基于Wi-Fi指纹的室内定位方法：通过标准粒子群优化算法

Wi-Fi Fingerprint-Based Indoor Localization Method via Standard Particle Swarm Optimization.

作者信息

Zheng Jin, Li Kailong, Zhang Xing

机构信息

School of Architecture and Art, Central South University, Changsha 410083, China.

College of Electrical and Information Engineering, Hunan University, Changsha 410082, China.

出版信息

Sensors (Basel). 2022 Jul 5;22(13):5051. doi: 10.3390/s22135051.

DOI:10.3390/s22135051

PMID:35808546

原文链接:https://pmc.ncbi.nlm.nih.gov/articles/PMC9269854/

Abstract

With the continuous development and improvement in Internet-of-Things (IoT) technology, indoor localization has received considerable attention. Particularly, owing to its unique advantages, the Wi-Fi fingerprint-based indoor-localization method has been widely investigated. However, achieving high-accuracy localization remains a challenge. This study proposes an application of the standard particle swarm optimization algorithm to Wi-Fi fingerprint-based indoor localization, wherein a new two-panel fingerprint homogeneity model is adopted to characterize fingerprint similarity to achieve better performance. In addition, the performance of the localization method is experimentally verified. The proposed localization method outperforms conventional algorithms, with improvements in the localization accuracy of 15.32%, 15.91%, 32.38%, and 36.64%, compared to those of KNN, SVM, LR, and RF, respectively.

摘要

随着物联网（IoT）技术的不断发展和完善，室内定位受到了广泛关注。特别是，基于Wi-Fi指纹的室内定位方法因其独特优势而得到了广泛研究。然而，实现高精度定位仍然是一个挑战。本研究提出将标准粒子群优化算法应用于基于Wi-Fi指纹的室内定位，其中采用了一种新的双面板指纹同质性模型来表征指纹相似度，以实现更好的性能。此外，通过实验验证了该定位方法的性能。所提出的定位方法优于传统算法，与KNN、SVM、LR和RF相比，定位精度分别提高了15.32%、15.91%、32.38%和36.64%。