Localization Of WiFi Access Points Based On The Rayleigh Lognormal Model

With the development of IEEE 802.11 network infrastructures,WiFi access points(APs)have been widely used in large streets and small lanes.Therefore,acquiring the locations of WiFi APs can not only plays a vital role in various WiFi related applications,such as localization,security,deployment and so on,but also inspires the emergence of novel applications.Most existing studies adopted the well-known lognormal distance path loss(LDPL)model,which only reflects large-scale fading during WiFi signal propagations but ignores small-scale fading induced by pervasive multipath effects.In this paper,we tackle the problem of AP localization based on the Rayleigh lognormal model which characterizes the influences of both large-scale fading and small-scale fading,and introduce particle filtering to estimate the location of a target AP.To enhance the applicability of this method,GPS in outdoor environments and WiFi fingerprinting in indoor environments can be leveraged to approximately label the locations of RSS measurements.Furthermore,the pedestrian dead reckoning(PDR)technique is employed to improve the accuracy of the location labels by incorporating the estimates of step length,step count and heading into the particle filtering.In addition,a particle area dynamic adjustment strategy(PADAS)is presented to reduce biases in AP localization estimates by dynamically refining the area where particles are distributed.Extensive experiments are designed and carried out in typical indoor and outdoor scenarios.It is shown that,if accurate location labels are available,theproposed method is able to achieve the localization accuracy of 2.48 m,2.55 m,7.80 m in three indoor scenarios and 4.21 m,5.03 m,6.61 m in three outdoor scenarios,which are significantly better than the existing solutions based on the LDPL model by14.13%-70.38%;otherwise,the enhanced method by using GPS/WiFi fingerprinting,PDR and PADAS is able to achieve the localization accuracy of 3.24 m in the indoor office and 6.57 m in the outdoor street.In addition,an Android APP is developed to validate the feasibility of the proposed algorithm on smartphones,demonstrating that it is totally acceptable to deploy the proposed method in practice.In conclusion,the proposed AP localization method has greatly improved the accuracy and practicability compared with traditional methods,and will evidently contribute to the applications and researches based on the locations of WiFi APs.