Indoor CSI fingerprint localization based on tensor decomposition

Indoor Wi-Fi localization methods based on the Received Signal Strength (RSS) are widely used because of the low computational complexity and strong applicability. Compared with the RSS, the Channel State Information (CSI) can provide the multi-channel subcarrier phase and amplitude information to better describe the signal propagation path. Thus, the CSI becomes one of the most commonly used signal features in indoor Wi-Fi localization. Compared to the CSI-based geometric localization method, the fingerprint-based localization method has advantages of easy implementation and high accuracy. Based on this, this paper proposes an indoor CSI fingerprint localization approach based on tensor decomposition. Specifically, we combine the tensor decomposition algorithm based on the Parallel Factor (PARAFAC) analysis model with the Alternating Least Squares (ALS) iterative algorithm to reduce the interference of the environment. Then, we use the tensor wavelet decomposition algorithm for feature extraction and obtain the CSI fingerprint. Finally, distinguishing from the traditional localization algorithm based on machine learning, this paper establishes a localization model based on the Partial Least Squares Regression (PLSR) algorithm to predict position coordinates. Experimental results show that the proposed approach is with the high localization accuracy and good fingerprint collection efficiency.