Adaptive ECN Thresholds for Egress Queue Congestion Control

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current network congestion control methods using the Explicit Congestion Notification (ECN) mechanism lack flexibility, leading to either excessive queue depth and delayed data packet transmission or suboptimal resource utilization due to low transmission rates.

Innovation Solution

A network device dynamically adjusts the ECN threshold for egress queues based on statistical information from past time periods, considering factors like packet transmission rate, queue depth change, and priority flow control back pressure to optimize congestion control.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If an excessively high ECN threshold is set, then the queue depth of the egress queue in the network device is comparatively large, but the data packet transmission delay becomes longer

Engineering Contradiction:
Improvequeue depthVSAvoidtransmission delay
Core Design Contradiction:
Quantity of substanceVSLoss of time

Solution Approach 1:

The patent applies dynamics by making the ECN threshold adaptive rather than static. The network device dynamically adjusts the ECN threshold based on real-time queue depth monitoring and historical statistical information. When queue depth exceeds the threshold, the device learns from the congestion event and adjusts the threshold for future periods, enabling the system to adapt to changing traffic patterns and minimize transmission delay while maintaining adequate queue depth.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent implements feedback mechanisms where the network device continuously monitors queue depth and uses this information to adjust the ECN threshold. The device collects statistical information about queue depth over time periods, processes this feedback, and modifies the threshold accordingly. This closed-loop feedback system enables the device to respond to congestion conditions and optimize the balance between queue depth and transmission delay.

Inventive Principle:
Principle #23Feedback

2Loss of time

If an excessively low ECN threshold is set, then the transmit end sends a data packet at a lower rate, but network resource utilization becomes lower

Engineering Contradiction:
Improvetransmission delayVSAvoidnetwork resource utilization
Core Design Contradiction:
Loss of timeVSProductivity

Solution Approach 1:

The patent makes the ECN threshold dynamic by using historical statistical information to adapt the threshold value. Instead of using a fixed low threshold that limits transmission rate, the device learns from past congestion patterns and adjusts the threshold to allow higher transmission rates when conditions permit, thereby improving network resource utilization while still preventing congestion.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The network device uses feedback from monitoring queue depth and transmission patterns to optimize the ECN threshold. By analyzing statistical information from previous time periods, the device determines appropriate threshold values that balance transmission rate and resource utilization. This feedback-driven approach prevents unnecessarily low thresholds from limiting network productivity.

Inventive Principle:
Principle #23Feedback

3Device complexity

If a fixed ECN threshold is used, then the congestion control is simple, but flexibility in network congestion control becomes comparatively low

Engineering Contradiction:
Improvecongestion control mechanismVSAvoidflexibility in congestion control
Core Design Contradiction:
Device complexityVSAdaptability or versatility

Solution Approach 1:

The patent transforms the static ECN threshold into a dynamic parameter that adapts to different network conditions. The network device implements a learning mechanism that processes historical statistical information about queue depth and congestion events, then adjusts the ECN threshold accordingly. This dynamic approach maintains reasonable complexity while dramatically improving flexibility and adaptability to varying traffic patterns and network states.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The network device performs self-adjustment of the ECN threshold by automatically collecting statistical information, analyzing congestion patterns, and modifying the threshold without external intervention. This self-service capability enables the system to adapt to changing conditions autonomously, providing flexibility while keeping the control mechanism relatively simple through automated decision-making.

Inventive Principle:
Principle #25Self-service

Data Source

PatentEP3961981B1Method and device for congestion control, communication network, and computer storage medium
Publication Date: 2025.09.10 HUAWEI TECH CO LTD
  • EP3961981B1 patent drawingFigure 1~2
  • EP3961981B1 patent drawingFigure 3~4
  • EP3961981B1 patent drawingFigure 5~6

AI summary

This application discloses a congestion control method and apparatus, a communications network, and a computer storage medium, and relates to the field of network technologies. The method includes: A network device first obtains statistical information of a target egress queue within a first time period, where the target egress queue is any target egress queue in the network device; then determines an explicit congestion notification ECN threshold for the target egress queue within a second time period based on the statistical information of the target egress queue within the first time period, where the second time period is chronologically subsequent to the first time period. When a queue depth of the target egress queue exceeds the ECN threshold within the second time period, the network device sets an ECN mark for a data packet in the target egress queue. In this application, the network device dynamically adjusts the ECN threshold for the egress queue, thereby improving flexibility in network congestion control.