Joint Resource Allocation and Antenna Selection In the Uplink of OFDMA Networks

TL;DR

This paper proposes an iterative optimization approach for joint resource allocation and antenna selection in uplink OFDMA networks, effectively solving a complex non-convex MINLP problem to improve network throughput.

Contribution

It introduces a novel iterative two-step method combining Hungarian algorithm and DC programming for joint sub-channel, power, and antenna optimization in uplink OFDMA networks.

Findings

Outperforms exhaustive search in simulations

Effectively handles non-convex MINLP problem

Improves network throughput

Abstract

The problem of joint sub-channel, power control and antenna selection in the uplink of OFDMA networks is investigated. The corresponding optimization problem is a non convex mixed integer non-linear problem(MINLP). To tackle the problem, an iterative two-step optimization problem is devised. First,~Having fixed the selected antenna as well as the associated power of each user,~a sub-channel assignment problem using the so-called Hungarian method is proposed. Second, the joint power and antenna selection problem is extended into a more elegant form through adding a penalty term to the objective function forcing the relaxed boolean variables associated with antenna selection problem to take binary values. To this end, the problem is restated as a difference of two concave function (DC programming) and is effectively solved in an iterative manner. The foregoing procedure is iteratively repeated until approaching a fixed network throughput. Simulation results demonstrate the superiority of the proposed method w.r.t. the exhaustive search with fixed power control mechanism.