Error Determination Device for Moving Image Stream Continuity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for real-time moving image distribution over IP networks, such as those using stream distribution systems, face challenges in accurately determining errors in the stream of pictures, leading to potential failures in reproducing correct moving images even if IP packets are received without loss.
Innovation Solution
An error determination device that acquires and compares target time information for each picture with reference time information, determining if the pictures are consecutive by checking the absolute difference against a threshold, ensuring correct reproduction of the moving image.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If only IP packet loss detection and correction methods (FEC/ARQ) are used, then packet transmission reliability is improved, but stream error detection capability deteriorates
Solution Approach 1:
The patent segments the error detection process into two independent parts: IP packet level detection (using FEC/ARQ) and stream picture level detection (using the proposed PTS continuity check). This segmentation allows each layer to handle its specific error type effectively, with the stream-level detection catching errors that packet-level methods miss.
Solution Approach 2:
The patent introduces PTS (Presentation Time Stamp) as an intermediary parameter to bridge the gap between packet transmission and picture reproduction. By monitoring PTS continuity, the system can detect stream errors without directly examining picture content, enabling error detection at the stream level while maintaining packet-level error handling.
2Reliability
If stream error detection is added to IP packet error detection, then error detection completeness is improved, but system complexity increases
Solution Approach 1:
The patent extracts only the essential time information (PTS) from the stream for error detection, rather than analyzing the entire picture data or complex stream structure. This extraction approach enables simple arithmetic comparison operations while maintaining comprehensive error detection capability.
Solution Approach 2:
The system uses its own existing PTS parameter, which is already present in the stream for timing synchronization purposes, to perform error detection. This self-service approach eliminates the need for additional external detection mechanisms, reducing system complexity while improving error detection completeness.
3Manufacturing precision
If strict PTS continuity checking is performed, then picture reproduction accuracy is improved, but tolerance to incidental errors decreases
Solution Approach 1:
The patent performs partial error detection by checking only the PTS continuity condition, rather than verifying all aspects of picture reproduction. This partial action provides sufficient error detection for the critical timing aspect while avoiding false alarms from incidental errors in other non-critical parameters.
Solution Approach 2:
The patent changes the detection parameter from direct picture content comparison to PTS time stamp arithmetic comparison. This parameter change makes the detection more robust to incidental errors in picture data while maintaining sensitivity to actual timing errors that would affect reproduction accuracy.
Data Source
AI summary
A processing-coefficient decision unit decides a processing coefficient based on a format of the stream. A target-PTS decision unit multiplies an input PTS value by the processing coefficient, to set this as a target PTS used to determine the continuity of the pictures. A reference-PTS calculation unit sets the target PTS as an initial value of a reference PTS, sequentially adds an increment corresponding to the PTS value per picture, and calculates the reference PTS which is compared with the target PTS. A difference calculation unit calculates a difference between the target PTS and the reference PTS. A threshold determination unit compares an absolute value of the difference with the processing coefficient being a threshold and determines whether the absolute value of the difference is less than the processing coefficient.