Entropy-Based NLOS Channel Identification for Geolocation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Geolocation technologies face significant challenges in urban and indoor environments due to non-line of sight (NLOS) propagation issues, which introduce positive biases and significant localization errors, affecting the reliability and accuracy of positioning systems.
Innovation Solution
A method that identifies the transmission channel condition by estimating the entropy of the channel impulse response (CIR) to distinguish between line of sight (LOS), NLOS with a detectable direct path, and NLOS without a detectable direct path, using entropy-based algorithms for robust NLOS identification and mitigation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If geolocation systems use TOA, TDOA, or AOA techniques, then positioning can be achieved in open environments, but localization accuracy deteriorates significantly in urban and indoor environments due to NLOS propagation
Solution Approach 1:
The system performs preliminary NLOS detection and identification before final positioning calculation. By analyzing channel characteristics (multipath components, delay spread, signal strength) in advance, the system identifies NLOS conditions and excludes or corrects affected measurements before computing the final position, preventing NLOS errors from degrading localization accuracy
Solution Approach 2:
The system changes the parameters used for positioning by selecting different measurement combinations based on detected channel conditions. When NLOS is detected, the system switches from using direct TOA/TDOA/AOA measurements to alternative parameters such as RSSI-based positioning or hybrid methods that incorporate NLOS mitigation algorithms, thereby maintaining accuracy in obstructed environments
2Device complexity
If all channel measurements are assumed to be LOS, then positioning calculation is simplified, but significant positioning errors occur due to NLOS corrupted measurements
Solution Approach 1:
The system performs preliminary NLOS detection and identification before final positioning calculation. By analyzing channel characteristics (multipath components, delay spread, signal strength) in advance, the system identifies NLOS conditions and excludes or corrects affected measurements before computing the final position, preventing NLOS errors from degrading localization accuracy
Solution Approach 2:
The system introduces an intermediary NLOS detection module that acts as a mediator between raw channel measurements and the positioning algorithm. This module analyzes channel impulse response, multipath characteristics, and signal strength to identify NLOS conditions, then provides corrected measurement sets to the positioning algorithm, thereby maintaining accuracy without requiring complex modifications to the core positioning mathematics
3Measurement precision
If existing NLOS identification techniques like RMS delay spread or kurtosis are used, then some NLOS detection capability is achieved, but accuracy is insufficient due to limited statistical information utilization
Solution Approach 1:
The system employs a multi-functional approach that simultaneously extracts and utilizes multiple statistical characteristics from the channel impulse response including mean excess delay, RMS delay spread, kurtosis, and signal strength. By combining these multiple indicators in a unified NLOS detection framework, the system achieves more accurate and reliable NLOS identification than single-metric methods, fully harnessing the statistical information contained in the channel measurements
Data Source
Figure 1(a)~1(c)
Figure 2(a)~2(c)
Figure 3~4
AI summary
This invention relates to methods and devices for channel identification. The invention is particularly concerned with techniques for non-line of sight channel identification. In embodiments of the invention the methods and devices are used for channel identification in wireless geolocation systems. Embodiments of the invention make use of an entropy estimation of the channel to distinguish channel conditions and in particular to identify line-of-sight and non-line-of-sight channels and which can be used to solve the NLOS problem of determining relative distances between transmitter and receiver. In particular embodiments an entropy estimation of the channel impulse response (CIR) is used to construct a robust entropy-based channel identification technique. As a result, more accurate localization in indoor and other multipath environments may be possible.