On low complexity cooperative spectrum sensing for cognitive networks

This paper presents a practical system design approach for cooperative spectrum sensing in cognitive sensor networks. An optimization problem is formulated, where the objective is to choose appropriate number of samples used in local energy calculation and linear combination weights for a global fusion center that together maximize global spectrum detection probability. Depending on the local information available to the fusion center and secondary users, practical system design is proposed in high fusion signal to noise ratio (SNR) regime, which has minimal implementation complexity and negligible performance loss, thus provides an efficient system design alternative in practice. Simulation results are presented to verify the analytical results.