Joint resource allocation in SWIPT-based multi-antenna decode-and-forward relay networks

TL;DR

This paper investigates joint resource allocation in a SWIPT-based multi-antenna relay network, proposing optimal and low-complexity antenna clustering schemes to maximize end-to-end data rate.

Contribution

It introduces a novel joint optimization framework for PS ratios and power allocation in multi-antenna relays, including a practical antenna clustering method with linear complexity.

Findings

Closed-form optimal solution for PS ratios and power allocation.

A greedy clustering algorithm with near-optimal performance.

Insights into the trade-offs between hardware cost and system performance.

Abstract

In this paper, we consider relay-assisted simultaneous wireless information and power transfer (SWIPT) for two-hop cooperative transmission, where a half-duplex multi-antenna relay adopts decode-and-forward (DF) relaying strategy for information forwarding. The relay is assumed to be energy-free and needs to harvest energy from the source node. By embedding power splitting (PS) at each relay antenna to coordinate the received energy and information, joint problem of determining PS ratios and power allocation at the multi-antenna relay node is formulated to maximize the end-to-end achievable rate. We show that the multi-antenna relay is equivalent to a virtual single-antenna relay in such a SWIPT system, and the problem is optimally solved with closed-form. To reduce the hardware cost of the PS scheme, we further propose the antenna clustering scheme, where the multiple antennas at the relay are partitioned into two disjoint groups which are exclusively used for information decoding and energy harvesting, respectively. Optimal clustering algorithm is first proposed but with exponential complexity. Then a greedy clustering algorithms is introduced with linear complexity and approaching to the optimal performance. Several valuable insights are provided via theoretical analysis and simulation results.