A Survey of Using Sign Function to Improve The Performance of EASI Algorithm

Blind signal processing (BSP) is attractive in the signal processing. Blind source separation (BSS) is one of the main components in blind signal processing. In this paper, the basic model and mathematical principle of BSS was presented. One of BSS algorithms, EASI (equivariant adaptive separation via independence), was reviewed. Then a new algorithm based on EASI was proposed in order to reduce complexity of algorithm computation. The results of simulation experiment prove the superior performance than conventional EASI both in convergence rate and computation complexity.