Network Congestion Control via Queue Buffer Synchronization

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing congestion control methods in networks lead to unequal sending rates among network nodes, resulting in unnecessary network bandwidth waste and reduced network utilization due to improper resource allocation.

Innovation Solution

A congestion control method that synchronizes queue buffer statuses among network nodes using a coupling vector, adjusting each node's queue buffer status based on the state vector and coupling coefficients to achieve equal sending rates and shared congestion risk.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If existing congestion control methods (PQM or AQM) are used to manage queue status of individual routers, then congestion can be controlled, but sending rates of routers become unequal causing network bandwidth waste and reduced network utilization

Engineering Contradiction:
Improvecongestion controlVSAvoidnetwork utilization
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent merges the congestion control decisions of multiple routers by synchronizing their queue buffer statuses. Each router combines its local queue status with the state vector of other coupled routers to achieve consistent sending rates across the network, transforming individual isolated control into coordinated collective control that eliminates bandwidth waste.

Inventive Principle:
Principle #5Merging (Combining)

2Reliability

If a router decreases sending rate due to detected packet loss, then congestion is avoided, but network resources become unavailable and bandwidth is wasted

Engineering Contradiction:
Improvecongestion avoidanceVSAvoidbandwidth waste
Core Design Contradiction:
ReliabilityVSLoss of energy

Solution Approach 1:

The patent implements a feedback mechanism where routers continuously exchange queue buffer status information with coupled routers. Each router adjusts its sending rate based on the synchronized state vector that reflects the combined status of all coupled routers, enabling precise congestion response that avoids both over-reaction (bandwidth waste) and under-reaction (congestion).

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP3306879B1Congestion control method and network node
Publication Date: 2020.03.11 HUAWEI TECH CO LTD
  • EP3306879B1 patent drawingFigure 1~2
  • EP3306879B1 patent drawingFigure 3
  • EP3306879B1 patent drawingFigure 4~5

AI summary

The present invention discloses a congestion control method and a network node, and relates to the field of communications technologies, so as to resolve existing problems of network resource waste and network utilization decrease. The congestion control method provided in embodiments includes: performing step a of obtaining, by a first network node, a current queue buffer status of each of N second network nodes; performing step b of combining, by the first network node, current queue buffer statuses of the N second network nodes and a current queue buffer status of the first network node into a state vector; performing step c of updating, by the first network node, the current queue buffer status of the first network node according to the state vector and a coupling vector; and repeating steps a to c till an absolute value of a difference between each of the queue buffer statuses of the N second network nodes and the queue buffer status of the first network node is less than or equal to a preset threshold.