Professional Documents
Culture Documents
28 of observation samples used to estimate the kurtosis. γ = σσ2s |h|2 is the instantaneous SNR and θ =
n
29
30 tan−1 Im[h]
Re[h]
is the instantaneous phase angle of the channel. The values of ai,j s for 16-QAM constellation
31
32
33 are evaluated and given in the Table 1. It is important to note that unlike second order statistics which
34
35 requires only the instantaneous SNR, the higher order statistics depend on the phase information of the
36
37
38
channel also.
39
40
41
42 TABLE I
43
44 M ODULATION CONSTANTS FOR 16-QAM CONSTELLATION
45
46
47 Modulation Scheme Modulation Constants
48
49
24ρ8
n 96ρ8
n 46.08ρ8
n −48.96ρ8
n 33.28ρ8
n
50 16-QAM a00 = N
, a10 = N
, a20 = N
, a21 = N
, a30 = N
,
51
52 128.64ρ8 10.33ρ8 −1.93ρ8 1.74ρ8 24ρ8
n n n n n
53 a31 = N
, a40 = N
, a41 = N
, a42 = N
, c00 = N
54
55
56
57
58
59
60
61
62
63
64
65
19
1
2
3 A PPENDIX B
4
5 U NBIASED PROPERTY OF THE K URTOSIS ESTIMATE
6
7
8 The kurtosis estimate in (5) is an unbiased estimate. This can be proved by taking the expectation on
9
10
11 both sides of (5),
12 Nobs Nobs 2
13 1 X 4 1 X 2
14 E[k̂] = E |r(i)| − 2 |r(i)|
15
Nobs i=1 Nobs i=1
16 Nobs Nobs Nobs
17 1 X 4 1 X 2 1 X 2
=E |r(i)| − 2 |r(i)| |r(j)| ,
18 Nobs i=1 Nobs i=1 Nobs j=1
19
20 Nobs Nobs N obs
1 X 4 1 X X
2 2
21 =E |r(i)| − 2 2
|r(i)| |r(j)| , (40)
22 Nobs i=1 Nobs i=1 j=1
23
24 where, E[·] is the expectation operator. Taking the expectation operator inside in summation, in (40), and
25
26
27 combining the like terms and unlike terms, we obtain,
28
Nobs
29 1 X 4 2 4 2 2
30 E[k̂] = E[|r(i)| ] − 2 Nobs E[|r(i)| ] + (Nobs − 1){E[|r(i)| ]} , (41)
31 Nobs i=1 Nobs
32
33 Since the expectation operator is independent of the index i, we can rewrite (41) as,
34
35
2 4 1
36 E[k̂] = 1− E[|r| ] − 2 1 − {E[|r|2 ]}2 . (42)
37 Nobs Nobs
38
39 For large Nobs , (42) can be approximated as E[k̂] ≈ k, and hence, k̂ is an asymptotically unbiased estimate.
40
41
42 It has been verified by simulations that for low to medium SNR values, choosing around Nobs = 50 samples
43
44 are sufficient to produce an almost unbiased estimate, (E[k̂ = k).
45
46
47
48 R EFERENCES
49
50
51 [1] Federal Communications Commission, “Spectrum Policy Task Force,” Rep. ET docket no. 02-135, Nov. 2002.
52
53 [2] S. Haykin, “Cognitive Radio: Brain-Empowered Wireless Communications”, IEEE journal on Selected Areas in Communications, vol.
54
23, no. 2, pp. 201–220, Feb. 2005.
55
56 [3] A. Sahai, N. Hoven, and R. Tandra, “Some fundamental limits on Cognitive Radio”, Proc. Allerton Conf. on Commununictions, control,
57
58 and computing, Monticello, Oct. 2004.
59
60 [4] Y. Hur, J. Park, W. Woo, K. Lim, C.-H. Lee, H. S. Kim, and J. Laskar, “A wideband analog multi-resolution spectrum sensing (MRSS)
61
62 technique for cognitive radio (CR) systems”, in Proc. IEEE Int. Symp. Circuit and System, Greece, May 21–24, 2006, pp. 4090-4093.
63
64
65
20
1
2
3 [5] J. Lunden, V. Koivunen, A. Huttunen, and H. V. Poor, “Spectrum sensing in cognitive radios based on multiple cyclic frequencies”, in
4
5 Proc. of the Second Int. Conf. on Cognitive Radio Oriented Wireless Networks and Communications (CrownCom), Orlando, FL, USA,
6
7 July 31–August 3, 2007.
8
[6] A. Sahai and D. Cabric, “Spectrum sensing: fundamental limits and practical challenges”, in IEEE International Symposium on New
9
10 Frontiers in Dynamic Spectrum Access Networks (DySPAN 05), Baltimore, Md, USA, Nov. 2005.
11
12 [7] Y. Zeng, Y.C. Liang, A.T. Hoang, and R. Zhang, “A Review on Spectrum Sensing for Cognitive Radio: Challenges and Solutions”,
13
14 EURASIP Journal on Advances in Signal Processing, vol. 2010, Article ID 381465.
15
16 [8] W. A. Gardner, “Exploitation of spectral redundancy in cyclostationary signals”, IEEE Signal Processing Magazine, vol. 8, no. 2, pp.
17
18 14-36, 1991.
19
20 [9] V. I. Kostylev, “Energy detection of a signal with random amplitude”, in Proc. IEEE Int. Conf. Commun., New York, Apr. 28–May 2,
21
2002, pp. 1606–1610.
22
23 [10] Y. Zeng and Y. C. Liang,“Covariance based signal detections for cognitive radio”, in Proceedings of the 2nd IEEE International
24
25 Symposium on New Frontiers in Dynamic Spectrum Access Networks (DySPAN 07), pp. 202-207, Dublin, Ireland, April 2007.
26
27 [11] X. Chen and S. Nagaraj, “Entropy Based Spectrum Sensing in Cognitive Radio”, IEEE Wireless Telecommunications Symposium, 2008,
28
29 pp. 57–61, April. 2008.
30
31 [12] A. Mody, “Spectrum Sensing of the DTV in the Vicinity of the Pilot Using Higher Order Statistics”, doc.: IEEE 802.22-07/0370r1,
32
Aug. 2007.
33
34 [13] S. J. Shellhammer, “Spectrum Sensing in IEEE 802.22”, IAPR Wksp. Cognitive Info. Processing, in Santorini, Greece, Jun. 2008.
35
36 [14] Y. Sun, Y. Liu, X. Tan, “Spectrum Sensing for Cognitive Radio Based on Higher-Order Statistics”, 4th International Conference on
37
38 Wireless Communications, Net- working and Mobile Computing, pp. 1–4, Oct. 2008.
39
40 [15] F. F. Digham, M.-S. Alouini and M. K. Simon, “On the energy detection of unknown signals over fading channels”, Proc. IEEE, Int.
41
42 Conf. Commun., Anchorage, AK, USA, pp. 3575–3579, May. 2003.
43
44 [16] W. Zhang, R. K. Mallik and K. B. Letaief, “Cooperative Spectrum Sensing Optimization in Cognitive Radio Networks”, Proc. IEEE
45
ICC, Beijing, China, pp. 3411–3415, May. 2008.
46
47 [17] J. M. Mendel, “Tutorial on Higher-Order Statistics(Spectra) in Signal Processing and System Theory: Theoretical Results and some
48
49 Applications”, Proc. IEEE, vol. 79, pp. 278–305, Mar. 1991.
50
51 [18] J. A. R. Fonollosa, “Sample Cumulants of Stationary Processes: Asymptotic Results”, IEEE transactions on Signal Processing, vol.
52
53 43, no. 4, April. 1995.
54
55 [19] A. Swami and B. M. Sadler, “Hierarchical Digital Modulation Classification Using Cumulants”, IEEE transactions on Communications,
56
vol. 48, no. 3, Mar. 2000.
57
58 [20] A. V. Dandawate and G. B. Giannakis, “Asymptotic Properties and covariance expressions of k-th order sample moments and cumulants”,
59
60 Twenty-Seventh Asilomar Conference on Signals, Systems and Computers, vol. 2, pp. 1186–1190 , Nov. 1993.
61
62 [21] H. V. Poor, An Introduction to Signal Detection and Estimation, 2nd ed., Springer Verlag, 1994.
63
64
65
21
1
2
3 [22] A. Ghasemi and E. S. Sousa, “Collaborative spectrum sensing for opportunistic access in fading environments”, Proc. IEEE Symp.
4
5 New Frontiers in Dynamic Spectrum Access Networks (DySPAN05), Baltimore, USA, pp. 131–136, Nov. 2005.
6
7 [23] I. S. Gradshteyn and I. M. Ryzhik, Table of Integrals, Series and Products, 7th ed., San Diego, CA: Academic, 2007.
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
22
1
2
3
4
5
6
7
8
9
10
11 10
−1
12
13
Pm,avg
14
15
16
17
18 −2
19 10
20 Kurtosis based method
21 Energy based method
22
23 −15 −10 −5 0 5 10
24 SNR(dB)
25
26
27
Fig. 1. Comparison of energy and proposed kurtosis based spectrum sensing method in terms of probability of miss, for α0 =0.01 and
28
29 Nobs =350.
30
31
32
33
34
35
36
37 10
−1
38
39
40
41
42
43
44
45
Pm,avg
−2
46 10
47
48
49
50
51
52
Average SNR = 8.82dB
53
54 Average SNR = 7.32dB
−3
10
55 10
−1
10
0
56 Pf,avg
57
58
59
Fig. 2. Complementary ROC curves for two different average SNR values using Nobs = 500 samples for kurtosis based scheme.
60
61
62
63
64
65
23
1
2
3
4
5 −1
6 10
7
8
9
−2
10 10
11
12
13
Pm,avg
14 10
−3
15
16
17 Kurtosis : 1 user
18 −4 Energy : 1 user
10
19 Kurtosis : 2 users
20 Energy : 2 users
21 Kurtosis : 4 users
22 Energy : 4 users
23 0 5 10 15
24 Average SNR(dB)
25
26
27
Fig. 3. Probability of miss vs average SNR for collaborative scheme using OR-fusion rule, for Nobs = 350, which depicts the considerable
28
29 improvement offered by kurtosis based scheme over energy based scheme for 4 collaborative CRs.
30
31
32
33
34
35
36
37 −1
10
38 Simulated
39 Analytical
40
41
42
43
44 −2
10
45
Pm,avg
46
47
48
49
50
−3
51 10
52
53
54
55 5 10 15 20 25 30
Average SNR(dB)
56
57
58
59
Fig. 4. Curves of analytical upper bound on probability of miss, derived in Section V and that obtained by simulation for Nobs =50.
60
61
62
63
64
65