A hybrid multi-objective optimisation for energy efficiency and better coverage in underwater wireless sensor networks
Underwater wireless sensor networks (UWSNs), which benefit ocean surveillance applications, marine monitoring and underwater target detection, have advanced substantially in recent years. However, existing deployment solutions do not satisfy the deployment of mobile underwater sensor nodes as a sto...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English English |
| Published: |
2022
|
| Subjects: | |
| Online Access: | https://eprints.ums.edu.my/id/eprint/35662/1/24%20PAGES.pdf https://eprints.ums.edu.my/id/eprint/35662/2/FULLTEXT.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Underwater wireless sensor networks (UWSNs), which benefit ocean surveillance applications, marine monitoring and underwater target detection, have advanced substantially in recent years. However, existing deployment solutions do not satisfy the deployment of mobile underwater sensor nodes as a stochastic system. Internal and external environmental problems concern maximum coverage in the deployment region while minimising energy consumption. To fill this gap, this research proposes and implements a multi-objective optimisation solution to balance conflicts concerning node deployment objectives. First, this research analyses the existing mobile underwater node deployment algorithms to identify the significant problems in existing solutions. Next, it establishes the research problems by implementing various existing algorithms using comparative analysis. Based on that analysis, this research suggests a hybrid algorithm: the Multi-Objective Optimisation Genetic Algorithm based on Adaptive Multi-Parent Crossover and Fuzzy Dominance (MOGA-AMPazy). The method adapts the original Non-Dominated Sorting Genetic Algorithm II (NSGA-II) by introducing a hybridisation of adaptive multi-parent crossover genetic algorithm and fuzzy dominance-based decomposition techniques. The algorithm introduces the fuzzy Pareto dominance concept to compare two solutions and uses the scalar decomposition method when one solution cannot dominate the other in terms of the fuzzy dominance level. The solution also proposes adaptive multi-parent crossover (AMP) to balance exploration and exploitation with new offspring, changing the number of parents involved in the crossover based on the execution of the new generation. The solution is further improved by introducing prospect theory to guarantee convergence through risk evaluation. The results obtained are then analysed to assess the proposed solution's performance in obtaining each deployment objective's optimal value. Finally, the proposed algorithm's effectiveness regarding node coverage, energy consumption, Pareto-optimal value, and algorithm execution time is validated using three Pareto-optimal metrics: including inverted generation distance (IGD), hypervolume, and diversity. Furthermore, this research utilises five commonly used two-objective ZDT test instances as benchmark tests, namely ZDT-1, ZDT-2, ZDT-3, ZDT-4, and ZDT-6. These tests use specific problem characteristics to impose the underlying proposed solution as well as three other systems. Pareto-optimal values obtained indicate that the proposed solution has almost complete coverage involving the actual Pareto front. Furthermore, all analysis and evaluation attributes indicate that the MOGA-AMPazy deployment algorithm can handle the multi-objective underwater sensor deployment problem better than other solutions. Thus, MOGA-AMPazy provides an efficient and comprehensive deployment solution for mobile sensor nodes in UWSNs. This study makes several noteworthy contributions to the body of knowledge concerning UWSNs, and it provides an excellent multi-objective representation to decision-makers or mission planners to monitor the region of interest (Rol). |
|---|