility function, multi layer mobility management protocols are needed [4].

Fig. 1. Mobility function

Since CR devices are expected to use a wide range of spectrum bands, a numerous set of frequency-separated spectrum bands might be sensed. These sets of available spectrum bands may be heterogeneous in terms of data rate and delay characteristics [6]. Consequently, channel switching may lead to significant change in network capacity. On the other hand, since the period and frequency of spectrum handoff is not exactly predictable, spectrum handoff may be miscounted as congestion condition. Spectrum handoff also affects the end-to-end network performance [7]. III.1. Spectrum handoff procedure Spectrum handoff process must be performed based on intelligent algorithms in order to determine the candidate spectrum based on the channel characteristics and spectrum availability. According to literature, the normal spectrum handoff process can be divided into three steps (Fig. 2). These three steps are as follows: PU detection and preparation time: In this step, PU reclaims its right to capture the licensed channel which has been occupied by the SU. CR users sense (in-band sensing) the spectrum bands by individual or cooperative sensing to detect the PU’s presence. Then, SU will prepare to perform spectrum handoff. Next, SU

International Review on Computers and Software, Vol.7, n. 6

S. Nejatian, et. Al. notifies its pair about interruption event and stops its communication within a predefined time. Spectrum scanning and handoff decision: Upon the first out-of-band sensing cycle, SU senses the spectrum bands in the pool, and determines the proper spectrum band. Coordination, reconfiguration and spectrum handoff execution: In this phase, SU reconfigures its RF frontend [9]. Then, it synchronizes the transmission schedule on the spectrum band, moves to new spectrum, and resumes its transmission. The SU may experience more than one interruption during its transmission. These interruptions lead to sets of consecutive spectrum handoffs. The above cycle is repeated, and a set of target channels will be determined consequently until the end of SU’s transmission.

Fig. 2. Spectrum Handoff procedure in CR

Achieving a fast and smooth spectrum handoff is extremely difficult, Due to the accidental appearance of the primary user. Hence, SU’s performance degradation will be happen [10]. The created delay for sensing, decisionmaking, handshaking, and channel switching leads to unwanted gaps in transmission. The total created handoff delay for performing one spectrum handoff can be expressed as follows [9]: sen ( out  of  band ) D  d prep  d syn  d sen ( out of band ) tx  d dec  d recfg  d syn

The first element of the above equation is related to preparation time. The three-second items are related to the delay resulted from the second step, and the two latest items are related to the third step in spectrum handoff procedure.

Copyright © 2007 Praise Worthy Prize S.r.l. - All rights reserved

In case of consecutive spectrum handoffs, the SU’s data delivery time will be increased. III.2. Different schemes of spectrum handoff Based on the instance of spectrum handoff and channel allocation, spectrum handoff schemes can be divided into two schemes [12]: Reactive spectrum handoff: In this scheme, SU performs out-of-band spectrum sensing and the first step of the handoff process followed by PU detection [13, 14]. In this scheme, the SU determines the target channels on demand in order to perform spectrum handoff, which leads to unwanted delay and performance degradation in SU’s transmission. Proactive spectrum handoff: In this scheme, SU predicts the future spectrum availability based on longterm observed channel utilization statistics, and it intelligently schedules the channel usage before PU appearance. Once the PU appears and interrupts the SU, SU can change its spectrum band immediately to a predetermined band. By proactive spectrum handoff, the collision between SUs and PUs will be reduced significantly [10]. This scenario leads to a faster spectrum band switching, but it requires more complicated algorithms. Although the proactive handoff is more efficient than reactive sensing in terms of delay, the energy constraint is a serious challenge in this scheme. In the event of PU detection, proactive spectrum handoff is more efficient while, in terms of user mobility and channel quality degradation, which immediate band switching is not necessary, the reactive spectrum handoff is more suitable [24].

IV.

Mobility function and spectrum handoff in infrastructure-based CR networks

In infrastructure-based networks, when the base station detects the spectrum mobility in the time domain or space domain, CR user evacuates the current spectrum band and captures a new frequency band (Fig. 3). During the handoff procedure and switching time, base station reduces the effects on the performance of upper layer protocols and maintains the level of needed quality service for user application by management function [8].

International Review on Computers and Software, Vol.7, n. 6

S. Nejatian, et. Al.

Fig. 3. Functional block diagram for spectrum mobility in infrastructure-based CR networks

In the following, we briefly describe the proposals belonging to spectrum handoff in infrastructure-based networks. Some papers investigated the proactive and reactive schemes, and modeled them in order to analyze their effects on the SU’s performance. The authors of [11, 12] have investigated a Markov transmission model integrating preemptive resume priority M/G/1 queuing network model in order to model reactive and proactive spectrum handoff. The proposed model in [11] is used in order to characterize and model the reactive spectrum handoff as well as the delay, which resulted from multiple spectrum handoffs in reactive scheme. The effect of sensing, handshaking, switching and waiting time on the multiple handoff delay is investigated in this paper. The analytical results of this paper are used in order to design the admission control policies for meeting the latency needs of the SU. The proposed analytical framework in [12] is used to compare and optimize the two major spectrum handoff scenarios under different conditions. The interrupted user’s traffic load coming from other channels is considered in a double channel switching circumstance. The authors analyzed the situation either the proactive or reactive handoff be used based on sensing time. The interaction between channels is also characterized by considering the traffic load of interrupted users who switch to another channel. In [15], a preemptive resume priority M/G/1, queuing model is also introduced in order to analyze the total service time of SU for proactive-decision spectrum handoff in CR networks. Based on this analytical model, the authors discussed the pre-determined target channel selection by proactive spectrum handoff in order to minimize the total service time with multiple spectrum handoffs. In this paper, the total service time is considered as the period from the packet transmission starting instant until the end of transmission with multiple spectrum handoff. Since the predetermined target channel might no longer available, this paper suggests the determination of optimal target channel

Copyright © 2007 Praise Worthy Prize S.r.l. - All rights reserved

sequences to minimize total service time. A sub-optimal greedy algorithm is suggested for target channel selection in order to reduce the complexity of the target channel selection greedy algorithm, which increases by the number of channel. Based on the proposed analytical model, it was shown that the sub-optimal algorithms can decrease the total service time with lower complexity compare to a randomly selection method. In [16], a proactive spectrum access algorithm is proposed in order to predict the future spectrum availability based on past channel history. In this paper, two channel determination schemes and switching scheme are proposed. The proposed proactive algorithm decreases the handoff latency and disruptions to PU by a downturn searching order of the channel idleness probability under various channel characteristics. The authors of [17] investigated the voluntary spectrum handoff, which is based on PU’s activity and considers the potential channel idle durations as well as the expected transmission time for starting the unforced spectrum handoff. The voluntary spectrum handoff estimates the PU’s activity and decreases the temporary disruption time in SU’s communication. In order to cope with time varying nature of available spectrum bands, the PU’s spectrum usage is calculated by defining two sensing period, which are the sensing window and history window. During the sensing window, the spectrum utilization is monitored, the SU’s activity is registered in a spectrum server, and this information is used for deducing the past PU’s activity when it is needed. The SUs calculate the PU’s activity by using the information of the on time sensing window as well as spectrum server information. In order to determine the voluntary spectrum handoff instance, the spectrum handoff time namely the residual spectrum lifetime is estimated for the current and target channels by two different spectrum selection algorithms, and the best channel is captured while the voluntary spectrum handoff is started before PU’s appearance. Fig. 4 illustrates the proposed method in [17].

International Review on Computers and Software, Vol.7, n. 6

S. Nejatian, et. Al.

Fig. 4. Architecture of proposed voluntary spectrum handoff in [17]

In [18], an analytical model of spectrum handoff as well as a basic model for licensed spectrum access in CR is investigated based on Markov chain analysis. Based on this model the probability of forced disconnection, blocking probability, and system throughput are derived. In order to decrease the probability of forced disconnection for CR transmission, a channel selection scenario is also proposed. The introduced channel selection scheme, which is based on a channel reservation, creates a tradeoff between forced disconnection and blocking according to the quality of service needed. The results have shown that the throughput of CR users significantly increases by the channel reservation method. The authors of [19] have discussed a number of errors in the Markov model and analysis of its parameters introduced in [18]. The correct expression of mentioned parameters is investigated, and authors claimed that there are significant differences between their simulation results and the results of [18] both in terms of magnitude and behavior. In [20], a scheme based on task allocation model in the ant colony is introduced for dynamic spectrum allocation. This scheme helps CR user to capture the best available channel based on the PU’s idle time. In this paper, the best available channel is defined as the channel that can facilitate low handoff rate and high link maintenance probability. The simulation results show that the proposed method reduces the handoff rate as well as the handoff latency by capturing the best available channel and in simple operation, which need no coordination with other CR users. The study [21] has proposed a handoff scheme based on the prediction model in which the channels are selected considering the low occupancy by PUs. The available spectrum bands are noncontiguous and will be distributed over an expanded range of both licensed and unlicensed spectrum band [4]. Hence, in order to perform successful spectrum handoff, the reliable and distributed spectrum bands scanning and sensing is an essential need for CR users [22]. Almost all papers in the literature perform spectrum handoff by the assumption that all the channels in the spectrum band can be properly

Copyright © 2007 Praise Worthy Prize S.r.l. - All rights reserved

sensed and scanned by the CR user. According to [23], the full scanning and sensing of spectrum bands is not practicable in realistic conditions because of excessive time requirement. In order to estimate the network condition by partially spectrum sensing, the authors of [23] use the partially observable Markov decision process. Based on this scheme, a partially observable Markov decision processbased spectrum handoff model is also proposed in order to determine the best available channel as the next target channel. Determination of the target channel based on partial observation leads to minimum waiting time at each spectrum handoff. Fig. 5 shows the basic concept of this paper. In [24], an adaptive channel reservation method is proposed, which intelligently scans the wide spectrum band by two effective circular-bidirectional and dual sensing methods. A pointer is used in this scheme in order to show the reserved channels as well as to facilitate the faster and smoother spectrum handoff. After PU detection, SU must notify its pair of the interruption events. It is also necessary to synchronize the transmission schedule on target channel before transmitting a packet between transmitter and receiver. Using a spectrum band without coordination with other SUs can cause link establishment failure. Therefore, the network coordination has a significant effect on the performance of the SUs. SUs then have to find a common channel and establish a link [25]. In [13, 17 and 26], the network rendezvous issue is solved by using a common control channel. In these papers, regulatory policy, channel determination and neighbor sensing message are considered to be sent in common control channels. In [27], the coordination scheme and signaling mechanism, before and during transmission is introduced, which includes initial link establishment and dynamic link maintenance. The authors introduce the communication system based on the interleaved OFDM as a promising option for initial link establishment.

International Review on Computers and Software, Vol.7, n. 6

S. Nejatian, et. Al.

Fig. 5. Partially observable Markov decision process-based spectrum handoff

The authors in [28] proposed an in-band signaling mechanism named ESCAPE. The paper presents the basic design of the mechanism in physical, link, and network layer. The main objective of this scheme is fast and reliable spectrum band evacuation in multiple cognitive radios. Evacuation time as well as failure probability and false alarm are used as main performance metrics from the perspective of PU and SU, respectively. Hence, the length of spreading code, message repetition time, and the threshold for warning message, considering the performance constraints is the main parameters for protocol designing.

V.

Mobility function and spectrum Handoff in CR Ad hoc networks

As previously mentioned, a fast and smooth spectrum handoff is extremely challenging in cognitive radio because of the randomness PUs’ presence. In ad hoc networks, there

is no central entity or spectrum broker to arrange and control the spectrum handoff. Therefore, the spectrum handoff in ad hoc networks is more challenging [10]. According to [8], spectrum handoff in ad hoc networks occurs when the PU is detected; SU’s connection is missed due to its mobility, or when channel quality is degraded. Detection of any of these events by spectrum sensing, neighbor discovery, or routing protocol leads to spectrum mobility and handoff initiation. There is an inherent dependency between spectrum handoff and routing in CR ad hoc networks. Spectrum handoff needs to interact with routing protocols in order to find the link failure (Fig. 9). On the other hand, the constructed route at the network layer should not have any effect on PU’s activities. Once the PU is detected, the protocol should either switch the affected channel, or recover the whole route. In case, there is no available channel for the next hop during the spectrum handoff, rerouting must be performed [49]. Therefore, for ad hoc networks, in terms of handoff delay the route recovery time must be added to the previously mentioned spectrum handoff delay. As illustrated in Fig. 6, routing protocols must be also considered in Ad hoc network mobility function in order to recover the link damage on end-to-end route. Managing the mobility events in Ad hoc network cannot be performed as efficiently as the infrastructure-based networks because of the inexistence of central entity, the time and location varying spectrum availability as well as more complicated topology.

Fig. 6. Functional block diagram for spectrum mobility in Ad hoc networks

VI.

Integrated Mobility and Handoff Management

Since, the available frequency bands shows different characteristics in CR networks, current radio conditions and PU activities must be considered for spectrum band characterization. In addition, a dynamic decision scheme is also needed, which considers the spectrum sensing and channel characteristics for maximizing the efficient CR transmission [35]. Available spectrum bands vary over time

Copyright © 2007 Praise Worthy Prize S.r.l. - All rights reserved

as well as with SU movement. In order to show the effect of these different events on the cognitive radio system and handoff decision making, the following analytical model is proposed. Table. 1 shows the different parameters used in this analytical model. In order to have a comprehensive handoff management, the effects of the SU’s mobility and channel transmission range on the probability of successful routing must be considered. In channel heterogeneity with different channel transmission ranges, the distance between SUs must be also considered. Suppose that there is the total

International Review on Computers and Software, Vol.7, n. 6

S. Nejatian, et. Al. number of C channels in a heterogeneous network, which are classified into L types according to their different transmission ranges. Different channels occupy different spectrum bands. The number of available channels of each type at each node is cl in which l={1,2,…,L}. It means that c=c1+c2+…+cL. The transmission range of channels of type l is Rl. Considering this channel classification; we can model the channel quality degradation. Two SUs can use a channel of type l for communication when their distance is less than Rl. When the SUs are moving and their distance become longer than Rl, they must change their channel, and choose a channel with a transmission range longer than Rl. In case that their distance becomes shorter than Rl, they have the more chances to capture higher number of channels for communicating. We illustrate the p as the probability of availability of a special channel between two nodes. The Pcat is also defined as the probability of single channel availability between two nodes, which is equal to: 2

Pcat  p .



From (1), we have:

The probability of single channel availability at a node p, is equal to Poff. So we have:

p

 

Substituting (4) in (6), we have: i 1

cj  2 c  Pcat ,c   P ( Ri )(1  (1  ( ) ) j 0 ) (7)   i 1 in which c0  0 L

In case that the PU’s activity on different channels is not same, for example, on channel k we have:

Pon, k 

k  k  k

Poff 

&

k  k  k

(8)

In this case, the (4) changes to the (9) as follows: i 1

c L

Pcat ,c   P ( Ri )(1  i 1

Pcat  1  p 2 .

(6)

cj j 0

 k 1

(1  (

k )2 ))  k  k

(9)

The node transmission range is RT. We suppose that R1<…..
i 1

In which the P(Ri) defines as the probability of communicating of two nodes with expected hop length less than Ri and longer than Ri-1 with R0=0. In (4), p is the probability of single channel availability at a node. This parameter shows the PU’s activity. We considered the p as a constant parameter, but we can consider different traffic activity model for PUs. By considering an alternating renewal two state birth-death process with the death rate  and birth rate β. Since the arrival time of each user is independent, the transmission of each user follows the Poisson arrival process. It means that the length of off and on periods have an exponential distribution with mean 1 /  and 1/  . The probability of PU activity can be estimated as follows:

Pon 

 

&

Poff 

 

(5)

Copyright © 2007 Praise Worthy Prize S.r.l. - All rights reserved

forward direction in the disk with radius RT is equal to e and we have:

P

finding a node in disk withradius RT

 1  exp ( N ) 2



N 2

,

(12

Considering the random variable d denotes the distance between a pair of transmitter and receiver nodes and the fd (Ri) represents its probability density function (pdf). The probability distribution function of d is as following:

Fd ( Ri )  Pr (d  Ri ) 

1  exp( R 2i / 2)

1 e in which 0  Ri  RT



N 2

(13)

We have:

Pr (a  X  b)  FX (b)  FX (a)

(14)

So we can conclude that:

International Review on Computers and Software, Vol.7, n. 6

S. Nejatian, et. Al.

Pr( Ri 1  d  Ri )  Fd ( Ri )  Fd ( Ri 1 ) 2



1  exp( R i / 2)



1 e 1 e 2 exp( R i 1 / 2)  exp( R 2i / 2)



N  2



1  exp( R

2 i 1

/ 2)

N  2

(15)

N

1 e 2 0  Ri 1  Ri  RT & R0  0

in which

Finally, we have: L

Pcat ,c  

exp( R 2i1 / 2)  exp( R 2i / 2)

i 1

1 e



N 2

The (16) shows the effect of different factors on the probability of channel availability, which is the main factor and basic element in cognitive radio networks. Based on (16), the PU’s activity, SU’s mobility, and channel heterogeneity must be considered in spectrum handoff and mobility management protocols. Hence, an integrated mobility function which considers the spectrum mobility in both time and space domain, must be investigated. In other words, for designing a mobility management protocol, the spectrum mobility, user mobility, and channel quality degradation must be considered jointly [9]. Fig. 7 shows the integrated mobility and handoff management concept and its functionalities.

i1

c

cj j 0

(1 



(1  (

k 1

k )2 ))  k  k

(16)

Fig. 7. Functional block diagram for integrated mobility and handoff management

Two main problems related to integrated mobility and handoff management are [9]: Dynamic spectrum availability: Spectrum availability changes over time and space because of PU activities and SU’s movement. Broad range of the available spectrum bands: The available spectrum bands are expanded over a large frequency range, and they are not adjacent. In [35], a spectrum handoff scheme is developed in which the optimal spectrum band is chosen based on the combination of estimated transmission time, the PU presence probability and spectrum availability time. A cooperative spectrum sensing method is used in order to predict the spectrum idleness. In this paper, the authors use a geo-location method in order to consider the handoff in the space domain. The simulation results indicate that the proposed spectrum handoff outperforms conventional methods in terms of handoff delay and transmission

Copyright © 2007 Praise Worthy Prize S.r.l. - All rights reserved

efficiency. In [36], a spectrum heterogeneity-based spectrum sharing scheme is investigated in which the SUs are free to move (Fig. 8). Prediction of the spectrum utilization time, which is the main metrics for spectrum sharing, is also modeled based on the SU’s mobility and channel transmission range. Channel allocation is done based on a usage threshold time in order to avoid consecutive channel handoff resulting from the short spectrum utilization time. The simulation results show that both PU’s activities and SU’s mobility are crucial in consecutive channel handoff and link available time. In this paper, the joint routing and spectrum allocation is also investigated in order to keep the link between SU and base station when the spectrum handoff occurs, and there is no extra link for communication in a single hop scheme. In [37], different mobility prediction schemes are evaluated based on their accuracy. Then, the mobility prediction is used in order to improve the bandwidth efficiency and cognitive radio scalability.

International Review on Computers and Software, Vol.7, n. 6

S. Nejatian, et. Al.

Fig. 8. Proposed spectrum heterogeneity-based spectrum sharing scheme in [36]

Using the spectrum pooling notion is extremely common in cognitive radio networks. The general idea of spectrum

pooling is based on gathering spectrum bands belonging to various owners together, and creating a public pool [38]. In [9], the authors investigate an integrated spectrum aware mobility management based on spectrum pooling in order to eliminate the heterogeneous spectrum availability in CR cellular networks. This integrated scheme covers the spectrum mobility and user mobility management. In this paper, the spectrum usage and stochastic connectivity model are used in order to determine appropriate spectrum handoff type and target cell. In terms of user mobility management, a handoff decision making based on switching cost mechanism is also used to minimize service quality degradation. Fig. 9 shows the main concepts of this paper.

Fig. 9. Proposed integrated spectrum aware mobility management based on the spectrum pooling in [9]

VII. CR spectrum handoff; characterization and metrics

VII.1. Number of spectrum handoff and total handoff delay

In order to characterize spectrum handoff, the main metrics are categorized into two groups [14]. The first group namely short-term performance metrics consists of link maintenance probability and switching delay. The second group, which consists of the number of spectrum handoff and SU connection termination probability, is known as long-term performance metrics. As Fig. 10 shows, the short-term metrics influence the long-term metrics. In this part, an overview of these metrics and their related works in literature is provided.

Since the SU’s transmission might last for a period of time, the sequential spectrum handoff is inevitable, and it may happen more than one time before a SU ends its transmission. Consecutive spectrum handoff leads to system performance degradation because of the out-ofband sensing and handoff overhead as well as longer total handoff delay. Handoff delay is the most important factor in the determination of mobility performance. Hence, it is necessary to increase SU’s transmission efficiency and minimize performance degradation by avoiding the consecutive spectrum handoffs, reducing the switching delay and consequently the total handoff delay. In order to overcome consecutive spectrum handoff and handoff latency problem, authors have proposed different methods in the literature. In [20], the authors reduce the spectrum handoff rate as well as handoff latency by using channel allocation algorithm based on task allocation model in the ant colony. In [23], the spectrum switching delay is reduced by using a spectrum handoff scheme based on partially sensing of frequency bands. In [15], the total service time is minimized by using a low-complexity greedy algorithm for target channel selection. In [39], the authors have

Fig. 10. Spectrum handoff performance metrics

Copyright © 2007 Praise Worthy Prize S.r.l. - All rights reserved

International Review on Computers and Software, Vol.7, n. 6

S. Nejatian, et. Al.

proposed a novel power control-based spectrum handoff in order to improve the transmission efficiency by reducing the number of spectrum handoff. In this scheme, SU does not switch to an idle spectrum band upon PU’s appearance unless its interference to PU exceeds a predetermined threshold. Upon PU arrival, the maximum transmission power, in which there is no interference for PU is calculated by SU. If the SU can connect to its receiver, the transmission performs; else the SU transfers its connection to another channel. In [31], the number of spectrum handoff is decreased by using the licensed band as operating band and unlicensed band as back up channels in the spectrum handoff event. In [35], the number of spectrum handoffs is reduced using the predicted channel usage time. In [40], the effect of spectrum mobility and total service time on the number of spectrum handoff is investigated. The study [41] proposed a tradeoff between whether waiting on the current channel until the end of PU’s transmission or performing spectrum handoff. In this new spectrum handoff strategy, handoff decision making is performed based on a cumulative probability which estimates the quality of the current channel. In this scenario, a backup channel is used for short time periods in which the PU behavior cannot be predicted. The simulation results show a significant decrease in the number of spectrum handoff compared to random and classic channel selection approaches. The authors of [42] have proposed a spectrum handoff scenario in the multicell system. This scenario is based on the cell outage estimation, which reduces the probability of cell outage occurrence, as well as the number of redundant spectrum handoffs. In this paper a novel handover scheme from low frequency to high frequency, which leads to decrease in cell coverage due to higher path loss in high frequency, is proposed to avoid cell outage. On the other hand, in order to avoid redundant intercell handoffs in case of handover from high to low frequency, another handoff scheme is also proposed. In the first scheme, an intercell handoff is performed before frequency changing to avoid cell outage that may result from coverage reduction. In the second scheme, the coverage is expanded by handoff from high frequency to low frequency in order to suspend the intercell handover of a user in the edge of a cell. The results imply on the reduced number of handoffs and outage probability. VII.2. Link maintenance and SU’s communication termination The link maintenance in CR spectrum handoff refers to the restructure of the SU’s link after PU’s appearance, and link maintenance probability is defined as the probability of successful link maintenance when the SU occupies the channel. Unsuccessful link maintenance leads to the termination of SU’s connection. The first paper in terms of link maintenance in cognitive radio is Copyright © 2007 Praise Worthy Prize S.r.l. - All rights reserved

[13]. In this paper, the authors investigate a general model for link maintenance in CR. A spectrum determination scheme is proposed in which only one sub channel from each spectrum band is assigned to each SU. According to this paper the lower the spectrum resources utilized for the communication, the lower the probability for the link maintenance and spectrum handoff in CR network. On the other hand, to maintain the link without spectrum band switching, the channel redundancy scheme is also used. The authors proved that redundancy solution does not necessarily work in CR networks. In [43], the restructure of the SU’s link is done by changing the modulation type and data corruption compensates by retransmission of the last data. The authors of [44] proposed the using of the backup channel for link maintenance. In this paper, the communication bands are monitored, and the available channel list is updated dynamically. Then, one or more of the idle channels are determined as the backup for supporting the current communication. The authors of [45] introduce an analytical model to study the performance of three various handoff scenarios in terms of link maintenance. They compare the non spectrum handoff scheme, spectrum handoff based on sensing, and spectrum handoff based on the predetermined channel list. The number of handoff trials for a desired link maintenance probability and effective data rate in each scheme is also calculated. The results show that the spectrum handoff based on sensing outperforms two other schemes in the case of high PU’s traffic. The authors of [35] show that the SU’s termination communication is dramatically decreased using channel usage time estimation. The performance metrics such as communication termination, as well as handover collision probabilities, are evaluated, and a handover control algorithm also introduced in [46]. In this paper, the channel bandwidth, which relies on network heterogeneity, is used as QoS parameter. The results of using this handover algorithm, which adaptively schedules the channel sensing based on using spectrum pooling notion, indicate that the network performance significantly improves, and communication is very stable in both synchrony and asynchrony networks. The authors of [47] used a combination of underlay transmission and multi cell spectrum handoff in order to overcome the finite transmission range in underlay transmission system. The multi cell scheme is used as complementary for the underlay transmission coverage. The optimum target cell is selected in order to avoid communication termination when the SU trespass the underlay limit. The results imply that the proposed method and algorithm reduces the link failure probability and communication termination while the QoS improves.

International Review on Computers and Software, Vol.7, n. 6

S. Nejatian, et. Al.

VIII. Challenges and open research issues In this part, the challenges and open research issue in CR spectrum handoff are introduced. VIII.1. Broad range of the dynamic available spectrum As mentioned previously, available spectrum bands vary over time as well as duo to SU movement. These bands show different characteristics in CR networks. Hence, a dynamic decision scheme is needed, which considers the spectrum sensing and time varying channel characteristics for maximizing the efficient CR transmission. In other words, for an effective spectrum handoff and mobility management, the dynamic spectrum availability must be deeply considered [48]. VIII.2. Optimization of in-band and out-of-band sensing Interference avoidance is an important issue in CR networks. Although longer spectrum sensing time during the CR transmission leads to less interference with PUs, the transmission time of CR user is decreased. Therefore, the issue of in-band sensing as well as it’s period is a challenging issue related to spectrum sensing, which must be more considered in order to have an accurate and efficient spectrum handoff. On the other hand, although the longer the out-of-band sensing, the more the available spectrum bands, it must be considered that the handoff delay will be decreased. Hence, finding an accurate and fast spectrum sensing is an important challenge, which reduces the spectrum handoff latency. VIII.3. Collision free and fair channel selection Channel selection is yet an open issue and more challenging in CR networks, especially in spectrum handoff. Channel allocation algorithm must consider the spectrum handoff delay and total service time without any collision while considering fairness for all SUs. Collision means that multiple SUs determine same channel for spectrum switching at the same time, which leads to data corruption and consequently, performance degradation [50].

can remain connected to its next hop forwarder by controlling its power, or it must switch to another spectrum band once it enters a new geographical rejoin while considering the spectrum mobility. The network connectivity must be guaranteed while moving, and during the spectrum handoff procedure. In this regard, using the accurate mobility prediction methods can be useful and efficient. Despite the significant effect of this issue in spectrum handoff, consideration is no more focused on the space domain fluctuating nature of spectrum bands. VIII.5. Joint routing and handoff in CR Ad hoc networks Optimal routs maintaining in CR Ad hoc networks is challenging because of the fluctuating nature of PU’s activity as well as CR node’s mobility. When a PU is detected, the routing protocol must make a decision on whether to switch the affected channel or perform rerouting. Therefore, an optimal joint routing and spectrum handoff protocol that adapts to the dynamic spectrum availability and node mobility considering end to end connectivity is necessary and challenging. VIII.6. Channel coordination and rendezvous As mentioned previously, channel usage without coordination and rendezvous between SU failures the link establishment. On the other hand, once a SU detects the PU, SU must inform its neighbors from this event in order to evacuate the spectrum to avoid interference. Hence, network coordination and rendezvous is essential for performance improvement. Almost all papers in the literature either do not consider the network coordination or establish it by using common control channel. For handoff in CR systems, establishing a common control channel is difficult because of the fluctuating nature of spectrum availability in time and space domain as well as the effect of PU traffic on common control channel. In brief, for improving the CR network performance, channel coordination must be further investigated.

In Ad hoc networks, the performance is dramatically related to node mobility. Spectrum handoff in the space domain happens once the SU changes its physical position. Maintaining the effective connectivity within SUs considering node mobility and spectrum handoff in the space domain is essential in order to provide quality of service requirements. For an example, a CR user in Ad hoc networks must determine whether it

VIII.7. Choosing the proper spectrum handoff scheme in an adaptable handoff framework Considering the vast development in wireless systems, CR, which is an emerging technology, can be considered as an important solution of scared resources. Although CR has been at the forehand of wireless research and development during the last decade, it has opened issues and research challenges that must be solved. Mobility is one of the most critical functions in cognitive radio because of its effect on network properties such as channel capacity, routing, connectivity and coverage. Despite the negative effects of the spectrum handoff and mobility on SU’s performance, it has been studied not as much as other

Copyright © 2007 Praise Worthy Prize S.r.l. - All rights reserved

International Review on Computers and Software, Vol.7, n. 6

VIII.4. Node's mobility considering and space domain handoff

S. Nejatian, et. Al.

functions and aspects of CR networks. For mobility management, in order to meet the QoS requirements, choosing the decision algorithm is also challenging despite the existence of different optimization schemes based on learning such as neural network and genetic algorithm. Novel solutions are also essential in order to ensure that the channel rendezvous and network coordination improves the CR network performance. We proposed an analytical model, which shows the effect of different factors on the handoff decision. Therefore, an integrated and comprehensive framework that considers all of these factors is an essential and unexplored necessity for handoff in CR networks.

IX.

Conclusion

Considering the vast development in wireless systems, CR, which is an emerging technology, can be considered as an important solution of scared resources. Although CR has been at the forehand of wireless research and development during the last decade, it has open issues and research challenges that must be solved. Mobility is one of the most important functions in cognitive radio because of its effect on network properties such as channel capacity, routing, connectivity, , and coverage. Despite the negative effects of the spectrum handoff and mobility on SU’s performance, it has been studied not as much as other functions and aspects of CR networks. For mobility management, In order to meet the QoS requirements, despite the existence of different optimization schemes based on learning such as neural network and genetic algorithm, choosing the decision algorithm is also challenging.

[5]

Alexander, W. Min, Kang, G. Shin, Impact of mobility on spectrum sensing in cognitive radio networks. Proceedings of the 2009 ACM workshop on Cognitive radio networks (Page: 13 Year of Publication: 2009).

[6]

Henrik, A., Olof, H., Ian, M., TCP over high speed variable capacity links: A simulation study for bandwidth allocation. Proceedings of the PIHSN (Page: 117-129 Year of Publication: 2002). Kaushik, R. Chowdhury, Marco, D. Felice, Ian, F. Akyildiz, TPCRAHN: A Transport Protocol for Mobile Cognitive Radio Ad hoc Networks. Proceedings of the INFOCOM, Rio de Janeiro. (Page: 2482 Year of Publication: 2009). W. Y. Lee, SPECTRUM MANAGEMENT IN COGNITIVE RADIO WIRELESS NETWORKS. Ph.D. Thesis, Georgia Institute of Technology, 2009. W. Y. Lee, S. Member, I. F. Akyildiz, Spectrum-Aware Mobility Management in Cognitive Radio Cellular Networks. IEEE Transactions on Mobile Computing, pp.1-14, 2011. Y. Song & J. Xie, ProSpect: A Proactive Spectrum Handoff Framework for Cognitive Radio Ad hoc Networks without Common Control Channel. IEEE Transactions on Mobile Computing, 2011. Wang, C. Wei, Wang, L. Chun, Adachi, F., Modeling and analysis for reactive-decision spectrum handoff in cognitive radio networks. Proceedings of the IEEE GlobeCom. (Page: 1 Year of Publication: 2010). Wang, L. Chun, Wang, C. Wei, Spectrum Handoff for Cognitive Radio Networks: Reactive-Sensing or Proactive-Sensins?, Proceedings of the Performance Computing and Communications International Conference, IEEE. (Page: 343 Year of Publication: 2008). Willkomm, D.; Gross, J.; Wolisz, A., Reliable link maintenance in cognitive radio systems. Proceedings of the first IEEE international symposium on DySPAN(Page: 371 Year of Publication: 2005). Zhang, Y., Spectrum Handoff in Cognitive Radio Networks: Opportunistic and Negotiated Situations. Proceedings of the IEEE International Conference on Communications (Page: 1 Year of Publication: 2009). Wang, C. Wang, Wang L. Chung, Modeling and Analysis for Proactive-Decision Spectrum Handoff in Cognitive Radio Networks, Proceedings of the 2009 International Conference on Communications (Page: 1 Year of Publication: 2009). L. Yang, L. Cao, H. Zheng, Proactive channel access in dynamic spectrum networks. Physical Communication, Vol. 1, n. 2, pp. 103111, 2008. Yoon, S.-un, Ekici, E., Voluntary Spectrum Handoff: A Novel Approach to Spectrum Management in CRNs, Proceedings of the IEEE international conference on communication (Page: 1 Year of Publication: 2010). X. Zhu, L. Shen, T.-shing. Yum, Analysis of Cognitive Radio Spectrum Access with Optimal Channel Reservation. IEEE Communications Letters, Vol. 11, n. 4, pp. 304-306, 2007. W. Ahmed, J. Gao, H. Suraweera, M. Faulkner, Comments on “analysis of cognitive radio spectrum access with optimal channel reservation.” IEEE Transactions on Wireless Communications, Vol. 8, n. 9, pp. 4488-4491, 2009. Huang, W., Chen, J., Li, S., A Channel Allocation Algorithm for Minimizing Handoff Rate in Cognitive Radio Networks. Proceedings of the 4th International Conference on Wireless Communications Networking and Mobile Computing (Page:1 Year of Publication: 2008) Hoyhtya, M., Pollin, S., Mammela, A., Classification-based predictive channel selection for cognitive radios, Procedings of IEEE International Conference on Communications (Page:1 Year of Publication:2010). Qiao, X., Tan, Z., Li, J., combined optimization of spectrum handoff and spectrum sensing for cognitive radio systems, Proceedings of 7th IEEE international conference on wireless communications, networking, and mobile computing (Page: 1 Year of Publication: 2011). Ma, R. Ting, Hsu, Y. Pin, Feng, K. Ten, A POMDP-Based Spectrum Handoff Protocol for Partially Observable Cognitive Radio Networks. Procedings of IEEE Wireless Communications and Networking Conference (Page: 1 Year of Publication: 2009).

[7]

[8]

[9]

[10]

[11]

[12]

[13]

[14]

[15]

[16]

[17]

Acknowledgements [18]

This work was supported in part by the Ministry of Higher Education (MOHE) Malaysia, and Research Management Center (RMC), University Technology Malaysia under GUP research grant no: Q.J130000.7123.01H99.

[19]

[20]

References [1]

[2]

[3] [4]

J. Mitola, G. Q. Maguire, Cognitive radio: making software radios more personal. Ieee Personal Communications, Vol. 6, n. 4, pp. 1318, 1999. J. Mitola, Cognitive Radio: An Integrated Agent Architecture for Software Defined Radio Dissertation. Scientific American, Vol. 294, n. 3, pp. 66-73, 2000. J. Mitola, Cognitive Radio Architecture Evolution. Proceedings of the IEEE, Vol. 97, n. 4, pp. 626-641, 2009. I. Akyildiz, W. Lee, M. Vuran, S. Mohanty, NeXt generation/dynamic spectrum access/cognitive radio wireless networks: A survey. Computer Networks, Vol. 50, n. 13, pp. 21272159, 2006. Elsevier.

Copyright © 2007 Praise Worthy Prize S.r.l. - All rights reserved

[21]

[22]

[23]

International Review on Computers and Software, Vol.7, n. 6

S. Nejatian, et. Al.

[24] I. F. Akyildiz, W.-Y. Lee & K. R. Chowdhury, CRAHNs: Cognitive radio Ad hoc networks. Ad hoc Networks, Vol. 7, n. 5, pp. 810-836, 2009. [25] Song, Y., Xie, J., Common Hopping based Proactive Spectrum Handoff in Cognitive Radio Ad Hoc Networks. Proceedings of IEEE global telecommunications conference (Page: 1 Year of Publication: 2010). [26] Subramani, S., Armour, S., Kaleshi, D., Zhong, F., Spectrum scanning and reserve channel methods for link maintenance in cognitive radio systems. Proceedings of VTC Spring IEEE Vehicular Technology Conference (Page: 1944 Institute of Year of Publication: 2008). [27] Han, C, Wang, J, Li, S., A Spectrum Exchange Mechanism in Cognitive Radio Contexts. Proceedings of 17th IEEE International Symposium on Personal Indoor and Mobile Radio Communications (Page: 1 Year of Publication: 2006). [28] Liu, X., Ding, Z., ESCAPE: A Channel Evacuation Protocol for Spectrum-Agile Networks. Proceedings of 2nd IEEE International Symposium on New Frontiers in Dynamic Spectrum Access Networks (Page: 292 Year of Publication: 2007). [29] Feng, W., Cao, J., Zhang, C., Liu, C., Joint Optimization of Spectrum Handoff Scheduling and Routing in Multi-hop Multiradio Cognitive Networks. Proceedings of 29th IEEE International Conference on Distributed Computing Systems (Page: 85 Year of Publication: 2009). [30] G. Ko, A. Franklin, S.-jin You, J.-suk . Pak, M.-sun. Song, C.-joo. Kim, Channel management in IEEE 802.22 WRAN systems. IEEE Communications Magazine, Vol. 48, n. 9, pp. 88-94, 2010. [31] M. Kalil, F. Liers, T. Volkert, A. Mitschele, A Novel Opportunistic Spectrum Sharing Scheme for Cognitive Ad hoc Networks. ECEASST, Vol. 17, 2009. [32] M. A. Kalil, H. Al-mahdi, A. Mitschele-thiel, Spectrum Handoff Reduction for Cognitive Radio Ad hoc Networks. Channels, pp. 1036-1040, 2010. [33] Giupponi, L., Perez, I., Fuzzy-based Spectrum Handoff in Cognitive Radio Networks. Proceedings of 3rd International Conference on Cognitive Radio Oriented Wireless Networks and Communications CrownCom (Page: 1 Year of Publication: 2008). [34] Y. Song, J. Xie, Performance Analysis of Spectrum Handoff for Cognitive Radio Ad hoc Networks without Common Control Channel under Homogeneous Primary Traffic. Compare A Journal Of Comparative Education, pp. 3011-3019, 2011. [35] J. Duan, Y. LI, An optimal spectrum handoff scheme for cognitive radio mobile Ad hoc networks. Advances in electrical and computer enginnering, Vol. 11pp. 11-16, 2011. [36] Ning, G., Cao, X., Duan, J., Chowdhury, K. R., A Spectrum Sharing Algorithm Based on Spectrum Heterogeneity for Centralized Cognitive Radio Networks, Proceedings of 73rd IEEE conferences on vehicular technology (Page: 1 Year of Publication: 2011). [37] Butun, I., Talay, C., Altilar, T., Khalid, M., Sankar, R., Impact of mobility prediction on the performance of Cognitive Radio networks, Proceedings of IEEE Wireless Telecommunications Symposium WTS (Page: 1 Year of Publication: 2010). [38] T. A. Weiss, F. K. Jondral, Spectrum pooling: an innovative strategy for the enhancement of spectrum efficiency. IEEE Communications Magazine, Vol. 42, n. 3, pp. 8-14, 2004. [39] D. Lu, X. Huang, C. Liu, Adaptive Power Control Based Spectrum Handover for Cognitive Radio Networks. Wireless Communications, pp. 1-5, 2010. [40] Liu, H. Jie, Wang, Z. Xu, Li, S., Yi, M., Study on the performance of spectrum mobility in cognitive wireless network, Procedings of 11th IEEE Singapore International Conference on Communication Systems (Page: 1010 Year of Publication: 2008). [41] Lertsinsrubtavee, A., Malouch, N., Fdida, S., Spectrum Handoff Strategy Using Cumulative Probability in Cognitive Radio Networks. Proceedings of 3rd International Congress on Ultra Modern Telecommunications and Control Systems and Workshops (Page: 1 Year of Publication: 2011). [42] O. Jo & D.-ho. Cho, Seamless spectrum handover considering differential path-loss in cognitive radio systems. IEEE Communications Letters, Vol. 13, n. 3, pp. 190-192, 2009.

Copyright © 2007 Praise Worthy Prize S.r.l. - All rights reserved

[43] Tian, J., Bi, G., A New Link Maintenance and Compensation Model for Cognitive UWB Radio Systems. Proceedings of 6th International Conference on ITS Telecommunications (Page: 254 Year of Publication: 2006). [44] Shi, Q., Taubenheim, D., Kyperountas, S., Gorday, P., Correal, N., Link Maintenance Protocol for Cognitive Radio System with OFDM PHY. Proceedings of 2nd IEEE International Symposium on New Frontiers in Dynamic Spectrum Access Networks (Page: 440 Year of Publication: 2007). [45] Wang, L. Chun., Anderson, C., On the performance of spectrum handoff for link maintenance in cognitive radio, Proceedings of 3rd IEEE International Symposium on Wireless Pervasive Computing (Page: 670 Year of Publication: 2008). [46] L. Zhang, G. Zheng, D. Jia, Spectrum Pooling-Based Vertical Handover for Heterogeneous Cognitive Radio Networks with QoS Constrains. Journal of Convergence Information Technology, Vol. 5, n. 5, pp. 156-169, 2010. [47] Xie, X., Yang, G., Ma, B., Spectrum handoff decision algorithm with dynamic weights in cognitive radio networks, Proceedings of IEEE Global Mobile Congress (Page: 1 Year of Publication: 2011). [48] Halleh, H., N. Fisal, S.K. Syed-Yusof., Joint Channel Estimation and Power Allocation for Cognitive WPMCM UWB System, International Review on Computers and Software, Vol. 6, n. 4, pp. 570-575, 2011. [49] Huang, X., Lu, D., Li, P., Fang, Y., Coolest Path: Spectrum Mobility Aware Routing Metrics in Cognitive Ad Hoc Networks, Proceedings of 31st IEEE International Conference on Distributed Computing Systems (Page: 182 Year of Publication: 2011) [50] Halleh, H., N. Fisal, S.K. Syed-Yusof., Wavelet Based NBI Mitigation Approach for Cognitive UWB Systems, International Review on Computers and Software, Vol. 6, n. 1, pp. 18-24, 2011. 1

Faculty of Electrical Engineering, University Technology Malaysia, Skudai, Malaysia. [email protected], 2 Faculty of Electrical Engineering, Sadjad Higher Education Institute, Mashad, Iran.

Authors’ information Samad Nejatian received his B. Eng in Electrical engineering from Systan and Baluchestan University of Iran in 2003. He got M. Sc in Telecommunication engineering from Sadjad university of Mashhad in 2006. He is currently doing his Ph.D at Universiti Teknologi Malaysia in Telamatic research group. His research interest is intelligent system and cognitive radio. [email protected] phone no; +60127519848 Sharifah Kamilah Bt Syed Yusof received BSc (cum laude) in Electrical Engineering from Geoge Washington University USA in 1988 and obtained her MEE and Ph.D in 1994 and 2006 respectively from UTM. She is currently working as associate professor with Faculty of Electrical Engineering, UTM, and collaborating with TRG laboratory. Her research interest includes Wireless communication, Software define Radio and Cognitive radio.She is a member of Eta Kappa Nu (HKN), and Phi Beta Kappa society. Nurul Mu’azzah Abdul Latiff is a senior lecturer at Faculty of Electrical Engineering, Universiti

International Review on Computers and Software, Vol.7, n. 6

S. Nejatian, et. Al.

Teknologi Malaysia (UTM), Malaysia. She received the B.Eng. degrees from Universiti Teknologi Malaysia in 2002, and M.Sc. and P.hD degrees from Newcastle University, UK, in 2003 and 2008 respectively. She is also a member of Telecommunications Research Group under Centre of Excellence in Telecommunication Technology, UTM. Currently she and her research team is working on MAC layer and routing layer in cognitive radio network. Her research interests also include communication protocols for mobile ad hoc network, wireless sensor network, wireless mesh network and optimization of wireless telecommunication. Vahid Asadpour received the Ph.D. degree in biomedical engineering from the Polytechnique University, Tehran. He is a Professor of Biomedical and Electrical Engineering at the Sadjad Institute of Higher Educations, Mashhad and Azad University, Mashhad Branch. His research area is biomedical signal processing, medical instrumentation, audiovisual signal processing and data classification. His recent research has been concerned with the development of robust methods for audio-visual identification, sleep apnea monitoring, pain detection and applications of Extended Kalman Filter (EKF) and Hidden Markov Model (HMM) in data classification. He has co-led internship projects with IRAN Development and Renewal Organization (IDRO) and Khorasan Regional Electrical Center (KREC). Norsheila Fisal received her B.Sc. in Electronic Communication from the University of Salford, Manchester, U.K. in 1984. M.Sc. degree in Telecommunication Technology, and PhD degree in Data Communication from the University of Aston, Birmingham, U.K. in 1986 and 1993, respectively. Currently, she holds university professor position at the Faculty of Electrical Engineering, UTM and is director of UTM MIMOS Center of Excellence in Telecommunication Technology, Universiti Teknologi Malaysia. Her Professional Affiliations are with Member of Institute Electrical and Electronic Engineers (IEEE), Member of Malaysian National Confederation of Computers (MNCC), Member of Institute For Information Processing (IFIP) and Member of Board of Engineer Malaysia.

Copyright © 2007 Praise Worthy Prize S.r.l. - All rights reserved

International Review on Computers and Software, Vol.7, n. 6