Theses in Progress

Bachelor's Theses

Quantum Key Distribution

Quantum Key Distribution

Short Description:
Describe and analyze QKD.

Description

Quantum key distribution (QKD) is a secure communication method which implements a cryptographic protocol involving components of quantum mechanics. The goal of the Bachelor Thesis is to decribe and analyze the algorithm. Furthermore, a program simulating QKD should be developed.

Prerequisites

Linear Algebra

Supervisor:

Student

Marcel Tri-Tai Chu

Quantum Games

Quantum Games

Keywords:
Quantum Information, Algorithm

Short Description:
Analyze Quantum Games

Description

There are games which had a greater winning chance with entanglement. Goal of this thesis is to analyze this game and calculate examples.

Prerequisites

Linear Algebra, Information Theory

Supervisor:

Student

Reza Khasbullah

Automatization of a breakout box for the Chinese DC charging standard

Automatization of a breakout box for the Chinese DC charging standard

Description

Laut chinesischem DC-Ladestandard enthält die DC-Ladeverbindung zusätzlich zu den Hochvolt-Leitungen eine Steckererkennung (CC1 und CC2), 12V Zusatzspannung, einen Schutzleiter und CAN-Leitungen zur Kommunikation zwischen EVSE und EV. Um die Verbindung und Kommunikation zwischen EVSE und EV zu überprüfen und zu variieren wird eine Breakout-Box benötigt, welche in das Kabel zwischen EVSE und EV zwischengeschalten wird. Diese ermöglicht es die CAN-Kommunikation und die Leitungen zu manipulieren. Ziel der Bachelorarbeit ist es eine vorhandene manuelle Breakout-Box zu automatisieren. Die Automatisierung soll die Trennung der CAN-, Zusatzspannungs- und der Leitung der Steckererkennung ermöglichen sowie die automatisierte Messung der Zusatzspannung und Spannung der Steckererkennungsleitungen beinhalten. Die Steuerung soll über einen CAN-Bus erfolgen. Dazu muss ein Multi-Purpose-Modul (MPM) implementiert, die Steuerung über den CAN-Bus definiert und die vorhandene Breakout-Box umgebaut werden. Zusätzlich soll eine vorhandene CANoe-Konfiguration der Breakout-Box um die Automatisierung erweitert werden um anschließend bereits vorhandene manuelle Testfälle zu automatisieren.

Contact

P3 systems GmbH

Heilbronner Straße 86

70191 Stuttgart

Germany

konstantin.weber@p3-group.com

+49 151 724 222 46

Supervisor:

Norbert Hanik - Dr. Konstantin Weber (P3 systems GmbH)

Student

Iliana Paspaleva

Implementation of an Acoustic Data Transmission System

Implementation of an Acoustic Data Transmission System

Description

Matlab implementation of an acoustic data transmission system.

Supervisor:

Julian Renner

Student

Ludwig Lohmer

The Weight Enumerators of Product Codes

The Weight Enumerators of Product Codes

Keywords:
weight enumerator, input-output weight enumerator, input-redundancy weight enumerator, distance spectrum

Description

For product codes (PCs), it is very easy to find the minimum distance and its multiplicity once those of the component codes are known. However, the characterization of its complete distance spectrum is a largely open problem even if those of the component codes are given. In [1], they provide a feasible method to compute the weight enumerators (WEs) of PCs where at least one component code is a single parity check (SPC) code. In addition, an extension to compute the exact input-output weight enumerator (IO-WE) for 2-dimensional PCs with SPC component codes and a lower bound for those with more dimensions are available in [2].

The student will understand the ideas presented there and develop a software to compute them efficiently.

References:

[1] http://ieeexplore.ieee.org/document/512606/

[2]  ttps://dspace.library.uvic.ca/bitstream/handle/1828/5487/Rankin_DavidM_EURASIPJAdvSignalProcess_2005.pdf?sequence=1&isAllowed=y

Prerequisites

Necessary:

  • Basics of channel coding
  • C++ (The applicant can suggest an alternative with a good reason)

Related Courses:

  • Channel Coding
  • Channel Codes for Iterative Decoding
  • Coding Theory for Storage and Networks

Contact

mustafa.coskun@tum.de

Supervisor:

Mustafa Coskun

Student

Enes Aksoy

Master's Theses

Path-Loss Average Techniques for the Nonlinear Fourier Transform

Path-Loss Average Techniques for the Nonlinear Fourier Transform

Keywords:
nonlinear Fourier transform, solitons, optical transmission systems, nonlinearity

Short Description:
In this thesis path-loss average techniques for transmission with modulation in the nonlinear Fourier domain are evaluated.

Description

In an attempt to improve achievable rates of optical communication systems in the high input power regime, modulation via the nonlinear Fourier transform (NFT) has attracted some attention in recent years. Since the NFT was conceived for the deterministic lossless nonlinear Schrödinger equation (NLSE), the fiber loss present in realistic optical communication systems degrades the achievable data rates for NFT aided communication systems. Some mitigation techniques are known that can be used to mitigate the negative effect of fiber loss during propagation.

In this thesis the student is is required to make him-/herself familiar with the fundamental concepts of the nonlinear Fourier transform (NFT) and implement algorithms for the forward and backward transforms. Then a purely solitonic transmission systems should be evaluated with and without path-loss averaged (PLA) approaches for different fiber models and amplification schemes.

Prerequisites

Lecture: Optical Communication Systems

Other: Matlab

Contact

benedikt.leible@tum.de

Supervisor:

Student

Thomas Göttsberger

Compression of Quantized Compressive Sensing Measurements

Compression of Quantized Compressive Sensing Measurements

Description

This thesis investigates compression algorithms for the quantized measuremends of a compressive sensing system.

Supervisor:

Student

Rami Ezzine

Modeling Complex Communication Channels Using Generative Adversarial Networks

Modeling Complex Communication Channels Using Generative Adversarial Networks

Description

We use Generative Adversarial Networks (GAN) to model various channels that might arise in a communications system.

Supervisor:

Student

Amir Hossein Rezaeitabar

Capacity Bounds for Amplitude-Constrained MIMO channels

Capacity Bounds for Amplitude-Constrained MIMO channels

Description

.

Prerequisites

.

Supervisor:

Andrei Nedelcu - (Politecnico di Milano)

Student

Saswati Mitra

Concatenated Codes for Error Correction in DNA Storage

Concatenated Codes for Error Correction in DNA Storage

Description

Encoding information into synthetic DNA is a novel approach for data storage. Due to its natural robustness and size in molecular dimensions, it can be used for long-term and very high-density archiving of data. Since the DNA molecules can be corrupted by thermal processes and the writing/reading process of DNA molecules can be faulty, it is necessary to encode the data using error-correcting codes. Due to the channel model concatenated codes are a suitable candidate for efficient error correction.

The student will analyze existing schemes for error correction in DNA storage based on concatenated codes and develop improved methods using soft information in the outer code, unequal error protection and list recovery. The improvements will be analyzed anlytically and by simulations.

Prerequisites

- Channel coding, basic probability theory, experience in programming

- Optional: Coding theory for storage and networks

Supervisor:

Student

Yuxuan Shi

Kanalschätzung für leitungsgebundene Kanäle bei zeitvarianten Störungen

Kanalschätzung für leitungsgebundene Kanäle bei zeitvarianten Störungen

Description

- Theoretische/analytische Betrachtung von Kanalschätz-
verfahren für OFDM-Systeme
- Simulation eines trainingsbasierten Kanalschätzers in Matlab
- Implementierung und Test des Kanalschätzverfahrens in
VHDL und Synthese der digitalen Schaltung in einem FPGA
- Funktionsverifikation anhand ein

Contact

Supervisor:

Karl-Tasnad Kernetzky - Dr. Thomas Weidinger (Rosenberger)

Student

Matthias Lechner

Secure Identification for Gaussian Channels

Secure Identification for Gaussian Channels

Description

In next generation connectivity systems, which rely on robust and low-latency information exchange, there exists communication tasks in which the Ahlswede/Dueck identification scheme is much more efficient
than Shannon’s transmission scheme. In this thesis one should provide a coding scheme for secure identification and determine the secrecy
capacity of the Gaussian Channel.

Prerequisites

Information Theory, Signal Theory

Supervisor:

Student

Wafa Labidi

Protected Code-based Post-Quantum Security

Protected Code-based Post-Quantum Security

Description

The foreseeable breakthrough of quantum computers represents a risk for communication which uses public-key cryptography. In order to prepare for such an event, embedded devices must integrate post-quantum cryptography, a set of algorithms based on mathematical problems that remains secure even in the presence of the quantum computers. Code-based is one of the most promising post-quantum cryptography. However, the implementation of code-based cryptography has two main challenges: i) satisfy performance and power constraints; and ii) resist side-channel attacks, which uses leakages derived from the implementation (timing, power or electromagnetic characteristics) to retrieve the secret information. The goal of this thesis is to implement a protected version (resistant to side-channel attack) of the newest version of the Gabidulin-Paramonov-Tretjakov code-based post-quantum cryptosystem. This Master thesis will be supervised by Dr. Johanna Sepúlveda (Chair of Security in Information Technology) and Prof. Dr.-Ing. Antonia Wachter-Zeh (Professorship for Coding for Communications and Data Storage).

Supervisor:

Julian Renner, Antonia Wachter-Zeh

Student

Robin Winzler

Simulation of Improved Staircase Code Decoding

Simulation of Improved Staircase Code Decoding

Description

Staircase codes, as introduced in 2011 by Smith et. al. [1], are a hardware friendly code design for error correction in optical communication systems. However, the choices of parameters such as block size and code rate that achieve a desired output bit error rate are limited by the error floor of the decoder. A new and improved decoder has been devised [2], allowing for staircase codes with a scope of new parameters to be considered for optical communication. While estimations show the significant improvements, the high throughput required to simulate the error floor can only be achieved with an efficient and parallelisable implementation. The main goal of the thesis is the implementation  in VHDL and simulation on an FPGA of the new decoder  in order to provide further evidence for the estimated performance.

[1] https://arxiv.org/abs/1201.4106
[2] https://arxiv.org/abs/1704.01893

Prerequisites

interest in channel coding, knowledge in VHDL

Supervisor:

Student

Yichao Li

Forschungspraxis or MSCE Internships

Efficient decoding of Reed-Muller/polar codes over binary erasure channel

Efficient decoding of Reed-Muller/polar codes over binary erasure channel

Description

The student will understand the main principles of Reed-Muller and polar codes. Then, s/he will implement an efficient decoder for binary erausre channel.

Supervisor:

Mustafa Coskun

Student

Zeliha Asena Kirik

Preparing a data-path for Optical Downlink

Preparing a data-path for Optical Downlink

Description

The work is on developing or modifying subsystems from the Osiris4CubeSat project at DLR, which is about developing an optical payload with the primary goal of demonstrating high speed optical downlinks between a CubeSat and the Earth. 

Supervisor:

Ginni Khanna, Norbert Hanik - Anil Morab Vishwanath (DLR)

Student

Lorenzo Castelvero

GPU Based Decoding of QC LDPC Codes

GPU Based Decoding of QC LDPC Codes

Description

In this research internship, the student will develop a GPU based belief propagation (BP) decoder for layered message passing. Extensive simulation campaigns and throughput comparisons to legacy decoders will be conducted.

Supervisor:

Student

Patrick Willner

Message Passing Decoding of BMERA Codes

Message Passing Decoding of BMERA Codes

Description

In this research internship, the student will implement a belief propagation (BP) decoder for message passing decoding of BMERA codes, also known as convolutional polar codes. For this, the student will first start with a conventional  BP decoder for polar codes and then take the necessary steps to adapt the algorithm for BMERA codes. Eventually, a numerical performance evaluation by means of frame error rate simulations should be made.

Prerequisites

Channel Coding, Channel Codes for Iterative Decoding, Information Theory

Supervisor:

Student

Zvezdana Kuzmanovic

List Viterbi Decoder for Short Convolutional Codes

List Viterbi Decoder for Short Convolutional Codes

Description

Short convolutional codes with large memory are best codes for short blocklengths at the expense of high complexity [1]. The student should understand first the Viterbi algorithm and then implement list Viterbi decoder [2].

[1] https://arxiv.org/pdf/1610.00873.pdf

[2] http://www2.ensc.sfu.ca/people/faculty/cavers/ENSC805/readings/42comm02-seshadri.pdf

Prerequisites

Necessary:

  • Basics of channel coding
  • Matlab or C++

Related Courses:

  • Channel Coding
  • Channel Codes for Iterative Decoding

Contact

mustafa.coskun@tum.de

Supervisor:

Mustafa Coskun

Student

Wenqi Jia

PAT using Polar Codes

PAT using Polar Codes

Description

Over the fading channels, channel coefficients should be estimated to use most of the decoding algorithms, e.g., Viterbi, BP, etc. In this internship the student is expected to understand the Polar Codes together with the successive cancellation decoding algorithm and use them in a PAT scheme over block-fading channels.

Supervisor:

Mustafa Coskun - (DLR)

Student

Marvin Xhemrishi

GPU-Implementation of the Split-Step Fourier Method with CUDA

GPU-Implementation of the Split-Step Fourier Method with CUDA

Keywords:
split-step Fourier method, parallel programming, programming in C, CUDA, code performance optimization and evaluation

Short Description:
In this thesis the split-step Fourier method has to be implemented utilizing GPU computation via CUDA.

Description

Within the scope of this thesis, the split-step Fourier method (SSFM), a numerical method solving the nonlinear Schrödinger equation (NLSE), will be implemented. The implementation has to be conducted, using the CUDA framework for parallel implementation on graphical processing units (GPUs), in the programming language C. The main focus is on reducing the computation time for the SSFM, by utilizing the massive parallel computational capabilities of GPUs. Even though Matlab can natively utilize GPU hardware for certain operations, the reduction in overall computation time is far below the expected reduction with an implementation fully written in C and called from Matlab (e.g. via Mex). Over the course of this thesis, the student will get familiar with the following concepts: split-step Fourier method, parallel programming, programming in C, CUDA, code performance optimization and evaluation.

Prerequisites

There are no special requirements needed to apply for this thesis. Still some knowledge in the following fields is beneficial: Optical Communication Systems, Basics in C programming (or basics in any other programming language), Matlab (Mex)

Contact

benedikt.leible@tum.de (Please attach an overview of your current grades to the email)

Supervisor:

Student

Thomas Göttsberger

Error-Correction for Partially Stuck Memory Cells

Error-Correction for Partially Stuck Memory Cells

Description

.

Supervisor:

Student

Venkatesh Satagopan

Internships

Implementation of Belief Propagation Decoding of Polar Codes

Implementation of Belief Propagation Decoding of Polar Codes

Description

.

Supervisor:

Student

Stefan Hägele

Porting Flash Programs of the lntWWW to HTML5 and Javascript

Porting Flash Programs of the lntWWW to HTML5 and Javascript

Description

Porting Flash Programs of the lntWWW to HTML5 and Javascript

Supervisor:

Student

Carolin Mirschina

Deep Learning Spezialisierung

Deep Learning Spezialisierung

Description

.

Supervisor:

Antonia Wachter-Zeh - (Luminovo Artificial Intelligence GmbH)

Student

Philipp Wolters

Automatisierung der Datenanalyse

Automatisierung der Datenanalyse

Description

.

Supervisor:

Gerhard Kramer - Zhang Xu (Ford Technology Development Center)

Student

Zibo Zhou

Funknetzoptimierungdes Vodafone Mobilfunknetzes

Funknetzoptimierungdes Vodafone Mobilfunknetzes

Description

.

Supervisor:

Gerhard Kramer - Georg Rötzer (Vodafone GmbH)

Student

Martin Göllner

Installation der Bodenlage Stuttgart im Center Langen

Installation der Bodenlage Stuttgart im Center Langen

Description

.

Supervisor:

Student

Stefan Schäffler

SDxR Legacy Waveform Integration

SDxR Legacy Waveform Integration

Keywords:
SDxR Legacy Waveform Integration

Short Description:
SDxR Legacy Waveform Integration

Description

SDxR Legacy Waveform Integration

Prerequisites

SDxR Legacy Waveform Integration

Supervisor:

Gerhard Kramer - (ROHDE & SCHWARZ GmbH & Co. KG München)

Student

Constantin Runge

Channel Model Simulator

Channel Model Simulator

Short Description:
Investigation of Winner II channel model

Description

The Student investigates the Winner II channel model.

Supervisor:

Student

Yosri Ben haj ali

Student Assistant Jobs

C Implementation of the Faure-Loidreau Public-Key Cryptosystem

C Implementation of the Faure-Loidreau Public-Key Cryptosystem

Description

The realistic threat of a quantum supercomputer has motivated research on post-quantum cryptography. The Faure-Loidreau Public-Key Cryptosystem is considered to be secure against attacks of quantum computers.

Supervisor:

Julian Renner

Student

Marvin Xhemrishi