Data Routing for Mobile Internet of Things Applications

The Internet of things (IoT) represents a new era of networking, it envisions the Internet of the future where objects or “Things” are seamlessly connected to the Internet providing various services to the community. Countless applications can benefit from these new services and some of them have already come to life especially in healthcare and smart environments. The full realization of the IoT can only be achieved by having relevant standards that enable the integration of these new services with the Internet. The IEEE 802.15.4, 6LoWPAN and IPv6 standards define the framework for wireless sensor networks (WSN) to run using limited resources but still connect to the Internet and use IP addresses. The Internet engineering task force (IETF) developed a routing protocol for low-power and lossy networks (LLN) to provide bidirectional connectivity throughout the network, this routing protocol for LLNs (RPL) was standardized in RFC6550 in 2012 making it the standard routing protocol for IoT. With all the bright features and new services that come with the futuristic IoT applications, new challenges present themselves calling for the need to address them and provide efficient approaches to manage them. One of the most crucial challenges that faces data routing is the presence of mobile nodes, it affects energy consumption, end-to-end delay, throughput, latency and packet delivery ratio (PDR). This thesis addresses mobility issues from the data routing point of view, and presents a number of enhancements to the existing protocols in both mesh-under and route-over routing approaches, along with an introduction to relevant standards and protocols, and a literature review of the state of the art in research. A dynamic cluster head election protocol (DCHEP) is proposed to improve network availability and energy efficiency for mobile WSNs under the beacon-enabled IEEE 802.15.4 standard. The proposed protocol is developed and simulated using CASTALIA/OMNET++ with a realistic radio model and node behaviour. DCHEP improves the network availability and lifetime and maintains cluster hierarchy in a proactive manner even in a mobile WSN where all the nodes including cluster heads (CHs) are mobile, this is done by dynamically switching CHs allowing nodes to act as multiple backup cluster heads (BCHs) with different priorities based on their residual energy and connectivity to other clusters. DCHEP is a flexible and scalable solution targeted for dense WSNs with random mobility. The proposed protocol achieves an average of 33% and 26% improvement to the availability and energy efficiency respectively compared with the original standard. Moving to network routing, an investigation of the use of RPL in dynamic networks is presented to provide an enhanced RPL for different applications with dynamic mobility and diverse network requirements. This implementation of RPL is designed with a new dynamic objective-function (D-OF) to improve the PDR, end-to-end delay and energy consumption while maintaining low packet overhead and loop-avoidance. A controlled reverse-trickle timer is proposed based on received signal strength identification (RSSI) readings to maintain high responsiveness with minimum overhead, and consult the objective function when a movement or inconsistency is detected to help nodes make an informed decision. Simulations are done using Cooja with different mobility scenarios for healthcare and animal tracking applications considering multi-hop routing. The results show that the proposed dynamic RPL (D-RPL) adapts to different mobility scenarios and has a higher PDR, slightly lower end-to-end delay and reasonable energy consumption compared to related existing protocols. Many recent applications require the support of mobility and an optimised approach to efficiently handle mobile nodes is essential. A game scenario is formulated where nodes compete for network resources in a selfish manner, to send their data packets to the sink node. Each node counts as a player in the noncooperative game. The optimal solution for the game is found using the unique Nash equilibrium (NE) where a node cannot improve its pay-off function while other players use their current strategy. The proposed solution aims to present a strategy to control different parameters of mobile nodes (or static nodes in a mobile environment) including transmission rate, timers and operation mode in order to optimize the performance of RPL under mobility in terms of PDR, throughput, energy consumption and end-to-end-delay. The proposed solution monitors the mobility of nodes based on RSSI readings, it also takes into account the priorities of different nodes and the current level of noise in order to select the preferred transmission rate. An optimised protocol called game-theory based mobile RPL (GTM-RPL) is implemented and tested in multiple scenarios with different network requirements for Internet of Things applications. Simulation results show that in the presence of mobility, GTM-RPL provides a flexible and adaptable solution that improves throughput whilst maintaining lower energy consumption showing more than 10% improvement compared to related work. For applications with high throughput requirements, GTM-RPL shows a significant advantage with more than 16% improvement in throughput and 20% improvement in energy consumption. Since the standardization of RPL, the volume of RPL-related research has increased exponentially and many enhancements and studies were introduced to evaluate and improve this protocol. However, most of these studies focus on simulation and have little interest in practical evaluation. Currently, six years after the standardization of RPL, it is time to put it to a practical test in real IoT applications and evaluate the feasibility of deploying and using RPL at its current state. A hands-on practical testing of RPL in different scenarios and under different conditions is presented to evaluate its efficiency in terms of packet delivery ratio (PDR), throughput, latency and energy consumption. In order to look at the current-state of routing in IoT applications, a discussion of the main aspects of RPL and the advantages and disadvantages of using it in different IoT applications is presented. In addition to that, a review of the available research related to RPL is conducted in a systematic manner, based on the enhancement area and the service type. Finally, a comparison of related RPL-based protocols in terms of energy efficiency, reliability, flexibility, robustness and security is presented along with conclusions and a discussion of the possible future directions of RPL and its applicability in the Internet of the future.