An improved proportionate NLMS algorithm based on the l0 norm

The proportionate normalized least-mean-square (PNLMS) algorithm was developed in the context of network echo cancellation. It has been proven to be efficient when the echo path is sparse, which is not always the case in real-world echo cancellation. The improved PNLMS (IPNLMS) algorithm is less sensitive to the sparseness character of the echo path. This algorithm uses the l
1
norm to exploit sparseness of the impulse response that needs to be identified. In this paper, we propose an IPNLMS algorithm based on the l
0
norm, which represents a better measure of sparseness than the l
1
norm. Simulation results prove that the proposed algorithm outperforms the original IPNLMS algorithm.