Distributed Antenna Position Estimation for Urban NLOS Ranging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current localization systems face challenges in urban environments due to signal degradation, multipath effects, and obstructions, leading to reduced accuracy and reliability, particularly in applications like autonomous driving and VRU protection, with UWB performance being affected by NLOS conditions.
Innovation Solution
A method integrating a distributed antenna system with neural network models, such as CNNs and RNNs, for enhanced position estimation, leveraging spatial and temporal patterns to improve accuracy and robustness, especially in complex environments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional localization systems (GPS, lidar) are used in urban environments, then the systems can provide basic positioning functionality, but the accuracy and reliability are significantly reduced due to signal degradation, multipath effects, and obstructions
Solution Approach 1:
The system segments the positioning task by using multiple distributed antennas instead of a single antenna, allowing the system to process signals from multiple spatial locations independently and combine results to overcome obstructions and multipath effects
Solution Approach 2:
The patent introduces neural network models as intermediary processing layers between raw signal measurements and final position estimates. These neural networks learn to compensate for signal degradation and identify reliable signal paths, effectively mediating between the degraded urban signals and accurate positioning requirements
2Measurement precision
If UWB technology is used for ranging procedures, then high precision and low interference characteristics are achieved, but performance is significantly affected by Non-Line-of-Sight (NLOS) conditions leading to increased errors and data loss
Solution Approach 1:
The patent combines measurements from multiple distributed antennas with neural network processing to create a hybrid positioning system. This merging allows the system to leverage the high precision of UWB ranging while compensating for NLOS errors through spatial diversity and learned correction models
Solution Approach 2:
The system changes the parameter of signal processing by applying neural network-based signal processing that adapts to NLOS conditions. The neural networks learn optimal parameter adjustments for ranging calculations under different propagation conditions, transforming reliable LOS measurements into robust NLOS-tolerant estimates
3Measurement precision
If distributed antenna systems with multiple antennas are deployed to improve positioning accuracy, then multiple reference points are provided for signal reception, but the device complexity and system cost increase
Solution Approach 1:
The patent makes the distributed antenna system multi-functional by using the same antenna array for both traditional communication functions and enhanced positioning. The neural networks process signals for multiple purposes, allowing the hardware to serve multiple functions without proportional increases in complexity
Solution Approach 2:
Instead of deploying a completely new dedicated positioning infrastructure, the system copies and repurposes existing distributed antenna structures (such as those already present in V2X communication systems) for positioning applications, avoiding the need for entirely new hardware deployments
Data Source
AI summary
A method for enhanced position estimation using a distributed antenna system. The method includes: receiving input data based on at least one measurement resulting from a ranging procedure using a distributed antenna system, wherein the ranging procedure specifies measures for determining a distance dx depending on a signal propagation regarding the ranging procedure; providing at least one neural network model, wherein the at least one neural network model specifies a spatial pattern recognition and/or a temporal sequence to process the input data; combining the input data using the at least one neural network model to provide an enhanced position estimate; and providing a position result based on the combining.