Xiaoyu ZHANG, Jinghui CHEN, Chen DONG, Guorong HE

A method for indoor Wi-Fi location based on improved back propagation neural network

In order to achieve high precision on indoor location, a Wi-Fi indoor location method based on improvedback propagation (BP) neural network is proposed. The classical BP neural network is optimized in real time by theant colony optimization algorithm. Meanwhile, the momentum term is introduced to construct an improved four-layerBP neural network model. The model uses the Wi-Fi signal feature as the input of the BP neural network and succeedsin the area classification under multiple Wi-Fi signal features. The experimental results demonstrate that the improvedBP neural network can increase the classification accuracy of the classifier effectively, and achieve a high-precision indoorarea location. Furthermore, it performs better practical results while ensuring the time complexity. The advantagesof this method are high practicability, low cost, high prediction classification accuracy, and robust stability, which canachieve the efficient classification of the short-range area.

PDF

___

[1] Lee K, Nam Y, Min SD. An indoor localization solution using Bluetooth RSSI and multiple sensors on a smartphone. Multimedia Tools and Applications 2018; 77 (10): 12635-12654. doi: 10.1007/s11042-017-4908-2
[2] Jiang X, Chen Y, Liu J, Gu Y, Hu L. FSELM: fusion semi-supervised extreme learning machine for indoor localization with Wi-Fi and Bluetooth fingerprints. Soft Computing 2018; 22 (11): 3621-3635. doi: 10.1007/s00500- 018-3171-4
[3] Xu H, Wu M, Li P, Zhu F, Wang R. An RFID indoor positioning algorithm based on support vector regression. Sensors 2018; 18 (5): 1504-1518. doi: 10.3390/s18051504
[4] Xu H, Ding Y, Li P, Wang R, Li Y. An RFID indoor positioning algorithm based on bayesian probability and k-nearest neighbor. Sensors 2017; 17 (8): 1806-1822. doi: 10.3390/s17081806
[5] Bozkurt KS, Yazici A, Gunal S. An f-score-weighted ındoor positioning algorithm ıntegrating wifi and magnetic field fingerprints. Mobile Information Systems 2018; 2018: 1-8. doi: 10.1155/2018/7950985
[6] Bi J, Wang Y, Cao H, Wang Y. A method of Wi-fi indoor positioning based on omnidirectional fingerprint database. Bulletin of Surveying and Mapping 2018; 2018 (2): 25-29. doi: 10.13474/j.cnki.11-2246.2018.0038
[7] Ban R, Kaji K, Hiroi K, Kawaguchi N. Indoor positioning method integrating pedestrian Dead Reckoning with magnetic field and WiFi fingerprints. In: IEEE 2015 Eighth International Conference on Mobile Computing and Ubiquitous Networking (ICMU); Hakodate City, Hokkaido, Japan; 2015. pp. 167-172.
[8] Zhou R, Li Z, Luo L. WiFi-pedestrian dead reckoning fused indoor positioning based on particle filtering. Journal of Computer Applications 2016; 36 (5): 1188-1191,1200. doi: 10.11772/j.issn.1001-9081.2016.05.1188
[9] Shen X, Bai Z, Dong Y, Dong Y. WiFi indoor localization algorithm based on dynFWA-SVM. Transducer and Microsystem Technology 2018; 37 (4): 121-124,128. doi: 10.13873/J.1000-9787(2018)04-0121-04
[10] Li J, He X, Cai Y, Xu Q. Method of Wi-fi indoor location based on k—means and random forest. Engineering of China 2017; 24 (4): 787-792. doi: 10.14107/j.cnki.kzgc.15C6.0543
[11] Chang Q, Velde SVD, Wang W, Li Q, Hou H et al. Wi-Fi fingerprint positioning updated by pedestrian dead reckoning for mobile phone indoor localization. In: The 6th China Satellite Navigation Conference (CSNC 2015); Xian, China; 2015. pp. 729-739.
[12] Wang J, Zhang X, Gao Q, Yue H, Wang H. Device-free wireless localization and activity recognition: A deep learning approach. IEEE Transactions on Vehicular Technology 2017; 66 (7): 6258-6267. doi: 10.1109/TVT.2016.2635161
[13] Liao JK, Chiang KW, Tsai GJ, Chang HW. A low complexity map-aided fuzzy decision tree for pedestrian indoor/outdoor navigation using smartphone. In: IEEE 2016 International Conference on Indoor Positioning and Indoor Navigation (IPIN); Alcala de Henares, Spain; 2016. pp. 1-8.
[14] Fang Y, Deng Z, Xue C, Jiao J, Zeng H et al. Application of an improved k nearest neighbor algorithm in WiFi indoor positioning. In: The 6th China Satellite Navigation Conference (CSNC 2015); Xian, China; 2015. pp. 517-524.
[15] Xiao F, Guo Z, Zhu H, Xie X, Wang RC. AmpN: Real-time LOS/NLOS identification with WiFi. In: IEEE 2017 International Conference on Communications (ICC); Paris, France; 2017. pp. 1-7.
[16] Moallem P, Ayoughi SA. Improving backpropagation via an efficient combination of a saturation suppression method and momentum term. Neural Network World 2010; 20 (2): 207-222.
[17] Zhang N, Wu W, Zheng G. Convergence of gradient method with momentum for two-layer feedforward neural networks. IEEE Transactions on Neural Networks 2006; 17 (2): 522-525. doi: 10.1109/tnn.2005.863460
[18] Fan J, Wang Z, Qian F. Research progress structural design of hidden layer in BP artificial neural networks. Control Engineering of China 2005; 12 (5): 105-109. doi: 10.14107/j.cnki.kzgc.2005.s1.036
[19] Hong B, Jin F, Gao Q. Multi-layer feedforward neural network based on ant colony system. Journal of Harbin Institute of Technology 2003; 35 (7): 823-825. doi: 10.3321/j.issn:0367-6234.2003.07.017
[20] Li E, Yang P, Sun X. Improved algorithm of BP neural networks based on the Activation function with four adjustable parameters. Microelectronics & Computer 2008; 25 (11): 89-93. doi: 10.19304/j.cnki.issn1000- 7180.2008.11.023
[21] Liu Q, Chen G, Liu X, Zhan J. Research on initialization algorithms of weights and biases of BP neural network. Journal of Southwest China Normal University 2010; 35 (6): 137-141. doi: 10.13718/j.cnki.xsxb.2010.06.050
[22] Rohra JG, Perumal B, Narayanan SJ, Thakur P, Bhatt RB. User localization in an ındoor environment using fuzzy hybrid of particle swarm optimization & gravitational search algorithm with neural networks. In: Proceedings of Sixth International Conference on Soft Computing for Problem Solving, SocProS 2016; Thapar University, Patiala, India; 2016. pp. 286-295.
[23] Li X, Tong Z, Guo E, Luo X. Quantifying spatiotemporal dynamics of solar radiation over the northeast China based on ACO-BPNN Model and Intensity Analysis. Advances in Meteorology 2017; 2017: 1-15. doi: 10.1155/2017/1042603