Optimized Cooperative Spectrum Sensing Algorithms in Cognitive Radio Networks

Spectrum sensing in cognitive radio (CR) is a basic requirement to protect the primary user (PU) from bad interference. However, sensing has two main challenges. Firstly, the CR is required to sense the presence of the PU under very low signal-to-noise ratio (SNR) that prolongs the sensing time. Secondly, some CR nodes may suffer from deep fading and shadowing effects, which may produce false sensing results. Cooperative spectrum sensing (CSS) is supposed to overcome these challenges but at the expense of extra energy consumption due to sensing and reporting the results to the fusion center (FC). This energy is dependent on the sensing time of CR nodes, the selection of the FC and the number of CR sensing nodes. In this paper, we formulate three optimization problems, and we propose three independent heuristic algorithms each of which is capable of solving one of the optimization problems by specifying the FC, the cooperative sensing nodes, the sensing time and detection threshold for the sensing nodes. The optimization objective in these problems is minimizing sensing and reporting energy, minimizing sensing time or maximizing throughput under the assumption of different SNR value between the PU and CR nodes. Simulation results show that our approach achieves better performance than other existing approaches in terms of total sensing time, total sensing energy and achieved throughput.