Network Time Synchronization One-Way Transfer Time Accuracy

Solution Overview

Problem

Current network time synchronization methods, such as NTP, IRIG, and PTP, face challenges in accurately calculating one-way transfer time due to variability caused by cache loading, data structure setup time, and scheduling variability, leading to errors in clock synchronization across networked devices.

Innovation Solution

The method involves initiating a first sequence to load data structures and prime scheduling, followed by a second sequence to calibrate timing, ensuring that the completion of the first sequence occurs just in time for the reception of a synchronization message, thereby reducing uncertainty in transfer time calculations and improving accuracy across various network time synchronization protocols like NTP and PTP.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional network time synchronization methods (NTP, IRIG, PTP) are used to estimate one-way transfer time, then clock synchronization can be achieved, but accuracy deteriorates due to variability caused by cache loading, data structure setup time, and scheduling variability

Engineering Contradiction:
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The patent applies preliminary action by executing a warm-up sequence before the actual time synchronization measurement. This warm-up sequence includes loading data structures into cache memory and priming scheduling mechanisms to eliminate variability caused by cold starts. By performing these actions beforehand, the system ensures that when the actual measurement occurs, the cache and scheduling are already in an optimized state, thereby reducing timing variability and improving measurement precision.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent implements continuity of useful action by maintaining active data structures and scheduling mechanisms throughout the warm-up period. Instead of allowing the system to idle or reset between measurements, the warm-up sequence keeps the relevant components actively engaged and ready. This continuous activity ensures that the timing characteristics measured during synchronization are representative of normal operating conditions rather than transient cold-start states.

Inventive Principle:
Principle #20Continuity of useful action

2Measurement precision

If data structures are loaded and scheduling is primed to reduce variability, then transfer time calculation accuracy improves, but system complexity increases due to the additional warm-up sequence

Engineering Contradiction:
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The warm-up sequence is implemented as a preliminary action that prepares the system state before actual synchronization measurements. By consolidating the complexity into this preliminary phase, the patent reduces the complexity of the main synchronization measurement process itself. The warm-up sequence handles the complex tasks of cache loading and scheduling priming separately, allowing the actual timing measurement to be simpler and more straightforward.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent segments the synchronization process into distinct phases: a warm-up sequence for loading data structures and priming scheduling, followed by the actual time synchronization measurement phase. This segmentation allows each phase to be optimized independently - the warm-up phase handles complexity-related tasks while the measurement phase focuses on precise timing. By separating these functions, the overall system manages complexity more effectively.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentUS9756153B2Method for improving accuracy in computation of one-way transfer time for network time synchronization
Publication Date: 2017.09.05
  • US9756153B2 patent drawing
  • US9756153B2 patent drawing
  • US9756153B2 patent drawing

AI summary

A method for improving accuracy in the computation of a one-way transfer time between two networked devices. In one aspect, variability in time transfer latency that is caused by cache loading, data structure setup time, and scheduling variability in software is reduced by initiating a first sequence of loading data structures into cache and priming scheduling, and then initiating a second sequence of calibrating the timing of a subsequent synchronization message so that the completion of the first sequence occurs just in time for the reception of the synchronization message. The method is applicable for any network time synchronization protocol, including Network Time Protocol (NTP) and Precision Time Protocol (PTP).