Exponential Hyperbolic Cosine Robust Adaptive Diffusion Strategy for Distributed Estimation

This paper presents a robust adaptive diffusion algorithm based on exponential hyperbolic cosine cost function to enhance the distributed estimation performance. Mathematical analysis shows the convergence criteria of the algorithm. Performance analysis of the proposed algorithm is carried out in terms of convergence speed and steady-state expected mean square error (EMSE), and also the performance is compared with other diffusion algorithms called d-LMS and d-NMVC. The robustness of the proposed algorithm is shown by simulation results for Gaussian noisy environment as well as for different non-Gaussian noisy environments.