Block Channel Analysis and Adaption Approach for Distributed Blind Equalization over Sensor Networks
Sulin Chi, Tetsuya Shimamura · 2024
In the study of signal estimation, the performance of distributed blind equalization over wireless sensor network (WSN) is significantly affected by channel conditions. However, there has been few studies investigating the effect of channel conditions on distributed blind equalization over WSN. To address this problem, this paper considers that different channel and noise variance conditions for distributed blind equalization. Initially, a smaller step-size is used to mitigate the noise enhancement effect that is caused by a large stepsize under ill-channel conditions. Then, a block channel analysis and adaption approach is proposed to analyze the performance of distributed blind equalization under current condition of channel. This approach determines whether to adjust a smaller step-size to a large value and re-designs the weight for each sensor by analyzing the corresponding channel conditions repeated over $\mathcal{Q}$ input data samples of equalizer. In the computer simulations, the average mean square error (MSE) is used to assess the performance of distributed blind equalization. Compared to the conventional method, the proposed method demonstrates improved performance for the distributed blind equalization.