Finished Theses

61 Entries found


TCP performance in wireless multi-hop networks (WMNs) is hard to achieve due to losses on the wireless channel, interferences and limited resources at individual nodes. Recent research has proposed a simple neural network (NN) structure with one input layer, two hidden layers, and one output layer that efficiently applies congestion control and that results in significant performance improvements compared to conventional TCP variants [1].

Further, NeuroEvolution of Augmenting Topologies (NEAT) is a method based on evolutionary algorithms that can outperform fixed-topology NNs in reinforcement learning tasks. We expect that NEAT may improve the performance of manually crafted NNs like iTCP even further.

This project addresses the problem of energy-efficient data dissemination from a source node to all other nodes in a wireless multi-hop network. Mahdi Mousavi et al. from the Communications Engineering Lab at TU Darmstadt have devised a decentralized algorithm towards this goal that is based on game theory [1]. While simulation results have shown that this mechanism significantly outperforms other conventional flooding mechanisms, its practical applicability still remains unexplored.

TETRA wird im kommerziellen Umfeld zur Kommunikation auf Werksgeländen eingesetzt. Im Behörden-Umfeld wird TETRA von den Organisationen und Behörden mit Sicherheitsausaufgaben (Polizei, Feuerwehr, Rettungsdienst) als primäres Kommunikationsmittel eingesetzt und hat dort den Analogfunk abgelöst. Die Sicherheit wie auch die Ausfallsicherheit dieser Kommunikation ist von größter Wichtigkeit, da darüber Notfälle bearbeitet werden. Das BOS-TETRA System ist vom Entwicklungsstand ca. 2 Jahre hinter den kommerziellen Anwendungen. Jedoch wurden weder kommerzielle noch die behördlichen TETRA-Installationen ausgiebigen öffentlichen Sicherheitsüberprüfungen unterzogen. In dieser Arbeit wurde die interne Kommunikation einer Motorola-Basisstation untersucht und im Anschluss nach Fehlern in der Implementierung gesucht. Die Fehlersuche wurde mittels Fuzzing durchgeführt. Dabei wurde ein Fehler gefunden, der das Base Radio (BR) zum Neustart bringt.

Network Coding has many positives properties that make it especially suitable for Wireless Multihop Networks [1]. Network Coding can be used to increase the effective capacity of the network, by coding (simplest form: bit-wise XOR) together packets of different flows and forwarding them in a single broadcast transmission to their intended receivers, e.g., [2]. It can also be used within a single flow to improve forward error correction (FEC) and, thus, increase transmission reliability, e.g., [3]. Unfortunately, systems based on Network Coding are easy targets for a number of attacks, and even easier to disrupt than protocols based on traditional forwarding [4].

Knowledge of global network state is crucial for several innovative network optimization techniques. Essentially, incorporating knowledge about the overall network state into locally made decisions at decentralized nodes might improve the overall network performance. A node might for instance perform transitions between network mechanisms that are optimized for certain network conditions. However, 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. Therefore, we seek to extend a node's view into the network by means of machine learning techniques.

Visible Light Communication (VLC) is a technology that uses light as the communication medium. Light offers a much higher bandwidth than the currently used traditional wireless communication technologies. This combined with the fact that it does not cause electromagnetic interference, makes VLC an ideal candidate to augment Radio Frequency (RF) technology by creating small VLC Private Area Networks (VPAN). Open source Visible Light Communication (OpenVLC) is an open-source project for conducting research in VLC technology. It is based on BeagleBone Black that has a hardware extension attached to it. It offers a Linux driver that can be used to control the hardware that communicates using an LED and a PD. However, the driver has some serious stability issues and design problems. In this thesis, we re-write the device driver for the OpenVLC platform that would implement parts of the IEEE 802.15.7 standard. The primary goal is to make the driver stable, flexible to upgrades, well documented and boosting a proper Medium Access Control (MAC) layer implementation.

Abstract of final thesis:

The Apple Watch provides the ability to automatically unlock a device running macOS when in proximity. The underlying proprietary protocol is called Auto Unlock (AU) and differs from other smart locking techniques. It uses a combination of two wireless technologies: Bluetooth Low Energy (BLE) and IEEE 802.11, to facilitate secure proximity detection. In this work we analyze the protocol by using reverse engineering and dynamic debugging. We show that AU uses both standardized protocols as well as proprietary techniques to implement a secure distance bounding protocol. With this knowledge, we discuss attack vectors and conduct a successful Man-in-the-Middle (MitM) attack on the protocol. Furthermore, we provide a starting point to allow implementations on other platforms by specifying the protocol and establish the foundation for further attacks.

Todays wireless networks are imagined to rely on radio communication. With the emergence of light-emitting diodes (LEDs) for illumination, the possibility arose to modulate light at high frequencies to transmit data, termed as visible light communication (VLC). However, VLC still needs to mature, faces problems as such as portability and thus, is proposed to coexist with ubiquitous Wi-Fi. There is little practice-oriented research on this topic, as VLC in itself is not established and hardware capable of both technologies does not exist. Before hybrid networks can be adopted in commercial products, testbeds are required for development of standards.

In this thesis, VLC hardware from prior works was modified, simulated, and built. With the constructed VLC  channel, measurements of link blockage were carried out. Handover from VLC to Wi-Fi on layer 2 was designed and simulated, based on the previously made measurements. Furthermore, a combined VLC+Wi-Fi testbed was designed based on two software-defined radio (SDR) platforms. The design shows that a testbed can be realized at reasonable effort. In addition, the proposed testbed underlines the wide design space of heterogeneous networks.

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

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