期刊
APPLIED SOFT COMPUTING
卷 76, 期 -, 页码 726-743出版社
ELSEVIER SCIENCE BV
DOI: 10.1016/j.asoc.2018.12.022
关键词
Minimal exposure path; Probabilistic coverage model; Wireless sensor network; Genetic algorithm; Grid-base algorithm
资金
- Vietnam National Foundation for Science and Technology Development (NAFOSTED) [DFG 102.01-2016.03]
A well-known method for evaluating the coverage quality of Wireless Sensor Networks (WSNs) is using exposure as a measure, especially in barrier coverage problems. Among all studies related to exposure, discussions regarding the Minimal Exposure Path (MEP) problem have dominated research in recent years. The problem aims to find a path on which an intruder can penetrate through the sensing field with the lowest probability of being detected. This path along with its exposure value enables network infrastructure designers to identify the worst-case coverage of the WSN and make necessary improvements. Most prior research worked on the MEP problem under the assumption that there are no environmental factors such as vibration, temperature, etc., which causes errors in practical WSN systems. To overcome this drawback, we first formulate the MEP problem based on Probabilistic Coverage Model with noise (hereinafter PM-based-MEP) and introduce a new definition of the exposure metric for this model. The PM-based-MEP is then converted into a numerically functional extreme with high dimension, non-differentially and non-linearity. Adapting to these characteristics, we propose two approximation methods, GB-MEP and GA-MEP, for solving the converted problem. GB-MEP is based on the traditional grid-based method which is fine-tuned by several tweaks, and GA-MEP is formed by the genetic algorithm with a featured individual representation and an effective combination of genetic operators. Experimental results on numerous instances indicate that the proposed algorithms are suitable for the converted PM-based-MEP problem and perform well regarding both solution accuracy and computation time compared with existing approaches. (C) 2018 Elsevier B.V. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据