An improved variable step size LMS adaptive filtering algorithm and its analysis

Aiming at the shortcoming of the former algorithms that the performance is not very good when signal to Interference ratio (SIR) appeared on the primary input is low, this paper proposes an improved algorithm of variable step size LMS adaptive filtering algorithm, establishes a new non-linear functional relationship between step factor and system output error and analyzes the affection of interference to the step factor in the improved algorithm. The theoretical analysis indicates that in the condition of low SIR, the improved algorithm is insensitive to the interference and has preferable convergence performance over the former algorithms. Computer simulation results confirm the theoretical analysis.