Generalized Maximum Correntropy Algorithm with Affine Projection Based on Sigmoid Cost Function

The affine projection algorithm based on generalized maximum correlation entropy is widely used due to its high robustness. However, it is difficult to be applied in the field of echo cancellation because of its poor robustness performance in sparse systems. In this paper, in order to improve the robustness of such algorithms in sparse systems, the sigmoid generalized maximum correlation entropy affine projection algorithm (SAPGMC) is proposed by invoking the improved sigmoid activation function as the cost function, which reduces the influence of impulse interference in sparse systems, and at the same time, a l1J-paradigm penalty term is added to the algorithm, so that the algorithm can be successfully applied to the field of echo cancellation. The simulation results demonstrate the superiority of the algorithm's performance in sparse systems.