Iterative Mode-Dropping for the Sum Capacity of MIMO-MAC with Per-Antenna Power Constraint

TL;DR

This paper introduces an iterative mode-dropping algorithm that efficiently optimizes input signals to maximize the sum capacity of MIMO-MAC systems under per-antenna power constraints, demonstrating fast convergence and capacity differences.

Contribution

The paper presents a novel iterative mode-dropping algorithm specifically designed for MIMO-MAC with per-antenna power constraints, improving capacity optimization methods.

Findings

Algorithm converges quickly in simulations

Capacity under per-antenna constraints differs from sum power constraints

Effective optimization of input covariance matrices

Abstract

We propose an iterative mode-dropping algorithm that optimizes input signals to achieve the sum capacity of the MIMO-MAC with per-antenna power constraint. The algorithm successively optimizes each user's input covariance matrix by applying mode-dropping to the equivalent single-user MIMO rate maximization problem. Both analysis and simulation show fast convergence. We then use the algorithm to briefly highlight the difference in MIMO-MAC capacities under sum and per-antenna power constraints.