Steady-State Performance Analysis of Incremental Variable Tap-Length Algorithm in WSNs Under Noisy Links condition

Abstract

Recently proposed distributed incremental fractional tap-length (FT) variable-length least mean square (LMS) technique do not consider noisy links errors, which occur during the communication of local estimations between nodes. In this paper, we study the noisy links effect on the performance of this algorithm. We derive a mathematical formulation for the steady-state length at each node. Our derived relationship shows how steady-state tap-length is affected by noisy links. Simulations confirm that there is a good match between the theory and simulated tap-length. Furthermore, the critical result is that, as the noise level increases, the steady-state tap-length decreases compared to the ideal link version. However, in low noise conditions, this length is still larger than the optimal filter length.