Robust general Kalman filter for echo cancellation

The Kalman filter is widely used in many practical applications. In this paper, we motivate the use of a more general form of the Kalman filter in the context of echo cancellation. This algorithm is derived by considering, at each iteration, a block of time samples instead of one time sample as it is the case in the conventional approach. The variance of the near-end signal explicitly appears within the algorithm, which allows us to better control it. Using a proper estimation of this parameter, the general Kalman filter is robust to large near-end signals (like double-talk). Simulation results indicate the appealing performance of the proposed algorithm.