Mobile Sensor Verification Using Vehicle Context and Time Sampling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for verifying the reliability of sensor data from mobile devices, particularly in driving scenarios, face challenges in ensuring that the data is accurate and reliable, which is crucial for preventing collisions and managing insurance discounts.
Innovation Solution
A method and system that involves receiving initial and subsequent sensor data from a mobile device, comparing the data to determine if it was collected while the device is in a vehicle, and transmitting notifications if the data is unreliable, with the use of machine learning algorithms to set conditions based on historical data and user patterns.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If sensor data is collected and transmitted from mobile devices to monitor driving behavior, then driving risks can be reduced and insurance discounts can be managed, but the reliability and accuracy of the sensor data cannot be ensured
Solution Approach 1:
The system performs preliminary actions by receiving and analyzing sensor data at multiple time points (first sensor data at first time, second sensor data at second time) before making reliability determinations. This allows the system to verify vehicle context in advance before using the data for insurance or safety decisions, ensuring data reliability while maintaining a manageable verification process through structured temporal sampling.
2Measurement precision
If the system verifies sensor data reliability by analyzing multiple time points and conditions, then data accuracy improves, but the time and computational resources required increase
Solution Approach 1:
The system implements periodic action by collecting sensor data at structured time intervals (first time, second time following by time interval) rather than continuously. This periodic sampling approach maintains measurement precision for verifying driving context while significantly reducing the time and computational resources required compared to continuous monitoring, as the system only needs to analyze discrete time points to determine reliability.
3Adaptability or versatility
If the system uses machine learning algorithms to analyze historical data and set conditions for reliability verification, then the ability to detect genuine driving patterns improves, but the complexity of the verification process increases
Solution Approach 1:
The system applies preliminary action by pre-training machine learning algorithms with historical sensor data to establish baseline driving patterns and reliability conditions before actual verification occurs. This allows the system to develop adaptive recognition capabilities for genuine driving patterns in advance, so that during operational verification, the complexity is managed through pre-established models rather than real-time complex analysis.
Data Source
AI summary
Method and system for verifying a reliability of sensor data received from a mobile device of a user are disclosed. For example, the method includes receiving first sensor data collected and/or generated by one or more sensors of the mobile device from an application installed on a mobile device of a user at a first time, receiving second sensor data collected and/or generated by the one or more sensors of the mobile device from the application at a second time, determining whether the mobile device is in a vehicle that the user is driving during a time interval based at least upon the first sensor data and the second sensor data, and in response to the mobile device not being in the vehicle that the user is driving during the time interval, transmitting a notification to the mobile device indicating that the application does not work properly.