| With the rapid development of wireless communication technology and intelligent equipment,indoor localization technology has been developed unprecedentedly.At present,there are many representative indoor localization technologies,such as Wi-Fi fingerprinting technology,inertial navigation technology,Bluetooth beacon technology,ultra wideband positioning technology and so on.Among the many indoor localization technologies,Wi-Fi fingerprinting has become one of the most popular technologies because of its low cost,good compatibility with existing smart devices and other advantages.However,because of the large mobility of people and variable Interference sources in indoor environment and unstable wireless signals the fingerprint algorithm cannot detect and eliminate the disturbed signal sources,if there are interference signal sources in the matching algorithm,it will seriously affect the accuracy of indoor localization.Therefore,it is necessary to study the fingerprint matching algorithm and improve the fingerprint algorithm so that it can detect and eliminate the interference signal source,so as to improve the practicability of Wi-Fi fingerprinting technology.Based on this problem,this paper proposes one algorithm based on standard deviation and the other weighted improvement method based on fingerprint algorithm in order to improve the accuracy and accuracy of indoor positioning,and through the actual test,comparing location accuracy of the fingerprint matching algorithm,the algorithm based on standard deviation and the weighted improvement method based on fingerprint algorithm in different indoor environment.In addition,this paper discusses the value of K in the fingerprint algorithm,and tests the signal emission law of wireless signals in open environment and indoor environment.On the basis of summing up the previous research results and combining the research of this paper,the following research results are proposed:(1)No matter in the strong signal region or weak signal area,using the fingerprint algorithm of K=2 can achieve indoor positioning accuracy of 1-7m,and the probability of positioning accuracy within 4m is more than 80%.(2)In the open environment,the distance resolution of Wi-Fi signal intensity is 1-3m,the effective range is less than 15 m,and the effective value can be used to locate more than-65 dbm.In the case of indoor,the range resolution of Wi-Fi signal intensity is lower than that in open environment,and the stability and signal intensity values are higher than those in open environment.(3)The obstructions reflects radio signal in the indoor environment,so in the indoor environment the relationship between signal intensity and distance is not consistent with the signal attenuation formula,if using the Wi-Fi signal strength ranging technique,the signal attenuation formula must be improved.In addition,when the K value is 2,the positioning accuracy of the fingerprint algorithm is obviously better than that of the fingerprint algorithm when the K value is 1.(4)The localization accuracy of the three algorithms in the strong signal region is better than that in the weak signal region In addition,the improved algorithms in different conditions can effectively improve the precision and accuracy of indoor positioning.In the strong Wi-Fi signal environment,the improved fingerprint algorithm based on standard deviation can eliminate the errors form the Wi-Fi signal instability and the occlusion of Wi-Fi signal by personnel flow,so in such environment using the improved fingerprint algorithm based on standard deviation can improve the precision and accuracy of indoor positioning.In the weak Wi-Fi signal environment,the improved fingerprint algorithm based on weight can suppress the weak Wi-Fi signal effect on accuracy,increase the strong effect of Wi-Fi signal on the accuracy,so in such environment the improved fingerprint algorithm based on weight can guarantee the precision and accuracy of indoor positioning. |
PDF Full Text Request