Research On WiFi Indoor Localization Method Based On Location Fingerprint

In recent years,with the rapid development of wireless communication and artificial intelligence technologies,mobile intelligent terminals have increasingly become an indispensable part of people’s daily life,which makes indoor location-based services(LBS)more and more significant.At the same time,WiFi network becomes more and more popular,and it has spread throughout various places including offices,schools,and shopping malls,and has become the primary means of indoor communication in people’s study,work,leisure and entertainment life.Therefore,indoor localization using WiFi has also received great attention.However,there are many interferences in indoor environments that affect the propagation of WiFi signals,resulting in poor localization accuracy and stability.This thesis focuses on the above issues and investigates WiFi fingerprint localization methods,and the major contributions of this thesis are summarized as follows:(1)A fingerprint localization method based on deep learning is proposed.In the proposed method,a received signal strength indication(RSSI)fingerprint database is firstly established,which can be used to train a neural network based on a convolutional neural network(CNN)and a long short-term memory(LSTM)network,then the RSSI data of the target node(TN)are preprocessed and input into the network model to predict the initial estimated position;Finally,based on the RSSI fingerprint database,three reference nodes closest to the RSSI of the TN are determined according to the Euclidean distance,and the corresponding coordinates are weighted with the coordinates of the initial estimated position to obtain a better estimated position.Simulation and measured results show that compared with the existing methods,the proposed method can obtain higher localization accuracy and better robustness.(2)A K-means fingerprint localization method based on particle swarm optimization(PSO)and cosine similarity matching is proposed.In the proposed method,a RSSI fingerprint database is firstly established,and the global optimal initial clustering centers of K-means are found using PSO,then the RSSI fingerprints are clustered according to the cosine similarity.At the localization stage,the RSSI data of the TN are preprocessed and matched according to the cosine similarity to obtain its location.Simulation and measured results show that by using the proposed method,the localization accuracy can be significantly improved,and the real-time requirements of localization can also be met.