WetLinks: a Large-Scale Longitudinal Starlink Dataset with Contiguous Weather Data

Dominic Laniewski,Eric Lanfer,Bernd Meijerink,Roland van Rijswijk-Deij,Nils Aschenbruck

2024-02-26

Abstract:Low Orbit Satellite (LEO) networks such as Starlink promise Internet access everywhere around the world. In this paper, we present WetLinks - a large and publicly available trace-based dataset of Starlink measurements. The measurements were concurrently collected from two European vantage points over a span of six months. Consisting of approximately 140,000 measurements, the dataset comprises all relevant network parameters such as the upload and download throughputs, the RTT, packet loss, and traceroutes. We further augment the dataset with concurrent data from professional weather stations placed next to both Starlink terminals. Based on our dataset, we analyse Starlink performance, including its susceptibility to weather conditions. We use this to validate our dataset by replicating the results of earlier smaller-scale studies. We release our datasets and all accompanying tooling as open data. To the best of our knowledge, ours is the largest Starlink dataset to date.

Networking and Internet Architecture

What problem does this paper attempt to address?

The problem that this paper attempts to solve is to construct a large - scale, long - term Starlink measurement data set and analyze its performance, especially the impact of weather conditions on its performance. Specifically, the objectives of the paper are as follows:

1. **Construct a large - scale data set**: Create a large - scale longitudinal Starlink data set named WetLinks. This data set contains approximately 140,000 measurement results collected within six consecutive months from two European sites (Osnabrück in Germany and Enschede in the Netherlands), covering all relevant network parameters, such as upload and download throughput, round - trip time (RTT), packet loss rate, and route tracing.

2. **Combine weather data**: Place professional weather stations beside two Starlink terminals to collect high - precision weather data in order to analyze the impact of weather conditions on Starlink performance.

3. **Verify the validity of the data set**: Verify the quality and reliability of the data set by reproducing the results of earlier, smaller - scale studies.

4. **Open data and tools**: Publicly release the data set and all supporting tools so that other researchers can use this data for further research and verification.

### Main contributions
- **Data set scale**: WetLinks is the largest Starlink data set to date, containing approximately 80,000 measurements from Osnabrück and 60,000 measurements from Enschede.
- **Weather impact analysis**: Combined with high - precision weather data, analyzed the impact of precipitation on Starlink download throughput.
- **Data set verification**: Verified the validity of the data set by reproducing the results of earlier studies.
- **Tool release**: Provided supporting tools for merging different data sources (performance measurements, weather data, etc.).

### Background and related work
- **Starlink overview**: Starlink is a low - Earth - orbit (LEO) satellite Internet service provided by SpaceX, aiming to provide Internet access worldwide.
- **Existing research**: Most previous studies have focused on the preliminary performance evaluation of Starlink, but the data sets in these studies have problems such as small scale, non - standard methods, and unclear description of measurement conditions.

### Measurement setup
- **Measurement locations**: Set up measurement points on the rooftops of university buildings in Osnabrück, Germany, and Enschede, the Netherlands.
- **Equipment configuration**: Each measurement point includes a Starlink user terminal (“dishy”), a Starlink router configured in passthrough mode, a VM running Ubuntu as a measurement client, and a Froggit DP2000 weather station placed directly beside the user terminal.
- **Data collection**: Conduct throughput and RTT measurements every 3 minutes and route tracing measurements every 6 minutes. At the same time, collect high - precision weather data.

### Analysis results
- **Throughput analysis**: The median download throughput is approximately 230 Mbit/s in Osnabrück and approximately 250 Mbit/s in Enschede. The median upload throughput is approximately 16.16 Mbit/s in Osnabrück and approximately 17.66 Mbit/s in Enschede.
- **Packet loss rate analysis**: More than 20% of the measurement results contain packet loss, with the minimum packet loss rate being 0.4% and the maximum packet loss rate being 84.4%.
- **Delay analysis**: The average RTT is 62.93 ms in Osnabrück and 64.49 ms in Enschede. Most of the measurement results have an RTT within the 1 σ range.
- **Time impact**: Different time periods during the day have a significant impact on network parameters, especially during the morning and evening peak hours.
- **Weather impact**: Rainfall has a weakly negative correlation with download throughput, with correlation coefficients of - 0.145 (Osnabrück) and - 0.201 (Enschede) respectively.

Through these analyses, the paper not only provides large - scale Starlink performance data, but also deeply explores the impact of weather conditions on satellite communications, providing valuable data resources for future research.