Disruption-tolerant networks (DTNs) have a wide range of applications, including emergencies where traditional communication infrastructure has been destroyed, remote rural deployments where communication infrastructure does not exist, and environmental monitoring in which animals are equipped with sensors and transmit data whenever they come into contact with a base station. Using the de-facto DTN protocol standard, Bundle Protocol version 7, nodes transmit data using Convergence Layer Protocols, which serve as abstractions for the underlying communication technology. In this article, we introduce QUICL , a novel convergence layer for disruption-tolerant networks. QUICL is built on the QUIC transport protocol, which offers advantages over TCP in a disruption-tolerant setting. In particular, it improves congestion control, supports multiplexing, ensures reliable transmission, effectively manages unstable networks, and encrypts traffic by default. Our implementation, already merged upstream, is based on the free and open-source DTN7-go -protocol suite and the QUIC-go -library. Our experimental evaluation shows that even in challenging situations such as bundle transmission over 63 hops with a packet loss of 30% on each hop, QUICL still delivers data where most other DTN software/convergence layer combinations fail to transmit any data.

Abstract Search Anything is presented, a novel approach to perform similarity search in images. In contrast to other approaches to image similarity search, Search Anything enables users to utilize point, box, and text prompts to search for similar regions in a set of images. The region selected by a prompt is automatically segmented, and a binary feature vector is extracted. This feature vector is then used as a query for an image region index, and the images that contain the corresponding regions are returned. Search Anything is trained in a self-supervised manner on mask features extracted by the FastSAM foundation model and semantic features for masked image regions extracted by the CLIP foundation model to learn binary hash code representations for image regions. By coupling these two foundation models, images can be indexed and searched at a more fine-grained level than finding only entire similar images. Experiments on several datasets from different domains in a zero-shot setting demonstrate the benefits of Search Anything as a versatile region-based similarity search approach for images. The efficacy of the approach is further supported by qualitative results. Ablation studies are performed to evaluate how the proposed combination of semantic features and segmentation features together with masking improves the performance of Search Anything over the baseline using CLIP features alone. For large regions, relative improvements of up to 9.87% in mean average precision are achieved. Furthermore, considering context is beneficial for searching small image regions; a context of 3 times an object’s bounding box gives the best results. Finally, we measure computation time and determine storage requirements.

Wireless sensor networks (WSNs) can automate data sensing tasks. To ensure redundancy and manage network connectivity issues, a sensing node stores a copy of the gathered data. Since this data may contain sensitive personal or business information, protecting privacy and preventing unauthorized access is crucial. We introduce the Write-only File System (WoFS), a novel encryption system for WSNs that secures data without user interaction, even if a sensor node is stolen. WoFS utilizes either symmetric encryption with volatile keys via a ratchet mechanism or asymmetric encryption. Asymmetric encryption, while slower, allows operation post-reboot, unlike the ratchet-based method. Our experiments show that WoFS achieves write speeds of 200 MB/s or higher, making it suitable for WSN applications. All developed software and artifacts are available under a permissive open-source license.

Fires are among the most dangerous and life-threatening emergency situations that might occur in residential settings. Therefore, smoke detectors are required in many rooms of a house in many European countries, such as the Netherlands, the UK, Germany, and Austria. We present a novel smart smoke detector that can be used for communication in emergency situations. In addition to playing a sound when it detects smoke, it provides (i) an acoustic message interface to call for help or leave a message, (ii) a mesh-based wireless communication interface to allow communication even if infrastructure is damaged, and (iii) an accelerometer to detect earthquakes.

Recent advances in low-cost microcontrollers have enabled innovative smart home applications. However, existing systems typically consist of single-purpose devices that only report sensed data to a controller. Given the potential for residential emergencies, we propose to integrate emergency detection systems into smart home environments. We present an ad-hoc distributed sensor network (DSN) designed to detect five common residential emergencies: fires, gas and water leakages, earthquakes, and intrusions. Our novel approach combines diverse sensors with a voting-based consensus algorithm among multiple nodes, improving accuracy and reliability over traditional alert systems. The consensus algorithm employs a majority rule with weighted votes, allowing adjustments for various scenarios. An experimental evaluation confirms our approach’s effectiveness in accurately detecting emergencies while demonstrating reliability in mitigating node failures, ensuring system longevity, and maintaining robust communication. Additionally, our approach significantly reduces power consumption compared to alternatives.

Abstract Most machine learning methods for animal recognition in camera trap images are limited to mammal identification and group birds into a single class. Machine learning methods for visually discriminating birds, in turn, cannot discriminate between mammals and are not designed for camera trap images. The authors present deep neural network models to recognise both mammals and bird species in camera trap images. They train neural network models for species classification as well as for predicting the animal taxonomy, that is, genus, family, order, group, and class names. Different neural network architectures, including ResNet, EfficientNetV2, Vision Transformer, Swin Transformer, and ConvNeXt, are compared for these tasks. Furthermore, the authors investigate approaches to overcome various challenges associated with camera trap image analysis. The authors’ best species classification models achieve a mean average precision (mAP) of 97.91% on a validation data set and mAPs of 90.39% and 82.77% on test data sets recorded in forests in Germany and Poland, respectively. Their best taxonomic classification models reach a validation mAP of 97.18% and mAPs of 94.23% and 79.92% on the two test data sets, respectively.

We present a novel multi-hop data dissemination protocol for wireless networks that minimizes the total energy consumption across an entire network by minimizing the transmission power at each hop. It is based on a game-theoretic model, constructs a spanning tree topology in a decentralized manner, and is usable in practice. We evaluate the protocol via simulation and a pratical implementation on a testbed of 75 Raspberry Pis, demonstrating that a total energy reduction of up to 90% can be achieved compared to a simple broadcast protocol.

Disruption-tolerant networks (DTNs) have a wide range of applications, such as emergencies where traditional communication infrastructure has been destroyed, remote rural deployments where communication infrastructure has never existed, and environmental monitoring where animals are equipped with sensors and transmit data whenever they come into contact with a base station. Using the de-facto DTN protocol standard, i.e., Bundle Protocol version 7 (BPv7), nodes transmit data via so-called Convergence Layer Protocols (CLPs) that act as general abstractions for the underlying communication technologies. BPv7 specifies MTCP and TCPCL as the two current CLPs for DTNs. However, both of them have different but equally undesirable shortcomings in terms of functionality, complexity, performance, and reliability. In this paper, we present QUICL, a novel CLP for DTNs. QUICL is based on the QUIC transport protocol and fully leverages QUIC's advantages over TCP-based transport protocols in a DTN environment. In particular, QUICL provides improved congestion control, allows multiplexing, ensures reliable transmission, effectively manages unreliable links, and uses encryption by default. Our prototypical implementation, already merged upstream, is based on the free and open-source DTN7-go protocol suite and the QUIC-go implementation. Our experimental evaluation shows that even with 30% packet loss, QUICL can still deliver data with minimal CPU overhead in scenarios where most other DTN/CLP combinations fail to transmit any data successfully.

We present a novel smart trap for counting olive moths using Internet of Things principles and deep learning algorithms. The smart trap takes a picture of the captured insects once per day and processes it using a deep convolutional neural network to detect “Prays oleae” insects and count their number. Then, the results are transferred via a wireless connection to a backend cloud server. The proposed smart trap is designed to reduce power consumption as much as possible. Two deep neural network models (i.e., YOLO V5 and YOLO V7) are employed to detect and count Prays oleae insects, using our newly created Prays oleae dataset, PraysDB. Our experimental results demonstrate the detection quality, energy efficiency, and computational performance of our smart trap.

A network for environmental monitoring typically requires a large number of sensors. If a longer service life is intended, it is essential that the deployed sensor systems can be upgraded without modifying hardware. Often, these networks rely on proprietary hardware/software components tailored to the desired functionality, but these could technically also be used for other applications. We present a demo of mechanism interception, a novel approach to unobtrusively add or modify the functionality of an existing networked system, in our case a TreeTalker, without touching any proprietary components. We demonstrate how a cloud infrastructure can be unobtrusively replaced by an edge infrastructure in a wireless sensor network. Our results indicate that mechanism interception is a compelling approach for our scenario to provide previously unavailable functionality without modifying existing components.

Networked systems and applications are often based on proprietary hardware/software components that manufacturers might not be willing to adapt or update if new requirements arise. We present mechanism interception, a novel approach to unobtrusively add or modify functionality to/of an existing networked system or application without touching any proprietary components. Behavioral changes are achieved by functionality-enhancing yet unobtrusive interceptors, i.e., components introduced between systems and their environments adding or updating mechanisms. We illustrate our approach by unobtrusively adding a vertical handover mechanism between Wi-Fi and LTE to a mobile end device without disconnecting TCP sessions. Our results indicate that mechanism interception is a compelling approach to achieve improved service quality and provide previously unavailable functionality.

During disasters, existing telecommunication infrastructures are often congested or even destroyed. In these situations, mobile devices can form a backup communication network for civilians and emergency services using disruption-tolerant networking (DTN) principles. Unfortunately, such distributed and resource-constrained networks are particularly susceptible to a wide range of attacks such as terrorists trying to cause more harm. In this article, we presentRESCUE, a resilient and secure device-to-device communication framework for emergency scenarios that provides comprehensive protection against common attacks.RESCUEfeatures a minimalistic DTN protocol that, by design, is secure against notable attacks such as routing manipulations, dropping, message manipulations, blackholing, or impersonation. To further protect against message flooding and Sybil attacks, we present a twofold mitigation technique. First, a mobile and distributed certificate infrastructure particularly tailored to the emergency use case hinders the adversarial use of multiple identities. Second, a message buffer management scheme significantly increases resilience against flooding attacks, even if they originate from multiple identities, without introducing additional overhead. Finally, we demonstrate the effectiveness ofRESCUEvia large-scale simulations in a synthetic as well as a realistic natural disaster scenario. Our simulation results show thatRESCUEachieves very good message delivery rates, even under flooding and Sybil attacks.

Mobile edge computing based on cloudlets is an emerging paradigm to improve service quality by bringing computation and storage facilities closer to end users and reducing operating cost for infrastructure providers (IPs) and service providers (SPs). To maximize their individual benefits, IP and SP have to reach an agreement about placing and executing services on particular cloudlets. We show that a Nash Bargaining Solution (NBS) yields the optimal solution with respect to social cost and fairness if IP and SP have complete information about the parameters of their mutual cost functions. However, IP and SP might not be willing or able to share all information due to business secrets or technical limitations. Therefore, we present a novel iterative bargaining approach without complete mutual information to achieve substantial cost reductions for both IP and SP. Furthermore, we investigate how different degrees of information sharing impact social cost and fairness of the different approaches. Our evaluation based on the mobile augmented reality game Ingress shows that our approach achieves up to about 82% of the cost reduction that the NBS achieves and a cost reduction of up to 147% compared to traditional Take-it-or-Leave-it approaches, despite incomplete information.

BACKGROUND: DNA is a promising storage medium for high-density long-term digital data storage. Since DNA synthesis and sequencing are still relatively expensive tasks, the coding methods used to store digital data in DNA should correct errors and avoid unstable or error-prone DNA sequences. Near-optimal rateless erasure codes, also called fountain codes, are particularly interesting codes to realize high-capacity and low-error DNA storage systems, as shown by Erlich and Zielinski in their approach based on the Luby transform (LT) code. Since LT is the most basic fountain code, there is a large untapped potential for improvement in using near-optimal erasure codes for DNA storage. RESULTS: We present NOREC4DNA, a software framework to use, test, compare, and improve near-optimal rateless erasure codes (NORECs) for DNA storage systems. These codes can effectively be used to store digital information in DNA and cope with the restrictions of the DNA medium. Additionally, they can adapt to possible variable lengths of DNA strands and have nearly zero overhead. We describe the design and implementation of NOREC4DNA. Furthermore, we present experimental results demonstrating that NOREC4DNA can flexibly be used to evaluate the use of NORECs in DNA storage systems. In particular, we show that NORECs that apparently have not yet been used for DNA storage, such as Raptor and Online codes, can achieve significant improvements over LT codes that were used in previous work. NOREC4DNA is available on https://github.com/umr-ds/NOREC4DNA . CONCLUSION: NOREC4DNA is a flexible and extensible software framework for using, evaluating, and comparing NORECs for DNA storage systems.

Google and Apple have jointly provided an API for exposure notification in order to implement decentralized contract tracing apps using Bluetooth Low Energy, the so-called “Google/Apple Proposal”, which we abbreviate by “GAP”. We demonstrate that in real-world scenarios the current GAP design is vulnerable to (i) profiling and possibly de-anonymizing infected persons, and (ii) relay-based wormhole attacks that basically can generate fake contacts with the potential of affecting the accuracy of an app-based contact tracing system. For both types of attack, we have built tools that can easily be used on mobile phones or Raspberry Pis (e.g., Bluetooth sniffers). The goal of our work is to perform a reality check towards possibly providing empirical real-world evidence for these two privacy and security risks. We hope that our findings provide valuable input for developing secure and privacy-preserving digital contact tracing systems.

SUMMARY: The development of de novo DNA synthesis, polymerase chain reaction (PCR), DNA sequencing and molecular cloning gave researchers unprecedented control over DNA and DNA-mediated processes. To reduce the error probabilities of these techniques, DNA composition has to adhere to method-dependent restrictions. To comply with such restrictions, a synthetic DNA fragment is often adjusted manually or by using custom-made scripts. In this article, we present MESA (Mosla Error Simulator), a web application for the assessment of DNA fragments based on limitations of DNA synthesis, amplification, cloning, sequencing methods and biological restrictions of host organisms. Furthermore, MESA can be used to simulate errors during synthesis, PCR, storage and sequencing processes. AVAILABILITY AND IMPLEMENTATION: MESA is available at mesa.mosla.de, with the source code available at github.com/umr-ds/mesa_dna_sim. CONTACT: [email protected]. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

BACKGROUND: Sequencing of marker genes amplified from environmental samples, known as amplicon sequencing, allows us to resolve some of the hidden diversity and elucidate evolutionary relationships and ecological processes among complex microbial communities. The analysis of large numbers of samples at high sequencing depths generated by high throughput sequencing technologies requires efficient, flexible, and reproducible bioinformatics pipelines. Only a few existing workflows can be run in a user-friendly, scalable, and reproducible manner on different computing devices using an efficient workflow management system. RESULTS: We present Natrix, an open-source bioinformatics workflow for preprocessing raw amplicon sequencing data. The workflow contains all analysis steps from quality assessment, read assembly, dereplication, chimera detection, split-sample merging, sequence representative assignment (OTUs or ASVs) to the taxonomic assignment of sequence representatives. The workflow is written using Snakemake, a workflow management engine for developing data analysis workflows. In addition, Conda is used for version control. Thus, Snakemake ensures reproducibility and Conda offers version control of the utilized programs. The encapsulation of rules and their dependencies support hassle-free sharing of rules between workflows and easy adaptation and extension of existing workflows. Natrix is freely available on GitHub ( https://github.com/MW55/Natrix ) or as a Docker container on DockerHub ( https://hub.docker.com/r/mw55/natrix ). CONCLUSION: Natrix is a user-friendly and highly extensible workflow for processing Illumina amplicon data.

Detection and segmentation of macrophage cells in fluorescence microscopy images is a challenging problem, mainly due to crowded cells, variation in shapes, and morphological complexity. We present a new deep learning approach for cell detection and segmentation that incorporates previously learned nucleus features. A novel fusion of feature pyramids for nucleus detection and segmentation with feature pyramids for cell detection and segmentation is used to improve performance on a microscopic image dataset created by us and provided for public use, containing both nucleus and cell signals. Our experimental results indicate that cell detection and segmentation performance significantly benefit from the fusion of previously learned nucleus features. The proposed feature pyramid fusion architecture clearly outperforms a state-of-the-art Mask R-CNN approach for cell detection and segmentation with relative mean average precision improvements of up to 23.88% and 23.17%, respectively.

In personalized medicine, a challenging task is to identify the most effective treatment for a patient. In oncology, several computational models have been developed to predict the response of drugs to therapy. However, the performance of these models depends on multiple factors. This paper presents a new approach, called Q-Rank, to predict the sensitivity of cell lines to anti-cancer drugs. Q-Rank integrates different prediction algorithms and identifies a suitable algorithm for a given application. Q-Rank is based on reinforcement learning methods to rank prediction algorithms on the basis of relevant features (e.g., omics characterization). The best-ranked algorithm is recommended and used to predict the response of drugs to therapy. Our experimental results indicate that Q-Rank outperforms the integrated models in predicting the sensitivity of cell lines to different drugs.

Emerging network architectures like Information-Centric Networking (ICN)offer simplicity in the data plane by addressing named data. Such flexibility opens up the possibility to move data processing inside network elements for high-performance computation, known as in-network processing. However, existing ICN architectures are limited in terms of (i)in-network processing and (ii)data plane programming abstractions. Such architectures can benefit from Complex Event Processing (CEP), an in-network processing paradigm to efficiently process data inside the data plane. Yet, it is extremely challenging to integrate CEP because the current communication model of ICN is limited to consumer-initiated interaction that comes with significant overhead in number of requests to process continuous data streams. In contrast, a change to producer-initiated interaction, as favored by CEP, imposes severe limitations for request-reply interactions. In this paper, we propose an in-network CEP architecture, INETCEP that supports unified interaction patterns (consumer- and producer-initiated). In addition, we provide a CEP query language and facilitate CEP operations while increasing the range of applications that can be supported by ICN. We provide an open source implementation and evaluation of INETCEP over an ICN architecture, Named Function Networking, and two applications: energy forecasting in smart homes and a disaster scenario.

Task scheduling in cloud environments is the problem of assigning and executing computational tasks on the available cloud resources. Effective task scheduling approaches reduce the task completion time, increase the efficiency of resource utilization, and improve the quality of service and the overall performance of the system. In this paper, we present a novel task scheduling algorithm for cloud environments based on the Heterogeneous Earliest Finish Time (HEFT) algorithm, called experiential HEFT. It considers experiences with previous executions of tasks to determine the workload of resources. To realize the experiential HEFT algorithm, we propose a novel way of HEFT rank calculation to specify the minimum average execution time of previous runs of a task on all relevant resources. Experimental results indicate that the proposed experiential HEFT algorithm performs better than HEFT and the popular Critical-Path-on-a-Processor (CPOP) algorithm considered in our comparison.

In this paper, we present a convolutional gated recurrent neural network (CGRNN) to predict epileptic seizures based on features extracted from EEG data that represent the temporal aspect and the frequency aspect of the signal. Using a dataset collected in the Children’s Hospital of Boston, CGRNN can predict epileptic seizures between 35 min and 5 min in advance. Our experimental results indicate that the performance of CGRNN varies between patients. We achieve an average sensitivity of 89% and a mean accuracy of 75.6% for the patients in the data set, with a mean False Positive Rate (FPR) of 1.6 per hour.

This paper presents a deep learning approach for depressive episode detection on mobile devices, called DL4DED. It is based on a convolutional neural network and a long short-term memory network to identify the status of a patient’s voice extracted from spontaneous phone calls. To run DL4DED on mobile devices, two neural network model compression techniques are used: quantization and pruning. DL4DED protects data privacy, since it can be executed on a patient’s smartphone. Our proposal is validated on the DAIC-WOZ database. The obtained results show that the accuracy of DL4DED with model compression is only slightly lower than the accuracy of DL4DED without model compression. Furthermore, our experiments indicate that the power consumption of DL4DED is reasonably low.

We present a novel data-driven approach to perform smooth Wi-Fi/cellular handovers on smartphones. Our approach relies on data provided by multiple smartphone sensors (e.g., Wi-Fi RSSI, acceleration, compass, step counter, air pressure) to predict Wi-Fi connection loss and uses Multipath TCP to dynamically switch between different connectivity modes. We train a random forest classifier and an artificial neural network on real-world sensor data collected by five smartphone users over a period of three months. The trained models are executed on smartphones to reliably predict Wi-Fi connection loss 15 seconds ahead of time, with a precision of up to 0.97 and a recall of up to 0.98. Furthermore, we present results for four DASH video streaming experiments that run on a Nexus 5 smartphone using available Wi-Fi/cellular networks. The neural network predictions for Wi-Fi connection loss are used to establish MPTCP subflows on the cellular link. The experiments show that our approach provides seamless wireless connectivity, improves quality of experience of DASH video streaming, and requires less cellular data compared to handover approaches without Wi-Fi connection loss predictions.

Computation offloading is often used in mobile cloud, edge, and/or fog computing to cope with resource limitations of mobile devices in terms of computational power, storage, and energy. Computation offloading is particularly challenging in situations where network connectivity is intermittent or error-prone. In this paper, we present OPPLOAD, a novel framework for offloading computational workflows in opportunistic networks. The individual tasks forming a workflow can be assigned to particular remote execution platforms (workers) either preselected ahead of time or decided just in time where a matching worker will automatically be assigned for the next task. Tasks are only assigned to capable workers that announce their capabilities. Furthermore, tasks of a workflow can be executed on multiple workers that are automatically selected to balance the load. Our Python implementation of OPPLOAD is publicly available as open source software. The results of our experimental evaluation demonstrate the feasibility of our approach.

In the aftermath of large-scale disasters, critical communication infrastructure is often destroyed. Ad hoc networks can restore wireless communication with basic functionalities, especially for civilians in the affected areas. However, as humans form groups and tend to stay around important locations like shelters in such situations, the network is highly intermittent. Autonomous Unmanned Aerial Vehicles can act as controllable and highly mobile data carriers between separated network clusters to enable delay-tolerant inter-cluster communication. A possible heterogeneous set of usable aerial vehicles and the necessity to adapt the system to various environments requires the system to be highly flexible. Combined with severe constraints in energy usage or number of available vehicles, there is a need to increase the efficiency of inter-cluster flights. In this paper, we present approaches to optimize inter-cluster flights based on communication performance and energy efficiency. The resulting optimization model is a powerful tool for operators to adjust system settings to match required demands.

Information-centric disruption-tolerant networks (ICN-DTNs) are useful to re-establish mobile communication in disaster scenarios when telecommunication infrastructures are partially or completely unavailable. In this paper, we present opportunistic named functions, a novel approach to operate ICN-DTNs during emergencies. Affected people and first responders use their mobile devices to specify their interests in particular content and/or application-specific functions that are then executed in the network on the fly, either partially or totally, in an opportunistic manner. Opportunistic named functions rely on user-defined interests and on locally optimal decisions based on battery lifetimes and device capabilities. In the presented emergency scenario, they are used to preprocess, analyze, integrate and transfer information extracted from images produced by smartphone cameras, with the aim of supporting the search for missing persons and the assessment of critical conditions in a disaster area. Experimental results show that opportunistic named functions reduce network congestion and improve battery lifetime in a network of battery-powered sensors, mobile devices, and mobile routers, while delivering crucial information to carry out situation analysis in disasters.

In this paper, we present a holistic approach to provide fault tolerance for distributed reactive programming. Our solution automatically stores and recovers program state to handle crashes, automatically updates and shares distributed parts of the state to provide eventual consistency, and handles errors in a fine-grained manner to allow precise manual control when necessary. By making use of the reactive programming paradigm, we provide these mechanisms without changing the behavior of existing programs and with reasonable performance, as indicated by our experimental evaluation.

Augmented Reality (AR) is an approach to enrich the real world with additional information. It allows users to interact with virtual objects that are linked to locations in the real world. In the area of mobile computer games, AR is quite demanding for managing resources in communication networks, since in-game points of interest typically lead to high network loads, whereas network utilization is otherwise below the average. Predicting the number of in-game actions of mobile AR games can help to scale the game back-end reasonably without over- or under-provisioning of resources. Context information like weather or available Wi-Fi access points can play a key role in estimating or predicting the number of in-game actions. In this paper, we analyze a comprehensive dataset that contains players' actions of one of the most popular mobile AR games, Ingress. The dataset entails more than 23.9 million player actions over a period of 17 months, as well as additional contextual information. Our analysis shows a highly significant relationship between context factors and in-game actions, which explains large parts of users' behavior in terms of when, where and how often they play the game. By combining four different context factors (i.e., time, user, physical and computing context), our analysis can explain up to 84.44% of the variation in the number of in-game actions.

In this paper, we present a novel approach to solve the labeled point cloud superpositioning problem for performing structural comparisons of protein binding sites. The solution is based on a parallel evolution strategy that operates on large populations and runs on GPU hardware. The proposed evolution strategy reduces the likelihood of getting stuck in a local optimum of the multimodal real-valued optimization problem represented by labeled point cloud superpositioning. The performance of the GPU-based parallel evolution strategy is compared to a previously proposed CPU-based sequential approach for labeled point cloud superpositioning, indicating that the GPU-based parallel evolution strategy leads to qualitatively better results and significantly shorter runtimes, with speed improvements of up to a factor of 1,500 for large populations. Binary classification tests based on the ATP, NADH, and FAD protein subsets of CavBase, a database containing putative binding sites, show average classification rate improvements from about 92 percent (CPU) to 96 percent (GPU). Further experiments indicate that the proposed GPU-based labeled point cloud superpositioning approach can be superior to traditional protein comparison approaches based on sequence alignments.

Software-defined networking paradigms have found their way into wireless edge networks, allowing network slicing, mobility management, and resource allocation. This paper presents dynamic role assignment as a novel approach to software-defined network topology management for wireless edge devices, such as laptops, tablets and smartphones. It combines the centralized control of wireless Network Interface Controller (NIC) modes with Network Function Virtualization (NFV) to integrate network topology transitions as well as network service and application service placement within a single mechanism. Our proposal is evaluated with respect to latency, bandwidth, and power consumption of the edge nodes. The experimental results show significant differences in both bandwidth (up to 18%) and power consumption (up to 15%) for playing different roles, and when using (a) a web proxy and (b) an intrusion prevention system as examples of application services.

In this paper, we present MiniWorld, a novel distributed network emulator. It is based on full virtualization using QEMU/KVM, offers three network backends for emulating both wired and wireless communication, and provides several mobility patterns as well as distance-based link quality models. A snapshot boot mode is offered for accelerated booting of identical environments and repeating emulation runs. To decrease runtimes, MiniWorld supports distributed emulation across multiple computers, based on a resource-aware virtual machine (VM) scheduler. Experimental results demonstrate the performance of MiniWorld with respect to VM boot times, network bandwidth, round trip times, and topology switching times.

To cope with the varying delay and bandwidth requirements of today's mobile applications, mobile wireless networks can profit from classifying and predicting mobile application traffic. State-of-the-art traffic classification approaches have various disadvantages: port-based classification methods can be circumvented by choosing non-standard ports, protocol fingerprinting can be confused by the use of encryption, and current supervised learning methods for analyzing the statistical properties of network flows try to detect predefined classes, such as e-mail or FTP traffic, learned during training. In this paper, we present a novel approach to unsupervised traffic flow classification using statistical properties of flows and clustering based on a neural auto encoder. A novel time interval based feature vector construction and a semi-automatic cluster labeling method facilitate traffic flow classification independent of known traffic classes. An experimental evaluation on real data captured over a period of four months is presented. The obtained results show that 7 different classes of mobile traffic flows are detected with an average precision of 80% and an average recall of 75%.

During disasters, existing telecommunication infrastructures are often congested or even destroyed. In these situations, mobile devices can be interconnected using wireless ad hoc and disruption-tolerant networking to establish a backup emergency communication system for civilians and emergency services. However, such communication systems entail serious security risks, since adversaries may attempt to steal confidential data, fake notifications of emergency services, or perform denial-of-service (DoS) attacks. In this paper, we present SEDCOS, a secure device-to-device communication system for disaster scenarios. SEDCOS allows new users to join the network during disasters, mitigates flooding DoS attacks, and offers role revocation for detected adversaries to withdraw their permissions and exclude them from group communication. SEDCOS mitigates flooding DoS attacks and offers role revocation for detected adversaries to withdraw their permissions and exclude them from group communication. SEDCOS mitigates flooding DoS attacks and offers role revocation for detected adversaries to withdraw their permissions. We demonstrate the effectiveness of SEDCOS by large-scale network simulations.

Text spotting, i.e., the localization and recognition of text occurrences in natural images and videos, is a challenging problem in computer vision. Important applications of text spotting are: video indexing, retrieval and search, the support of blind persons in finding their way in unknown environments and reading street and shop signs for autonomously driving vehicles or car license plate recognition in traffic control systems. Recently, the use of deep convolutional neural networks as well as recognizing whole words instead of single characters led to significant progress in the field of text spotting. Furthermore, there is a tendency towards merging and unifying the single steps of the text spotting pipeline. Despite current progress, it is still difficult to recognize in-scene text in comparison to overlaid text. This work gives an overview of the state-of-the-art in text spotting and highlights current breakthroughs and trends.

We consider a data dissemination scenario in a wireless network with selfish nodes. A message available at a source node has to be disseminated through the network in a multi-hop manner. In order to incentivize a node to forward the source's message to others, a forwarding cost is paid to a forwarder by its respective receiver. In the case of multicast transmission, the cost is shared among the receivers using the Shapley value (SV). Moreover, a node may exploit the maximal ratio combining (MRC) technique to receive the message from multiple transmitting nodes. In this paper, we show that in a game theoretic framework, the optimal decision of a node for receiving the message with minimum cost can be achieved by solving a linear optimization problem. In addition, we propose an algorithm by which truthfulness is a dominant strategy for the nodes and thus, fair cost allocation is guaranteed. Simulation results show that our proposed algorithm shares the cost of data dissemination among the nodes of a network in a fair manner. Compared to previous algorithms, the proposed algorithm can reduce the total cost paid by the nodes in the network for receiving messages.

CavBase is a database containing information about the three-dimensional geometry and the physicochemical properties of putative protein binding sites. Analyzing CavBase data typically involves computing the similarity of pairs of binding sites. In contrast to sequence alignment, however, a structural comparison of protein binding sites is a computationally challenging problem, making large scale studies difficult or even infeasible. One possibility to overcome this obstacle is to precompute pairwise similarities in an all-against-all comparison, and to make these similarities subsequently accessible to data analysis methods. Pairwise similarities, once being computed, can also be used to equip CavBase with a neighborhood structure. Taking advantage of this structure, methods for problems such as similarity retrieval can be implemented efficiently. In this paper, we tackle the problem of performing an all-against-all comparison using CavBase, consisting of more than 200,000 protein cavities, by means of parallel computation and cloud computing techniques. We present the conceptual design and technical realization of a large-scale study to create a similarity database called CavSimBase. We illustrate how CavSimBase is constructed, is accessed, and is used to answer biological questions by data analysis and similarity retrieval.

Ensuring the software quality of mobile applications with respect to performance, robustness, energy consumption, security and privacy is an important problem for a growing researcher and developer community. In this paper, we present Dynalize, a Platform-as-a-Service cloud for the dynamic analysis of mobile applications. It allows researchers and developers to investigate mobile applications at runtime in a virtual device cloud and to publish the performed analyses as web services. In contrast to existing approaches, it makes use of container virtualization on top of Infrastructure-as-a-Service instances, enabling dynamic provisioning and fast deployment of dynamic analyses. A custom container layout and a novel storage solution on the virtual server layer ensures cost- and runtime-efficient large-scale analyses of thousands of apps. The applicability of Dynalize is demonstrated by a security analysis of about 6,000 Android applications. Experiments on container startup, virtual device to container throughput and different storage back ends show the feasibility of the proposed approach.

Cloud computing has become a widely used paradigm in many IT domains such as e-health. It offers several advantages to the users, e.g. elasticity, flexibility and the rapid sharing of a huge set of digital data. However, many security and privacy concerns still pose significant challenges. In particular, the most identified problem is how to enforce the user's security policy in the access control of the outsourced data. In fact, cloud environments does not provide facilities to support high level defined security policies. For instance, the swift storage component of openstack supports only fine grained access control to execute a specific action on a specific defined object. In this paper, we designed and implemented a middleware to provide high level security policies while using such swift fine grained primitives. An e-health collaborative application dedicated for remote diagnosis is used to illustrate the suggested approach.

Flocculation of yeast cells is an important phenomenon that often occurs in beverage production and applications of white biotechnology. In this paper, a novel model for cell movement and cell-cell interaction to simulate yeast cell flocculation is presented. To simulate this process with acceptable runtimes, a GPU implementation based on OpenCL and Java is described. The implementation allows us to track the cell movement in a detailed manner by a 3D visualization during execution. Experimental results indicate that the GPU implementation is up to factor of 736 faster than a multithreaded C/C++ implementation on a multi-core workstation for simulations with up to 20.000 yeast cells. Moreover, the GPU implementation requires only up to 225 milliseconds to perform simulations with up to 1.000.000 yeast cells.

The goal of the provider with respect to dynamic resource allocation in cloud computing is to maintain application performance according to service level agreements while reducing electrical power costs. To achieve this goal, we present a resource manager that optimizes a utility function expressing the trade-off between the conflicting objectives of maintaining application performance and reducing power costs. It is based on an artificial neural network (ANN) to find the best resource allocation to virtual machines that optimizes the utility function. To provide support for a potentially large number of virtual machines, we present a distributed version of the resource manager consisting of several ANNs in which each ANN is responsible for modeling application performance and power consumption of a single VM while exchanging information with other ANNs to coordinate resource allocation. Simulated and real experiments show the effectiveness of the distributed ANN resource manager over static allocation, a centralized version and a distributed non-coordinated version.

This paper presents a cross-layer self-healing approach for Cloud computing environments, based on the Complex Event Processing method. It analyzes monitored events to detect performance-related problems and performs action to fix them without human intervention. Our proposal makes use of novel analysis rules, derived from a comprehensive study of the relationships between monitored metrics across multiple Cloud layers. The results of our study are used to define and optimize the analysis rules and identify the causes of performance-related problems. The results of several experiments demonstrate the benefits of the proposed approach in terms of speeding up the analysis without affecting the quality of the diagnosis.

Cloud monitoring and analysis are challenging tasks that have recently been addressed by Complex Event Processing (CEP) techniques. CEP systems can process many incoming event streams and execute continuously running queries to analyze the behavior of a Cloud. Based on a Cloud performance monitoring and analysis use case, this paper experimentally evaluates different CEP architectures in terms of precision, recall and other performance indicators. The results of the experimental comparison are used to propose a novel dynamic CEP architecture for Cloud monitoring and analysis. The novel dynamic CEP architecture is designed to dynamically switch between different centralized and distributed CEP architectures depending on the current machine load and network traffic conditions in the observed Cloud environment.

Many Android apps have a legitimate need to communicate over the Internet and are then responsible for protecting potentially sensitive data during transit. This paper seeks to better understand the potential security threats posed by benign Android apps that use the SSL/TLS protocols to protect data they transmit. Since the lack of visual security indicators for SSL/TLS usage and the inadequate use of SSL/TLS can be exploited to launch Man-in-the-Middle (MITM) attacks, an analysis of 13,500 popular free apps downloaded from Google's Play Market is presented.

In this paper, we present an approach to provide different execution environments within Omnivore, our peer-to-peer based scheduling system for operating a pool of unused desktop computers as a desktop Grid, and integrating the desktop Grid into a larger Grid environment via the Grid Way meta-scheduler. The proposed approach is based on extending Omnivore to support virtualization technologies. A new plug-in infrastructure allows us to plug in different kinds of execution modules (including virtual machines) and select the most adequate execution environment for a job at run time. Since each job runs in its own virtual machine containing the required operating system, software and data, administrative efforts are reduced, security is improved, ease of use of the resources is enhanced, and utilization of the resources is increased. Experimental results are presented to indicate that there is a small overhead in using the proposed approach, but for long running jobs it is negligible.

Reliable video content analysis is an essential prerequisite for effective video search. An important current research question is how to develop robust video content analysis methods that produce satisfactory results for a large variety of video sources, distribution platforms, genres, and content. The work presented in this article exploits the observation that the appearance of objects and events is often related to a particular video sequence, episode, program, or broadcast. This motivates our idea of considering the content analysis task for a single video or episode as a transductive setting: the final classification model must be optimal for the given video only, and not in general, as expected for inductive learning. For this purpose, the unlabeled video test data have to be used in the learning process. In this article, a transductive learning framework for robust video content analysis based on feature selection and ensemble classification is presented. In contrast to related transductive approaches for video analysis (e.g., for concept detection), the framework is designed in a general manner and not only for a single task. The proposed framework is applied to the following video analysis tasks: shot boundary detection, face recognition, semantic video retrieval, and semantic indexing of computer game sequences. Experimental results for diverse video analysis tasks and large test sets demonstrate that the proposed transductive framework improves the robustness of the underlying state-of-the-art approaches, whereas transductive support vector machines do not solve particular tasks in a satisfactory manner.

The idea of using image and video data available in the World-Wide Web (WWW) as training data for classifier construction has received some attention in the past few years. In this paper, we present a novel incremental and scalable web-supervised learning system that continuously learns appearance models for image categories with heterogeneous appearances and improves these models periodically. Simply specifying the name of the concept that has to be learned initializes the proposed system, and there is no further supervision afterwards. Textual and visual information on web sites are used to filter out irrelevant and misleading training images. To obtain a robust, flexible, and updatable way of learning, a novel learning framework is presented that relies on clustering in order to identify visual subclasses before using an ensemble of random forests, called random savanna, for subclass learning. Experimental results demonstrate that the proposed web-supervised learning approach outperforms a support vector machine (SVM), while at the same time being simply parallelizable in the training and testing phases.

In this paper, an approach for improving the energy efficiency of infrastructure-as-a-service clouds is presented. The approach is based on performing live migrations of virtual machines to save energy. In contrast to related work, the energy costs of live migrations including their pre- and post-processing phases are taken into account, and the approach has been implemented in the Eucalyptus open-source cloud computing system by efficiently combining a multi-layered file system and distributed replication block devices. To evaluate the proposed approach, several short- and long-term tests based on virtual machine workloads produced with common operating system benchmarks, web-server emulations as well as different MapReduce applications have been conducted. The results indicate that energy savings of up to 16 percent can be achieved in a productive Eucalyptus environment.

In this paper, a novel scheduling algorithm for Cloud-based workflow applications is presented. If the constituent workflow tasks are geographically distributed - hosted by different Cloud providers or data centers of the same provider - data transmission can be the main bottleneck. The algorithm therefore takes data dependencies between workflow steps into account and assigns them to Cloud resources based on the two conflicting objectives of cost and execution time according to the preferences of the user. Our implementation is based on BPEL, an industry standard for workflow modeling, and does not require any changes to the standard. It is based on, but not limited to, the Active BPEL engine and Amazon's Elastic Compute Cloud. To automatically adapt the scheduling decisions to network-related changes, the data transmission speed between the available resources is monitored continuously. Experimental results for a real-life workflow from a medical domain indicate that both the workflow execution times and the corresponding costs can be reduced significantly.

A common approach in Infrastructure-as-a-Service Clouds or virtualized Grid computing is to provide virtual machines to customers to execute their software remotely. While giving full super user permissions eases the installation and use of a customer's software, it may lead to security issues. Providers usually delegate the task of keeping virtual machines up to date to the customer, while the customer expects the provider to perform this task. Consequently, a large number of virtual machines (either running or dormant) are not patched against the latest software vulnerabilities. The approach presented in this paper deals with this problem by helping users as well as providers to keep virtual machines up to date. Prior to the update step, it is crucial to know which software is actually outdated. While this task seems trivial, developing a solution that takes care of multiple, different software repositories and identifies the correct packages is a challenging task. The Update Checker presented in this paper identifies outdated software packages in virtual machines, even if the virtual machines are installed with different repositories. The paper presents the design, the implementation and an experimental evaluation of the approach.

One of the challenges of Infrastructure-as-a-Service Clouds is how to dynamically allocate resources to virtual machines such that quality of service constraints are satisfied and operating costs are minimized. The tradeoff between these two conflicting goals can be expressed by a utility function. In this paper, a two-tier resource management approach based on adequate utility functions is presented, consisting of local controllers that dynamically allocate CPU shares to virtual machines to maximize a local node utility function and a global controller that initiates live migrations of virtual machines to other physical nodes to maximize a global system utility function. Experimental results show the benefits of the proposed approach in Cloud computing environments.

The number of web services capable of processing multimedia data is growing. Typically, a multimedia web service realizes only a specific algorithmic processing step, such as video decoding. Thus, it is desirable to compose several web services hosted on different sites into a new value-added workflow. However, the transfer of large amounts of multimedia data within workflows based on SOAP as the prevalent communication paradigm between web services induces redundant data transfers. In previous work, we have presented a reference technique called Flex-SwA that solves this problem. However, its usage is accompanied by additional software development efforts that have to be repeated when a new service or client is implemented. In this paper, we present an aspect-oriented programming approach that significantly reduces these software development efforts. The solution allows developers to easily extend existing multimedia web services with the capability of efficient data transmission without modifying the implementations of the original services, while at the same time the advantages of SOAP web services are still maintained. Experimental results for a distributed video analysis workflow demonstrate the feasibility of the presented approach.

Live migration of virtual machines is an important issue in Cloud computing environments: when physical hosts are overloaded, some or all virtual machines can be moved to a less loaded host. Live migration poses additional challenges when virtual machines use local persistent storage, since the complete disk state needs to be transferred to the destination host while the virtual machines are running and hence are altering the disk state. In this paper, several approaches for implementing and synchronizing persistent storage during live migration of virtual machines in Cloud infrastructures are presented. Furthermore, the approaches also enable users to migrate swap space, which is currently not possible on most virtual machine hypervisors. Finally, measurements regarding disk synchronization, migration time and possible overheads are presented.

The commercial success of Cloud Computing and recent developments in Grid Computing have brought platform virtualization technology into the field of high performance computing. Virtualization offers both more flexibility and security through custom user images and user isolation. In this paper, we present an approach for combined malware detection and kernel root kit prevention in virtualized Cloud Computing environments. All running binaries in a virtual instance are intercepted and submitted to one or more analysis engines. Besides a complete check against a signature database, live introspection of all system calls is performed to detect yet unknown exploits or malware. Furthermore, to prevent that an intruder retains persistent control over a running instance after a successful compromise, an in-kernel root kit prevention approach is proposed. Only authorized and thus trusted kernel modules are allowed to be loaded during runtime, loading of unauthorized modules is no longer possible. Finally, the performance of the presented solutions is evaluated.

In this paper, a novel approach to prevent accidental or deliberate data breaches is presented. The proposed approach provides platform, network and offline security. Data is categorized as sensitive or insensitive, and the corresponding applications are isolated by using virtualization technology. Data theft or accidental loss is prevented by encrypting virtual hard disks and by introducing a multi-lane network architecture. If no connection to a corporate network is available, an offline mode handles data transfer and encryption. Authentication is managed by applying a biometric feature vector in association with a smart card setup. The approach increases security without disrupting the everyday work routines of users. An implementation based on Virtual Box and Java Card is presented. A performance evaluation of the critical components is provided.

In this paper, an approach to assign BPEL workflow steps to available resources is presented. The approach takes data dependencies between workflow steps and the utilization of resources at runtime into account. The developed scheduling algorithm simulates whether the makespan of workflows could be reduced by providing additional resources from a Cloud infrastructure. If yes, Cloud resources are automatically set up and used to increase throughput. The proposed approach does not require any changes to the BPEL standard. An implementation based on the ActiveBPEL engine and Amazon's Elastic Compute Cloud is presented. Experimental results for a real-life workflow from a medical application indicate that workflow execution times can be reduced significantly.

In this paper, we present an approach to automate the planning of an endovascular stent graft for abdominal aortic aneurysms (AAAs), which are treated with bifurcated prosthesis (Y-stents) when located close to the iliac bifurcation. During the intervention, the folded Y-stent graft — consisting of several parts — is inserted via the iliac region and expanded inside the patient's body. The first step of the proposed approach is to detect different body landmarks with a statistical method. In the next step, these landmarks are used to calculate two vascular centerlines, which provide multiplanar reformatting (MPR) slices that are used for an automatic segmentation of the artery walls. The segmented artery walls provide the manufacturer specific measures to choose an adequate bifurcated prosthesis. In a final step, the expansion of the stent is simulated in the patient's data. Results for 50 abdominal aortic aneurysm cases provided by computed tomography angiography (CTA) acquisitions are successfully verified by a virtual stenting expert.

The MapReduce pattern popularized by Google has successfully been utilized in several scientific applications. In this paper, it is investigated whether a MapReduce approach utilizing on-demand resources from a Cloud is beneficial to perform simulation tasks in the area of Systems Biology and whether it can be seamlessly integrated into a service-oriented scientific workflow framework. In particular, an Amazon Elastic Map Reduce Cloud implementation of the 13C-MFA (Metabolix Flux Analysis) Monte Carlo bootstrap approach aimed at the integration into an existing BPEL-based scientific workflow system is presented. A comparison of a 64 node MapReduce cluster with a single node computation approach reveals a total performance gain up to a factor of 14, with a total cost for on-demand resources of $11. The most critical factor in terms of performance is I/O, i.e. our application suffers from the fact that I/O operations on many small files are expensive using Amazon S3 and the Hadoop DFS.

The adoption of service-oriented architectures based on web services in industrial application domains enables the seamless integration of business software with manufacturing tasks and promises increased interoperability. To be able to invoke a manufacturing task running on a programmable logic controller (PLC) as a web service from a business application, the SOAP4PLC engine has been proposed. In this paper, an extension of the SOAP4PLC engine is presented to invoke a web service based business application from a PLC application. A use case is shown in which the extended SOAP4PLC engine is used to interface a PLC application concerned with realizing the functionality of a charging station for electrical vehicles with a web service based accounting system.

We present an interactive visualization system for the analysis of Gaussian mixture speaker models. The system exhibits the inner workings of the model intuitively by visualizing graphical representations of its parameters and of the underlying acoustical data at the same time. This enables the exploration of new modeling possibilities in the context of speaker clustering tasks.

In this paper, a fast and robust graph-based approach for boundary estimation of fiber bundles derived from Diffusion Tensor Imaging (DTI) is presented. DTI is a non-invasive imaging technique that allows the estimation of the location of white matter tracts based on measurements of water diffusion properties. Depending on DTI data, the fiber bundle boundary can be determined to gain information about eloquent structures, which is of major interest for neurosurgery. DTI in combination with tracking algorithms allows the estimation of position and course of fiber tracts in the human brain. The presented method uses these tracking results as the starting point for a graph-based approach. The overall method starts by computing the fiber bundle centerline between two user-defined regions of interests (ROIs). This centerline determines the planes that are used for creating a directed graph. Then, the mincut of the graph is calculated, creating an optimal boundary of the fiber bundle.

The Gaussian Mixture Model (GMM) is often used in conjunction with Mel-frequency cepstral coefficient (MFCC) feature vectors for speaker recognition. A great challenge is to use these techniques in situations where only small sets of training and evaluation data are available, which typically results in poor statistical estimates and, finally, recognition scores. Based on the observation of marginal MFCC probability densities, we suggest to greatly reduce the number of free parameters in the GMM by modeling the single dimensions separately after proper preprocessing. Saving about 90% of the free parameters as compared to an already optimized GMM and thus making the estimates more stable, this approach considerably improves recognition accuracy over the baseline as the utterances get shorter and saves a huge amount of computing time both in training and evaluation, enabling real-time performance. The approach is easy to implement and to combine with other short-utterance approaches, and applicable to other features as well.

Speech processing is typically based on a set of complex algorithms requiring many parameters to be specified. When parts of the speech processing chain do not behave as expected, trial and error is often the only way to investigate the reasons. In this paper, we present a research methodology to analyze unexpected algorithmic behavior by making (intermediate) results of the speech processing chain perceivable and intuitively comprehensible by humans. The workflow of the process is explicated using a real-world example leading to considerable improvements in speaker clustering. The described methodology is supported by a software toolbox available for download.

The commercial success of Cloud computing and recent developments in Grid computing have brought platform virtualization technology into the field of high performance computing. Virtualization offers both more flexibility and security through custom user images and user isolation. In this paper, we deal with the problem of distributing virtual machine (VM) images to a set of distributed compute nodes in a Cross-Cloud computing environment, i.e., the connection of two or more Cloud computing sites. Ambrust et al. identified data transfer bottlenecks as one of the obstacles Cloud computing has to solve to be a commercial success. Several methods for distributing VM images are presented, and optimizations based on copy on write layers are discussed. The performance of the presented solutions and the security overhead is evaluated.

The Business Process Execution Language for Web Services (BPEL4WS) is the de facto standard for the composition of web services into complex, valued-added workflows in both industry and academia. Since the composition of web services into a workflow is challenging and error-prone, several graphical BPEL4WS workflow editors have been developed. These tools focus on the composition process and the visualization of workflows and mainly address the needs of web service experts.To increase the acceptance of BPEL4WS in new application domains, it is mandatory that non web service experts are also empowered to easily compose web services into a workflow. This paper presents the Domain-Adaptable Visual Orchestrator (DAVO), a graphical BPEL4WS workflow editor which offers a domain-adaptable data model and user interface. DAVO can be easily customized to domain needs and thus is suitable for non web service experts.

In this paper, a software system for supporting stenting in the vascular domain is presented. The system covers all treatment phases from diagnosis to follow-up examinations. During the preoperative phase, the system supports the physician by suggesting the date and the kind (open surgery, minimally invasive) of intervention based on segmenting the patient's CT data. Therapy planning is additionally supported by a computer-aided stent simulation. Using virtual stenting, it is possible to simulate stents of different manufacturers in the preoperative CT data of the patient. As a result, it can be decided whether a chosen stent has proper dimensions and should be used during the following intervention. The intraoperative phase is supported by visualizing the selected stent from the planning phase at the requested position. After stenting, regular follow-up examinations are necessary to detect stent migration and endoleaks. These time-consuming procedures are also supported by the developed system.

BPEL is the de facto standard for business process modeling in today's enterprises and is a promising candidate for the integration of business and Grid applications. Current BPEL implementations do not provide mechanisms to schedule service calls with respect to the load of the target hosts. In this paper, a solution that automatically schedules workflow steps to underutilized hosts and provides new hosts using Cloud computing infrastructures in peak-load situations is presented. The proposed approach does not require any changes to the BPEL standard. An implementation based on the ActiveBPEL engine and Amazon's Elastic Compute Cloud is presented.

In this paper, a novel architecture for a streaming intrusion detection system for Grid computing environments is presented. Detection mechanisms based on traditional log-files or single host databases are replaced by a streaming database approach. The streaming architecture allows processing of temporal attack data across multiple sites and offers the potential for performance benefits in large scale systems, since data is processed during its natural flow and only stored as long as necessary for analysis. Two cross-site example attacks in a Grid environment and the streaming detection logic for these attacks are presented to illustrate the approach. Experimental results of a prototypical implementation are presented.

Web services are supported by almost all major software vendors, but nevertheless there is still a certain barrier that prevents a broader user community to actually use them. The barrier is the lack of appropriate clients offered in conjunction with the services. This paper presents a Web Service Browser that automatically generates a dynamic user interface when the user browses to the location of the service description and additionally handles the invocation of the service. To ease the use of the service, the browser takes care of data management by using an implementation of the Flex-SwA architecture. Results are presented to the user in a human-readable manner. When the result contains multimedia data, an audio or video player is used to present the result. Use cases demonstrate the benefits of the browser. With the Web Service Browser, web services simply become a usable component offered in the WWW.

Web services usually have functional as well as non-functional properties. Functional properties, such as the WSDL description, are usually static, whereas non-functional properties are often dynamic and thus vary over time. One of these non-functional properties are prices for using a web service or the resources it consumes. It is desirable to dynamically set prices depending on criteria such as the time of day or usage patterns and to expose pricing information in several ways. In this paper, we introduce an extended version of our previously proposed temporal policy language to handle these requirements. The extension provides the possibility of adding the exposition of validity periods to service users by weaving an attribute from the temporal policy namespace to WS-Policies. Furthermore, a schema for temporal policies and a state diagram are introduced. A use case from the area of pricing high performance computing resources is presented to demonstrate that exposing validity periods to service users enables them to automatically estimate how long they can use computing resources for a given price.

High performance computing resources are currently mainly used by computer scientists or domain experts. With the upcoming Cloud computing infrastructures and the vast amount of data available in the World Wide Web, such computing resources become interesting for end users who want to develop their own computationally demanding applications. In this paper, a service-enabled mashup editor called MIRO is presented to assist end users in the task of developing distributed applications utilizing high performance computing resources. MIRO allows the combination of popular web applications with Grid and Cloud services. The separation of the view into a user and developer view allows both the user and the developer to easily work with the editor. Two use cases are presented to show how Flickr and YouTube search can be combined with multimedia analysis services.

If legacy code has to be integrated into an application, it is often necessary to call this code available as source code written in a particular programming language or available in binary format for a particular computing platform from another programming language or from a remote machine. For this reason, wrapping code has to be developed for each source code library or binary code to be integrated. This paper presents an extensible framework that supports legacy code integration by modeling legacy code not only in a way that is programming (language) independent, but also by supporting different input and output types and bindings. This aim is achieved by the use of an integrated plug-in mechanism.

Speaker clustering is the task of grouping a set of speech utterances into speaker-specific classes. The basic techniques for solving this task are similar to those used for speaker verification and identification. The hypothesis of this paper is that the techniques originally developed for speaker verification and identification are not sufficiently discriminative for speaker clustering. However, the processing chain for speaker clustering is quite large - there are many potential areas for improvement. The question is: where should improvements be made to improve the final result? To answer this question, this paper takes a biomimetic approach based on a study with human participants acting as an automatic speaker clustering system. Our findings are twofold: it is the stage of modeling that has the highest potential, and information with respect to the temporal succession of frames is crucially missing. Experimental results with our implementation of a speaker clustering system incorporating our findings and applying it on TIMIT data show the validity of our approach.

The adoption of service-oriented architectures based on web services in industrial automation promises increased interoperability and flexibility. However, industrial automation requires real-time processing, i.e. a task has to be processed within a specific deadline, which is a large obstacle for utilizing web services in this domain.The Time-Constrained Services (TiCS) framework meets the demands of industrial automation and empowers automation engineers to develop, deploy, publish, compose, and invoke time-constrained web services. This paper presents the TiCS real-time SOAP engine for industrial PCs called SOAP4IPC. It permits the execution of web services in real-time. The architecture of the SOAP4IPC engine, implementation details, and experimental results are discussed.

The use of service-oriented architectures based on web services in the manufacturing layer of industrial enterprises yields vertical integration and promises increased interoperability and flexibility. Unfortunately, two main obstacles complicate the use of web services in the manufacturing layer. First, the hardware/software used in this layer differs from the hardware/software used in other layers. Second, the manufacturing layer is maintained by automation engineers who typically are not familiar with web services. This paper presents the first SOAP engine for programmable logic controllers to advance the use of web services in the manufacturing layer. The engine offers a low memory footprint to respect the low computational power of programmable logic controllers and allows to export web services automatically without intervention of an automation engineer.

In this paper, TrueIP--a system to prevent IP spoofing using identity-based cryptography--is presented. TrueIP is based on a new identity-based signature scheme to allow verification of an IP address without relying on a certificate or a public key infrastructure. It does not require changes or restrictions to the Internet routing protocol, is incrementally deployable, and offers protection from denial-of-service attacks based on IP spoofing. Implementation issues for practical deployment are discussed. Measurements of the TrueIP computation times for signature generation and verification are presented. Furthermore, the management overhead and bandwidth consumption to achieve proof of legitimate IP address possession and verification is compared with a standard Public Key Infrastructure approach using X.509 certificates signed by a Certificate Authority.

The video coding standard H.264 supports video compression with a higher coding efficiency than previous standards. However, this comes at the expense of an increased encoding complexity, in particular for motion estimation which becomes a very time consuming task even for today's central processing units (CPU). On the other hand, modern graphics hardware includes a powerful graphics processing unit (GPU) whose computing power remains idle most of the time. In this paper, we present a GPU based approach to motion estimation for the purpose of H.264 video encoding. A small diamond search is adapted to the programming model of modern GPUs to exploit their available parallel computing power and memory bandwidth. Experimental results demonstrate a significant reduction of computation time and a competitive encoding quality compared to a CPU UMHexagonS implementation while enabling the CPU to process other encoding tasks in parallel.

BPEL is the de-facto standard for business process modeling in today's enterprises and is a promising candidate for the integration of business and Grid applications. While BPEL works well for traditional web services, it has a number of drawbacks with respect to the more complex world of WSRF- based Grid computing, especially where security is concerned. In this paper, a solution that extends the BPEL security approach to encompass secure Grid application interactions is presented. The proposed approach is capable of handling both web service and Grid service resources and their corresponding security mechanisms. The BPEL language is extended by security-related settings. An implementation of a GSI-compliant BPEL engine that can also manage the lifetime of proxy certificates is presented.

Dedicated servers remain to be a common constituent of Grid job scheduling architectures, forcing site administrators to make compromises between administrative expenses and system reliability. Apart from requiring administrative attention, dedicated servers create single points of failure and should not be subjected to network churn. This paper presents the design and implementation of Omnivore, a fully decentralized job scheduling system, built on a peer-to-peer based meta-scheduler. Omnivore is able to cope both with node failures and network churn, eliminating the need for central administration and continuous resource availability. It is integrated into the Grid landscape (especially the Globus Toolkit 4) by means of the GridWay meta- scheduler to provide scalable distributed scheduling, replicated storage and system monitoring capabilities. Results obtained from an experimental evaluation of our implementation show that Omnivore is both scalable and resilient in the presence of node failures and network churn.

A Web service has several functional properties (e.g. its operations) and non-functional properties (e.g. quality of service and security parameters). Functional properties are usually static, whereas non-functional properties are often dynamic and thus vary over time. To describe properties with time constraints, the paper introduces WS-TemporalPolicy. WS-TemporalPolicy empowers a service developer to attach a validity period to the properties described in a WS-policy. The generation, validation, storage and retrieval, and deployment process of temporal policies is supported by the Temporal Policy Runtime Environment. Implementation issues and two use cases are presented to illustrate the use of temporal policies.

Simulating turbulence in fluids is a fascinating part of physics which requires a high amount of computational power. Since for transitional Reynolds numbers each simulation run can be performed on a single contemporary CPU, turbulence studies are ideally suited for distributed computing where each node performs a simulation for a single initial condition. The approach presented in this paper makes use of unused computational power by integrating a dynamically changing set of possibly unreliable desktop PCs into a grid infrastructure of attentively administered dedicated cluster resources. The basic idea is to use peer-to-peer (P2P) technology for managing the set of computers and develop a "bridge" to interface the P2P network with a grid meta-scheduler which in turn interfaces with the grid middleware. This eliminates the need for central administration and continuous resource availability. It provides distributed scheduling, replicated storage and system monitoring capabilities. Experimental results obtained from an evaluation of our implementation show that our approach is both scalable and resilient in the presence of node failures and network churn.

This paper presents a grid browser as a familiar environment for accessing a service-oriented grid. To relieve the service developer from developing a graphical user interface and a grid service client, and to relieve the service provider from installing and maintaining a grid portal, the proposed grid browser can be used instead of a grid portal. A grid browser is a Web browser that renders WSDL files similar to HTML files. Surfing to the location of the WSDL description of a grid service results in the automatic generation of a graphical frontend in the grid browser which a user can easily fill out to invoke a service or submit a job. Furthermore, data transfers are integrated into the grid browser by integrating an implementation of the Flex-SwA architecture, such that data transmission and service invocation can be done efficiently in a single step. An implementation of the proposed grid browser to improve the usability of a service-oriented grid is presented. A use case points out the advantages of the grid browser.

This paper suggests to seamlessly adopt a service-oriented architecture based on Web services throughout an industrial enterprise as a standardized, homogeneous communication backbone, from the business layer down to the manufacturing layer. Since manufacturing processes typically have time constraints, especially real-time constraints, particular attention has to be paid to the description of such time constraints within a Web service and the timely execution of time-constrained Web services within the proposed infrastructure. Furthermore, an outline of the time-constrained services (TiCS) framework - a framework which empowers automation engineers to develop, deploy, publish, compose and invoke time-constrained services - and a prototypical implementation of two main components of the TiCS framework, namely the TiCS Wizards and the TiCS Real-time Repository, are presented.

The adoption of service-oriented architectures based on Web services in industrial automation promises increased interoperability and flexibility. The orchestration of existing Web services to workflows is a challenging task which is complicated by the fact that manufacturing processes have time constraints, especially real-time constraints. This paper presents the time-constrained services (TiCS) Modeler which supports the assisted orchestration of BPEL4WS workflows with time constraints. The presented prototypical implementation is based on a formal derivation of the time constraints of a workflow.

The Grid computing paradigm is aimed at providing seamless access to different kinds of resources, such as compute clusters, data, special appliances and even people. Like most complex IT systems, Grid middleware systems exhibit a number of security problems, and there will always be attacks that are unknown and can circumvent even the best security measures and intrusion detection systems. This creates the requirement that Grid environments should be equipped with intrusion tolerance mechanisms as well as with the traditional intrusion prevention and intrusion detection mechanisms. In this paper, we present a new intrusion tolerance approach which improves the security of stateful WSRF Grid servers against stealth attacks. The proposal is based on a novel server rotation strategy utilizing paravirtualization to close attack windows for stateful service-oriented Grid headnode servers. A flexible plugin based rotation manager deals with the complex issue of stateful connections to the Grid server, and a database connector is utilized to detach service state from the rotating functional components of the Grid server. A prototypical implementation based on the Globus Toolkit 4 is presented.

Workflows of web services orchestrated by the Business Process Execution Language (BPEL) have been successfully used in many business applications. Although these technologies were not originally designed for multimedia processing, they offer advantages to speed up the development of distributed multimedia analysis applications by allowing the composition or reconfiguration of existing services. However, in the case of service-oriented distributed video content analysis, a huge amount of binary data has to be transferred between different services. As a consequence, service orchestration based on BPEL leads to a performance bottleneck due to indirect message and data transport: the workflow engine receives results (which are potentially very large) from finished services and passes them to a subsequent service. In this paper, we present two novel approaches based on our previously developed Flex-SwA framework to model the binary data transmission between services in BPEL workflows. The proposed approaches circumvent the performance bottleneck at the orchestrating engine and provide efficient possibilities to transfer large data amounts as well as large data units. The first approach models the data flow in BPEL; the services exchange data directly. The second approach models the data flow outside of the BPEL engine and shifts it completely to the Flex-SwA framework. Experimental results for a video analysis workflow demonstrate the advantages of the proposed approaches.

In this paper, an approach for managing behaviour trust of participants in Grid computing environments is presented. By considering the interaction process among participants in Grid environments similar to an industrial production process, we argue that through the use of statistical methods of quality assurance it is possible to monitor the behaviour of Grid participants and discover deviations in order to assess the behaviour trust of the participants.

In this paper, we present a fast and robust algorithm for centerline extraction in blood vessels. The algorithm is suitable for catheter simulation in CT data of blood vessels. It creates an initial centerline based on two user-defined points (start- and endpoint). For curved vessel structures, this initial centerline is computed by Dijkstra's shortest path algorithm. For linear vessel structures, the algorithm directly connects the start- and the endpoint to get the initial centerline. Thereafter, this initial path will be aligned in the blood vessel, resulting in the vessels centerline (i.e. an optimal catheter simulation path). The alignment is done by an active contour model combined with polyhedra placed along it. Results of the proposed centerline algorithm are demonstrated for CTA with variations in anatomy and location of pathology.

There is no universal approach for the routing problem in mobile ad hoc networks: Depending on mobility or data traffic patterns, resource constraints, environmental conditions or delivery requirements in terms of reliability, latencies or available bandwidth, protocol designs and optimizations focus on different aspects. Changes in the usage context have an effect on the suitability or efficiency of the utilized routing protocol, such that switching to another protocol intended to work in the altered environment might be desired. In this paper, we present a novel approach to automatically configure and adaptively switch between routing protocols at runtime by coordinating local preferences using a fault-tolerant distributed voting algorithm. It has been implemented as an extension to a routing software running on Linux and Unix-like systems.

In this paper, we present the design and implementation of a filtering approach for XML documents which is based on XPath expressions and Aspect-Oriented Programming (AOP). The class of XPath expressions used allows for branching, wildcards and descendant relationships between nodes. For the embedding of simple paths into XPath expressions, a dynamic programming approach is proposed. The AOP paradigm, which provides a means for encapsulating crosscutting concerns in software, is introduced to integrate the filtering approach in the broader context of event-based parsing of XML documents using SAX.

Change detection in XML documents is an important task in the context of query systems. In this paper, we present CustX- Diff, a customizable change detection approach for XML documents based on X-Diff [6]. CustX-Diff performs the change detection operation simultaneosly with the XPath based filtering of XML document parts. The class of XPath expressions used is the tree patterns subset of XPath. For the embedding of simple paths into XPath expressions during the difference operation, a dynamic programming approach is proposed. Comparative performance results with respect to the original X-Diff [6] approach demonstrate the efficiency of the proposed method.

Speaker clustering is the task of assigning a unique label to all speech segments in a video uttered by the same speaker. There are two key challenges: processing speed and robustness in the presence of noise. In this paper, we present an approach to significantly improve the processing speed of a hierarchical speaker clustering algorithm by using the earth mover's distance (EMD) as the distance measure. By extending the well-known MIXMAX speaker model such that the EMD can be applied, noise robustness is achieved. Experimental results show that the runtime of the proposed EMD approach decreases by more than a factor of 120 compared to a likelihood ratio based distance measure while the clustering performance remains nearly the same

The performance of video analysis and indexing algorithms strongly depends on the type, content and recording characteristics of the analyzed video. Current video indexing approaches often make use of thresholding techniques or supervised learning which requires labeling of possibly large training sets. Furthermore, the application of the same training model or parameters might lead to a suboptimal indexing accuracy for a given video. In this paper, we propose to use a novel self-supervised learning framework for robust video indexing to address this issue. Based on an initial classification result for a given video, the best features are selected by adaboost and are then used to train SVM (support vector machine) classifiers, all on the given video. Finally, a specialized ensemble of classifiers is employed for the given video for decision making. Experimental results show that a state-of-the-art video cut detection approach can be significantly improved by the self-supervised learning approach

SOAP is the standard protocol for message exchange in Web service environments. As an XML-based protocol, SOAP is not suitable for the transmission of large amounts of binary data. This fact has been addressed by the SOAP messages with attachments specification, which regulates the transfer of a SOAP message together with an arbitrary number of binary attachments composed within a MIME multipart/related message. Although this leads to a reduction of transmission overhead, Web service communication using SOAP messages with attachments still lacks communication and processing flexibility. In this paper, we present a novel and more flexible way of handling attachments in SOAP-based Web service environments. In contrast to SOAP messages with attachments, our approach offers message forwarding without additional communication cost and demand-driven evaluation and transmission of binary data, thus providing the opportunity to save time by overlapping service execution and data transmission

In service-oriented grid computing, great emphasis is placed on platform independence and cross-platform interoperability, at the price of a performance overhead incurred by the middleware and the high level programming languages typically utilized for developing software services. Reconfigurable hardware has been used in many areas of computing to improve the performance of applications by realizing performance critical parts in hardware. Typically, this is done in an application specific way, creating a custom solution for the project at hand for a specific reconfigurable hardware system. In this paper, we introduce a generic architecture in which grid services can be dynamically transformed and run on reconfigurable hardware in a dynamic environment in which different types of reconfigurable hardware systems are present. Three approaches - static design time integration, dynamic run time integration and transparent dynamic run time integration -are presented for integrating such on-demand "hardware services" into a service-oriented grid environment

Retrieving information about the occurrences of persons in a video is an important task in many video indexing and retrieval applications. The problem is to answer the question "In which shots and scenes does person X appear?". In this paper, we present an automatic video annotation system with respect to a person's appearance based on state-of-the-art algorithms for face detection, tracking and recognition. In contrast to many related approaches, knowledge about the persons in a given video is not assumed in advance. Adaboost is employed after an initial clustering of faces to select the best features describing a person's face. These features are then used to train new classifiers based only on the faces extracted from the video under consideration. Several possibilities to train Adaboost and support vector machine (ensemble) classifiers directly on a video are compared. Finally, experimental results demonstrate the effectiveness of correcting in-plane face rotation and of the employed self-supervised learning method

Detection and localization of text in videos is an important task towards enabling automatic content-based retrieval of digital video databases. However, since text is often displayed against a complex background, its detection is a challenging problem. In this paper, a novel approach based on fuzzy cluster ensemble techniques to solve this problem is presented. The advantage of this approach is that the fuzzy clustering ensemble allows the incremental inclusion of temporal information regarding the appearance of static text in videos. Comparative experimental results for a test set of 10.92 minutes of video sequences have shown the very good performance of the proposed approach with an overall recall of 92.04% and a precision of 96.71%

The accurate recognition of text that appears in images/videos using analytical character recognition methods is often very difficult, despite the fact that the text might be correctly localized, segmented and binarized. This is mainly due to changing features of the text such as various fonts, or noise factors embedded in the image which are inherited from the complex background. In this paper, we treat the problem of comparing text images for content-based retrieval purposes, by presenting a holistic approach to this issue. First, the shape of text is represented by estimating the salient points in the text image. Then, alignment shape methods are used to establish the correspondence of the salient points. Finally, a measure is suggested to compute the dissimilarity between two text images based on the generated correspondence. Empirical evaluation of the proposed holistic comparison method has demonstrated its very good performance

In typical on demand grid computing scenarios, services from different organisations can potentially run in the same Web service engine on a single grid node, making intra-engine service security vital for any production system. In this paper, a solution to the problem of intra-engine inter-service security for ad hoc grid environments based on WSRF is presented. To ensure that only authorized access to grid services is possible from within other services' code, a dynamic group enabled sandboxing approach within Apache Axis is proposed to protect dynamically deployed grid services. It relies on the features provided by a hot deployment service developed for ad hoc grids. A prototypical implementation of the hot deployment service and the intra-engine service security approach based on the Globus Toolkit 4 (GT4) is used to demonstrate the feasibility of our approach.

In this paper, we present several extensions of our previous work on combining the multidimensional scaling technique and the reorderable matrix method to visualize time-varying matrices: (a) the Sammon mapping is employed as another dimension reduction technique that in contrast to multidimensional scaling pays more attention to small distances; (b) a novel method for the interactive colored visualization of covariances/correlations is presented; (c) the K-means clustering algorithm is used and its results are directly visualized in the mentioned dimension reduction plots; (d) a novel view, namely the visualization of the timely evolution of the cluster membership, is proposed. The latter is based on calculating accumulated adjacency matrix that gathers the information regarding membership of objects in clusters for each point of time. The color visualization of this matrix allows the investigation of changes in cluster memberships and possible outliers, i.e. objects that change clusters frequently. Results are presented by visualizing sensitivity matrices generated during the simulation of metabolic network models.

MOTIVATION: Supporting the evolutionary modeling process of dynamic biochemical networks based on sampled in vivo data requires more than just simulation. In the course of the modeling process, the modeler is typically concerned not only with a single model but also with sequences, alternatives and structural variants of models. Powerful automatic methods are then required to assist the modeler in the organization and the evaluation of alternative models. Moreover, the structure and peculiarities of the data require dedicated tool support. SUMMARY: To support all stages of an evolutionary modeling process, a new general formalism for the combinatorial specification of large model families is introduced. It allows for automatic navigation in the space of models and excludes biologically meaningless models on the basis of elementary flux mode analysis. An incremental usage of the measured data is supported by using splined data instead of state variables. With MMT2, a versatile tool has been developed as a computational engine intended to be built into a tool chain. Using automatic code generation, automatic differentiation for sensitivity analysis and grid computing technology, a high performance computing environment is achieved. MMT2 supplies XML model specification and several software interfaces. The performance of MMT2 is illustrated by several examples from ongoing research projects. AVAILABILITY: http://www.simtec.mb.uni-siegen.de/ CONTACT: [email protected].

In this paper, we present a solution to the probl of dynamically deploying grid service factories onto computing nodes running an implentation of the Open Grid Services Infrastructure (OGSI). By providing a non-intrusive Hot Deployment Service (HDS), we extend the service-oriented grid computing paradigm, as it is defined by the Open Grid Services Architecture (OGSA), to provide a more dynamic ad hoc grid environment. Service-oriented grid middleware utilizing the HDS enables organizations or interorganizational communities to form an ad hoc grid to harness unused and scattered resources of an existing IT-infrastructure. The availability of the HDS also improves the capabilities to manage existing grid systs based on the Globus Toolkit 3, which is a vital requirent for the adoption of service-oriented grid systs in production environments.

We present a novel approach to visualize time-varying matrices. This approach is based on combining multidimensional scaling and the reorderable matrix method. An adapted version of multidimensional scaling which allows the construction of similarity plots for columns/rows of time-varying matrices is proposed. In addition, we have extended the reorderable matrix method to allow the visual exploration of time-varying matrix data in a tabular form for being able to verify the results of MDS and possibly discover new patterns in data. The benefits of our approach are illustrated by showing visualizations of sensitivity matrices generated during simulations of metabolic network models.

Tracking superimposed text moving across several frames of a video is relevant for exploiting its temporal occurrence for effective video content indexing and retrieval. In this paper, an approach is presented that automatically detects, localizes and tracks text appearing in videos. The proposed approach consists of two steps: (1) unsupervised text detection and localization in each Nth frame to monitor new text events, i.e. text appearing in a video for the first time; (2) text tracking within a group of pictures (GOP) using MPEG motion vector information extracted directly from the compressed video stream. Comparative experimental results for a set of videos are presented to show the benefits of our approach.

The segmentation of video sequences into shots is the first step towards video content analysis. Two kinds of shot boundaries can be distinguished: abrupt scene changes ("cuts") and gradual transitions. In this paper, we present a technique to reduce the error rates of cut detection algorithms based on pixel-wise or histogram-based frame difference metrics when operating directly on compressed MPEG video data. The proposed approach, called "frame difference normalization" (FDN), intends to eliminate the effects of a specific frame pattern in MPEG streams responsible for causing such errors. Experimental results will be presented to demonstrate the benefits of our proposal and its superiority over a more general noise filter. Furthermore, the proposed method is not limited to a particular algorithm but it is applicable to an entire class of cut detection algorithms.

A system allowing to control an electrically powered wheelchair without using the hands is introduced. HaWCoS -- the "Hands-free" Wheelchair Control System -- relies upon muscle contractions as input signals. The working principle is as follows. The constant stream of EMG signals associated with any arbitrary muscle of the wheelchair driver is monitored and reduced to a stream of contraction events. The reduced stream affects an internal program state which is translated into appropriate commands understood by the wheelchair electronics. The feasibility of the proposed approach is illustrated by a prototypical implementation for a state-of-the-art wheelchair. Operating a HaWCoS-wheelchair requires extremely little effort, which makes the system suitable even for people suffering from very severe physical disabilities.

In this paper, we present the AnimalScript visualization language. This scripting language uses the flexibility of the Animal system and provides many additional new graphic primitives and animation effects that go beyond the traditional Animal GUI features.AnimalScript can easily be configured by changing the content of a registration file. Users may also have multiple registration files, as AnimalScript will always use the first registration file it finds. AnimalScript can easily be extended with additional features without needing to read, let alone change, any existing code.

The design of an optimal control strategy for a hybrid system is a matter of growing interest in computational engineering. The solution of optimization problems in most engineering disciplines often requires efficient parallel optimization algorithms to solve these kinds of problems in reasonable time. Instead of introducing parallelism to selected components of an existing sequential algorithm, the algorithm proposed in this paper is aimed at utilizing the available computational resources efficiently throughout the course of the optimization. To assure a certain level of efficiency the algorithm can be adapted to the available resources and the dimension of the problem to be solved. The features of this inherently parallel algorithm are described, and the parallel performance is analyzed by means of a scalability analysis. To demonstrate the use of the algorithm for the solution of optimization problems in computational engineering, two problems from groundwater engineering are solved.

The presence of stochastic or deterministic trends in economic time series can be a major obstacle for producing satisfactory predictions with neural networks. In this paper, we demonstrate the effects of nonstationarity on neural network predictions using the time series of the mortgage loans purchased in the Netherlands. We present different preprocessing techniques for removing nonstationarity, and evaluate their properties by producing multi-step predictions using a linear stochastic forecasting model and a neural network. The results indicate that detecting nonstationarity and selecting an appropriate preprocessing technique is highly beneficial for improving the prediction quality.

In this paper, we present Animal, a new tool for developing animations to be used in lectures. Animal offers a small but powerful set of graphical operators. Animations are generated using a visual editor, by scripting or via API calls. All animations can be edited visually. Animal supports source and pseudo code inclusion and highlighting as well as precise user-defined delays between actions. The paper evaluates the functionality of Animal in comparison to other animation tools.

Algorithm animation has received much interest over the last few years.In this paper, we discuss the experiences gained in integrating animations into introductory computer science courses with large audiences of more than 200 students.After providing a short introduction to the animation tool we developed, we describe why and how we used animations in our lectures and present some example animations.

The fitness landscape of the graph bipartitioning problem is investigated by performing a search space analysis for several types of graphs. The analysis shows that the structure of the search space is significantly different for the types of instances studied. Moreover, with increasing epistasis, the amount of gene interactions in the representation of a solution in an evolutionary algorithm, the number of local minima for one type of instance decreases and, thus, the search becomes easier. We suggest that other characteristics besides high epistasis might have greater influence on the hardness of a problem. To understand these characteristics, the notion of a dependency graph describing gene interactions is introduced. In particular, the local structure and the regularity of the dependency graph seems to be important for the performance of an algorithm, and in fact, algorithms that exploit these properties perform significantly better than others which do not. It will be shown that a simple hybrid multi-start local search exploiting locality in the structure of the graphs is able to find optimum or near optimum solutions very quickly. However, if the problem size increases or the graphs become unstructured, a memetic algorithm (a genetic algorithm incorporating local search) is shown to be much more effective.

In this paper, a fitness landscape analysis for several instances of the quadratic assignment problem (QAP) is performed, and the results are used to classify problem instances according to their hardness for local search heuristics and meta-heuristics based on local search. The local properties of the fitness landscape are studied by performing an autocorrelation analysis, while the global structure is investigated by employing a fitness distance correlation analysis. It is shown that epistasis, as expressed by the dominance of the flow and distance matrices of a QAP instance, the landscape ruggedness in terms of the correlation length of a landscape, and the correlation between fitness and distance of local optima in the landscape together are useful for predicting the performance of memetic algorithms-evolutionary algorithms incorporating local search (to a certain extent). Thus, based on these properties, a favorable choice of recombination and/or mutation operators can be found. Experiments comparing three different evolutionary operators for a memetic algorithm are presented.

A memetic algorithm (MA), i.e. an evolutionary algorithm making use of local search, for the quadratic assignment problem is presented. A new recombination operator for realizing the approach is described, and the behavior of the MA is investigated on a set of problem instances containing between 25 and 100 facilities/locations. The results indicate that the proposed MA is able to produce high quality solutions quickly. A comparison of the MA with some of the currently best alternative approaches-reactive tabu search, robust tabu search and the fast ant colony system-demonstrates that the MA outperforms its competitors on all studied problem instances of practical interest.

CARDWATCH, a database mining system used for credit card fraud detection, is presented. The system is based on a neural network learning module, provides an interface to a variety of commercial databases and has a comfortable graphical user interface. Test results obtained for synthetically generated credit card data and an autoassociative neural network model show very successful fraud detection rates.

The prediction of the volatility of financial time-series is very important for the evaluation and pricing of options and the development of option trading strategies. In this paper, a neural network for predicting the volatility of the German Bund future is presented. Its performance is compared to that of a nonlinear GARCH model.

The replication of data objects in a large computer network is a difficult task which cannot be approached by simply employing the techniques used in small networks, because the high replication factors possible raise new issues which must be addressed. The paper presents a solution to the problem of managing replicas in a large scale environment. The solution is based on a multi-level quorum algorithm for maintaining the consistency of replicas, a probabilistic addressing mechanism for efficiently locating replicas in the system and an efficient scheme for handling dynamic changes in the number of replicas. The feasibility of this approach is demonstrated by presenting performance measurements in a simulated network.>

Neither low-level mechanisms such as semaphores nor higher-level mechanisms such as path expressions provide a simple means of solving synchronisation problems involving the scheduling of processes or classes of processes according to different priorities. This paper presents a new set of primitives which are easy to use and simple to implement. Their use is described in terms of the familiar reader–writer problem and the general scheduling problem involving arbitrary levels of priority with support for pre-emption and shared access by certain process classes. An efficient implementation, which reduces to a minimum the number of calls required to the process scheduler, is then described.