Background Application Activity Detection via Bandwidth Pattern Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing audience measurement methods cannot accurately determine the activity state of background applications on consumer devices due to limitations in accessing application status, leading to background applications being assumed inactive.
Innovation Solution
The system detects distinct bandwidth usage patterns to identify active background applications, tracking activity status and crediting applications for both foreground and background usage by monitoring bandwidth usage and generating log files for crediting engines.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If background applications are assumed inactive due to limitations in accessing application status, then device complexity and power consumption are reduced, but measurement precision and accuracy of audience measurement are worsened
Solution Approach 1:
The patent introduces bandwidth usage as an intermediary indicator to indirectly determine application activity state. Instead of directly accessing application status (which requires complex system permissions), the system monitors network bandwidth consumption patterns to infer whether background applications are active, thus resolving the contradiction between measurement accuracy and system complexity
Solution Approach 2:
The patent replaces the mechanical approach of directly querying application status through operating system APIs with a statistical approach using bandwidth usage patterns. By substituting direct status detection with indirect statistical inference based on network traffic characteristics, the system achieves accurate activity detection without requiring complex system access mechanisms
2Measurement precision
If direct access to application status is enabled to accurately detect background application activity, then measurement precision is improved, but ease of operation and system stability are worsened due to increased system complexity and potential conflicts
Solution Approach 1:
The patent uses bandwidth usage as a mediator that operates independently of direct application status access mechanisms. This intermediary approach allows accurate activity detection without interfering with application execution or requiring privileged access, thereby maintaining system stability while improving measurement precision
Solution Approach 2:
The system leverages bandwidth usage data that is already being consumed by applications themselves to determine their activity state. By using the applications' own network consumption patterns as the measurement signal, the system avoids creating additional system complexity or operational conflicts while achieving accurate detection
3Measurement precision
If bandwidth usage monitoring is implemented to detect background application activity, then measurement precision is improved, but use of energy by the measurement system is worsened
Solution Approach 1:
The system uses bandwidth usage data that is inherently generated by applications during their normal operation. By leveraging this already-consumed resource rather than creating new monitoring mechanisms, the system achieves accurate detection with minimal additional energy expenditure
Solution Approach 2:
The patent changes the measurement parameter from direct application status (which requires complex system access and higher computational overhead) to bandwidth usage patterns (which are easier to monitor and process). This parameter transformation enables accurate detection while reducing the energy required for measurement operations
Data Source
AI summary
Examples disclosed herein include means for comparing bandwidth usage of an application executing in a background of a device to a threshold to determine a state of the application as one of active or inactive, means for logging event records associated with the application, and means for crediting a duration of background execution of the application. In disclosed examples, the means for crediting is to determine whether the bandwidth usage pattern is spiked or continuous based on a first event record representative of background execution of the application being started, update a second event record to be representative of the background execution of the application being stopped when the bandwidth usage pattern is spiked and a timestamp of the second event record exceeds a temporal activity window, and determine the duration of the background execution of the application based on the first event record and the second event record.