Efficient Processing of Spatio-Temporal Joins on IoT Data

As the Internet of Things (IoT) has become widespread, the demand for storing and querying data generated by things (e.g., moving sensors) is growing to obtain more useful information. One of the emerging queries on such IoT data is the spatio-temporal join, which joins data generated by different things but generated at (almost) the same time and location. In this paper, we propose an efficient method for processing spatio-temporal joins on IoT data. The proposed method divides the 3D spatio-temporal space into small, equal-sized spaces, called cells. As data is generated by things, the proposed method maintains the information about which thing's data are in which cells. When a spatio-temporal join between specified things is requested, the proposed method first identifies cells, each of which has data of all the specified things within or near it. The proposed method then retrieves only the data within or near the identified cells and performs the join only between the retrieved data. Consequently, compared with previous methods where the processing cost increases rapidly as the size of data or the number of things being joined increases, the processing cost is greatly reduced. The experimental results on a real IoT dataset show that the proposed method significantly reduces the execution time compared with the existing methods.