Globecom 2012 - Signal Processing for Communications Symposium
An Energy-Efficient Cooperative Spectrum Sensing Scheme for Cognitive Radio Networks †
Nan Zhao† , F. Richard Yu‡ , Hongjian Sun§ , and Arumugam Nallanathan§
School of Information and Communication Engineering, Dalian University of Technology, Dalian, China ‡ Computer Engineering, Carleton University, Ottawa, ON, K1S 5B6, Canada § Department of Electronic Engineering, King’s College London, London, WC2R 2LS, UK Email:
[email protected]; Richard
[email protected]; {hongjian.sun, arumugam.nallanathan}@kcl.ac.uk
Abstract—Rapidly rising energy costs and increasingly rigid environmental standards have led to an emerging trend of addressing “energy efficiency” aspect of wireless communication technologies. Cognitive radio can play an important role in improving energy efficiency in wireless networks. In this paper, we propose an energy-efficient and time-saving one-bit cooperative spectrum sensing scheme, which has two stages. If the signal-to-noise ratio (SNR) is high or no primary user exists, only one stage of coarse spectrum sensing is needed, by which the sensing time and energy are saved. Otherwise, the second stage of fine spectrum sensing will be performed to increase the spectrum sensing accuracy. Furthermore, only one-bit decision is sent by each secondary user to minimize the overhead. Plenty of simulation is performed, and the results show that the sensing time and energy consumption are both reduced significantly in the proposed scheme.
I. I NTRODUCTION Cognitive radio (CR) has attracted significant attention as a promising technology to overcome the spectrum shortage problem caused by the current inflexible spectrum allocation policy [1]. In CR networks, secondary users (SUs) should sense the radio environment, and adaptively choose transmission parameters according to sensing outcomes to avoid the interference to primary users (PUs). Hence, it is a fundamental issue in CR networks that SUs should be able to efficiently and effectively detect the presence of PUs [2]. The existing spectrum sensing techniques can be broadly divided into four categories: matched filter detection [3], cyclostationary detection [4], and energy detection [5]. In matched filtering detection and cyclostationary detection, SUs should have some knowledge about the primary signal features. Entropy-based detection can solve the noise uncertainty of the spectrum sensing through information entropy. Energy detection has been widely applied, since it does not require a priori knowledge of the primary signals, and has much lower computational complexity than the other three detection schemes. Hence, we focus on energy detection throughout this paper. Spectrum sensing is a tough task especially when signalto-noise ratio (SNR) is low. To improve the performance of spectrum sensing, cooperative spectrum sensing (CSS), where This work was supported by the Fundamental Research Funds for the Central Universities, China Postdoctoral Science Foundation under 2012M510806, and the National Natural Science Foundation of China (NSFC) under Grant 61201224.
individual SUs sense the spectrum and send the information to a fusion centre to obtain the final decision, has been studied extensively [6]–[9]. In conventional hard combination CSS [6], only one-bit decision is sent to the fusion centre by each SU, and its overhead is minimum; however, its performance can still be improved. Soft combination CSS scheme [7] has the optimal performance through using the accurate sensing results from different SUs; however, its overhead is large, which makes it difficult to be implemented in practical networks. A two-bit overhead combination CSS scheme is proposed in [8], [9], which is a tradeoff between hard combination and soft combination CSS. In [10], a three-threshold decision based CSS (TTD-CSS) scheme is proposed, in which the second local decision bit is sent to the fusion centre after the failure of the first cooperation to eliminate the sensing failure. However, the performance of TTD-CSS is not improved much compared with the conventional hard combination CSS. In [11], a twostage spectrum sensing scheme is proposed for multi-channel sensing. It can decrease the average channel sensing time by allowing the spectrum detector to focus on the channels that are more likely to be vacant. A two-stage sensing scheme using energy detection in the first stage and cyclostationary detection in the second stage is designed in [12], and the second stage detection is performed when the decision metric is greater than the threshold. In [13], a two-stage sensing scheme is proposed to minimize the sensing time. Although the above two-stage spectrum sensing schemes can achieve better performance than the single-stage schemes [11]–[13], none of them considers cooperative spectrum sensing. A two-stage two-bit CSS (TSTBCSS) is given in [9], and its performance is improved over the conventional hard combination CSS; however, it uses two-bit overhead and its sensing time and energy consumption can still be reduced. Although some works have been done in cooperative spectrum sensing, most of them focus on achieving high data rate for SU networks while avoiding the interference to PU networks. Consequently, many of these techniques significantly increase system overhead and energy consumption. However, rapidly rising energy costs and increasingly rigid environmental standards have led to an emerging trend of addressing “energy efficiency” aspect of wireless communication technologies [14]. CR can play an important role in improving
3607
energy efficiency in wireless networks, and thus the “green” requirements of CR should be well satisfied [15]. In this paper, we propose an energy-efficient and timesaving one-bit CSS (TSEEOB-CSS) scheme, which has two stages in the spectrum sensing process. If the SNR is high or no PU exists, only one stage of coarse spectrum sensing is needed, by which the sensing time and energy are saved. Otherwise, the second stage of fine spectrum sensing will be performed to increase the spectrum sensing accuracy. In addition, only one bit decision is sent to the fusion centre to minimize the overhead. Simulation results are presented to show that the sensing time and energy consumption are both reduced significantly in the proposed scheme. The rest of this paper is organized as follows. In Section II, the system model of the cognitive spectrum underlay network is described, and the energy detection and cooperative spectrum sensing schemes are introduced. In Section III, the proposed TSEEOB-CSS algorithm is introduced, parameter setting is discussed, and the performance is analyzed. Some simulation results are discussed in Section IV. Finally, we conclude this study in Section V. II. S YSTEM M ODEL We consider cooperative spectrum sensing in a centralized CR network consisting of a cognitive base station (fusion centre) and a number of SUs. In the network, each SU sends its sensing data to the base station, and the base station combines the sensing data from different SUs and makes the final decision on the presence or absence of the PUs. We assume the sensing data are sent from the SUs to the base station free of error throughout this paper. In this section, energy detection and cooperative spectrum sensing are introduced. A. Energy Detection
Pf
= Pr (T > λ|H0 ) = Q (λ − 1)
Pd
= Pr (T > λ|H1 ) = Q (λ − γ − 1)
N 2
,
N 2(γ + 1)2
(3)
,
(4)
where Q(.) is the Q-function. B. Cooperative Spectrum Sensing We consider a CR network composed of K SUs and a base station (fusion centre). We assume that each SU performs energy detection independently and then sends the local decision to the base station, which will fuse all available local decision information to infer the absence or presence of the PU. In the conventional hard combination CSS scheme, each cooperative partner i makes a binary decision based on its local observation and then forwards its one-bit decision Di (Di = 1 stands for the presence of the PU, and Di = 0 stands for the absence of the PU) to the base station. At the base station, all 1-bit decisions are fused together according to the logic rule, and the final decision can be obtained as K ≥k H1 Y = (5) Di λ H1 2 T = (2) |x (n)| λ1 + Δ, sends the local decision D1i = 1 to the fusion centre indicating that PUs exist; if T 1i < λ1 − Δ, sends the local decision D1i = 0 to the fusion centre indicating that no PU exists; if λ1 − Δ < T 1i < λ1 + Δ, nothing will be sent. Step 2: The first stage local decisions D1i are fused at the fusion centre, and the final decision DF can be obtained as ⎧ ⎨1, More than K/2 SUs indicate presence of PU DF= 0, More than K/2 SUs indicate absence of PU (6) ⎩ Final decision cannot be obtained, Otherwise. If the final decision DF can be obtained, DF is sent to each SU. If the final decision DF cannot be obtained, nothing will be done. Step 3: If the final decision DF is received by the SUs, goes to step 6. If the final decision DF is not received by the SUs after τ period, performs the second stage fine energy detection with 2Ns samples. Assume τ T1 in Fig. 1, and thus τ can be ignored compared with T1 and T2 . Step 4: Local decision D2i (i = 1, 2, . . . , K) is obtained through the second stage fine energy detection as 1 T 2i > λ 2 (7) D2i = 0 T 2i < λ 2 , where T 2i is the second stage local sensing result of SUi using energy detection. Then the local decisions D2i are sent to the fusion centre. Step 5: The second stage local decisions D2i are fused at the fusion centre, and the final decision DF can be obtained according to ⎧ K ⎨ 1 D2i ≥ K/2 DF = (8) i=1 ⎩ 0 Otherwise. DF is sent to each SU, and goes to step 6.
There are three important parameters in the proposed TSEEOB-CSS algorithm, λ1 , λ2 and Δ. These parameters affect the detection performance greatly. Thus, we discuss the parameters setting, and give some rules as follows. (1) The rules in setting Δ Theorem 1: For a fixed value of λ1 , the larger the value of Δ is, the better the detection performance can be achieved with longer sensing time. Proof: Assume the square of the sampled signal x(n), 2 |x(n)| , follows a distribution with mean μ and variance σ 2 . The energy detection result can be described as T =
N 1 2 |x (n)| , N n=1
(9)
which follows Gaussian distribution with mean μ and variance σ 2 /N according to the central limit theorem. Therefore the probability that the final decision obtained at each SU after the first stage coarse detection can be expressed as λ1 +Δ (x−μ)2 1 − √ P =1− (10) e 2σ2 /N dx. 2πN σ λ1 −Δ Hence, when Δ is larger, and the probability (1 − P ) indicating the need of second stage fine detection correspondingly becomes larger, which means the sensing time becomes longer. Also, the detection performance becomes better for the probability of second stage detection is larger. Theorem 2: Δ should be smaller as the sample number Ns of the first stage coarse detection becomes larger. Proof: Assume the background noise ω(n) in (1) follows Gaussian distribution with zero mean and unit variance (i.e., 2 ω(n) ∼ N (0, 1)), and thus the square of ω(n), |ω(n)| , follows chi-square distribution with 1 degree of freedom. The 2 mean of |ω(n)| is 1, and its variance is 2. The first stage coarse energy detection result can be expressed as T =
Ns 1 2 |ω (n)| , Ns n=1
(11)
when no PU exists, and it follows Gaussian distribution with 2 mean 1 and variance σH0 = 2/Ns according to the central limit theorem. We set
Δ = aσH0 = a 2/Ns , (12) where a is a positive constant. When the number of samples becomes larger, e.x., Ns2 = 2Ns , the variance of the first stage coarse energy detection result changes to 2/(2Ns ) = 1/Ns ,
3609
1 0.9 0.8
Pd
0.7 Energy detection, 768 samples CSS, 256 samples CSS, 768 samples TTD−CSS,VOTING TSTB−CSS, VOTING
0.6 0.5 0.4
Fig. 2.
TSEEOB−CSS, λ1=1.125
The distribution of the first stage energy detection result.
TSEEOB−CSS, λ1=1.100
0.3
and hence Δ should be set as a 1/Ns accordingly. Therefore, the larger the number of samples is, the smaller the value of Δ should be. (2) The rules in setting λ1 λ1 should be set according to requirement of the false alarm probability (Pf ). In Case I as shown in Fig. 2, λ1 is set relatively low (e.g., below 1). If no PU exists, the distribution of the first stage energy detection result is show in Fig. 2. The probability that the detection result is larger than λ1 + Δ is so large that Pf cannot be set small (e.g., Pf = 0.01). In Case II as shown in Fig. 2, λ1 is set relatively high (e.g., above 1). The probability that the detection result is smaller than λ1 −Δ is so large that Pf cannot be set large (e.g., Pf = 0.5). For Pf above 0.5 is not meaningful in practical networks, λ1 is usually set above 1. Furthermore, if certain Pf can be achieved, the detection performance is better when λ1 is smaller. However, when λ1 becomes smaller, the probability indicating the need of second stage fine detection becomes larger, which means the sensing time becomes longer. (3) The rules in setting λ2 After λ1 and Δ are set, λ2 can be set according to the required Pf through measurements. C. Analysis of Time-Saving and Energy-Efficiency Performance When there are PUs in the network, the first stage coarse energy detection result T follows a distribution with the mean of 1 + μs . If the SNR of PU signal is larger, 1 + μs will be larger, and thus the probability that T > λ1 + Δ will also become larger. This means the probability indicating the need of second stage fine detection becomes smaller, and thus the sensing time and energy consumption will be reduced due to the low probability of the need of second stage fine detection. When no PU exists in the network, the probability that T < λ1 − Δ is relatively large because λ1 is usually set above 1. Therefore, the probability indicating the need of second stage fine detection is relatively large, and the sensing time and energy consumption will also be reduced greatly when no PU exists.
0.2 −16
Fig. 3.
TSEEOB−CSS, λ1=1.050 −14
−12
−10 SNR (dB)
−8
−6
−4
Detection performance against SNR comparison when Pf = 0.1.
From the above analysis, we can conclude that the sensing time and energy consumption is reduced greatly when the SNR of PU signal is high or no PU exists in the proposed algorithm, and it enables a “green” CR network. IV. S IMULATION R ESULTS AND D ISCUSSIONS In this section, we illustrate the performance of the proposed TSEEOB-CSS using computer simulations. We assume that the signal of PU is BPSK modulated, and the baseband symbol rate fb is equal to 1Mbps. The sampling frequency fs at SUs is 64MHz. In TSEEOB-CSS algorithm, the number of samples in the first stage, Ns , is set to 256, and Δ is set to 0.1, which is equal to 1.13σH0 . We compare the detection performance of TSEEOB-CSS algorithm with that of TTD-CSS algorithm in [10] and TSTBCSS algorithm in [9]. In TTD-CSS algorithm, λ is set to 1.007 and Δ is set to 0.1, the number of samples, N , is set to 768, and VOTING rule is used. In TSTB-CSS algorithm, the number of samples in the first and second stages is both set to 768/2 = 384, λ is set to 1.034, and Δ is set to 0.1. The detection performance of the energy detection and the hard combination cooperative spectrum sensing scheme is also compared. The detection probability (Pd ) of these algorithms is compared in Fig. 3, with unit power of the background noise and Pf = 0.1. From Fig. 3 we can see that the performance of TSEEOBCSS algorithm with λ1 = 1.125, 1.100, and 1.050 is all better than that of 768-sample energy detection algorithm and 256-sample hard combination CSS algorithm. The larger the value of λ1 is, the better the performance TSEEOBCSS algorithm can achieve. The detection performance of TSEEOB-CSS algorithm with λ1 = 1.1000 is close to that of TTD-CSS algorithm, TSTB-CSS algorithm and 768-sample hard combination CSS algorithm, and its performance is even better than that of these algorithms when λ1 is equal to 1.050.
3610
70 1
λ1=1.125 60
0.8
Percent of reduced energy (%)
Probability of second−stage detection
0.9
0.7 0.6 0.5 0.4 0.3
λ1=1.125
0.2
λ1=1.100
0.1
λ1=1.050
0 −24
−20
−18
−16
−14 −12 SNR (dB)
−10
λ1=1.050
50
40
30
20
10
0 −24 −22
λ1=1.100
−8
−6
−4
Fig. 4. Probability of the second stage detection performance comparison with different λ1 values.
A remarkable advantage of the proposed TSEEOB-CSS algorithm is that it can save sensing time and energy consumption greatly, which will be discussed in the following. The probability of second stage fine detection under different SNRs is analyzed in Fig. 4. We can see that although the detection performance with λ1 = 1.125 is worse than that with λ1 = 1.100 and 1.050 (from Fig. 3), the probability of second stage detection is the lowest and close to 0.2 with the shortest sensing time and smallest energy consumption. Hence, we can conclude that the sensing time is shorter, the energy consumption is smaller, and the detection performance is worse with larger λ1 . In practical CR networks, if the sensing time or energy consumption is the most important factor to be considered, we can use large λ1 ; otherwise, λ1 should be smaller. Using the probability of second stage fine detection obtained from Fig. 4, we can calculate the percent of the reduced processing energy by TSEEOB-CSS algorithm under different SNRs, and the results are shown in Fig. 5. From Fig. 5, we can see that when SNR is larger than -5dB, the percentage of the reduced energy is more than 60, and the maximal percentage of the reduced energy is 66.67. When no PU exists in the network, the energy consumption can also be reduced by 50%, 35% and 15% when λ1 is set to 1.125, 1.100, and 1.1050, respectively. Thus, the proposed algorithm can improve energy efficiency significantly. V. C ONCLUSIONS In this paper, we have presented a two-stage time-saving and energy-efficient cooperative spectrum sensing algorithm for CR networks. In the algorithm, only one stage of sensing is needed when the SNR is high or there is no primary user. One-bit decision is sent by each SU to minimize the overhead. We analyzed the design parameters in the proposed algorithm. Simulation results were presented to show the effectiveness of the proposed algorithm.
−22
−20
−18
−16
−14 −12 SNR (dB)
−10
−8
−6
−4
Fig. 5. Energy efficiency performance comparison of the proposed algorithm with different λ1 values.
R EFERENCES [1] S. Haykin, “Cognitive radio: Brain-empowered wireless communications,” IEEE J. Sel. Areas Commun., vol. 23, no. 2, pp. 201–220, Feb. 2005. [2] S. Haykin, D. J. Thomson, and J. H. Reed, “Spectrum sensing for cognitive radio,” Proc. IEEE, vol. 97, no. 5, pp. 849–877, May. 2009. [3] H. V. Poor, An Introduction to Signal Detection and Estimation. New York: Springer-Verlag, 1994. [4] M. Ghozzi, F. Marx, M. Dohler, and J. Palicot, “Cyclostationaritybased test for detection of vacant frequency bands,” in Proc. Int. Conf. Cogn. Radio Oriented Wirel. Netw. Commun., pp. 43–47, Jun. 2006. [5] D. Cabric, S. M. Mishra, and R. W. Brodersen, “Implementation issues in spectrum sensing for cognitive radios,” in Proc. Asilomar Conf. on Signals, Systems, and Computers, pp. 772–776, Nov. 2004. [6] G. Ganesan and Y. Li, “Cooperative spectrum sensing in cognitive radio networks,” in Proc. Int. Symp. New Front. Dynamic Spec. Netw., pp. 137– 143, Nov. 2005. [7] J. Ma and Y. Li, “Soft combination and detection for cooperative spectrum sensing in cognitive radio networks,” in Proc. IEEE Global Telecommun. Conf., pp. 3139–3143, Nov. 2007. [8] M. Mustonen, M. Matinmikko, and A. Mammela, “Cooperative spectrum sensing using quantized soft decision combining,” in Proc. Int. Conf. Cogn. Radio Oriented Wirel. Netw. Commun., pp. 1–5, Jun. 2009. [9] N. Zhao, “A novel two-stage entropy-based robust cooperative spectrum sensing scheme with two-bit decision in cognitive radio,” Wirel. Pers. Commun., to be publeshed. [10] H. X. Xia, G. P. Zhang, and H. B. Xu, “A flexible cooperative spectrum sensing scheme for cognitive radio networks,” IEICE Electron. Express, vol. 8, no. 8, pp. 542–548, Apr. 2011. [11] S. Fazeli-Dehkordy, K. N. Plataniotis, and S. Pasupathy, “Twostage spectrum detection in cognitive radio networks,” in Proc. IEEE Int. Conf. Acoust. Speech Signal Process., pp. 3118–3121, Mar. 2010. [12] S. Maleki, A. Pandharipande, and G. Leus, “Two-stage spectrum sensing for cognitive radios,” in Proc. IEEE Int. Conf. Acoust. Speech Signal Process., pp. 2946–2949, Mar. 2010. [13] P. R. Nair, A. P. Vinod, and A. K. Krishna, “A fast two stage detector for spectrum sensing in cognitive radios,” in Proc. IEEE Veh. Tech. Conf. Fall, pp. 1–5, Sep. 2011. [14] G. Y. Li, Z. Xu, C. Xiong, C. Yang, S. Zhang, Y. Chen, and S. Xu, “Energy-efficient wireless communications: tutorial, survey, and open issues,” IEEE Wireless Comm., vol. 18, no. 6, pp. 28–35, Dec. 2011. [15] G. Gur and F. Alagoz, “Green wireless communications via cognitive dimension: an overview,” IEEE Network, vol. 25, no. 2, pp. 50–56, Mar. 2011.
3611
