A novel Tunnel Positioning Approach via Long Term Evolution Cellular Signal

Abstract Currently, Radio Frequency Identification System (RFIS), ZigBee and Ultra-Wide Band (UWB) methods are mainly used to positioning in enclosed space. But they require complex hardware layout high costs, resulting the inability meet needs of environments. Therefore, we designed a novel tunnel approach via Long Term Evolution (LTE) cellular signal. This includes: signal acquisition, data preprocessing, feature database construction, model training real-time positioning. In preprocessing stage, adopt 3sigma Kalman filtering filter outliers noise, use information gain rate select effective features. phase, combination K-Weighted Nearest Neighbor (KWNN) Support Vector Regression (SVR) is for tunnel. To verify approach, did an experiment using from actual The experimental results show that this has better accuracy than FK-NN [1] TSVR [2].