• Your own topic – English only

You have the possibility until the beginning of the semester to come up with your own topic and find a supervisor who is willing to mentor more students.

• Location privacy – English only

Location-based services (LBSs) have become an essential part of our daily lives. In such services, users offer their (precise) locations to service providers in return of benefiting from the service. However, offering location data to service providers put users' privacy at huge risk. Often these locations are associated with points of interest (POIs) of the users. Therefore, service providers are able to infer users' private behavior by knowing these POIs with a relatively high degree of certainty. For this reason, the adoption and deployment of location privacy protection mechanisms (LPPMs) are essential to protect users' privacy.
In this seminar, you will investigate and discuss the existing LPPMs as well as the privacy metrics that reflect how much privacy a user gains when applying a protection mechanism.

• Network Security Breaches – English only

The goal of this seminar is the outlining of popular network security breaches (2-3 examples).  Subsequently, state-of-the-art protection or detection approaches against these presented breaches should be explained.

Leonard Bradatsch

• First Packet Authentication – English only

All three of the above stated techniques describe an authentication process that starts with the very first packet a client sends to an network entry node when entering a network or accesing a service. This kind of authentication can be performed without having the communication parties any messages exchanged before the authentication happened.
The seminar paper should describe these techqniues and should outline the different existing approaches to perform such a first packet authentication.

Leonard Bradatsch

• Accessibility in Modern Web Applications – English only

Due to "Statistisches Bundesamt" 9.52% of germany's population had a severe disability in 2019. Many of various disabilities e.g. visual impairement influence the user interaction with web applications. Although most of the impact of a disability can be compensated with accessible tools like special input devices and client-software, such tools are heavily relying on extra information from the application itself.
This seminar paper should summarize and describe current concepts, tools and frameworks for development of an accessible modern web application.

• State of the art in biological molecular dynamics simulation software – English only

Parallel molecular dynamics (MD) simulations are critical to researching the interactions of hundreds of millions of atoms
and molecules in biomolecular systems like DNA transcription, protein assembly and many more.
MD simulations are used in molecular medicine and drug research or to create visualizations of biological systems,
which further our understanding of life's secrets.
MD frameworks such as NAMD and GENESIS are highly parallelized, high-performance software packages
which can scale up to hundreds or even many thousands of CPU and GPU cores.
This seminar should give an overview of the basic operating principles of molecular dynamics simulation software,
including examples and state of the art research on the topic.

Gerhard Habiger

• CheapBFT: a resource-efficient replication approach – English only

State-machine replication is a technique to deploy mulitple server replicas at once and let them do the same on behalf of client requests. In case of faults the service is still available. Even more, the service can tolerate arbitrarily misbehaving replicas (up to a certain number) by forming a quorum of votes. This needs typically 3f+1 replicas if we want to tolerate up to f faulty replicas. CheapBFT is another approach that needs less resources and still provides fault tolerance against arbitrarily misbehaviour, so called Byzantine failures. This seminar work is supposed to characterise CheapBFT in comparison to traditional approaches against Byzantine faults by state-machine replication.

There is only a single scientific paper describing CheapBFT. However, the candidate student is requested to introduce state-machine replication by having a look at papers about fundamental aspects and related traditional approaches.

Franz J. Hauck

• BFT Consensus Protocols: An Overview – English only

Byzantine fault tolerant (BFT) consensus protocols enable a set of distributed nodes to agree on a series of values, even if a minority of the nodes is faulty or malicious.
Originally developed as a fundamental mechanism for state machine replication (SMR), they have gained a lot of attentation from blockchain researchers recently, because they provide an efficient alternative to PoW mechanisms in (permissioned) blockchains.
However, since blockchains are operated in different network environments compared to SMR systems, a new generation of consensus protocols has evolved in recent years.
The task of this seminar is to provide a gentle introduction to the underlying consensus problem and a comparison of different BFT consensus protocols.

Alexander Heß

• Attacks on Automotive Systems – English only

Over the last 12 years, a large number of security attacks on automotive systems were discovered and published.
This Seminar topic should provide an overview over these attacks, identify common structures, present particularly noteworthy examples and conclude with some lessons learned. A survey article on automotive attacks is provided as a starting point.

Frank Kargl

• Secure over-the-air updates for vehicles – English only

Modern vehicles have become complex software products that need to be maintained for several decades in order to ensure their safe operation on the road. Typically, software is updated much more frequently than a car is taken to the shop for maintainance and therefore other ways to deploy software updates need to be considered. Over-the-air updates allow quick and cheap deployment of new or updated software to a large vehicle fleet, but security aspects need to be considered very carefully for obvious reasons.
In this seminar, you will identify and compare different proposals for implementing over-the-air updates for vehicles from a security perspective.

Dominik Mauksch

• Concepts and techniques in Library Isolation – English only

Different applications are isolated in modern operating systems so that they can only exchange information via predefined interfaces, e.g. shared memory. However, libraries used in applications usually run in the application context and thus with the same privileges and address space as the main application. They therefore increase the attack surface of an application, e.g. in the case of return-oriented programming attacks.
In this seminar, you should have a look into different approaches of in-application and library isolation techniques and compare them, e.g. with respect to their performance impacts and security assumptions.

Dominik Mauksch

• Evolution of the Hypertext Transfer Protocol – English only

Since HTTP was first developed at CERN in 1989 it has seen many changes. While it still serves its original purpose, HTTP also became a ubiquitous application protocol with a variety of uses that go far beyond the transportation of hypertext documents. But not only the applications of HTTP have changed drastically in the last 32 years, the protocol stack has also evolved. Nowadays more than 90% of the HTTP traffic is encrypted using TLS and while the adoption of HTTP/2 is steadily increasing, the IETF is in the process of finalizing the HTTP/3 protocol with 71% of running web browsers already supporting it.
This seminar should explore the history and evolution of HTTP and discuss old and new challenges and issues of the protocol and how newer HTTP versions address previous issues.

• Authentication in Web Applications – English only

Modern web application development encompasses a variety of approaches to implement authentication and session handling, ranging from traditional password-based authentication and cookie-based session handling to multi-factor authentication and complex authentication protocols, such as OAuth, OpenID, and SAML. Depending on the requirements of the application some approaches are better suited than others. This seminar should explore the authentication design space and compare stateful and stateless session handling approaches.

Echo Meißner

• Differential Privacy – English only

Differential Privacy (DP) is a privacy enhancing technology (PET) which should help to increase the privacy of users submitting data. Very simply spoken, during the data gathering phase noise is added to the value of each user in such a way that the individual user cannot be distinguished of others, and in the evaluation phase, the noise is taken into consideration to calculate certain analytics over the whole dataset. This oversimplified example was only used to explain the basics of the idea behind this PET, which has more specifics and potential.
In this seminar, you will take a deeper look into differential privacy, explain it with the help of an example, and present a simple framework using DP in a current application.

Michael Wolf

• The Boring Loop - a critical analysis – English only

Elon Musk opened recently the Las Vegas Convention Center Loop which was build by his company "The Boring Company". This project was created to speed up travel time in cities between specific locations, but also show that the cost of tunnel drilling can be reduced drastically. With the Loop in Vegas now finished, theses promises can be examined critically, as well as the concept of individual transport through tunnels.
In this seminar, you will have to compare the cost and performance of the Vegas Center Loop with other tunnel projects in the USA, Germany and other countries. Additionally you have to compare this concept with other transportation methods, like subway, bus, E-Scooter, etc...

Michael Wolf

• Secure Multi-Party Computation – English only

The goal of Secure Multi-Party Computation (MPC) is to enable parties to work together without ever knowing one another's confidential information. It plays an important role in solving security and privacy issues and there are many examples of where it can be helpful.
The aim of this seminar is to investigate MPC with respect to both theoretical and practical aspects and look at uses in real world applications.

Migena Ymeraj

• Distributed Machine Learning – English only

Due to the poor scalability and efficiency of learning algorithms, Machine Learning cannot handle large-scale data. This issue gave rise to Distributed Machine Learning. Even though it is a promising line of research, it still faces a lot of challenges.
The goal of this seminar is to discover the importance of Distributed Machine Learning, while comparing it with traditional Machine Learning environments and investigating its challenges.

Migena Ymeraj

• Containerization Technologies – English only

Kubernetes, Docker Swarm, Openshift, Portainer, Apache Mesos and others are in everyones mind. They've been made to revolutionize how to separate several applications and software stacks from each other. But did you ever had a look at the underlying technologies?
In this survey the student is asked to examine, compare and summarize the main differences between many (at least five) different OS-level virtualization technologies currently existing in modern desktop and server environments. As a first task, the student needs to introduce the main ideas behind the concepts of such user space instances and its differences to other approaches like hardware-based virtualization. Maybe, also an historical overview can be given as a guide for the reader. In a second step the student has to examine the differences of the proposed technologies and summarize them according to their dis-/advantages. Finally, the student has to conclude which technology might be the best for which use cases, e.g. working in privacy-aware, secure or speed-optimized environments.

Thomas Bläsing

• Timing Attacks - An Overview – English only

Classical cryptographic research deals with adversaries of polynomially bounded computational power. However, this attack model is not always realistic. In particular, an attacker may be able to measure the time it takes to run cryptographic algorithms. As often the running time of an algorithm depends on its input, this can be used to gather various information about the inputs to the algorithm or its internal workings. In a security context, sensitive data such as encryption keys and passwords may be recoverable by measuring the run time of programs. Examples are the run time of multiplications in the RSA algorithm, the POODLE and Lucky Thirteen Attack on TLS, as well as various  forms of Cache timing attacks (PRIME+PROBE, EVICT+TIME).
In this seminar, the student should give an overview of timing attacks, thereby explaining at least one example in depth. Further, some mitigations against timing attacks  should be discussed.

Henning Kopp (Schutzwerk GmbH)

• Padding Oracles – English only

Block ciphers only work on inputs that are a multiple of the cipher’s block length in commonly used modes such as CBC. As data usually comes in arbitrary lengths, inputs to these algorithms need to be extended (padded) to a multiple of the block length. This seemingly simple problem lead to quite huge cryptographic problems. Depending on the scenario, it is even possible that an attacker is able to completely decrypt the ciphertext or encrypt data of their choosing.
Your paper and presentation should explain the details of how a padding oracle works and how it can be used for decryption and ciphertext forgery. Moreover, you should give an overview of the problems this attack scenario caused as well as how the security community tries to avoid it nowadays.

Martin Lang (BMW Car IT)

• Secure In-Car Communication – English only - Master only (RTDS)

Modern cars contain an abundance of different ECUs controlling different aspects of the car's functionality. Along with an ever rising number of sensors and actuators, more and more control of the car is handed over to digital equipment. The rising complexity of these systems also leads to an increasingly large attack surface. Assuring the integrity and authenticity of in-car communication is therefore critical for the safety of driver, passengers, and other road users.
The target of this seminar is to analyse the current state of ECU communication security of different in-car protocols and bus systems and collecting and documenting potential challenges for the design of a secure ECU platform.

Thomas Lukaseder (Escrypt)

Sorted by Topics

Automotive
- Attacks on Automotive Systems (Frank  Kargl)
- Secure over-the-air updates for vehicles (Dominik Mauksch)
- Concepts and techniques in Library Isolation  (Dominik Mauksch)
- Secure In-Car Communication (Thomas Lukaseder)
- The Boring Loop - a critical analysis (Michael Wolf)

Cryptography
- Secure Multi-Party Computation (Migena Ymeraj)
- Padding Oracles (Martin Lang BMW)
- Timing Attacks - An Overview (Henning Kopp Schutzwerk)

Distributed Systems
- State of the art in biological molecular dynamics simulation software (Gerhard Habiger)
- CheapBFT: a resource-efficient replication approach (Franz J. Hauck)
- BFT Consensus Protocols: An Overview (Alexander Heß)

Network Security
- Network Security Breaches (Leonard Bradatsch)
- First Packet Authentication (Leonard Bradatsch)

Privacy
- Location privacy (Ala'a Al-Momani)
- Differential privacy (Michael Wolf)
- Distributed Machine Learning (Migena Ymeraj)

Software Engineering
- Accessibility in Modern Web Applications (Eugen Frasch)
- Evolution of the Hypertext Transfer Protocol (Echo Meißner)
- Authentication in Web Applications (Echo Meißner)
- Containerization Technologies (Thomas Bläsing)

• Your own topic – English only

You have the possibility until the beginning of the semester to come up with your own topic and find a supervisor who is willing to mentor more students.

• Differential privacy – English only

Statistics about user data are used heavily for improving services, performing marketing studies, and many other purposes. The acquisition and processing of potentially sensitive information poses serious privacy threats to individuals participating in, e.g., surveys. Such information might include birthdate, gender, religion, and nationality, from which identifying an individual becomes a trivial task. Moreover, inferring more sensitive information about a specific individual becomes also possible. To solve issue, many service providers, e.g., Apple and Google, started using the so-called “Differential Privacy” to preserve their users’ privacy.
In this seminar, you will investigate differential privacy with respect to both theoretical and practical aspects. In addition to addressing the privacy guarantees of differential privacy, your task will then be addressing how major tech companies are implementing differential privacy nowadays.

• Location privacy – English only

Location-based services (LBSs) have become an essential part of our daily lives. In such services, users offer their (precise) locations to service providers in return of benefiting from the service. However, offering location data to service providers put users' privacy at huge risk. Often these locations are associated with points of interest (POIs) of the users. Therefore, service providers are able to infer users' private behavior by knowing these POIs with a relatively high degree of certainty. For this reason, the adoption and deployment of location privacy protection mechanisms (LPPMs) are essential to protect users' privacy.
In this seminar, you will investigate and discuss the existing LPPMs as well as the privacy metrics that reflect how much privacy a user gains when applying a protection mechanism.

• Network Security Breaches – English only

The goal of this seminar is the outlining of popular network security breaches (2-3 examples).  Subsequently, state-of-the-art protection or detection approaches against these presented breaches should be explained.

Leonard Bradatsch

• First Packet Authentication – English only

All three of the above stated techniques describe an authentication process that starts with the very first packet a client sends to an network entry node when entering a network or accesing a service. This kind of authentication can be performed without having the communication parties any messages exchanged before the authentication happened.
The seminar paper should describe these techqniues and should outline the different existing approaches to perform such a first packet authentication.

Leonard Bradatsch

• Accelerating Techniques for 3D Ray Tracing – English only

Although 3D ray tracing was already used in the 70s to render simple animations, it got popular due to the gaming industry the past 10 years.
One of the reasons for its slow development as rendering technique is that ray tracing needs much more computational power than other rendering techniques to provide results in real-time.
The goal of this seminar is to look on the multiple algorithms, techniques and strategies that have been developed in the past to accelerate ray tracing.

Eugen Frasch

• The Elixir Programming Language – English only

Elixir is a functional and concurrent programming languague for developing highly scalable and reliable applications.
The goal of this seminar is to provide an introduction to Elixir, describe its main features and provide an example to demonstrate its fault tolerant characteristics.

Eugen Frasch

• State Machine Replication in the Libra Blockchain – English only

In 2021, Facebook wants to start a cryptocurrency called Diem, former Libra. This topic is supposed to summarise the underlying blockchain operations and present details in the seminar report and presentation. Starting from a given article the student should investigate how the blockchain works. As it has a closed group of participants, a technique called state-machine replication is used. Here all participants have to agree how to build the next block of the blockchain. The agreement protocol will be one aspect of this topic.
Beside technical aspects, which clearly have the main focus, the student may also want to deal with social and economic aspects of the new currency.

Franz J. Hauck

• Evolution of the Paxos Algorithm – English only

The Paxos algorithm enables a group of nodes to agree on a single value, even if the network communication or individual nodes temporarily drop out. While the original algorithm was layed out multiple decades ago, several modified and extended versions have been proposed over the years.
The goal of this seminar is to outline the core aspects of the Paxos algorithm, and discuss the modifications of a few of the more recent versions.

Alexander Heß

• The Hyperledger Project – English only

Hyperledger is an open-source project that comprises a set of tools and frameworks for building open-source blockchains.
The goal of this seminar is to provide an overview of the project's scope, capabilities, and discuss it's relevance for real-world applications.

Alexander Heß

• Secure over-the-air updates for vehicles – English only

Modern vehicles have become complex software products that need to be maintained for several decades in order to ensure their safe operation on the road. Typically, software is updated much more frequently than a car is taken to the shop for maintainance and therefore other ways to deploy software updates need to be considered. Over-the-air updates allow quick and cheap deployment of new or updated software to a large vehicle fleet, but security aspects need to be considered very carefully for obvious reasons.
In this seminar, you will identify and compare different proposals for implementing over-the-air updates for vehicles from a security perspective.

Dominik Mauksch

• Securing CAN bus – English only

Controller Area Network (CAN) – despite its age and bandwith limitations – is still a widely used bus standard to distribute safety-critical signals in modern vehicles. Research has shown that CAN does not have sufficient security measures to ensure basic security goals and therefore can be used by attackers to compromise the safety of a car and its passengers.
In this seminar, you should have a look into different approaches that make CAN more secure and compare them regarding e.g. their effectiveness, efficiency and real-time capability.

Dominik Mauksch

• State of the Art of Web Application Security – English only

The field of web applications is constantly and rapidly evolving, but so are attacks targeting them. For this reason the World Wide Web Consortium (W3C) assembled a working group to develop technical and policy mechanisms to improve the security for applications on the Web. In recent years, this Web Application Security Group proposed various drafts for mechanisms of which some have been refined into W3C recommendations and are now implemented in all major browsers (such as CSP and SRI). The establishment of Let's Encrypt laid the ground stone of a widespread adoption of TLS, which was further refined in TLS 1.3 and supplemented by new technologies, such as certificate transparency (CT).
This seminar should give an overview of current web application security techniques, discuss W3C recommendations and other developments in the area of web application security, and practical implications for current web applications.

• DevOps Monitoring using Grafana and Prometheus – English only

An important aspect of modern continuous delivery lifecycles is the constant monitoring of deployed applications.
This is not only used to assess the health of applications, but can also provide an overview of user activity and reveal performance issues in a production environment.
Nowadays, developers can rely on established and ready-to-use solutions instead of developing their own monitoring software.
This seminar should introduce the aspects of modern DevOps monitoring and give an overview of a typical monitoring stack (e.g., the open source software Prometheus and Grafana).
This overview should highlight the challenges that arise when monitoring different layers of an application deployment (such as hardware, network, database, and application metrics).

Echo Meißner

• Password Managers – English only

In the current time, users have multiple accounts on different platforms. To abide security guidelines, they should have a different password for each account resulting in a many passwords to remember. Password managers help with this task by storing this sensitive information in an encrypted database, protected by a master password. However, once in a while security issues about password managers make the news.
In this seminar, password managers (also theoretical ones) should be compared to each other by the security they provide and if security breaches have been reported. Based on this, the advantages of password managers need to be evaluated to the disadvantages they have.

Michael Wolf

• Secure Multi-Party Computation – English only

The goal of Secure Multi-Party Computation (MPC) is to enable parties to work together without ever knowing one another's confidential information. It plays an important role in solving security and privacy issues and there are many examples of where it can be helpful.
The aim of this seminar is to investigate MPC with respect to both theoretical and practical aspects and look at uses in real world applications.

Migena Ymeraj

• Distributed Machine Learning – English only

Due to the poor scalability and efficiency of learning algorithms, Machine Learning cannot handle large-scale data. This issue gave rise to Distributed Machine Learning. Even though it is a promising line of research, it still faces a lot of challenges.
The goal of this seminar is to discover the importance of Distributed Machine Learning, while comparing it with traditional Machine Learning environments and investigating its challenges.

Migena Ymeraj

• Containerization Technologies – English only

Kubernetes, Docker Swarm, Openshift, Portainer, Apache Mesos and others are in everyones mind. They've been made to revolutionize how to separate several applications and software stacks from each other. But did you ever had a look at the underlying technologies?
In this survey the student is asked to examine, compare and summarize the main differences between many (at least five) different OS-level virtualization technologies currently existing in modern desktop and server environments. As a first task, the student needs to introduce the main ideas behind the concepts of such user space instances and its differences to other approaches like hardware-based virtualization. Maybe, also an historical overview can be given as a guide for the reader. In a second step the student has to examine the differences of the proposed technologies and summarize them according to their dis-/advantages. Finally, the student has to conclude which technology might be the best for which use cases, e.g. working in privacy-aware, secure or speed-optimized environments.

Thomas Bläsing

• State of the Art in Attacking Hash Functions – English only

Hash functions such as MD5 or SHA1/2/3 are usually employed to ensure the integrity of a file by computing a fingerprint of the data. Common applications include  white/blacklists, signatures, as well as data bases for forensic analysis.
One major requirement of a secure hash function is that it should be hard to compute collisions, i.e., to find two files with the same hash value. The notion of hardness can be further formalized, but intuitively means that there should not be an algorithm which computes collisions faster than brute-force.
For some hash functions, such as SHA1 or MD5, computing collisions can be done faster than brute-force. These hash functions are considered insecure, but are still in use. As an example, SHA1 is used in git. Still, exploitation of hash collisions is only rarely seen in the wild. One reason is that the output of fast algorithms for finding collisions are usually bytes of high entropy and do not conform to valid file formats.
The goal of this seminar is to give an overview of the current attacks against hash functions (fastcoll, unicoll, hashclash, ...) as well as their constraints (e.g., identical/chosen prefix collision). Depending on the personal interest and previous knowledge of the participant, these attacks can be explored in more depth, or alternatively the exploitation of hash collisions can be described.

Henning Kopp (Schutzwerk GmbH)

• Padding Oracles – English only

Block ciphers only work on inputs that are a multiple of the cipher’s block length in commonly used modes such as CBC. As data usually comes in arbitrary lengths, inputs to these algorithms need to be extended (padded) to a multiple of the block length. This seemingly simple problem lead to quite huge cryptographic problems. Depending on the scenario, it is even possible that an attacker is able to completely decrypt the ciphertext or encrypt data of their choosing.
Your paper and presentation should explain the details of how a padding oracle works and how it can be used for decryption and ciphertext forgery. Moreover, you should give an overview of the problems this attack scenario caused as well as how the security community tries to avoid it nowadays.

Martin Lang (BMW Car IT)

• Secure In-Car Communication – English only

Modern cars contain an abundance of different ECUs controlling different aspects of the car's functionality. Along with an ever rising number of sensors and actuators, more and more control of the car is handed over to digital equipment. The rising complexity of these systems also leads to an increasingly large attack surface. Assuring the integrity and authenticity of in-car communication is therefore critical for the safety of driver, passengers, and other road users.
The target of this seminar is to analyse the current state of ECU communication security of different in-car protocols and bus systems and collecting and documenting potential challenges for the design of a secure ECU platform.

Thomas Lukaseder (Escrypt)

Sorted by Topics

Blockchain
- State Machine Replication in the Libra Blockchain (Franz J. Hauck)
- Evolution of the Paxos Algorithm (Alexander Heß)
- The Hyperledger Project (Alexander Heß)

Automotive
- Secure over-the-air updates for vehicles (Dominik Mauksch)
- Securing CAN bus (Dominik Mauksch)
- Secure In-Car Communication (Thomas Lukaseder)

Privacy
- Differential privacy (Ala'a Al-Momani)
- Location privacy (Ala'a Al-Momani)
- Password Managers (Michael Wolf)
- Distributed Machine Learning (Migena Ymeraj)

Cryptography
- Secure Multi-Party Computation (Migena Ymeraj)
- Padding Oracles (Martin Lang BMW)
- State of the Art in Attacking Hash Functions (Henning Kopp Schutzwerk)

Software Engineering
- State of the Art of Web Application Security (Echo Meißner)
- DevOps Monitoring using Grafana and Prometheus (Echo Meißner)
- Containerization Technologies (Thomas Bläsing)
- The Elixir Programming Language  (Eugen Frasch)
- Accelerating Techniques for 3D Ray Tracing (Eugen Frasch)

Network Security
- Network Security Breaches (Leonard Bradatsch)
- First Packet Authentication (Leonard Bradatsch)

• Your own topic – English only

You have the possibility until the beginning of the semester to come up with your own topic and find a supervisor who is willing to mentor more students.

✘ Pattern-Selection Methods – English only

Security and Privacy engineering have gained a lot of attention recently. Many methodologies and tools have been proposed to assist practitioners coming up with secure and privacy-enhanced systems. Security and privacy patterns are considered among the backbones to introduce such systems.
In this seminar, you will investigate the current selection methods to select suitable and fitting security and privacy patterns according to, e.g., their properties and the system's context they are envisioned to be deployed in. According to the analysis you carry on, you are expected to have a clear overview of what propoerties such selection-methods are based on and how patterns were classified accordingly.

✘ Network Security Breaches – English only

The goal of this seminar is the outlining of popular network security breaches (2-3 examples).  Subsequently, state-of-the-art protection or detection approaches against these presented breaches should be explained.

Leonard Bradatsch

✘ First Packet Authentication / Firewall Poking / Single Packet Authentication – English only

All three of the above stated techniques describe an authentication process that starts with the very first packet a client sends to an network entry node when entering a network or accesing a service. This kind of authentication can be performed without having the communication parties any messages exchanged before the authentication happened.
The seminar paper should describe these techqniues and should outline the different existing approaches to perform such a first packet authentication.

Leonard Bradatsch

✘ SimGrid – English only

The increased popularity of distributed systems over the last years also increased the need for test environments to evaluate and validate distributed applications.
Unfortunately, the setup of a distributed system for application testing is expensive and time consuming. A solution for this problem is frameworks that are able to simulate a distributed system or at least their requirements. Such a simulation framework is SimGrid.
The goal of this seminar is to outline the functionalities of SimGrid, its advantages and drawbacks and also how it is used in application development.

Eugen Frasch

✘ Real-Time in Distributed Systems – English only

The term "Real-Time" is one of the most confusing terms in computer-science. However from spaceflight to medical systems real-time systems have an high impact on our daily life.
This seminar should have a closer look to real-time distributed systems and their development process.

Eugen Frasch

✘ OpenWorm – English only

The OpenWorm project is an international open science project, working on a complete simulation of a multicellular organism called Caenorhabditis elegans. This roughly 1mm long worm was the first animal that had its whole genome fully sequenced, and is also, as of today, the only organism of which the full 'connectome' is known, i.e., of which the full map of all connections of its 302 neurons has been mapped. A full simulation of the worm's neural network as well as creating a full model of the physical movement of the worm are among the project's goals, and have spurred the creation of dedicated physics engines aimed at simulating biological systems.
This seminar should give a brief overview of the entire OpenWorm project, its current status and recent activities, and further present one of its core concepts (e.g., the simulation engine, or the neurological simulation model, etc.) in greater detail.

Gerhard Habiger

✘ Swarm Intelligence – English only

Swarm intelligence is a hot topic especially in robotics and UAV research. With multiple cooperative agents, it is possible to solve tasks that would otherwise be infeasible or very costly to achieve.
The goal of this seminar is to give an introduction to swarm intelligence and to examine further aspects of reasearch in this area in more detail. The focus could be on security, safety and/or privacy of robotic swarm agents, but the seminar is not strictly limited to this direction.

✘ Break-glass Access Control Systems in Medical Devices – English only

With health-related data being considered as very sensitive under the EU's General Data Protection Law and medical devices performing life-critical operations on the human body, we expect data and functionalities in healthcare devices to be secured in the best way possible by encryption, pseudonymization and access control among others.
In the case of emergency, however, break-glass mechanisms may be needed to bypass security mechanisms that would otherwise prevent proper treatment.
The goal of this seminar is to introduce modern break-glass access control mechanisms used in healthcare and discuss their implications on security, safety and privacy.

Matthias Matousek

• Surveying Peripheral Sensors in Context of Mobile Crowdsensing – English only

In this seminar report, the students are required to survey the current state of peripheral sensors that can be coupled with smartphones to further accurate the mobile crowdsensing applications. These peripheral sensors can be coupled with smartphones using Bluetooth technology or wifi. The students would be further required to study the current state of coupling technologies. In addition to the general perspective, the students will survey the peripheral sensors, their technology, and limitations within the context of mHealth (mobile health).

Muntazir Mehdi

✘ State of the Art of Web Application Security – English only

The field of web applications is constantly and rapidly evolving, but so are attacks targeting them. For this reason the World Wide Web Consortium (W3C) assembled a working group to develop technical and policy mechanisms to improve the security for applications on the Web. In recent years, this Web Application Security Group proposed various drafts for mechanisms of which some have been refined into W3C recommendations and are now implemented in all major browsers  (such as CSP and SRI). The establishment of Let's Encrypt layed the ground stone of a widespread adoption of TLS, which was further refined in TLS 1.3 and supplemented by new technologies, such as certificate transparency (CT).
This seminar should give an overview of and discuss W3C recommendations, other developments in the area of web application security, and practical implications for current web applications.

✘ Trusted Execution Environments – English only

Trust management is a central aspect of computer security. For instance, an operating system uses sandboxes to protect itself and other applications from viruses and malicious software, and cryptography is used to protect data in transit and at rest. With the advent of cloud computing, even the hardware that executes a particular software is not always considered trustworthy. Trusted Execution Environments (TEEs) try to relieve of the need to fully trust the hardware, by adding a secure area to the CPU that can guarantee code/data confidentiality and integrity through cryptographic means. Hence, protecting an application from untrusted hardware, software, and even privileged attackers (i.e., the operating system). Several TEE implementations can already be used today, such as Intel SGX and ARM TrustZone. While the former proprietary implementations often expect trust in the vector, open-source alternatives that address this weak point are already in development. 
In this seminar, you will investigate TEEs, highlight use-cases for this technology, and compare prominent representatives for their features and shortcomings.

Echo Meißner

✘ Quaternions and their Application in Computer Science – English only

Quaternions are the next extension of complex numbers which are complete regarding multiplication and division. They are often shown using three imaginary units: i,j and k. Quaternions are useful to describe rotations and motions in 3D space and are therefore applied in a variety of computer science fields, e.g., 3D game engines and robotics.
The seminar should convey a sensible understanding of quaternions and how to interpret them, as well as give an overview of their applications in computer science.

David Mödinger

✘ Time constraints of Security in CACC – English only

One application of Corporate Adaptive Cruise Control (CACC) is platooning where vehicles drive very close after each other to reduce the air resistance and therefore reduce fuel consumption. This, however, comes with safety risks due to the reduced distance gap to the following vehicle and therefore reduced reaction time. If the vehicles drive 100 km/h (~30m/s), a safety distance of 50m is required by German law. When this distance is now reduced to 10m or less in CACC, only a third of a second reaction time is available. During this time, a message send from the leading vehicle, needs to be processed by both vehicles, the leading and the ego vehicle. E.g. by encrypting, signing, verifying, ...
Your task is to look at the time each task will take (theoretically and if possible practically) and evaluate if it is enough time to execute these steps with proper security in mind.

Michael Wolf

✘ Multi Agent Based Simulations – English only

PaySim, a Mobile Money Payment Simulator simulates money transactions between users based on Multi Agent Based Simulation (MABS). It also generates data that can be used to test algorithms which should detect suspicious activities or fraud. This generated data is based on real financial data, which cannot be published for security reasons. In order to use or train the detection algorithms on real data, the synthetic information should be as similar as possible to the real one but not exactly the same.
Your job is to present an overview of MABS, what they are and how they are used.

Michael Wolf

✘ Secure Multi-Party Computation – English only

The goal of Secure Multi-Party Computation (MPC) is to enable parties to work together without ever knowing one another's confidential information. It plays an important role in solving security and privacy issues and there are many examples of where it can be helpful.
The aim of this seminar is to investigate MPC with respect to both theoretical and practical aspects and look at uses in real world applications.

Migena Ymeraj

✘ Distributed Machine Learning – English only

Due to the poor scalability and efficiency of learning algorithms, Machine Learning cannot handle large-scale data. This issue gave rise to Distributed Machine Learning. Even though it is a promising line of research, it still faces a lot of challenges.
The goal of this seminar is to discover the importance of Distributed Machine Learning, while comparing it with traditional Machine Learning environments and investigating its challenges.

Migena Ymeraj

✘ Protocol Identification – English only

Network management and security require knowledge about the communication in transit to make meaningful decisions. An example for this is a network intrusion detection system that needs to decide about benign and malicious network traffic based only on the observable protocol packets. Proprietary encapsulation, encryption, missing
protocol specification, and privacy concerns often do not allow for deep packet inspection and therefore prevent recognition of the full variance of observed protocols. Protocol Identification (PI) is an approach to be able to identify the type of protocol in network communications based on alternative methods, such as statistical analyses.
The goal of this seminar topic is to determine and discuss common methods in PI and the quality of results they currently can provide.

Stephan Kleber (Daimler TSS)

✘ Differential Cryptanalysis – English only

Modern symmetric encryption algorithms such as AES use alternating permutations and substitutions in order to achieve the notions of confusion and diffusion. The  substitutions - essentially a lookup table - are the only nonlinear operations. In differential cryptanalysis these substitution components, or s-boxes, are analyzed by tracking the operations of the encryption on a difference of input data. In an ideal cipher, given a fixed input difference, any possible output difference has the same probability of  occurrence. However that is mathematically impossible. If the deviations from the ideal probability are too large, the encryption key can be recovered using a chosen plaintext  attack.
In the seminar, the basics of differential cryptanalysis of a single round cipher should be explained, together with the necessary background. Interested students may  explain how to extend the attack to multiple rounds.

Henning Kopp (Schutzwerk GmbH)

✘ Trusting Trust Revisited: Preventing Software Supply Chain Attacks Using Modern Methods – English only

Ken Thompson's famous 1984 Turing Award Lecture on trusting your compiler is still relevant today, perhaps more than ever. XcodeGhost showed that these attacks are not just theory, and the Snowden documents tell us that not only criminals are considering these attacks.
Diverse Double Compilation and Reproducible Builds have been proposed to detect such attacks, and various projects have started performing reproducibility testing. For example, Debian has now reached 94.5 % reproducibility in its package archive for the bullseye version. However, trusting a central party to certify that software is reproducible may not be sufficient in defense against state actors. Mechanisms similar to certificate transparency for TLS certificates, such as CHAINIAC, have been proposed to achieve the same transparency goals for software udpates.
What current state of the art mechanisms exist to defend against software supply chain attacks? Where do those mechanisms have gaps and/or require significant effort by trusted notaries or volunteers? What practical problems prevent us from building everything reproducible?

Clemens Lang (BMW Car IT)

✘ An Introduction to Elliptic Curve Cryptography and ECDSA – English only

Elliptic Curve Cryptography (ECC) has become a prominent alternative to classical asymmetric crypto systems. Due to the mathematical properties of elliptic curves, key lengths can typically be shorter at the same security level compared to, e.g., RSA. This makes ECC particularly interesting for low-power devices in IoT applications or can reduce network load.
Your work should give an overview of the basic mathematical principles behind elliptic curve cryptography and show similarities to already existing cryptography systems (e.g., DSA and ECDSA). Moreover, the past decade has shown that ECDSA is a relatively fragile crypto algorithm. Although ECDSA is not broken if used correctly, there are many things that can go wrong around parameter selection. Your work should provide an overview of the known problems, an outlook towards the recent LadderLeak problem and what the EdDSA algorithm does differently in order to reduce this problematic aspect of EC-cryptography.

✘ Zero Trust Networks – English only

Traditional security infrastructures rely heavily on a perimeter based security model, which creates a perimeter between an untrusted zone (usually the internet) and a trusted zone (usually the internal network). However, this security model has a few disadvantages, such as the ability to deal with insider attacks, and attackers that have breached the perimeter and have gained access to the trusted network zone. An alternative concept are zero trust networks, which treat the entire network as untrusted.
The goal of this seminar is to give an overview of zero trust networks and discuss its strengths and weaknesses, as well as existing concepts and implementations.

Dominik Lang (ditis)

✘ Vehicular Intrusion Detection – What can we learn from Network IDS? – English only

Intrusion Detection Systems (IDS) in Networks have a long tradition and are an essential tool to detect and, in many cases, also defend against attacks. With increasingly complex intra-vehicular networks, IDS are also becoming more and more common within vehicles. However, the characteristics of these networks vary considerably and the transferability of experience is accordingly limited.
In this seminar, we want to look into what we can learn from other Intrusion Detection Systems and what can be transferred to the automotive realm.

Thomas Lukaseder (Escrypt)

Sorted by Topics

AI / ML
- Distributed Machine Learning (Migena Ymeraj)
- OpenWorm (Gerhard Habiger)
- Swarm Intelligence (Matthias Matousek)

Automotive
- Time constraints of Security in CACC (Michael Wolf)
- Vehicular Intrusion Detection – What can we learn from Network IDS? (Thomas Lukaseder)

Privacy and Computer Science
- Pattern-Selection Methods (Ala'a Al-Momani)
- State of the Art of Web Application Security (Echo Meißner)
- Quaternions and their Application in Computer Science (David Mödinger)
- Trusting Trust Revisited: Preventing Software Supply Chain Attacks Using Modern Methods (Clemens Lang BMW)

Cryptography
- Trusted Execution Environments (Echo Meißner)
- Secure Multi-Party Computation (Migena Ymeraj)
- An Introduction to Elliptic Curve Cryptography and ECDSA (Martin Lang BMW)
- Differential Cryptanalysis (Henning Kopp Schutzwerk)

Distributed Systems
- SimGrid (Eugen Frasch)
- Real-Time in Distributed Systems (Eugen Frasch)
- Surveying Peripheral Sensors in Context of Mobile Crowdsensing (Muntazir Mehdi)

Network Security
- Protocol Identification (Stephan Kleber)
- Network Security Breaches (Leonard Bradatsch)
- First Packet Authentication / Firewall Poking / Single Packet Authentication (Leonard Bradatsch)
- Zero Trust Networks (Dominik Lang)

• Your own topic – English only

You have the possibility until the beginning of the semester to come up with your own topic and find a supervisor who is willing to mentor more students.

✘ Location Privacy – English only

Location-based services (LBSs) have become an essential part of our daily lives. In such services, users offer their (precise) locations to service providers in return of benefiting from the service. However, offering location data to service providers put users' privacy at huge risk. Often these locations are associated with points of interest (POIs) of the users. Therefore, service providers are able to infer users' private behavior by knowing these POIs  with a relatively high degree of certainty. For this reason, the adoption and deployment of location privacy protection mechanisms (LPPMs) are essential to protect users' privacy. In this seminar, you will investigate and discuss the existing LPPMs as well as the privacy metrics that reflect how much privacy a user gains when applying a protection mechanism.

✘ Privacy in Ride Hailing Services – English only

Online taxi services, which are also known as ride-hailing services (RHSs), are becoming more and more popular. People rely on such services on daily basis. Potentially, some of the origins or destinations of such trips may be sensitive and reveal additional information about the user of a RHS including their behavior. Thus, RHSs in their current setting pose some serious privacy risks to the users. Recently, there has been many proposals for privacy-enhanced ride hailing systems. In this seminar, you will investigate and discuss such privacy-enhanced systems while addressing the way they achieve the privacy enhanced features and the utility loss of each.

✘ Privacy Patterns Landscape – English only

Privacy engineering has gained a lot of attention recently. Many methodologies and tools have been proposed to assist practitioners coming up with privacy-enhanced systems. Privacy patterns are considered one of the backbones to introduce such privacy-enhanced systems. In this seminar, you will investigate current privacy patterns and analyse them from architectural, design, and system perspective in a similar way security patterns were analyzed.
According to the analysis you carry on, you are expected to classify the privacy patterns into, among others, application architecture patterns, application design patterns, and system-wide architecture patterns, providing a holistic landscape of the current status of privacy patterns.

✘ Programming Models for the Internet of Things  – English only

IoT, the infamous Internet of Things, provides and interesting example of a heterogenous distributed network: Many different nodes with very different capabilities and properties, act together as one application. From powerful cloud instances over edge gateways to sensors, developers desire a well integrated programming environment. There are some noteable attempts to provide a programming model for this case. The goal of this seminar is to probide an overview over common approaches or concrete programming models used in practice or proposed by academia.

David Mödinger

✘ State of the Art of Web Application Security – English only

The field of web applications is constantly and rapidly evolving, but so are attacks targeting them. For this reason the World Wide Web Consortium (W3C) assembled a working group to develop technical and policy mechanisms to improve the security for applications on the Web. In recent years, this Web Application Security Group proposed various drafts for mechanisms of which some have been refined into W3C recommendations and are now implemented in all major browsers  (such as CSP and SRI). This seminar should give an overview of and discuss these recommendations and their practical implications for current web applications.

✘ Trusted Execution Environments – English only

Trust management is a central aspect of computer security. For instance, an operating system uses sandboxes to protect itself and other applications from viruses and malicious software, and cryptography is used to protect data in transit and at rest. With the advent of cloud computing, even the hardware that executes a particular software is not always considered trustworthy. Trusted Execution Environments (TEEs) try to relieve of the need to fully trust the hardware, by adding a secure area to the CPU that can guarantee code/data confidentiality and integrity through cryptographic means. Hence, protecting an application from untrusted hardware, software, and even privileged attackers (i.e., the operating system). Several TEE implementations can already be used today, such as Intel SGX and ARM TrustZone. While the former proprietary implementations often expect trust in the vector, open-source alternatives that address this weak point are already in development. 
In this seminar, you will investigate TEEs, highlight use-cases for this technology, and compare prominent representatives for their features and shortcomings.

✘ Resource Scheduling in Cloud Computing – English only

With increased popularity of Cloud Computing the approach of treating multiple nodes as one big resource unit came up. This allows to run multiple different applications on one cluster at the same time. The biggest challenge is to schedule the processes of the applications without overstress the cluster or slowing down one of the applications.
The goal of this seminar is to look at the different scheduling approaches and their use cases in cloud computing.

Eugen Frasch

✘ Securing Smartphones – English only

In this day and age, almost everyone owns a smartphone and takes it with them wherever they go. These devices contain a lot of personal data; thus, securing these devices is very important.

The goal of this seminar is to give an overview of the security architectures and mechanisms implemented in modern smartphones (for example based on iOS and/or Android) and to research solutions and proposals from academic literature.

✘ Range Proofs – English only

Confidential transactions in crypto currencies require range proofs to detect integer overflows. Any output amount of a transaction has to be a positive integer. As storage is valuable on block-chains, the goal is to reduce the size as much as possible. Recent advances in bulletproofs reduce the size significantly.
The paper should compare the different existing methods and point out how the improvements are achieved.

✘ An Introduction to Reinforcement Learning – English only

Reinforcement Learning (RL) encompasses a broad field of machine learning techniques aimed at enabling machines to tackle complex problems like video games, robotics or financial systems.
There are several sub-fields of RL like Deep RL or Apprenticeship Learning which use different paradigms to produce intelligent agents capable of solving difficult problems.
This seminar should give an overview over RL in general, describe its main ideas and features, and either provide a comprehensive set of examples where RL succeeded (or didn't succeed), and/or attempt to solve a a small problem to demonstrate how RL works.

Gerhard Habiger

Every (new) network protocol (e.g., TCP or NetFlow) and device (e.g., switches or routers) needs to be tested. The main task of this seminar is to outline different modern testing approaches. In what way do researches test network protocols and devices. In what network environment is the protocol/device tested? What traffic is used? How often are test runs repeated?
The seminar paper should outline modern approaches and state the pros and cons of the presented methodologies.

✘ Network Security Breaches – English only

The goal of this seminar is the outlining of popular network security breaches (2-3 examples).  Subsequently, state-of-the-art protection or detection approaches against these presented breaches should be explained.

✘ Recent Advances in Game AI – English only

The goal of this seminar is to survey recent advances in game AI research. A lot of interesting progress and discoveries have been made in the last years of machine learning and artificial intelligence research in the context of games.

✘ Cryptographic Accumulators – English only

Cryptographic Accumulators are comparable to cryptographic hashes; but instead of creating a digest of a single element, multiple values can be accumulated into a single digest. Afterwards it is possible to prove if individual elements are contained in the digest or not. This basic construct can be used to build very interesting applications, from secure signatures to electronic cash systems. The goal of this seminar is to present the idea behind cryptographic accumulators and how they can be applied to such applications.

✘ Messenger Security – English only

Messenger apps are among the most common communication forms. Almost every person with a smartphone uses one or more messengers. As messages often contain very private information, the security and privacy of such messenger services is crucial for users.
In this seminar, popular messaging apps and their protocols are examined and evaluated for their security properties, such as end-to-encryption, etc.

• Time constraints of Security in CACC – English only

One application of Corporate Adaptive Cruise Control (CACC) is platooning where vehicles drive very close after each other to reduce the air resistance and therefore reduce fuel consumption. This, however, comes with safety risks due to the reduced distance gap to the following vehicle and therefore reduced reaction time. If the vehicles drive 100 km/h (~30m/s), a safety distance of 50m is required by German law. When this distance is now reduced to 10m or less in CACC, only a third of a second reaction time is available. During this time, a message send from the leading vehicle, needs to be processed by both vehicles, the leading and the ego vehicle. E.g. by encrypting, signing, verifying, ...
Your task is to look at the time each task will take (theoretically and if possible practically) and evaluate if it is enough time to execute these steps with proper security in mind.

Michael Wolf

✘ Internet of Things: A Security Perspective – English only

IoT devices have been in the new for both, the huge incease in numbers and spread in different areas, but also for having weak security and being abused by botnets. During the development phase of IoT articles, the security perspective specifically for this domain is missing in many cases. For example, these items have lower hardware requirements than regular computers, which limits the use of hard encrpytion algorithms.
The student has to give a broad overview of IoT devices with an introduction to their basic architecture. Then, the student has to find security threats specifically for this domain, with examples. Last, the student has to propose some protection mechanisms, which should be implemented to mitigate said threats.
 

Michael Wolf

✘ Mobile Sensing and Smartphone Apps for Hearing Healthcare – English only

Mobile Sensing often focuses on the aspects of sensor data collection and analysis applied particularly for the purposes of education, diagnosis, treatment, or monitoring.
The aim of this seminar is to study current developments in mobile sensing and smartphone apps, specifically applied in the domain of hearing healthcare.

Muntazir Mehdi

• Surveying Peripheral Sensors in Context of Mobile Crowdsensing – English only

In this seminar report, the students are required to survey the current state of peripheral sensors that can be coupled with smartphones to further accurate the mobile crowdsensing applications. These peripheral sensors can be coupled with smartphones using Bluetooth technology or wifi. The students would be further required to study the current state of coupling technologies. In addition to the general perspective, the students will survey the peripheral sensors, their technology, and limitations within the context of mHealth (mobile health).

Muntazir Mehdi

✘ The Signal Messaging Protocol – English only

WhatsApp, Wire, Facebook Messenger and Signal (among others) implement end-to-end encryption using the Signal Messaging Protocol. This protocol implements several uncommon and desirable security properties, such as "future secrecy", "post-compromise security", or "message repudiation". These features are enabled by the underlying key exchange and message exchange algorithms Extended Triple Diffie-Hellman (X3DH) and the Double Ratchet Algorithm. Various security research groups (Cohn-Gordon et al. in 2019, Frosch et al. in 2016, Kobeissi et al. in 2017) have analyzed the Signal Messaging Protocol and have given the protocol design positive reviews.
Which interesting security guarantees does the Signal Protocol provide? How are those properties achieved? What weaknesses have been identified in the protocol?

Clemens Lang (BMW Car IT)

✘ The state of post-quantum RSA  – English only

The publication of Shor’s quantum computer integer factorization algorithm in 1994 is often understood as the beginning of the end of RSA. More than 25 years later, quantum computers are still far away from executing Shor’s algorithm on real world problems. Nevertheless, the recent advances in quantum computer technology indicate that there is a real threat the probably still most-widely used public-key cryptography algorithm out there. But not only quantum computers advanced but also the research on cryptography. More recent research on RSA shows that its parameters can be tuned in a way that quantum attacks are infeasible while the regular RSA operations on are still feasible on classical computers (albeit costly).
Your work should give an overview of RSA’s problems with quantum computers and what can be done about it at which cost. A basic understanding of the mathematical background of RSA is needed to write this paper but can be acquired while working on the topic.

✘ Timing Attacks - An Overview – English only

Classical cryptographic research deals with adversaries having polynomially bounded computational power. However, this attack model is not always realistic. In particular, an attacker may be able to measure the time it takes to run cryptographic algorithms. As often the running time of an algorithm depends on its input, this can be used to gather various information about the inputs to the algorithm or its internal workings. In a security context, sensitive data such as encryption keys and passwords may be recoverable by measuring the run time of programs. Examples the multiplications in the RSA algorithm, the POODLE and Lucky Thirteen Attack on TLS, as well as various forms of Cache timing attacks (PRIME+PROBE, EVICT+TIME).

In this seminar, the student should give an overview of timing attacks, thereby explaining at least one example in depth. Further, some mitigations against timing attacks should be discussed.

Henning Kopp (Schutzwerk GmbH)

✘ Security Assessment of the Open Charge Point Protocol – English only

The Open Charge Point Protocol (OCPP) specifies the communication between charge points for electric vehicles and the energy provider. It authenticates the user to authorize the payment of the consumed energy for charging the vehicle.

The protocol specification contains a number of security relevant design decisions, that are disputable. Thus, this seminar paper should highlight possible security issues in the design and discuss them. Moreover, a survey of existing works that address OCPP security should accompany the discussion. Finally, possible improvements should be proposed.

Stephan Kleber (Daimler TSS)

Sorted by Topics

Automotive
- Time constraints of Security in CACC (Michael Wolf)
- Security Assessment of the Open Charge Point Protocol (Stephan Kleber DaimlerTSS)

Cryptography
- Range Proofs (Felix Engelmann)
- Cryptographic Accumulators (Matthias Matousek)
- The state of post-quantum RSA (Martin Lang BMW)
- The Signal Messaging Protocol (Clemens Lang BMW)
- Timing Attacks - An Overview (Henning Kopp Schutzwerk)
- Trusted Execution Environments (Echo Meißner)

Distributed Systems/ IoT/ Web
- Resource Scheduling in Cloud Computing (Eugen Frasch)
- Programming Models for the Internet of Things (David Mödinger)
- State of the Art of Web Application Security (Echo Meißner)
- Internet of Things: A Security Perspective (Michael Wolf)

Mobile Health and Security
- Mobile Sensing and Smartphone Apps for Hearing Healthcare (Muntazir Mehdi)
- Surveying Peripheral Sensors in Context of Mobile Crowdsensing (Muntazir Mehdi)
- Securing Smartphones (Felix Engelmann)

Network Security
- Analysis of Modern Network Testing Approaches (Leonard Bradatsch)
- Network Security Breaches (Leonard Bradatsch)

AI / ML
- Recent Advances in Game AI  (Matthias Matousek)
- An Introduction to Reinforcement Learning (Gerhard Habiger)

Privacy
- Location Privacy (Ala'a Al-Momani)
- Privacy in Ride Hailing Services (Ala'a Al-Momani)

• Your own topic – English only

You have the possibility until the beginning of the semester to come up with your own topic and find a supervisor who is willing to mentor more students.

Every (new) network protocol (e.g., TCP or NetFlow) and device (e.g., switches or routers) needs to be tested. The main task of this seminar is to outline different modern testing approaches. In what way do researches test network protocols and devices. In what network environment is the protocol/device tested? What traffic is used? How often are test runs repeated?
The seminar paper should outline modern approaches and state the pros and cons of the presented methodologies.

✘ Attacks on Machine Learning – English only

In recent years we have seen an enormous increase in the use of Machine Learning -- especially Deep Learning. While the achievements with machine learning are certainly phenomenal, there are also issues that come with it. One issue is the susceptibility of machine learning against malicious attacks.
The goal of this seminar is to survey and compare state-of-the-art research in attacks against machine learning. 

✘ Attacks on Platoons – English only

Platooning is an application of Cooperative Intelligent Transport Systems (C-ITS). These consist of vehicles exchanging messages over VANETs in order to coordinate their actions. This dependability on the messages broadcasted wirelessly makes the vehicles susceptible to certain attacks.
In this seminar, existing attacks, proposed attacks and maybe new invented attacks against a vehicle platoon should be gathered and presented.

Michael Wolf

✘ Attribute-Based Credentials – English only

Attribute-Based Credentials (ABCs), or Anonymous Credentials, are a way to prove the possession of attributes to another party without disclosing one's identity. This seemingly simple achievement can be utilized for a huge number of privacy-preserving applications, such as privacy-preserving age checks or anonymous payments.
The goal of this seminar is to analyze and summarize a number of relevant publications about ABCs.

✘ Browser Privacy – English only

The Browser is the entry for many people to the internet. Therefore it should protect the user not only against attacks, but also the privacy of the users. Though there are some standards, and many browser are using the Blink Enginge, each browser handles the privacy of their users differently.
In this seminar, a chosen set of browsers should be compared how they protect the privacy of their users, not only by the default settings, but also by the options they give their users (configuration, modification, addons).

Michael Wolf

✘ Buffer overflows are dead - long live the buffer overflow  – English only

With the widespread introduction of the non-executable bit feature at the beginning of this century, classical buffer overflow exploits that override the stack directly with executable code were eliminated. For a while, it seemed as if this huge class common to programs written in system programming languages such as C or C++ are eliminated. However, it did not take long until Return-to-libc or more generally Return-Oriented-Programming appeared and buffer overflow exploits were back in business as the exutable code is not located on the stack anymore. In recent years, there were multiple ideas how these kinds of attacks can be mitigated as well with an acceptable performance overhead. Most prominently, there is the idea of Control Flow Integrity (CFI) and shadow stacks.
Your paper/talk should provide an overview of the recent research and development regarding these techniques.

✘ Contemporary Design of a CSPRNG Using the Example of Fortuna – English only

Cryptographically secure pseudo-random number generators (CSPRNGs) are used for generating random keys in cryptosystems such as encryption or signature algorithms. Consequently, they are a vital part of every cryptosystem, as the overall security greatly depends on the entropy of the keys. A modern CSPRNG such as Fortuna has various interesting properties beyond its output being uniformly random distributed. As an example, it is possible to recover from an attack where the full internal state of the CSPRNG is leaked.
The goal of this seminar is to outline the design space of contemporary CSRPNGs. The focus should be on an in-depth explanation of the CSPRNG Fortuna which is currently used in Linux' /dev/urandom.

Henning Kopp (Schutzwerk GmbH)

✘ Data Entanglement – English only

In Coding and Data Reliability, e.g., censorship resistance, research applies a concept called data entanglement. For this technique parts of a file are combined with parts of another file to reconstruct a lost file or to make fully removing a file from a system expensive.
The goal of this seminar is to look at one or more applied schemes for data entanglement and its uses in real world applications.

David Mödinger

✘ Differential Privacy – English only

Statistics about user data are used heavily for improving services, performing marketing studies, and many other purposes. The acquisition and processing of potentially sensitive information poses serious privacy threats to individuals participating in, e.g., surveys. Such information might include birthdate, gender, religion, and nationality, from which identifying an individual becomes a trivial task. Moreover, inferring more sensitive information about a specific individual becomes also possible. To solve issue, many service providers, e.g., Apple and Google, started using the so-called “Differential Privacy” to preserve their users’ privacy.
In this seminar, you will investigate differential privacy with respect to both theoretical and practical aspects. In addition to addressing the privacy guarantees of differential privacy, your task will then be addressing how major tech companies are implementing differential privacy nowadays.

✘ Mobile Sensing and Smartphone Apps for Hearing Healthcare – English only

Mobile Sensing often focuses on the aspects of sensor data collection and analysis applied particularly for the purposes of education, diagnosis, treatment, or monitoring.
The aim of this seminar is to study current developments in mobile sensing and smartphone apps, specifically applied in the domain of hearing healthcare.

Muntazir Mehdi

✘ Network Security Breaches – English only

The goal of this seminar is the outlining of popular network security breaches (2-3 examples).  Subsequently, state-of-the-art protection or detection approaches against these presented breaches should be explained.

✘ OAuth and OpenID (Connect) – English only

Along with the modern web and it's services come a lot of passwords and accounts. OAuth, OpenID and OpenID Connect are some of the single sign-on frameworks and protocols that strive to make authentication and authorization manageable in a secure way.
The goal of this seminar is to give an overview of these technologies, how they interact, and potential security considerations.

Dominik Lang (ditis)

✘ Overview of IoT Frameworks and Platforms – English only

The Internet of Things (IoT) hype is in full force, cheap sensors are abundantly available, and big companies as well as private consumers can build up their smart device networks with ease.
Thanks to this hype, many tools and frameworks for building and managing IoT installations exist. Some examples include software frameworks like Node-RED, Flogo or Eclipse Kura, the entire AWS, Google or Microsoft IoT Platforms, API-based platforms like Zetta, and many many more.
This seminar should give an overview over some of these solutions, describe their purposes and features, and either compare a selection of these tools or implement a minimal running IoT example using one platform of choice.

Gerhard Habiger

✘ QuisQuis – English only

QuisQuis is a new design for anonymous cryptocurrencies, based on updateable public keys. The problem of current anonymous blockchain systems, is the inability to prune old outputs, as they are all needed to verify future transactions. QuisQuis achieves the anonymity by re-randomising public keys, so that all inputs can be considered spent. The protocol employs multiple new techniques.
The paper should give an overview of the system and compare it to other anonymous cryptocurrencies.

✘ Range Proofs – English only

Confidential transactions in crypto currencies require range proofs to detect integer overflows. Any output amount of a transaction has to be a positive integer. As storage is valuable on block-chains, the goal is to reduce the size as much as possible. Recent advances in bulletproofs reduce the size significantly.
The paper should compare the different existing methods and point out how the improvements are achieved.

✘ Resource Scheduling in Cloud Computing – English only

With increased popularity of Cloud Computing the approach of treating multiple nodes as one big resource unit came up. This allows to run multiple different applications on one cluster at the same time. The biggest challenge is to schedule the processes of the applications without overstress the cluster or slowing down one of the applications.
The goal of this seminar is to look at the different scheduling approaches and their use cases in cloud computing.

Eugen Frasch

✘ Secure Multiparty Computation – English only

Secure multiparty computation (MPC) enables several parties, each with a private input, to compute a joint function without exposing their input to each other. This attracts many application scenarios, e.g. training a machine learning model on private input from several companies/hospitals, where privacy is crucial. Within the last decade, the development of MPC has been on the practical side and many efficient implementations have been proposed.
In this seminar, you will investigate those practical implementations of secure MPC while addressing the features of each including the suggested functionality, and check whether there is any real-world usage of each. Then, selected implementations are to have in-depth comparison among each other.

✘ Smart Home Security – English only

Home automation has been a common theme of science fiction for a long time but has recently become reality. Newly constructed homes often come with pre-installed smart home features based on proprietary hardware/software and the same can be retrofitted to older buildings. Additionally, a huge community of DIY enthusiasts develops open source and open hardware alternatives. However, security is often not the main priority of these products or can be difficult to implement correctly, due to hardware and power constraints. Numerous examples, such as smart door and garage locks, show that this is becoming a serious issue. 
This seminar should outline the differences in conventional IT security and IoT security and survey the security of common home automation software (e.g., home-assistant), protocols (e.g., MQTT), and devices (e.g., ESP8266).

✘ Surveying Peripheral Sensors in Context of Mobile Crowdsensing – English only

In this seminar report, the students are required to survey the current state of peripheral sensors that can be coupled with smartphones to further accurate the mobile crowdsensing applications. These peripheral sensors can be coupled with smartphones using Bluetooth technology or wifi. The students would be further required to study the current state of coupling technologies. In addition to the general perspective, the students will survey the peripheral sensors, their technology, and limitations within the context of mHealth (mobile health).

Muntazir Mehdi

✘ Time constraints of Security in CACC – English only

One application of Corporate Adaptive Cruise Control (CACC) is platooning where vehicles drive very close after each other to reduce the air resistance and therefore reduce fuel consumption. This, however, comes with safety risks due to the reduced distance gap to the following vehicle and therefore reduced reaction time. If the vehicles drive 100 km/h (~30m/s), a safety distance of 50m is required by German law. When this distance is now reduced to 10m or less in CACC, only a third of a second reaction time is available. During this time, a message send from the leading vehicle, needs to be processed by both vehicles, the leading and the ego vehicle. E.g. by encrypting, signing, verifying, ...
Your task is to look at one proposed protocol (e.g. IEEE 802.11p) and cross reference the steps required by this protocol with estimations of how long each step will probably take. If no official estimation is available, make one yourself and explain it. Optionally, you can compare this total processing time with that of other sensors, like camera with image processing.

Michael Wolf

✘ Trusted Execution Environments – English only

Trust management is a central aspect of computer security. For instance, an operating system uses sandboxes to protect itself and other applications from viruses and malicious software, and cryptography is used to protect data in transit and at rest. With the advent of cloud computing, even the hardware that executes a particular software is not always considered trustworthy. Trusted Execution Environments (TEEs) try to relieve of the need to fully trust the hardware, by adding a secure area to the CPU that can guarantee code/data confidentiality and integrity through cryptographic means. Hence, protecting an application from untrusted hardware, software, and even privileged attackers (i.e., the operating system). Several TEE implementations can already be used today, such as Intel SGX and ARM TrustZone. While the former proprietary implementations often expect trust in the vector, open-source alternatives that address this weak point are already in development. 
In this seminar, you will investigate TEEs, highlight use-cases for this technology, and compare prominent representatives for their features and shortcomings.

✘ What did we learn in almost 20 years of TLS? – English only

In 2018, the latest version of TLS - TLS 1.3 - was finally finished and published. The discussions and decisions that were made during the development phase are interesting to follow. In an attempt to design the most secure version of TLS that is nowadays possible, the IETF working group decided to deprecate and remove quite some parts of older TLS releases that have a rather troubling history. As part of this work, we want to follow the development of TLS 1.3 and try to learn from it for the design of cryptographic (network) protocols in general.
Towards this goal, we follow two questions. What do we learn about protocol design by understanding the removed parts in TLS 1.3? It defines a new key agreement protocol, removes old cryptographic algorithms and changes the defaults. What are the underlying problems that lead to these decisions? What can we learn when designing cryptographic protocols on our own? What can we learn from the new features in TLS 1.3? What is the general direction in which the community is moving? Which security aspects have grown more important nowadays? Which building blocks are considered generally better than others?

Clemens Lang (BMW Car IT)

Sorted by Topics

Automotive
- Time constraints of Security in CACC (Michael Wolf)
- Attacks on Platoons (Michael Wolf)

Cryptocurrencies
- Range Proofs (Felix Engelmann)
- QuisQuis (Felix Engelmann)

Distributed Systems
- Resource Scheduling in Cloud Computing (Eugen Frasch)

Internet of Things
- Overview of IoT Frameworks and Platforms (Gerhard Habiger)
- Smart Home Security (Echo Meißner)

Mobile Health
- Mobile Sensing and Smartphone Apps for Hearing Healthcare (Muntazir Mehdi)
- Surveying Peripheral Sensors in Context of Mobile Crowdsensing (Muntazir Mehdi)

Network Security
- Analysis of Modern Network Testing Approaches (Leonard Bradatsch)
- Network Security Breaches (Leonard Bradatsch)
- What did we learn in almost 20 years of TLS? (Clemens Lang BMW)

Penetration Testing
- Buffer overflows are dead - long live the buffer overflow (Martin Lang BMW)
- Attacks on Machine Learning (Matthias Matousek)

Privacy
- Differential Privacy (Ala'a Al-Momani)
- Secure Multiparty Computation (Ala'a Al-Momani)
- Attribute-Based Credentials (Matthias Matousek)
- OAuth and OpenID (Connect) (Dominik Lang)
- Data Entanglement (David Mödinger)

Security Mechanisms in HW and OS
- Trusted Execution Environments (Echo Meißner)
- Contemporary Design of a CSPRNG Using the Example of Fortuna (Henning Kopp)