Intelligent traffic routing management algorithm for wireless sensor networks.
Tshilongamulenzhe, Tshimangadzo Mavin
Tshilongamulenzhe, Tshimangadzo Mavin
Citations
Altmetric:
Abstract
Over the past decade, Wireless Sensor Networks (WSNs) have gained attention worldwide for transmitting packets from one location to another. WSNs are formed by sensor nodes (SNs) interconnected with one another through wireless links. SNs in WSNs may be deployed randomly in any location; and they have the potential to move around in the network environment. The deployment areas include the healthcare, military, home, and irrigation environments. In addition, these WSNs are primarily used to monitor temperature, humidity, and pressure, using a variety of SNs. Packet transmission in WSNs allows network users to access information wherever they are, depending on network coverage. However, packet transmission in WSNs comes with various challenges such as packet loss, packet delay, decrement of network throughput, and many others. These challenges are normally caused by head-of-line congestion that occurs in WSNs during packet transmission. Head-of-line congestion occurs when the first packet holds up a line of packets during many-to-one packet routing. The existing algorithms used for packet traffic management have failed to minimize head-of-line congestion in WSNs. This is because the existing algorithms lack proper decision-making based on packet flow to avoid packet congestion in WSNs. Such is the case within the agricultural environment. As a result, the study designed the intelligent traffic routing management (ITRM) algorithm by integrating machine learning congestion control (MLCC) and load balancing routing mechanism (LBRM). Network Simulator-2 (NS-2) has been used to test the effectiveness of the proposed ITRM algorithm when compared with the MLCC and LBRM. For low-priority packets, the NS-2 simulation results demonstrated that the proposed ITRM algorithm has an average of 0.27% packet loss, 0.4% packet delay, and 98.3% network throughput which is more than MLCC and LBRM when simulation time starts at 30 seconds, with less than 10 bytes of packet size. The NS-2 simulation results for high-priority packets in the agricultural environment demonstrated that the proposed ITRM algorithm has an average of 1.3% packet loss, 1.06% packet delay, and 95.7 % network throughput when packet size is 10 bytes in WSNs.
The simulation results from the NS-2 simulator demonstrated that the proposed ITRM algorithm performs better than the MLCC and LBRM in terms of packet loss and packet delay; and performs much better in terms of the network throughput. As a result, the quality of services (QoS) has been improved in WSNs within the agricultural environment.
Description
Thesis submitted in partial fulfilment of the requirements for the degree of Doctor of Computing in Information Technology, Department of Information Technology, Faculty of Information and Communication Technology at the Tshwane University of Technology.
Date
2022-08-01
Journal Title
Journal ISSN
Volume Title
Publisher
Tshwane University of Technology.
Research Projects
Organizational Units
Journal Issue
Keywords
Wireless Sensor Networks., Sensor Nodes., Intelligent Traffic Routing Management algorithm., Packet transmission., Network throughput., Network Simulator-2., Agricultural environment.