Classification and Estimation of Global Network Properties in Wireless Multi-Hop Networks

Bachelor Thesis, Master Thesis

open


Type
Analysis: 2
Empiricism: 8
Implementation: 8
Literature Research: 3

Motivation

Knowledge of global network state is crucial for several innovative network optimization techniques. However, these techniques are often evaluated in simulation environments with omniscient knowledge about the network at individual nodes, which is not realistic in practical scenarios. In fact, an individual node's scope of the network is limited in practice since it is able to overhear the wireless channel only locally, and explicit notification about global network state would result in large overhead.

In this thesis project, you are going to engineer features and learning algorithms that allow nodes to gain knowledge about distant parts of a network just by overhearing the wireless channel. The difficulty is to identify features that comprise valuable information from distant nodes, which we believe might be feasible since multi-hop packet transfers may implicitly allow to monitor how distant nodes interact with the network.

 This topic is for you if you are interested in machine learning, wireless networks, and practical experimenting. The project might be co-supervised by another researcher from the collaborative research center MAKI, who is specialized either in the field of topology control, autonomous agents or machine learning techniques.


Goal

  • Literature review: Identify network optimization techniques that rely on global network knowledge and extract their requirements.
  • Define goals: Make a list of network properties that should be classified or estimated. This might comprise larger network topologies, membership of topologies, node count, flow and traffic characteristics, or network scenarios consisting of combinations thereof.
  • Identification of features: Identify potential features that can be obtained by traffic monitoring. Features that comprise relevant information about distant nodes might for instance be obtained by inspecting packet headers of the higher layers (e.g., network layer and transport layer).
  • Feature engineering and machine learning: Select and engineer features that can be obtained by overhearing the wireless channel.
  • Implementation: Run experiments with a network simulator (e.g., ns-3) and tune your algorithms. Finally, you may evaluate your approach also in a practical wireless testbed.


Supervisor:

Research Areas: Sichere Mobile Netze



Back


Contact

Prof. Dr.-Ing. Matthias Hollick

Technische Universität Darmstadt
Department of Computer Science
Secure Mobile Networking Lab 

Mornewegstr. 32 (S4/14)
64293 Darmstadt, Germany

Phone: +49 6151 16-25472
Fax: +49 6151 16-25471
office@seemoo.tu-darmstadt.de

Affiliations

A A A | Drucken Drucken | Impressum Impressum | Sitemap Sitemap | Suche Suche | Kontakt Kontakt | Webseitenanalyse: Mehr Informationen
zum Seitenanfangzum Seitenanfang