Edinburgh Napier University

  At the heart of the Internet of Things (IoTs) are the Low-power and Lossy networks (LLNs), a collection of interconnected battery-operated and resource-constrained tiny devices that enable the realization of a wide range of applications in multiple domains. For an efficient operation, such networks require the design of efficient protocols especially at the network layer of their communication stack. In this regards, the Routing Protocol for LLNs (RPL) has been developed and standardised by the IETF to fulfil the routing requirements in such networks. Proven efficient in tackling some major issues, RPL is still far from being optimal in addressing several other routing gaps in the context of LLNs. For instance, the RPL standard lacks in a scalable routing mechanism in the applications that require bidirectional communication. In addition, its routing maintenance mechanism suffers from relatively slow convergence time, limiting the applicability of the protocol in time-critical applications, and a high risk of incorrect configurations of its parameters, risking the creation of sub-optimal routes. Furthermore, RPL lacks in a fair load-distribution mechanism which may harm both energy and reliability of its networks. Motivated by the above-mentioned issues, this thesis aimed at overcoming the RPL’s weaknesses by developing more efficient routing solutions, paving the way towards successful deployments and operations of the LLNs at different scales. Hence, to tackle the inefficiency of RPL’s routing maintenance operations, a new routing maintenance algorithm, namely, Drizzle, has been developed characterized by an adaptive, robust and configurable nature that boosts the applicability of RPL in several applications. To address the scalability problem, a new downward routing solution has been developed rendering RPL more efficient in large-scale networks. Finally, a load-balancing objective function for RPL has been proposed that enhances both the energy efficiency and reliability of LLNs. The efficiency of the proposed solutions has been validated through extensive simulation experiments under different scenarios and operation conditions demonstrating significant performance enhancements in terms of convergence time, scalability, reliability, and power consumption.