Search

For the current requirements in knowledge and technology transfer, solutions exist that focus exclusively on certain areas and no longer do justice to today's complexity. For example, research information systems primarily offer a marketing function, and transfer and idea portals offer a possibility for collecting and exploiting ideas that requires a great deal of human resources. This paper deals with the design of a framework that links existing systems and approaches in a meaningful way and, based on the findings of innovation research, improves the co-creative collaboration of actors from science, business and society in an it-based way. Therefore, the authors propose the CCDTP-framework based on a requirements analysis.

We want to motivate and inspire children to learn languages with the help of augmented reality. In this individual case study, we evaluate the potential of our Augmented Reality learning assistant “ARTranslate”. We conduct a individual case study by letting children do everyday shopping with the help of our augmented reality learning app. We evaluate these repeatedly with queries in the following days. The respective retention rates range from 26% to 92%. We were able to show that children autonomously learn foreign language terms with the help of our augmented reality learning initiator. To start the initial motivation when learning a new language, ARTranslate has delivered strong indications of high effectiveness.

This contribution focuses on a newly developed “Framework for Iterative Data Cleansing and Migration” (FiDM), which describes a process for achieving a quality-assured research information system, which could be implemented as a prototype. The open source solution VIVO, an ontology-based linked data software, was used to test the applicability of FiDM and to identify key requirements and findings for the general migration process. The use of FiDM supports public institutions in the partial standardization of the research management process and the integration of existing research data as well as metadata using appropriate research information systems. While ensuring data quality, networking between science, industry and the public can be established and the visibility of research achievements can be increased. FiDM was implemented and evaluated as an example at University of Applied Sciences Mittweida as part of the migration process to the RIS VIVO.

There are many ways to improve the quality of indoor localization. A big lever here is the quality and quantity of training data. For the evaluation of algorithms, fully annotated and repeatable accurate training data is essential. In order to reduce the personnel effort for manual annotations, it is advisable to generate these training data automatically. This offers the additional advantage of changing existing scenes in small steps. This enables a finely granular evaluation of algorithms trained with it. For this purpose we developed a Modular Synthesizer for Annotated Test Sets (MSATS). With MSATS we generate large data sets of completely annotated image data based on publicly available 3D CAD models. The scenes rendered automatically in small-step increments using multidimensional, modular parameter fields. The object annotations are determined from the projected virtual 2D viewer plane using a raycast method and coupled to the 3D data. The composed scenes are presented in orbitals. The resulting sets of images allow us to evaluate object detectors in detail and compare their output with the MSATS output.

Standard language texts pose significant challenges for some people because they do not understand the content due to the linguistic complexity. Transforming them into a simplified linguistic form can reduce these communicative hurdles. Easy-to-read texts enable people with learning disabilities to receive texts independently and thus form an important basis for equal participation in social life. In order to ensure a high level of comprehensibility of easy-to-read texts, simplifications are made not only to the language but also to the content. During manual translation, the simplifications made are checked by target group readers. In the case of (semi-)automated intelligibility checking, the check by target group readers is omitted. This paper focuses on the integration of the target audience of easy-to-read texts into the text review process. A web-based interface adapted to the reception needs is created and a rating system is integrated, through which people with learning disabilities can independently give feedback on the text comprehensibility. The interface will be evaluated in the form of a user study with brain-injured people. The results obtained indicate that both the web-based interface and the rating system are suitable for actively involving target group readers in the text review sub-process. The results of the study form the basis for the development of a system that takes into account the real text comprehension of the target group readers in addition to linguistic features when annotating texts in plain language.

Especially in times of distance learning, educational institutions need to diversify their digital teaching methods. Often, they offer only few opportunities for interaction, which is, however, essential for learning motivation. To integrate motivational aspects into digital teaching, this study tests digital games for their compatibility with a learning objectives catalogue of economics. Numerous studies have already shown that serious games have positive effects on learning motivation. In this study, however, not only serious games are considered, but also entertainment games. The main findings include that, on average, there is a 57% thematic overlap with the learning objectives catalogue. 20% of the objectives are explained in a way that makes abstract application possible. The thematic overlap of serious and entertainment games is nearly the same. The results show that there is a lack of games that are compatible with the considered learning catalogue. Remarkably, on average no major differences regarding knowledge transfer were found between the analysed serious and entertainment Games.

At the political and social level, easy-to-read texts are regarded as an important tool for creating barrier-free communication. In some countries, the right to comprehensible information is enshrined in law, which means that more and more standard-language texts have to be translated into easy-to-read texts. To support this intralingual translation process, previous approaches address the automation of specific sub-processes and thus focus on specific actors in the translation process. Interaction between those involved in the translation process is mostly decentralized and takes place using various applications. There is a lack of a holistic system that connects all actors involved in the process and allows interactions between the participants. This paper focuses on the implementation of a web-based assistance system and suitable interfaces that allow the representation of sub-processes of the intralingual translation process, taking into account the actors involved, and enable interactions between the participants. In order to store the data generated during the translation process in a structured and clear manner, various data management approaches could be used in this context and evaluated by means of functional tests. The results indicate that the interfaces between the actors have been optimized by the interaction components integrated in the system. Based on this, user-oriented further developments of web-based assistance systems can be carried out in the context of the intralingual translation process, whereby the system is to be used in real application contexts in the future in order to elicit the need for further assistance functions.

Computer and video games are the subject of intensive and wide-ranging research within the scientific community. The spectrum ranges from the analyses of the effects of games on the propensity to violence, well-being or on the the learning process. Previous works largely concluded positive effects of games on the learning behaviour, particularly at the motivational level. However, the analysis has mainly focused on specially designed educational games, not on games from the entertainment sector that are established on the market. Here, we address this gap by identifying suitable games and examining them for intersections with a catalogue of economical learning objectives. For this purpose, this catalogue is transformed into an adapted criteria catalogue and examined on the basis of five selected games. The main findings include that on average there is a 56% thematic overlap between the learning objectives catalogue, while 18% of the objectives were explained in a way that an abstracted application is possible even outside the gaming environment. Those findings highlight the potential value of these games for an application within the educational sector while suggesting the necessity for adjustments and additions to better address the objectives.

The shift towards digital teaching is leading to an increased need for interactive teaching methods. Game-Based Learning combines teaching content with a motivating application. The learning simulation Finanzmars uses elements from Game-Based Learning to prepare the contents of a classical lecture in the field of economics for a learning game. The simulation is aimed at students of business administration or similar courses of study. The contents can be edited and extended directly by the teaching staff using an external configuration tool. The knowledge transfer is flanked by Micro Learning, systematic introductions to increasingly complex subjects, interaction with game elements and increased motivation through in-game successes such as a premium currency. The player finds himself in an economic simulation with the goal of exploiting the resources of Mars in a profitable and optimized way. In the game, construction, upgrading and repair of buildings, the development of future technology through research, the export of resources and expansion to other celestial bodies are available. The functionalities are based on possible courses of action in the real world, whose mechanisms play a central role in the lecture on the topic of finance. By means of logging and the evaluation of in-game objectives, the teachers can track the progress of the students. Our evaluation of 12 Master's students in Business Administration shows that there is an increase in economic expertise. Furthermore, the evaluation of the associated standardized AttrakDiff questionnaire according to DIN EN ISO 9241–11 certifies that the application is action-oriented and user-friendly. The description of the environment with regard to the course of studies and the teaching concept was largely done in the publication “Finanzmars im Kosmos von Blended Learning” by Marc Ritter, Christian Roschke and Volker Tolkmitt, who received the Best Paper Award in Teaching at the CARF Lucerne Conference in 2019. This publication focuses on the systematic elaboration of game design and implementation with a view to increasing learning success.

Current research in the domain of video activity detection focuses on real-time activity detection. This includes multiple approaches in the mobile environment, such as the detection of correct motion sequences in the sports and health area or in safety-relevant environments. Current trends focus on the use of 3D CNNs. This work describes a approach to combine the results of a human skeleton point detector with an LSTM on mobile devices. Frameworks for pose detection are combined with LSTMs for activity detection with sensor data, optimized for the mobile area. The resulting system allows the direct detection of a person pose and the classification of activities in a video recorded with a smartphone. The successful application of the system in several field tests shows that the described approach works in principle and can be transferred to a resource-limited mobile environment by optimization.

In order for indoor positioning services to be able to assert themselves across a broad front, simple processes with minimal costs and low user resistance are preferable. This is where our contribution kicks in: We present a method to detect individual positions within buildings by the use of locally induced distortions of the earth’s magnetic field with a smartphone alone, without any additional technology to locate the 2D position. To compensate for the lack of exact 2D coordinates, we fuse the sensor data into a gravitational magnetic vector (GMV) and perform a temporal classification based on a recurrent network. In this work we investigate the applicability of Long-Short-Term-Memory (LSTM) networks to find cross-correlations over a time frame. The trained models are available in a smartphone application. With this application the recognition rates of the locations are analyzed.

Annotating image data is one of the most time-consuming parts of the training of machine learning algorithms. With this contribution, we are looking for a solution that decreases the time needed for annotating images of the human retina created by Optical coherence tomography (OCT). As a first step, we use a simple annotation tool to test whether the sorting of images by their predicted amount of parts that contain anomalies decreases the time needed for annotation without increasing the number of annotation mistakes. The predictions are made by a convolutional neural network (CNN) that was trained on a previously annotated image set. We investigated the annotation behaviour in two groups of five subjects each. The first group received the (OCT) images in the order of recording, the second group sorted by the number of predicted anomalies. We observed a significant increase in annotation speed in the subjects of the second group while the quality of annotation remained at least stable.

We would like to offer users an easy way to generate smart pre-selected high quality Ground Truth Data for location detection.

Virtual environments offers the possibility to simulate the real world as well as the effects like optical illusions. They are images with special arranged content to pretend the human visual system to observe non-existent content for instance. In this study we use several well known optical illusions implemented in real world and in a virtual environment. They were present to subjects to explore their properties and their perception within the different environments.

In this paper, we investigate the human decision making process in a virtual environment. The main goal is to find influencing optical and behavioral economical factors for intuitive decisions in Virtual Reality. We therefore place test persons in a virtual corridor with six visually and technically varying doors. The experimental task is to open one of the doors without overthinking or hesitating too long. This is repeated several times while randomizing and recombining optical features in every iteration. Our data shows that most of the introduced determinants actually do have an impact in the user’s decision. We can also observe different intensities of impact depending on the factor. It appears that color is by far the most influential component, followed by the complexity of the doors opening process. In contrast, position, spotlights and color brightness show only marginal correlation with choices made by the user.

The management of complex application development such as video game development warrants a separate consideration. Recurring creative tasks or process chains must be handled simultaneously to tasks of the domains of programming, implementation and testing of all project-relevant results. These drastically divergent domains create a field of tension between inter-task dependencies and resulting temporary team- or competence-related bottlenecks. Current project management approaches such as Scrum, Kanban lack the focussed linearity und structurization of recurring process chains while on the other hand rigid project management systems impede the developer’s empowerment and through incorrect montitoring or prioritization hinder the solution efficiency within the project. This paper discusses a new approach that combines several project management methodologies to address the issues identified. Based on an analysis, a concept is developed and tested prototypically in the form of a project management system. The so called “AStar Project Management” incorporates specific areas from established agile project management methods combined with classic waterfall structures.

In recent years, various methods for the automated recognition of human activities based on the analysis of sensor data using machine learning approaches have been researched. The continuously increasing hardware performance of mobile devices such as smartphones and wearables and the ongoing development of existing algorithms have resulted in steadily higher recognition rates. These latest advances resulted in an growing demand for intelligent and context sensitive mobile applications. However, the generation of valid ground truth information with a suitable quality remains a major challenge. In addition, there is currently no standardized procedure for generating an activity classifier for the use in custom application areas. The holistic workflow introduced in this paper focuses on the recognition of activities using mobile devices. For this purpose, the generation of a ground truth information by recording and annotating sensor data is described. The generated data set is used for transfer learning in a machine learning framework and the resulting model is the basis for a mobile real-time classification application. The data source is a current study of the University of Applied Sciences Mittweida.

The combination of person and location recognition provides numerous new fields of possible applications, such as the development of approaches for detecting missing persons in public spaces using real-time monitoring. It is necessary to use frameworks of both domains on given data sets and to merge the acquired results. For this purpose Thomanek et al. [11] developed an evaluation and management system for machine learning, which allows the interconnection of different frameworks and the fusion of result vectors [11]. This paper discusses the EMSML in terms of interfaces and components to develop a generic workflow that supports the integration of different frameworks for people and location recognition. In this context, the focus is on the required adaptation of existing frameworks to the implemented infrastructure. A generic workflow concept can be deduced from the analysis results. This concept can be applied to two typical frameworks for evaluation and implemented as prototypes. Subsequently, developed test cases are used to demonstrate the functional validity of the prototypes and the applicability of the concept.

Many modern systems use image understanding components to inspect, observe and react. Often, the training is realized and limited to manually annotated real world data, but dangerous or resource expensive scenarios are rare. We create a solution to overcome these limitations and reduce the manual annotation process by producing synthetic scenarios of arbitrary content and composition.

Learning management systems (LMS) are increasingly being used in the academic field as a supplement to face-to-face learning and are intended to effectively support knowledge transfer through individually integrable e-learning modules. In practice, however, it can be observed that LMS often does not live up to these expectations. Although the learning effectiveness of LMS is discussed in the literature, it is sometimes limited to the system level. Previous studies on the learning effectiveness of LMS have dealt in particular with the usability and adaptivity of the system. However, the one-sided focus on the system seems insufficient in the context of learning effectiveness, since motivational factors at the learner level are largely ignored. This study addresses the motivational factors in the use of LMS by students and uses a prototype to determine which e-learning modules have a positive effect on learning motivation. For this purpose, an e-learning prototype is developed on an LMS with various e-learning modules, and the extent to which the individual modules have an effect on learning motivation is investigated. The results of the study show that e-learning modules which demonstrate individual learning progress are suitable for promoting the learning motivation of students when using LMS.

The analysis of video footage regarding the identification of persons at defined locations or the detection of complex activities is still a challenging process. Nowadays, various (semi-)automated systems can be used to overcome different parts of these challenges. Object detection and their classification reach even higher detection rates when making use of the latest cutting-edge convolutional neural network frameworks. Integrated into a scalable infrastructure as a service data base system, we employ the combination of such networks by using the Detectron framework within Docker containers with case-specific engineered tracking and motion pattern heuristics in order to detect several activities with comparatively low and distributed computing efforts and reasonable results.

The number of application areas for object recognition are on the rise. Usable data are often limited or not well prepared to adapt to research. For image-based recognition, extensive training data is required in order to achieve precise object recognition with good repeatability.

Many systems use visual devices to detect, inspect and analyze persons, scenes, and properties of objects. Often, they use samples to learn relevant indicators to reach a high level of quality of the appropriated operation. Nevertheless, collecting samples and annotate the relevant parts may be a hard, expensive and error prone task in same fields of use. To overcome this problem we create a system to generate synthetic scenarios based on predefined and exact definitions of the content as well as the sample production process. To demonstrate the usability we apply a scenario with a humanoid with known activity and with various environment objects to different systems for visual detection and analysis.

Video surveillance of security-relevant areas is being used ever more frequently. Because of limited human resources, they are usually only checked for the presence of problematic activity after a specific event has occurred. An approach to the solution is provided by automated systems that are capable of detecting and analyzing movement sequences of objects including persons. Even though solutions already exist for scene recognition [3, 4, 8, 9], their architecture, their problem-specific domain and the nature of the systems make it difficult to integrate new activities or better algorithms. A system structure based on decentralized process instances and their communication via defined interfaces would instead enable a simpler expansion of the system. This paper describes the determination of activities in videos using decentralized frameworks and their interconnection via standardized interfaces. All components act autonomously and provide their data via a central location. Based on this approach, the modular system can be used for a wide variety of applications in the context of machine learning.

Online customer reviews are not only an important decision-making tool for customers, they are also used by e-commerce providers as a source of information to analyze customer satisfaction. In order to reduce the complexity of evaluation comments, written reviews are additionally represented by evaluation stars in many evaluation systems. Numerous studies address the sentiment recognition of written reviews and view polarity recognition as a binary or ternary problem. This study presents the first results of a holistic approach, which takes up the combination of customer reviews with evaluation points realized in platform-dependent evaluation systems. Sentiment analysis is regarded as a quinary classification problem. In this study, 5,000 customer evaluations are analyzed with lexicon-based sentiment analysis at document level with the target to predict the evaluation points based on the determined polarity. For sentiment analysis the data mining tool RapidMiner is used and the categorization of the sentiment polarity is realized by using different NLP techniques in combination with the sentiment dictionary SentiWordNet. The supervised learning algorithms k-Nearest Neighbor, Naive Bayes and Random Forest are used for classification and their classification quality is compared. Random Forest achieves the most accurate results in conjunction with NLP techniques, while the other two classifiers provide worse results. The results suggest that a stronger scaling of polarity requires a stronger differentiation between classes and thus a more intensive lexical preprocessing.

Over the last decade, the smartphone has become an important part of our everyday lives and is used even in the most rural regions. The use cases of this device go far beyond communication for which they were mainly designed. It can, e.g., be used for activity recognition or serve as a measuring device. However, for these use cases, smartphones might suffer difficulties such as jittering. Jittering is the deviation between the time an event occurs and the time the event is reported by the device. In this paper, we investigate the influence of environmental conditions such as heat as well as the processor load and memory usage on jittering. With this work, we want to contribute to a better understanding of jittering and its influencing factors.

We present a novel cryptanalytic method based on graph traversals to show that record linkage using Bloom filter encoding does not preserve privacy in a two-party setting. Bloom filter encoding is often suggested as a practical approach to medical data aggregation. This attack is stronger than a simple dictionary attack in that it does not assume knowledge of the universe. The attack is very practical and produced accurate results when experimented on large amounts of name-like data derived from a North Carolina voter registration database. We also give theoretical arguments that show that going from bigrams to n-grams, n > 2, does not increase privacy; on the contrary, it actually makes the attack more effective. Finally, some ways to resist this attack are suggested.

The UCT algorithm has gained popularity for use in AI for games, especially in board games. This paper assess the performance of UCT-based AIs and the effectiveness of augmenting them with a lookup table containing evaluations of games states in the game of Chinese Checkers. Our lookup table is only guaranteed to be correct during the endgame, but serves as an accurate heuristic throughout the game. Experiments show that using the lookup table only for its endgames is harmful, while using it for its heuristic values improves the quality of play. This work is performed on a board with 81 locations and 6 pieces, which is larger than previous work on lookup tables in Chinese Checkers. It is a precursor to using the 500 GB full-game single-agent data on the full-size board with 81 locations and 10 pieces.

Intrusion Detection Systems (IDS) have been widely deployed in practice for detecting malicious behavior on network communication and hosts. False-positive alerts are a popular problem for most IDS approaches. The solution to address this problem is to enhance the detection process by correlation and clustering of alerts. To meet the practical requirements, this process needs to be finished fast, which is a challenging task as the amount of alerts in large-scale IDS deployments is significantly high. We identifytextitdata storage and processing algorithms to be the most important factors influencing the performance of clustering and correlation. We propose and implement a highly efficient alert correlation platform. For storage, a column-based database, an In-Memory alert storage, and memory-based index tables lead to significant improvements of the performance. For processing, algorithms are designed and implemented which are optimized for In-Memory databases, e.g. an attack graph-based correlation algorithm. The platform can be distributed over multiple processing units to share memory and processing power. A standardized interface is designed to provide a unified view of result reports for end users. The efficiency of the platform is tested by practical experiments with several alert storage approaches, multiple algorithms, as well as a local and a distributed deployment. Copyright © 2011 John Wiley & Sons, Ltd.

Contemporary applications of failure criteria frequently incorporate two-dimensional or simplified three-dimensional methodologies for prediction of stresses. Motivation behind the development of a new multi-dimensional failure criterion is due mainly to the lack of a sufficiently accurate mathematical tool that accounts for the behavior of brittle material with anisotropic properties. Such a criterion should be able to provide a reliable maximum load estimate so that design of the structure is not penalized in terms of excessive weight requirements. The failure criterion developed is represented by a fracture surface in a six-dimensional stress space. The criterion is applied for failure prediction of SR-200 beryllium sheet structures, a non-homogeneous orthotropic material used widely in space applications. Two experiments are used to verify the criterion.

Modern attacks are using sophisticated and innovative techniques. The utilization of cryptography, self-modified code, and integrated attack frameworks provide more possibilities to circumvent most existing perimeter security approaches, such as firewalls and IDS. Even Application Layer Gateways (ALG) which enforce the most restrictive network access can be exploited by using advanced attack techniques. In this paper, we propose a new attack for circumventing ALGs. By using polymorphic and encrypted shellcode, multiple shellcode stages, protocol compliant and encrypted shell tunneling, and reverse channel discovery techniques, we are able to effectively bypass ALGs. The proposed attack consists of four phases with certain requirements and results. We implemented the initial shellcode as well as the different stages and conducted the practical attack using an existing ALG. The possibility to prevent this attack with existing approaches is discussed and further research in the area of perimeter security and log management is motivated.

In response to the emerging deployment of IPv6 on network devices, this paper proposes the integration of IPv6 on Lock-Keeper, an implementation of a high level security system for preventing online attacks. It is designed to permit the secure data exchange over physically separated networks in an IPv4-based environment. A new intercommunication module is added to manage IPv4/IPv6 handoff inside the Lock-Keeper, which provides several benefits. First, the Lock-Keeper gains the flexibility to work in IPv4/IPv6 environments. Second, an application layer gateway to bridge IPv4 and IPv6 networks is achieved. Third, the IP-layer protocol isolation is realized inside the Lock-Keeper to enhance the security of the protected network by exchanging data between physically separated networks using different IP protocols.

Intrusion Detection Systems (IDS) are widely deployed in computer networks. As modern attacks are getting more sophisticated and the number of sensors and network nodes grows, the problem of false positives and alert analysis becomes more difficult to solve. Alert correlation was proposed to analyze alerts and to decrease false positives. Knowledge about the target system or environment is usually necessary for efficient alert correlation. For representing the environment information as well as potential exploits, the existing vulnerabilities and their Attack Graph (AG) is used. It is useful for networks to generate an AG and to organize certain vulnerabilities in a reasonable way. In this paper, we design a correlation algorithm based on AGs that is capable of detecting multiple attack scenarios for forensic analysis. It can be parameterized to adjust the robustness and accuracy. A formal model of the algorithm is presented and an implementation is tested to analyze the different parameters on a real set of alerts from a local network.

Scanning is essential for gathering information about the actual state of computer systems or networks. Therefore, it is always taken as the first step of potential attacks against targets. In certain cases, scanning itself is categorized as an attack. Scanning can on the other hand be used for the right purposes, for example, checking the system configurations, verifying firewall rules, proofing security polices, as well as monitoring the large scale network environment. From this point of view, scanning is an effective method for system or network management, security measurement and auditing. To visualize, analyze, and finally evaluate the data gathered by scanners, Attack Graph plays an important role. High quality information about the target system or network is the prerequisite for constructing the attack graph. However, different implementations of scanners have different capabilities and always result in different kinds of outputs. These outputs are usually heterogeneous and not machine-readable, which makes the further analysis a challenging task. In this paper, we examine common types of scanners and demonstrate how to combine multiple types of scanners. The results of all the involved scanners are integrated into a well-designed and consistent data structure, which can not only be well interpreted by human security specialists but also be directly fed into an attack graph construction tool.

Intrusion Detection Systems (IDS) have been widely deployed in practice for detecting malicious behavior on network communication and hosts. The problem of false-positive alerts is a popular existing problem for most of IDS approaches. The solution to address this problem is correlation and clustering of alerts. To meet the practical requirements, this process needs to be finished as soon as possible, which is a challenging task as the amount of alerts produced in large scale deployments of distributed IDS is significantly high. We identify the data storage and processing algorithms to be the most important factors influencing the performance of clustering and correlation. We propose and implement the utilization of memory-supported algorithms and a column-oriented database for correlation and clustering in an extensible IDS correlation platform. The utilization of the column-oriented database, an In-Memory Alert Storage, and memory-based index tables leads to significant improvements on the performance. Different types of correlation modules can be integrated and compared on this platform. A plugin concept for Receivers provides flexible integration of various sensors and additional IDS management systems. The platform can be distributed over multiple processing units to share memory and processing power. A standardized interface is designed to provide a unified view of result reports for end users. The efficiency of the proposed platform is tested by practical experiments with several alert storage approaches, different simple algorithms, as well as local and distributed deployment.

Intrusion Detection Systems (IDS) have been used widely to detect malicious behavior in network communication and hosts. IDS management is an important capability for distributed IDS solutions, which makes it possible to integrate and handle different types of sensors or collect and synthesize alerts generated from multiple hosts located in the distributed environment. Sophisticated attacks are difficult to detect and make it necessary to integrate multiple data sources for detection and correlation. Attack graph (AG) is used as an effective method to model, analyze, and evaluate the security of complicated computer systems or networks. The attack graph workflow consists of three parts: information gathering, attack graph construction, and visualization. This paper proposes the integration of the AG workflow with an IDS management system to improve alert and correlation quality. The vulnerability and system information is used to prioritize and tag the incoming IDS alerts. The AG is used during the correlation process to filter and optimize correlation results. A prototype is implemented using automatic vulnerability extraction and AG creation based on unified data models.

Modern attacks are using sophisticated and innovative techniques. Teaching and training programs have to focus on the practical aspects of security, i.e., on attack techniques and defense methods, such as Intrusion Detection Systems (IDS). Team exercises provide an effective method to increase practical experience in security, as each team can gather knowledge on offensive as well as defensive techniques while learning from each other. The proposed laboratory provides effective security training by offering practical scenarios to practice attack and defense. The scenarios are parameterized and configured automatically to improve manageability. By using virtualized components as well as dedicated network infrastructure components, we can create complex scenarios that reflect real network environments. We designed and implemented several scenarios with various levels of difficulty. Finally, the teaching experiences for students and teachers are described based on the execution of the scenarios in multiple training sessions.

Simulation is an efficient way to model a real system by a computer program and is used to study and evaluate the characteristics or behaviors of the system. In this paper, we present an effective simulation tool that is designed for simulating the working procedure of the Lock-Keeper system, which is a high level security device for physically separating two networks. Due to the special mechanism of data exchange within the Lock-Keeper system, it is a challenging task to specify the Lock-Keeper's performance for a given application scenario. To get enough performance data, the intuitive way is to practically conduct numerous testing experiments and then analyze their results, which is extremely time consuming. Therefore, we are motivated to design and implement a simulator to predict the Lock-Keeper performance theoretically. Compare with most of available network simulation tools, the proposed simulator is capable of simulating the transfer of application layer data, i.e., file based data streams. The simulator is built based on a simple model of the Lock-Keeper data exchange procedure. Information on a target application scenario can be specified within an XML file, which is used as the input for the later calculation. Several kinds of reports are generated by this specialized simulator to indicate how the data is exchanged through Lock-Keeper. To verify the simulation results, we conduct several experiments, which practically test data transfer for the scenarios using the real Lock-Keeper system. The comparison between theoretical simulation and practical testing proves the effectiveness of our proposed simulation tool.

This paper addresses the challenges to formally specify the vulnerability information and unify text-based vulnerability descriptions, which might be available in various commercial, governmental, or open source vulnerability databases, into a generic information model. Our motivation is to utilize the remodeled vulnerability data for automating the construction of attack graph, which has been recognized as an effective method for visualizing, analyzing, and measuring the security of complicated computer systems or networks. A formal data structure is proposed based on a comprehensive conceptual analysis on normal computer infrastructure and related vulnerabilities. The newly proposed vulnerability representation, which contains most of meaningful properties extracted from textual descriptions of actual vulnerability items, can be directly fed into the reasoning engine of attack graph tools. A lightweight information extraction mechanism is designed to automatically transform textual vulnerability descriptions into the proposed data structure. Several Reader and Writer plugins are implemented to enable the communication with known vulnerability repositories.

Intrusion Detection Systems (IDS) have been used widely to detect malicious behaviors in network communication and hosts. IDS management is an important capability for distributed IDS solutions, which makes it possible to integrate and handle different types of sensors or collect and synthesize alerts generated from multiple hosts located in the distributed environment. Facing new application scenarios in Cloud Computing, the IDS approaches yield several problems since the operator of the IDS should be the user, not the administrator of the Cloud infrastructure. Extensibility, efficient management, and compatibility to virtualization-based context need to be introduced into many existing IDS implementations.Additionally, the Cloud providers need to enable possibilities to deploy and configure IDS for the user. Within this paper, we summarize several requirements for deploying IDS in the Cloud and propose an extensible IDS architecture for being easily used in a distributed cloud infrastructure.

Performance is a critical aspect for all kinds of security solutions. The Lock-Keeper is a high-level security solution which implements the concept of physical separation. To evaluate and predict the performance of Lock-Keeper, the internal data transfer process has to be analyzed and simulated. As the Lock-Keeper consists of three independent systems, connected through a hardware-based switch, with file-based communication, the internal data transfer process is relatively complex. In this paper, we propose a mathematical model which can be used to formally represent most of the Lock-Keeper data transfer scenarios. The model is verified by practical experiments for three different application scenarios. We conclude that the performance of the Lock-Keeper system highly depends on the application scenarios.

Efficient Intrusion Detection System (IDS) management is a prominent capability for distributed IDS solutions, which makes it possible to integrate and handle different types of sensors or collect and synthesize alerts generated from multiple hosts located in a loosely coupled environment. Extensibility is the main requirement for most of IDS management systems. The concept of virtualization has been introduced into many popular IDS implementations due to the advantage on isolation and fast recovery in case of being compromised. Advanced capability for combining these newly emerged Virtual Machine (VM) based IDS approaches is another requirement for IDS management. This paper proposes an extensible IDS management architecture based on a new design of Event Gatherer component. By using the known IDS standard IDMEF and a plug-in concept, the Event Gatherer ensures flexibility and compatibility.Experiments are carried out to demonstrate the extensibility and virtualization-compatibility of the proposed IDS management architecture.

Intrusion Detection System (IDS) management is an important component for most distributed IDS solutions. One of the main requirements is extensibility, which enables the integration of different types of IDS sensors as well as the deployment in different kinds of environments. Lock-Keeper is a simple implementation of the high level security idea, "Physical Separation". It works as a sluice to exchange data between two networks without having to establish a direct and physical connection. To enhance the security of the Lock-Keeper system itself, it is necessary to deploy IDS sensors on Lock-Keeper components. This paper proposes an extensible IDS management architecture, which can be easily integrated on the special hardware platform of Lock-Keeper. Unified interface and communication between different integrated IDS sensors are designed using the known IDS standard, IDMEF, and realized as several kinds of plugins, such as handlers, receivers, and senders. A prototype of implementation is presented and some practical experiments are carried out to show the extensibility and applicability of the proposed architecture.

Attack graph is used as an effective method to model, analyze, and evaluate the security of complicated computer systems or networks. The attack graph workflow consists of three parts: information gathering, attack graph construction, and visualization. To construct an attack graph, runtime information on the target system or network environment should be monitored, gathered, and later evaluated with existing descriptions of known vulnerabilities. The output will be visualized into a graph structure for further measurements. Information gatherer, vulnerability repository, and the visualization module are three important components of an attack graph constructor. However, high quality attack graph construction relies on up-to-date vulnerability information. There are already some existing databases maintained by security companies, a community, or governments. Such databases can not be directly used for generating attack graph, due to missing unification of the provided information. This paper challenged the automatic extraction of meaningful information from various existing vulnerability databases. After comparing existing vulnerability databases, a new method is proposed for automatic extraction of vulnerability information from textual descriptions. Finally, a prototype was implemented to proof the applicability of the proposed method for attack graph construction.

The paper proposes a new computational approach for predicting the dynamic response characteristic of magnetic actuators, e.g. a long-stroke linear actuator. This method is based on a comprehensive magnetic energy formulation. Using the magnetic co-energy as a function of position of the armature and magnetomotive force, a system of first order differential equations can be found which relates the electrical supply to the magnetic behavior and the movement of the plunger.

Archiving large datasets requires parallel processing of both data and metadata for timely execution. This paper describes the work in progress to use various processing techniques, including multi-threading of data and metadata operations, distributed processing, aggregation, and conditional processing to achieve increased archival performance for large datasets.

The intention of Chapter 9 is to provide an practical approach to the design of fractional-horsepower drive systems. Here the system view and its function is always in focus. First of all common requirements and specifications for drive systems are classified and an introduction to safety-related demands from authorities is given. A systematic procedure for solving drive tasks is introduced and common types of motion, e.g. constant speed, point-to point control, reversing motion, and path control are categorised. Most important aspects of motor selection are explained and equations for conversion of mechanical drive parameters are given. The comparison of dc-motors and stepper motors for positioning tasks is discussed in detail. In the following seven different examples of the design of small drive systems are handeled in detail, e.g. disk storage drive, piston pump, or light pointer. Throughout this chapter the same systematic and stepwise approach for concept decision, dimensioning, and verification is applied. The intention behind is to support the reader with learning on practical applications.