Research On High Usability Wi-Fi Indoor Positoning Techniques Based On Android Smartphone

With the growing demand for location-based services,especially for indoor location-based services,and the rapid development of Wi-Fi communication technology,indoor positioning technology based on Wi-Fi has become a research hotspot.In positioning technology based on Wi-Fi,the Wi-Fi-based Received Signal Strength Indication(RSSI)indoor positioning technology can be located only by using the most common smartphones and wireless network resources available in life.With the advantages of easy to implement and low-cost,so now the indoor positioning technology based on Wi-Fi are mostly use RSSI observables.But there are still many areas for improvement in this technology.Based on this,this paper focuses on the research and improvement of RSSI-based fingerprinting algorithm to improve the usability of Wi-Fi based indoor positioning technology on Android phones.Bayesian fingerprinting positioning,a classic Wi-Fi-based indoor positioning method,consists of two phases: radio map learning and position inference.Thus far,the application of Bayesian fingerprinting positioning is limited due to its poor usability;radio map learning requires an adequate number of observables at each reference point,long-term fieldwork,and high development and maintenance costs,also bring problems such as low efficiency of radio map learning,large amount of calculation and limited positioning accuracy.The research in this paper mainly focuses on the key issue of how to collect RSSI observables efficiently.To improve the usability and positioning accuracy of Wi-Fi-based positioning algorithms is the research goal.In this paper,based on a statistical analysis of actual RSSI observables,the real distribution of RSSI observables is almost entirely non-Gaussian,but is closer to the classical Weibull signal model in the radio propagation model.Therefore,a Weibull–Bayesian density model is proposed to represent the probability density of Wi-Fi RSSI observables.The main contributions of this paper are summarized as follows:(1)In the real-time positioning stage,this paper proposes the strongest AP judgment floor method based on Wi-Fi.Before positioning,it will first determine which floor the user is on,and then choose the radio map of the corresponding floor for positioning.This method is complete the Wi-Fi-based indoor positioning system and enhances the usability.(2)By analyzing the spatial distribution characteristics of RSSI observables in indoor environment,a Weibull model is proposed to represent the probability of Wi-Fi RSSI observables,which could improve the accuracy of the fitting RSSI observables distribution and thus improve the accuracy and usability of the algorithm.(3)In this paper,a Weibull–Bayesian density model is proposed to represent the probability density of Wi-Fi RSSI observables.The model can be calculated with fewer samples,can calculate the probability density with a higher accuracy than the traditional histogram method.Furthermore,the parameterized Weibull model can simplify the radio map by storing only three parameters that can restore the whole probability density,which can greatly reduce the complexity of the radio map without losing any useful information.(4)The improved Bayesian positioning inference is performed in the positioning phase using probability density rather than the traditional probability distribution of predefined RSSI bins.The proposed method calculates the posterior probability using the Bayesian density model and a run-time dynamically defined bin according to adds 5dB and 5dB of the RSSI observables of the received AP in the positioning phase,and a new algorithm based on RSSI for Wi-Fi fingerprinting is proposed.The proposed method was implemented on an Android smartphone,and the performance was evaluated in different indoor environments.Results revealed that the proposed method enhanced the usability of Wi-Fi Bayesian fingerprinting positioning by requiring fewer RSSI observables and improved the positioning accuracy by 19–32% in different building environments compared with the classic histogram-based method,even when more samples were used.In general,the new method proposed in this paper exhibits good prospects.However,the presented results are only for Wi-Fi fingerprinting positioning,and the proposed method has not been integrated with pedestrian dead reckoning(PDR)or other localization sources;hence,the positioning accuracy is not fully up-to-date.In the future,we will conduct additional studies on multisource integrated positioning.For instance,the fusion of Wi-Fi fingerprinting with PDR and maps will result in a better positioning accuracy.