Proximity Tracker using Received Signal Strength, Particle Filter and Extended Kalman Filter

Abstract

Received Signal Strength Indicator (RSSI) is a relative variable related to signal strength. The strength of the signal transmitted is used in continuous close-proximity tracking applications. The approximate exact location of a node is estimated using various kinds of filtering techniques. The demand for close-proximity tracking or location and proximity integration increases with the society’s dependency to mobile devices but using such geolocation services are power-hungry and is directly affected by atmospheric conditions. Present tracking services are made relative to the earth’s axis. To address these gaps, this study focuses on integrating Extended Kalman Filter (EKF) and Particle Filter (PF) into a system of Bluetooth-enabled nodes that are capable of relative positioning in three-dimensional space. Three nodes made of Raspberry Pi Zero are tracked by a parent device made of Raspberry Pi 3 Model B. The nodes’ coordinates are displayed in the parent device’s dashboard. Detection of the nodes are done using the library BIueZ. These nodes broadcast themselves to the parent device to determine their range and location using a 750-meter range Bluetooth dongle. The device is tested in both open areas, a memorial park and a beach resort. Using spiral method, the weakest RSSI value measured is -142dB at 776.43 meters and -23dB at 0 meter. RSSI value in an area with obstruction or interference is -168dB at a distance of693.1 meters. The output shows very promising results indicating effectiveness and efficiency in the field of proximity tracking.