Neural-Network ToA Estimation for NLOS Channel Impulse Responses
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing RF positioning systems face challenges in accurately estimating time of arrival (ToA) due to impairments from non-line-of-sight (NLOS) and obstructed-line-of-sight (OLOS) conditions, synchronization errors, motion fading, clock drifts, and environmental changes, leading to localization errors and high computational complexity.
Innovation Solution
A method using a neural network or function approximator, trained on channel impulse responses (CIRs), infers ToA by classifying mobile antennas based on movement profiles and adapting to changing conditions, reducing computational effort through intermittent retraining and classification into plausible and implausible groups.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional ToA estimation methods are used, then computational complexity is high, but accuracy in NLOS and OLOS conditions is poor
Solution Approach 1:
The patent replaces traditional signal processing methods (autocorrelation, cross-correlation, spectral analysis) with a neural network-based system. The neural network is trained on simulated CIR data and directly outputs ToA estimates, eliminating the need for complex mathematical computations during real-time operation. This substitution of mechanical/mathematical processing with neural network inference significantly reduces computational complexity while improving accuracy in challenging propagation conditions.
Solution Approach 2:
The patent uses simulated channel impulse responses (CIRs) as copies of real-world propagation conditions to train the neural network. By training on synthetic data that replicates various NLOS and OLOS scenarios, the network learns to recognize patterns and estimate ToA accurately without requiring complex real-time analysis of actual signals, thereby simplifying the computational process.
2Adaptability or versatility
If neural network is trained continuously to adapt to environmental changes, then adaptability improves, but computational effort increases
Solution Approach 1:
The patent implements periodic retraining of the neural network at predetermined time intervals rather than continuous training. This periodic action allows the system to adapt to environmental changes while controlling computational resources. The network is retrained at specific moments when environmental conditions change significantly, balancing adaptability with computational efficiency.
Solution Approach 2:
The patent performs preliminary training of the neural network on simulated data before deployment. This preliminary action prepares the network with general knowledge of various propagation conditions, allowing it to perform accurately in real-world scenarios without requiring constant retraining. The simulated training data pre-learning enables the network to handle diverse environmental conditions efficiently during actual operation.
Data Source
AI summary
Methods and systems related to neural networks or other function approximators operate for training a neural network is provided, or another function approximator, for inferring a predetermined time of arrival of a predetermined transmitted signal on the basis of channel-impulse-responses, CIRs, of transmitted signals between a mobile antenna and a fixed antenna, the method having: obtaining a channel impulse response condition characteristic, CIRCC, descriptive of channel impulse responses of transmitted signals associated with mobile antenna positions within a reach of the fixed antenna; generating, by simulation, a training set of simulated CIRs which are associated with different times of arrival in one or more simulated scenes, and which fit to the CIRCC; training the neural network, or other function approximator, using the simulated CIRs and the different associated times of arrivals to obtain a parametrization of the neural network, or other function approximator, associated with the CIRCC.