Block-Variable-Adaptive Approach for Distributed Blind Equalization over Wireless Sensor Network

Recently, the distributed blind equalization in wireless sensor network (WSN) is commonly employed for signal estimation. However, the performance of distributed blind equalization is highly sensitive to the condition of channel, especially for ill transmission channel condition, leading to inaccuracies in predicting the transmitted data signal. In order to mitigate the impact of the ill transmission channel condition on others, a block-variable-adaptive approach is proposed to redesign the weight for each sensor node and adjust the step-size for distributed blind equalization under the current condition of channel based on the analysis of eigenvalue spread of received signal at each sensor node every $\mathcal{K}$ data samples. In the computer simulations, we use the mean-square error to access the effectiveness of distributed blind equalization. Compared with the conventional approach, the proposed approach demonstrates a lower MSE and faster convergence.