Set-Membership Adaptive Soft Combining for Distributed Cooperative Spectrum Sensing

This paper presents a new adaptive soft combiner implemented on cognitive radio nodes for distributed cooperative spectrum sensing. The key feature of the proposal is the use of the standard set-membership normalized least-mean squares (SMNLMS) algorithm, which enjoys good tracking capabilities, while entailing low computational burden. Indeed, simulation results show that the computational savings can be as high as 37% for 8 cooperating secondary users, as compared to another previously proposed data-selective soft combiner from the literature.