| Autor: |
Al-Nader, Issam, Lasebae, Aboubaker, Raheem, Rand, Ngondi, Gerard Ekembe |
| Zdroj: |
Sensors (14248220); 3/15/2024, Vol. 24 Issue 6, p1-17, 17p, 2 Diagrams, 4 Charts, 6 Graphs |
| Abstrakt: |
The multi-objective optimization (MOO) problem in wireless sensor networks (WSNs) is concerned with optimizing the operation of the WSN across three dimensions: coverage, connectivity, and lifetime. Most works in the literature address only one or two dimensions of this problem at a time, except for the randomized coverage-based scheduling (RCS) algorithm and the clique-based scheduling algorithm. More recently, a Hidden Markov Model (HMM)-based algorithm was proposed that improves on the latter two; however, the question remains open if further improvement is possible as previous algorithms explore solutions in terms of local minima and local maxima, not in terms of the full search space globally. Therefore, the main contribution of this paper is to propose a new scheduling algorithm based on bio-inspired computation (the bat algorithm) to address this limitation. First, the algorithm defines a fitness and objective function over a search space, which returns all possible sleep and wake-up schedules for each node in the WSN. This yields a (scheduling) solution space that is then organized by the Pareto sorting algorithm, whose output coordinates are the distance of each node to the base station and the residual energy of the node. We evaluated our results by comparing the bat and HMM node scheduling algorithms implemented in MATLAB. Our results show that network lifetime has improved by 30%, coverage by 40%, and connectivity by 26.7%. In principle, the obtained solution will be the best scheduling that guarantees the best network lifetime performance as well as the best coverage and connectedness for ensuring the dependability of safety-critical WSNs. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
