Packet-Level Traffic Modeling with Heavy-Tailed Payload and Inter-Arrival Distributions for Digital Twins

TL;DR

This paper introduces a compact, hybrid packet-level traffic generator for digital twins that accurately models heavy-tailed packet sizes and inter-arrival times using a combination of hidden Markov models and mixture density networks, outperforming recent neural approaches.

Contribution

The paper presents a novel hybrid traffic generator combining HMMs and Student-t mixture models to efficiently and accurately reproduce complex traffic patterns with heavy tails, suitable for digital twin deployment.

Findings

Outperforms recent neural network and transformer-based generators in most metrics.

Uses significantly fewer parameters, around 0.2 MB, enabling deployment in resource-constrained environments.

Accurately models heavy-tailed inter-arrival times and payload distributions across diverse traffic types.

Abstract

Digital twins of radio access networks require packet-level traffic generators that reproduce the size and timing of packets while remaining compact and easy to recalibrate as traffic changes. We address this need with a hybrid generator that combines a small hidden Markov model, which captures buffering, streaming, and idle states, with a mixture density network that models the joint distribution of payload length and inter-arrival time (IAT) in each state using Student-t mixtures. The state space and emission family are designed to handle heavy-tailed IAT by anchoring an explicit idle state in the tail and allowing each component to adapt its tail thickness. We evaluate the model on public traces of web, smart home, and encrypted media traffic and compare it with recent neural network and transformer based generators as well as hidden Markov baselines. Across most datasets and metrics, including average per-flow cumulative distribution functions, autocorrelation based measures of temporal structure, and Wasserstein distances between flow descriptors, the proposed generator matches the real traffic most closely in the majority of cases while using orders of magnitude fewer parameters. The full model occupies around 0.2 MB in our experiments, which makes it suitable for deployment inside digital twins where memory footprint and low-overhead adaptation are critical.