003 Systeme
Refine
Year of publication
Document Type
- Doctoral Thesis (26)
- Report (4)
- Working Paper (2)
- Book (1)
- Conference Proceeding (1)
Has Fulltext
- yes (34)
Keywords
- Simulation (5)
- Ingenieurwissenschaften (3)
- Maschinenbau (3)
- Produktionstechnik (3)
- - (2)
- Analyse (2)
- Energy Awareness (2)
- Hybridsimulation (2)
- Intelligentes Stromnetz (2)
- Telemedizin (2)
Institute
- Technische Fakultät (15)
- Department Informatik (10)
- Department Maschinenbau (3)
- Fachbereich Wirtschaftswissenschaften (2)
- Lehrstuhl für Fertigungsautomatisierung und Produktionssystematik (FAPS) (2)
- Department Elektrotechnik-Elektronik-Informationstechnik (1)
- Lehrstuhl für Konstruktionstechnik (KTmfk) (1)
- Naturwissenschaftliche Fakultät (1)
- Philosophische Fakultät und Fachbereich Theologie (1)
- Rechts- und Wirtschaftswissenschaftliche Fakultät (1)
Die zunehmende Vernetzung von Maschinen und Anlagen, steigende Heterogenität von Kommunikationsteilnehmern und die Konvergenz von IT- und OT-Applikationen im Zuge von Industrie 4.0 stellen Planende von industriellen Kommunikationsnetzwerken vor große Herausforderungen. Bisher lassen sich industrielle Kommunikationsnetzwerke erst nach Vollendung der Engineering-Phase hinreichend validieren. Dies führt zu hohen Kosten durch nicht entdeckte Fehler in der frühen Phase des Lebenszyklus einer automatisierten Produktionsanlage. Ziel dieser Arbeit ist es, eine Validierung des geplanten Kommunikationsnetzwerks bereits in der frühen Planungs- und Design-Phase einer automatisierten Produktionsanlage zu ermöglichen. Hierfür wird das Co-Simulationsframework PLANET eingeführt, welches auf Basis von applikationsbezogenen Kommunikationsanforderungen und einer Kombination aus Materialfluss- und Kommunikationssimulation das industrielle Kommunikationsnetzwerk validiert.
Dazu wird eine Strukturierung industrieller Applikationen hinsichtlich Kommunikationsanforderungen anhand eines Kommunikationstechnologie-Portfolios vorgenommen. Durch die Modellierung von applikationsbezogenen Kommunikationsanforderungen erfolgt die Definition einer Beschreibungssprache in der Verwaltungsschale zur Verwendung in Simulationen. Des Weiteren wird auch die Gesamtarchitektur der Co-Simulation vorgestellt, welche diese Informationen nutzt und anhand einer prototypischen Implementierung die Ergebnisse validiert.
The integrated treatment of planning problems which are usually considered separately and sequentially has been studied for a long time, due to the better solutions one may find in the extended decision space of an integrated problem. This thesis considers the integrated locomotive scheduling and driver assignment problem in rail freight transport. We also consider the generalization of this problem, which we call graph 2-list-colouring with compatibility constraints. The motivation to study this problem originates from our collaboration with DB Cargo Polska within the ROMSOC Programme. The thesis consists of two parts. Part I focuses on modelling and solving the integrated locomotive scheduling and driver assignment problem in rail freight transport. After a literature review, we present a novel optimization model for the problem studied and a way to improve its formulation. Next, we introduce the decomposition-based solution approach we derive for the problem. To ensure the global feasibility of the solutions to the decomposed subproblems, we devise four classes of valid inequalities. We also develop a presolve heuristic. We then test our algorithm against two sets of instances. In general, the methods presented enabled the creation of locomotive timetables and driver assignments in less than two hours. In Part II, we study a generalization of the integrated locomotive scheduling and driver assignment problem, which we call graph 2-list-colouring with compatibility constraints. We begin by defining the problem studied and putting it in the context of other, more famous combinatorial problems. Then we present two formulations for the problem and discuss how they may be tightened. We also study a case for which we may find a complete polyhedral description. Next, we present a decomposition-based solution approach which adapts the algorithm introduced in Part I. We then test the performance of our method against a set of standard instances drawn from the literature, which were appropriately modified. Altogether, our work is a practical contribution to the solvability of the integrated locomotive scheduling and driver assignment problem. We also show how the developed method may be extended to successful use in more general graph-theoretic contexts.
The present thesis aims to get a fundamental understanding of how the molecular characteristics of a solute and solvent influence the thermophysical properties of mixtures consisting of a liquid and a dissolved gas. Due to their relevance as working fluids in chemical and energy engineering, accurate data for the viscosity, interfacial tension, and mutual diffusion coefficient are required for the efficient design and optimization of related processes. In this work, light scattering experiments, which can determine the thermophysical properties in macroscopic thermodynamic equilibrium, are combined with equilibrium molecular dynamics (EMD) simulations at temperatures up to 573 K. EMD simulations are able to predict multiple thermophysical properties by the analysis of the molecular motion in macroscopic thermodynamic equilibrium. The combination of EMD simulations with light scattering experiments allows for the validation of simulations as well as an interpretation of the results by studying the influence of molecular characteristics and the fluid structure on the thermophysical properties.
Design and Implementation of an Automated Performance Modeling Toolkit for Regular Loop Kernels
(2023)
Analytical performance modeling helps software developers understand performance behavior and, as a result, guides development towards more efficient and faster software. Resulting models are the distilled knowledge and experience from performance and architecture analysis, poured into a structured form. Their construction and application is a complex, tedious and labor-intensive task, that requires in-depth knowledge of modern architectures and comes with potential pitfalls. This research aims to lower the entry barrier of using performance models to allow large-scale performance model studies and make analytic models accessible to code generation and similar use-cases. The focus lies on–but is not strictly limited to–regular loop kernels, which are defined by long-running loop nests with data-content independent performance behavior.
Techniques and implementations for static prediction of performance model contributions have been developed. They specifically cover the throughput of instruction execution and transfer volumes in cache architectures. The developed prediction methods and used performance models support Intel, AMD and multiple ARM-based multi-core architectures, six of which are presented and validated in this work. While GPGPUs are similar, they are out of the scope of this work. Four main stand-alone software components were developed: Kerncraft, generalized Layer Conditions (gLC), pycachesim and the Open Source Architecture Code Analyzer (OSACA). In the Kerncraft tool, static high-level code analysis is used to bring together individual data transfer and instruction throughput predictions into an explanatory comprehensive runtime and performance prediction, using the Execution-Cache-Memory and Roofline models. The built-in measurement functionality supports model validation and kernel benchmarking. The generalized Layer Condition model analysis gives an analytical closed-form cache behavior model for stencil-like kernels, which is used to predict cache utilization. The pycachesim cache simulator predicts cache behavior in more complex cases and for arbitrary kernels. OSACA comes with x86 and AArch64 support and can analyze the block throughput, critical path and loop-carried dependency.
Execution modeling (OSACA), cache modeling (pycachesim and gLC) and automatic performance modeling (Kerncraft) are evaluated and discussed using extensive validated studies, going far beyond what is practical to manually model. The presented research results reveal performance effects that were not classified, modeled or even known before. This lays the groundwork for further architecture performance research and model refinements, deepening our understanding of performance behavior on modern architectures. The published open source projects are already used and integrated with other research projects and software development utilities, indicating the relevance of this work beyond the academic fields of performance modeling and architecture analysis.
For thousands of years, people have depicted the real world through geographic maps or portraits of people, landscapes, and machines for different reasons, like representing complex environments simply and understandably or depicting detailed aspects of reality. Due to the increasing digitalization and connectivity over the last years, it is possible to continuously collect more and more data describing the real world, including interactions between environments, objects, and people, and to gain more precise, comprehensive, and completely new insights. Activity trackers for the collection and analysis of health data, smart home systems for the monitoring and energy-efficient control of light, temperature, and household appliances, or the collection and analysis of production data in industrial manufacturing for the monitoring, prediction, and optimization of production processes are used. An emerging approach in this context for creating virtual data-based representations of real objects is the Digital Twin Concept, examined by the research presented in this thesis. Digital Twins are virtual models that mirror physical objects throughout their lifecycle, and real-time connections between the physical and virtual worlds allow Digital Twins to monitor and control physical objects from any location. Physical objects can be any living or non-living object, such as aircraft, manufacturing equipment, cars, buildings, wind turbines, food, or even people.
Companies see great potential in Digital Twins, and more and more organizations are using Digital Twins or plan to do so. In addition to the interest of practitioners, Digital Twins are being intensively studied as research objects. However, the discussion on Digital Twins in the scientific literature shows that there is currently no consolidated understanding of Digital Twins or the Digital Twin Concept. Nevertheless, the interest in Digital Twins has grown strongly in business and research in recent years. Therefore, specific findings and solutions from research are becoming increasingly important to enable organizations to apply Digital Twins in meaningful and efficient ways. This thesis aims to contribute to the understanding and application of Digital Twins by addressing two research areas. In the context of the first research area, a rigorous multi-level literature review was conducted to contribute to the understanding of Digital Twins. Two studies were conducted in the second research area to contribute to the application of Digital Twins.
The first research area contains a rigorous multi-level literature review consisting of three studies. First, a preliminary study was conducted to develop a search and selection strategy for scientific literature and, based on this, a review sample. In the second study, a review of reviews was conducted based on the identified literature reviews from the review sample. As a result, it could be shown that the number of literature reviews has increased significantly since 2018 and that the research is distributed across various outlets and disciplines, with a clear focus on the manufacturing domain. Furthermore, a content analysis was conducted that revealed four research areas containing different aspects of the research on Digital Twins, which are examined more closely. Since the literature reviews did not cover a large part of the publications in the review sample, a third study was conducted. In the third study of this research area, a scoping review, an analysis of the Digital Twin research community, and a citation analysis were conducted to analyze the primary studies of the review sample. The literature analysis showed that the research could be well structured according to seven different research topics, which are examined and characterized in detail.
Two studies were conducted on applying Digital Twins in the second research area. In the first study, an analysis of Digital Twin application cases was conducted to provide a comprehensive and structured overview. Based on a set of application cases retrieved from previous scientific literature, the study identified six dimensions with a total of 24 characteristics to classify Digital Twin application cases. Furthermore, a comprehensive overview of selected Digital Twin application cases is presented. In the second study, an IT system based on the Digital Twin Concept was implemented to gain detailed insights into the design and implementation of Digital Twin systems in an enterprise setting. Therefore a Digital Twin introductory project of an industry partner was accompanied over two years according to scientific principles. Within this project, an IT system based on the Digital Twin Concept was implemented, and its validity and utility were demonstrated and evaluated. Five application cases were realized and applied to production settings for the demonstration and evaluation. The study presents a detailed description of the artifact's conceptual architecture and technical implementation.
In summary, this thesis provides valuable insights into two specific research areas in the field of Digital Twins. It provides a comprehensive picture of the research landscape and presents detailed insights into the application of Digital Twins. Therefore this thesis provides a solid basis for further research on the possibilities and challenges of converging real and virtual worlds.
Invasive Computing
(2022)
Invasive computing is a paradigm for designing and programming future parallel computing systems. For systems with 1,000 or more cores on a chip, resource-aware programming is of utmost importance to obtain high utilisation as well as computational, energy and power efficiency. Invasive computing provides a programmer explicit handles to specify and argue about resource requirements desired or required in different phases of execution: In an invade phase, an application asks the operating system to allocate a set of processor, memory and communication resources to be claimed. In a subsequent infect phase, the parallel workload is spread and executed on the obtained claim of resources. Finally, if the degree of parallelism should be lower again, a retreat operation frees the claim again, and the application resumes a sequential execution. To support this idea of self-adaptive and resource-aware programming, not only new programming concepts, languages, compilers, and operating systems were needed to be developed, but also revolutionary architectural changes in the design of MPSoCs (multiprocessor systems-on-a-chip) to efficiently support invasion, infection, and retreat operations. This book gives a comprehensive overview of all aspects of invasive computing.
Die Verschmelzung der physikalischen und virtuellen Welt prägt die heutigen Lebens- und Arbeitswelten. Neue technische Lösungen meist in Form smarter Produkt-Service-Systeme - sollen individuelle Bedürfnisse und Bedarfe adressieren. Adoptionsprognosen in diesem Kontext erfordern zuverlässige Daten, Informationen und geeignete Analyseinstrumente für die Vorhersage individuellen Verhaltens. Speziell in Vorimplementierungsphasen sind Adoptionsprognosen ein wertvolles Instrument zur Vorhersage künftiger Ereignisse oder Situationen.
Diese Arbeit bietet einen interdisziplinären Einblick in Adoptions- und Akzeptanzprozesse unter Zugrundelegung emotionaler Antizipationen im Sinne der Simulation der zukünftigen Realität menschlicher Entscheidungsprozesse. Die auf Basis eines mehrphasigen Mixed-Methods-Ansatzes gehobenen Erkenntnisse der Arbeit zeigen, dass das thesenbasiert konzipierte und empirisch validierte Affective Adoption Forecasting Framework A²FORE den aktuellen Herausforderungen der Abschätzung individueller Adoptionsentscheidungen vor Markteinführung neuartiger smarter Produkt-Service-Systeme mit besonderem Fokus auf vulnerablen Konsumenten im Gesundheitswesen begegnen kann. A²FORE spiegelt die Nutzbarmachung antizipatorischer Emotionen als Prognoseelement für den Erfolg smarter Produkt-Service-Systeme wider. Dies entspricht der praxisbezogenen Forderung der Entwicklung eines handhabbaren Modells zur Adoptionsprognose auf Basis eines User Acceptance Testings im Rahmen von Vorimplementierungsphasen.
Komplexität ist eine der großen Herausforderungen unserer Zeit. Daher müssen Ingenieure besonders in der Entwicklung die stetig steigende Komplexität in diversen Domänen beherrschen, um den Produkt- und damit Unternehmenserfolg zu realisieren. Für die zweckmäßige Modellierung der vielen verschiedenen Elemente und Relationen komplexer Systeme sind domänenübergreifende, integrative und damit holistische Modellierungsansätze erforderlich, welche jeweils kontextbezogen nützliche, mathematische Analyse- sowie Visualisierungsmöglichkeiten erlauben um Entwicklern eine multikriterielle Bewertung bestimmter Handlungsalternativen zu ermöglichen. Derzeit bestehende Ansätze eignen sich hierfür jedoch unter anderem aufgrund methodischer Defizite, ihres hohen Modellierungsaufwands, eingeschränkter Analyse- und Visualisierungsmöglichkeiten oder ihrer geringen Unterstützung bei der multikriteriellen Bewertung nur eingeschränkt.
Daher wird in der vorliegenden Arbeit ein Konzept vorgestellt, mit dem Entwickler beim holistischen Komplexitätsmanagement in der Produktentwicklung unterstützt werden, indem kontextbezogen jeweils geeignete Methoden für die integrierte Modellierung, Analyse, Visualisierung und Bewertung komplexer Systeme für eine zielgerichtete Entscheidungsfindung zur Verfügung gestellt werden. Zur effektiven und effizienten Anwendung der Methodik wird ein IT-Werkzeug präsentiert, das Entwickler bei deren Aufgaben im Rahmen des holistischen Komplexitätsmanagements unterstützt.
In this thesis, we investigate Trusted Execution Environments (TEEs), isolated execution contexts that provide integrity and confidentiality guarantees for security-critical code and data. In particular, we study TEEs based on ARM TrustZone (TZ) which is a Central Processing Unit (CPU) extension for ARM chipsets. The hardware-enforced isolation offered by TZ is deployed on almost all modern mobile devices to support a wide range of security-critical use cases. These use cases include user authentication, mobile banking, and Digital Rights Management (DRM).
The focus of our work is the analysis of software-based design and implementation flaws in TZ TEEs. First, we review the previously unexplored TEE implementation of a market-leading mobile device manufacturer and unveil several critical design flaws. These flaws make us realize that TZ is not an all-in-one-solution, but rather a construction kit for TEEs. An end product’s capabilities to guarantee integrity and confidentiality, highly depend on the system designer’s choices of hardware and software components. We demonstrate that a well-designed system cannot be compensated by security-by-obscurity practices, especially concerning confidentiality guarantees.
Then, we identify a pattern of bugs affecting all commercially used TEEs. This pattern’s root cause is an Application Programming Interface (API) design that defaults to a dangerous type-confusion bug within Trusted Applications (TAs) if specific validations are accidentally missing or imprecise. Due to the severity of the resulting bugs and the wide adoption of this API, we introduce TEECheck, a static analysis framework capable of automatically detecting these type-confusion bugs in commercially-used TEEs. The large-scale study we conduct using TEECheck yields numerous affected TAs underlining the relevance of our findings.
Next, to address the problem of memory corruption bugs in TEEs in general, we propose TEEzz. TEEzz is a novel, automated fuzzing framework capable of effectively fuzzing production TAs and the first framework addressing the challenges occurring when fuzzing TEEs on real devices. Our system uses an automated, dynamic-analysis-based technique for inferring field types of messages, and their dependencies, to facilitate type-aware black-box mutational fuzzing of stateful TAs. We thoroughly evaluate TEEzz against other state-of-the-art fuzzing techniques applied to production TAs. Using TEEzz, we discovered and disclosed more than 50 unique bugs, where, at least, one of which was a previously unknown zero-day bug. The high number of triggered bugs indicates that memory corruptions within TAs are not rare, but occur rather frequently.
Further, we investigate a novel re-hosting approach for TAs to enable dynamic analyses without the limited access to the TEE internals, which usually is exclusive to the manufacturer. Our re-hosting solution, called SimTA, provides dynamic analysis capabilities to examine security-critical patches in TAs. Using SimTA, we re-discover the bug of a known CVE and unveil two previously unknown critical issues silently patched by the vendor.
Moreover, we elaborate on the severity of memory corruption bugs in TEEs by examining exploits on real devices. This perspective of an adversary allows us to conclude that commercially used TEEs are lacking behind the state-of-the-art regarding exploit mitigation techniques. As real-world exploits are difficult to reproduce but are relevant for industry, researchers, and educators, we provide a set of real-world inspired vulnerable examples and their corresponding exploits targeting a free and open-source TEE platform.
Lastly, given we have overcome all design and implementation flaws discussed before, this thesis gives an opportunistic outlook by proposing two new application scenarios based on TZ TEEs. First, we present TEEMo, a solution for the trusted monitoring of peripheral configurations. TEEMo securely reports peripheral configurations of embedded devices (i.e., optical network elements in our case) to remote log servers. Finally, we introduce TEE-BI, a TEE-based translation table introspection framework for Android devices that can reliably and stealthily extract cryptographic keys of secured communications from running applications. We examine TEE-BI in the context of remote forensic investigations and conclude that it provides higher evidential value for acquired data from suspects’ devices than other commonly used investigation practices.
This thesis investigates novel anti-forensic techniques for hiding malicious activity and proposes counter strategies for conducting robust digital analysis through virtualization technology. We begin by surveying the current landscape of memory acquisition, a technique extensively used during forensic investigations. In order to evade analysis, malware nowadays incorporates sophisticated anti-forensics, which hinder the analysis process. We present advances in anti-forensics by introducing new methods for hiding memory from analysis tools to expand existing knowledge. The final part demonstrates analysis techniques that provide resilience against anti-forensics.
First, we define a universal taxonomy of different methods for acquiring a system’s memory, many of which have proven to be vital against modern malware threats. Then, based on this taxonomy, we comprehensively survey the field of modern memory acquisition, abstracting from both Operating Systems (OSs) and specific hardware architectures. Finally, we unveil the limitations of today’s acquisition techniques and conclude that most approaches are prone to anti-forensics, enabling malware to subvert the analysis process and escape the investigation.
In the second part, we introduce new approaches that hide memory from forensic applications, preventing analysts from accessing the content of specific regions. On the one hand, we manipulate the memory management subsystem of different OSs to alter the memory view of live forensic tools. In addition, we demonstrate different strategies to detect these subversion techniques, providing a possibility to improve respective tools. With Styx, on the other hand, we present a powerful rootkit technique that leverages hardware-based virtualization to counter even robust acquisition methods. Styx subverts the highly privileged hypervisor layer to take complete control over a system without introducing detectable modifications. Furthermore, to prevent acquisition software from noticing the rootkit’s memory footprint, Styx locates in particular memory regions reserved for device mappings. As these regions are not always entirely consumed by devices, the resulting offcuts serve our rootkit as a perfect hiding spot. Furthermore, by simulating invalid address ranges which are not accessible to a processor, Styx deceives forensic tools with a tampered view on these leftovers.
Finally, we demonstrate the design of anti-forensic resilient systems which enable a forensically robust analysis through virtualization. We first present SEVGuard which protects (forensic) applications from malicious threats operating at the highest privilege levels. Based on virtualization and encryption features of modern processor architectures, SEVGuard provides a secure execution environment that enforces confidentiality and integrity of existing applications by encrypting their memory and processor state. Instead of protecting an application, StealthProbes hides the analysis component from the examined target, giving analysts the chance to inspect its functionality without risking the sample to notice the investigation. Our system stealthily instruments an application’s memory and hides these modifications, leveraging the latest virtualization features and exploiting cache incoherencies that arise from memory address translations. Furthermore, StealthProbes integrates a transparent function-level tracer for enabling deep insight into an application’s runtime behavior. As a result, even programs that thwart the analysis by enforcing code integrity are stealthily dissectable. For achieving a forensically sound investigation, the actual deployment or execution of a forensic method must not alter the state of an analyzed system. With HyperLeech, we present a minimally invasive approach which uses Direct Memory Access (DMA) to stealthily deploy a forensic hypervisor through external Peripheral Component Interconnect Express (PCIe) hardware. The hypervisor transparently virtualizes the running target system, serving analysts as a stealthy and privileged execution layer for all kinds of forensic tasks. Without causing a notable impact on the target’s state, HyperLeech enables forensic methods to execute without the risk of destroying evidence or alerting malware.
Writing well-maintainable parallel programs
that efficiently utilize many processor cores is still a
significant challenge. Threads are hard to use, and so are
event-based schemes. Furthermore, threads are affected by
the blocking anomaly, that is, the loss of parallelism when
threads execute a blocking system call—often resulting
in low core utilization and unnecessarily high response
times. This paper introduces pseudo-blocking system calls
built upon modern asynchronous queue-based system-call
techniques (like Linux’s io_uring) circumventing the
blocking anomaly. They are similar to Go’s programming
model, where one develops against a blocking interface to
keep the code structure clean. However, instead of using
synchronous non-blocking system calls as the underlying
technique, our approach internally uses an asynchronous
queue-based interface. We further present a novel architec-
ture for concurrency platforms, like Cilk and Go, enabling
low latencies and high throughput via pseudo-blocking
system calls. Finally, we discuss future OS enhancements
that would improve our proposed architecture. We imple-
mented and evaluated a concurrency platform based on
the concept of pseudo-blocking system calls. Our platform
can outperform state-of-the-art systems like Go by 1.17 ×
in a file-content search benchmark. It is able to increase
the throughput of a echo-server benchmark by 4 % when
compared to Go, and by 17.8 % when compared to Rust’s
Tokio while improving the tail latency.
Organizations deploy gamification in CSCW systems to enhance motivation and behavioral outcomes of users. However, gamification approaches often cause competition between users, which might be inappropriate for working environments that seek cooperation. Drawing on the social interdependence theory, this paper provides a classification for gamification features and insights about the design of cooperative gamification. Using the example of an innova-tion community of a German engineering company, we present the design of a cooperative gamification approach and results from a first experimental evaluation. The findings indicate that the developed gamification approach has positive effects on perceived enjoyment and the intention towards knowledge sharing in the considered innovation community. Besides our conceptual contribu-tion, our findings suggest that cooperative gamification may be beneficial for cooperative working environments and represents a promising field for future research.
Wireless power transfer (WPT) technology is envisioned to revolutionize the powering of wireless devices making them truly wireless. Conventional wired-recharging or battery replacement of wireless devices are not only costly and inconvenient, but can be dangerous in hazardous environments or even impossible for powering biomedical implants. This dissertation focuses on far-field WPT using radio frequency (RF) signals. The dual capability of RF signals to transfer both information and power motivates the integration of WPT and communication systems.
The main focus of this dissertation is on the analysis and design of communication systems with WPT. These systems can be classified into two broad categories, namely simultaneous wireless information and power transfer (SWIPT) systems and wireless powered communication (WPC) systems. In particular, in SWIPT systems, the same RF signal is used to transfer both information and power, whereas in WPC systems, an energy harvesting (EH) node harvests energy from an RF signal over a WPT downlink channel to facilitate its communication with a desired receiver over a wireless information transfer uplink channel. In the following, we summarize the main contributions of this dissertation.
Multi-User Scheduling in SWIPT: In SWIPT systems, practical circuit limitations prevent a device from fully utilizing the same RF signal for both information decoding and EH. Hence, information decoding and EH need to be separated in time, power, or space. This creates a tradeoff between the rate of information transfer to the device and the amount of energy that can be harvested by the device, the so-called rate-energy tradeoff. In a multi-user SWIPT system, the users share the downlink channel to receive both information and energy and each user is interested in optimizing its rate-energy tradeoff. In this dissertation, we propose a novel framework for the online scheduling of multiple users in a downlink SWIPT system, where the users’ rate-energy tradeoff can be controlled. In particular, we aim to answer the following fundamental question: “How should a SWIPT transmitter schedule multiple heterogeneous users in a common downlink channel to achieve desired information transfer rates and harvested powers at the users while ensuring fairness in the resource allocation among the users?”
Energy Buffer-Aided WPC systems: Most works in the WPC literature assume that EH nodes consume the harvested energy for wireless information transfer in a slot by slot manner. However, the system performance can be improved if EH nodes store the harvested energy in energy buffers (e.g. batteries) and optimize the use of the stored energy for future communication tasks. Hence, in this dissertation, we aim to answer the following question: “In WPC systems, how can an EH node with an energy buffer optimize its uplink transmit power such that the communication outages with its desired receiver are minimized?” In particular, we propose two simple online uplink transmission policies that do not require knowledge of the amount of harvested energy in the downlink nor of the uplink channel state information. Accordingly, we analyze the probability of communication outages assuming Nakagami-m fading channels and show that the proposed transmission policies provide a near-optimal outage probability performance.
Nonlinear Circuit-based Saturation-Aware EH Model: Most works in the WPT literature assume a linear EH model for the EH receiver. In practice, however, RF EH circuits exhibit a nonlinear input-output characteristic. In particular, the amount of harvested power at the EH receiver depends on both the power and the waveform (shape) of the transmit RF signal. This nonlinearity motivates the design of transmit RF signals that enable efficient WPT. To this end, we develop a novel nonlinear EH model by analyzing a single-diode half-wave rectifier circuit, which is commonly used for WPT applications. The proposed EH model captures the saturation of the output power for high input RF powers. Accordingly, we obtain a closed-form expression for the harvested power whose accuracy is confirmed by circuit simulations.
Capacity of SWIPT Systems for Nonlinear EH Model: The proposed nonlinear EH model is applied in a SWIPT system with the objective of optimizing the transmit RF waveform for maximization of the system rate-energy region. In particular, we aim to answer the following question: “What is the optimal distribution of the transmit RF signal that maximizes the rate of information transfer to an information decoding receiver while simultaneously satisfying the EH requirements of multiple nonlinear EH receivers?”. Specifically, we study the conditional capacity of a SWIPT system subject to constraints on the average and peak powers at the transmitter and nonlinear EH constraints at the EH receivers. We prove that the optimal distribution of the transmit signal amplitude that maximizes the rate-energy tradeoff is unique, discrete, and finite. In addition, we propose a suboptimal input distribution whose performance closely approaches that of the optimal one.
Various problems in practice are mathematically described by a regression model in which the linear relation between possibly noisy outputs and their corresponding inputs is to be learned. An effective approach for linear regression is the regularized form of the least-squares method commonly abbreviated by RLS. This scheme specifies the linear relation, such that a penalized version of error given by the linear model is minimized. The penalty describes information known in priori about the model.
Due to the growing interest in high-dimensional applications, recent studies often desire to characterize RLS in the large-system limit. Nevertheless, for various forms of RLS, basic analytical tools fail to address this task. For these forms, numerical simulations further deal with high computational complexity. Hence, assessing the performance by numerical approaches is also intractable. The main objective of this dissertation is to address this issue by using analytical tools from statistical mechanics. To this end, RLS solution of a linear regression model is interpreted as the ground state of a spin glass whose macroscopic parameters are derived via the replica method. Following the relation between the spin glass and its corresponding regression problem, the asymptotic distortion achieved by RLS is analytically derived for an arbitrary distortion function. The asymptotic characterization of RLS leads to a general form of the decoupling principle. This result states that the regression outputs given by RLS in a high-dimensional linear model are statistically described via a decoupled scalar setting. When the corresponding spin glass exhibits replica symmetry, this decoupled setting reduces to a scalar system with independent additive Gaussian noise and recovers earlier results in the literature. For cases in which replica symmetry breaks, the Gaussianity and independency of decoupled noise are violated. Using the derivations via the replica method, the decoupled noise term is further characterized for cases with broken replica symmetry with an arbitrary number of breaking steps.
Linear regression has a diverse scope of applications from statistics and communications to information theory, signal processing and machine learning. To enlighten applications of this study, two examples of linear regression via RLS are investigated in this dissertation. The first example considers the problem of sparse recovery. Using the asymptotic results, the large-system performance of various recovery algorithms is studied and compared to the results in literature. For algorithms with tractable computational complexity, investigations confirm validity of the derivations. The results further characterize those recovery algorithms whose performance has remained unknown due to their intractable complexity. In the second example, multiple-input multiple-output (MIMO) precoding is investigated. In this respect, the problem of MIMO precoding is first interpreted as linear regression and an RLS-based scheme is proposed. The large-system performance of this scheme is then investigated by means of the asymptotic results and compared to the benchmark. Investigations depict close agreement of asymptotic characterization with finite-dimensional simulations and demonstrate significant performance enhancements in some particular cases.
Die heutige Zeit unterliegt mehr denn je einem strukturellen Wandel, im Zuge dessen Wissen zunehmend an Bedeutung gewinnt. Während in den 1990er Jahren noch Produktionsfaktoren wie Arbeit, Standort und vor allem Kapital von Relevanz sind, ist das Wissen eines Unternehmens heute ein wettbewerbsentscheidender strategischer Faktor. Durch die Konzentration der unternehmerischen Aufgaben auf diese entscheidende Ressource muss sichergestellt werden, dass das für den Geschäftserfolg notwendige Wissen marktorientiert aufgebaut, abgesichert und optimal genutzt wird. Bestehende Ansätze des Wissens¬managements vernachlässigen jedoch die Herausforderungen der produzierenden Industrie, berücksichtigen die menschliche Kommunikation nicht ausreichend und bilden Wissen nicht ganzheitlich ab.
Ziel der Arbeit ist es, die spezifischen Anforderungen der produzierenden Industrie für den Wandel hin zu Wissensunternehmen zu analysieren und in einem ganzheitlichen Wissens-managementkonzept abzubilden. Die systematische Analyse der Einflüsse auf das Wissensmanagement, wie Hindernisse, Erfolgsfaktoren, Methoden und Systeme, schafft ein Verständnis darüber, wie Wissen in der produzierenden Industrie optimal genutzt werden kann. Durch die Abbildung von Wissen, dessen Beziehungen sowie der darin enthaltenen Logik in einem ontologiebasiertes Wissensmanagement-Framework werden Unternehmen in der formalen Erfassung der erfolgskritischen Ressource unterstützt. Die adäquate Integration der Technologie der künstlichen Intelligenz und des maschinellen Lernens ermöglicht es Mitarbeitern zudem, mit Ontologien als Wissensrepräsentationsform aufwandsarm zu interagieren.
Darüber hinaus zeigt die Arbeit auf, wie das ontologiebasiertes Wissensmanagement-Framework in einer realen Fertigung angewendet und dadurch eine optimale Ausgestaltung des gegenwärtigen und zukünftigen Ressourceneinsatzes gewährleistet werden kann.
In view of increasing the efficiency of development processes in the field of software and systems engineering, model-driven techniques are coming into ever more widespread use. On the one hand, the abstract graphic models help to master the complexity of the system under development. On the other hand, formal models serve as the source for analysis and automated synthesis of a system. Thereby, model-based transformation engines and generators allow the specifications to be defined platform- and target-language-independently and to be automatically mapped to the desired target platform.
Test-driven development is a promising approach in the field of agile software development. In this method, the development process is based on relatively short iteration cycles with preceding test specifications. Due to the fact that the actual implementation is consistently carried out in compliance with the previously written tests, this method leads to a higher test coverage at early development stages and thus contributes to the overall quality assurance of the resulting system.
This thesis introduces the concept of Test-driven Agile Simulation(TAS) as a consistent evolution of the systems engineering methods through the combination of test- and model-driven development techniques with the model-driven simulation. With the help of simulations, performance evaluations and validation of the modeled system can be carried out in the early stages of the development process, even when no program code or a fully implemented system is yet available. The primary goal of this approach is to combine the advantages of the above techniques to enable a holistic model-based approach to systems engineering with improved quality assurance. In particular, special attention is thereby paid to the modeling and validation of the overall system, taking into account the effects of communication between its components.
The whole approach is founded upon the widespread and established standards for Model-Driven Architecture(MDA) provided by the Object Management Group(OMG). Using the OMG's standard modeling language UML in combination with the specialized extension profiles, it is possible to specify requirements, system models as well as test models in a uniform, formal, and standard-compliant manner. The creation and presentation of the essential elements of such specifications are largely done with the help of graphical diagrams, such as class, composition structure, state, and activity diagrams. In order to facilitate behavioral modeling using detailed activity diagrams, TAS provides support for the textual activity language Alf, which is also a standard provided by OMG.
UML models can be used at different levels of abstraction for specification as well as for analysis. In the TAS approach, these models are automatically transformed into an executable simulation code which can then be executed to ensure primarily the required behavior of the system and the correctness of the tests. In this way, by running tests early on the simulated model, the mutual validation of the system and test specification is performed. The simulation of the modeled system also provides insights into the expected dynamic behavior of the system in terms of functional as well as non-functional properties.
For supporting the TAS approach, a versatile integrated tool environment is provided by our framework SimTAny. The framework offers seamless support for modeling, transforming, simulating, and testing UML-based specification models. In addition to the modeling methodology of TAS, the realization of the framework itself is largely based on the standards of the OMG. Thereby, model-based approaches and standardized transformation languages have a wide application in different components of SimTAny. Other helpful features of SimTAny include traceability of requirements across modeled elements and automatically generated code artifacts, as well as the management and design of simulation experiments. The service-oriented architecture of the framework also makes it possible to met the challenges in the distributed development processes. This also leads to the simplifications in terms of functionality extension of the framework itself and its integration into existing development environments and processes.
Molecular communication (MC) is a new communication engineering paradigm. The main characteristic of MC systems is that information is embedded in the properties of signaling molecules, e.g., their number, type, and time of release by the transmitter, which is in contrast to the conventional communication systems that embed information in the properties of electromagnetic waves, e.g., their amplitude, frequency, and phase. MC systems are envisioned to enable revolutionary applications, e.g., sensing of a target substance in biotechnology, targeted drug delivery in medicine, and monitoring of oil pipelines or chemical reactors in industrial applications. The main focus of this dissertation is on the communication theoretical design and analysis of MC systems. In particular, we develop efficient channel estimation, symbol synchronization, and detection schemes for MC systems as outlined in the following.
Channel Estimation: Knowledge of the channel impulse response (CIR) is typically needed for communication system design. In the MC literature, it is typically assumed that either the CIR is known or a specific, usually oversimplified, MC channel model is adopted and the model parameters are estimated. In this dissertation, a general training-based CIR estimation framework is developed for MC systems which is not limited to a particular MC system model and does not require knowledge of channel parameters.
Symbol Synchronization: Symbol synchronization refers to the estimation of the start of a symbol interval and is needed for reliable detection. Therefore, in this dissertation, several symbol synchronization schemes are developed that take into account the practical challenges of MC systems, namely that the transmitter may not be equipped with an internal clock and may not be able to emit molecules with a fixed release frequency.
Coherent Detection: Coherent detectors use knowledge of the channel state information (CSI) for data recovery. MC systems are subject to signal-dependent diffusion noise due to the molecule counting process, inter-symbol interference, and external interfering molecules. Taking these impairments into account, the optimal signal-to-noise-plus-interference ratio maximizing filter is derived as a function of the instantaneous CSI of the MC system. It is shown that unlike for the additive white Gaussian noise channel in conventional wireless communications, the simple channel correlator is not optimal for molecule counting receivers.
Non-Coherent Detection: Non-coherent detection schemes are desirable for application in MC systems since they do not require knowledge of the instantaneous CSI which can be difficult to obtain in practical MC scenarios. Therefore, several non-coherent multiple-symbol detectors are proposed in this dissertation. It is shown that if the number of symbols exploited for detection is sufficiently large, the performance of the proposed detectors approaches that of the optimal coherent detector which assumes perfect CSI knowledge.
CSI-Free Detection: Although the proposed non-coherent detectors do not need instantaneous CSI, statistical CSI is still needed. To relax this requirement, it is shown in this dissertation that a class of codes, namely constant-composition (CC) codes, facilitate maximum likelihood (ML) sequence detection at the receiver without requiring instantaneous or statistical CSI. The corresponding CSI-free detector is derived and shown to have a simple structure.
It is noted that for all of the above estimation/detection scenarios, the corresponding optimal ML (or maximum a posteriori) estimator/detector is derived and serves as a performance upper bound. In addition, several suboptimal low-complexity estimators/detectors are proposed that may be more suitable for applications in practical MC systems with limited computational capability. Performance analysis and comprehensive simulation results are provided to assess the effectiveness of all proposed estimators/detectors.
The selection of information on the internet is causing massive concerns and discussion. Critics fear the polarization of internet users and warn of echo chambers and filter bubbles. Especially search engines and social networking sites are being held responsible: as information intermediaries they are positioned between content producers and recipients, selecting and rating content. However, empirical findings on selection and bias in internet information intermediaries are scarce and disparate. Moreover, the preselection of information by intermediaries is not a specific characteristic of the digital world, but is equally done by traditional intermediaries such as mass media, bookshops or libraries. This paper summarizes the current state of research on selection and bias in traditional and internet information intermediaries. Different concepts, definitions and implications of the term information intermediary are analyzed in the context of the adjacent term gatekeeper. Furthermore, empirical findings on the theories of filter bubbles, echo chambers and fragmentation are summed up, as well as the state of knowledge on selection in different types of intermediaries. As a basis for further research, two models for analyzing selection and bias in intermediaries are introduced: Bozdag’s Filter Model and Jürgens & Stark’s Intermediary Effects Model. Lastly, aspects that distinguish selection in internet information intermediaries from selection in traditional contexts are compiled. This paper concludes that the production or reinforcement of bias and fragmentation by internet intermediaries is not sufficiently proven. Moreover, there is a research gap on whether traditional intermediaries have comparable effects. Further research requires a comprehensive analysis on the selection mechanisms in both traditional and internet information intermediaries.
Library updates, program errors, and maintenance tasks in general force developers to apply the same code change to different locations within their projects. If the locations are very different to each other, it is very time-consuming to identify all of them. Even with sufficient time, there is no guarantee that a manual search reveals all locations. If the change is critical, each missed location can lead to severe consequences. The manual application of the code change to each location can also get tedious. If the change is larger, developers have to execute several transformation steps for each code location. In the worst case, they forget a required step and thus add new errors to their projects.
To support developers in this task, this thesis presents the recommendation system ARES. It leads to more accurate recommendations compared to previous approaches. ARES achieves this by conserving variations in the training examples in more detail due to its pattern design and by an improved handling of code movements. With the tool C3, this thesis also presents an extension to ARES that allows the extraction of training examples from code repositories. In combination, both tools create a recommendation system that automatically learns code recommendation patterns from repositories.
ARES, C3, and similar tools rely on lists of edit operations to express code changes. However, creating compact (i.e., short) lists of edit operations from data in repositories is difficult. As previous approaches produce too long lists for ARES and C3, this thesis presents a novel tree differencing approach called MTDIFF. The evaluation shows that MTDIFF shortens the edit operation lists compared to other state-of-the-art approaches.
Ein effizientes Metamodell des Car2Car Kommunikationsaufkommens abhängig von der Umgebungstopologie
(2018)
Car-to-Car/Car2Car (C2C) Kommunikation wird seit langer Zeit in der Forschung
betrachtet. Jetzt hat die Volkswagen (VW) Aktiengesellschaft (AG) den Serieneinsatz
von Car2Car Kommunikation ab 2019 angekündigt. Deren Motivation ist die Steigerung
der Fahrsicherheit durch eine Zunahme der Informationen, die ein Fahrzeug
über sein Umfeld erhält. Die Car2Car Kommunikation ermöglicht unter anderem den
Austausch von Informationen, die Sensoren nicht detektieren können. So empfängt
ein Fahrzeug, das auf eine Kreuzung zufährt, die Position und Geschwindigkeit von in
Kürze kreuzenden und durch Gebäude verdeckten Fahrzeugen per Car2Car Kommunikation.
Dies ermöglicht eine frühzeitige Warnung des Fahrers des empfangenden
Fahrzeuges oder ein automatisiertes Eingreifen in dessen Fahrzustand. Für den Erfolg
dieser Funktion ist die Verfügbarkeit der Car2Car Daten ausschlaggebend.
Aktuell gibt es keine recheneffiziente Methode zur Emulation eines realistischen
Vehicular Ad-Hoc Networks (VANETs) unter Echtzeitbedingungen. Folglich kann der
Einfluss der Umgebung auf die Verfügbarkeit von Car2Car Daten nicht während Erprobungsfahrten
untersucht werden. Für diesen Anwendungsfall sind vorhandene Simulationsmethoden
aufgrund ihres Detailgrades zu rechenintensiv. In der vorliegenden
Arbeit wird ein effizientes Metamodell vorgestellt, das den Test der Verfügbarkeit
von Car2Car Daten für Fahrzeugfunktionen in Echtzeit ermöglicht. Das Metamodell
basiert auf einem Neuronalen Netz, das das Kommunikationsaufkommen innerhalb
eines VANETs aus den topologischen Eigenschaften des umliegenden Stadtgebietes
ableitet. Aufgrund dessen ist für die echtzeitfähige Simulation eines umgebungsspezifischen
VANETs keine komplexe Berechnung von Sender-Empfänger-Beziehungen
erforderlich, sondern deren Kommunikationserfolg wird effizient geschätzt.
Ein Vergleich des Metamodells mit dem Stand der Technik zeigt, dass damit
eine Lösung zur echtzeitfähigen Simulation von umgebungsspezifischem Car2Car
Kommunikationsaufkommen präsentiert wird, die bisher nicht verfügbar war. Das
Metamodell bietet sowohl die erforderliche Modellierungsgüte als auch die notwendige
Echtzeitfähigkeit und ermöglicht damit eine effiziente Entwicklungsmethodik für
Car2Car basierte Fahrzeugfunktionen. Des Weiteren bildet die vorgestellte Metamodellierungsmethodik
eine Grundlage für vielschichtige, weitere Forschungsarbeiten.