Energy-efficient and reliable routing techniques for machine-to-machine communications

Abstract

In recent years, Machine-to-Machine (M2M) communications has emerged as ubiquitous communication among large number of embedded devices. M2M communications networks are typically composed of thousands of highly resource constraint devices connected with dynamic and lossy wireless links with predominantly unpredictable characteristics, categorized as Low-Power and Lossy networks (LLNs). Routing is of paramount importance for M2M communications, as the data has to be relayed via multiple hops before it reaches destination. The Internet Engineering Task Force (IETF) has standardized the IPv6 Routing Protocol for Low Power and Lossy Networks (RPL) as predominate candidate for M2M communications. In fact, the severe resource constraints and unreliable wireless medium of LLNs have driven the routing protocols towards an energy-ecient, reliable, stable, and scalable design. Power conservation and reliable data delivery are the two main primary goals to design routing protocols. Though, it is extremely challenging to achieve both simultaneously, especially for LLNs. This thesis studies, designs, implements and evaluates the performance of energy-ecient and reliable routing techniques for M2M communications. Specifically, this thesis rst presents a review of the state-of-the-art of M2M communications with focus on the enabling wireless communication technologies, standardization activities, and envisioned research opportunities. Furthermore, with a comprehensive study on RPL, a framework design of RPL is proposed. Aiming to simultaneously achieve high reliability and energy conservation in LLNs, suits of vi Abstract routing protocols are proposed. Extensive experimental studies are conducted in NS-3 to validate the eectiveness and exibility of proposed protocols. This thesis proposes a hybrid RPL (HyRPL) aiming to provide reliable routing support with signicant reduction of routing overhead. In HyRPL, the necessity for route discovery is determined by measuring the cost of pre-established route. With the design of a route constraint, HyRPL eectively balances the trade-o between utilizing the pre-established route and on-demand route discovery in LLNs. This thesis also proposes a cluster-parent based RPL (CRPL), which eectively leverages path diversity so as to improve the reliability and reduce the number of retransmissions. With the objective to minimize the total number of transmissions, CRPL allows each node to select an optimal forwarder-set with a low-complexity algorithm. Moreover, an ecient loss recovery scheme for the detection and retransmission of lost data packets is designed to avoid duplicate transmissions. This thesis also proposes a novel energy-ecient and region-based routing protocol, called ER-RPL. In contrast to traditional routing protocols, where all the nodes are required for route discovery, ER-RPL only requires a subset of nodes. By exploiting the region feature of static M2M communication networks, ER-RPL achieves energy conservation without compromising the reliability. Both the theoretical analy...