This article was originally published on July 13, 2020.
Editor’s Note 1: This content is contributed by Paolo Novellini, Principal Engineer at Xilinx and member of the ITSF Steering Group (International Time and Synchronization Forum)
Editor’s Note 2: ITSF (International Time and Synchronization Forum) is one of the two most important world-wide conferences about synchronization. The author of this white paper, Paolo Novellini, is a member of the steering committee and represents Xilinx at the ITSF. You can find his bio here.
Synchronization Technology is Pervasive
Synchronization technology is pervasive in many industry sectors: finance, telecom, industrial, automotive, and aerospace & defense. These markets have several applications that heavily rely on synchronization.
While geo-localization is probably one of the most well-known applications relying on this technology, it is not the only one. Many synchronization techniques are available; ultimately, they can be classified into two main categories:
- Based on packets
- Based on GNSS (Global Navigation Satellite System)
This white paper describes different synchronization techniques, highlighting the advantages and the disadvantages of each.
Implementing Custom PTP Schemes and Latency Measurement on Xilinx Platforms
Let’s look at one scenario. Precision Time Protocol (PTP) works at layer 2 in the Open System Interconnection (OSI) model, to be as close as possible to the physical medium, and hence, gain accuracy. The goal of PTP is to transfer time with minimal error from a network node to another node.
When the link latency is symmetrical, the system assumes that timescales in the master and slave are aligned until a next correction is measured. The slave node can correct its time scale by the correction mechanism. If the link latency is asymmetrical, it translates into an error in the slave synchronization. Beyond asymmetry, the precision at which timestamps are generated dominates the overall synchronization precision.
System designers can use a standard platform and, at the same time, achieve their desired precision, using their unique ideas and creativity. The Xilinx platforms, such as FPGAs, SoC, and ACAPs, are fully programmable and allow system architects and designers to implement their PTP schemes and latency measurement architectures. Ultimately, the PTP accuracy is not limited by the hardware platform itself, but by the designer's architecture. While achieving desired PTP accuracy, the system also hosts custom applications in a single device. When using the Xilinx platforms, it is a true paradigm shift in PTP applications.
How can I unlock the power of all the programmability and flexibility I get from Xilinx platforms?
Download this White Paper to get plenty of examples and ideas on how to achieve different levels of PTP accuracy. These will help you as a base to trigger your creativity and ideas.