Kapalı alan konum belirleme ve navigasyon tekniklerinin farklı uygulama alanlarında etkin kullanımına ilişkin bir süreç modeli yaklaşımı

Abstract

Bu tez çalışmasında kapalı mekan konum belirleme teknikleri, bu teknikler arasındaki farklar, performans kriterleri incelenmiştir. Kapalı mekan konum verilerinin kullanıldığı harita üretim teknikleri ve sektörde kullanılan çözümler incelenmiştir.Konum bulma algoritmalarından triangülasyon, trilaterasyon ve parmak izi (fingerprint) yöntemleri ile en kısa yol algoritmalarından Bellman Ford, Dijkstra,Floyd-Warshall Ford, A* arama algoritmaları incelenmiş ve bu yöntemlerin farklı bina türlerine göre avantajları ve dezavantajları belirlenmiştir. Uygulamada en yaygın kullanılan kapalı mekan konum belirleme algoritmaları triangülasyon ve parmak izi yöntemleridir. Ayrıca uygulamada bluetooth cihazları ile kapalı mekan konum belirleme WiFi teknolojisine göre daha yaygın tercih edilmektedir. Düşük maliyet ve mobil cihazlarda uyum sorunu olmamasından dolayı ekonomik ve pratik olarak avantajlı olduğu görülmektedir. Kapalı mekan konum belirleme çözümlerinde şimdiye kadar belirlenmiş bir standart bulunmadığı için pek çok firma kendine göre yazılım çözümleri geliştirmektedir. Benzer bir sorun işletim sistemi geliştiren firmaların cihazlar arası haberleşmek için kendi belirledikleri standartlarda yazılım çözümleri geliştirdikleri ve bu cihazların teknolojinin ilerlemesi ile farklı cihazlar ile haberleşmesi ihtiyacı pek çok sorunu beraberinde getirmiştir. Kapalı mekan navigasyon çözümlerinde OSI Katmanları referans alınarak altı katmandan oluşan standart önerisi yapılmıştır. Bu katmanlardan Fiziksel Katman, Network Katmanı, Sinyal Katmanı, Algoritma, Konumsal Katman ve LBS Katmanı olarak adlandırılmış ve veri iletişim sırasının bu katmanları üzerinden sırası ile işlenmesi gerektiği anlatılmıştır. Ayrıca Kapalı Mekan Konum belirmele tekniklerinde performans kriterleri üzerinde detaylı incelemer yer almaktadır. Doğruluk, Hassasiyet, Kapsama,Uyumluluk ve Ölçeklenebilirlik olarak belirlenen bu performans kriterleri incelenmiştir. Kapalı mekan konum belirleme işi için yol haritası çıkarılmış, bu yol haritası planlama,veri toplama,harita üretimi ve konumsal hizmet süreci olarak dört ana başlıkta incelenmiştir. Bu yol haritasında bir bina için navigasyon hizmeti sürecindeki tüm adımlar madde madde sıralanmış, işin başlangıçtan hizmet sürecine kadar olan tüm kısımlar detaylandırılmıştır.Konum belirleme ile birlikte kapalı mekan harita üretimi de önem kazanmaktadır. Bu konuda OGC'nin de standart oluşturma çalışmaları bulunmaktadır. Hatta yakın zamanda OpenStreetMap (OSM) uygulamasının harita çizim uygulaması olan OSM for Java (JOSM) için farklı indoor araçları geliştirilmiştir. Bu araçlar ile IndoorGML standartlarına uygun kapalı mekan harita üretiminde kolaylık sağlamaktadır.

In this thesis, indoor location techniques, differences between these techniques and performance criteria are examined. Map production techniques used in indoor location data and solutions used in the sector are examined.The search algorithms of triangulation, trilateration and fingerprint methods and shortest path algorithms such as Bellman Ford, Dijkstra, Floyd-Warshall Ford and A* and their advantages and disadvantages have been determined according to different building types. The most common indoor location algorithms used in practice are triangulation and fingerprinting methods. In addition, indoor location with bluetooth devices is preferred more widely than WiFi technology in practice. It appears because of the low cost and lack of compatibility issues in mobile devices. Since there is no established standard for indoor positioning systems, many companies are developing their own software solutions. A similar problem is that companies that develop operating systems have developed software solutions to communicate between their devices, and these devices needs to communicate with different devices with the progress of technology has brought with it many problems. In the indoor navigation solutions, a standard proposal consisting of six layers was made taking OSI Layers as reference. These layers are called Physical Layer, Network Layer, Signal Layer, Algorithm, Positional Layer and LBS Layer and it is explained that the data communication sequence must be processed through these layers in turn. Also detailed analysis on the performance criteria in the Indoor positioning systems techniques is given. These performance criteria, which are determined as Accuracy, Sensitivity, Coverage, Compatibility and Scalability, have been examined. A road map has been prepared for indoor location determination work, and this road map has been examined in four main topics as planning, data collection, map production and location service process. In this roadmap, all the steps in the navigation service for a building are itemized, detailing everything from the start to the service period.Indoor location map production is also important with location determination. In this regard, OGC also has standardization studies. In addition, recently, different indoor tools have been developed for OSM for Java (JOSM), the map drawing application of OpenStreetMap (OSM) application. With these tools, it provides convenience in the indoor map production according to the IndoorGML standards.In this thesis study, fingerprinting, triangulation and calculation methods based on RSSI have been investigated from indoor positioning methods. As a result of the analysis, requirements analysis should be done firstly for Indoor positioning systems studies. How much energy is consumed by the devices to be installed, what accuracy the system should aim for, and so on. The technologies, hardware and algorithms that can be used according to the infrastructure information may differ. The accuracy and visualization of the map used in the user interface is as important as the least method. The maps in the user interface may be 2D or 3D, but more importantly, the mentioned maps must be built on the basis of a spatial data standard such as OGC. In practice many companies develop their system based on the local coordinate system. Although this is not a problem for small areas, it creates problems in large and complex areas. In addition, outdoor and indoor locations relationships can not be established directly with this type of approach, and this compromises the integrity of the navigation. Furthermore, the use of city and building standards such as CityGML, BIM, and IndoorGML will become more prominent for global coordinate systems, and interoperability problems will arise with systems developed based on the local coordinate system.Wireless radio technology is most commonly used in indoor environments and the easiest accessible WiFi technology to achieve indoor positioning accuracy of about 5-15 meters can be achieved. Although WiFi technology can be used directly in some cases without the need for infrastructure, higher accuracy is not sufficient for applications that require it. Considering the wireless technologies covered in the thesis study, it was determined that Bluetooth technology is a suitable solution for Indoor positioning systems due to its superiority to other technologies and the properties explained in the thesis. Indoor positioning applications where Bluetooth technology is used can achieve accuracy of 1-5 meters. The fact that Bluetooth technology consumes a small amount of energy since version 4 has led BLE 4.x and 5.0 technologies to be preferred in closed areas.Although Bluetooth technology consists of 79 frequency bands, only 40 can be used by BLE. On the other hand, it is observed that a large number of signal transmitters are used in practice in order not to have areas outside the scope of signal transmission in wide areas. In this case, the number of signals per unit area from different signal transmitters increases and temporal differences occur in instantaneous data transmission as it fills the range of 40 frequencies. The mentioned temporal differences also cause the algorithm to account for the wrong position.The best example of this and similar situations is that indoor navigation solutions in fairgrounds do not work efficiently. In addition to signal transmitters in the exhibition area, the BLE signals in the personal devices of the visitors also increase the number of signals per unit area. For this reason, the use of the triangulation method is recommended because the application of the fingerprint method in such application areas will not produce accurate results. The triangulation method can not be developed with fewer signal transmitters than the fingerprint method. Especially with Bluetooth v5 technology, the distance between the signal data and the receiver is 4 times higher than the v4 technology, which will reduce the number of transmitters in wide area.Indoor positioning systems are not only used for positioning and navigation purposes. Especially when integrated with location-based services, many applications can be developed that create added value. Examples of these are process improvement in the retail sector, customer satisfaction, and the fact that people in the banking sector are able to get a queue number without physical action when they arrive in the bank.