An Efficient Spectrum Mobility Management Strategy in Cognitive Radio Networks

An Efficient Spectrum Mobility Management Strategy in Cognitive Radio Networks
Prabhjot Kaur
Department of Electronics and Communication Institute of Technology and Management Gurgaon, India [email protected]
Abstract—In this paper, we propose to design a spectrum mobility strategy using Fuzzy Logic System for cognitive radio networks (CRN). The proposed strategy enables cognitive radios (CR) to vacate the spectrum if primary user (PU) needs it back or to adjust its transmit power in order to avoid interference with PU or nearby CR. Using our scheme, CR switches between the bands only if it is not able to modify its transmit power within the tolerable interference limits. Thus, our work is divided in two modules, one with the priority to control CR transmit power within tolerable range and second to switch to another frequency band in order to avoid interference. Simulated results show that using our proposed fuzzy power control scheme, we can decrease transmit power consumption and achieve lower number of spectrum handovers. Keywords-Cognitive Radio, fuzzy logic, power control, spectrum holes, spectrum mobility

Moin Udin, Arun Khosla
Dr. B.R. Ambedkar National Institute of Technology Jalandhar, India

I.

INTRODUCTION

To be able to transmit and receive in a particular band, the service provider or the user obtains spectrum usage license from the respective government / regulatory body. In present scenario, the spectrum is distributed among service providers on mutually exclusive / non shared basis. This way of spectrum distribution is called fixed spectrum assignment and the licensed user as primary user (PU). As per the observations of federal communication commission (FCC) of USA [1], many of the frequency bands are unoccupied most of the times. This underutilization of spectrum motivated the researchers to think in terms of spectrum holes and defined in [3] as A spectrum hole is a band of frequencies assigned to a primary user, but, at a particular time and specific geographic location, the band is not being utilized by that user. In 2000, J. Mitola [2] at first introduced a device called cognitive radio (CR) that utilizes spectrum holes to communicate wherever and whenever needed, thus increasing spectrum utilization efficiency. CR is built upon software defined radio (SDR) platform. The basic feature of an SDR is its reconfigurability to different physical layer attributes [4]. The efforts for standardizing the air interface for CR is on going and first draft for the standard IEEE 1900.4 is also available now [5]. Another standard being developed is IEEE 802.22. This is based on underutilized frequencies in TV band [6]. Main focus among all these activities, efforts and

standardization is to enable CR to sense the spectrum holes effectively, communicate as required without causing interference to PU and to increase spectrum utilization efficiency. Protection to PU can be provided in two ways, one to return the frequency as and when demanded back by the PU else control the transmit power of CR in a way that interference temperature limit is not violated. To obtain high signal to noise ratio for better data rate, transmit power should be high. However, increasing transmit power level has an undesirable effect of increasing interference. Thus, it is not possible to represent the overall system performance with single index of performance. Rather tradeoffs for decision to control transmit power or switch to another frequency needs to be adopted. In this paper, we propose to tackle this issue in computationally traceable fashion using fuzzy logic. A hierarchical model is defined where two fuzzy logic systems (FLS) are used, one to control the transmit power and another to decide if handover to another frequency is required or not. Our work is different from the prior art available in fuzzy based CR networks as follows. A collaborative spectrum sensing scheme in a CR network is proposed in [8]. In [9], cross layer optimization architecture between medium access control (MAC) and transport layer is presented without employing spectrum sensing or access schemes. An opportunistic spectrum access scheme is proposed in [11]. However our work focuses on efficient power control and spectrum handoff scheme. In [7], a fuzzy based power control mechanism is designed for secondary user (SU) to coexist with the primary user (PU/s) for better spectrum utilization and lesser transmission power consumption as compared to the fixed power control. In our opinion, the three antecedents considered in [7] are related to each other and convey redundant information. In our scheme, power control is optimized with only two antecedents sufficient to decide the power control. Further handover strategy is also designed using fuzzy logic in our work. Another fuzzy based work [10] is done for spectrum handoff decision to be taken by SU in a decentralized network. Spectrum handover is initiated when interference at PU exceeds certain threshold or if QoS at SU is not satisfactory. This also forms the basis for the algorithm proposed in our work. However, we consider power modification as one of the antecedents in deciding the spectrum mobility. The rest of the paper is organized as follows. In section II,

This work is sponsored by Bureau of Research and Institutional Development, All India Council for Technical Education, New Delhi, India.

978-1-4577-0183-2/09/$26.00 ©2011 IEEE

we briefly introduce cognitive radios and fuzzy logic systems. The proposed algorithm is described in section III and the implemented architecture of our work is described in section IV. Finally the results are discussed in section V. Conclusions and future scopes are presented in section VI. II. PRILIMINARIES

A. Cognitive Radios The under utilization of licensed bands has motivated the development of CR [1]-[6]. A CR can reliably sense wide bandwidth, detect spectrum holes and use these holes for communication as and when required only if it does not interfere with PU. CR in this context is also referred to as an SU. The air interface for CR is based on following four main procedures: Spectrum sensing: detection of spectrum holes with the help of spectrum sensing techniques as transmitter detection, interference based detection, and cooperative detection. Spectrum management: acquiring the best available spectrum to meet user communication requirements. The function includes spectrum analysis and then selecting the band according to user requirements. Spectrum mobility: mobility occurs when CR changes its frequency band upon detection of PU signal. CR needs to switch to another frequency, maintaining seamless communication requirements during the transition to better spectrum. Spectrum sharing: once a CR knows its transmitting frequency, it informs its receiver about the band chosen. Besides, a fair spectrum scheduling method is to be provided. It can be regarded to be similar to generic MAC problems in existing systems. In this paper, we focus on two of the above mentioned procedures. Firstly, the spectrum management to control transmit power in order to avoid interference and maintain quality of service (QoS) for passing instructions to change over the frequency, and secondly, spectrum mobility for switching to another band when interference or QoS deterioration exceeds tolerable limits. B. Spectrum mobility and management The next generation networks target to use the spectrum in a dynamic manner by allowing the CR, to operate in the best available frequency band. To obtain the best available channel, a CR has to capture the best available spectrum using spectrum sensing defined earlier. Spectrum mobility arises in two cases as: x Current channel conditions become worse. This may be purely due to the channel conditions or due to increase in interference with adjacent users. x A PU appears in the network The protocols for different layers of the CR network stack must adapt to the channel parameters of the operating

frequency and should be transparent to the spectrum handoff and the associated latency. Thus the purpose of spectrum mobility or handover is to make sure that such transitions are made smoothly and as soon as possible such that the applications running on a CR perceive minimum performance degradation during a spectrum handoff [14]. Mobility management protocols are required to accomplish the spectrum mobility functionalities. These protocols learn in advance about the spectrum handoffs and make sure that the ongoing communications undergo only minimum performance degradation. III. PROPOSED STRATEGY We illustrate our proposed spectrum mobility management algorithm through a flow chart in Fig 1. In a network comprising SU and PU, an SU has CR capabilities to detect spectrum holes and is reconfigurable. As PU is licensed user, it is given priority for transmission and communication on its frequency only takes place if its performance is not affected and SU does not cause any interference. Our strategy focuses on the idea that instead of Start

Estimate PU transmitted and Received Power

FLS 1 to compute power modification required in transmit power of SU

Adjust SU transmit power as per modification required

Estimate QoS with modified power

Is QoS met and interference tolerable? Yes Fuzzy Logic to decide spectrum handover

No

Spectrum handover takes place

Figure 1. Power control and spectrum handover algorithm

returning the band to PU whenever it wants and switching to another available band, CR should try to modify transmit power in order to avoid interference with PU. As illustrated in figure 1, SU at first estimates the power it receives from PU and compares it with the power being transmitted by the PU. This comparison will help estimate the power modification required in the transmit power of SU. However, if power transmitted becomes very low, signal to noise ratio (SNR) will drop and also the rate of transmission. In effect, QoS will be deteriorated. Spectrum handover will be the only choice here. Thus, first we try to reduce the interference between SU and PU and depending upon various QoS parameters, we make SU decide whether or not spectrum handover should take place. IV.
IMPLEMENTATION OF PROPOSAL

(a)

The proposed strategy of section III is implemented using hierarchical FLS [12]. Power modification is determined with the help of FLS 1 and spectrum handover decision is taken using FLS 2. The block diagram of the implemented hierarchical FLS is presented in figure 2. A. FLS1 for power control technique We compute the percentage change in power required for SU transmission in order to avoid interference with PU using the model as shown in figure 2. The two antecedents used here are as below. Antecedent 1: RPower.as power received by SU from PU. Antecedent 2: TPower as actual power transmitted by PU. We assume that data rate of PU is deterministic for SU which helps deducing SNR of PU and consequently the TPower of PU. Transmission and reception ranges of subscriber station (SS) of IEEE 802.16 [13] are considered here to define the membership function ranges of TPower and RPower respectively. Membership function for RPower is defined with the help of received signal strength indicator (RSSI) measured at SS. From a succession of RSSI measurements, the SS derives and updates the estimates of RSSI mean and standard deviation. The range over of these measurements extend 3 dB on each side beyond the –40 dBm to –123 dBm limits for the final averaged statistics that are reported [13]. Also, SS receiver is capable of decoding a maximum on-channel signal of –30 dBm. Thus, we consider a random generation of TPower and RPower in the range of -84 dBm to 43 dBm and -123 dBm to 30 dBm respectively for an SU. RPower Rsu FLS 1 For Power Modified Pmod

(b)

(c)
Figure 3. Membership functions for FLS 1 (a) antecedent 1(b) antecedent 2 (c) consequence

Both these antecedents and the consequence are characterized by three fuzzy sets T (x) where x represents antecedent and consequence variable as shown in (1)

T ( RPower ) T (TPower ) {Low, Medium, High}

T ( P mod ified )

(1)

FLS 2 For Spectrum mobility

PHandover

TPower

The corresponding membership functions are shown in figure 3. We compare RPower with TPower in order to access the interference level, which in this case helps estimate the modification required in transmit power of SU. A complete set of logical relations between RPower and TPower is made to build rule base for FLS 1 as given in Table 1. As an example to justify our logic, consider rule 7, RPower is high and TPower is low, this corresponds to a situation where interference level is very high thus, power modification should be high in order to reduce SU transmit power to avoid interference with PU.

Figure 2. Block diagram of the implemented FLS system

TABLE I Rule No. RPower

RULE BASE FOR FLS 1 TPower Pmodified Rule No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27

TABLE 2 RPower low low low low low low low low low medium medium medium medium medium medium medium medium medium high high high high high high high high high

RULE BASE FOR FLS 2 Pmodified low low low medium medium medium high high high low low low medium medium medium high high high low low low medium medium medium high high high Rsu low medium high low medium high low medium high low medium high low medium high low medium high low medium high low medium high low medium high PHandover PNo PNo PYes No PNo PNo No No PNo PYes PYes Yes PNo PYes PYes PNo PNo PYes Yes Yes Yes PNo Yes Yes PYes PYes Yes

1 2 3 4 5 6 7 8 9

low low low medium medium medium high high high

low medium high low medium high low medium high

low low medium low medium medium medium high
high

B. FLS2 for spectrum mobility Second FLS is designed to find the probability of handover with two antecedents and one consequence. One of the inputs RPower is same as that of FLS1. Consequence of FLS1 is another input to FLS2. The third antecedent is bit rate of SU, Rsu considered as an indicator of QoS for SU. The bit rate chosen is 96 Mb/s with a roll off factor of 0.25, channel size of 25 MHz, 64-QAM modulation scheme corresponding to IEEE 802.16 standard [13]. Rsu is characterized with a term set of three fuzzy sets as shown in (2) and the output linguistic variable Pmodified is characterized into four fuzzy sets as shown in (3). Membership functions for (2) and (3) are as represented in figure 4b and 4c respectively.

PNo = probably no, PYes = probably yes

T ( Rsu ) {Low, Medium, High} T ( PHandover ) {No, PNo , PYes, Yes}

(2) (3)

Where PNo and PYes are probably no and probably yes respectively. The corresponding logical rule base so designed is presented

in table 2. We present the proof of our logic considering rule 8, where antecedent 1 representing interference is low, antecedent 2 representing QoS is high and modification in power is medium, there is no handover required. This corresponds to a condition for no handover when there is tolerable interference and high QoS. Similarly we logically design the complete set of relations and corresponding consequence to obtain complete rule base. V.
RESULTS AND DISCUSSION

(a)

(b)
Figure 4. (a) membership functions for rate Rsu as antecedent 2 (b) membership function for handover probability as consequence

Different parameters considered for simulations are based on IEEE 802.16 standard with their values same as discussed in section IV A and B. Percentage modification made in transmit power of SU with variation in power transmitted by the PU and that of received at SU is shown in figure 5(a) and (b) respectively. The results show lower modification level of SU transmit power with TPower and RPower, steep increase in modification when RPower and TPower shift to medium range and gradual increase in this level while TPower and RPower are high. Thus our strategy helps SU to modify its transmit power gradually rather than in absolute fixed steps. This helps CR to adapt to channel conditions efficiently. The modification level in SU transmit power helps it decide if handover is required or not. The determined possibility of handover with variation in RPower, Pmodified and data rate of SU is as shown in figure 6. A reduction in handover frequency by 50%, 50% and 80% is obtained with an individual variation with RPower, Rsu and Pmodified respectively. This is proved by the fact that handover probability is above 60 % (as defined by its membership function) for half set of readings corresponding to RPower and Rsu and for 20% of the set of Pmodified as shown in figure 6. Observations also show that

80 70 60 50 40 30 20 10 0 -83 -82 -80 -18 -18 2 3 21 35 44 Tranm it pow er of PU (Tpow er)

Power Modification (Pmodified in %)

80 70 60 50 40 30 20 10 0 -120 -122 -122 -92 -85 -85 -60 -60 -54 -44

Power Modification (Pmodified in %)

Recieved Pow er of PU at SU (RPow er)

(a)

(b)

Figure 5. Percentage modification in transmit power of SU with varying (a) transmit power of PU (b) received power of PU at SU

80 Handover Probability (% )

80 H an o v er p ro b ab ility ( % ) 70 60 50 40 30 20 10 0 0 20 40 60 Pow er Modification (%) 80
Handover Probability (%)

80 70 60 50 40 30 20 10 0 54 18 20 52 88 94 52 64 15 38 Data rate (Mbps)

70 60 50 40 30 20 10 0 15 18 20 38 52 52 54 64 88 94 Recieved Pow er of PU at SU Rsu (dBm )

(a)

(b)

(c)

Figure 6. Handover probability achieved due to variation in (a) received power of PU at SU (b) power modification done in transmit power of SU as per output received using FLS1 (c) data rate of SU

the times a handover is made, the power modified is also above 50 %. Thus using our strategy, a handover takes place when either QoS is not met or the interference limit is above tolerable limit. Gradual modification in power and corresponding lesser number of handovers is the major contribution of this work. VI.
CONCLUSION AND FUTURE SCOPE

ACKNOWLEDGMENT Prabhjot Kaur thanks Bureau of Research and Institutional Development, All India Council for Technical Education, New Delhi, India for grant received under research promotion scheme vide File No. 8023/BOR/RID/RPS-65/2007-08. REFERENCES
[1] Federal Communication Commission (FCC), “Unlicensed operation in TV Broadcast Bands,” Notice for Proposed Rule Making, ET Docket No. 04-113, May 25, 2004. J. Mitola, “Cognitive radio: An integrated agent architecture for software defined radio,” PhD Dissertation, Royal Inst. Technol. (KTH), Stockholm, Sweden, 2000. S. Haykin, “Cognitive radio: Brain-empowered wireless communications,” IEEE Journal on Selected Areas in Communications, vol. 23, no. 2 , pp.201-220, Feb. 2005. Mitola, J., III; Maguire, G.Q., Jr. “Cognitive radio: making software radios more personal,” IEEE Personal Communications, Volume 6, Issue 4, Aug 1999 pp. 13 – 18. Hiroshi Harada , Stanislav Filin, “White Space Communication Enabled by IEEE Standard 1900.4”, Feburary, 2009. C. Cordeiro, K. Challapali, D. Birru, and S. Shankar, “IEEE 802.22: The first worldwide wireless standard based on cognitive radios,” New Frontiers in Dynamic Spectrum Access Networks, 2005.

An efficient power control and handover strategy has been proposed in this paper. The simulated results show that the number of handovers are reduced if SU transmit power is controlled within the interference tolerable limits. However, this being one of the initial proposals of the use of fuzzy logics in CR, it does not encompass learning approaches to optimize the proposed FLS, and the fuzzy rule base is selected based on the logical reasoning. Thus a closer optimized scenario will be created in future with an adaptation to a real time heterogeneous network. Further, it is assumed that data rate of PU is deterministic for SU which helps deducing SNR of PU and consequently the TPower of PU. In such scenarios, the proposed scheme seems fit for point-to-point PU communication scenarios but probably not for the systems with many undetectable receivers such as TV broadcast etc.

[2]

[3]

[4]

[5] [6]

[7]

[8]

[9]

DySPAN 2005.2005 First IEEE International Symposium, pp.328-337, Nov. 2005. Le, H.-S.T. Qilian Liang, “ An Efficient Power Control Scheme for Cognitive Radios,” IEEE conference on Wireless Communications and Networking, WCNC March, 2007. pp. 2559-2563 Wendong Yang, Yueming Cai and Youyun Xu, “A fuzzy collaborative spectrum sensing scheme in cognitive radio,” Proceedings of 2007 International Symposium on Intelligent Signal Processing and Communication Systems Nov.28-Dec.1, 2007 Xiamen, China, pp. 566-569 Baldo, Nicola; Zorzi, Michele, “Fuzzy Logic for Cross-layer Optimization in Cognitive Radio Networks,” 4th IEEE Consumer Communications and Networking Conference, CCNC 2007. Jan. 2007, pp: 1128 – 1133 and IEEE Communications Magazine, April 2008. pp. 64-71.

[10] L. Giupponi, Ana I. Pérez-Neira, “Fuzzy-based Spectrum Handoff in Cognitive Radio Networks,” 3rd International Conference on Cognitive Radio Oriented Wireless Networks and Communications, CrownCom 2008. pp. 1-6. [11] Hong-Sam T. Le1 and Hung D. Ly2, “Opportunistic Spectrum Access Using Fuzzy Logic for Cognitive Radio Networks,” Second International Conference on Communications and Electronics, ICCE 2008, June 2008. pp. 240 – 245. [12] G. Klir and B. Yuan, Fuzzy sets and fuzzy logic: theory and applications. Upper Saddle River, NJ, USA: Prentice-Hall PTR, 1995. [13] IEEE standard for Local and metropolitan area networks Part 16: Air Interface for broadband Wireless Access systems, IEEE Standard 802, 29 May, 2009. [14] Cognitive Radio. [Online]. Available : http://www.iut.ac.ir/~omidi/SDR07/07_CognitiveRadio/home.htm