Poster Sessions

April 3 - 6, 2017
San Jose, CA
DoubleTree San Jose

Poster Sessions
Tuesday, April 4: 10:05 - 11:05 AM
Wednesday, April 5: 9:40 - 11:10 AM

A Small Time Budget Footprint Solution Enables Cost-Effective Next Generation
of Telecom Network

Alejandro Gonzalez, Telecommunication Engineer, Seven Solutions
The next generation of Telecom network imposes highly demanding requirements for time transfer. Some technologies claim to require a target of a hundred nanoseconds. This time budget needs to be smartly split between the different network elements as primary reference clocks, core, metro, backhaul and nodes. By using ultra-accurate time transfer solutions the upgrade of existing networks and deployment of new ones can be addressed in a cost-effective way. For instance, deploying enhanced PTP with high accuracy at the core network can increase the time budget available for remaining devices. This allows keeping older quipment in non-core network layers while achieving a global time budget at an affordable cost. In this contribution we discuss how the emergent solution of White Rabbit can address this target. View poster here.

How to Have the Best Time
Harlan Stenn, President, Network Time Foundation
If you can't trust your time you can't trust your data Network Time Foundation is home of the Internet's best time software NTF's open source solutions for accurate network time include:

  • The NTP Project
  • Network Time Security (NTS)
  • Ntimed
  • Linux PTP
  • Certification and Compliance
View poster here.

Preparing to Extend Commercial Telecom PTP Across the United States
Lee Cosart, Senior Technologist, Microsemi
Marc Weiss, Consultant, NIST

This poster discusses our preparations for a second phase of a joint project employing commercial equipment to send national timing signals through a telecommunication network. As presented at WSTS 2016, this experiment was previously performed to connect the UTC (NIST) time scale located in Boulder, Colorado with the UTC (USNO) Alternate Master Clock time scale located at Schriever Air Force Base in Colorado via a telecommunication provider's optical network. Timing signals using the Precision Time Protocol (PTP) were sent in the usual two-way fashion, but each one-way delay was measured, because we had UTC time scales at both ends of the network that were within 10 ns of each other. For a second Phase of this project, a circuit is now being developed to link the UTC (NIST) time scale in Boulder, Colorado with a telecom office in Chicago. While UTC (NIST) will time the end of the circuit in Boulder, UTC from GPS will be used to time the circuit at the Chicago endpoint. Ultra-long haul equipment will be used to link the circuit, and the one-way delays of PTP signals will be measured in each direction. We will be using the same protocol, transporting a Gb Ethernet over the OTN to transfer the PTP time packets. This commercial transport mechanism was shown capable of backing up GPS for time transfer at the 100 ns level on the circuit across Colorado. We will subsequently determine if this technique can be used as a service across the country. We summarize the history of the project, present updated data from our ongoing Phase I work, and show measurements from the CenturyLink lab that model the Boulder-Chicago circuit. View poster here.

Rethink Oscillator Classification for Time Holdover and Mobile Synchronization
Jeff Gao, Director, Product Marketing, SiTime
Clock systems are historically categorized into stratum levels of accuracy according to GR-1244 standards. This classification scheme has worked well for systems where frequency synchronization and holdover are critical. However, this scheme no longer provides sufficient and relevant performance metrics for designers of phase and time synchronization applications such as cloud RAN and mobile backhaul that rely on IEEE 1588 and other packet based synchronization protocols. This presentation proposes a new classification scheme that defines clock performance in terms of time and phase synchronization and holdover capabilities under real life operating profiles. Major contributing factors such as ?F/?T, critical to a given time and phase synchronization capability, are also discussed. The overall objective of this proposal is to make the clock classification more meaningful by enabling engineers to efficiently specify the most optimal clock performance without having to use stratum level as a proxy to time and phase holdover performance.

Satellite Time and Location (STL) Service as an Augmentation or Alternative to GNSS
for Critical Timing Applications

Lisa Perdue, Applications Engineer, Spectracom
The Satellite Time and Location Service provided by Satelles on the Iridium Satellite System offers a new reference signal to be used in critical timing applications. The service provides a 500ns to UTC accuracy, allowing it to be used as a reference in critical timing applications. Compared to the GNSS signals, which are weak and susceptible to interference, the STL service provides a higher power and authenticable service to add robustness to systems historically using only GNSS as a primary reference. In this paper we explore using the STL service as a standalone timing reference and as an augmentation to GNSS as a reference. The STL receiver is used with a time and frequency server to provide timing signals such as 1PPS, NTP, and PTP. Reference monitoring allows real time analysis of GNSS, STL, and the holdover oscillator references so there is no interruption in service should the primary reference be lost. Two configurations are tested; GNSS used as a primary reference and STL is used as a backup, and STL is used as the primary reference. Results on the timing signal accuracies generated by the time and frequency server are presented. View the poster here.

The Robustness Upgrading of GNSS DO
Kunihiko Hashimoto Design Engineer, Furuno Electric Electric
Using GNSS references for timing synchronization provides high precision and is economical. However, GNNS reliability is vulnerable to various disturbances. Furuno Electric proposes several technical counter measures to GNSS vulnerabilities. Especially, the ability to maintain 1.5µsec holdover timing accuracy is important. We achieved 1.5µsec / 24hours holdover under ◿T=20℃ environment temperature condition change by using OCXOs. View the poster here.

Time and Frequency Metrology of Commercial White Rabbit Hardware
Jeff Sherman, Physicist, NIST
White Rabbit is an open source project started at CERN which extends the Ethernet physical layer protocol for the purpose of high-performance time and frequency transfer over standard telecom fiber optics. We evaluate the suitability of a commercial White Rabbit implementation for transfer of primary UTC(NIST) reference signals (10 MHz and pulse-per-second) over distances of several hundred meters. With tight environmental control, we observed 100 fs-level time transfer stability over 3 hours of averaging using the 10 MHz signal. In a deployment between laboratory buildings, the pulse-per-second signal stability in loopback was 1 ps after one hour of averaging; residual temperature coefficients of order 1 ps/K were present in either the White Rabbit equipment or measurement apparatus. View the poster here.

Timing Challenges in the Smart Grid
Cuong Nguyen, Lead, Smart Grid Testing and Certification, NIST
This poster presentation provides a summary of the challenges of wide area clock synchronization and potential solutions based on the outcomes of the NIST/IEEE workshop on Timing Challenges for Smart Grid. NIST and IEEE-SA conducted the workshop to gather inputs from stakeholders to identify and analyze the practical challenges that are currently being experienced in wide area time synchronization in current measurement and control deployments. Timing-related barriers can also prevent the power industry from realizing increased automation in measurement and control strategies. In the U.S. power grid, where each interconnection stretches over a large spatial expanse, achieving correct timing can be challenging. Timing needs include one microsecond synchronization to a traceable time and frequency reference for data fusion. Issues include, but are not limited to, GPS and communication infrastructure as well as concerns for reliability and resilience when a reference source becomes unavailable. In addition, dynamic distributed measurement and control systems are time sensitive. In distributed control where scheduling of resources, seamlessly orchestrated coordination, and hard deadlines must be met, measurement data, state estimators and control commands are temporally valid for a specific, sometimes brief, duration. In order to provide the common time reference for correct temporal behavior, all systems must be synchronized to a traceable time and frequency source. Participants at the workshop delved further into wide area precision clock synchronization issues in power systems based on their implementation of advanced sensor and control systems. They also discussed potential timing needs for future power systems. Another key topic was how to collaborate and prioritize key research and standards activities. View the poster here.

Ultra Stable OCXOs with Enhanced Capabilities
Cyril Datin, R&D Manage, Rakon
This poster presentation provides a summary of the challenges of wide area clock synchronization and potential solutions based on the outcomes of the NIST/IEEE workshop on Timing Challenges for Smart Grid. NIST and IEEE-SA conducted the workshop to gather inputs from stakeholders to identify and analyze the practical challenges that are currently being experienced in wide area time synchronization in current measurement and control deployments. Timing-related barriers can also prevent the power industry from realizing increased automation in measurement and control strategies. In the U.S. power grid, where each interconnection stretches over a large spatial expanse, achieving correct timing can be challenging. Timing needs include one microsecond synchronization to a traceable time and frequency reference for data fusion. Issues include, but are not limited to, GPS and communication infrastructure as well as concerns for reliability and resilience when a reference source becomes unavailable. In addition, dynamic distributed measurement and control systems are time sensitive. In distributed control where scheduling of resources, seamlessly orchestrated coordination, and hard deadlines must be met, measurement data, state estimators and control commands are temporally valid for a specific, sometimes brief, duration. In order to provide the common time reference for correct temporal behavior, all systems must be synchronized to a traceable time and frequency source. Participants at the workshop delved further into wide area precision clock synchronization issues in power systems based on their implementation of advanced sensor and control systems. They also discussed potential timing needs for future power systems. Another key topic was how to collaborate and prioritize key research and standards activities.

HOST SPONSORS

ATIS
NIST