Investigation into energy efficient wireless sensor networks for smart monitoring and sensing applications
Date of Issue2015
School of Electrical and Electronic Engineering
Intelligent Systems Centre
Technological advances in wireless electronics and algorithms have made it possible for us to have a plethora of assorted Wireless Sensor Network (WSN) nodes with diverse applications. Several recent developments have shown promise for the distinct nodes to be integrated and coordinate with each other despite having dissimilar hardware and protocol stacks. Some of the problems associated with the WSNs that have proven to be a major hurdle in their ubiquitous acceptance as the benchmark monitoring solutions are: energy consumption, scalability, reliability and security. All of these aspects demand thorough research; moreover, they also have a trade-off between each other, for instance, having a reliable system may demand an increase in energy consumption. Decreasing the energy consumption will in addition affect the range and scalability of the network. In this thesis the author has concentrated the research on integrating Radio-Frequency Identification (RFID) nodes with the traditional WSNs, placing more emphasis on reducing overall energy consumption of monitoring applications. Due to the diversity of the available nodes, their requirements and resources are drastically different. This feature introduces a challenge to reduce the energy consumption of an overall application composed of these hybrid platforms. In particular, some challenging in-pipe monitoring applications have been proposed for industries utilizing RFIDs and other sensor nodes, mainly because of the high range of low frequency RFID tags and their magnetic induction property made it suitable for metallic in-pipe monitoring. In order to reduce energy consumption and increase the network life-time of the proposed inspiring monitoring scenarios, a realistic estimate of the energy consumptions of various nodes deployed in one application with given MAC protocols is absolutely necessary. Experimenting with actual sensor networks to determine energy consumption is both time-consuming and costly, particularly when these diverse nodes are deployed in one application. Normally, simulators are used to estimate the performance of these nodes. Despite the abundance of WSN simulators offered today, none can simulate the energy efficiency given several types of nodes and situations in one application. Hence, a simulation platform is needed for a blended deployment of distinct types of WSNs and RFIDs to access application performance requirements while curbing energy consumption to improve the application's lifetime. In order to overcome this lack, a distinctive WSN simulator was developed, which is a dynamic, platform independent, rapid prototyping and interoperable multi-environment simulation platform for RFID and WSNs. Future works involve extending the simulator by improving it in terms of both its performance and functions. Finally, the techniques of Cognitive Radio (CR) and Cooperative Networks (CN) are investigated to enhance the range and performance of the monitoring applications. These techniques include Energy Detector, used in CRs spectrum sensing, and re-generative relays, used in CN. These techniques improve performance and reliability for long range in-pipe communications.
DRNTU::Engineering::Electrical and electronic engineering