A bias removal blind modulation classification algorithm for MIMO channel

With the development of wireless communication, modulation classification (MC) has wide applications both in military and civilian communication, such as anti-interference, adaptive modulation and spectrum management. This paper proposes a bias removal blind MC algorithm (B-MCA) based on the preprocessing of the mixtures, the bias removal equivariant adaptive separation via independence (EASI) algorithm, the bias cancellation of the estimation signals and the high order cumulants (HOC) based classification algorithm. The advantage of the proposed algorithm B-MCA is that, by considering the actual communication scenarios, B-MCA not only can realize the MC of the mixtures in the multiple-input-multiple-out (MIMO) channel, but also has a good recognition rate in the low signal-to-noise ratio (SNR) case. Simulation results show that B-MCA has good performance of convergence rate and classification accuracy.