Understanding Propagation Delay
Propagation delay is the interval required for an electrical or optical signal to travel from one point in a network to another. Unlike processing delays that occur within equipment, propagation delay is purely a function of distance and medium characteristics.
The medium's properties significantly influence signal speed. In free space or vacuum, electromagnetic waves travel at approximately 3 × 10⁸ m/s. However, within cables and fibres, signals move slower due to the dielectric material's refractive index. Copper twisted-pair cables typically support velocities of 60–77% of light speed, whilst fibre-optic cables often achieve 65–99% depending on the specific type.
Distance compounds this effect. A transatlantic submarine cable spanning 6,000 km introduces delays of roughly 30–50 milliseconds. By contrast, signals within a data centre traverse metres and incur microsecond-scale delays. For most casual internet users, this delay is imperceptible; for high-frequency traders executing orders, a single millisecond determines profit or loss.
Propagation Delay Formula
Propagation delay is calculated as the ratio of distance to propagation speed in the medium.
Propagation Delay = Distance ÷ Propagation Speed
Velocity Factor = Propagation Speed ÷ Speed of Light
Distance— Physical separation between transmitter and receiver, measured in metres or kilometres.Propagation Speed— Speed at which the signal travels through the transmission medium, typically expressed in metres per second.Velocity Factor— Ratio of actual propagation speed to the speed of light in vacuum; typically ranges from 0.6 to 0.99 depending on cable type.
Practical Example: New York to London
Imagine transmitting a financial order from New York to a server in London, separated by approximately 5,567 km (great-circle distance). Using copper-based transatlantic cable with a velocity factor of 0.77:
- Propagation speed = 0.77 × 299,792,458 m/s ≈ 230.8 × 10⁶ m/s
- Distance = 5,567 km = 5,567,000 m
- Propagation delay = 5,567,000 m ÷ (230.8 × 10⁶ m/s) ≈ 24.1 milliseconds
This delay is unavoidable. No routing optimisation or compression can reduce it below the speed-of-light limit. In high-frequency trading, where transactions occur in microseconds, such latency is critical to risk management and execution strategy.
Common Pitfalls and Considerations
Several factors often trip up engineers and network designers when estimating propagation delay.
- Confusing propagation delay with latency — Propagation delay is only one component of total network latency. Processing delays in routers, queuing delays in buffers, and transmission delays (time to push all bits onto the wire) add substantially to end-to-end delay. A 24 ms propagation delay may become 100+ ms once queuing and processing are included.
- Ignoring actual cable routing — The great-circle distance between two cities is not the same as cable distance. Undersea cables follow seafloor topography, avoiding hazards and following economical paths. The actual distance may be 10–30% longer than straight-line calculations suggest, increasing delay accordingly.
- Assuming uniform propagation speed — Different cable segments use different materials and designs. A fibre-to-the-premises network may have 0.67 velocity factor in some sections and 0.95 in others. Using a single average value introduces compounding errors over long paths with heterogeneous infrastructure.
- Overlooking cross-talk and signal integrity — At very high frequencies or over extreme distances, signal degradation forces retransmissions and error correction, effectively increasing delay even though physical propagation speed remains constant. Budget margins matter in critical systems.
Applications Across Industries
Telecommunications: Network designers calculate propagation delays to budget total latency, synchronise clock signals, and allocate bandwidth reserves for acknowledgements and retransmissions.
Financial services: High-frequency trading firms spend millions on low-latency infrastructure. Saving even 1 ms of propagation delay by upgrading to microwave links or optimising cable routes directly affects profitability.
Online gaming: Competitive players experience perceived lag when propagation delay exceeds 100–150 ms. Server placement, content delivery networks, and peering agreements all aim to minimise this delay.
Aerospace and satellite communications: Geostationary satellites orbiting 36,000 km above Earth introduce inherent 240+ ms round-trip delays. Mission-critical systems must account for this in real-time control and feedback loops, often requiring ground stations or relay constellations.