Design And Implementation Of The Indoor Localization Algorithm Based On WiFi

The popularity of smartphones has fostered a growing interest in Location Based Service (LBS). However, the foremost issue for LBS is how to accurately obtain the user’s position information. With the assistance from Global Positioning System (GPS), the location of people at outdoor regions can be easily determined with a certain level of precision. But indoor localization still remains a challenge. While indoor location information is imperative for the realization of mobile LBS applications, recently research on indoor localization systems that rely on received signal strength (RSS) fingerprint has been attracting much attention.It is preferable to utilize the current existing wireless communication infrastructure to get users’position. Recently, received signal strength (RSS) fingerprint system based on WiFi has begun to gain in popularity. However, the complexity and cost of fingerprint map building which requires an extensive and thorough site-survey process limit their application and development.. Moreover, the static radio map is also vulnerable to environmental dynamics and device heterogeneity. In this article, we propose a novel approach for localization addressing some major technical issues posed in RSS fingerprint. The innovation of our solutions is that we utilize the "neighbor relative" RSS(NR-RSS) to search the fingerprint database and adopts a Markov-chain Model to assist positioning. Our proposed scheme can also realize calibration-free positioning access different devices. With the assistance of NR-RSS, taking into account the effects of the pedestrian’s history status, our efficient scheme can dampen the unpredictable signal fluctuation problem in indoor environment and obtain quite high accuracy.In this article, we first introduce the development and related work of indoor localization and then we discuss the shortage and main problems of current indoor localization systems. After that, we describe the system architecture and our localization algorithm. And then we designed and implemented our localization algorithm on the Android client and the remote server and display the performance of our localization system compared with others. Finally, this article summarizes our work, and points out the direction of the next phase of work and improvement objectives.