Adaptive UWB Indoor Positioning System

Non-Line of Sight (NLOS) and multipath propagation of a wireless signal can significantly impact the performance of an Ultra Wide-Band (UWB) based indoor localization system. While UWB can pinpoint the location of its tag to the centimeter level, it struggles in obstructed conditions. The existing error mitigation algorithms to improve the localization accuracy of UWB are either by using specialized positioning techniques or by compensating errors through NLOS error mitigation techniques. However, they may require additional measurements or depend on prior knowledge and assumptions. To tackle this problem, this paper proposes an adaptive algorithm that does not require prior data and leverages accessible parameters. The proposed method adapts to environmental and signal conditions to ultimately select an appropriate localization technique based on the signal-to-noise ratio (SNR) and the path loss exponent or can also be referred to as attenuation exponent of the transmitted signal. Using the proposed algorithm, we simulate a UWB localization system and explore its performance under a system parameter, which is SNR, and an environmental parameter, which is a path loss exponent. The results show approximately 10 % improvement at moderately NLOS conditions. The paper sets the basis for future real-world experiments and opens up new approaches that look into the different signal transmission and reception real-time parameters to achieve higher accuracy and efficiency.