Master's theses of year 2021
Theses and projects (PhD, MSc, BSc, Project)
-
Passive OS-Fingerprinting with Netflow Data.
3
2021.
BibTeX Entry
@misc{brie21, author = {Aldo Brießmann}, title = {{Passive} {OS-Fingerprinting} with {Netflow} {Data}}, year = {2021}, key = {brie21}, month = {3}, school = {Ludwig-Maximilians-Universität München}, supervisors = {Tobias Appel}, type = {Masterthesis}, } -
Ende-zu-Ende-Verschlüsselung unter Nutzung des Quantenschlüsselaustauschs.
12
2021.
BibTeX Entry
@misc{kast22, author = {Swantje Kastrup}, title = {{Ende-zu-Ende-Verschlüsselung} unter {Nutzung} des {Quantenschlüsselaustauschs}}, year = {2021}, key = {kast22}, month = {12}, school = {Ludwig-Maximilians-Universität München}, supervisors = {Nils gentschen Felde and Sophia Grundner-Culemann}, type = {Masterthesis}, } -
Pixel-based 2 DoF Synthesis of 360° Viewpoints with Flow-Based Interpolation.
3
2021.
PDF
Abstract
Virtual Reality technology allows users to experience virtual environments by interacting with them, and navigating within them. These environments tend to be either meticulously modeled in 3D by hand, or recorded using 360° cameras. The advantage of using 360° images is that a high level of realism is achievable with relatively little effort. However, the use of 360° images generally limits users to a single viewpoint or forces them to "jump" between different viewpoints. In order to improve the viewing experience, image-based rendering, or image-based synthesis, aims to create novel viewpoints based on captured viewpoints, in the best case enabling a user to navigate freely and naturally within a scene. There are a number of different approaches to image-based synthesis, many of which use some form of feature correspondence to extract the scene geometry from the captured images. While using the scene geometry makes it possible to synthesize novel views, it can also be problematic, since inaccurate scene geometry can lead to severe artefacts, and accurate scene geometry is difficult to obtain unless dedicated depth sensors are used. In order to reduce the number and severity of these artefacts, this thesis proposes a pixel-based 2-DoF synthesis algorithm that combines basic reprojection with flow-based interpolation. Instead of estimating scene geometry, a sphere is used as a proxy for inaccurately estimated scene geometry. To mitigate the artefacts caused by the inaccurate geometry, flow-based interpolation is used to generate viewpoints with more accurate perspectives in a method called "flow-based blending". A proof-of-concept implementation of the approach is presented and tested with a select set of parameters, using different virtual and real scenes. The synthesized images are then evaluated based on mathematical error metrics, as well as on visible artefacts. The results of the evaluation show that in the majority of cases where the basic method produces significant artefacts, the synthesis using flow-based blending improves the accuracy of the results.BibTeX Entry
@misc{klet21, author = {Rosalie Kletzander}, title = {{Pixel-based} 2 {DoF} {Synthesis} of 360° {Viewpoints} with {Flow-Based} {Interpolation}}, year = {2021}, pdf = {https://bib.nm.ifi.lmu.de/pdf/klet21.pdf}, abstract = {Virtual Reality technology allows users to experience virtual environments by interacting with them, and navigating within them. These environments tend to be either meticulously modeled in 3D by hand, or recorded using 360° cameras. The advantage of using 360° images is that a high level of realism is achievable with relatively little effort. However, the use of 360° images generally limits users to a single viewpoint or forces them to "jump" between different viewpoints. In order to improve the viewing experience, image-based rendering, or image-based synthesis, aims to create novel viewpoints based on captured viewpoints, in the best case enabling a user to navigate freely and naturally within a scene. There are a number of different approaches to image-based synthesis, many of which use some form of feature correspondence to extract the scene geometry from the captured images. While using the scene geometry makes it possible to synthesize novel views, it can also be problematic, since inaccurate scene geometry can lead to severe artefacts, and accurate scene geometry is difficult to obtain unless dedicated depth sensors are used. In order to reduce the number and severity of these artefacts, this thesis proposes a pixel-based 2-DoF synthesis algorithm that combines basic reprojection with flow-based interpolation. Instead of estimating scene geometry, a sphere is used as a proxy for inaccurately estimated scene geometry. To mitigate the artefacts caused by the inaccurate geometry, flow-based interpolation is used to generate viewpoints with more accurate perspectives in a method called "flow-based blending". A proof-of-concept implementation of the approach is presented and tested with a select set of parameters, using different virtual and real scenes. The synthesized images are then evaluated based on mathematical error metrics, as well as on visible artefacts. The results of the evaluation show that in the majority of cases where the basic method produces significant artefacts, the synthesis using flow-based blending improves the accuracy of the results.}, key = {klet21}, month = {3}, school = {Ludwig-Maximilians-Universität München}, supervisors = {Markus Wiedemann}, type = {Masterthesis}, } -
Optimal anonymization under varying degrees of public knowledge.
5
2021.
BibTeX Entry
@misc{koch21, author = {Andre Koch}, title = {{Optimal} anonymization under varying degrees of public knowledge}, year = {2021}, key = {koch21}, month = {5}, school = {Ludwig-Maximilians-Universität München}, supervisors = {Sophia Grundner-Culemann and Zinterhof}, type = {Masterthesis}, } -
Hybrid Data Race Detection using CLR Profiling and IL Instrumentation.
6
2021.
BibTeX Entry
@misc{dief21, author = {Daniel Diefenthaler}, title = {{Hybrid} {Data} {Race} {Detection} using {CLR} {Profiling} and {IL} {Instrumentation}}, year = {2021}, key = {dief21}, month = {6}, school = {Ludwig-Maximilians-Universität München}, supervisors = {Karl Fuerlinger and Felix Mößbauer (Siemens) and Dr. Christian Kern (Siemens)}, type = {Masterthesis}, } -
Communication Offload Strategies for Multithreaded Distributed Memory Applications.
6
2021.
BibTeX Entry
@misc{pera21, author = {Sebastian Peralta Friedburg}, title = {{Communication} {Offload} {Strategies} for {Multithreaded} {Distributed} {Memory} {Applications}}, year = {2021}, key = {pera21}, month = {6}, school = {Ludwig-Maximilians-Universität München}, supervisors = {Karl Fuerlinger and Roger Kowalewski and Pascal Jungblut}, type = {Masterthesis}, } -
Design and Development of a Container-based Framework for Machine Learning Inference in the Edge-to-Cloud Computing Continuum.
6
2021.
BibTeX Entry
@misc{mari21, author = {Stefan Marinov}, title = {{Design} and {Development} of a {Container-based} {Framework} for {Machine} {Learning} {Inference} in the {Edge-to-Cloud} {Computing} {Continuum}}, year = {2021}, key = {mari21}, month = {6}, school = {Ludwig-Maximilians-Universität München}, supervisors = {Andre Luckow}, type = {Masterthesis}, } -
Conception and Implementation of a Realtime Asset Database based on Netflow Data.
6
2021.
BibTeX Entry
@misc{sila21, author = {Mario Can Silaci}, title = {{Conception} and {Implementation} of a {Realtime} {Asset} {Database} based on {Netflow} {Data}}, year = {2021}, key = {sila21}, month = {6}, school = {Technische Universität München}, supervisors = {Daniel Weber and Tobias Appel and Simon Bauer}, type = {Masterthesis}, } -
Real-time Volumetric Cloud Visualization of Meteorological Simulation Data.
7
2021.
BibTeX Entry
@misc{kenk21, author = {Alexander Kenkenberg}, title = {{Real-time} {Volumetric} {Cloud} {Visualization} of {Meteorological} {Simulation} {Data}}, year = {2021}, key = {kenk21}, month = {7}, school = {Ludwig-Maximilians-Universität München}, supervisors = {Daniel Kolb}, type = {Masterthesis}, } -
A Survey of Three RISC-V Simulators for the Evaluation of Embedded Hypervisors.
9
2021.
BibTeX Entry
@misc{fran21b, author = {Caroline Frank}, title = {A {Survey} of {Three} {RISC-V} {Simulators} for the {Evaluation} of {Embedded} {Hypervisors}}, year = {2021}, key = {fran21b}, month = {9}, school = {Ludwig-Maximilians-Universität München}, supervisors = {Tobias Lindinger and Jan Schmidt}, type = {Masterthesis}, } -
Hybrides Zugriffsberechtigungskonzept für WebMaDa.
9
2021.
BibTeX Entry
@misc{edin21, author = {Wolfgang Christian Edinger}, title = {{Hybrides} {Zugriffsberechtigungskonzept} für {WebMaDa}}, year = {2021}, key = {edin21}, month = {9}, school = {Ludwig-Maximilians-Universität München}, supervisors = {Corinna Schmitt}, type = {Masterthesis}, } -
ID-based signing with Group-IKEv2 in Constrained Networks.
2
2021.
BibTeX Entry
@misc{hasl21, author = {Konrad Haslberger}, title = {{ID-based} signing with {Group-IKEv2} in {Constrained} {Networks}}, year = {2021}, key = {hasl21}, month = {2}, school = {Ludwig-Maximilians-Universität München}, supervisors = {Tobias Guggemos and Corinna Schmitt}, type = {Masterthesis}, } -
Optimization Approaches for Quantum Circuits using ZX-calculus.
11
2021.
PDF
Abstract
Quantum computing is a promising research field in modern computer science as quantum computers have the potential to solve some computationally hard problems much faster than classical computers. Instead of bits, quantum computers use two-state quantum mechanical systems called qubits which have some useful properties such as entanglement and superposition that have no bit equivalent. Similar to classical computers where bits are modified by Boolean gates on circuits, qubits can be modified by quantum-logical gates on quantum circuits. While in theory there are many applications for quantum computers from a variety of different research areas, in practice most them cannot yet be realized on up to date quantum computers as the size of quantum circuits is strongly limited by hardware and physical constraints. Therefore, it is crucial to optimize quantum circuits as far as possible in terms of size and complexity. This work focuses on quantum circuit optimization using the ZX-calculus, a recently developed graphical language designed to simplify reasoning about quantum systems. Quantum circuits can be optimized in an intuitive and efficient way by transforming them to equivalent ZX-diagrams and using rules of the ZX-calculus for diagram simplification. However, some rules can modify ZX-diagrams in such a way that the re-extraction of a quantum circuit is no longer possible. Moreover, rules that simplify ZX-diagrams can still increase the size of the underlying circuits. Due to these problems, most of the existing ZXcalculus based optimization approaches become inefficient with increasing quantum circuit complexity. In our work we develop different strategies to improve those approaches. In particular, we introduce heuristics to estimate the optimization benefit gained by certain ZX-rules. This allows treating ZX-diagram simplification as a classical search problem where heuristics can be applied to guide the sequence of rule applications towards minimal circuits. We implement different heuristic-based algorithms like greedy search, random search and simulated annealing in the open-source library PyZX where we test them against existing strategies on circuits of variant size. The results show that using heuristics in diagram simplification often leads to better overall optimization results, especially when optimizing large and complex quantum circuits. Our algorithms can be further improved in several aspects like runtime or consideration of hardware topology.BibTeX Entry
@misc{stau21, author = {Korbinian Staudacher}, title = {{Optimization} {Approaches} for {Quantum} {Circuits} using {ZX-calculus}}, year = {2021}, pdf = {https://bib.nm.ifi.lmu.de/pdf/stau21.pdf}, abstract = {Quantum computing is a promising research field in modern computer science as quantum computers have the potential to solve some computationally hard problems much faster than classical computers. Instead of bits, quantum computers use two-state quantum mechanical systems called qubits which have some useful properties such as entanglement and superposition that have no bit equivalent. Similar to classical computers where bits are modified by Boolean gates on circuits, qubits can be modified by quantum-logical gates on quantum circuits. While in theory there are many applications for quantum computers from a variety of different research areas, in practice most them cannot yet be realized on up to date quantum computers as the size of quantum circuits is strongly limited by hardware and physical constraints. Therefore, it is crucial to optimize quantum circuits as far as possible in terms of size and complexity. This work focuses on quantum circuit optimization using the ZX-calculus, a recently developed graphical language designed to simplify reasoning about quantum systems. Quantum circuits can be optimized in an intuitive and efficient way by transforming them to equivalent ZX-diagrams and using rules of the ZX-calculus for diagram simplification. However, some rules can modify ZX-diagrams in such a way that the re-extraction of a quantum circuit is no longer possible. Moreover, rules that simplify ZX-diagrams can still increase the size of the underlying circuits. Due to these problems, most of the existing ZXcalculus based optimization approaches become inefficient with increasing quantum circuit complexity. In our work we develop different strategies to improve those approaches. In particular, we introduce heuristics to estimate the optimization benefit gained by certain ZX-rules. This allows treating ZX-diagram simplification as a classical search problem where heuristics can be applied to guide the sequence of rule applications towards minimal circuits. We implement different heuristic-based algorithms like greedy search, random search and simulated annealing in the open-source library PyZX where we test them against existing strategies on circuits of variant size. The results show that using heuristics in diagram simplification often leads to better overall optimization results, especially when optimizing large and complex quantum circuits. Our algorithms can be further improved in several aspects like runtime or consideration of hardware topology.}, key = {stau21}, month = {11}, school = {Ludwig-Maximilians-Universität München}, supervisors = {Tobias Guggemos and Wolfgang Gehrke and Sophia Grundner-Culemann}, type = {Masterthesis}, } -
Towards Quantum-Resistant MACSec using EAP-TLS.
1
2021.
PDF
Abstract
In the last 40 years, quantum computing developed from an exclusively theoretical description of a quantum Turing machine to real-world implementations with various technologies and capabilities. Linking this development to digital computer's rapid development in the 20th century, a quantum computer with practical implications on industry and everyone's daily life seems within reachable bounds. There are many reasons to seek such a practical implementation. From algorithmic improvements to completely new technical possibilities like quantum teleportation, a quantum computer promises to solve certain tasks faster than possible with a classical digital computer. Besides the benefits such a computer could provide, it would also have a significant impact on cryptography. Modern cryptography protocols in general and especially the field of public-key cryptography relies on mathematical problems that are believed to be intractable. Famous examples of such algorithms are the RSA and the Diffie-Hellman (DH) key exchange protocol and its variant, Elliptic Curve Diffie-Hellman (ECDH). Today, nearly all encrypted messages in the modern Web are bootstrapped by either one of these protocols. A serious flaw in these cryptosystems would have a massive impact on the confidentiality of user data. This is where quantum computing comes into play. In 1999 Peter Shor published his famous algorithm, which uses a quantum computer to break both the RSA and the DH problem. Since Shor was able to show that both algorithms run in polynomial time, the foundation of modern cryptography is questioned. Luckily, even more than 20 years later, there is no implementation of a quantum computer available that could be used to break cryptographic keys of reasonable size. While this may not be true in the more distant future, an ever-growing effort was introduced to find alternative, quantum-safe cryptosystems for which no such attack exists. This work focuses on the adaptation and evaluation of such algorithms for IEEE 802.1X and IEEE 802.1AE. IEEE 802.1X focuses on the mutual authentication of clients in IEEE 802.1 Ethernet networks. For this purpose, asynchronous digital signature schemes are used that are directly affected by Shor's algorithm. Furthermore, 802.1X uses public-key cryptography and key exchanges to agree on a symmetric key between the clients and the connected network equipment. This work provides a design for a quantum-safe implementation of EAP-TLS, which can be used in IEEE 802.1X to mitigate attacks that involve a quantum computer. An extensive evaluation of the performance of different signature and key exchange algorithms is provided, and as a proof-of-concept, a real-world implementation is benchmarked with selected post-quantum and classical algorithms.BibTeX Entry
@misc{loes21, author = {Robin Lösch}, title = {{Towards} {Quantum-Resistant} {MACSec} using {EAP-TLS}}, year = {2021}, pdf = {https://bib.nm.ifi.lmu.de/pdf/loes21.pdf}, abstract = {In the last 40 years, quantum computing developed from an exclusively theoretical description of a quantum Turing machine to real-world implementations with various technologies and capabilities. Linking this development to digital computer's rapid development in the 20th century, a quantum computer with practical implications on industry and everyone's daily life seems within reachable bounds. There are many reasons to seek such a practical implementation. From algorithmic improvements to completely new technical possibilities like quantum teleportation, a quantum computer promises to solve certain tasks faster than possible with a classical digital computer. Besides the benefits such a computer could provide, it would also have a significant impact on cryptography. Modern cryptography protocols in general and especially the field of public-key cryptography relies on mathematical problems that are believed to be intractable. Famous examples of such algorithms are the RSA and the Diffie-Hellman (DH) key exchange protocol and its variant, Elliptic Curve Diffie-Hellman (ECDH). Today, nearly all encrypted messages in the modern Web are bootstrapped by either one of these protocols. A serious flaw in these cryptosystems would have a massive impact on the confidentiality of user data. This is where quantum computing comes into play. In 1999 Peter Shor published his famous algorithm, which uses a quantum computer to break both the RSA and the DH problem. Since Shor was able to show that both algorithms run in polynomial time, the foundation of modern cryptography is questioned. Luckily, even more than 20 years later, there is no implementation of a quantum computer available that could be used to break cryptographic keys of reasonable size. While this may not be true in the more distant future, an ever-growing effort was introduced to find alternative, quantum-safe cryptosystems for which no such attack exists. This work focuses on the adaptation and evaluation of such algorithms for IEEE 802.1X and IEEE 802.1AE. IEEE 802.1X focuses on the mutual authentication of clients in IEEE 802.1 Ethernet networks. For this purpose, asynchronous digital signature schemes are used that are directly affected by Shor's algorithm. Furthermore, 802.1X uses public-key cryptography and key exchanges to agree on a symmetric key between the clients and the connected network equipment. This work provides a design for a quantum-safe implementation of EAP-TLS, which can be used in IEEE 802.1X to mitigate attacks that involve a quantum computer. An extensive evaluation of the performance of different signature and key exchange algorithms is provided, and as a proof-of-concept, a real-world implementation is benchmarked with selected post-quantum and classical algorithms.}, annote = {quasimodo}, key = {loes21}, month = {1}, school = {Ludwig-Maximilians-Universität München}, supervisors = {Tobias Guggemos and Sophia Grundner-Culemann}, type = {Masterthesis}, }Additional Infos
quasimodo -
Quantum-safe Signature Scheme for IKEv2 based on Isogenies.
1
2021.
PDF
Abstract
A quantum-safe Signature Scheme for IKEv2 based on Isogenies With the field of quantum computing emerging fast and as a result of technology enhancements, it only seems to be a matter of time, until quantum computers will be able to break a large number of cryptographic algorithms that secure the internet of today. Although quantum computers are currently not powerful enough to actually execute attacks on those cryptographic algorithms, now is the right time to think about how to prevent this potential future threat. Post-quantum cryptography is the field of research which deals with that kind of threat. The IPSec protocol-suite is one of the protocols securing the internet and it is prone to be broken with the rise of sufficiently powerful quantum computers. Ongoing work exists to secure the initial key exchange of IPSec done by the IKEv2, from which all keys for further encrypted communication are derived. This prevents attackers to capture IPSec-packets in transit and use a quantum computer to decrypt those packages later. The authentication mechanism of IKEv2 nevertheless can be attacked by exploiting the incorporated non-quantum-resistant signature scheme. This enables an attacker to to claim a false identity when establishing an IPSec-connection. This work aims to provide a version of IKEv2 extended by a signature scheme that is resistant to attacks driven by quantum computers. This signature scheme relies on the problem of finding isogenies between supersingular elliptic curves and is assumed to be quantum-resistant. We give an introduction to isogeny based cryptography, including a brief explanation of the mathematical foundations it is based on. Further we discuss signature schemes based on isogenies which evolve from Unruh's transformation of non-interactive zero knowledge proofs. The work contributes a detailed analysis of requirements for PQ-Signature IKEv2, to comply with constraints that exist within the IKEv2-protocol and also to ensure interoperability and backwards compatibility. The resulting protocol specification reflects the elaborated requirements and allows two peers that try to communicate via IPSec to authenticate in a quantum-safe manner. A proof-of-concept implementation based on the aforementioned protocol is proposed. Finally an extensive evaluation of the protocol design as well as of the implementation is given, discussing the advantages and limitations of the new approach.BibTeX Entry
@misc{juer21, author = {Markus Jürgens}, title = {{Quantum-safe} {Signature} {Scheme} for {IKEv2} based on {Isogenies}}, year = {2021}, pdf = {https://bib.nm.ifi.lmu.de/pdf/juer21.pdf}, abstract = {A quantum-safe Signature Scheme for IKEv2 based on Isogenies With the field of quantum computing emerging fast and as a result of technology enhancements, it only seems to be a matter of time, until quantum computers will be able to break a large number of cryptographic algorithms that secure the internet of today. Although quantum computers are currently not powerful enough to actually execute attacks on those cryptographic algorithms, now is the right time to think about how to prevent this potential future threat. Post-quantum cryptography is the field of research which deals with that kind of threat. The IPSec protocol-suite is one of the protocols securing the internet and it is prone to be broken with the rise of sufficiently powerful quantum computers. Ongoing work exists to secure the initial key exchange of IPSec done by the IKEv2, from which all keys for further encrypted communication are derived. This prevents attackers to capture IPSec-packets in transit and use a quantum computer to decrypt those packages later. The authentication mechanism of IKEv2 nevertheless can be attacked by exploiting the incorporated non-quantum-resistant signature scheme. This enables an attacker to to claim a false identity when establishing an IPSec-connection. This work aims to provide a version of IKEv2 extended by a signature scheme that is resistant to attacks driven by quantum computers. This signature scheme relies on the problem of finding isogenies between supersingular elliptic curves and is assumed to be quantum-resistant. We give an introduction to isogeny based cryptography, including a brief explanation of the mathematical foundations it is based on. Further we discuss signature schemes based on isogenies which evolve from Unruh's transformation of non-interactive zero knowledge proofs. The work contributes a detailed analysis of requirements for PQ-Signature IKEv2, to comply with constraints that exist within the IKEv2-protocol and also to ensure interoperability and backwards compatibility. The resulting protocol specification reflects the elaborated requirements and allows two peers that try to communicate via IPSec to authenticate in a quantum-safe manner. A proof-of-concept implementation based on the aforementioned protocol is proposed. Finally an extensive evaluation of the protocol design as well as of the implementation is given, discussing the advantages and limitations of the new approach.}, annote = {quasimodo}, key = {juer21}, month = {1}, school = {Ludwig-Maximilians-Universität München}, supervisors = {Tobias Guggemos and Sophia Grundner-Culemann}, type = {Masterthesis}, }Additional Infos
quasimodo -
Experimental Examination of Distributed Conflicts in Software Defined Networks.
10
2021.
PDF
Abstract
This thesis explores policy conflicts in Software Defined Networks (SDN) between multiple applications and across multiple nodes in a network. The research questions comprise a concept for reproducible experimental infrastructures to examine conflicts and the development of new applications that enforce policies. In experiments, the new implementations are deployed in combination with existing applications in order to discover new conflict classes. For any discovered conflict classes, detection mechanisms are required. The role of presented implementations while taking into account security in SDN is evaluated. We examine existing work and the technical background to the presented approach in order to answer these topics. A software architecture for experiment automation is devised, and a set of designed and random network topologies serve as input for the experimental process. The results include new conflict classes based on the tested networks and applications, detection algorithms for the conflicts and an experimental architecture that facilitates further research. Three approaches to generate random networks that display realistic properties and three new control applications are presented. Insights on the validity of discovered distributed conflicts and the soundness of presented detection methods as well as experiment automation are gained by a second set of experiments on additional, untested networks. The results advance the understanding of distributed conflicts in SDN and promote future approaches to conflict resolution and interpretation. Diese Arbeit untersucht Richtlinienkonflikte in Software Defined Networks (SDN) zwischen mehreren Anwendungen und mehreren Knoten in einem Netz. Die Forschungsfragen umfassen ein Konzept für reproduzierbare experimentelle Infrastrukturen zur Untersuchung von Konflikten und die Entwicklung neuer Anwendungen, die Richtlinien durchsetzen. In Experimenten werden die neuen Implementierungen in Kombination mit bestehenden Anwendungen eingesetzt, um neue Konfliktklassen zu entdecken. Für alle erkannten Konfliktklassen sind Erkennungsmechanismen erforderlich. Die Rolle der vorgestellten Implementierungen unter Berücksichtigung der Sicherheit in SDN wird evaluiert. Wir untersuchen bestehende Arbeiten und die technischen Hintergründe des vorgestellten Ansatzes, um diese Themen zu beantworten. Eine Softwarearchitektur für automatisierte Experimente wird entwickelt, und eine Reihe von entworfenen und zufälligen Netzwerktopologien sind die Basis für den experimentellen Prozess. Die Ergebnisse umfassen neue Konfliktklassen, Erkennungsalgorithmen für die Konflikte und eine experimentelle Architektur, die weitere Forschung ermöglicht. Drei Ansätze zur Generierung von Zufallsnetzwerken, die realistische Eigenschaften aufweisen, und drei neue SDN-Anwendungen werden vorgestellt. Erkenntnisse über die Validität entdeckter verteilter Konflikte und die Zuverlässigkeit der vorgestellten Erkennungsalgorithmen und die Automatisierung werden durch eine zweite Reihe von Experimenten an zusätzlichen, ungetesteten Netzwerken gewonnen. Die Ergebnisse fördern das Verständnis verteilter Konflikte in SDN und zukünftige Ansätze zur Konfliktlösung und -interpretation.BibTeX Entry
@misc{reye21, author = {Nicholas Reyes}, title = {{Experimental} {Examination} of {Distributed} {Conflicts} in {Software} {Defined} {Networks}}, year = {2021}, pdf = {https://bib.nm.ifi.lmu.de/pdf/reye21.pdf}, abstract = {This thesis explores policy conflicts in Software Defined Networks (SDN) between multiple applications and across multiple nodes in a network. The research questions comprise a concept for reproducible experimental infrastructures to examine conflicts and the development of new applications that enforce policies. In experiments, the new implementations are deployed in combination with existing applications in order to discover new conflict classes. For any discovered conflict classes, detection mechanisms are required. The role of presented implementations while taking into account security in SDN is evaluated. We examine existing work and the technical background to the presented approach in order to answer these topics. A software architecture for experiment automation is devised, and a set of designed and random network topologies serve as input for the experimental process. The results include new conflict classes based on the tested networks and applications, detection algorithms for the conflicts and an experimental architecture that facilitates further research. Three approaches to generate random networks that display realistic properties and three new control applications are presented. Insights on the validity of discovered distributed conflicts and the soundness of presented detection methods as well as experiment automation are gained by a second set of experiments on additional, untested networks. The results advance the understanding of distributed conflicts in SDN and promote future approaches to conflict resolution and interpretation. Diese Arbeit untersucht Richtlinienkonflikte in Software Defined Networks (SDN) zwischen mehreren Anwendungen und mehreren Knoten in einem Netz. Die Forschungsfragen umfassen ein Konzept für reproduzierbare experimentelle Infrastrukturen zur Untersuchung von Konflikten und die Entwicklung neuer Anwendungen, die Richtlinien durchsetzen. In Experimenten werden die neuen Implementierungen in Kombination mit bestehenden Anwendungen eingesetzt, um neue Konfliktklassen zu entdecken. Für alle erkannten Konfliktklassen sind Erkennungsmechanismen erforderlich. Die Rolle der vorgestellten Implementierungen unter Berücksichtigung der Sicherheit in SDN wird evaluiert. Wir untersuchen bestehende Arbeiten und die technischen Hintergründe des vorgestellten Ansatzes, um diese Themen zu beantworten. Eine Softwarearchitektur für automatisierte Experimente wird entwickelt, und eine Reihe von entworfenen und zufälligen Netzwerktopologien sind die Basis für den experimentellen Prozess. Die Ergebnisse umfassen neue Konfliktklassen, Erkennungsalgorithmen für die Konflikte und eine experimentelle Architektur, die weitere Forschung ermöglicht. Drei Ansätze zur Generierung von Zufallsnetzwerken, die realistische Eigenschaften aufweisen, und drei neue SDN-Anwendungen werden vorgestellt. Erkenntnisse über die Validität entdeckter verteilter Konflikte und die Zuverlässigkeit der vorgestellten Erkennungsalgorithmen und die Automatisierung werden durch eine zweite Reihe von Experimenten an zusätzlichen, ungetesteten Netzwerken gewonnen. Die Ergebnisse fördern das Verständnis verteilter Konflikte in SDN und zukünftige Ansätze zur Konfliktlösung und -interpretation.}, key = {reye21}, month = {10}, school = {Ludwig-Maximilians-Universität München}, supervisors = {Cuong Ngoc Tran and Reinhard Gloger (LRZ)}, type = {Masterthesis}, } -
Analysis of Quantum Gate Collection and Decomposition Strategies for Optimizing Large Quantum Circuits.
12
2021.
BibTeX Entry
@misc{murs21, author = {Thomas Murschall}, title = {{Analysis} of {Quantum} {Gate} {Collection} and {Decomposition} {Strategies} for {Optimizing} {Large} {Quantum} {Circuits}}, year = {2021}, key = {murs21}, month = {12}, school = {Ludwig-Maximilians-Universität München}, supervisors = {Tobias Guggemos and Sophia Grundner-Culemann}, type = {Masterthesis}, } -
Towards Measuring Discrete Entropy under Continuous Transformations: A Case Study by Mapping into the Original Domain.
12
2021.
BibTeX Entry
@misc{groe21, author = {Bernhard Groettrup}, title = {{Towards} {Measuring} {Discrete} {Entropy} under {Continuous} {Transformations:} A {Case} {Study} by {Mapping} into the {Original} {Domain}}, year = {2021}, key = {groe21}, month = {12}, school = {Ludwig-Maximilians-Universität München}, supervisors = {Hayk Shoukourian}, type = {Masterthesis}, }
Disclaimer:
This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All person copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder.