Robust Non-parametric Sparse Distributed Regression Over Wireless Networks

Abstract The classical distributed sparse regression based on least square error is sensitive to the outliers in the desired data. In this manuscript, we consider the rank based estimator named minimum Wilcoxon norm for developing robust non-parametric sparse regression over distributed adaptive networks. The convergence analysis of the proposed algorithm is analyzed using asymptotic linearity of rank test. Exhaustive simulation studies show that the proposed methods are robust against outliers in the desired data and exploits sparsity, hence performs better than the existing methods if the parameter of interest is sparse. The proposed algorithms are validated for three different applications namely distributed parameter estimation, tracking and distributed power spectrum estimation.