Dual-Queue DOCSIS Gateway for Latency Reduction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current communication networks, particularly DOCSIS networks, face significant latency issues due to queuing delays and media access delays, which are exacerbated by traditional TCP protocols and shared-medium scheduling, leading to suboptimal performance in managing upstream traffic.

Innovation Solution

The implementation of a low latency DOCSIS AQM scheme that utilizes a dual-channel queue structure with strict priority queuing, where latency-sensitive traffic is separated from latency-insensitive traffic, and an upstream scheduler that proactively manages bandwidth based on estimated demand, enabling more frequent grants and predictive scheduling.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of time

If traditional TCP protocols and shared-medium scheduling are used, then network coverage and compatibility are maintained, but latency increases due to queuing delays and media access delays

Engineering Contradiction:
ImprovelatencyVSAvoidqueue management complexity
Core Design Contradiction:
Loss of timeVSDevice complexity

Solution Approach 1:

The patent segments the single traffic queue into multiple queues (e.g., high-priority queue and low-priority queue) to differentiate latency-sensitive traffic from latency-insensitive traffic. This segmentation allows selective scheduling where critical traffic receives preferential treatment, reducing overall latency while maintaining manageable complexity through structured queue organization.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies local quality by assigning different service qualities to different queues. High-priority queues receive guaranteed service levels and preferential scheduling, while low-priority queues use best-effort service. This localized quality differentiation reduces latency for critical applications without requiring complete system redesign, balancing performance improvement with operational complexity.

Inventive Principle:
Principle #3Local quality

2Loss of time

If separate networks (mobile networks, DOCSIS networks, Wi-Fi networks) operate independently, then network autonomy and control are maintained, but internetwork visibility and communication cause higher latency

Engineering Contradiction:
Improveinternetwork latencyVSAvoidnetwork integration capability
Core Design Contradiction:
Loss of timeVSAdaptability or versatility

Solution Approach 1:

The patent implements a universal queue management system that can handle multiple network types (DOCSIS, mobile, Wi-Fi) within a single gateway device. The scheduling mechanism is designed to work across different network protocols and traffic types, providing unified latency management while maintaining the autonomy of individual network segments. This multi-functional approach reduces internetwork latency without sacrificing network-specific adaptability.

Inventive Principle:
Principle #6Universality (Multi-functionality)

3Productivity

If TCP flows send traffic faster than the link rate, then bandwidth utilization is maximized, but queue overflows occur causing packet drops and increased latency

Engineering Contradiction:
Improvebandwidth utilizationVSAvoidqueuing delay
Core Design Contradiction:
ProductivityVSLoss of time

Solution Approach 1:

The patent applies preliminary action through proactive traffic shaping and rate limiting mechanisms that regulate traffic entering the queues before congestion occurs. By pre-controlling the rate at which traffic is admitted to high-priority and low-priority queues, the system prevents queue overflow and packet drops while maintaining high bandwidth utilization. This anticipatory approach ensures smooth traffic flow and reduces latency without sacrificing productivity.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS11356380B1Systems and methods for queue protection
Publication Date: 2022.06.07 CABLE TELEVISION LAB INC
  • US11356380B1 patent drawing
  • US11356380B1 patent drawing
  • US11356380B1 patent drawing

AI summary

A scheduling device for managing a packet queue of a communication gateway includes a receiving portion configured to receive data packets according to at least one communication protocol, a processor, and classification module configured to separate the received data packets into a first traffic queue and a second traffic queue separate from the first traffic queue. The first traffic queue includes a low latency service flow classified to have strict priority. The second traffic queue includes a primary service flow classified as having a classic priority. The classification module separates the received data packets so that those with a first indicator are sent to the first traffic queue and those without the first indicator are sent to the second traffic queue.