Abstract: To improve the spectrum sensing performance, this paper proposes cooperative spectrum sensing that fuses the sensing information from every cognitive radio with soft fusion based on weighting. To reduce interference to the primary user and improve throughput of cognitive radio, two kinds of weight selection
methods that minimize interference and maximize throughput are proposed. As large weighting is allocated to the cognitive radios with high SNR or low channel gain to the coordinator, proportion of the sensing information sent by the cognitive radio with high SNR to the fusion information at the coordinator is increased. Lost information due to channel fading is compensated. Simulation results show that performance of the algorithm is better than those based on SNR weighting and those without weighting. The proposed algorithm can also reduce influences of the fading channel on detection performance. The results show that, by selecting appropriate weights, interference can be reduced and throughput increased.

Key words: cooperative spectrum sensing, energy detection, soft fusion, cognitive radio