Power Optimisation and Relay Selection in Cooperative Wireless Communication Networks

Abstract

Cooperative communications have emerged as a significant concept to improve reliability and throughput in wireless systems. In cooperative networks, the idea is to implement a scheme in wireless systems where the nodes can harmonize their resources thereby enhancing the network performance in different aspects such as latency, BER and throughput. As cooperation spans from the basic idea of transmit diversity achieved via MIMO techniques and the relay channel, it aims to reap somewhat multiple benefits of combating fading/burst errors, increasing throughput and reducing energy use. Another major benefit of cooperation in wireless networks is that since the concept only requires neighbouring nodes to act as virtual relay antennas, the concept evades the negative impacts of deployment costs of multiple physical antennas for network operators especially in areas where they are difficult to deploy. In cooperative communications energy efficiency and long network lifetimes are very important design issues, the focus in this work is on ad hoc and sensor network varieties where the nodes integrate sensing, processing and communication such that their cooperation capabilities are subject to power optimisation. As cooperation communications leads to trade-offs in Quality of Services and transmit power, the key design issue is power optimisation to dynamically combat channel fluctuations and achieve a net reduction of transmit power with the goal of saving battery life. Recent researches in cooperative communications focus on power optimisation achieved via power control at the PHY layer, and/or scheduling mechanism at the MAC layer. The approach for this work will be to review the power control strategy at the PHY layer, identify their associated trade-offs, and use this as a basis to propose a power control strategy that offers adaptability to channel conditions, the road to novelty in this work is a channel adaptable power control algorithm that jointly optimise power allocation, modulation strategy and relay selection. Thus, a novel relay selection method is developed and implemented to improve the performance of cooperative wireless networks in terms of energy consumption. The relay selection method revolves on selection the node with minimum distance to the source and destination. The design is valid to any wireless network setting especially Ad-hoc and sensor networks where space limitations preclude the implementation of bigger capacity battery. The thesis first investigates the design of relay selection schemes in cooperative networks and the associated protocols. Besides, modulation strategy and error correction code impact on energy consumption are investigated and the optimal solution is proposed and jointly implemented with the relay selection method. The proposed algorithm is extended to cooperative networks in which multiple nodes participate in cooperation in fixed and variable rate system. Thus, multi relay selection algorithm is proposed to improve virtual MIMO performance in terms of energy consumption. Furthermore, motivated by the trend of cell size optimisation in wireless networks, the proposed relay selection method is extended to clustered wireless networks, and jointly implemented with virtual clustering technique. The work will encompass three main stages: First, the cooperative system is designed and two major protocols Decode and Forward (DF) and amplify and forward (AF) are investigated. Second, the proposed algorithm is modelled and tested under different channel conditions with emphasis on its performance using different modulation strategies for different cooperative wireless networks. Finally, the performance of the proposed algorithm is illustrated and verified via computer simulations. Simulation results show that the distance based relay selection algorithm exhibits an improved performance in terms of energy consumption compared to the conventional cooperative schemes under different cooperative communication scenarios.