An Alternation Diffusion LMS Estimation Strategy over Wireless Sensor Network

This paper presents a distributed estimation strategy called alternation diffusion LMS estimation (AD-LMS) to estimate an unknown parameter of interests from noisy measurement over wireless sensor network. It is useful in the wireless sensor networks where robustness and low consumption are desired features. Diffusion LMS is introduced in this estimation strategy to improve the performance and reduce the communication burden. With the proposed strategy, whether each node distributes its estimation depends on an alternative parameter. The node only exchanges its estimation when the instant time meets some conditions. Next, each node combines the estimations of neighbors with its own estimation using combination coefficients upon the topology of the network. At last, the nodes update their estimations with a normalized LMS algorithm. The proposed AD-LMS strategy is compared to standard diffusion strategy. The results show that they achieve exactly the same coverage rate and nearly the network performance (network MSD and steady-state MSD) of standard diffusion strategy while reducing the communication burden significantly.