Cooperative Multi-Cell MIMO Downlink Precoding With Finite-Alphabet Inputs
This work investigates the design of linear precoders in cooperative multi-cell MIMO downlink coverage for finite-alphabet source signals. Ancient design of multi-cell MIMO downlink precoder depends on Gaussian input assumption, that could result in performance loss when true data inputs consist of discrete non-Gaussian symbols. This work presents optimized precoders for finite-alphabet input by maximizing the total rate underneath per-base station power constraints. Specifically, we propose 2 distributed algorithms: a finite-alphabet signal Gaussian interference gradient projection algorithm and a block diagonalization alternating optimization algorithm while supporting interference cancellation. Our numerical results demonstrate considerable performance gain in terms of approximate transmission knowledge rate as well as decoded bit error rate over precoding schemes designed by using the non-realistic Gaussian input assumption.
Did you like this research project?
To get this research project Guidelines, Training and Code... Click Here