Low Latency DOCSIS Agent Traffic Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current DOCSIS systems lack effective methods to identify and prioritize low-latency traffic, leading to increased latency and jitter, especially in applications like video teleconferencing and online gaming, due to the indiscriminate treatment of different traffic types within a single service flow.
Innovation Solution
The implementation of a cloud-based control plane Low Latency DOCSIS (LLD) agent that identifies characteristics or 'fingerprints' of low-latency traffic and communicates these to a network-hosted LLD agent, which processes packets to direct them to appropriate service flows, using techniques such as Type of Service (ToS) field marking and Access Control Lists, enabling separate queuing for latency-sensitive and non-latency-sensitive traffic.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If traffic is carried over a single service flow without differentiation, then device complexity is reduced, but latency and jitter increase for latency-sensitive applications
Solution Approach 1:
The patent segments traffic into different service flows based on latency requirements. It creates a first service flow for latency-sensitive traffic and a second service flow for non-latency-sensitive traffic, allowing differentiated handling without requiring complex per-packet classification at each network node.
Solution Approach 2:
The patent introduces an intermediary classification mechanism at the network edge (CMTS or RPD) that marks packets with latency sensitivity indicators before they enter the core network. This intermediary step enables downstream nodes to handle traffic appropriately without requiring full classification capabilities at each node.
2Loss of time
If client devices mark latency-sensitive traffic, then low-latency performance is improved, but hardware compatibility and ease of operation deteriorate due to gateway-specific implementations
Solution Approach 1:
The patent moves the classification function from client devices to an intermediary network element (CMTS or RPD). The client device simply sends traffic to the network, and the intermediary classifies and marks packets based on destination IP addresses or other network-layer information, making the solution independent of specific client hardware or software.
Solution Approach 2:
The network infrastructure performs self-service classification by automatically identifying latency-sensitive traffic based on destination addresses and applying appropriate markings. This eliminates the need for client devices to implement proprietary classification software or hardware modifications.
3Loss of time
If traffic classification is performed at the CPE gateway, then latency-sensitive traffic identification is improved, but adaptability to different access technologies deteriorates
Solution Approach 1:
The patent implements a universal classification mechanism at the CMTS or RPD that works across different access technologies (DOCSIS, PON, 5G). The classification based on destination IP addresses and service flow assignment is technology-agnostic, allowing the same approach to be applied regardless of the underlying access network type.
4Loss of time
If separate service flows are created for latency-sensitive traffic, then latency and jitter are reduced, but device complexity and configuration difficulty increase
Solution Approach 1:
The patent segments service flows into distinct categories (latency-sensitive and non-latency-sensitive) with predetermined characteristics. This segmentation allows network nodes to automatically route traffic to the appropriate service flow based on simple packet markings, reducing the complexity of real-time service flow management while maintaining low latency performance.
Data Source
AI summary
Devices, systems, and methods that identify and mark traffic intended to pass through a network with low latency relative to other traffic through the network. The devices, systems, and methods may include a first low latency DOCSIS (LLD) agent that identifies characteristics of low latency packets and/or a second LLD agent that uses the characteristics to identify low latency traffic and mark it for low latency treatment in the network.