Abstract
An indoor trajectory is the path of an object moving through corridors and stairs inside a building. There are various types of technologies that can be used to reconstruct the path of a moving object and detect its position. GPS has been used for reconstruction in outdoor environments, but for indoor environments, mobile devices with embedded sensors are used. An accelerometer sensor and a magnetometer sensor are used to detect human movement and reconstruct the trajectory on a single floor. In an indoor environment, there are many activities that will create the trajectory similar to an outdoor environment, such as passing along the corridor, going from one room to another, and other activities. We need to analyse trajectories to obtain the movement patterns, understand the most frequently visited places or paths used as well as the least frequented ones. Furthermore, we can utilize movement patterns to obtain a better building design and layout. The latest studies focus on reconstructing the trajectory on a single floor. However, actual indoor environments are comprised of multi-floors and multibuildings. The purpose of this paper is to reconstruct a trajectory in an indoor multi-floor environment. We have conducted extensive experiments to evaluate the performance of our proposed algorithms in a campus building. The result of our experiment shows that the height of the building can be detected using a barometer sensor that gives an atmospheric pressure reading which is then transformed by setting the range value according to the number of floors, enabling the sensors to detect activity in a multi-floor building. The readings obtained from the magnetometer sensor can be used to reconstruct the trajectory similar to the real path based on the direction and degree of direction. The system accuracy in recognizing steps in a multi-floor building is about 84%.