A cooperative spectrum sensing scheme using adaptive fuzzy system for cognitive radio networks

Cognitive radio represents a promising design paradigm for the next-generation of wireless networks that can enhance utilization of the scarce spectrum available. It allows dynamic spectrum access over temporarily unused spectrum holes that are identified by spectrum sensing. In order to enhance the sensing accuracy by taking advantage of the spatial diversity in multiple cognitive radio user networks, cooperative spectrum sensing has been proposed in the literature. However, most existing cooperative spectrum sensing schemes require knowledge of the noise variance and signal-to-noise ratio of primary signal at each node, which can add a significant transmission overhead. This paper proposes an adaptive fuzzy system-based scheme for cooperative spectrum sensing without the need for a priori network information such as the noise variance, channel state information, and characteristics of existing primary signals. Energy observations at nodes are used for online learning from the environment to make a reliable global sensing decision at a central node without further information. Simulation results show that the proposed scheme outperforms the equal-gain combination based scheme, and matches the optimal soft combination scheme in terms of sensing accuracy.