Overview

The CESNET-MINER22-TS dataset [1] contains Flow Time Series derived from IP flows in the CESNET-MINER22 dataset [2]. The data was collected by monitoring the national research and educational network CESNET2 and transformed into time-series representations suitable for temporal pattern analysis.

Dataset Metadata

PropertyValue
TypeRecreated dataset
CategoryTime Series
Primary TaskCryptomining Detection
Source DatasetCESNET-MINER22
Source NetworkCESNET2 (national research and educational network)

Flow Time Series

Definition

Flow Time Series are sequences of observations based on Network Dependencies. A Network Dependency is defined as:

Long-term communication between device pairs in which a service is provided by one node to another.

Construction Process

  1. IP Flow Collection: Raw flows captured from CESNET2 network
  2. Dependency Identification: Group flows by communicating device pairs
  3. Time-Series Creation: Generate sequential observations from flow patterns
  4. Feature Extraction: Compute temporal and statistical features

Advantages for Cryptomining Detection

Flow Time Series enable detection of:

  • Periodic Behaviors: Regular communication patterns characteristic of mining
  • Long-term Patterns: Sustained connections typical of mining pools
  • Service Relationships: Client-server mining pool interactions
  • Temporal Anomalies: Deviations from normal communication patterns

Relationship to CESNET-MINER22

AspectCESNET-MINER22CESNET-MINER22-TS
FormatIP FlowsFlow Time Series
GranularityIndividual flowsAggregated flow sequences
Analysis TypeFlow-level classificationTemporal pattern detection
FeaturesPer-flow attributesPeriodic behavior features

Research Applications

The CESNET-MINER22-TS dataset supports research in:

  • Periodic Pattern Detection: Identifying rhythmic communication patterns
  • Temporal Analysis: Understanding time-dependent cryptomining behaviors
  • Behavioral Classification: Distinguishing mining from legitimate services
  • Feature Engineering: Developing temporal features for improved detection

How to Cite

@inproceedings{plny2022decrypto,
  title={DeCrypto: Finding Cryptocurrency Miners on ISP Networks},
  author={Pln{\`y}, Richard and Hynek, Karel and {\v{C}}ejka, Tom{\'a}{\v{s}}},
  booktitle={Nordic Conference on Secure IT Systems},
  pages={139--158},
  year={2022},
  organization={Springer}
}

Download

[1] Josef Koumar, Richard Plný, & Tomáš Čejka. (2023). CESNET-MINER22-TS: Periodic Behavior Features of Cryptomining Communication [Data set]. Zenodo.
DOI: 10.5281/zenodo.8033351

References

[2] Richard Plný, Karel Hynek, & Tomáš Čejka. (2022). CESNET-MINER22 (1.0) [Data set]. Zenodo.
DOI: 10.5281/zenodo.7189293