An Improved NLMS Algorithm for Interference Cancellation

The problem of transceiver isolation of jammer on small platform will seriously affect its reconnaissance and receiving at the same time. Assuming that the coupling path of interference signals is sparse, the commonly used system identification least mean square (LMS) algorithm is not accurate enough to identify the system transfer function and achieve an ideal isolation effect when solving the problem of transceiver isolation. In order to solve this problem, many LMS algorithms based on sparse constraints have been proposed. On this basis, this paper makes use of the advantages of variable step size method for system identification, and combines it with sparse constrained NLMS algorithm, a NLMS algorithm based on variable step size sparse constraint is proposed. The algorithm makes full use of the sparsity of the interference coupling path between the receiving antenna and transmitting antenna, and can accurately identify the attenuation coefficient of the coupling path. The theoretical analysis and simulation results show that this method can achieve good isolation in sparse time-invariant environment, and can solve the problem of transceiver isolation on small platforms.