Sensor fusion for indoor positioning system through improved RSSI and PDR methods

Abstract

Given the growth of the Internet of Things (IoT) and smart home appliances, the concept of the Indoor Positioning System or IPS has considerably risen. The major application of IPS is in locating people in roofed places. In this type of positioning, accuracy has always been the most important challenge. The main focus of this article is to improve the indoor positioning of people. First, two positioning approaches are proposed, Radio Signal Strength Indicator (RSSI) and Pedestrian Dead Reckoning (PDR). In the RSSI positioning part, the distance between the participating nodes is calculated based on the low-fluctuating values of RSSI. A novel filter called Weight-Based Optimization (WBO) is developed to optimize these raw RSSI values. Moreover, the path loss model parameters, which are location-dependent, are calculated by employing Particle Swarm Optimization (PSO) to convert the resulting RSSI values to distances. In the PDR positioning part, the accelerometer sensor data is used to detect the person’s steps, and the compass sensor data is used to detect the heading direction. Finally, the results of RSSI and PDR methods are combined using a sensor fusion approach. The positioning accuracy resulting from the proposed IPS approach is 68 cm, which is far better than the existing methods.