Original title: From Stochastic Foundations to Emerging Frontiers: A Comprehensive Survey on Packet Jitter Modeling, Measurement, and Metrics in IP Networks
Juan Eduardo Macias-Rivera, José Ricardo Gómez Rodríguez, Ana Gabriela Castañeda-Miranda, Remberto Sandoval-Aréchiga +2 · September 2026
When you stream video or play games online, packets of data travel across the network. If the time between these arriving packets keeps changing, you get jitter. Jitter hurts the quality of real-time services, yet engineers measure and model it in fragmented ways. Researchers surveyed studies from 1990 through August 2026 to see how jitter research evolved. They found models shifted from simple single-queue equations to complex, data-driven approaches across whole networks. The way you define jitter changes your answers. Testing 1440 network windows, one common formula reported 8.17 ms, another hit 52.36 ms, and a third peaked at 6.14 ms. Open challenges remain in unified models, high-resolution measurements, and security.
Imagine timing a bus ride to decide when to leave for school. You could measure from your front door to the stop, from when the bus pulls away, or take the average of your last few rides — and each method gives a different answer for whether the bus is late. One method says the bus is early, another says it is very late, and you cannot tell which is true. In this paper, the same internet traffic is measured with different ways of measuring delay, and the choice of method changes the reported number just as the choice of timing method changes the bus result.
Anyone relying on real-time internet services — video calls, cloud gaming, remote surgery, industrial control — depends on jitter staying low. This survey shows the same traffic can look nearly problem-free or badly degraded depending on which jitter metric an operator uses, so comparing reports across tools and providers is currently unreliable. That is why the field is pushing for standard, distribution-aware reporting and open benchmarks.
This is a survey, not a new measurement campaign; the RIPE Atlas illustration is a single reproducible example, not a general characterization of internet jitter. The abstract does not claim to solve the open problems it lists, and it does not propose a new metric or model. It also does not cover every possible jitter definition; it synthesizes existing work and identifies gaps.
Ask anything about this paper. Answers come from its abstract, and say so when the abstract doesn't cover your question.
From Stochastic Foundations to Emerging Frontiers: A Comprehensive Survey on Packet Jitter Modeling, Measurement, and Metrics in IP Networks — Juan Eduardo Macias-Rivera, José Ricardo Gómez Rodríguez, Ana Gabriela Castañeda-Miranda et al. — Telecom — doi:10.3390/telecom7050124
Abstract licence: CC BY
Explanation generated by Synthesis from the abstract — it is not the original paper.