Sensor Data Verification Using Latent Embeddings

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing systems face challenges in verifying the authenticity of sensor data, particularly in large-scale scenarios, as they struggle to differentiate between genuine and fictitious data, leading to difficulties in confirming the provision of geographical location-based services such as transportation.

Innovation Solution

A network system generates test embeddings from sensor data and compares them with reference embeddings to verify the likelihood of actual service provision, using latent dimension space representations and machine learning techniques to determine similarity scores, thereby identifying potential spoofing and ensuring data accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If the system processes large amounts of raw sensor data to verify service provision, then verification accuracy improves, but computational cost increases

Engineering Contradiction:
Improveverification accuracyVSAvoidcomputational cost
Core Design Contradiction:
Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The patent extracts only the most relevant features from raw sensor data (acceleration, velocity, position, heading) to create condensed data representations. This extraction approach maintains verification accuracy by focusing on discriminative features while reducing computational cost by eliminating redundant data processing.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The system transforms sensor data parameters into different representations (e.g., converting raw GPS coordinates into distance and bearing calculations, transforming acceleration data into velocity profiles). These parameter transformations enable efficient comparison and verification while reducing computational complexity.

Inventive Principle:
Principle #35Parameter changes

2Measurement precision

If the system uses detailed sensor data analysis to differentiate fictitious from genuine data, then detection accuracy improves, but processing complexity increases

Engineering Contradiction:
Improvedetection accuracyVSAvoidprocessing complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the verification process into distinct modular components: data collection from multiple sensors, feature extraction, route comparison, and anomaly detection. Each module handles a specific aspect of verification, reducing overall processing complexity while maintaining high detection accuracy through specialized processing at each stage.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system introduces intermediate data representations (processed sensor features, calculated route parameters, comparison metrics) that mediate between raw sensor data and final verification decisions. These intermediaries simplify the comparison process by providing standardized formats for analysis while preserving discriminative information.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Productivity

If the system verifies sensor data at large scale, then service verification coverage improves, but processing time increases

Engineering Contradiction:
Improveverification throughputVSAvoidprocessing time
Core Design Contradiction:
ProductivityVSLoss of time

Solution Approach 1:

The system performs preliminary processing of sensor data during service execution, continuously collecting and pre-processing sensor readings as the provider moves. This preliminary action prepares data for verification before it is needed, enabling faster large-scale verification by having processed data ready for comparison rather than processing everything after service completion.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP4132030B1Verifying sensor data using embeddings
Publication Date: 2024.07.10 UBER TECHNOLOGIES INC
  • EP4132030B1 patent drawingFigure 1
  • EP4132030B1 patent drawingFigure 2
  • EP4132030B1 patent drawingFigure 3A

AI summary

A network system analyzes data samples using embeddings based on, for example, symbolic representations of the data samples or representations in latent dimension space. The network system coordinates providers who provide geographical location-based services to users. The network system may receive data samples from the client device of a provider. For instance, a sensor of the client device captures the data samples during a transportation service along a particular route. To verify that the data samples accurately indicate the location or movement of the provider, the network system can generate a test embedding representing the data samples and compare the test embedding with a reference embedding. The reference embedding is generated based on data samples captured for other similar services, e.g., corresponding to providers who also provided transportation services along the same particular route.