Use this URL to cite or link to this record in EThOS:
Title: Service embedding in IoT networks
Author: Al-Shammari, Haider Qays Saeed
ISNI:       0000 0004 8504 5848
Awarding Body: University of Leeds
Current Institution: University of Leeds
Date of Award: 2019
Availability of Full Text:
Access from EThOS:
Access from Institution:
The Internet of Things (IoT), also called the Internet of Everything, is a new technology that has realised the paradigm of a global network of things capable of interacting with each other. IoT nodes can generate enormous amounts of data, perform certain analyses, and make decisions to provide efficient and smart services. These data are used to discover and resolve various issues and to provide value-added services to the users. The data analysis and decision making may be embedded into IoT devices to ensure that the decision-making takes place at the data source (i.e. the sensor node). The IoT infrastructure is composed of numerous heterogeneous devices used and deployed by various applications and services. The recent IoT architectures are designed as service- or event-driven paradigms, and there is no unified IoT architecture to correlate and integrate the data from physical nodes and services. The proposed IoT solutions use private techniques and cause several problems related to technology compatibility, information sharing, service management, and network upgrading. All these obstructions are impeding the development of IoT. In order to integrate various resources and services into a comprehensive system, these is a need for an efficient architecture that hides such heterogeneity from higher-level applications, provides interoperability for information exchange with other IoT devices, and considers different service scenarios, application-based demands, and recent technologies. Energy efficiency is considered a key enhancing factor in IoT where the sensing, processing, and communication of a huge number of IoT devices consume substantial amounts of energy. Energy consumption is considered to be a sustainability issue with respect to IoT devices, as these devices are powered by low-power sources or batteries, which impede the continuous operation of IoT systems. Recent proposals have increased the energy efficiency of processors and networks through a further development of energy-efficient paradigms. In this thesis, these motivations were considered to propose and develop a framework for service embedding in IoT networks to enhance the network performance and reduce both the cost of devices and the power consumption. Furthermore, a service-oriented architecture (SOA) was designed as viable middleware between a user's applications and an IoT physical layer, and interoperability between heterogeneous IoT devices was achieved. This SOA enabled the abstraction of IoT device functions that could then be translated into basic services, which in turn could be composed of complex services and exploited to the upper application layer. Energy efficiency was also considered a key enhancing factor in the proposed framework, as processing and communication power consumption is a sustainability issue in IoT systems. The objective of minimising the power consumption resulted in a framework that selected only low-power-consumption nodes and routes. The constraints introduced while minimising power consumption affected networkrelated issues such as traffic latency because of mutually dependent factors including the traffic volume and the routes selected to consolidate the traffic and hence save power. The proposed framework also optimised the selection of the routes for the traffic between the source and the destination to enhance the optimality of power saving and latency minimisation. Furthermore, we extended the proposed framework to include the fog and the cloud as on-demand access processing resources, as real-time services require a high processing speed and considerable data storage. We investigated the impact of the processing latency and the coexistence constraints of the processing and network power consumption. The cloud and the fog provided an ideal processing solution for IoT devices and services. In IoT networks, a node can cause temporary outages. A smart building consists of interconnected sensors, controllers, and actuator devices. If a sensor of a monitoring and control system fails, the controller may receive an incorrect signal, which may result in the failure of the entire system. Hence, node and traffic resilience were investigated in this thesis. Furthermore, recent resilience schemes were evaluated in terms of their power consumption and mean traffic latency. As a result, in this thesis, a novel traffic resilience technique was developed to enhance network performance and reduce network power consumption.
Supervisor: Elmirghani, Jaafar ; Elgorashi, Taisir Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available