Indoor Location Estimation Based on Local Magnetic Field via Hybrid Learning

Abstract

In this chapter, the magnetic field samples over more than 2 months inside an office building presents a finding that there exist the relative stable measurements for a single location and the relative obvious difference in most of locations. Under this phenomenon, a hybrid learning method based on the local magnetic field measurements is proposed. (1) Kalman filter is firstly utilized to smooth the initial samples in order to obtain the stable data. (2) Classification programs by Extreme learning machine (ELM) is introduced to model the relationship between the measurements and physical locations, and then four potential positions can be chosen for a special measurement. (3) The optimal location is finally confirmed in view of those four selections by using K-nearest neighbor (KNN) algorithm. A series of experiments and comparisons with other five methods were implemented to validate the feasibility and superiority of this technique for improving the positioning accuracy.