Research On Trajectory Analysis On Indoor Moving Objects

Trajectories of moving objectsrecord the positionsof objects that change with time, which are usually represented as aseries of locations in a certain time period. In recent years, the development of RFID, Wi-Fi, and Bluetooth technologies makes it possible forindoor positioning. According to a recent survey, people spend80percentages of their lives in indoor space. Therefore, with the development of indoor positioning technologies, indoor location-based services will be the trend in the future. Indoor location-based services need to analyze the trajectories of indoor moving objects to satisfy the needs of users. But currently researches on trajectory analysis on indoor moving objectsare still at a very early stage, and many problems and issues remain unsolved.In this paper, we focus on two issues related with indoor trajectory analysis, namle the generation of simulated trajectories of indoor moving objects, and indoor hotspots query. As deploying real indoor positioning devicesis time and money consuming, it is not feasible to construct different environment settings for various indoor applications. Aiming to provide experimental trajectory data for various indoor space related algorithms, we present a flexible tool to generate trajectories for indoor moving objects, which is named IndoorSTG (Indoor Spatiotemporal Trajectory Generator). IndoorSTG can simulate different indoor environments using various elements and virtual positioning devices. Meanwhile, it can generate semantic-based trajectories for indoor moving objects in a specific indoor space. Further, in this paper we analyze the trajectories of many indoor moving objects, and propose object-based indoor hotspot query method. Finally, we use the trajectories dataset of indoor moving objects generated by IndoorSTG tool for validation.The main contributions this paper can be summarized as follows:(1) We present a flexible tool to generate trajectories for indoor moving objects, which is named IndoorSTG. IndoorSTG can simulate different indoor environments using various elements including rooms, doors, corridors, stairs, elevators, and virtual positioning devices such as RFID or Bluetooth readers. Meanwhile, it can generate semantic-based trajectories for indoor moving objects in a specific indoor space. At first, according to the characteristics of a variety of real-world indoor moving objects and correlation between moving objects and indoor locations, moving objects and indoor location can carry classification, and we can establish a moving object-position relationship. For instance, in a college, teachers or postgraduates are usually associated with their office or lab, who spend more time in moving in their related locations. In order to more accurately simulate a variety of complex indoor environments, indoorSTG allows users to set the indoor parameter configuration and to complete the layout of the indoor environment. Thus IndoorSTG has better scalability and universality.(2) We present an effective algorithm to answer indoor hotspot region (iHR) queries.We first define the problem of indoor hotspot region queries. Then, we propose the algorithm for trajectory-based indoor hotspot region queries that takes into account the specialproperties of indoor trajectories. We start from establishinga staying model for indoor moving objects based on their trajectories. This model is used toremovesome redundant trajectories to improve the efficiency of the following analysis. After that, we build a matrix maintaining the relations between objects and locations. Finally, we calculate the hotness of candidate hotspot areas on the basis of the matrix, and find out all hotspot regions in indoor environment. The results of comparison experimentsdemonstrate the effectiveness of the algorithm.