UWB Distance Estimation Error Mitigation via Machine Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing localization systems face challenges in accurately estimating distances between wireless devices in harsh environments due to non-line-of-sight (NLOS) propagation effects, which introduce positive biases in distance calculations, affecting localization performance.
Innovation Solution
The use of machine learning techniques, such as support vector machines and Gaussian processes, to identify and mitigate NLOS effects by characterizing waveforms and extracting features like energy, rise time, and delay spread, allowing for accurate distance estimation and obstruction detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If traditional distance estimation methods are used in NLOS conditions, then the system can operate in harsh environments, but distance estimation accuracy deteriorates due to positive biases
Solution Approach 1:
The patent introduces an intermediary classification system that first identifies whether a signal is LOS or NLOS before performing distance estimation. This intermediary step uses machine learning classifiers to analyze signal characteristics and determine propagation conditions, allowing the system to apply appropriate correction methods for NLOS signals thereby maintaining accuracy in harsh environments
Solution Approach 2:
The patent changes the parameter estimation approach by using machine learning models to predict and correct NLOS bias. Instead of directly estimating distance from signal strength, the system estimates propagation conditions first, then applies learned correction parameters to obtain accurate distance measurements in NLOS conditions
2Measurement precision
If machine learning techniques are used to identify and mitigate NLOS effects, then distance estimation accuracy is improved, but system complexity increases
Solution Approach 1:
The patent applies preliminary action by pre-training machine learning models offline using labeled LOS/NLOS data. The classification and mitigation functions are learned in advance during a training phase, allowing the deployed system to simply evaluate pre-trained models rather than performing complex learning operations in real-time, thus reducing operational complexity while maintaining high accuracy
Solution Approach 2:
The patent substitutes traditional signal processing methods with machine learning-based classification and regression models. Instead of using conventional threshold-based LOS/NLOS detection, the system employs learned models that automatically adapt to environmental characteristics, achieving superior accuracy with relatively simple evaluation operations
Data Source
AI summary
Ultra-wide bandwidth (UWB) transmission is a promising technology for indoor localization due to its fine delay resolution and obstacle-penetration capabilities. However, the presence of walls and other obstacles present a significant challenge in terms of localization, as they result in positively biased distance estimates. Measurement campaigns with FCC-compliant UWB radios can quantify effects of non-line-of-sight (NLOS) propagation. Features of waveforms measured during a campaign can be extracted for use in distinguishing between NLOS and line-of-sight situations in embodiments of the present invention. Embodiments further include classification and regression methods based on machine learning that improve the localization performance while relying solely on the received signal. Applications for systems employing an example embodiment of the invention include indoor or outdoor search and recovery with high accuracy and low cost.