Scaling Resilient Timing Through Standards Based Terrestrial Networks
Evan Alexander, George Zampetti, Eric Colard, Jim Olsen, Lee Cosart, Greg Wolff, Microchip Technology Inc.
Location: Ballroom C
Date/Time: Monday, Jun. 1, 4:45 p.m.
Precise and continuous time synchronization is a foundational dependency for critical infrastructure sectors including telecommunications, energy, transportation and defense. Today, Global Navigation Satellite Systems (GNSS) serve as the primary source of absolute time for these systems. However, GNSS signals are inherently vulnerable to jamming, spoofing, adjacent-band emissions and solar weather events. These vulnerabilities increasingly represent a single point of failure for infrastructure that depends on nanosecond-level timing accuracy. Simply adding more GNSS receivers scales cost and attack surface rather than resilience. Whereas, timing architectures that rely on custom hardware, specialized optics, or non-standard network elements slow adoption and limit deployment at scale.
This presentation argues that resilient timing does not require the reinvention of network architectures, but rather leverages proven commercial technologies implemented using open, international standards. Doing so represents the lowest risk and fastest path to deploying timing resilience across operational critical infrastructure networks.
Recent advances in High Accuracy Time Transfer (HA-TT), standardized in ITU-T G.8271.1/Y.1366.1, enable sub-nanosecond terrestrial time distribution over optical fiber using Precision Time Protocol (PTP) combined with Synchronous Ethernet (SyncE). The HA-TT implementations examined in this work utilize standard Ethernet interfaces, commercial off-the-shelf (COTS) SFPs , and existing fiber infrastructure. This approach enables rapid deployment and scaling using proven commercial equipment.
The presentation will detail recent experiments measuring the performance of HA-TT, demonstrating a maximum time error of approximately 5 ns over distances approaching 800 km across ten network hops, corresponding to roughly 500 picoseconds of time error per hop. These results enable absolute time to be distributed terrestrially with accuracy comparable to local GNSS reception, while eliminating the requirement for GNSS receivers at every timing endpoint. This capability supports centralized or regional timing architectures and enables precise timing delivery into environments where GNSS is unreliable or unavailable, such as underground facilities, dense urban centers, or jammed regions.
When combined with Enhanced Primary Reference Time Clock (ePRTC) systems as defined in ITU-T G.8272.1, HA-TT, enables resilient UTC-traceable timing networks. ePRTC systems integrate cesium atomic clocks with advanced holdover algorithms to maintain traceability to authoritative time sources such as UTC(USNO) or UTC(NIST) . Experimental results demonstrate holdover performance better than 100 ns over 40 days, enabling infrastructure networks to continue operating with predictable and bounded timing error during extended GNSS outages.
The implementation of ePRTC and HA-TT exists in the virtual Primary Reference Time Clock (vPRTC) architecture, which is a highly secure and resilient network-based timing architecture that has been developed to meet the expanding needs of modern critical infrastructures. A key resilience advantage of vPRTC is geographic diversity and redundancy. GNSS receivers, atomic clocks, and timing distribution nodes do not need to be co-located and can instead be separated by tens or hundreds of kilometers in a bidirectional or dual-ring network configuration. This spatial separation significantly reduces the likelihood that localized jamming, spoofing, or physical attacks affect both the authoritative time source and downstream infrastructure simultaneously, while the bidirectional HA-TT mitigates single points of failure such as fiber cuts.
The presentation concludes by describing a proposed terrestrial-based national timing architecture that consists of a geographically distributed grid of protected autonomous time scales across the United States. Interconnected via fiber-based HA-TT in a multi-node vPRTC configuration, such a network could technically enable precise, UTC-traceable time independent of GNSS at the point of delivery, with traceability provided via terrestrial connection to authoritative UTC time scales.