In indoor non-line-of-sight (NLOS) environments, wireless signals are often disrupted by obstacles, multipath effects, and other factors, which severely impact the performance and accuracy of positioning systems. To address this issue, an indoor positioning algorithm is proposed that integrates the K nearest neighbor (KNN) algorithm with the least squares (LS) method. First, the received signal strength (RSS) fingerprint database is corrected to reduce error fluctuations caused by NLOS conditions, providing a more stable foundation for positioning. Then, the KNN algorithm is applied to the corrected fingerprint library to compute an initial position estimate for the target, leveraging proximity-based signal characteristics in the target’s vicinity. Finally, the LS method is used to refine the initial position estimate from KNN, minimizing residual errors and enhancing positioning accuracy. Experimental results demonstrate that, in NLOS environments, the proposed algorithm significantly outperforms standalone KNN and LS methods, achieving superior accuracy and positioning performance. This integrated approach shows promise for improving accuracy in indoor positioning systems under challenging NLOS conditions.

Yujie Yang. Fusion Localization Algorithm for Non-Line-of-Sight Environment. Academic Journal of Computing & Information Science (2024), Vol. 7, Issue 11: 142-149. https://doi.org/10.25236/AJCIS.2024.071119.

[1] Xiansheng Guo, Shilin Zhu, Lin Li, Fangzi Hu & Nirwan Ansari.(2019). Accurate WiFi Localization by Unsupervised Fusion of Extended Candidate Location Set. IEEE Internet of Things Journal(2),2476-2485.

[2] Guo Xiansheng, Li Lin, Ansari Nirwan & Liao Bin. (2018). Knowledge Aided Adaptive Localization via Global Fusion Profile. IEEE Internet of Things Journal(2),1081-1089.

[3] Mehrabian Hamidreza & Ravanmehr Reza. (2023). Sensor fusion for indoor positioning system through improved RSSI and PDR methods. Future Generation Computer Systems, 254-269.

[4] Xiansheng Guo, Nirwan Ansari, Fangzi Hu, Yuan Shao, Nkrow Raphael Elikplim & Lin Li. (2020). A Survey on Fusion-Based Indoor Positioning. IEEE Communications Surveys & Tutorials(1), 566-594.

[5] Naveed Salman, Mounir Ghogho & Andrew H. Kemp. (2014).Optimized Low Complexity Sensor Node Positioning in Wireless Sensor Networks. IEEE Sensors Journal(1), 39-46.

[6] Ritabroto Ganguly, Manjarini Mallik & Chandreyee Chowdhury.(2024). Design of a knowledge distillation network for wifi-based indoor localization. Multimedia Tools and Applications (prepublish), 1-17.

[7] Bohang Chen, Jun Ma, Lingfei Zhang, Zhuang Xiong, Jinyu Fan & Haiming Lan. (2023). An improved weighted KNN fingerprint positioning algorithm. Wireless Networks(6),6011-6022.

[8] Wang Zhiyuan, Huang Huiling, Li Haoran, Chen Ziyang, Han Jun & Pu Jixiong. (2023).Non-line-of-sight imaging and location determination using deep learning. Optics and Lasers in Engineering

[9] Jingxue Bi, Yunjia Wang, Xin Li, Hongxia Qi, Hongji Cao & Shenglei Xu.(2018).An Adaptive Weighted KNN Positioning Method Based on Omnidirectional Fingerprint Database and Twice Affinity Propagation Clustering. Sensors(8),2502-2502.

[10] A. Anastasijević.(2013).An Indoor Positioning Realisation for GSM using Fingerprinting and kNN. Telfor Journal(1),82-86.