An enhanced evolutionary algorithm for requested coverage in wireless sensor networks
Wireless sensor nodes with specific and new sensing capabilities and application requirements have affected the behaviour of wireless sensor networks and created problems. Placement of the nodes in an application area is a wellknown problem in the field. In addition, high per-node cost as well as ne...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Published: |
2016
|
| Subjects: | |
| Online Access: | http://eprints.utm.my/id/eprint/84068/1/KamalJadidyAvalPFC2016.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Wireless sensor nodes with specific and new sensing capabilities and application requirements have affected the behaviour of wireless sensor networks and created problems. Placement of the nodes in an application area is a wellknown problem in the field. In addition, high per-node cost as well as need to produce a requested coverage and guaranteed connectivity features is a must in some applications. Conventional deployments and methods of modelling the behaviour of coverage and connectivity cannot satisfy the application needs and increase the network lifetime. Thus, the research designed and developed an effective node deployment evaluation parameter, produced a more efficient node deployment algorithm to reduce cost, and proposed an evolutionary algorithm to increase network lifetime while optimising deployment cost in relation to the requested coverage scheme. This research presents Accumulative Path Reception Rate (APRR) as a new method to evaluate node connectivity in a network. APRR, a node deployment evaluation parameter was used as the quality of routing path from a sensing node to sink node to evaluate the quality of a network deployment strategy. Simulation results showed that the behaviour of the network is close to the prediction of the APRR. Besides that, a discrete imperialist competitive algorithm, an extension of the Imperialist Competitive Algorithm (ICA) evolutionary algorithm was used to produce a network deployment plan according to the requested event detection probability with a more efficient APRR. It was used to reduce deployment cost in comparison to the use of Multi-Objective Evolutionary Algorithm (MOEA) and Multi-Objective Deployment Algorithm (MODA) algorithms. Finally, a Repulsion Force and Bottleneck Handling (RFBH) evolutionary-based algorithm was proposed to prepare a higher APRR and increase network lifetime as well as reduce deployment cost. Experimental results from simulations showed that the lifetime and communication quality of the output network strategies have proven the accuracy of the RFBH algorithm performance. |
|---|