GNSS Receiver Neural Network for Doppler Error Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
GNSS receivers experience significant errors in position and velocity calculations due to multipath effects, particularly in urban environments where signals are reflected off buildings, leading to distorted pseudorange and Doppler measurements.
Innovation Solution
The use of machine learning techniques, specifically training neural networks with features extracted from GNSS signals, to predict range rate errors and correct Doppler measurements, thereby improving velocity measurement accuracy in GNSS receivers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional GNSS signal processing is used, then the receiver can obtain position and velocity solutions, but the measurements are significantly distorted by multipath effects in urban environments
Solution Approach 1:
The patent introduces an intermediary machine learning model that processes raw GNSS measurements and outputs corrected measurements. This intermediary layer filters out multipath effects by learning the relationship between signal characteristics and measurement errors, thereby improving velocity measurement accuracy without requiring changes to the fundamental GNSS signal processing architecture
Solution Approach 2:
The patent replaces traditional mechanical signal processing methods with a machine learning-based approach. Instead of using conventional filtering and correction algorithms, the system uses trained neural networks to predict and correct measurement errors, substituting computational mechanics with intelligent algorithms that adapt to varying multipath conditions
2Measurement precision
If machine learning models are trained and deployed in GNSS receivers, then velocity measurement accuracy is significantly improved, but the device complexity increases
Solution Approach 1:
The patent applies preliminary action by training the machine learning model offline before deployment. All complex training operations are performed in advance using labeled data, and the trained model is then deployed to the GNSS receiver. This approach separates the complex training phase from the operational phase, reducing the computational burden and complexity within the receiver itself while maintaining high measurement accuracy
Solution Approach 2:
The patent changes the parameters of the signal processing system by introducing learned correction terms based on signal characteristics. Instead of using fixed processing parameters, the system dynamically adjusts corrections based on features extracted from the received signals, allowing adaptive improvement of measurement accuracy without requiring a complete redesign of the receiver architecture
Data Source
AI summary
Machine learning techniques are used to compute predicted range rate errors in a GNSS receiver. In one embodiment, training data is computed to provide true range rate error data for a set of received GNSS signals. A system extracts features from the set of received GNSS signals and uses the extracted features and the true range rate error data to train a model (e.g., a set of one or more neural networks) that can produce predicted range rate errors for use in correcting measurements. The trained set of one or more neural networks can be deployed in GNSS receivers and used in the GNSS receivers to correct Doppler measurements using the predicted range rate errors provided by the trained set of one or more neural networks.