Distributed Blind Equalization in Networked Systems.

Ying Liu,Yunlong Cai
DOI: https://doi.org/10.1109/icassp.2017.7952967
2017-01-01
Abstract:In this paper, we study the problem of distributed blind equalization in single-input multi-output (SIMO) systems, wherein the channels of networked systems share some similarities. This corresponds to a multi-task optimization problem. To tackle this problem, an adaptive distributed generalized Sato algorithm (d-GSA) using the diffusion cooperation rule is proposed. In the proposed d-GSA, only the scalar of equalizer output is combined and transmitted among neighbors, which significantly reduces the cost of computation and communication. The performance of d-GSA is analyzed theoretically and verified by numerical simulations. Results show that the d-GSA outperforms the corresponding non-cooperative GSA.
What problem does this paper attempt to address?