철도환경에서 BLE Beacon과 보정 알고리즘을 이용한 실내 위치 인식 시스템

Abstract

Developments in indoor positioning and information guidance solutions are being carried out actively based on beacons recently. The main factors of these technologies are the distribution of smartphones and the development of info- communications. In the Internet of Things (IoT) market leading to hyper-connectivity, Bluetooth SIG announced Bluetooth 5.0 in December 2016, anticipating to expand and distribute the potential possibilities of the beacon. It is anticipated that machine to machine (M2M) and Internet of Things (IoT) will be more deep rooted in our lives. However, the products of Bluetooth 4.x version are common, being commercially serviced. Therefore, it will take some time for replacements by Bluetooth 5.0 version are made completely. Such beacons are utilized through store advertisements, issuance of service coupons and indoor location search. It is anticipated in the future, various services will be provided in offline stores, including automatic payment services and services for the social minorities. Therefore, the IoT environment must be enhanced by solving the problems inherent in Bluetooth 4.x version, with the accurate and detailed indoor location positioninglocalization as the premises for high-quality services. In this paper, for indoor localization utilizing beacons, the received signal strength indication (RSSI) values of the beacon measured in railroad environment selected as the test bed was adjusted by grafting various filters, and through function analysis, the indoor location positioning appropriate for railroad environments was researched. As a method for correction, well-known digital filter was used for pre-processing filtering process. The measured values of RSSI value cannot be trusted due to various factors of noises and attenuation. Such RSSI measured values were adjusted by selecting a filter with the smallest amount of positioning error to have the closest value to the path loss model of the line of sight (LOS) environment. Also, function evaluations were made by applying multilateration method and KNN based fingerprint method. It is hoped that based on the results of this paper, the accuracy of location positioning services in various railroad environments would be enhanced.