Abstract
It is known that superposition signaling in Gaussian interference networks is capable of improving the achievable rate region. However, the problem of maximizing the rate gain offered by superposition signaling is computationally prohibitive, even in the simplest case of two-user single-input single-output interference networks. This paper examines superposition signaling for the general multiple-input multiple-output broadcast Gaussian interference networks. The problem of maximizing either the sum rate or the minimal user’s rate under superposition signaling and dirty paper coding is solved by a computationally efficient path-following procedure, which requires only a convex quadratic program for each iteration but ensures convergence at least to a locally optimal solution. Numerical results demonstrate the substantial performance advantage of the proposed approach.
Original language | English |
---|---|
Pages (from-to) | 4646-4654 |
Journal | IEEE Transactions on Communications |
Volume | 65 |
Issue number | 11 |
Early online date | 24 Jul 2017 |
DOIs | |
Publication status | Published - 15 Nov 2017 |