Signal Detection Based on Auto - Correlation and Kurtosis

The spectrum sensing problem has augmented new scenarios with cognitive radio and opportunistic spectrum access concepts. It is one of the most challenging issues in cognitive radio systems. In this paper, we present a novel technique to sense and blindly detect OFDM signal based on auto-correlation and kurtosis. We carryout performance analysis of the proposed approach at various channel conditions and effect of increase in sample length of the sensed signal. Further, threshold value to decide the existence of OFDM signal is calculated. In addition, for 100 % detection rate, we perform an analysis of various pilot allocation strategies in OFDM signal and consequently its effect on detector performance. Detector performance improves further by using frequency spaced – all time (combo-pattern) pilot allocation in OFDM signal by 1dB to 3dB as compared with lattice and block type pilot allocation strategies. Experimental results show that the proposed approach can be successfully used to detect OFDM signal blindly in cognitive radio with 0 % false alarm rate. [Niu QQ, Kim JY, Chaudhry A, Min SH, Choi SH, Kim HJ, Kim CJ. Robust OFDM Signal Detection Technique Based on Auto-Correlation and Kurtosis. Life Sci J 2013;10(1):1589-1598] (ISSN: 1097-8135). http://www.lifesciencesite.com. 234