AI-Based Master Clock Selection for Adaptive PTP Synchronization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing IEEE 1588 Best Master Clock (BMC) selection procedure is platform-dependent, lacks sophisticated decision-making, and is cumbersome due to its reliance on conditional statements, limiting flexibility and scalability across heterogeneous hardware platforms.
Innovation Solution
Implementing an Artificial Intelligence (AI)/Machine Learning (ML) model to analyze PTP announce messages, determining the best master clock based on distance and similarity with pre-defined reference parameters, enhancing decision-making efficiency and compatibility across various systems.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the traditional BMC algorithm is used for master clock selection, then the selection process is simple and based on clear conditional rules, but the system lacks adaptability to heterogeneous hardware platforms and sophisticated decision-making capabilities
Solution Approach 1:
The patent replaces the traditional mechanical BMC algorithm (based on conditional statements and platform-specific logic) with an AI/ML-based selection mechanism. The slave clock uses machine learning models to analyze PTP announce messages and select the best master clock, enabling platform-independent operation across heterogeneous hardware while providing sophisticated adaptive decision-making capabilities.
2Ease of operation
If the BMC algorithm uses multiple conditional statements for parameter comparison, then the selection logic is explicit and deterministic, but the procedure becomes cumbersome and difficult to implement across different platforms
Solution Approach 1:
The patent substitutes the cumbersome conditional statement-based BMC algorithm with an AI/ML-based selection process. Instead of implementing complex if-else logic for comparing PTP parameters across different platforms, the slave clock employs machine learning models that automatically learn optimal selection criteria from data, simplifying implementation while maintaining sophisticated decision-making.
Solution Approach 2:
The patent transforms the selection process from comparing discrete parameter values using conditional statements to analyzing parameter patterns and relationships through AI/ML models. This allows the system to handle various PTP parameters (clock class, accuracy, offset log variance, priority values) in a unified, platform-independent manner by learning from training data rather than following rigid comparison rules.
3Reliability
If the traditional BMC algorithm is implemented, then the master clock selection is deterministic and easy to understand, but the system lacks dynamic adaptation to changing network conditions
Solution Approach 1:
The patent replaces the static, deterministic BMC algorithm with a dynamic AI/ML-based selection system. The slave clock uses machine learning models that can adapt to changing network conditions by learning from historical PTP announce messages and real-time observations, while maintaining reliable synchronization through trained prediction capabilities.
Solution Approach 2:
The patent implements feedback mechanisms where the AI/ML model continuously learns from PTP announce messages and synchronization performance. The system uses determined distances and similarities between actual PTP parameters and reference parameters to dynamically adjust master clock selections, enabling adaptation to changing network conditions while maintaining high synchronization reliability.
Data Source
AI summary
Embodiments of the disclosure describe a method 300 for synchronizing a slave clock with a predicted best master clock. The method 300 includes receiving, at the slave clock, a plurality of PTP announce messages. The method 300 includes determining at least one of a distance and a similarity between each of a plurality of parameters in the received PTP announce message from each master clock and a corresponding plurality of pre-defined reference parameters associated with a reference master clock. The method 300 includes predicting the best master clock among the plurality of master clocks for the clock synchronization. The best master clock is predicted based on the determined at least one of the distance and the similarity between each of the plurality of parameters in the received PTP announce message from each master clock and the corresponding plurality of pre-defined reference parameters associated with the reference master clock.