Rate Adaptation Negotiation for Congestion Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network congestion control methods, such as TMMBR, often result in oscillating bitrate adjustments due to unsynchronized decision frequencies and intervals between receiver-based and sender-based rate adaptation techniques, leading to unpredictable quality of service, buffering, packet delays, and losses.
Innovation Solution
Configuring clients to use different rate adaptation methods based on session setup negotiations, such as using TMMBR for semi-static adaptation and SCReAM, GCC, or NADA for dynamic adaptation, to synchronize bitrate adjustments and improve network performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple rate adaptation methods (TMMBR, SCReAM, GCC, NADA) are used in parallel, then dynamic rate adaptation capability is improved, but bitrate stability deteriorates due to oscillating behavior
Solution Approach 1:
The patent segments the rate adaptation functionality into distinct modules (TMMBR, SCReAM, GCC, NADA) that can be independently selected and configured. Each method operates as a separate entity that can be enabled or disabled based on network conditions and requirements, allowing the system to use only one method at a time rather than running multiple methods simultaneously, thus avoiding oscillating behavior while maintaining adaptability.
Solution Approach 2:
The patent implements dynamic selection of rate adaptation methods based on current network conditions. The system can switch between different adaptation methods (TMMBR for receiver-based, SCReAM/GCC/NADA for sender-based) depending on whether the network is congested, the type of traffic, and other real-time parameters. This dynamic approach allows the system to adapt to changing conditions without the instability caused by running multiple methods in parallel.
2Device complexity
If receiver-based rate adaptation (TMMBR) is used, then implementation simplicity is improved, but adaptability to network conditions deteriorates
Solution Approach 1:
The patent creates a universal rate adaptation framework that can accommodate both receiver-based (TMMBR) and sender-based (SCReAM, GCC, NADA) methods within the same system. The framework provides a common interface and configuration mechanism that supports multiple adaptation approaches, allowing the system to select the most appropriate method based on network conditions while maintaining a unified implementation structure that does not significantly increase complexity.
3Productivity
If sender-based rate adaptation (SCReAM, GCC, NADA) is used, then network performance is improved, but synchronization with receiver-based methods deteriorates
Solution Approach 1:
The patent segments the rate adaptation decision-making process by assigning it exclusively to either the sender or the receiver based on the selected method. When sender-based methods (SCReAM, GCC, NADA) are chosen, the sender independently makes rate decisions without attempting to synchronize with receiver-based TMMBR messages. This segmentation eliminates the synchronization conflicts that arise when multiple methods with different decision frequencies operate simultaneously, while still achieving high network performance through the chosen sender-based approach.
Data Source
AI summary
A method of configuring a first client device to utilize at least one type of rate adaptation method in a negotiated session with a second client device. The first client determines whether a first type of rate adaptation method and a generic feedback format are to be used in the negotiated session. Upon determining that only the first type of rate adaptation method is to be used, the first client device is configured to utilize the first type of rate adaptation method for both semi-static and dynamic rate adaptation. Upon determining that both are to be used, the first client device is configured to utilize the first type for semi-static rate adaptation, and to utilize for dynamic rate adaptation a preferred rate adaptation method selected from a plurality of rate adaptation method types that provide improved performance for dynamic rate adaptation compared to the first type of rate adaptation method.