Block Adjustment Approach of Distributed Blind Estimation in Wireless Sensor Network

Distributed estimation in wireless sensor network (WSN) is a crucial application of in-network processing for unknown parameter estimation, which has received much attention in both science and engineering. However, the training signal is required to be known at the receiver when the unknown parameter of interest is estimated. In most practical applications, it is challenging to obtain the knowledge of the training signal in advance. Distributed blind equalizer is developed to estimate the transmitted data signal without the knowledge of the training signal, which is essential in many practical applications. However, the performance of distributed blind equalization is easily susceptible to transmission channels, especially when a large step-size is employed, which leads to noise enhancement. In this paper, a smaller initial step-size is used to mitigate the impact of noise enhancement. Furthermore, a block adjustment approach is proposed, based on the analysis of estimation error of distributed blind equalization every <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$M$</tex> data samples, to improve the prediction accurate of distributed blind equalization under different channels. Furthermore, we also analyze the effect of noise enhancement on distributed blind equalization for different data samples. Finally, the average mean-square error (MSE) is used to evaluate the performance of distributed blind equalization. Computer simulations demonstrate that the proposed approach provides better performance compared with the conventional method.