Performance Analysis of Distributed Incremental LMS Algorithm in Non-stationary Environment

In this paper, we analyse the transient and steady-state performance of the distributed incremental least mean-square (DILMS) algorithm. theoretical relations which explain how transient and steady-state performance of DILMS algorithm (in terms of mean-square error (MSE)) is affected by non-stationary environment will be achieved. As the results show, theoretical and simulations are very close together. simulation results show that step size versus the error index curves of the non-stationary environment are not linear, and have an optimum step size for minimize the error

Keywords:

Adaptive filter , Distributed estimation , Energy conservation , Distributed Incremental LMS algorithm