摘要
为了提升矿井WiFi指纹定位系统的定位精度和覆盖范围,提出了一种基于改进粒子群算法和遗传算法的优化定位方案,通过改进的粒子群算法动态调整粒子惯性权重,精确优化路径损耗指数和环境噪声参数,从而更好地反映复杂矿井环境下的信号传播特性。同时,遗传算法用于优化基站部署,结合改进的加权K最近邻(WKNN)算法,进一步提升定位精度。实验结果表明,改进的遗传算法和WKNN算法使得定位误差中位数降至1.62 m。该方法显著提升了矿井定位精度和系统鲁棒性,保障了矿井的安全生产。
To enhance the accuracy and coverage of WiFi fingerprinting localization systems in mines,this paper proposesd an optimized localization scheme based on an improved Particle Swarm Optimization(PSO)and Genetic Algorithm(GA).The improved PSO dynamically adjusts the particle inertia weight to precisely optimize the path loss exponent and environmental noise parameters,thereby better reflecting the characteristics of signal propagation in the complex mining environment.Meanwhile,GA was used to optimize base station deployment,and in combination with an improved Weighted K-Nearest Neighbor(WKNN)algorithm,further improves localization accuracy.Results:Experimental results show that the median positioning error is reduced to 1.62 meters by using the improved GA and WKNN algorithm.This approach significantly enhances the localization accuracy and system robustness in mining environments,ensuring safe production in mines.
作者
胡青松
苗家珲
罗渝嘉
张元生
霍羽
李世银
HU Qingsong;MIAO Jiahui;LUO Yujia;ZHANG Yuansheng;HUO Yu;LI Shiyin(School of Information and Control Engineering,China University of Mining and Technology,Xuzhou 221116,China;State Key Laboratory of Intelligent Opti mized Manufacturing in Mining&Metallurgy Process,Beijing 100160,China)
出处
《有色金属(矿山部分)》
2025年第4期18-25,共8页
NONFERROUS METALS(Mining Section)
基金
国家自然科学基金资助项目(52474185)
矿冶过程智能优化制造全国重点实验室开放研究基金(BGRIMM-KZSKL-2023-1)。
关键词
信号与信息处理
粒子群算法
遗传算法
群智能
目标定位
矿井智能化
signal and information processing
particle swarm optimization
genetic algorithm
swarm intelligence
object localization
intelligent mining
