Robust Transmissions in Wireless Powered Multi-Relay Networks with Chance Interference Constraints

TL;DR

This paper develops a joint optimization framework for signal beamforming and energy harvesting in wireless powered multi-relay networks, balancing throughput maximization with probabilistic interference constraints.

Contribution

It introduces a novel joint optimization approach for beamforming and energy harvesting strategies under chance interference constraints in wireless powered relay networks.

Findings

Proposed algorithms achieve near-optimal throughput and energy efficiency.

Effective handling of non-convex optimization problems via polyblock approximation.

Enhanced network performance through joint signal and energy cooperation.

Abstract

In this paper, we consider a wireless powered multi-relay network in which a multi-antenna hybrid access point underlaying a cellular system transmits information to distant receivers. Multiple relays capable of energy harvesting are deployed in the network to assist the information transmission. The hybrid access point can wirelessly supply energy to the relays, achieving multi-user gains from signal and energy cooperation. We propose a joint optimization for signal beamforming of the hybrid access point as well as wireless energy harvesting and collaborative beamforming strategies of the relays. The objective is to maximize network throughput subject to probabilistic interference constraints at the cellular user equipment. We formulate the throughput maximization with both the time-switching and power-splitting schemes, which impose very different couplings between the operating parameters for wireless power and information transfer. Although the optimization problems are inherently non-convex, they share similar structural properties that can be leveraged for efficient algorithm design. In particular, by exploiting monotonicity in the throughput, we maximize it iteratively via customized polyblock approximation with reduced complexity. The numerical results show that the proposed algorithms can achieve close to optimal performance in terms of the energy efficiency and throughput.