Collide, Collate, Collect: Recognizing Senders in Wireless Collisions

Bachelor Thesis by Johannes Lauinger

Submitted: August 10th, 2017

Advisor: Prof. Dr.-Ing. Matthias Hollick
Supervisor: Robin Klose, M.Sc.

Secure Mobile Networking Lab
Department of Computer Science
Technische Universität Darmstadt

Citation

Cite this work as follows:

• Lauinger, Johannes Tobias. "Collide, Collate, Collect: Recognizing Senders in Wireless Collisions." B.Sc. Thesis. Technische Universität Darmstadt, 2017.

BibTex:

@mastersthesis{lauinger2017,
    type    = {B.Sc. Thesis},
    author  = {Lauinger, Johannes Tobias},
    title   = {Collide, Collate, Collect: Recognizing Senders in Wireless Collisions},
    school  = {Technische Universität Darmstadt},
    year    = {2017},
    pdf     = {https://tubama.ulb.tu-darmstadt.de/253/1/lauinger2017-sender-recognition.pdf}
}

Note: the full-text PDF version is only accessible from within the TU Darmstadt network.

Repository Structure

The implementation directory contains the Matlab code that I wrote to conduct experiments for this thesis. Some files, starting with seemoo_, can however not be used directly. This is due to a dependency to an internal library from the SEEMOO Lab at TU Darmstadt that I couldn't include in this repository. Furthermore, there is a couple of calls to functions starting with ieee_80211_. The implementation of those functions is missing.

The LaTeX files and assets for the thesis document, as well as my presentations, are located in the thesis directory. The final thesis document is included in PDF format in the top-level directory.

Note: During the work on this thesis, there were two different repositories for implementation and thesis. I attempted to merge those, so there may be some strange things and/or consistency errors in the Git history.

Some data has been redacted, for example the MAC addresses that I captured with airodump-ng in data/. This is due to obvious privacy concerns.

Abstract

With wireless mobile IEEE 802.11a/g networks, collisions are currently inevitable despite effective counter measures. This work proposes an approach to detect the MAC addresses of transmitting stations in case of a collision, and measures its practical feasibility. Recognizing senders using cross-correlation in the time domain worked surprisingly well in simulations using Additive White Gaussian Noise (AWGN) and standard Matlab channel models.

Real-world experiments using software-defined radios also showed promising results in spite of decreased accuracy due to channel effects. During the experiments, various Modulation and Coding Schemes (MCSs) and scrambler initialization values were compared. Knowledge about which senders were transmitting leading up to a collision could help develop new improvements to the 802.11 MAC coordination function, or serve as a feature for learning-based algorithms.

Zusammenfassung

In drahtlosen mobilen Netzwerken nach den IEEE 802.11a/g Standards sind Kollisionen trotz wirkungsvoller Gegenmaßnahmen nicht vollständig zu vermeiden. Diese Arbeit stellt einen Ansatz zur Erkennung der MAC-Adressen der beteiligten Sender bei einer Kollision vor und untersucht, inwiefern das Verfahren in der Praxis funktioniert. Über Kreuzkorrelation im Zeitbereich funktionierte die Erkennung in Simulationen unter Additivem Weißen Gaußschen Rauschen (AWGN) und verschiedenen Standard-Kanalmodellen von Matlab erstaunlich gut.

Praktische Experimente mit Software-Defined Radios zeigten ebenfalls vielversprechende Ergebnisse, wenn auch die Genauigkeit der Erkennung durch Kanaleffekte beeinträchtigt wurde. Bei den Experimenten wurden verschiedene Modulation and Coding Schemes (MCSs) und Scrambler-Initialisierungen verglichen. Die Kenntnis über die beteiligten Sender bei einer Kollision könnte zur Verbesserung der Koordinierungsfunktion oder als Feature für lernbasierte Verfahren verwendet werden.

License

Copyright (c) 2017 Johannes Lauinger

Thesis Document and Source Code

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License.

Matlab Implementation

Licensed under the terms of the GNU GENERAL PUBLIC LICENSE, Version 3.