Mobile Phone-Based Indoor Positioning Method Using Adaptive Compressed Sensing for Mobile-Collected Fingerprints

Objectives The location method based on received signal strength indication (RSSI) fingerprint requires building a fingerprint database of the location area in the offline stage. The traditional static fingerprint collection method is time-consuming and labor-intensive to establish and update the fingerprint database, and the location consistency is easily affected by the terminal difference (such as the difference of the received signal caused by the different hardware of the fingerprint collection phone and the location phone), which hinders the large-scale application of this method.

Methods To address these problems, we collect RSSI fingerprints during mobile walking and build a corresponding mobile collection fingerprint database, and construct feature vectors according to the characteristics of mobile collection fingerprints. Sparse feature representation of mobile-collected fingerprints is proposed, and a fingerprint matching indoor location model is established based on adaptive compressed sensing algorithm.

Results The experimental results show that the fingerprint collection efficiency is improved by 90.83%, the average location error is 1.96 m, the root mean square error is 2.75 m, and the location consistency difference error is improved by 32.67% on average.

Conclusions The proposed method outperforms the existing algorithms in terms of fingerprint collection efficiency, location accuracy and location consistency of different phones.