Sensor Data Collection for Fraud Detection in Electronic Devices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing electronic devices face challenges in efficiently collecting and processing sensor data for fraud detection while minimizing resource consumption and maintaining user privacy, as continuous data collection can burden device resources, consume power, and expose sensitive information.
Innovation Solution
Implementing a data collector module that selectively activates and deactivates data collection based on page risk, using a SharedWorker module to retain and transmit data, and employing data obfuscation to maintain anonymity, with a Defensive Module that operates on a dedicated processing thread and transmits data periodically to conserve resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If continuous sensor data collection is implemented for fraud detection, then fraud detection capability is improved, but device resource consumption increases
Solution Approach 1:
The system implements periodic data collection and transmission instead of continuous operation. The data collector module activates to collect sensor data at specific intervals and transmits aggregated data periodically to the server, reducing continuous resource consumption while maintaining fraud detection effectiveness through sufficient data sampling over time.
Solution Approach 2:
The system dynamically adjusts data collection behavior based on risk assessment. The Defensive Module evaluates page risk levels and dynamically modifies collection intensity - increasing data gathering when fraud risk is detected and reducing or suspending collection when risk is low, optimizing the balance between detection capability and resource usage.
2Measurement precision
If comprehensive sensor data is collected for fraud detection, then detection accuracy is improved, but user privacy is compromised
Solution Approach 1:
The system extracts and transmits only essential fraud-relevant features from comprehensive sensor data rather than transmitting all raw data. The data collector identifies and extracts specific patterns and characteristics necessary for fraud detection while leaving sensitive personal information on the device, achieving accurate detection without full privacy exposure.
Solution Approach 2:
The system introduces data aggregation and anonymization as intermediary processing steps between data collection and server transmission. The Defensive Module processes raw sensor data locally to create aggregated, anonymized representations that preserve fraud detection accuracy while removing personally identifiable information, acting as a privacy-protecting intermediary layer.
3Speed
If data is transmitted frequently to the server, then real-time fraud detection is improved, but network bandwidth consumption increases
Solution Approach 1:
The system implements periodic batch transmission of aggregated data instead of continuous real-time streaming. Data is collected and accumulated locally over defined intervals, then transmitted in periodic batches to the server, reducing network bandwidth consumption while maintaining timely fraud detection through efficient use of transmission windows.
Solution Approach 2:
The system performs preliminary data aggregation and preprocessing locally before transmission. The data collector module prepares and consolidates sensor data in advance during collection periods, reducing the volume and frequency of network transmissions required while ensuring detection readiness when data is sent to the server.
Data Source
AI summary
System and method of collecting and processing data in electronic devices. A sensors data collector collects measurements from at least an accelerometer and a gyroscope of an electronic device. A data-loss prevention module operates to pass these measurements, immediately upon their collection, to a supplemental locally-running processing thread which retains the measurements even after a refresh of a web-page in which the measurements were collected, and which transmits the measurements to a remote server even after refresh of the web-page in which the measurements were collected. Non-global scope of functions is utilized, to reduce security exposure. An asynchronous SharedWorker module is utilized, to alleviate congestion of computing resources of the electronic device. Data obfuscation and encoding is utilized to maintain anonymity of user-entered data while still allowing a remote server to ensure the integrity of data received from the electronic device.