Low-Latency FEC Buffering for Live Stream Error Recovery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing forward error correction (FEC) methods introduce latency, making it undesirable to wait for several seconds before viewing new video streams, especially when switching channels, and are less effective with smaller time windows.
Innovation Solution
Systems and methods that output source data at a rate less than the source stream, build a buffer, and perform FEC decoding to output packets at the source stream rate, allowing for quick output and error correction across large data windows with partial or no error correcting capability initially.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If FEC is applied to a time window of x seconds, then error correction capability is improved, but output latency increases
Solution Approach 1:
The system performs preliminary buffering of source data before FEC decoding. The buffer accumulates data at a reduced output rate, allowing the system to prepare error correction capabilities in advance without delaying the initial output. This resolves the contradiction by separating the data accumulation phase from the decoding phase, enabling low-latency output while maintaining strong FEC protection for large time windows.
2Reliability
If the time window is increased to 30 seconds or more, then FEC recovery capability from larger outages is improved, but the delay before viewing new video streams increases
Solution Approach 1:
The buffer performs preliminary data accumulation at a rate slower than the source stream, preparing sufficient data for large-window FEC decoding without waiting for the full time window to elapse. This allows the system to achieve both long time window protection and minimal delay before viewing new streams.
Solution Approach 2:
The system dynamically adjusts the output rate from the buffer, initially outputting at a reduced rate to build the buffer quickly, then transitioning to full-rate output once sufficient data is accumulated. This dynamic rate adjustment enables the system to provide immediate viewing capability while maintaining the ability to perform FEC decoding over large time windows.
Data Source
AI summary
Systems, methods, apparatus, and computer program products for providing forward error correction with low latency to live streams in networks are provided. One example method includes receiving source data at a first rate, outputting the source data at a rate less than the first rate, collecting the source data in a buffer, FEC decoding the source data, thereby generating decoded data; and outputting the decoded data at a rate equal to the first rate, either after collecting the source data in the buffer for a predetermined time duration or after collecting a predetermined amount of the source data in the buffer.