Adaptive Voltage Scaling (AVS) is a well-known technique in ASICs. However, the requirements for its application in FPGAs are quite different. AVS uses run-time knowledge to determine a favorable voltage/frequency operating point. There are many suggestions in the literature for the sensors to be used to gain this run-time knowledge. However, the specific placement and routing of each sensor instance is either ignored or manually defined by hard macros or directed routing. In this paper, we propose a new approach to sensor placement. We are able to omit strict constraints like directed routing and give more freedom to the EDA tool without compromising sensor accuracy by using a proper sensor calibration. As a result, we are able to avoid placement and routing conflicts between sensors and application design. This allows the use of larger sensors with significant routing delay. An extensive experimental evaluation proves the validity of our approach.

The Message Queuing Telemetry Transport (MQTT) protocol is a de-facto standard for machine-to-machine (M2M) communication in the Internet of Things (IoT) domain. However, since it is limited to TCP at transport layer, it is not possible to run MQTT over UDP, which would be desirable for real-time capable publish/subscribe. MQTT for Sensor Networks (MQTT-SN), which is a lightweight version of MQTT, allows the use of UDP, however, is strongly coupled to MQTT by the current specification and cannot be operated independently of it. In this work, we therefore propose a standalone broker implementation of MQTT-SN. In initial measurements, we can show that our approach is much more performant than the approach described in the specification and that our implementation compares very well with the protocols CoAP Pub/Sub and MOTT.

Time-Sensitive Networking (TSN) disrupts realtime communication technology by making IEEE Ethernet realtime-capable. For time-triggered, hard realtime traffic, TSN provides standardized mechanisms to reserve communication paths as well as individual transmission time slots for data frames. By leveraging these means in a precomputed network schedule, TSN allows for bounded end-to-end delays and minimal jitter. Not being part of the IEEE standard, corresponding scheduling algorithms are an active field of research. Unfortunately, due to differing model assumptions, evaluation setups, and key metrics, a fair comparison of schedulers is impossible so far. In this paper, we present a systematic and reproducible approach to benchmark TSN schedulers. First, we provide a scheduler taxonomy that enables to cluster schedulers by their characteristics. Second, we analyze interactions of input parameters and scheduler results to derive a benchmarking parcour for quantitative comparisons. Finally, we use the approach to benchmark existing schedulers and show subtle interaction effects. This way, our approach enables—for the first time-comparability between schedulers, fueled by the public availability of our benchmarking scenarios.

The growing capabilities of wireless communication technologies such as WLAN-based systems enable their successful adoption in an ever-increasing range of applications. Particularly in the domain of the internet of things, the trend towards wireless interconnection is driven by its improved scalability and flexible, low-cost deployment compared to wired systems such as Ethernet. To handle the growing complexity of their design, developers need a process for early performance evaluation of these communication-intensive networked embedded systems. For this, established network simulation frameworks like ns-3 and ${OMNeT}++$ are often utilized. These frameworks require developers to model every part of their system, including the protocol stack and the target application they are developing. From the developers’ perspective, a workflow that would enable them to use real target software as much as possible is desirable. The available solutions to couple real target software with a network simulation, however, often exhibit several limitations. These range from a lacking detail level of the channel simulation to the requirement of access to the source code of every part of the system under design, and often prevent adoption of these coupling solutions. In this paper, we propose a data interface for the established simulator ns-3 that allows coupling to the Linux kernel at a very low level of the protocol stack, namely the lower WLAN MAC layer. This interface allows the performance evaluation of WLAN-based systems, in which real target code of almost the entire software stack can be used. With such an interface, developers can write real target software and test it under different wireless network scenarios and channel conditions, powered by a simulation using ns-3. We show the practicality of our interface by comparing it to a simulation solely performed in ns-3, as well as to another framework for coupling real target software to a channel simulation.

Approximate or inexact arithmetic is a promising approach towards lower power consumption for applications that can tolerate a certain amount of imprecision. As human perception is limited in its precision, this applies to image and audio processing. Beyond, other applications like neuronal networks or AI processing can benefit from such arithmetic as well, as they are inherently tolerant to a certain amount of inaccuracy. One of the most critical components of arithmetic circuits regarding power, delay, and area are multipliers. Various sophisticated approaches towards approximate multipliers are already published for ASICs. However, such ASIC approaches are under-performing in conjunction with the specific Lookup-Table (LUT)-based design of FPGAs. As FPGAs gain in importance for applications like signal processing, there is a substantial lack of approximate design methodology for FPGAs. We propose an approach towards approximate signal processing that is specifically tailored towards the LUT-based hardware of FPGAs. It allows for significant performance improvements while lowering the energy demands. While introducing an insignificant average relative error of just 0.14%, we achieve a 45.9% area reduction in terms of LUTs while decreasing the delay by 30.6% compared to the Xilinx Vivado multiplier IP core. Our proposed design is open source and available at https://github.com/niemann-c/approx-mult-for-fpga.

Due to the increasing complexity of distributed embedded systems, the development process demands early substantiation of design decisions for the targeted system. Virtual prototypes enable an early validation of hardware/software systems and they have proven to be very beneficial for target software development. However, these prototypes often do not include models for inter-device communication aspects of distributed systems. Especially the use of wireless communication has a significant impact on the overall system behavior and performance. This paper introduces a SystemC model of an essential part of IEEE 802.11 Wireless LAN communication, the Enhanced Distributed Channel Access (EDCA) protocol. In comparison to widely used network simulations, the modeling with SystemC promises an easier integration into existing virtual prototypes. EDCA is the main part of the lower medium access layer (MAC) of IEEE 802.11 and interfaces to the physical-layer communication as well as to non-time-critical higher MAC-layer functions. We conducted experiments with an additional simple physical communication channel model to compare our solution in terms of precision and performance to the well-established network simulator ns-3. We show that our model differs by $\approx 0.3 \%$ of retransmitted frames and an accompanied difference of the overall simulated transmission time of $\approx 0.45 \%$. It meets the overall expectations for the stochastic EDCA protocol. Moreover, the presented model has a significantly better simulation performance with $3 \cdot 10^{-5}$ s wall-clock time per frame, which is at least two orders of magnitude better than with the compared ns-3 simulation.

Future applications of the Industrial Internet of Things will increasingly depend on the timely exchange of information and data. Time-Sensitive Networking, which is currently being standardized, therefore extends switched Ethernet networks by realtime communication capabilities offering deterministic message delays. In its time-triggered communication variant, a network path needs to be planned for each realtime data flow and corresponding transmission time slots have to be reserved on the network links along that path. In this paper, we present the first joint mathematical model for path routing and time slot scheduling that combines multicast support with an individual transmission scheduling on each link along a flow's path. We provide a formalization as an Integer Linear Programming (ILP) problem and discuss several optimizations to significantly reduce the ILP solver's runtime without affecting the solution quality. In a thorough evaluation, we show the applicability of our approach and analyze the effects and trade-offs of different objective functions.

In the age of digitalization, data protection plays an important role. Data is created, modified and shared over the Internet between data owners and data users. A key issue in this context is the respect of data sovereignty. The International Data Spaces (IDS) Reference Architecture has been developed in order to preserve data sovereignty. In this regard, Internet of Things (IoT) devices, such as sensors, play an important role for providing data. However, an IoT device is not capable of being integrated directly into the IDS by default. We propose an approach to enable the IDS for vendor independent IoT devices, using an open interoperability standard for the IoT specified by the Open Geospatial Consortium (OGC). This allows data owners to benefit from providing their own data while retaining control over it.

Today’s operating rooms consist of a multitude of medical devices. Nowadays, the interconnection of these devices is manufacturer-dependent. To overcome this lack of interoperability, the new IEEE 11073 Service-oriented Device Connectivity (SDC) family of standards has been developed. Plenty of use cases take advantages from the interconnection of medical devices among each other, based on SDC. In this paper, we focus on time-critical use-cases, like the activation of a high-frequency surgical device to emit power at the instrument triggered by a networked foot switch. The execution of such remote control operations, are dependent on deterministic data transmission. To fulfill these requirements we propose the usage of the new Time-sensitive Networking (TSN) standards IEEE 802.1. The combination of SDC and TSN enables semantic interoperability and hard real-time communication at the same network, based on standards. We provide a new approach to automatically configure the underlying TSN network by means of the medical devices’ self-description according to the SDC family of standards. Our approach shows that the required information for configuring the TSN network can be extracted from the self-description.

Clock (or time) synchronization is essential for many applications in the Industrial Internet of Things (IIoT). Hence, it is a vital research field and important field of standardization ambitions. The most accurate protocols like PTP and gPTP need specialized hardware to reach their maximum precision. Without this hardware, they cannot compensate massive packet delays. It was shown that approaches based on linear programming (LP) can mitigate this problem. However, changes in the clock frequency lead to nonlinear clocks, which are not well compensated by LP-based approaches. As a consequence, we propose the SLMT approach that uses LP, multicasts, and temperature compensation for time synchronization. To the best of our knowledge, SLMT is the first synchronization approach that combines LP and one-way exchange or multicasts, respectively. Consequently, it is efficient regarding the number of messages. Furthermore, to the best of our knowledge, SLMT is the first synchronization approach that combines LP with a temperature compensation in order to mitigate LP's conceptual drawback with nonlinear clocks. In an extensive evaluation and comparison to many state-of-the-art approaches, we show that SLMT outperforms these approaches, especially under harsh conditions like rapid temperature changes and unknown non-negligible network delays.

Industrial Internet of Things (IIoT) applications benefit from the knowledge of the device and user positions in a manifold way. Reliable indoor navigation combined with IIoT enables Location-Based-Services (LBS) such as assistance functions of moveable actuators. A crane which follows its operator can significantly increase the efficiency of the process. Safety mechanisms are also enhanced by positioning information. For example, exclusion areas where only automated devices are operating can be implemented. In this paper a novel localization framework is introduced to fuse sensor data from either absolute or relative positioning sources. The core of the framework is an Extended Kalman-Filter (EKF) architecture that is able to handle data from several different sources. Each localization source needs to fulfill requirements regarding data representation and structure defined by the framework, e.g., current state and variance. The approach is verified in a real world scenario with two different sensor types as information sources: Ultra Wide Band localization and Pedestrian Dead Reckoning. We show that the combination of these technologies improves the localization accuracy and evaluate advantages and drawbacks of this approach.

Tracking of goods, containers, and vehicles in harbors is a challenging task because seaports typically are in a secluded area with limited networking capability. Existing solutions use the combination of RFID tagging and Wireless Sensor Networks (WSN). A harbor is considered a harsh industrial environment with metallic components and surfaces. These conditions influence the wireless networking performance. In addition, harbors' areas vary from 500 ha to 7500 ha. Hence, the coverage range of wireless systems and the exposition to interference are considered. LoRa (Long Range) technology becomes a promising solution among other Low Power Wide Area Networks (LPWAN). Therefore, we investigate the LoRa technology to locate and track assets in harbors. Through ns-3 simulations on scalability, interval rate, and coverage range performance metrics, we evaluated the feasibility to use LoRa in seaports. In our experiments, we applied 1000 LoRa nodes within a radius of 2500 m to the gateway. The results exhibit a probability of successful transmission of 85 % in an interval of 300 sec.

Renewable energies lead to a decentralization of power generation but also to a destabilization of the power grid, as photovoltaic or wind turbine systems provide nearly zero inertia that is essential for a stable power grid. Leveraging consumer devices to support already present grid control systems counter the growing grid instability. These devices may adapt their power consumption continuously, rather than erratic on-off switching, to provide a grid-friendly stabilization effect. Simulations with a grid model of Continental Europe and multiple consumer load control schemes with different consumer-impact levels were performed to analyze stabilization effect and user tolerability. The results show effective stabilization by the tested consumer load control schemes during a reference incident, allowing them to be used for various devices and device groups. Additional, the simulations proved the scalability of the proposed control algorithms.

In Smart Buildings there is a large number of connected devices. Each of them is possibly vulnerable, so that an attacker could make use of a single smart device to run attacks in the Smart Building network. We propose a concept to partition the network into trust zones depending on the application layer, so that devices that are logically linked on application layer are able to communicate on MAC layer. A trusted device is used to bootstrap new devices and reconfigure existing trust zones. We restrict the communication as far as possible, so that the potential damage caused by a compromised device is limited. In principle, malicious behavior of devices could lead to an exclusion on MAC layer. The general concept is described using a typical IoT protocol stack containing IEEE 802.11s, IP, UDP/DTLS, and CoAP.

This survey addresses the question if existing realtime capable Internet technologies for wired communication can meet the requirements of future industrial IoT scenarios with a rising number of heterogeneous devices to be connected and increasing amounts of data to be exchanged in real-time. The development towards an industrial IoT is further referred to as Industry 4.0 in Germany and Industrial Internet in the USA, respectively. We first investigate selected widespread technologies at all layers of the ISO/OSI model with respect to their realtime capability, scalability and dynamic reconfiguration, standard compliance and platform complexity as well their capability to integrate non-real-time devices. On the one hand, we note that TSN technology at physical and link layer is standardized but exhibits very high platform complexity for the switches and is thus costly. Subsequently, it is hence discussed if purely software-based approaches can enable RT communication over Ethernet. Moreover, even though TSN-enabled network components can enforce real-time behavior, scheduling and routing algorithms for computing the respective network configuration are not part of the TSN standards. Those algorithms could be executed on a central SDN controller to achieve high performance and realtime capability, however, the scalability of such a centralized approach is limited by the fact that corresponding algorithms have exponential computational complexity. Hence, one of the future research directions outlined proposes to trade off distributed against centralized scheduling and routing approaches with regard to scalability and dynamic reconfiguration, real-time capability, and platform complexity. We conclude there is a need for the advancement of existing and for the development of new, possibly hybrid, real-time capable approaches that combine the advantages of centralized and distributed solution in order to meet all requirements.

Clock synchronization protocols such as the precision time protocol (PTP), which are used to synchronize components of distributed systems, are fundamental to enable timed and coordinated activities, e.g., in real-time applications within the industrial Internet of things (IIoT). In theory, PTP is able to achieve a precision on the order of nanoseconds. However, its practical accuracy remains limited by packet delay variations. In this paper, we hence present a novel approach to increase the synchronization precision of PTP. Our approach (PTP-LP) relies on PTP to obtain precise hardware timestamps taken during multiple synchronization periods. These timestamps establish the constraints for a Linear Programming (LP) solver that is used to estimate the clock differences between devices. Moreover, we propose the heuristic PTP-H that achieves comparable accuracy but is less computationally complex. We evaluate PTP-LP and PTP-H in comparison with two state-of-the-art approaches under various conditions in terms of clock stabilities and packet delay distributions. PTP-LP and PTP-H are fully compatible with existing standards and show to be in particular robust to varying packet delays. Especially, PTP-LP outperforms previous approaches in presence of a stable hardware clock and unknown non-negligible network delay, which are both realistic working conditions.

WLAN mesh networks are one of the key technologies for upcoming smart city applications and characterized by a flexible and low-cost deployment. The amendment IEEE 802.11s introduces low-level mesh interoperability at the WLAN MAC layer. On the physical layer, IEEE 802.11n introduced major improvements such as HT data rates, MIMO techniques, and frame aggregation. However, building large-scale 802.11n/s testbeds and reproducible setups is challenging and costly. On the other hand, existing attempts for down-scaling real-world setups are limited to works without support for 802.11n and 802.11s. We therefore present Mini-Mesh, a miniaturized indoor 802.11n/s testbed. Following a transmission range scaling approach, we deploy a 6×6-node mesh grid on an area of only 1 m2. We validate the applicability of our method via comparative measurements, exhibiting a deviation of less than 6 % between a scaled indoor and unscaled outdoor setup. Based on these results, we parameterize a path loss model helping us to estimate outdoor dimensions for arbitrary indoor mesh topologies.

Clock synchronization is an important issue in wired and wireless networks as a common time basis is essential for coordinated activities of nodes in distributed systems. Typical applications demanding precise synchronization are Industrial Internet scenarios having real-time constraints, Wireless Sensor Networks (WSNs) where the nodes communicate only for a short period and can sleep the remaining time to save energy, approaches based on Time Division Multiple Access (TDMA), and distributed measurements. The basic idea of our approach is to estimate all delays in a network. As a result, we can estimate the one-way delay between a reference node and all other nodes in the network. Consequently, we can use broadcast messages to synchronize the entire network. Utilizing a novel measurement method and a Java prototype implementation, we show that our approach achieves a high precision (≈ 123 μβ). Furthermore, it is highly scalable and platform independent. As our synchronization approach operates at the application layer, it is suitable for both wired and wireless networks.

This paper investigates research, made on semantic information models for building automation systems. It analyzes what information domains are covered to provide context, the vocabulary provided to describe building automation devices and functions, and how these models are structured. The intention is to find out good practices and try to identify trends, commonalities, differences, and possible next steps.

Vendor-independent interoperability is one of the key-enablers for medical devices in future operating rooms, intensive care units, and medical care in general. Using the paradigm of a Service-Oriented Architecture (SOA) is a promising approach realized by the new IEEE 11073 SDC family of standards. Standard compliant communication stacks will be used to build up systems of networked medical devices. The performance of the stack implementation is crucial for the usability in real-world medical environments. Therefore, we investigate the latency of currently available middleware stacks: SoftICE, openSDC, and OSCLib. The aim is to evaluate the suitability of the underlying concepts, understanding the communication behavior using different hard- and software platforms, and finding problems to support future development. For the latency measurements we build up a use-case independent testbed and instrument the libraries to get more information. On the one hand, our investigations substantiate the suitability of the underlying concept and the available middleware stack implementations. On the other hand, unexpected results occurred, like a strong dependency of communication latency on the combination of hardware platform, Java Virtual Machine (JVM), and JVM configuration and even a strong dependency on the intensity of exchanged data when using Java middleware implementations.

The new standard IEEE 802.11s enables vendor-independent wireless mesh networks based on the 802.11 WLAN technology. Transmission Control Protocol (TCP) is the most widespread transport protocol for reliable data delivery and still the basis for many network applications. TCP supports different mechanisms for flow and congestion control. However, designed for wired networks, it does not consider the dynamics of wireless networks and especially multi-hop wireless mesh networks. In addition, 802.11s provides own mechanisms such as Automatic Repeat Request (ARQ) for frame retransmissions to hide wireless loss from the upper layers. Being transparent to each other, retransmission schemes on both layers may interfere and operate redundantly, if not properly adjusted. We study the effects of ARQ retry limit variation on TCP throughput in a real-world multi-hop 802.11s test bed. As a result, we suggest ARQ adaptation based on the 802.11s standard's Airtime Link Metric (ALM) for path selection, serving as indicator for overall frame travel time. Our proposed approach solely relies on standard features and imposes no modifications to 802.11s or TCP.

The new standard amendment IEEE 802.11s enables low-level interoperability for future WLAN mesh networks. Support of the Hybrid Wireless Mesh Protocol (HWMP) and the Airtime Link Metric (ALM) for MAC-layer routing is mandatory. Its default distance vector routing mode facilitates scalability but also results in a limited network view per mesh node. Moreover, mesh mechanisms operate transparently to higher layers which makes the management and optimization of 802.11s networks a challenging task. Available on every standard-compliant node, ALM offers the potential to derive mesh topology information. We present AKadeMesh (Adaptive Kad-enhanced Mesh), a cross-layer approach specifically designed for 802.11s networks. It is based on the P2P protocol Kad and dynamically adapts its logical overlay to the physical mesh underlay by directly considering ALM. The resulting topology-aware P2P overlay is used to realize logical clustering for the distributed management of 802.11s networks, thereby maintaining unrestrained interoperability to the mesh standard. Our solution was implemented and evaluated in a real-world test bed. Results demonstrate its practical feasibility and verify the expected clustering benefit.

The microservice approach has created a hype in the domain of cloud and enterprise application business. Before, grown, monolithic, software has been pushed to the limits of maintainability and scalability. The microservice architecture approach utilizes the service oriented architecture together with best practices and recent developments in software virtualization to overcome those issues. One monolithic application is split up into a set of distributed services. Those are strongly decoupled to enable high maintainability and scalability. In this case an application is split up in a top down manner. In the internet of things, applications need to be put together from a set of small and independent services. Thus, creating value added services would require to freely combine services of different vendors to fully make use of the IoT's heterogeneity. Even though the direction is different, many of the requirements in microservices are similar to those of the internet of things. This paper investigates patterns and best practices that are used in the microservices approach and how they can be used in the internet of things. Since the companies using microservices have made considerations on how services have to be designed to work together properly, IoT applications might adopt several of these design decisions to improve the ability to create value added applications from a multitude of services.

Post-Dennard scaling CMOS technologies suffer from considerable degradation due to increasing electrical fields caused by the lack of further reduction of the supply voltage. This aspect of aging is widely disregarded so far and cannot be addressed at design time by adding static margins anymore. Instead, it needs to be counteracted effectively at run time over the entire device lifetime. For this purpose, dynamic runtime approaches for aging management are required, relying on detailed in formation regarding the current system state. In this paper we propose a novel aging monitoring mechanism providing that crucial information at a marginal resource overhead. The current device degradation is measured via the aging-dependent delay variation, which can be quantified in situ with built-in tests exploiting the strictly monotonic relation between supply voltage and propagation delay. Furthermore, we suggest to utilize the information gained this way for a proactive aging-aware task mapping.

Real-time systems play a major role in the realm of industrial automation. It is predicted for the number of smart interconnected devices that participate in such systems to grow significantly in the future. This development is also referred to as Industrial Internet of Things (IIoT) or Industry 4.0. The high number of devices results in highly distributed applications. Therefore, it is no longer sufficient for each device to be real-time capable. In fact, the influence of the communication on the overall timing behavior of the applications grows. Taking this into account, a variety of real-time capable Ethernet approaches called Industrial Ethernet (IE) emerged. However, the established IE solutions rely on proprietary hardware and/or non standard conform protocol adaptations. This leads to very expensive hardware and incompatibilities with other IE solutions or common Ethernet, and thus degrades the interoperability. HaRTKad describes a purely software-based P2P approach that allows deterministic communication over common Ethernet. Although it solely relies on well-known standards and allows real-time communication, it suffers from a multitude of problems. In this paper, the low network utilization, the handling of hash collisions and the traffic prioritization are revealed as the most significant limitations of HaRTKad and possible solutions to these problems are presented.

In the recent past, the development of applications and protocols for the Internet of Things (IoT) made a big leap forward. New approaches have emerged to adopt IoT technologies in the realm of industrial automation. This development is also referred to as Industrial Internet of Things (IIoT) or Industry 4.0. It is predicted for the number of smart interconnected devices participating in automation systems to grow significantly in the future. However, the industrial domain introduces new requirements for IoT technologies regarding the timeliness of interactions. Current IoT protocols, like the Constrained Application Protocol (CoAP), do not yet provide real-time behavior for the inter-device communication. In our previous work, we have already proposed a real-time extension for CoAP that enables deterministic network behavior through a TDMA-based approach. We have shown that the proposed mechanisms for time synchronization, time slot management, and access control can be realized purely software-based. However, a central instance is needed as a time server. This introduces a Single Point of Failure (SPoF) to the system, limiting the robustness and scalability of the approach. In this paper, we introduce a concept for a distributed time server for CoAP. The proposed concept includes a refined time synchronization mechanism as well as strategies to select multiple time servers and share information between them. Furthermore, the described amendments to the real-time extension are integrated into the lightweight platform-independent jCoAP communication stack and evaluated in a multi-device real-world test bed.

Current IoT protocols, like the Constrained Application Protocol (CoAP), do not yet provide real-time behavior for the inter-device communication. In this paper, we propose a real-time extension for the CoAP standard that defines interfaces for the time synchronization among nodes and the time slot management. These interfaces enable a controlled exclusive network access based on a Time Division Multiple Access (TDMA) approach. With this extension, it is possible to realize access control on the application layer without the modification of lower layer protocols. The described interfaces are prototypically implemented within the jCoAP communication stack and evaluated in a multi-device real-world testbed. In our prototype, we used established algorithms for the time synchronization. The results, reveal the weaknesses of the chosen synchronization algorithm. However, the interface definition allows the usage of more accurate algorithms.

In industrial automation environments, networks providing a reliable and timely data delivery are required. Fulfilling this need, Industrial Ethernet (IE) systems have established as an important networking technology in many application areas. Although there are several IE solutions on the market, all of these systems have notable drawbacks, like limited scalability or the introduction of a Single Point of Failure (SPoF). Therefore, we propose a novel IE system that is based on Software Defined Networking (SDN). Originally meant for data center and IT networks, the SDN concept offers features like central network management functions and a fine-grained traffic control that allows to support many applications with diverse requirements even in the same network. Thereby, SDN is also perfectly suited for complex automation environments. To guarantee RT data transmission as well as scalability and an efficient resource usage, our IE system uses a Medium Access Control (MAC) scheme that is based on a Time Division Multiple Access (TDMA) mechanism that is extended by simultaneous data transmissions on physically separate links. The enhanced TDMA mechanism is configured by a joint routing and scheduling algorithm that takes application requirements into account. Our theoretical analysis as well as results achieved with a prototype implementation of the system confirm the applicability of our concept in demanding automation environments with applications that require a worst case communication latency below 1 ms.

WLAN mesh networks are characterized by their flexible and low-cost deployment, scalability, and self-healing capabilities. The new WLAN standard IEEE 802.11s introduces low-level mesh interoperability. However, building large-scale real-world test beds and reproducible setups is challenging and costly. In the majority of research works, network simulation is preferred over practical measurements. Here, the main disadvantage exists in simplified device and protocol models restricting the comparability to practical implementations. In contrast, using device emulation still requires the simulation of wireless channel and environment models. Consequently, a combination of both emulation and simulation is needed to enable virtual prototyping of real applications and protocols in WLAN mesh networks. Nevertheless, the computation of complex wireless channel effects requires a decoupling of wall clock and simulation time. Therefore, we present ViPMesh, a virtual prototyping framework for IEEE 802.11s and its Linux reference implementation. ViPMesh relies on WLAN device emulation and nested virtualization using QEMU and Linux containers to support the analysis of real applications on top of an unmodified protocol stack. Adopting an alternative time source approach for QEMU, ViPMesh acts as discrete-event simulator. It further integrates channel and environment models with support for IEEE 802.11n MIMO techniques, high throughput modes, multi-channel operation, and node mobility. To the best of our knowledge, this is the first approach that combines the IEEE 802.11s reference implementation with the described simulation features. The functionality of ViPMesh is demonstrated in different example scenarios.

For the realization of decentralized communication infrastructures, P2P technology offers an excellent technological foundation to complement or replace existing centralized structures, which take the client-server network model as a basis. In particular, the distributed hash table-based P2P network Kad impresses with high lookup performance, scalability, and resilience against failures and attacks. In addition to these advantages, the Kad protocol needs an extension in terms of its lookup to allow for lookup determinism in an a priori unknown Kad network. Therefore, this paper introduces a dynamic search tolerance, which adapts itself to any network configuration autonomously at runtime. The search tolerance is periodically calculated and all nodes are provided with the updated value. Thereby, the Kad protocol meets the demands for deterministic lookups, which enables dependable and efficient data retrieval, qualifying it as reliable basis of communication infrastructures in, e.g., real-time environments. Results show the high performance of the proposed algorithms for a Kad network with up to 50,000 nodes.

The vision of the smart home is increasingly becoming reality. Devices become smart, interconnected and accessible through the Internet. In the classical building automation domain already a lot of devices are interconnected providing interfaces for control or collecting data. Unfortunately for historical reasons they use specialized protocols for their communication hampering the integration into newly introduced smart home technologies. In order to make use of the valuable information gateways are required. BACnet as a protocol of the building automation domain already can make use of IP and defined a way to represent building data as Web services in general, called BACnet/WS. But using full fledged Web services would require too much resources in the scenario of smart home thus we need a more resource friendly solution. In this work a Devices Profile for Web Services (DPWS) adaptation of the BACnet/WS specification is proposed. DPWS enables Web service conform communication with a focus on a small footprint, which in turn enables interdisciplinary communication of constrained devices.

Software-defined networking (SDN) is a principle for the flexible configuration of networks, which recently has aroused an increasing interest of researchers and companies. Today, the operation and configuration of networks as well as their adaptation to changing requirements represent a major challenge if legacy network management protocols are used due to their inability to provide network-wide configuration. Compatibility issues with legacy protocols or even proprietary protocols have further contributed to hardly reconfigurable and inefficient networks and thus prepare the way for the new concept called SDN. SDN implementations are mainly in the data center today. Implementations will find their way into broader networking applications over the next few years. However, even automation networks with highest QoS requirements can benefit from SDN support by achieving significantly more devices that meet these requirements. Therefore, this paper addresses the emulation of SDN-supported automation networks to examine possible design options. Limitations and possibilities when using the popular and widely accepted SDN emulator Mininet are analyzed.

In recent years, cloud computing, which provides users with network resources such as memory and computing power depending on the users' needs, has gained enormously in importance. Providers such as Amazon allow the users to access their cloud storage and computing resources by different interfaces. In this regard, guaranteeing compatibility is a severe issue as many different proprietary interfaces exist for accessing data in different clouds. The solution to this problem is the RESTful Cloud Data Management Interface (CDMI) standard, which has passed the ISO audit as first standard and is therefore evolving into the most common standard for accessing clouds. This paper investigates the combination of CDMI with a P2P-based storage and computing back-end in order to realize a self-organizing cloud for distributed data storage and processing called DuDE-Cloud. The basis is the self-organizing hash table (DHT)-based P2P storage and computing back-end called DuDE, which utilizes the DHT protocol Kad. DuDE is able to bundle existing storage and computing resources of different devices dynamically and has already proven its advantageous performance over centralized solutions. In a test scenario, resources of DuDE-Cloud were successfully accessed through CDMI using a GUI front-end thereby proving the proper functionality. As a result, a resilient high-performance distributed cloud solution is available that is compatible with other clouds.

The broad availability of WLAN-capable off-the-shelf hardware lets WLAN mesh networks appear as promising technology for future distributed wireless applications. Featuring automatic device discovery, interconnection and routing, they provide a higher scalability, flexibility, and robustness compared to common centralized WLAN infrastructures. Besides these advantages, characteristics such as variable network topologies and link qualities imply new technical challenges for administration and real-world operation. Adopted in late 2011, IEEE 802.11s appears as new WLAN standard amendment, enabling vendor-independent mesh networks based on the widespread WLAN technology. However, network monitoring and management fall out of the standardization scope and are therefore not specified. In this paper, we present a novel 802.11s management solution based on the SNMP protocol. It covers dynamic mesh bootstrapping, error recovery, status monitoring and remote configuration. The presented solution was implemented and evaluated in a real-world testbed comprising more than 10 mesh nodes.

The distributed computing of data is a challenging task in terms of the self-organizing task distribution and computing, especially if distributed computing systems are becoming very large and complex. Therefore, the distributed hash table (DHT)-based P2P system called DuDE has been developed to compute statistics of access nodes of Internet service providers in an efficient way. DuDE exploits the high failure resilience and scalability features of the DHT network Kad to achieve a high-performance distributed system, which avoids the bottlenecks of a centralized computing solution. To ensure highly available data, Reed-Solomon codes for reliable distributed data storage are utilized. For implementing DuDE, usual working steps of distributed computing have been extended to realize a highly scalable computing system. We have developed a simulation model for a large-scale DuDE network consisting of up to 9,000 access nodes for computing statistics. In this paper, simulation results are presented, which demonstrate that DuDE is able to almost linearly accelerate the distributed computing compared to a centralized solution while introducing low traffic overhead.

The term Internet of Things (IoT) describes a scenario where embedded systems are integrated into everyday objects, turning them into smart devices to assist the user in his everyday life. Each of these smart objects only offers a very limited amount of computational power since it is only specialized in a limited set of tasks. In order to achieve complex goals, the devices have to interact with each other. Therefore, they do not only need to be interconnected either by wire or through wireless technology but also need a set of common protocols to enable vendor-independent communication. In the past years, various protocols pursuing this objective have emerged. One of the most promising approaches is the Constrained Application Protocol (CoAP) because it offers high interoperability and very low communication overhead at the same time. Typical IoT applications include the observation and manipulation of their environment through sensors and actuators. Since the physical world is continuous in time and does not wait for calculations to finish, it is essential that the execution times of the applications stay within certain boundaries. These timing constraints are referred to as real-time requirements. However, current protocol implementations do not consider real-time requirements for IoT applications. In this paper, we introduce the jCoAP communication stack as a lightweight Java implementation of CoAP. We give a brief introduction to real-time communication and CoAP and provide insight in the design concept of jCoAP and the offered functionalities. Furthermore, a performance evaluation is done in order to point out the suitability of the jCoAP framework for real-time IoT applications.

The Internet of Things is a great vision but also an important challenge for the current research. Through, the connection of many different devices there are outstanding benefits such as an increasing automation and intelligent energy management. Nevertheless, the integration of legacy devices into the Internet of Things is a significant issue. Legacy meters, e.g., cannot provide metering data automatically and the installation of modern smart meters is cost-intensive and thus often unprofitable. An alternative approach is to extend legacy meters with an automatic meter reading system. Thus, they can be integrated into an intelligent system such as an efficient energy management. In this paper a robust and universal approach for automatic meter reading is proposed. The approach can be applied to water, gas, and electric meters and uses a Web service as standardized and open communication interface. The recognition rate of a prototype implementation is 98 % and the measurement duration is approx. 1.5 seconds. The entire system was implemented on a Raspberry Pi as a cost-efficient hardware platform.

The domain name system (DNS) includes infrastructures deployed by Internet service providers (ISPs) and third-party suppliers to ensure high responsiveness, resilience, and load sharing. This equipment implies high effort and energy for 24/7 operation. To facilitate cost reductions in this regard, P-DONAS—a peer-to-peer (P2P)-based DNS—organizes access nodes (ANs) of an ISP’s access network, which possess available resources, into a decentralized, self-organizing distributed hash table--based P2P network. Each AN acts as traditional DNS server and solely stores a piece of DNS data. DNS requests issued to an AN are resolved via P2P lookups while maintaining full compatibility with traditional DNS. The article discusses the application of P-DONAS as both a complement and an alternative to traditional DNS. Results from both simulations and a practical test arrangement prove P-DONAS’ high scalability and its performance comparable to that of a commercial DNS name server relieving this name server by 53% to 75% of DNS traffic.

The Internet of Things is becoming more and more relevant in industrial environments. As the industry itself has different requirements like (hard) real-time behavior for many scenarios, different solutions are needed to fulfill future challenges. Common Industrial Ethernet solution often leak scalability, flexibility, and robustness. Most realizations also require special hardware to guarantee a hard real-time behavior. Therefore, an approach is presented to realize a hard realtime network for automation scenarios using Peer-to-Peer (P2P) technology. Kad as implementation variant of the structured decentralized P2P protocol Kademlia has been chosen as base for the realization. As Kad is not suitable for hard real-time applications per se, changes of the protocol are necessary. Thus, Kad is extended by a TDMA-based mechanism. Additionally, to evaluate the performance an prototype is presented, which is realized on an embedded platform with a real-time operating system. Thereby, with the presented approach and a realized prototype it is possible to investigate the performance of a Kad network with hard real-time capabilities.

For companies in the automation industry, the development of real-time Ethernet to connect devices is of high economic interest to replace conventional fieldbus systems. Therefore, many approaches for adapting Ethernet to real-time requirements come from industrial applications. This is a challenging task as the original Ethernet standard IEEE 802.3 was not designed for real-time data transmission. Likewise, protocols basing on Ethernet like TCP, UDP, and IP do typically not consider real-time requirements. Hence, adaptations on several OSI layers become necessary to make the industrial system meet hard real-time requirements. For this purpose, a multitude of realtime capable Industrial Ethernet systems has been developed, which solve the problems of standard Ethernet- and TCP/IP- or UDP/IP-based communication in a variety of ways. This paper gives a summary of different Industrial Ethernet protocols for the real-time data transmission via Ethernet in automation environments. Advantages and disadvantages of these protocols are analyzed with regard to their sustainability in terms of their realtime capability, reliability, scalability, self-configuration of the network, and hardware requirements. Against the background of connected devices tremendously growing in number and computational power in the prospective “Industrial Internet”, consequences for future developments are drawn.

Web Services are becoming more and more relevant also in the domain of embedded devices as they are becoming an important aspect of the Internet of Things. Embedded devices, especially in the field of automation, require real-time behavior. The Devices Profile for Web Services defines WS-Standards for embedded systems. The WS-Eventing is one of this standards within DPWS and enables the distributions of events. However, WS-Eventing has a scalability problem as in ad-hoc networks the notification is transmitted sequential by an event source. Under these circumstances, it is not suitable for large-scale networks with a high amount of devices. Therefore, a new approach is presented to solve this issue by acquiring helping devices. The communication is only based on unicast IP communication and thus is easy to integrate into automation infrastructures and could be established over several subnets. This approach has been investigated with an experimental setup and shows the high scalability compared to the original notification mechanism. Additionally, the real-time capability is given due to the chosen platform and real-time operating system. Thus, the new approach enables the idea of Internet of Things in automation scenarios.

Web service technology becomes popular in a growing number of applications. The field of application reaches from smart home over factory automation up to smart cities. One of the main aspects of this technology is a dynamic device and service discovery. For this purpose, the WS-Discovery standard is used. WS-Discovery supports two modes: an infrastructure-based and an ad hoc mode. The infrastructure-based mode requires a service proxy, which represents a single point of failure. The ad hoc mode is based on the multicast and static response time values. Therefore, it shows a low scalability. In this work, we present a novel discovery approach, which bases on distributed hash tables (DHTs). DHTs are often utilized in peer-to-peer networks and represent a well-established technology. The search for devices and services is performed using hash values. The benefit of this approach is a high scalability, deterministic behavior, and flow control. The suggested approach can be applied for networks of any size. Since this approach uses unicast messages, it is not limited to local networks allowing the device and service discovery over multiple subnets or even over the Internet.

In this paper we describe a common problem when it comes to mutual authentication of devices in smart environments. Basically, there exist two approaches. In the brokered approach, two parties authenticate with the help of a common trusted party. However, this often introduces a single point of failure. The alternative approach lets the devices authenticate with each other directly, usually by exchanging a shared secret out of band. But this is only possible if both devices have the necessary user interface peripherals to establish an out-of-band channel. We will describe an abstract basic protocol to bridge the possible peripherals' gaps by authenticating indirectly with the help of nowadays basically omnipresent multimedia devices such as smart phones. After extending the proposed protocol for more complex use cases, we refine it to a practical authenticated key establishment protocol for indirect authentication. Finally we describe our freely available prototype implementation consisting of an internet-of-things-enabled light bulb and light switch and an Android app that comprise our proof of concept.

In this paper, a highly integrable Field Programmable Gate Array-based hardware design of multilayer perceptron as a realization of an artificial neural network is presented. Such a hardware solution ensures a deterministic behavior required for any hard real-time compositions. The integration into existing systems is achieved by the application of UDP/IP. %A developed protocol enables the hardware solution to act as a stand-alone device with no need for an additional host PC. Additionally, the presented design is highly flexible due to a parameterizable multilayer perceptron approach. However, most reconfigurations usually require a hard coded reimplementation, resynthesis, and the download of a new bit file to the target platform, which also requires an additional host PC. Contrary with the presented solution, it is possible to configure the multilayer perceptron's parameters during runtime via a software interface. This approach allows the multilayer perceptron to be adapted to nearly any application. The developed design combines the flexibility of a software solution to generate and comfortably reconfigure the multilayer perceptron as well as the high performance of a hardware solution. %The investigation of hardware utilization and performance of a running prototype stützt%Finally, the hardware utilization and performance are investigated. As proof of concept, a running prototype has been realized, which shows the design to be highly flexible and with good performance while the hardware resource consumption is kept minimal.

Device and service discovery is an essential part of automation networks in order to ensure the interoperability of devices and Plug&Play functionalities. Embedded Web services such as Devices Profile for Web Services are an upcoming technology for building and factory automation. For ad hoc discovery, the WS-Discovery standard is used. This approach avoids single points of failures. However, it shows low scalability due to static response time values. The rising number of devices leads to an increasing network load and especially to an increasing device processing load as a result of the discovery process. In this work, we evaluate the issues of ad hoc service discovery and suggest an optimized algorithm for dynamic data rate reduction in large scale networks. Moreover, the suggested improvements allow reducing peak data rates by more than 50% for constant timing parameters. Using the suggested approach, the ad hoc discovery can be applied in any network size. The desired peak data rates or response times of the system can be significantly reduced and dynamically adjusted according to the desired values. This approach guarantees a scalable ad hoc discovery mechanism and is backward compatible with legacy equipment.

The P2P-based system for the distributed storing of session data of Internet service providers' access nodes called P2P-based storage platform is presented. Session data is continuously changing due to customers connected to access nodes, i.e., it is highly volatile. However, it has to be stored persistently as it is required for data forwarding, traffic filtering, and deriving statistics. Failing access nodes must be able to restore their data after reentering the network. Today, the session data is stored in the access nodes' flash memory, which is limited in life time and size and intended for other purposes. Contrary, the P2P-based storage platform allows to store session data in the access nodes' available RAMs connected by the distributed hash table-based P2P network Kad. A Kad network consisting of 8,000 access nodes for distributed storage of session data is setup for simulation. The simulation results show the traffic overhead to be minimal, linear scalability, and prove the high availability of the stored session data.

In this paper, an approach to synchronize the P2P network Kad to be applied in automation scenarios is presented. The approach bases on a deterministic algorithm to synchronize the network, which is required for hard real-time applications. Todays Industrial Ethernet solutions include and support machine to machine communication in automation scenarios. However, instead of allowing direct communication, the communication relies on centralized structures, which are deficient in resilience and scalability However, instead of allowing direct communication, the communication relies on centralized structures. The presented decentralized approach benefits from nodes helping to synchronize the network. However, the higher the number of helping nodes the higher is the time deviation on the nodes of the network, which contrary results in a higher time error. Therefore, a trade-off between synchronization performance and time error has to be determined to meet predefined constraints depending on the application scenario. Moreover, the individual clock drift of every device is considered to define necessary re-synchronization intervals of the network. Additionally, the optimum number of nodes to synchronize the Kad based network has been identified and the resulting synchronization performance and generated traffic are determined. Furthermore, an approach is presented to handle the dynamic churn of nodes.

To date, Wireless Sensor Networks (WSN) require their own specific methodologies, tools, and technologies. With the rise of 6LoWPAN protocols (i.e., IPv6 over Low Power Wireless Personal Area Networks), WSNs can use IP to share a common network layer with other networks and the Internet for highly interoperable device communication. But additional efforts are necessary to develop new or adapt existing application layer protocols for IP-enabled WSNs. Hence, this paper investigates W3C SOAP Web Services (WS) in the context of WSNs. In particular, it is shown that the Devices Profile for Web Services (DPWS) can be used as an application layer protocol in WSNs.

Temperature management and signal integrity are two highly relevant challenges for nano scale CMOS devices. As temperature of integrated circuits affects the frequency of defect mechanisms' occurrence and consequently influences reliability, temperature monitoring is inevitable. To provide the necessary thermal sensors, different techniques are available. One is to use the temperature dependent speed of logic devices and perform a time to digital conversion. In this work, this approach is evaluated for the use in FPGAs. Furthermore, problems regarding signal integrity affect the reliability of highly integrated circuits. Therefore, we discuss the possibilities of crosstalk effects in FPGA and demonstrate a method for time to digital conversion in order to measure the impact of coupling capacitances in the interconnection structure of FPGAs. The main contribution of the introduced method is to enable simple post-production investigations for signal integrity of programmable devices.

Because of the aggressive scaling into the nanometer regime, degradation due to wearout significantly impairs design parameters. For instance, such wearout is caused by gate oxide breakdown, which decreases the operating lifetime of integrated circuits to an extent that cannot be neglected by circuit designers to date. In this paper, we introduce an approach which applies selective redundancy to different combinational designs in order to improve reliability as regards gate oxide breakdown. Therefore, the most vulnerable transistor stacks of standard cells are doubled based on activity and the propagation delay of the design. Finally, reliability improvements of up to 75% are presented that are gained with Spice simulations. Such improvements come at the price of overhead for area and power consumption as well as delay of at most 14%. However, it is interesting to notice that the initial delay penalty of our enhanced designs finally turn into a timing advantage, as the designs are more and more affected by wearout over time. Hence, this advantage translates into further reliability improvements when clock requirements are also considered. Besides, it needs to be noted that the presented strategies can additionally improve defect yield.

There are many applications that use low power wireless sensors to monitor people's vital signs or the environment around people. For mobile applications, smartphones with Internet access have become a commonly used tool to make this data remotely accessible. However, the yet presented architectures are quite complex and lack flexibility. Our aim was to investigate how existing communication architectures for smartphone/WPAN sensor ensembles can be simplified and to provide a transparent solution where the smartphone doesn't require knowledge about sensors for which it serves as an Internet gateway. We therefore present an IP-based communication architecture, that allows applications to easily integrate remote sensor data of mobile users based on existing Internet standards. In particular, our solution provides simple Web service-based APIs and does not require the smartphone to process sensor data. Instead, the smartphone becomes a transparent gateway as it only forwards IP packets.

The actual situation in smart metering is characterized by a coexistence of a large number of proprietary and open standards for wired as well as wireless communication. These standards show low or no interoperability to each other. Therefore, it is very difficult to integrate multi-vendor solutions using one sustainable holistic approach. The proposed approach is to use Web Service technology as an open widespread Internet standard for the creation of a heterogeneous network for smart metering devices. Smart metering is an emerging topic for realizing modern energy policies. Monitoring, analyzing, and controlling of power consumption is the precondition for optimizing energy strategies. This work focuses on the integration of Web Service technology in the area of smart meter communication. Furthermore, mechanisms for reducing Web Service traffic overhead by over 90% is presented.

What visions of technologies such as the Internet of Things (IoT), Pervasive Computing (PC) or Ambient Intelligence (AI) have in common is that they employ a very high amount of critically resource-constrained devices. This of course raises a whole set of new security challenges. Instead of proposing another middleware for IoT, PC or AI, and presenting a security approach for it we focus on existing, widely-adopted security mechanisms and concepts and give a comprehensive overview. Furthermore, we examine the Web Service security suite as an example for a comprehensive security framework on application level. We use an intentionally kept simple scenario including light bulbs and switches to demonstrate necessary security mechanisms and properties and to judge applicability of the presented existing security techniques and frameworks.

Progressive technology scaling raises the need for efficient VLSI design methods facing the increasing vulnerability to permanent physical defects, while considering power efficiency of resulting circuit implementations at the same time. Triple Modular Redundancy (TMR) represents a common method to encounter reliability problems, but has the drawback of increased area and power consumption. This work introduces a Low Power Redundant (LPR) design solution that targets the power penalty of TMR implementations. This is done by enhanced and new functional runtime capabilities for error detection and operation control. By exploiting the inherent modularity and parallelism of TMR, the LPR solution applies additional control logic to switch dynamically between compare phases (to indicate faults and their locations) and parallel operation (with reduced operation frequency). This allows power optimized circuit operation with full support for the treatment of permanent faults. Simulation results on different ALU implementations show a decrease of power consumption of up to 60% compared to conventional TMR. Furthermore, different strategies for the switching between operation modes are introduced that enable power efficient system operation in the presence of permanent physical defects. Moreover, significant reliability improvements are also achieved due to the adaptive use of the redundant modules.

The increasing level of device connectivity in today's homes and buildings enables numerous opportunities for home owners, building managers, device manufacturers and solution providers. Standardized communication protocols (e.g., ZigBee, Bluetooth) provide physical connectivity and thus serve as a basis for smart home applications. However, beyond physical connectivity, real interoperability to effectively develop such applications requires additional efforts. It requires harmonizing multiple protocol standards, dealing with a heterogeneous device landscape, different data formats, managing resource constraints of devices and providing means to react quickly when devices and applications leave or join the system. We propose to address these application-specific requirements in multiple layers of abstraction. Our research has shown that each layer is used by different types of developers (e.g., device supplier, service provider, home owner) with specific tool support. The paper provides a classification and detailed analysis of requirements that have to be addressed in order to enable application development in smart homes. It further proposes and analyzes WS4D-PipesBox, a multi-layer framework, to illustrate how applications could be developed using multiple layers of abstraction.

A huge momentum towards IP enabled Wireless Sensor Networks (WSN) appeared through the emerging 6LoW-PAN protocols (i.e. IPv6 over Low power Wireless Personal Area Networks). By usage of existing cross domain open standards in contrast of proprietary solutions, deployments of WSNs are not tailored too tight for specific applications. Nevertheless, 6LoWPAN is only the first step for the usage of internet protocols in WSNs. Still efforts on higher layers on top of 6LoWPAN are an urgent need to provide seamless connectivity and interaction of highly resource constrained devices with higher valued services. This paper describes a new approach to bind SOAP to the emerging Constrained Application Protocol (CoAP) protocol. Thereby CoAP provides a lightweight but reliable transport binding for SOAP based protocols. Compared to the widespread TCP based HTTP binding, round trip times can be reduced by 43% in an exemplary scenario. Combined with dedicated XML compressors like the Efficient XML Interchange format (EXI), existing heavy weight SOAP based protocols become also applicable in WSNs.

The P2P-based system for the distributed computing of statistics called DuDE is presented. High scalability and failure resilience features of P2P are exploited to achieve a high- performance distributed system, which avoids the bottlenecks of a centralized computing system. To ensure high data availability, a sophisticated algorithm for distributed data storage is integrated. Furthermore, an algorithm for global peer discovery is presented, which allows for finding all data assigned to peers without the need for a central instance. For the realization of DuDE, common working stages of distributed computing are extended to enable a highly scalable computing system based on P2P technology. Generated results from a test system show a nearly perfect linear speedup for distributed computing as well as high processor and memory relief compared to a centralized solution.

A huge momentum towards IP enabled Wireless Sensor Networks (WSN) appeared through the emerging 6LoWPAN protocols (i.e. IPv6 over Low power Wireless Personal Area Networks). By using existing cross domain open standards in contrast of proprietary solutions, technologies and deployments are not tailored too tight for specific applications. Nevertheless, 6LoWPAN is only the first step. Still efforts on higher layers on top of 6LoWPAN are an urgent need to provide the seamless connectivity and interaction of highly resource constrained devices with higher valued applications and services. This paper describes a new approach to bind SOAP to the emerging Constrained Application Protocol (CoAP) protocol. Thereby, CoAP provides a lightweight but reliable transport binding for SOAP. Compared to the widespread TCP based HTTP binding, round trip times can be reduced by 43% in an exemplary scenario because of the omitted expensive bidirectional TCP handshake mechanisms. Combined with dedicated XML compressors like EXI, SOAP based protocols become also applicable in WSNs.

In large wireless sensor networks, low energy consumption is a major challenge. Hence, deployed nodes have to organize themselves as energy efficient as possible to avoid unnecessary sensor and transceiver operations. The energy conserving operations are limited by the task of the network, usually the network has to guarantee complete functionality during its lifetime. The contribution of this paper completes the functionality-aware and energy-efficient clustering algorithm family MASCLE by two innovative algorithms. As already given by the MASCLE-algorithms, the proposed Hex-MASCLE algorithms combine advantages of temporal and spatial network fragmentation. In contrast to previous approaches, the shapes of the basic cells are given by regular hexagons, similar to honeycombs. In the present work, two possible versions for hexagon-based clustering with self-healing abilities are proposed and evaluated. As result, the applying sensor network achieve a significant improve of network lifetime. Additionally, the algorithms are more fault-tolerant against localization errors.

As transistor dimensions are shrinking into regions of only a few atomic layers, designers are faced with various problems including increased reliability and power issues. Since these problems are amplified by higher circuit temperatures, this paper proposes an approach for the fine-grained modeling of temperature distribution in many-core systems based on Networks-on-Chip. With this model, algorithms can be developed that consider the significant impact of temperature -e.g. on performance, power or reliability. To simulate the dynamic nature of temperature, the thermal properties of according integrated systems are modeled through the instantiation of equivalent RC-circuits. This approach exploits the dualism between electrical and thermal flows of energy. Finally, an application with system control for task mapping and power management exemplifies the proposed simulation methodology.

Several alternatives of mesh-type topologies have been published for the use in Networks-on-Chip. Due to their regularity, mesh-type topologies often serve as a foundation to investigate new ideas or to customize the topology to application-specific needs. This paper analyzes existing mesh-type topologies and compares their characteristics in terms of communication and implementation costs. Furthermore, this paper proposes BEAM (Border-Enhanced Mesh) - a mesh-type topology for Networks-on-Chip. BEAM uses concentration while necessitating only low-radix routers. Thereto, additional resources are connected to the outer boundaries of a conventional mesh. As a result, overall bandwidth is traded off against hardware overhead. In conclusion, simulation and synthesis results show that the conventional mesh stands out due to its communication performance, whereas clustered and concentrated topologies offer the least hardware overhead. BEAM ranges in between and is an option to balance hardware costs and communication performance.

Service-oriented architectures and systems designs are powerful concerning reusability of functional blocks and hiding implementation details from functional interfaces. But to compose a complete application, often central entities and engines are required for processing a specific sequence of service. In component-based designs the component itself is capable of describing both not only their offered services and interfaces but also dependencies on other services and interfaces to fulfill a complete task or application logic. This paper investigates on how to transfer and enhance exiting component-based approaches, already known from business applications, into the domain of embedded web services environments. Special focus is on the Devices Profile for Web Services (DPWS) technology which features service orientation also in device centric applications. This paper introduces a new approach to create applications, based on services provided by devices deployed with DPWS, in an abstract and dynamic way.

Wireless Sensor Networks (WSNs) have been of high interest during the past couple of years. One of the most important aspects of WSN research is location estimation. As a good solution of fine grained localization Reichenbach et al. introduced the Distributed Least Squares (DLS) algorithm, which splits the costly localization process in a complex precalculation and a simple postcalculation which is performed on constrained sensor nodes to finalize the localization by adding local knowledge. This approach lacks for large WSNs, because cost of communication and computation theoretically increases with the network size. In practice the approach is even unusable for large WSNs. This restriction have been overcome by scalable DLS (sDLS), which enabled to use the idea of DLS in large WSNs for the first time. Although, sDLS outperforms DLS for large networks, cost of communication and computation is initially higher for small networks, caused by data updates. The approach, presented in this work, dramatically reduces cost of communication of sDLS. Additionally, a new approach of distance estimation is applied to original DLS. In contrast to earlier simulations, this leads to improved localization, which is used for fairer comparison.

Wireless Sensor Networks (WSN) have attracted considerable research effort in the community during the past couple of years. One of the most challenging issues so far is the extension of network lifetime with regards to small battery capacity and self-sustained operation. Endeavors to save energy have been made on various frontiers, ranging from hardware improvements over medium access and routing protocols to network clustering and role changing strategies. In addition some authors studied failures in communication regarded as error detection. Yet, only weak attention has been paid to the detection of malicious nodes and its potential for lifetime extension. In this work, we present a short overview of detection and classification of malicious nodes in WSN and describe its potential in terms of network lifetime and reliability.

This demonstration shows the hardware prototype of the IPclip (IP calling line identification presentation) mechanism for IPv6 networks. IPclip is a mechanism, which provides Trust-by-Wire in IP-based networks by adding trustworthy location information to IP packets. It is implemented on an FPGA development board and configurable at runtime via a graphical configuration tool. We show IPclip's basic functionality in a localization scenario using an analysis tool and Google Earth and discuss several application scenarios during the demonstration.

We propose a measure to characterize the energy efficiency of algorithms for localization in wireless networks. The measure presented differs from previous approaches in that it is bounded and supports objective comparison. Furthermore, it corresponds to the general understanding that a high value should indicate high efficiency. Simulation results for localization using maximum likelihood estimation (MLE), for least squares estimation (LS) and for multidimensional scaling (MDS) are briefly discussed.

As sensor networks are deployed over various terrains, the complexity of their topology continues to grow. Holes in networks often cause existing geographic routing algorithms to fail. In this paper, we propose a novel geographic routing algorithm called Greedy Forwarding with Virtual Position (GF-ViP). We introduce virtual position as the middle position of all neighbors of a node. Instead of comparing nodes' geographic position, GF-ViP employs virtual position for selecting the next hop. Such virtual position reflects the neighborhood of a sensor node, as well as the tendency of further forwarding. For network with routing holes, GF-ViP significantly increases success rate of packet routing, while the overhead is kept low. Furthermore, multiple levels of virtual position can be obtained with localized iteration. We propose the Greedy Forwarding with Multi-level Virtual Position (GF-MVP) algorithm. According to various context information of a sensor network, different levels of virtual position can be used alternatively to increase success rate of packet routing in sensor networks.

For Service-oriented Architectures, Web Services are claimed as state of the art to connect business execution layers as well as networking devices. Additionally, the deployment of Wireless Sensor Networks became applicable over the last years. The usage of application layer gateways and proxy concepts allow the integration of these sensor networks into real world scenarios and existing networks that make use of Web Services. This paper presents a new approach to adapt and enhance the Devices Profile for Web Services to be applied in Wireless Sensor Networks directly. Thus, seamless connectivity between business layers, device level networks, and Wireless Sensor Networks are possible.

During the last years, recent technological advances enabled the development of tiny devices equipped with radio, micro controller and sensors, called sensor nodes. Collaborating networks of such devices, known as wireless sensor networks (WSNs), are subject of many researches. A couple of network centric tasks like data transmission and aggregation, self-organisation, localization and energy awareness have been focused in various publications. Researches, e.g. IPv6 over low power WPAN (6LoWPAN), are going to standardize the connection of sensor nodes with common Internet protocols. Nevertheless, until now proprietary techniques are used within most networks, especially for the coupling towards the outside world as well as the data management. Depending on WSN hardware and software, the gateway software as well as storage and presentation of data differs for most sensor networks. In this work we present an infrastructure which bases upon a generic gateway. Standardized dataset will be stored by this device using sensor Web enablement (SWE) and standard protocols. By use of devices profile for Web services (DPWS) plug and play abilities as well as a comfortable query interface for presentation is realized.

Service-oriented architectures (SOA) are nowadays a wide-spread software building approach. The fields of application ranges from embedded devices up to enterprise and business solutions. For distributed services and devices it makes sense to have a mechanism for remote administration to manage their life cycle. Technological standards like OSGi (see section 3) provide such a life cycle management, but device-oriented Web service technologies like the devices profile for Web services (DPWS) do not specifies such a mechanism as a standard. In this paper we present a service-based life cycle manager for an implementation of DPWS to manage the services of a device during runtime through remote reconfiguration. The advantage of the presented solution is the ability to be platform and language independent by using a script-based interaction between the life- cycle-manager and the individual services.

We propose a metric to characterize the energy efficiency of range-based localization algorithms in wireless sensor net-works. The metric proposed differs from previous approaches in that it is bounded and supports objective comparison of localization algorithms using simulations. The goal of the current work is to show that our metric achieves expected results for well-known localization algorithms and, there-fore, can be used to characterize and compare the energy efficiency. Simulation results for energy efficiency show that maxi-mizing the local likelihood yields highest energy efficiency whereas the linear least squares approach and the multi-dimensional scaling method exhibit a strong susceptibility when ranging errors are large.

In order to achieve energy conservation in WSNs, most topology management protocols use a subset of sensor nodes for global routing. Using fewer nodes results in a reduced connectivity of the network, which eventually increases the number of routing holes. Holes in networks often cause failures in message routing due to the local minimum problem. Therefore, traditional geographic routing protocols cannot be applied with such topology management protocols. In this paper, we propose a novel topology management protocol derived from the Geographical Adaptive Fidelity (GAF) protocol, called GAF with COnnectivity-awareness (GAF&Co). Instead of using virtual grids in GAF, our approach employs hierarchical hexagonal cells to avoid local minimums in WSNs. The purpose is to schedule redundant nodes into energy-saving mode, while maintaining the connectivity of a network for simple geographic routing protocols. Comparing to GAF, the number of cells as well as the overall energy consumption of a WSN also drops dramatically with the proposed protocol.

Modern sensor networks are deployed in various terrains of interest. As the complexity of their deployed areas is growing, existing geographic routing algorithms are facing challenges. Holes in networks often cause failures in message routing. Energy consumption, scalability, and routing efficiency are also key design challenges. In this paper, we propose a novel geographic routing algorithm called HOle-BYpassing routing with Context-AwareNess (HobyCan). Our approach locally sets up multiple detour paths to bypass almost all kinds of holes. Therefore, contours of holes are extended with multiple detour paths. According to various context information of a sensor network, such as the size of holes or the remaining energy of nodes, disjoint detour paths can be used alternatively to achieve optimal routing paths or load balance of the network. Simulation results demonstrate the performance of our algorithm, as well as the significance of context information as routing parameters.

In ubiquitous computing, localization of users in indoor environments is a challenging issue. On the one hand, localization data needs to have fine granularity to provide reasonable input for intention recognition and task planning. On the other hand, effects like multi-path interference and signal scattering of RF propagation in indoor environments reduces the accuracy of traditional wireless localization techniques. However, we prove that such adversary effects can be characterized and utilized conversely to localize the source of RF-scatter with passive UHF RFID. Since measuring spatial correlation requires many spatially separated transmitter and receiver pairs, cost-effective and unobtrusively attachable passive RFID-tags are especially suitable for this purpose. The passive tags are spatially distributed in a manner such that it is possible to infer the spatial correlation of received signal strength (RSS). The idea is to characterize the influence of user presence on RSS, and use such relationship for localization. Three localization algorithms are investigated which consist of a maximum likelihood estimator (MLE), and two linear least squares variants. Algorithms are applied to measurement data which we obtained in an indoor environment. The results evidences our idea of human localization in such bistatic RFID systems.

Using simple and cost-effective tags, passive RFID systems offer a promising aid for identifying and localizing objects and users in indoor environments. Although systems based on radio frequencies usually suffer from multi-path interference and signal scattering, we show that the characteristics of such interference and scattering can be analyzed with passive RFID. We present and analyze measurements of received signal strength (RSS) conducted in an indoor environment using a passive bistatic RFID-System. In order to characterize the influence of human presence on RSS, measurements were conducted for different user locations and orientations in an indoor deployment area. Finally, an analytical approximation of the relation between user location and RSS is presented in accordance to our measurement results.

Service-oriented architectures become more and more important in connecting devices with each other. The main advantages of service-oriented architectures are higher abstraction level and interoperability of devices. In this field Web services become the most important standard for communication between devices. But this upcoming technology is only available on powerful devices. Embedded hardware is often excluded from the deployment of Web services because of the lack of resources like computing power and memory. In this area also real-time capabilities for process control are required. This paper presents a new approach to handle Web services communication on deeply embedded hardware with the devices profile for Web services specification.

Hash functions have a space complexity of O(n) and a possible time complexity of 0(1). Thus, packet classifiers exploit hashing to achieve packet classification in wire speed. Especially evolvable hash functions can adapt to a changing classification data base. But hash functions do have an important flaw. Some of the hashed keys may result in a large number of collisions. If those keys occur frequently, overall performance of a hash based packet classifier suffers. Although there is only limited or no existing locality in the data to be computed by a packet classifier, utilizing a cache can solve this problem. Unlike classical caches known from microprocessors, which are not adequate for packet classification and lookup algorithms, the proposed cache bases on a different strategy. It caches only the keys producing the most collisions instead of the ones that occur most often. That way, a cache improves the worst case performance of a hash function-based classifier. Based on simulation results, we show that even the mean performance improves significantly.

During the last years, the Internet has grown into a mass-medium for communication and information exchange. Millions of people are using the Internet for business and in social life. Users can be reached easily and cost-effectively. Unfortunately the Internet's open structure is the reason for its frequent misuse for illegal and criminal actions such as dissembling phishing attacks. Thus, anti-phishing techniques are needed to recognize potential phishing threats. But mostly these techniques are only of reactive nature, are soon circumvented by expert frauds, or are not efficient enough. This paper describes an anti-phishing framework. A concept for trust management and a mechanism called IPclip are presented. The main idea of IPclip is to guarantee trust-by-wire in packet-switched networks by providing trustworthy location information along with every IP packet. This information is used as supplementary and trustworthy trigger to identify potential phishing threats. Besides, the proposed framework allows for tracing the threat's origin by using a set of location information.

Currently the developing process for enterprise applications is improved by the Service Oriented Architecture (SOA) paradigms. With SOAs the creation of modular and clearly defined software architectures at a high grade of interoperability and reusability is possible. For resource constraint networked devices the Devices Profile for Web Services (DPWS) specification adapt the SOA paradigms to create a framework for interoperable and standardized communication between embedded devices. This paper explains how special parts of DPWS can be adapted to provide real-time capabilities. So this paper shows how developers of real-time software can use the advantages of Web services and which adaptations are necessary to create Web services with real-time capabilities.

The capabilities of passive optical networks (PONs) are strongly influenced by the quality of the used dynamic bandwidth allocation (DBA) algorithm. DBA algorithms control the assignment of available upstream bandwidth to the users connected to the PON. In an oversubscribed environment, that poses a challenge regarding the selection of an appropriate DBA algorithm. Therefore, DBA algorithms are subject of continuous research. To support efficient development and realistic evaluation of DBA algorithms, an FPGA-based hardware evaluation platform is presented. It supports fast implementation and evaluation of both hardware- and software-based algorithms. The evaluation process is featured by software tools, which are used to control stimulus creation. Furthermore, the results of changes in simulated user traffic can be analyzed on a connected workstation.

During the last decades, the Internet has steadily developed into a mass medium. The target group radically changed compared to, e.g., the 90s. Because virtually everyone has access to the Internet, threats due to insecurity and anonymity reach critical levels and have to be tackled by both carriers and Internet Service Providers. Regaining trust-by-wire, comparable to classic fixed line telephones, could mitigate or even solve problems like Spam, Phishing, and the localization of VoIP emergency calls. This paper presents the hardware implementation of a new and highly flexible solution-Internet protocol-calling line identification presentation-which provides additional support for new services to restore peoplepsilas confidence into the Internet. Supported services are VoIP emergency calls, Spam detection and prevention, and phishing prevention. Already in the access network, the hardware adds unambiguous location information on the packetpsilas origin to IP packets. We document the hardware design of the solution. Furthermore, hardware consumption and performance of a prototype are presented.

This paper investigates wireless sensor networks where a small percentage of nodes are assumed to know their location a priori. These reference nodes enable absolute localization of other nodes in direct neighborhood. Having estimated their location, these nodes in turn provide their location to other nodes within transmission range. Therefore, location information spreads throughout the network. Consequently, in later state of the network, unknowns desiring to determine their location, or to improve it, will be able to choose from a large pool of nodes with known or estimated locations, which we refer to as beacons. We investigate a method to select a subset of beacons to minimize the error of localization. Regarding Cramer-Rao-Lower- Bound on localization error, the method proposed constitutes a significant improvement in comparison with the often used nearest-neighbors approach.

As the scale of modern sensor networks continues to grow, energy consumption, scalability and routing efficiency are becoming key design challenges. Network management plays an important role in achieving these goals. By decomposing a sensor network into smaller groups, clustering and its variants have been presented as efficient ways in network management. In this paper, we propose a dynamic, localized clustering approach derived from generic budget-based clustering techniques. The approach generates dynamic cluster sizes for a hierarchy of cluster heads, with respect to network context such as residual energy and activity rates of sensor nodes. We further refine the local estimated cluster sizes by using additional feedback during clustering process. Simulation results of stochastic deployment are used to demonstrate the performance of our algorithm, as well as the impact of context information as clustering parameters.

Leakage power dissipation is one of the most critical factors for the overall current dissipation and future designs. However, design techniques for the reduction of leakage power should not decrease design performance. Therefore, an enhanced Dual Vth/Dual Tox CMOS ap-proach is presented which applies mixed gates consisting of different transistor types. The paper introduces the new and fundamental idea of different gate types before the various possible configurations are analyzed. This is followed by extraction and exploration of design rules and recommendations. Simulations of modified ISCAS'85 designs show an average leakage reduction of 60% at constant performance compared to raw designs. This corresponds to an additional reduction of 20% compared to previous Dual Vth/Dual Tox CMOS approaches.

In many applications such as signal integrity checking of hardware prototypes or determining dynamic behavior of wideband amplifiers etc. analysis of periodic radio signals with high accuracy is desired. The straight forward approach is to sample the signal to be analyzed with twice its highest harmonic content. But as fundamental frequencies are often far into the MHz range its harmonics may well span into the upper MHz or even GHz range. This leaves two possibilities: first, obey the sampling theorem and sample at that rate. But ADCs able to sample in the GHz range are expensive, power hungry and offer 6 to 8 bits maximum. Sometimes, this is not an option at all. Second, try to live with under sampling accepting alias frequencies in baseband. As long as aliased harmonics do not overlap this presents an acceptable solution. If they overlap, however, nothing can be done once the signal is sampled. But there is one more option presented in this paper, namely the possibility of deliberate nonuniform sampling. If the sampling set is chosen in a convenient way, mostly offered by an additive random sampling (ARS) scheme, this opens up promising possibilities to extend the alias-free processing range into the far MHz or even GHz region.

Autonomous localization of nodes in wireless sensor networks is essential to minimize the complex self organization task and consequently enhancing the overall network lifetime. Recently, precise localization algorithms are impeded by multi path propagation of signals originated by reflections at walls or other objects in the environment. However, the localization error can be reduced by applying statistical methods that consider the mass of input information provided in large WSNs. In this paper we present the "Iterative DLS"-algorithm (iDLS), a new localization method that reduces the error of the initial position estimate by more than 46% and finally allows a precision of 3 m (fieldsize: 100 mtimes100 m) at noisy input. This is achieved by extending the known "distributed least squares"-algorithm by a refinement-phase, where sensor nodes provide their initial position to neighbors. This algorithm places an absolute minimum of computational requirement on the resource constrained sensor nodes.

In wireless sensor networks, the coarse grained localization is a method to compute the position of randomly distributed sensor nodes. Without optimizations, it provides low precision which heavily depends on the transmission range of base stations. In this paper, we propose novel optimizations of coarse grained localization with centroid determination (CGLCD) to determine the position of nodes more precisely. Our focus is to compute an optimal transmission range of all base stations and to reduce the total energy consumption. We present an analytic proof of a simple equation to determine the optimal transmission range in grid-aligned finite wireless sensor networks. Using this optimal transmission range, we reduced the positioning error about 80%. Thereby, nodes as well as base stations require lowest energy

Managing reconfigurable hardware resources at runtime is expected to be a new task for future operating systems. But due to the mixture of parallel and sequential parts of dynamically reconfigurable applications, it is not entirely clear so far, how to use and to program such systems. A new interpretation of dynamically reconfigurable applications is presented. It is shown, that the parallel computing concept of distributed object systems may be adapted for dynamically reconfigurable architectures. This approach answers many open questions concerning communication, interruption, and relocation of reconfigurable modules. It is explored by means of an extended Linux operating system in conjunction with a SystemC model of a dynamically reconfigurable FPGA

Positioning sensor nodes requires distance information to reference points. Due to resource limitations in sensor networks, distance determination in low-cost sensor nodes without additional hardware is difficult. Known techniques such as distance estimation based on received signal strength (RSSI) are mostly inaccurate or have limitations. We propose a new method to measure the distance between a transmitting node and a receiving node using the minimal transmission power. The determined distance is more precise than RSSI, has a low variance and is therefore particularly suitable for positioning. Finally, we implemented a demonstrator application using weighted centroid localization to show the practical implementation.

Obstacles are but pleasing for many aspects of large real-world sensor networks. Among other things, the presence of obstacles distort the sensor node localization process and might lead to costly routing because of unnecessary detours and/or dead ends. In order to relieve these problems, this paper proposes a distributed obstacle localization algorithm, called DOLfor short, in which the sensor nodes interact with each other mostly locally. The proposed algorithm is very resource and communication efficient in that all sensor nodes send only a small number of additional messages. Finally, the sensor network fine tunes the employed routing algorithm.

Today, an increasing number of customers subscribes for a high bandwidth Internet access. But not only communication speed is demanded. Quality-of-service moves more and more into the customers' focus. Both carriers and Internet Service Providers (ISPs) have increasing requirements derived from new services they want to offer to their customers. We present a new hardware solution is presented, which can satisfy many of these upcoming demands. This solution is highly flexible and can be adapted to various applications. The MAC address translation - MPLS user network interface (MATMUNI) provides the functionality of MAC address translation (MAT), multi protocol label switching-user network interface (MPLS-UNI), and a traffic manager (TM). This way, a module implemented on a single FPGA offers a wide range of functionality in an access network for low costs and high flexibility. The selection of an FPGA as implementation target offers the possibility to adapt to future demands towards functionality. Moreover, the all functional elements work with wire speed. Only a negligible delay is inserted into the datapath

Autonomous localization of nodes in wireless sensor networks is essential to minimize the complex self-organization task and to enhance network lifetime. Known techniques such as distance estimation based on received signal strength are often inaccurate and produce outliers. We propose a new method to measure a distance using the minimal transmission power between a transmitting node and a receiving node. The determined distance is very precise and has a low variance. It is therefore suitable for localization which is exemplary demonstrated for the approximate "weighted centroid localization" algorithm

Summary form only given. The collaboration of countless tiny sensor nodes in a network promises an enormous potential of novel applications. Advances in miniaturization and integration of electronic and mechanical components will enable sensor nodes with a size of a few cubic millimeters in the near future. At the same time, an ongoing price decline will allow the deployment of sensor networks covering thousands of nodes and, as a consequence, replace conventional wired sensors in many areas. Besides application-specific tasks of a node, the entire network requires a conformance to dynamical system requirements. The development focus changes from the single result of a sensor node to the cumulative result of the network. Consequentially, the following requirements for the design and implementation process of sensor networks arise: (i) sensor networks have to be self-organizing, (ii) sensor nodes must perform tasks of network maintenance, (iii) cooperative processing of tasks should lead to more precise results and new application fields, (iv) sensor networks require security mechanisms that are adaptive to environmental conditions, (v) all algorithms and protocols must be optimized with respect to resources, i.e. energy. The key constraint is energy. While many research teams are investigating better means to store, save, or generate electrical energy, progress is still slow. The paper addresses the inherent limitations and arising research opportunities of wireless sensor networks, both in terms of hardware and software issues.

Due to their layered approach, networks-on-chip (NoC) are a promising communication backbone in the field of heterogeneous dynamically reconfigurable systems. In this paper a future FPGA architecture is discussed having a hardwired NoC as an additional high-level routing resource. Instead of implementing on-chip interconnection with valuable reconfigurable resources, on top of this architecture, cost-efficient statically and dynamically reconfigurable systems can be built. The concept of such an FPGA is explored by means of an abstract SystemC model. This model not only implements the NoC but also permits a tile based dynamic reconfiguration. It is shown, that this approach advances the research on operating system support for dynamic reconfiguration in a new way.

Two economic factors are essential for the success of wireless sensor networks as new key technology: low-cost hardware and strong prototype applications. Although not perfect, Bluetooth can be a driving technology in this domain. We present a new data collection protocol (DCP) for wireless sensor networks. DCP is tailored to Bluetooth-based sensor nodes and therefore enables sensor network applications based on inexpensive hardware. DCP is scalable, robust, and not limited to piconet or scatternet structures.

Software development for wireless sensor networks requires novel programming paradigms and technologies. This article describes the concept of a new service oriented software architecture for mobile sensor networks. With this architecture, a flexible, scalable programming of applications based on an adaptive middleware is possible. The middleware supports mechanisms for cooperative data mining, self-organization, networking, and energy optimization to build higher-level service structures. The purpose of our research activities is the development of a framework, which radically simplifies the development of software for sensor network applications.

The realization of fast datapaths in signal processing environments requires fastest, power efficient logic styles with synchronous behavior. This paper presents a method to combine improvements on algorithm and logic level. To reduce the power consumption of dynamic logic, a method for using single-rail structures is presented including a new scheme to realize inverting logic functions. It is shown that such structure is most efficient when redundant number systems are utilized. These self-timed logic is integrated in a global clock system using the Asynchronous Chain True Single Phase Clock (AC-TSPC) logic resulting in a latch-free structure. Comparisons with other logic styles show the achievement potential. First simulations for a horizontal redundant adder slice show area and power savings of 40% and 30% compared to complementary Domino logic.

In some applications the observed samples are inherently nonuniform. In contrast to that in this paper we take advantage of deliberate nonuniform sampling and perform DSP where the classical approaches leave off. For instance think about mobile communication or digital radio. Deliberate nonuniform sampling promises increased equivalent sampling rates with reduced overall hardware costs. The equivalent sampling rate is the sampling rate that a uniform sampling device would require in order to achieve the same processing bandwidth. While the equivalent bandwidth of a realizable system may well extend into the GHz range its mean sampling rate is usually in the MHz range. Current existing prototype systems achieve 40 times the bandwidth of a classic DSP system that would operate uniformly (Artyukh et al. (1997)). Throughout the literature on nonuniform sampling (e.g. Bilinskis et al. (1992), Marvasti (2001), and Wojtiuk (2000)) many sampling schemes have been investigated. In this paper the authors discuss a nonuniform sampling scheme that is especially suited to be implemented in digital devices, thus, fully exploiting state-of-the-art ADC without violating their specifications. An analysis of the statistical properties of the algorithm is given to demonstrate common pitfalls and to prove its correctness.

In some applications the observed samples are inherently nonuniform. In contrast to that in this paper we take advantage of deliberate nonuniform sampling and perform DSP where the classical approaches leave off. For instance think about mobile communication or digital radio. Deliberate nonuniform sampling promises increased equivalent sampling rates with reduced overall hardware costs. The equivalent sampling rate is the sampling rate that a uniform sampling device would require in order to achieve the same processing bandwidth. While the equivalent bandwidth of a realizable system may well extend into the GHz range its mean sampling rate is usually in the MHz range. Current existing prototype systems achieve 40 times the bandwidth of a classic DSP system that would operate uniformly (cf. [Y. Artyukh, et al., Evaluation of Pseudorandom Sampling Process, 1997] and [Y. Artyukh, et al., virtual Oscilloscope of the DASP-Lab System 1997]). Throughout the literature on nonuniform sampling (e. g. [I. Bilinskis and A. Mikelsons, 1992], [F. Marvasti, 2001] and [J.J. Wojtiuk, 2000]) many sampling schemes have been investigated. In this paper the authors discuss a nonuniform sampling scheme that is especially suited to be implemented in digital devices, thus, fully exploiting state-of-the-art ADCs without violating their specifications. An analysis of the statistical properties of the algorithm is given to demonstrate common pitfalls and to prove its correctness.

This article describes the current work to extend the Java processor Java Silicon Machine (JSM) for usage in embedded systems. The JSM is a JavaCard processor supporting all JavaCard bytecodes. The JSM is a fully synthesizable 32 bit processor soft core with a very small footprint. The capability of it's integration in small embedded and automation systems is outlined. Special target platform is the SmartDev system which consists of a Java core interfacing to a wide variety of peripherals. SmartDev is intended to be used in mobile embedded systems for administrational, controlling and measurement purposes.

Sedlak (1987) proposed a modulo multiplication algorithm which is suitable for smart card implementation due to its low latency time. It is based on ZDN (Zwei-Drittel-N) arithmetic using an interleaved serial multiplication and reduction to calculate the product P=AB mod M. It can be shown that the maximum average reduction rate is theoretically limited to 3 bit/operation. In this paper we propose a modified left-to-right signed digit re-coding algorithm to receive an average shift of 4.5 bit/operation. Based on the presented ideas we also propose a modified reduction algorithm giving an average reduction rate of 4.5 bit/operation too. The speed up of our algorithms compared with the original algorithm is therefore 50%.

An efficient implementation of modular exponentiation, i.e., the main building block of many public key cryptographic devices, is achieved by algorithmic optimization of the Montgomery modular multiplication algorithm based on multiple precision such that pre-computation of N'/sub 0/=-N/sub 0//sup -1/ mod W can be avoided. This can be attained by modifications of the multiplier used.

In this paper we describe an approach for using digit online arithmetic in the field of neural network computation. Digit online, a serial most significant digit first arithmetic, shows significant advantages over all other digital implementations. The serial communication between the online modules make the implementation of connection intensive networks feasible. The accuracy of the computation is only loosely coupled with the chosen digit level range, which determine the necessary count of interconnections. Furthermore, the accuracy is eligible through the length of the processed digit vector. The goal of this paper is to develop a strategy for the implementation of different network models. The comparison with the results of other implementations illustrate the advantages of the digit online approaches and the suitability for the application in the field of neural networks.

Scheduling time impact on system performance increases especially when using dynamic priority algorithms, because of the enlarged computational effort at runtime. This overhead can be reduced by using dedicated hardware that does the time consuming computations necessary for scheduling. This can be a coprocessor capable of implementing dynamic scheduling algorithms which are, until now, rarely used because of their complex computations at schedule time. One of these algorithms is Least-Laxity-First (LLF). This is an optimal scheduling methodology that allows detection of time constraint violations ahead of reaching a task's deadline, but has the disadvantage of showing poor runtime behavior in some special situations ("thrashing"). In this paper, we present a universal deterministic scheduling coprocessor that implements the newly developed Enhanced Least-Laxity-First-algorithm (ELLF) which eliminates this disadvantage of LLF. Computation time of this device is rather a matter of time resolution than of the number of tasks.

No abstract available.

In 1996, about 600 million IC cards were manufactured worldwide. Due to very small die sizes (max. 25 mm/sup 2/) smartcards encounter more severe restrictions than conventional coprocessors. We study coprocessor architectures for very fast but area efficient modular exponentiation (FME) based on Montgomery multiplication. For assessment purposes we developed an evaluation board containing a 8051 microprocessor, a XILINX FPGA and RAM with variable bus width (8b to 32b). We evaluated these architectures in terms of the main design parameters to ease design decisions for smartcards in arbitrary technologies.

The CORDIC algorithm is used in many fields of signal processing for computation of elementary functions. Its main advantages are versatility and simplicity. When implemented in a word parallel pipeline it yields the highest possible throughput. However this solution is accompanied with increased hardware complexity and chip area requirements. The goal of this paper is to develop redundant CORDIC pipeline architectures yielding very low chip area. The speed does not decrease at all when compared with other proposals. Our novel architectures result in the smallest redundant CORDIC implementation known to the authors. It also exhibits considerably less gate switching activity thus also reducing power consumption.

One of the main problems of the CORDIC algorithm is the limited convergence domain, in which the functions can be calculated. Two different approaches can be employed to overcome this constraint: first, an argument reduction method and, second, an expansion of the CORDIC convergence domain. While the first approach requires significant processing overhead due to the need for divisions especially for tanh/sup -1/, the second technique achieves an increased but still limited convergence domain only. In this brief contribution, we present a unified division-free argument reduction method and a regular pipeline/array architecture for floating point or fixed point implementations which results in savings of computation time. In contrast to previous methods we avoid extra CORDIC arithmetic for realization of argument reduction.>

Several methods for increasing the speed of the CORDIC algorithm are presented. First, an improved method which guarantees a constant scale factor when employing redundant addition schemes is developed. Then, an architecture with increased parallelism which considerably reduces the CORDIC latency time and the amount of hardware is described.>