Journal of Computer Applications
Next Articles
Received:
Revised:
Online:
Published:
杨乐,张达敏
通讯作者:
基金资助:
Abstract: Abstract: An improved hunter-prey optimization (IHPO) method is proposed to improve network coverage in order to tackle the issue of broad coverage blind areas and uneven distribution of conventional wireless sensor networks. With the goal to improve population information interchange, cross-variation with dynamic proportional factors was first implemented during the prey position update stage. Second, adaptive α variation is included in the phase of updating the global ideal location to balance the algorithm's performance demands over time. Finally, the population is guided to complete dynamic reverse learning using the global optimal position of adaptive α variation perturbation, increasing the population's variety and capacity for global search. The network nodes optimized by IHPO have a greater coverage rate and are dispersed more uniformly in the WSN coverage challenge. The coverage rate may increase to 92.56% when the sensor perception capacity is insufficient, which is 18.95% greater than the original algorithm and 11.35% and 13.04% higher than the two upgraded algorithms IPSO and IGWO, respectively. The network working duration may be increased to 2500 cycles in the routing test time that the node energy consumption is more evenly distributed at the same.
Key words: hunter prey optimization algorithm, differential evolution, adaptive α variation, dynamic reverse learning, WSN coverage
摘要: 针对传统无线传感器网络节点部署覆盖盲区大、分布不均等问题,提出一种改进的猎人猎物优化算法(Improved Hunter-prey optimization,IHPO)优化网络覆盖。首先,在猎物位置更新阶段,引入差分进化思想借助动态比例因子进行交叉变异,增强种群信息交流;其次,在全局最优位置更新阶段由α稳定分布提出自适应α变异进行扰动,平衡不同时期算法性能需求;最后,利用自适应α变异扰动的全局最优位置引种群完成动态反向学习,增加种群全局搜索能力和多样性。在WSN覆盖问题中,使用IHPO优化的网络节点分布更均匀、覆盖率更高,在传感器感知能力不足时能达到92.56%覆盖率,对比原始算法提高18.95%,对比两种改进算法IPSO、IGWO分别提高11.35%、13.04%,同时部署后的节点能耗更均衡,在路由测试中网络工作时间可以延长至2500轮次。
关键词: 猎人猎物优化算法, 差分进化, 自适应α变异, 动态反向学习, WSN覆盖
CLC Number:
TP183
杨乐 张达敏. 改进猎人猎物优化算法在WSN覆盖中的应用[J]. 《计算机应用》唯一官方网站, DOI: 10.11772/j.issn.1001-9081.2023081208.
0 / Recommend
Add to citation manager EndNote|Ris|BibTeX
URL: https://www.joca.cn/EN/10.11772/j.issn.1001-9081.2023081208