Piracy Detector Condensing CDN Logs for OTT Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing anti-piracy solutions face challenges in efficiently utilizing CDN logs for over-the-top piracy detection due to the enormous amount of data, making it cost-prohibitive and impractical.
Innovation Solution
A piracy detector that extracts and transforms CDN log records into watch session records, using a real-time event processing framework to distribute and process data across multiple nodes, reducing the data volume and identifying anomalies through clustering, thereby enhancing the accuracy and efficiency of piracy detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If raw CDN logs are used for piracy detection, then detection accuracy is improved, but storage and processing costs become prohibitive
Solution Approach 1:
The patent extracts only the essential fields from raw CDN logs (client IP, user agent, timestamp, requested URL) to create condensed log records. This extraction process removes redundant information while preserving the key elements needed for piracy detection, thereby reducing data volume while maintaining detection accuracy.
Solution Approach 2:
The patent segments the piracy detection process into multiple stages: log collection, condensation, clustering, and analysis. By dividing the enormous log data into manageable segments processed in parallel across multiple computing nodes, the system can handle large volumes of data efficiently without requiring excessive storage or processing resources.
2Reliability
If all CDN log data is processed, then piracy detection completeness is improved, but processing time increases
Solution Approach 1:
The patent performs preliminary condensation of log data into a standardized format with only essential fields before the main piracy detection analysis. This preliminary processing step organizes the data in advance, enabling faster subsequent processing and clustering operations while ensuring no critical information is lost.
Solution Approach 2:
The patent processes log data in parallel batches across multiple computing nodes, processing only the necessary portions of data required for detection at any given time. This approach maintains detection completeness by covering all logs eventually, but reduces overall processing time through concurrent execution and selective processing of relevant data segments.
3Measurement precision
If detailed log analysis is performed, then piracy detection accuracy is improved, but processing complexity increases
Solution Approach 1:
The patent applies different processing treatments to different aspects of the log data. Essential fields like client IP and user agent undergo condensation and clustering, while other fields are either discarded or processed with appropriate granularity. This localized quality approach ensures detailed analysis where needed while avoiding unnecessary complexity elsewhere.
Solution Approach 2:
The patent introduces an intermediary condensation step that transforms raw CDN logs into a standardized, simplified format before feeding data to the piracy detection algorithms. This intermediary representation layer decouples the complexity of raw log parsing from the detection logic, making the overall system more manageable while preserving detection accuracy.
Data Source
AI summary
Techniques for detection over-the-top piracy are described. In some embodiments, a piracy detection method is performed at a server by a piracy detector. The piracy detector obtains records associated with requests for access from a plurality of client devices. The piracy detector further distributes the records to a plurality of nodes according to distribution keys extracted from the records, where each of the plurality of nodes receives a respective set of records associated with a respective distribution key and generates a set of respective watch session records based on the respective set of records. The piracy detector also generates watch session records associated with the distribution keys by aggregating the respective watch session records from the plurality of nodes. The piracy detector additionally identifies one or more pirated client devices among the plurality of client devices based on clusters established from the watch session records.