Asymptotically maximum likelihood detection for subcarrier mapping‐induced cyclostationary signatures in cognitive radio networks. (15th December 2011)

• Record Type: Journal Article
• Authors: Jiang, Yonglei; Hu, Honglin
• Is Part Of: International journal of communication systems. Volume 26:Number 8(2013:Aug.)
• Page Start: 1011
• Page End: 1027
• Publication Date: 2011-12-15
• Subjects: cognitive radio -- network coordination -- spectrum sensing -- cyclostationary signature -- subcarrier mapping -- maximum likelihood detection -- doubly selective fading; Telecommunication systems -- Periodicals; 621.382
• DOI: 10.1002/dac.1385
• Languages: English
• ISSNs: 1074-5351

Abstract

SUMMARY: In cognitive radio networks based on orthogonal frequency division multiplexing, the subcarrier mapping‐induced cyclostationary signatures are utilized for robust network coordination. However, the effect of network coordination depends on the reliable detection of such cyclostationary signatures. Moreover, doubly selective fading increases the difficulty of reliable detection. In this paper, we first derive an exact covariance matrix to characterize the statistics of the cyclostationary signatures induced by subcarrier mapping. Using the exact covariance matrix rather than its estimation adopted in conventional algorithms, we propose an asymptotically maximum likelihood detection algorithm. The detection performance of the proposed asymptotically maximum likelihood detection algorithm is analyzed and evaluated both in flat fading and doubly selective fading channels. Simulation results show that the proposed algorithm outperforms the conventional low‐complexity single‐cycle signature detection algorithm in terms of less observation period or smaller mapping set size. Copyright © 2011 John Wiley & Sons, Ltd. Abstract : The effect of cyclostationary signature‐based network coordination in cognitive radio networks depends on the reliable detection of such cyclostationary signatures. In this paper, we derive an exact covariance matrix to characterize the cyclostationary signatures induced by subcarrier mapping. Using the exact covariance matrix rather than its estimation, we propose an asymptotically maximum likelihood detection algorithm. Detection performance is analyzed and evaluated both in flat and doubly selective fading channels. Simulation results indicate that the proposed detector provides significant performance improvement.