Title :
Collaborative Quickest Spectrum Sensing via Random Broadcast in Cognitive Radio Systems
Author :
Li, Hushen ; Dai, Huaiyu ; Li, Chengzhi
Author_Institution :
Dept. of Electr. Eng. & Comput. Sci., Univ. of Tennessee, Knoxville, TN, USA
fDate :
7/1/2010 12:00:00 AM
Abstract :
Quickest detection is applied in spectrum sensing in cognitive radio systems when multiple secondary users collaborate with limited communication time slots. When the transmissions of sensing results are not coordinated to avoid confliction, random broadcast is used to exchange information. A necessary condition for the optimal broadcast probability, as a function of the log likelihood ratio of local observation, is obtained using variational analysis. To alleviate the difficulty of computing the optimal broadcast probability, a simple threshold broadcast scheme is proposed. Simulation shows that the proposed threshold broadcast scheme can achieve substantial performance gain (less than 60% in detection delay for the same false alarm rate) over schemes of random broadcast without regulation and single-user spectrum sensing.
Keywords :
Cognitive radio; Collaboration; Computational modeling; Delay; Frequency; Licenses; Performance gain; Radio broadcasting; Robustness; Wireless communication; Cognitive radio; quickest detection; random broadcast; spectrum sensing;
Journal_Title :
Wireless Communications, IEEE Transactions on
DOI :
10.1109/TWC.2010.07.091667