An Indoor Localization Approach Based on Deep Learning for Indoor Location-Based Services

The rapid increase in the demand of location based services (LBS) for indoor environments has attracted scholars to indoor localization based on fingerprinting due its high accuracy. In this paper, we propose our novel indoor localization approach based on fingerprints of Received Signal Strength Indicator (RSSI) measurements. We present our approach of fingerprint preparation and setup and how we utilized machine learning techniques using Long Short-Term Memory (LSTM) Neural Networks for location estimation. Our experimental results shows that our localization approach outperforms well-known existing approaches like the KNN and localization techniques.