Faster-than-nyquist signaling by sum-of-absolute-values optimization

Abstract

In this article, we propose a new signal detection method for faster-than-Nyquist signaling (FTNS) systems for high-speed data communications. The symbol detection problem is formulated as under-determined linear equations on a finite set. We transform the problem to the sum-of-absolute-values (SOAV) optimization, for which we derive a fast algorithm based on proximal splitting. The effectiveness of the proposed method is shown by simulations.