Vehicle Sensor Node Filtering for NLOS Device Localization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Localization routines using UWB or BLE transceivers in vehicles are inaccurate due to signal distortion and reflection by reflective objects in NLOS environments, leading to improper vehicle function access.
Innovation Solution
A method and system that determines target device distances using vehicle sensor nodes, identifies a reference distance, and assigns nodes to reflecting and non-reflecting sets based on threshold comparisons, using non-reflecting node data to improve localization accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If UWB or BLE transceivers are used to detect objects and perform localization routines, then vehicle function access can be provided, but localization accuracy deteriorates due to signal distortion and reflection by reflective objects in NLOS environments
Solution Approach 1:
The patent segments the sensor nodes into two distinct sets: reflecting sensor nodes and non-reflecting sensor nodes. This segmentation is based on comparing target device distances from multiple sensor nodes to identify which nodes are affected by signal reflection. By separating the nodes, the system can selectively use only the non-reflecting nodes for localization calculations, thereby maintaining reliability while improving measurement precision.
2Quantity of substance
If sensor data from all vehicle sensor nodes is used for localization, then more data is available for determining positional characteristics, but measurement precision deteriorates due to inclusion of distorted data from reflecting nodes
Solution Approach 1:
The patent extracts and removes the harmful reflecting sensor nodes from the dataset used for localization. By identifying nodes whose measured distances deviate significantly from the median distance (indicating reflection effects), the system extracts only the clean, non-reflecting sensor data for determining positional characteristics. This extraction process maintains sufficient data quantity while eliminating the precision-deteriorating distorted measurements.
3Adaptability or versatility
If signal reflection by objects is present in the environment, then NLOS conditions occur, but localization accuracy deteriorates due to distorted and reflected signals
Solution Approach 1:
The patent converts the harmful effect of signal reflection into a beneficial identification mechanism. By analyzing the distorted signals caused by reflection, the system identifies which sensor nodes are affected and excludes them from localization calculations. The reflection, while harmful to direct measurement, provides a detectable anomaly that enables the system to adaptively select only the reliable non-reflecting nodes, thereby maintaining environmental adaptability while preserving localization accuracy.
Data Source
AI summary
A method includes determining a plurality of target device distances between a plurality of vehicle sensor nodes of a vehicle and the target device, identifying a reference target device distance from among the plurality of target device distances based on a magnitude of the plurality of target device distances, and determining a threshold comparison distance based on the reference target device distance, a reference intranode distance, and a predetermined error distance. The method includes assigning the plurality of vehicle sensor nodes to one of a reflecting vehicle sensor node set and a non-reflecting vehicle sensor node set based on a comparison between the threshold comparison distance and the plurality of target device distances and determining a positional characteristic of the target device based on one or more target device distances associated with one or more vehicle sensor nodes from among the non-reflecting vehicle sensor node set.