Privacy Leak Detection via Android Instrumentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing mobile computing systems lack accurate and user-friendly methods to detect privacy leaks from third-party applications without requiring platform modifications or source code access, posing challenges due to framework code and callback functions.
Innovation Solution
A system that instruments Android applications to detect privacy leaks in real-time, employing static analysis and taint propagation techniques to identify legitimate and illegitimate data leaks, without modifying the platform, and allows users to configure detection for specific applications with minimal performance overhead.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If operating system modification is used to detect privacy leaks, then detection accuracy is improved, but device complexity and ease of operation deteriorate due to requiring root access and custom firmware
Solution Approach 1:
The patent introduces an intermediary component (instrumentation library and proxy application) that mediates between the privacy detection system and the target application. This intermediary layer enables accurate privacy leak detection without requiring operating system modification, as it intercepts and monitors data flows through the application's public APIs and data storage mechanisms, thereby resolving the contradiction between detection accuracy and ease of installation
Solution Approach 2:
The patent creates a copy of the application's data storage structure (using FileObserver to monitor file system changes and proxy classes to mirror data access patterns). This copying approach allows the system to track privacy-sensitive data flows accurately without modifying the original application or operating system, thus maintaining detection precision while avoiding the complexity of system-level modifications
2Measurement precision
If comprehensive instrumentation is applied to all code paths, then detection coverage is improved, but performance overhead increases
Solution Approach 1:
The patent applies local quality by instrumenting only specific code paths and data access points that are relevant to privacy leakage detection. Instead of comprehensively instrumenting all code, the system focuses on monitoring file system operations, network connections, and data access patterns in storage mechanisms, thereby achieving effective detection coverage with minimal performance overhead
Solution Approach 2:
The patent employs partial action by implementing selective monitoring of data flows based on predefined privacy-sensitive categories. The system instruments only the necessary components (FileObserver for file monitoring, proxy classes for specific data access patterns) rather than applying full instrumentation to all application operations, thus balancing detection coverage with acceptable performance overhead
3Productivity
If static analysis is used to identify leakage paths, then instrumentation overhead is reduced, but detection accuracy may decrease due to inability to handle dynamic callbacks
Solution Approach 1:
The patent implements feedback mechanisms through dynamic monitoring components (FileObserver, proxy application) that observe actual runtime behavior and data access patterns. This feedback loop allows the system to detect privacy leaks in dynamic callback scenarios that static analysis alone would miss, thereby maintaining high detection accuracy while benefiting from the efficiency of static analysis for initial instrumentation identification
4Measurement precision
If the system monitors all data access patterns, then privacy detection accuracy is improved, but performance overhead and device complexity increase
Solution Approach 1:
The patent applies segmentation by dividing the privacy monitoring system into distinct modular components: FileObserver for file system monitoring, proxy classes for data access interception, and category-based filtering for different types of data (images, videos, contacts). This segmented architecture enables accurate privacy detection across multiple data access patterns while managing system complexity through modular, independently configurable components
Data Source
AI summary
A system and method detect privacy leaks in applications of an operating system of a mobile device. An instrumentation module permits tracking of privacy-sensitive information without modification of a middleware of the operating system and a process virtual machine.