Description
Holdover is a term defined in 3GPP specifications (22.261, 22.878) to describe the state of a clock that was previously synchronized or syntonized to another clock (typically a primary reference or master clock) but has lost that reference. In holdover mode, the clock's frequency is determined in part using data acquired while it was synchronized to the reference, and it continues to operate within its accuracy requirements. This concept is essential for network synchronization, ensuring that network elements can maintain timing accuracy for a limited period after a reference failure, preventing service disruption.
In a telecommunications network, synchronization is vital for many functions, including radio frame timing, handover, and quality of service. Clocks are organized in a hierarchical manner, with a primary reference clock (PRC) at the top, distributing timing via synchronization chains (e.g., SyncE, IEEE 1588 PTP). When a clock loses its reference due to a link failure or network disturbance, it enters holdover. The clock uses its internal oscillator, which may be disciplined by data collected during normal operation, such as the history of phase and frequency adjustments. The quality of holdover depends on the oscillator's stability (e.g., OCXO, rubidium) and the algorithms used to predict frequency drift.
3GPP specifications define holdover in the context of 5G network synchronization requirements. For example, 22.261 (Service requirements for the 5G system) and 22.878 (Study on synchronization requirements) specify that network elements must support holdover to maintain timing accuracy for a certain period, often up to 24 hours or more, depending on the application. The holdover performance is characterized by metrics such as time error (TE), maximum time interval error (MTIE), and time deviation (TDEV). The clock must remain within specified limits (e.g., ±1.5 µs for certain 5G services) during holdover.
Holdover is particularly important for time-sensitive applications like industrial automation, smart grid, and financial trading, where precise timing is critical. In 5G, holdover ensures that base stations can continue to operate during temporary synchronization outages, avoiding service degradation. The 3GPP has studied synchronization requirements for various verticals, leading to the definition of holdover as a key capability. The evolution of holdover in 3GPP reflects the growing dependence on precise timing and the need for resilient synchronization architectures.
From an implementation perspective, holdover involves a combination of hardware (stable oscillator) and software (holdover algorithm). The algorithm may use temperature compensation, aging compensation, and learning from past behavior to minimize frequency drift. The clock may also use alternative references, such as GNSS, to re-enter synchronization. The 3GPP specifications provide a framework for defining holdover requirements and test methods, ensuring interoperability and performance. Overall, holdover is a fundamental concept for network synchronization resilience, enabling networks to ride through transient synchronization failures.
Purpose & Motivation
Holdover exists to provide resilience in network synchronization. In modern telecommunications networks, many elements rely on a continuous and accurate timing reference to function correctly. If a reference fails, even briefly, the network could experience service outages, dropped calls, or data corruption. Holdover allows a clock to maintain its accuracy for a period after losing its reference, using stored data and a stable local oscillator. This bridges the gap until the reference is restored or an alternative is found.
Historically, synchronization was often provided by dedicated T1/E1 lines or GPS. As networks evolved to packet-based timing (e.g., IEEE 1588 PTP), the risk of reference loss increased due to network congestion and failures. Holdover became a critical requirement to ensure that timing accuracy is maintained during transient outages. The 3GPP recognized this need in its study on synchronization requirements for 5G, leading to the definition of holdover in Release 18. The definition emphasizes that the clock must remain within its accuracy requirements, which are derived from the service requirements of the applications it supports.
The motivation for standardizing holdover in 3GPP is to ensure that all network elements can interoperate and provide consistent performance. By defining holdover in specifications like 22.261 and 22.878, 3GPP provides a common understanding and testable requirements. This is particularly important for 5G networks, which support diverse services with stringent timing needs. Holdover thus solves the problem of synchronization resilience, enabling networks to maintain service continuity and meet the demanding requirements of vertical industries.
What changed in its specifications
The latest release-to-release comparison of specifications that define Holdover. Each summary describes the whole specification, not only this term.
Rel-19 adds to TS 22.261 Indirect Network Sharing, 5G satellite access Phase 3 requirements, a 5G wireless sensing service clause, signalling traffic monitoring, SNPN interconnection clauses, and Localized Mobile Metaverse Services. It also extends disaster roaming to 5G-only UEs and adds Ambient IoT device clauses.
Full comparison in TS 22.261 →Release Timeline
Detected Changes Across Releases
from 3GPP Change RequestsSpecific changes extracted from the „Change history“ tables of 3GPP specifications (1 CRs across 1 releases). Complements the general historical overview above with the evidence-based evolution of this function.
Defining Specifications
3GPP specifications that define or reference Holdover, with the latest known release. Sourced from the 3GPP document catalog — see methodology.
| Specification | Title | Release |
|---|---|---|
| TS 22.261 vk81 | 5G Service and Operational Requirements | Rel-20 read |
| TR 22.878 vi20 | Technical Report on 5G Timing Resiliency | Rel-18 read |