Client-Side FEC Selection for Scalable ABR Error Resilience
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Adaptive bitrate (ABR) streaming over unreliable transport mechanisms faces challenges in error resilience due to limitations in existing forward error correction (FEC) techniques, which are often sub-optimal and require real-time client feedback, leading to scalability issues and inefficiencies.
Innovation Solution
A system and method for dynamic client-side selection of FEC data, where FEC information is generated on a per configuration setting basis and included in manifest files, allowing clients to monitor conditions and select suitable bitrates and FEC data without real-time client feedback, thereby optimizing error correction efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If FEC data is generated in real-time based on client feedback, then error correction effectiveness is improved, but system complexity and processing power requirements increase
Solution Approach 1:
The patent pre-generates multiple FEC datasets with different configuration settings (e.g., different FEC rates, coding schemes) before transmission. The client can then select from these pre-prepared options based on observed channel conditions, eliminating the need for real-time FEC generation and complex server-side processing while maintaining effective error correction.
Solution Approach 2:
The system dynamically adapts FEC parameters by allowing the client to select from multiple pre-configured FEC datasets based on real-time channel conditions. This dynamic selection approach achieves adaptability without requiring real-time FEC generation, reducing server processing complexity while maintaining reliability.
2Adaptability or versatility
If multiple FEC configuration settings are provided, then adaptability to different channel conditions is improved, but data transmission overhead increases
Solution Approach 1:
The patent segments FEC data into multiple configuration settings (different FEC rates, coding schemes) organized in a manifest file. The client receives only the necessary FEC configuration metadata through the manifest and selectively requests or applies specific FEC datasets based on observed channel conditions, avoiding transmission of all possible FEC variants and reducing overall overhead.
Solution Approach 2:
The system provides adaptability by offering multiple FEC configuration parameters (FEC rates, coding schemes) in the manifest file. The client changes these parameters dynamically based on channel conditions, selecting from pre-defined options rather than transmitting complete alternative datasets, thus achieving versatility with minimal overhead.
3Reliability
If real-time client feedback is required for FEC selection, then error correction optimization is improved, but scalability is reduced
Solution Approach 1:
The patent enables the client to autonomously select appropriate FEC datasets by monitoring channel conditions and comparing observed performance against multiple pre-configured FEC options. This self-service approach eliminates the need for real-time server-client feedback loops, allowing each client to independently optimize error correction without increasing server processing load, thereby improving scalability.
4Manufacturing precision
If FEC data is generated on a per-segment basis, then error correction precision is improved, but processing power requirements increase
Solution Approach 1:
The patent pre-generates FEC datasets for multiple segments with different configuration settings before transmission. The client selects from these pre-computed options based on observed channel conditions, achieving precise error correction for each segment without requiring real-time FEC generation and processing, thus reducing computational power requirements.
Data Source
AI summary
A client-side FEC selection system involves pre-generating FEC data for a plurality of media segment streams based on a number of FEC configuration settings at a server. Metadata relative to the FEC data may be provided to a client device via appropriate manifest files or other mechanisms, whereupon the client device is operative to select and request a suitable FEC data fragment responsive to monitoring various network characteristics.