Receiver-Side Congestion Window Control for Multi-Sender Networks

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing network communication protocols like TCP rely on sender-side congestion control mechanisms that are limited by round-trip-time measurements, leading to inefficiencies and potential conflicts between congestion control and flow control, and do not account for complete network conditions.

Innovation Solution

Implementing receiver-side congestion and flow control using receiver nodes to determine congestion windows based on packet loss, explicit congestion notifications, and one-way latency, allowing for more accurate and coordinated adjustments across multiple sender nodes.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If sender-side congestion control using round-trip-time measurements is implemented, then congestion control can be performed at the sender, but measurement precision and accuracy of network conditions are limited

Engineering Contradiction:

Improvecongestion control accuracyVSAvoidnetwork condition information

Core Design Contradiction:

Measurement precisionVSLoss of information

Solution Approach 1:

The patent inverts the traditional congestion control approach by moving the congestion control logic from the sender to the receiver. The receiver now determines the congestion window size based on complete network condition information (packet loss, ECN, one-way latency) and communicates this to the sender, thereby achieving more accurate congestion control while eliminating the information limitations of sender-side RTT measurements

Inventive Principle:

Principle #13The other way round (Inversion)

Solution Approach 2:

The patent implements enhanced feedback mechanisms where the receiver provides detailed network condition feedback to the sender through multiple indicators (packet loss detection, ECN marks, one-way latency measurements). This feedback loop enables the sender to adjust its congestion window based on comprehensive network state information, significantly improving measurement precision compared to traditional RTT-based feedback

Inventive Principle:

Principle #23Feedback

2Reliability

If sender-side congestion control is implemented, then congestion control can be performed, but conflicts with flow control occur

Engineering Contradiction:

Improvecongestion control effectivenessVSAvoidcontrol mechanism complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent merges congestion control and flow control into a unified receiver-side mechanism. The receiver now performs both functions simultaneously by determining the congestion window based on network conditions (congestion control) and receiver buffer status (flow control), eliminating conflicts between the two mechanisms and simplifying the overall control architecture

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

The receiver acts as an intermediary that coordinates both congestion control and flow control decisions. By centralizing these control functions at the receiver, the patent eliminates the conflicts that arise when separate sender-side mechanisms operate independently, while the receiver uses its complete view of network and receiver state to make coordinated decisions

Inventive Principle:

Principle #24Intermediary (Mediator)

3Measurement precision

If receiver-side congestion control is implemented, then more complete network condition information is used, but device complexity increases

Engineering Contradiction:

Improvenetwork condition detection accuracyVSAvoidreceiver node complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent enhances the universality of the receiver node by enabling it to perform multiple functions: traditional flow control, congestion window determination, packet loss detection, ECN processing, and one-way latency measurement. This multi-functional approach consolidates control logic at the receiver, improving measurement precision while managing complexity through functional integration rather than multiplication

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS12537774B2Systems and methods for congestion and flow control in networks

Publication Date: 2026.01.27 LEMON INC(GB)

AI summary

Congestion windows are controlled from a receiver in a communications network, and then provided to senders to control flow of packets within the network. The congestion window can be dynamically adjusted based on detection of one of packet loss, a congestion signal and a packet latency signal. The congestion window can be further adjusted for fairness or flow control based on bandwidth use. The sender node receives the congestion window and controls sending rates according to the congestion window. The sender node can determine packet loss and further adjust the congestion window.