Open Access Te Herenga Waka-Victoria University of Wellington
Browse
thesis_access.pdf (3.83 MB)

Event Reliability in Wireless Sensor Networks

Download (3.83 MB)
Version 2 2023-09-25, 02:08
Version 1 2021-12-07, 13:46
thesis
posted on 2023-09-25, 02:08 authored by Mahmood, Muhammad

Ensuring reliable transport of data in resource-constrained Wireless Sensor Networks (WSNs) is one of the primary concerns to achieve a high degree of efficiency in monitoring and control systems. The two reliability mechanisms typically used in WSNs are packet reliability and event reliability. Packet reliability, which requires all packets from all the sensor nodes to reach the sink, can result in wastage of the sensors' limited energy resources. Event reliability, which only requires that one packet related to each event reaches the sink, exploits the overlap of the sensing regions of densely deployed sensor nodes to eliminate redundant packets from nodes in close proximity that contain duplicate information about an event.  The majority of previous research in this area focuses on packet reliability rather than event reliability. Moreover, the research that does focus on event reliability relies on the sink to impose some form of control over the flow of data in the network. The sinks' centralized control and decision-making increases the transmission of unnecessary packets, which degrades overall network performance in terms of energy, congestion and data flow.  This thesis proposes a distributed approach to the control of the flow of data in which each node makes in-node decisions using data readily available to it. This reduces the transmission of unnecessary packets, which reduces the network cost in terms of energy, congestion, and data flow. The major challenges involved in this research are to: (i) accurately identify that multiple packets are carrying information about the same event, (ii) reliably deliver the packets carrying information about the unique event, (iii) ensure that enough information about the area of interest is reliably delivered to the sink, and (iv) maintain the event coverage throughout the network.  This thesis presents the Event Reliability Protocol (ERP) and its extension, the Enhanced Event Reliability Protocol (EERP). The protocols aim for the reliable transmission of a packet containing information about each unique event to the sink while identifying and minimizing the unnecessary transmission of similar redundant packets from nodes in the region of the event. In this way, the sensor nodes consume less energy and increase the overall network lifetime. EERP uses a multilateration technique to identify multiple packets containing similar event information and thus is able to filter redundant packets of the same event. It also makes use of implicit acknowledgment (iACKs) for reliable delivery of the packets to the sink node. The process is based on the hop-by-hop mechanism where the decisions are made locally by the intermediate nodes.  The thesis reports on simulations in QualNet 5.2 for verifying the accuracy of our event identification and event reliability mechanisms employed in the ERP and EERP. The results show that EERP performs better in terms of minimizing overall packet transmission and hence the energy consumption at the sensor nodes in a WSN. Also, the results for event identification mechanism and reliable event delivery show that EERP considerably improves upon other protocols in terms of unique events delivery.

History

Copyright Date

2019-01-01

Date of Award

2019-01-01

Publisher

Te Herenga Waka—Victoria University of Wellington

Rights License

CC BY-SA 4.0

Degree Discipline

Engineering

Degree Grantor

Te Herenga Waka—Victoria University of Wellington

Degree Level

Doctoral

Degree Name

Doctor of Philosophy

ANZSRC Type Of Activity code

3 APPLIED RESEARCH

Victoria University of Wellington Item Type

Awarded Doctoral Thesis

Language

en_NZ

Alternative Language

en

Victoria University of Wellington School

School of Engineering and Computer Science

Advisors

Welch, Ian; Andreae, Peter