Sum-Rate Maximization with Minimum Power Consumption for MIMO DF Two-Way Relaying: Part II - Network Optimization

TL;DR

This paper develops a network-level optimal power allocation algorithm for MIMO decode-and-forward two-way relaying, maximizing sum-rate while minimizing power consumption, considering asymmetry effects and complex network scenarios.

Contribution

It introduces a convex optimization-based algorithm for sum-rate maximization with minimal power in MIMO TWR, extending prior relay optimization to network-wide optimization.

Findings

Optimal power allocation can be achieved through convex reformulation in most cases.

Asymmetry in antennas, power limits, and channels negatively impacts sum-rate and efficiency.

Simulation confirms the effectiveness of the proposed algorithm and highlights asymmetry effects.

Abstract

In Part II of this two-part paper, a sum-rate-maximizing power allocation with minimum power consumption is found for multiple-input multiple-output (MIMO) decode-and-forward (DF) two-way relaying (TWR) in a network optimization scenario. In this scenario, the relay and the source nodes jointly optimize their power allocation strategies to achieve network optimality. Unlike the relay optimization scenario considered in part I which features low complexity but does not achieve network optimality, the network-level optimal power allocation can be achieved in the network optimization scenario at the cost of higher complexity. The network optimization problem is considered in two cases each with several subcases. It is shown that the considered problem, which is originally nonconvex, can be transferred into different convex problems for all but two subcases. For the remaining two subcases, one for each case, it is proved that the optimal strategies for the source nodes and the relay must satisfy certain properties. Based on these properties, an algorithm is proposed for finding the optimal solution. The effect of asymmetry in the number of antennas, power limits, and channel statistics is also considered. Such asymmetry is shown to have a negative effect on both the achievable sum-rate and the power allocation efficiency in MIMO DF TWR. Simulation results demonstrate the performance of the proposed algorithm and the effect of asymmetry in the system.