The advancement of wireless technology has led to an increase in the employment of wireless sensor networks (WSNs). Traditionally, WSNs are powered by batteries. However, the high power consump- tion and the need to change the batteries regularly has made these networks costly to maintain. The nodes in the WSNs are increasingly strained as power consumption increases and the batteries are depleted faster. This has consequently decreased the overall lifetime of the WSNs.
Although many energy-conserving techniques exist, for example energy-efficient medium access control and energy-efficient routing protocols, energy consumption remains one of the significant constraints in the development of WSNs. A natural solution to this constraint is harvesting energy from the environment. However, unlike conventional energy, energy harvested from the environment is random in nature, making it challenging to realise energy-harvesting transmission schemes. Although energy harvesting might be considered a solution to many problems, it brings about new challenges with regard to the usage and management of the energy harvested. Some of these challenges include uneven consumption of power in the network, resulting in dead nodes in some portion of the network
and the batteries used in the network are being affected negatively by the energy usage; they may consequently sustain the nodes for long or short periods. To analyse the usage and consumption of energy, a number of techniques have been proposed, namely; information theory, game theory and queueing theory.
By this time, the performance of the sensor nodes in WSNs has been analysed making use of a queueing-theoretic model for each sensor. The aforementioned model inadequately expresses the physical constraints, namely, the energy drawing process and the finite battery capacity.
This research focuses on developing a model that captures the harvesting, accumulation and dissipation of energy, utilising queueing theory. A rechargeable battery with a finite storage capacity will be used. To ensure that the battery does not lose its capability to store charge after being recharged repeatedly, the leaky bucket model is proposed to check the network data flow as the harvested energy in the WSN is analysed.
To capture real-world WSNs with energy harvesting in which there is energy leakage, the energy- harvesting sensor node performance is analysed with two assumptions: data transmission and energy leakage occurring and the token buffer being subjected to a threshold. The system had finite buffers for the data and energy. To make it possible to have some influence over the system performance measures a threshold is imposed on the token buffer.
Four models are developed: a basic model, a basic model with leakage incorporated, a basic model with leakage and priority incorporated and a basic model with leakage, priority and threshold incorporated. The developed models are simulated and results for the performance measures are obtained. / Dissertation (MEng (Computer Engineering))--University of Pretoria, 2021. / BWMC, NRF / Electrical, Electronic and Computer Engineering / MEng (Computer Engineering) / Unrestricted
| Identifer | oai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:up/oai:repository.up.ac.za:2263/80778 |
| Date | January 2021 |
| Creators | Angwech, Otim Patricia |
| Contributors | Alfa, Attahiru S., u10177109@tuks.co.za, Maharaj, Sunil |
| Publisher | University of Pretoria |
| Source Sets | South African National ETD Portal |
| Detected Language | English |
| Type | Dissertation |
| Rights | © 2019 University of Pretoria. All rights reserved. The copyright in this work vests in the University of Pretoria. No part of this work may be reproduced or transmitted in any form or by any means, without the prior written permission of the University of Pretoria. |
Page generated in 0.0014 seconds