Source-Aware Network Traffic Shaping for Congestion Mitigation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Communications networks face challenges in managing congestion and satisfying subscriber demands for media within limited bandwidth resources, especially as congestion levels change, with existing methods often impacting subscriber experiences undesirably.
Innovation Solution
Implementing a system with a gateway that detects congestion and identifies adaptive traffic streams, allowing for selective reduction in capacity allocation to these streams, which automatically adapt to lower quality, thereby mitigating congestion while maintaining network performance for non-adaptive traffic.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If network bandwidth resources are increased to satisfy subscriber demands, then subscriber satisfaction improves, but network cost and infrastructure complexity increase
Solution Approach 1:
The patent changes the parameter of traffic prioritization by dynamically identifying and prioritizing adaptive traffic streams over non-adaptive streams during congestion. This allows the network to maintain service quality for time-sensitive traffic while allowing adaptive traffic to degrade gracefully, effectively managing bandwidth without infrastructure expansion
Solution Approach 2:
The patent enables adaptive traffic streams to self-regulate their own quality of service. When congestion is detected, adaptive streams automatically adjust their transmission rates or quality parameters without requiring network intervention, allowing the network to serve non-adaptive traffic at full capacity while adaptive traffic self-manages its own degradation
2Loss of energy
If congestion management impacts all traffic types equally, then network load is reduced, but subscriber experience deteriorates due to quality degradation across all services
Solution Approach 1:
The patent applies different quality management strategies to different traffic types locally. Adaptive traffic streams are allowed to degrade in quality during congestion, while non-adaptive traffic streams maintain their quality standards. This local differentiation preserves subscriber experience for critical services while allowing non-critical adaptive services to absorb the congestion impact
Solution Approach 2:
The patent segments traffic into adaptive and non-adaptive categories, applying separate congestion management policies to each segment. This segmentation allows the network to reduce overall load through adaptive traffic while protecting the quality of non-adaptive traffic, thereby reducing network load without uniformly degrading subscriber experience
3Reliability
If bandwidth allocation is reduced during congestion, then network performance for critical traffic is maintained, but adaptive traffic quality decreases
Solution Approach 1:
The patent converts the potential harm of congestion into a benefit by using adaptive traffic's ability to degrade gracefully as a mechanism for congestion management. The adaptive nature of these streams, which could be seen as a limitation, becomes advantageous because they can automatically reduce their bandwidth consumption during congestion, thereby protecting network performance for critical non-adaptive traffic
Data Source
AI summary
Systems and methods are described for source-aware shaping of network traffic in a communications system. Embodiments can manage resource congestion within a communications network by preferentially adjusting certain types of traffic on a per-subscriber and/or per-application basis. For example, a subscriber can receive various types of traffic in respective streams, and the streams can be identified according to whether they carry adaptive traffic. When a congestion condition is detected, one or more of the identified streams can be squeezed or expanded, which can cause an associated content source to automatically adapt communication of the adaptive traffic (e.g., its quality) to the reduced or expanded capacity. Selectively squeezing those streams identified as adaptive can appreciably mitigate congestion, while maintaining a desired level of network performance for the subscriber.