Performance Analysis of Transversal Filter Combination for Feedback Suppression in Hearing Aids

This paper investigates the mean square performance of two transversal adaptive filters for the feedback cancellation model in digital hearing aids. This convex combination allows the update has to be performed independently with their errors generated during the adaptation process. The Gradient descent stochastic algorithm is considered to minimize the overall system error with the help of the mixing parameter $\delta(n)$. The complete analytical expressions of the combination have been derived and validated with the simulation results in the stationary environment. The proposed combination scheme shows better feedback cancellation performance, as illustrated in the results.