Joint cumulant estimate correction and decision for cooperative modulation classification by using multiple sensors

In this paper, several data fusion and soft decision fusion methods are proposed for cooperative automatic modulation classification (AMC) by using multiple sensors. A well-known AMC using fourth-order cumulant is considered. Also, a novel joint cumulant estimate correction is proposed. The AMC performance of proposed fusion methods with and without joint estimate correction is evaluated through the extensive Monte Carlo trials for different multipath fading channels, sensor spatial distributions, and application scenarios, and compared to those of the existing methods. Numerical results have confirmed huge performance improvement achievable with here proposed methods, especially in the non-idealized application scenarios.