RTT Evaluation Using One-Way Delay Under Delayed ACKs

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing methods for evaluating round-trip time (RTT) in high-throughput data transmission scenarios, particularly in wireless networks, suffer from inaccuracies due to reduced acknowledgement (ACK) frequency in delayed acknowledgement protocols, leading to skewed RTT statistical values and impaired data transmission performance.

Innovation Solution

The method infers RTT changes based on one-way delay (OWD) measurements at the receiving node, reducing dependence on per-packet acknowledgements and enhancing RTT estimation accuracy by using OWD to calculate RTT statistical values.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If delayed acknowledgement is used to reduce ACK transmission frequency, then bandwidth utilization improves, but RTT measurement accuracy deteriorates

Engineering Contradiction:

Improvebandwidth utilizationVSAvoidRTT measurement accuracy

Core Design Contradiction:

ProductivityVSMeasurement precision

Solution Approach 1:

The receiving node pre-calculates OWD statistical values (minOWD, maxOWD, avgOWD) for multiple packets before sending the ACK. This preliminary calculation ensures that even though ACK frequency is reduced, the RTT measurement accuracy is maintained by using pre-computed one-way delay statistics from multiple packets

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent introduces OWD (one-way delay) as an intermediary measurement to bridge the gap between reduced ACK frequency and RTT measurement accuracy. By calculating OWD at the receiving node and using it to derive RTT statistical values, the system maintains measurement precision without requiring frequent ACKs

Inventive Principle:

Principle #24Intermediary (Mediator)

2Measurement precision

If per-packet acknowledgement is used, then RTT measurement accuracy improves, but bandwidth utilization deteriorates

Engineering Contradiction:

ImproveRTT measurement accuracyVSAvoidbandwidth utilization

Core Design Contradiction:

Measurement precisionVSProductivity

Solution Approach 1:

The patent merges multiple OWD measurements into single statistical values (minOWD, maxOWD, avgOWD) that are then used for RTT calculation. By combining information from multiple packets into aggregated statistics, the system achieves accurate RTT measurement without requiring separate ACKs for each packet

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

Instead of requiring complete per-packet acknowledgements, the patent uses partial information (OWD statistical values from a subset of packets) to derive RTT measurements. The receiving node calculates OWD statistics from multiple received packets and includes these in periodic ACKs, providing sufficient measurement data without full per-packet acknowledgment overhead

Inventive Principle:

Principle #16Partial or excessive action

Data Source

PatentEP3832953B1System, method, and apparatus for evaluating round-trip time

Publication Date: 2025.12.24 HUAWEI TECH CO LTD

Figure 1(a)~1(c)
Figure 2~3(a)
Figure 3(b)~5

AI summary

This application discloses a round-trip time evaluation system, method, and apparatus, to improve accuracy of round-trip time evaluation. In the system, a first device sends a plurality of packets to a second device on a same path, where each of the plurality of packets carries a timestamp indicating a sending time of the packet. The second device determines, by calculating a statistical value of one-way delays of the plurality of packets, one timestamp in timestamps carried in the plurality of packets. The second device sends an ACK to the first device, where the ACK includes the determined timestamp and ACK time information. The first device determines one measurement value of an RTT of the path based on the timestamp in the ACK, the ACK time information, and a time at which the ACK is received.