UWB Smart Key Location Classification via Machine Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current passive entry systems using low-frequency radio technologies face challenges in accurately determining whether a smart key is inside or outside a motor vehicle due to interference from complex vehicle geometries and reflective environments, leading to unreliable location classification.
Innovation Solution
The use of machine learning models combined with environment and distance information, particularly through ultra-wideband (UWB) radio technology, to classify the location of a smart key by analyzing reception signals and accounting for environmental reflections, thereby improving the accuracy of smart key location determination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If low-frequency radio technology is used for passive entry systems, then the system can establish connection and perform basic authentication, but the location determination accuracy deteriorates due to interference from complex vehicle geometries and reflective environments
Solution Approach 1:
The patent introduces an intermediary classification process that acts as a mediator between the low-frequency connection establishment and the ultra-wideband location measurement. This classification system uses machine learning models to interpret the relationship between UWB reception power and smart key location, resolving the contradiction by providing an intermediate layer that translates raw signal data into reliable location classifications despite environmental interference
Solution Approach 2:
The patent changes the frequency parameter from low-frequency radio technology to ultra-wideband technology for the locating process. This parameter change enables more accurate distance measurement through time-of-flight calculations while using machine learning models to compensate for the increased sensitivity to environmental reflections, thereby improving location determination accuracy while maintaining connection reliability
2Measurement precision
If ultra-wideband radio technology is used to improve location accuracy, then the smart key position can be determined more precisely, but the system becomes more sensitive to environmental reflections and complex geometries
Solution Approach 1:
The patent converts the harmful effect of environmental reflections into a beneficial feature by using machine learning models trained on data that includes various reflection scenarios. Instead of trying to eliminate reflections, the system learns to recognize patterns in reflected signals and uses this knowledge to accurately determine smart key location even in the presence of complex vehicle geometries and reflective environments
Solution Approach 2:
The patent performs preliminary training of machine learning models using extensive measurement data collected in various environments before deployment. This preliminary action prepares the system to handle environmental reflections and complex geometries by pre-learning the relationships between UWB signal characteristics and smart key location, enabling accurate location determination when the system is actually in use
3Measurement precision
If machine learning models are used to classify smart key location, then the accuracy of location determination is improved, but the device complexity increases due to training data collection and model implementation
Solution Approach 1:
The patent implements a self-service approach where the system automatically collects training data during normal operation and continuously improves its machine learning models without requiring extensive manual intervention. The system uses real-world measurement data from various environments to train and refine its location classification algorithms, reducing the complexity burden on developers and deployment processes
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enhances the reliability of smart key location classification by accounting for environmental influences, reducing incorrect classifications and improving the robustness of the locating process in various settings, including complex geometries and reflective environments.
Implementation Method 1
determine environment information relating to the receiver... determining a relative position of the transmitter with respect to the receiver on the basis of the environment information and the reception signal... analyzing the reception signal with respect to a reflection at an environmental structure
Implementation Method 2
A precise time-of-flight (ToF) measurement can be carried out by using a very broad spectrum... Measuring the ToF then makes it possible to calculate the distance between the transmitter and the receiver using the constant of the speed of light
Data Source
AI summary
A method for a receiver locates an authentication unit of a motor vehicle. The method includes determining environment information relating to the receiver. The method also includes receiving a reception signal from a transmitter. The method further includes determining a relative position of the transmitter with respect to the receiver on the basis of the environment information and the reception signal.