Visual Positioning System for Anti-Spoofing Location Verification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing online systems face challenges in verifying the location of client devices to prevent spoofing, as sophisticated programs can model sensor output from multiple sensors, making it difficult to distinguish between genuine and falsified pose data.
Innovation Solution
The implementation of a visual positioning system (VPS) model that compares ostensible pose data with image data captured by the client device to determine if they match, using a VPS model to output candidate poses based on received images and taking disciplinary actions if the data does not align within a threshold, such as disabling services or deactivating accounts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If sensor data from multiple sensors is required to verify device pose, then reliability of location verification is improved, but device complexity increases
Solution Approach 1:
The patent replaces the mechanical sensor-based pose verification system with a visual positioning system that uses image processing and computer vision algorithms. Instead of relying on multiple physical sensors (accelerometers, gyroscopes, magnetometers) to determine device pose, the system uses images captured by the device camera to identify landmarks and calculate pose information, thereby reducing sensor complexity while maintaining verification reliability
Solution Approach 2:
The patent introduces a visual positioning system as an intermediary between the device and the location verification process. This intermediary system processes images to extract pose information, acting as a mediator that translates visual data into location verification data, thereby reducing direct reliance on multiple sensors while improving verification reliability
2Ease of operation
If sophisticated spoofing programs model sensor output, then ease of operation for spoofing is improved, but reliability of location verification deteriorates
Solution Approach 1:
The patent replaces sensor-based verification with vision-based verification, making spoofing more difficult because it requires manipulating visual content rather than simply modeling sensor outputs. The system uses image data processed through visual positioning algorithms to verify location, which is harder to spoof than sensor data
Solution Approach 2:
The patent implements a feedback mechanism where the system continuously verifies pose consistency between visual positioning results and reported pose data. When inconsistencies are detected, the system can trigger additional verification steps or alert operators, creating a feedback loop that actively counters spoofing attempts
3Measurement precision
If VPS model compares pose data with image data, then measurement precision of location verification is improved, but device complexity increases
Solution Approach 1:
The patent uses visual positioning technology that leverages pre-built maps and computer vision algorithms to achieve high-precision location verification. By replacing complex sensor fusion algorithms with vision-based landmark recognition and pose estimation, the system achieves precise measurement while reducing the complexity of onboard processing requirements
Solution Approach 2:
The patent employs pre-built visual maps and pre-processed landmark databases that are created beforehand. These preliminary resources are stored and reused during verification, eliminating the need for real-time complex processing and reducing device complexity while maintaining high measurement precision
Data Source
AI summary
An online system uses a visual positioning system (VPS) model to verify the location of a client device for anti-spoofing measures. The online system receives ostensible pose data and image data from the client device. This pose data and image data are ostensibly captured by the client device at the same time, or within some threshold time of each other. The online system determines whether they match according to a VPS model. The online system uses the VPS model to output candidate poses for the client device based on the received image data and compares those candidate poses to the pose in the received pose data. If the differences between the candidate poses and the pose from the received pose data exceed a threshold, the online system may determine that the received pose data and image data do not match and thus are likely being spoofed.