WLAN Indoor Positioning Algorithm Based On Signal Feature Extraction

With the popularity of smartphones and the development of mobile Internet,location-based service(LBS)has received increasing attention,it is often necessary to determine the position information of the users,especially in the complex indoor environment.The wireless local area network(WLAN)indoor positioning technology based on the received signal strength(RSS)take advantage of the hot spots which are deployed in the WLAN environment.This kind of technology can locate the wireless devices reliably and efficiently without another specialized hardware device and it has been widely concerned due to its high bandwidth,strong penetrating power,widely coverage,easy to access.In WLAN indoor environment,multipath effect and personnel ambulate irregularly disturb the signal,and shadow effect causes the loss of the signal.Due to the above several kinds of influence,the RSS values which are sampled from each access point(AP)on the same position have complex time-varying statistical properties.It reduces the WLAN indoor positioning accuracy based on RSS,and presents a difficult challenge for researchers.This articale takes reducing the influence of the RSS time-varying properties to the indoor positioning accuracy as research object,combines theoretical analysis and experimental contrast.The main achivments are summarized as the following:(1)Expound the basic principles and advantage of WLAN technology.Analyze the basic principles of position fingerprint,indoor location identification method,and the interference factors of influencing the indoor positioning accuracy.(2)Analyze the distribution of probability distribution of RSS,and contrary to the RSS time-varying properties,extract the signal characteristics of the original RSS by using linear discriminant analysis(LDA)and kernel direct discriminant analysis(KDDA).(3)Present the shuffled frog leaping algorithm(SFLA)and the least square support vector regression(LSSVR)algorithm.Study the two parameters optimization process of LSSVR based on SFLA and propose an indoor positioning algorithm based on KDDA and SFLA-LSSVR algorithm.The experimental results show that the positioning accuracy of the KDDA-SFLA-LSSVR algorithm is much superior to the WKNN,ANN,LSSVR algorithm under the condition of the same sampling numbers,and the number of RSS signal which is sampled from each AP is significantly reduced in the same positioning accuracy,and the proposed algorithm is a WLAN indoor positioning algorithm with good performance.