Research On WLAN Local Location Technology Based On Fingerprint And AP Selection

With the increasing use of data services and multimedia services in social life,location-based services(LBS)show a clear upward trend in people's needs.Especially in the underground parking lot,airport hall,shopping center and other scenes,often need to accurately locate the indoor objects,the traditional satellite positioning technology using microwave transmission,due to multipath interference,non-line-of-sight transmission,signal attenuation and other reasons,the signal is difficult Through the complex outer building,can not meet the requirements of indoor positioning.With the indoor wireless signal facilities are widely deployed,WLAN-based indoor positioning technology has become an important research and development direction in the field of positioning in recent years.In this paper,the WLAN indoor positioning technology in-depth study,first analyzed the status of WLAN indoor positioning technology research and development trends,from the access point(Access Point,AP)Received Signal Strength(Received Signal Strength,RSS)propagation characteristics Field measurement studies.On this basis,the feature fingerprint matching algorithm and the AP selection algorithm are analyzed,and the improved algorithm is proposed to improve the accuracy of indoor positioning.The concrete work is as follows:(1)Study the spread of RSS signal characteristics: in the actual positioning environment,the RSS signal for field measurement and analysis.The probability distribution of RSS is studied by sampling time,person walking,AP selection and sampling point position as variables.In addition,the influence of sampling grid size and AP number on the positioning effect was studied by design experiment.The number of fingerprint grid edge and the optimal AP selection were determined.(2)Research on fingerprint feature matching algorithm: An improved feature matching algorithm based on Bayesian algorithm is proposed.The concept of regional partitioning is introduced to cluster all the reference points in the positioning space,which solves the problem that the Naive Bayesian algorithm only estimates the position of the reference point.The validity of the improved algorithm is verified by the simulation experiment,and the cumulative probability in the range of 2 m is improved from 60% to 70% before the improvement.(3)Research on AP selection algorithm: An AP selection algorithm based on information gain is proposed.On the basis of the InfoGain algorithm,the difference between the adjacent APs is introduced,weighted with the information gain value,and the integrated identizability of the AP is quantified.The AP with the most position recognition capability is selected as the standard.Through the simulation experiment,the proposed AP selection algorithm can effectively remove the redundant AP information,and the feature matching algorithm combined to reduce the positioning complexity,compared with the InfoGain algorithm,2 m precision range of cumulative probability,from 78% to 85%.