Research On The Site Survey Analysis For WiFi Fingerprint-based Localization

In recent years,the growing demand for location-based services(LBS)has promoted the booming development of indoor localization technologies.With the gradual popularity of wireless local area network(WLAN)and mobile devices,WiFi fingerprint-based localization techniques have won wide attention from researchers due to their simple principles and universality.Although the performance analysis and optimization work for this technology has emerged,the mechanism of error formation is still unclear and optimizing localization performance is still extremely challenging,especially the relationship between the site survey for WiFi fingerprint-based localization and localization error is very complicated,the existing research work mainly relies on experimental methods and lacks theoretical support.To address these problems,this paper conducts an in-depth research in the following aspects.Firstly,a probabilistic formula is proposed to correctly distinguish any two reference points,and then extended to distinguish multiple reference points,which shows that the difference of fingerprints at any two locations is a key factor affecting the localization performance.Based on the Lognormal Distance Path Loss(LDPL)model,the mathematical relationship between the fingerprint difference between two reference points and the interval of reference points is analyzed,and then the expected localization error formula is obtained,which shows that the expected localization error grows gradually as the interval of reference points increases and the growth trend will gradually become faster.In addition,the specific form of the expected localization error in the one-dimensional and two-dimensional localization area is derived.Secondly,based on the theoretical analysis of the interval of reference points and the expected localization error,a method is proposed to obtain the optimal interval of reference points.Although decreasing the interval of reference points will reduce the localization error,it will also significantly increase the workload of site survey,so the paper defines a survey cost formula to weigh the localization error and the site survey overhead,and then analyzes the relationship between the interval of reference points and the survey cost to obtain the optimal interval of reference points in the current scenario.Finally,a large number of simulations and experiments are conducted to verify the theoretical analysis of the expected localization error formula and also verified the effectiveness of the method for obtaining the optimal interval of reference points.The influence of other parameters in the formula on the expected localization error and survey cost is discussed.In summary,this paper theoretically analyzes the relationship between interval of reference points and localization error,and proposes a method to obtain the optimal interval of reference point,which improves the research on WiFi fingerprint-based localization error analysis for site survey and also provides theoretical basis and methodological guidance for efficient site survey.