Research On Devices Diversity Of Wi-Fi Indoor Positioning Based On Fingerprint

With the popularity of mobile intelligent devices and the Internet,location based service(LBS)is becoming more and more important.Location information plays an irreplaceable role in modern people's life.Typical location services include navigation of peripheral resources,push of interest points,etc.GPS,BD,GLONASS outdoor positioning systems are very mature and have relatively high accuracy,but in complex indoor environment,the signal is blocked,interfered,even can't be received,so that using satellite positioning accuracy is very low.Therefore,many experts and scholars have proposed a variety of indoor positioning solutions.Wi-Fi indoor location based on location fingerprint is one of them.At present,the coverage rate of Wi-Fi is high,the positioning cost is low,and the equipment is simple,so the Wi-Fi indoor positioning technology based on location fingerprint has become a research hotspot.However,due to the Wi-Fi signal is vulnerable to be disturbed,so the same device at different times to test the signal strength(RSSI)is not consistent.The difference between the hardware and software of the different devices leads to the fact that these devices get the different signal intensity even if at the same time,the same place for the same AP.In this paper,the problem of device diversity of Wi-Fi indoor positioning is mainly studied.First of all,experiments and analysis for the difference of different devices at the same time the same place to collect the same AP signal strength are carried out,and the method of standard deviation is used to solve the problem of unstable fingerprint database.Secondly,according to the device diversity,leading to three variables,test coverage,fingerprint extraction coverage and the average Euclidean distance.Based on these three variables,the similarity matching method of these three variables is proposed to solve the positioning error caused by the device diversity.In addition,in order to determine the coefficient of similarity matching method,the experimental analysis method was used to determine the final coefficient.Finally,according to the difference of RSSI,the mean value compensation method is used to correct the RSSI value.The experiment results show that the proposed scheme can effectively relieve the problem of device diversity and has a positive effect on the application of Wi-Fi indoor positioning system.