Joint Scheduling and Duty Cycle Control Framework for Hierarchical Machine-to-Machine Communication Networks.

Abstract

This thesis presents a novel distributed optimisation framework for machine-tomachine (M2M) communication networks with dynamic traffic generation, heterogeneous applications and different device capabilities. The aim of the framework is to effectively manage the massive access of energy constrained M2M devices while satisfying different application requirements. The proposed framework has three control blocks which run at cluster heads and M2M gateways: i) The distributed duty cycle control that adapts to dynamic network traffics for IEEE 802.15.4 MAC layer protocol with stop-and-wait automatic repeat request (ARQ) and Go-Back-N ARQ schemes. ii) The cluster head control that applies dynamic programming (DP) and approximate dynamic programming (ADP) techniques to maximise single cluster utility while balancing the tradeoff between system performance and algorithm complexity. iii) The gateway control that applies network utility optimisation (NUM) and mixed integer programming (MIP) techniques to maximise the aggregated long-term network utility while satisfying different application requirements among clusters. Both theoretical and practical concerns are addressed by the proposed control framework. Simulation results show that the proposed framework effectively improve the overall network performance in terms of network throughput, energy efficiency, end-to-end delay and packet drop ratio.