Università degli Studi di Napoli Federico II Dipartimento di Ingegneria Elettrica e delle Tecnologie dell'Informazione (DIETI)

MIRAGE Project

MIRAGE is a reproducible architecture for mobile-app traffic capture and ground-truth creation. The human-generated datasets below have been collected by means of MIRAGE and are released openly to the research community.

Why mobile traffic

Mobile apps generate a large share of today's network traffic, and analysing it, to classify apps and activities, predict traffic or manage the network, requires data with a reliable ground truth. Traffic produced by scripted interactions does not reflect how people really use apps, while traffic captured in operational networks lacks reliable labels.

MIRAGE addresses both issues: it is a reproducible architecture to capture the traffic generated by human experimenters using real apps on real devices, and to label every flow with the app, and in the most recent datasets the activity, that generated it.

How MIRAGE works

Every MIRAGE dataset is collected with the same architecture, so that campaigns run in different years, with different apps and experimenters, produce traffic that can be compared and reused.

Best Paper Award — 4th IEEE International Conference on Computing, Communications and Security (ICCCS 2019), October 2019, Rome (Italy).

From capture to release

  1. 01Human experimentersVolunteers use real mobile apps on real devices, following the activities defined for each campaign.
  2. 02Traffic captureThe traffic generated by the devices is captured during each session of use.
  3. 03Ground truthInformation collected on the devices is used to label each flow with the app that generated it, and the activity performed.
  4. 04Open releaseThe labeled traffic is released openly to the research community, with the papers that describe it.

MIRAGE datasets

Human-generated mobile-app traffic with ground truth, captured with the MIRAGE architecture.

NewCC BY-NC-ND 4.0

MIRAGE-AppAct-2024

Traffic of 20 popular mobile apps, labeled with both the app and the specific activity performed (chat, audio and video calls, gaming, streaming).

  • Experimenters 240+
  • Apps 20
  • Devices 3
  • Format JSON
Details
CC BY-NC-ND 4.0

MIRAGE-COVID-CCMA-2022

Traffic of Communication and Collaboration Mobile Apps (CCMA) that bloomed with the COVID-19 pandemic, labeled with both the app and the activity performed.

  • Experimenters 150+
  • Apps 9
  • Devices 3
  • Format JSON, pickle
Details
CC BY-NC-ND 4.0

MIRAGE-Video

Traffic of 14 video mobile apps with associated ground truth, collected between June 2019 and March 2020.

  • Experimenters 280+
  • Apps 14
  • Devices 3
  • Period Jun 2019 – Mar 2020
Details
CC BY-NC-ND 4.0

MIRAGE-2019

The original MIRAGE dataset: human-generated mobile app traffic with associated ground truth, for advancing the state of the art in mobile traffic analysis.

  • Experimenters 280+
  • Apps 40
  • Devices 3
  • Award Best Paper, ICCCS 2019
Details

Papers

The MIRAGE architecture is described in the paper below; each dataset page lists the paper to cite for that dataset.

Giuseppe Aceto, Domenico Ciuonzo, Antonio Montieri, Valerio Persico, Antonio Pescapè, “MIRAGE: Mobile-app Traffic Capture and Ground-truth Creation”, 4th IEEE International Conference on Computing, Communications and Security (ICCCS 2019), October 2019, Rome (Italy).

Download

Before downloading

Please fill in the form below to download the dataset.