Fine-grained Starlink Throughput Variation Examined With State-Transition Modeling

Sep 15, 2026

Uploaded by Aditya

Leveraging a dataset of nearly half a billion packets with high-precision timestamps and sizes, the study extracts burst characteristics from Starlink’s Ethernet interface. Burst structure reflects physical-layer reception of OFDMA frames on the satellite link, enabling analysis of burst rates and inferred transitions between physical-layer rates. Results reveal structured rate-transition behavior, mainly between neighboring rates, and joint transitions of burst rate and duration that mostly occur within the same rate, often using an intermediate short burst. The paper ties burst rate, burst duration, and inter-burst silence to effective throughput.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages8

Reading Time20 min

AccessFree to download

Introduction

Data Collection

Burst Characteristics and Rate Transitions

Throughput Determinants

Fine-grained Starlink Throughput Variation Examined With State-Transition Modeling Johan Garcia Karlstad University / Icomera AB Email: [johan.garcia@kau.se](johan.garcia@kau.se) Simon Sundberg Karlstad University Email: [simon.sundberg@kau.se](simon.sundberg@kau.se) Anna Brunstrom Karlstad University Email: [anna.brunstrom@kau.se](anna.brunstrom@kau.se) Abstract—Leveraging a data set of almost half a billion packets with high-precision packet times and sizes, we process it to extract characteristics of the bursts emitted over Starlink’s Ethernet interface. The structure of these bursts directly reflect the physical layer receipt of OFDMA frames on the satellite link. We study these bursts by analyzing their rates, and by proxy the transition between different physical layer rates. The results highlight that there is definitive structure in the transition behavior, and we note specific behaviors such as particular transition steps associated with rate switching, and that rate switching occurs mainly to neighboring rates. We also study the joint burst rate and burst duration transitions, noting that transitions occur mainly within the same rate, and that changes in burst duration are often performed with an intermediate short burst in-between. Finally, we examine the configurations of the three factors burst rate, burst duration, and inter-burst silent time, which together determine the effective throughput of a Starlink connection. I. INTRODUCTION Low-Earth orbit (LEO) systems have made it possible to provide Internet connectivity across wide areas that were previously under-served in terms of wireless communication access. Such access has generally been provided by cellular networks with limit to their coverage in sparsely populated areas. Providing passengers onboard vehicles such as trains and buses with Internet connectivity when traversing areas of no, or poor, cellular connectivity is also a use case, as is the use of Starlink for 5G backbone connectivity [1], [2] . In contrast to cellular networks, the details of LEO systems, and SpaceX’s Starlink [15] in particular, is opaque and many system design decisions can only be inferred through measurements as the amount of publicly provided information on system details is limited. Several measurement studies have been performed to evaluate different aspects of the Starlink network [10], [11], [18],[20] . In this work, we in particular focus on the satellite downlink mechanisms that yield a particular observed throughput at the network layer. By utilizing a large data set of highprecision per-packet measurements, we study the behavior of the transmission bursts which are associated with how Starlink allocates system resources between concurrent users. Previous studies [5], [11] have noted the existence of a 15 second reconfiguration interval, as well as an OFDMA frame time of 1.33 ms. Our recent study [4] on physical layer details highlight that the data in these frames is transmitted in bursts observable at the network layer, and by analyzing the timings of these bursts insights on physical layer characteristics can be gained. The number of frames per burst in our measurements mostly vary between 1 and 7 . Between these bursts, there are inter-burst pauses, where no data is received by the user. On longer time scales (i.e. multi-burst time scales) we note that the throughput obtained by a user is a function of a) the physical transmission rate of the frames, b) the number of frames in a burst, and c) the inter-burst silent time. Throughput is thus a combination of the prevailing radio channel conditions as managed by (a) and the scheduling across competing users as managed by (b,c) . We here focus on an initial investigation of these factors with the aim to obtain additional knowledge regarding Starlink rate transition behavior. Building on our previous studies of Starlink burst rates we here study how the transition between rates are performed with

FAQs

Which factors determine the effective throughput in a Starlink connection?

Effective throughput is determined by the burst rate, burst duration, and inter-burst silent time. These three factors jointly capture how radio conditions and scheduling interact across concurrent users.

Sep 15, 2026

Uploaded by Aditya