稳健性(进化)
计算机科学
字错误率
室内定位系统
定位系统
人工智能
数学
生物化学
化学
加速度计
基因
操作系统
几何学
点(几何)
作者
Shiwu Xu,Yi Wu,Xufang Wang,Fen Wei
标识
DOI:10.1109/jlt.2023.3265171
摘要
Indoor positioning system based on visible light communication and location fingerprinting can achieve cm-level positioning accuracy, which has become an important candidate for high precision positioning. However, in the case of non-uniform and low-density location fingerprinting distribution, it will still lead to large errors and poor robustness. To solve this problem, an accurate visible light positioning (VLP) method based on location fingerprinting and meta-heuristic is proposed, and all possible cases of uneven distribution of location fingerprinting are considered by introducing the random selection rate. The effectiveness of the proposed VLP method is compared with other six VLP methods based on location fingerprinting. Simulation results show that when the signal-to-noise ratio is 20 dB, the average positioning error of the proposed VLP method is 0.51 cm, which is at least 80.23% lower than the other six VLP methods. Experimental results show that when the random selection rate is 50%, the average positioning error of the proposed VLP method is 3.11 cm, which is at least 41.87% lower than the other six VLP methods. The proposed VLP method can be applied to positioning scenarios with different location fingerprinting distributions. The research results can provide a new research idea for the fingerprinting weight calculation method.
科研通智能强力驱动
Strongly Powered by AbleSci AI