DESIGN AND ANALYSIS OF DISTRIBUTED COOPERATIVE ALGORITHMS FOR OPTIMISING THE PERFORMANCE OF FUTURE WIRELESS NETWORKS

  • Mohammed Bahbahani

Student thesis: Phd

Abstract

Cooperative communication networks, in which nodes share their resources to maximise the overall network performance, have recently attracted significant interests. Traditionally, the nodes only collaborated in multi-hop transmission at the network (NET) layer. However, cooperative diversity at the physical (PHY) layer can achieve capacity gains from the already wasted power in broadcast channels. Meanwhile, energy-efficient designs that enable perpetual network operation are crucial for future wireless applications, such as the Internet of Things (IoT). Hence, the thesis in hand concerns the design and analysis of cooperative techniques to enhance performance and lifetime of future wireless networks while addressing the practical aspects, such as algorithm complexity, decentralisation and cooperation incentives. To fulfil the above, we apply game theory to model the actions of autonomous nodes as they compete or cooperate in maximising their utilities, and in turn, the aggregate network performance. Our research includes a comprehensive analysis of cooperation at the various layers of the communication stack. At the physical layer, we solve the relay selection problem in a one-to-many decode-and-forward relay channel as a political coalition formation game. This algorithm is shown to yield a stable set of relays that achieves near-optimal sum-rate performance while allowing for a trade-off between performance and complexity through the formation of parties. In the next problem, the power-level selection in amplify-and-forward energy harvesting relay networks is approached via a repeated Bayesian Stackelberg game, in which the source (seller) and the relays (buyers) iteratively negotiate the price that maximises their utilities. The scheme only requires a statistical knowledge of channel, energy and data side information. The third problem concerns a cross-layer cooperative protocol, which integrates duty-cycling, clustering, energy harvesting and cooperative diversity to enhance the throughput while maintaining perpetual operation. In addition, a cooperative medium access control (MAC) protocol is designed, in which the relay node forwards the source’s packet during a designated sub-slot. The optimal cluster size is also optimised based on the application requirements. The protocol is evaluated and compared against other benchmark schemes using the realistic network simulator OMNET++. Finally, we present a novel cooperative application-layer protocol, in which smartphone users in a crowded stadium coordinate their cellular network usage to improve their average quality of experience (QoE). Besides, a cluster-based mesh network is designed using the phones’ Peer-to-Peer (P2P) connectivity to exchange common match-specific application data. Additionally, the selfish behaviour of users is modelled as a finitely repeated game, where each stage game is played with symmetric near-Nash strategies. Simulations performed in MATLAB confirm the improvement in QoE achieved by the proposed scheme, particularly when it is adopted by more than half of the fans.
Date of Award31 Dec 2018
Original languageEnglish
Awarding Institution
  • The University of Manchester
SupervisorKhairi Hamdi (Supervisor) & Emad Alsusa (Supervisor)

Keywords

  • cooperative communication
  • relay network
  • energy harvesting
  • game theory
  • stadium connectivity

Cite this

'