Title: Sum-rate maximisation for two-way AF MIMO relaying networks and random antenna selection

Authors: Rahima Sayeb; Noureddine Hamdi

Addresses: Communication Systems Lab, ENIT, Manar University, BP 37, LE BELVEDERE 1002 Tunis, Tunisia ' Department of Computer Science and Mathematics, INSAT, Carthage University, Centre Urbain Nord, BP 676, 1080 Tunis, Tunisia

Abstract: This paper considers a multiple-input multiple-output (MIMO) two-way relaying network, where two-source nodes exchange data via a half-duplex amplify-and-forward (AF) relay. In this paper, we propose an efficient random antenna selection strategy based on maximising the sum-rate and minimising the power consumption. The proposed strategy involves choosing the best transmit and receive antennas at the two sources based on the maximisation of the overall sum-rate. The key design criterion is to reduce transmit radio frequency chains in order to minimise the energy consumption of the system. In the first step, transmit and receive antennas will be selected randomly. In the second step, signal-to-noise ratios (SNRs) of randomly chosen antennas are compared with a predefined threshold γth. If the SNRs of the selected antennas are higher than the predefined threshold, these antennas will be chosen for the communication. Else another random selection is used. Also, we show that for the problem of sum-rate maximisation related to the introduced system, the optimal relay matrix have a general beamforming structure. Simulations demonstrate the performance of the proposed algorithm.

Keywords: multiple-input multiple-output; MIMO relaying networks; cooperation; AF models; two-way relaying; TWR; sum-rate maximisation; random antenna selection; energy consumption; signal-to-noise ratio; SNR; beamforming structure; simulation.

DOI: 10.1504/IJCNDS.2016.073407

International Journal of Communication Networks and Distributed Systems, 2016 Vol.16 No.1, pp.16 - 28

Accepted: 06 Jul 2015
Published online: 01 Dec 2015 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article