Optimal Multinode Sensing in a Malicious Cognitive Radio Network

Abstract

Spectrum sensing is an essential function in cognitive radio systems for dynamic spectrum access. Multinode sensing is a technique being used in cognitive radio networks to enhance the sensing performance using space diversity concept. The challenges in multinode spectrum sensing are the prediction of signal status in multiple frequency bands in a low signal-to-noise ratio (SNR) regime and sensing reliability. The weighted gain combining (WGC) and the equal gain combining are the two soft decision cooperative sensing techniques being used frequently in literature. In this paper, we introduce weighted gain cooperative sensing using differential evolution (DE) and adjusted box-plot methods to exalt the sensing reliability together with the sensing performance. The main advantage of the WGC method using DE is that it can generate optimal weights independent of received signal characteristics, which is an indispensable condition to realize the system in real time. The proposed optimal cooperative sensing method with entropy and cyclic features enhances the sensing performance, and it is less severe to noise uncertainties compared with the traditional sensing methods. It can detect the low SNR signals up to $-$ 24 dB at desired sensing performance ( $P_{f} = 0.1$ and $P_{d} = 0.9$ ) with a frame size of 256 and using five nodes in cooperation. It is a significant improvement for IEEE 802.22 WRAN systems, which work under low SNR regime.