Separation of complex signals with known source distributions in time-varying channels using optimum complex block adaptive ICA

This paper presents a novel realization of the complex block adaptive independent component analysis algorithm. The algorithm optimally updates the real and imaginary components of the weight vector independently. The new implementation is employed for the separation of complex signals with known source distributions, a scenario frequently encountered in practice. Under time-varying channel conditions, the performance of the proposed method is compared with the widely known Complex Fast-ICA. Simulation results show that this new technique exhibits superior performance in time varying channel conditions in terms of convergence speed. In addition, the performance of the proposed method is independent of the processing block length and is achieved without any additional cost in computational complexity.