Klaus D. Müller-Glaser

Application of ubiquitous computing in personal health monitoring systems.

A possibility to significantly reduce the costs of public health systems is to increasingly use information technology. The Laboratory for Information Processing Technology (ITIV) at the University of Karlsruhe is developing a personal health monitoring system, which should improve health care and at the same time reduce costs by combining micro-technological smart sensors with personalized, mobile computing systems. In this paper we present how ubiquitous computing theory can be applied in the health-care domain.

MORPHEUS: Heterogeneous Reconfigurable Computing

Reconfigurable architectures and NoC (network-on-chip) communication systems have introduced new research directions for technology and flexibility issues, which have been largely investigated in the last decades. Exploiting the flexibility of reconfigurable architectures, the run-time adap-tivity through run-time reconfiguration, opens a new area of research by considering dynamic reconfiguration. Since software parts of an embedded system can also be included into reconfigurable hardware by integration of an IP-based microcontroller, the reconfigurable architecture provides a flexible, multi-adaptive heterogeneous platform forHW/SW co-design. In this paper, we present the European integrated project MORPHEUS (1ST 027342). Its goal is to develop new heterogeneous reconfigurable SoCs with various sizes of reconfiguration granularity and to provide an integrated toolset of spatial and sequential design that can be used for mapping and execution of the target applications. Additionally a NoC approach is included in order to demonstrate the mentioned benefits and scalability for actual and future SoC design. The power of this approach will be demonstrated with four applications from the industrial environment.

Multiparadigm modeling in embedded systems design

Embedded electronic systems for monitoring and control of technical processes (electronic control unit-ECU) are systems comprised of heterogeneous components (hardware, software, sensors, actuators, power electronics), thus making high demands on their development. Describing different aspects and views of the whole system, subsystem, or component requires according modeling paradigms for requirements specification, design, hardware implementation, software code generation, verification, integration, and testing. The first part of the paper surveys characteristic ECU features and describes a design strategy and the related technology, bringing out the necessity of multiparadigm modeling. Examples from automotive ECU applications are used throughout the paper. With respect to the problem that currently available tools provide insufficient support, integration strategies for multiparadigm modeling based on multiple tools are surveyed in the second part, concluding with examples from our own research activities.

A system architecture for reconfigurable trusted platforms

For improving the security of embedded systems, trusted computing is a promising technology. For the area of microprocessors in general and personal computers in particular the Trusted Computing Group (TCG) has published detailed specifications. The resulting hardware has been available for some years. This contribution discusses the feasibility of deploying ideas from trusted computing in the domain of reconfigurable hardware, esp. FPGAs, and possible benefits and drawbacks. We give a proposal to use actually available FPGA technology to build a trusted platform on reconfigurable hardware. We also show how trusted computing can deal with partial dynamic reconfiguration while still allowing the user to fully exploit its potentials.

A backplane approach for cosimulation in high-level system specification environments

The design of microelectronic systems including hardware and software for open-loop and closed-loop control requires the combination of expertise from different domains. However, no integrated approach to the specification and design nor to the analysis and simulation of the overall system is available. The design of such complex and heterogeneous systems mandates a systematic, computer aided approach to requirements definition, specification and design as well as to the verification and validation of the results. A simulation backplane concept is presented for parallel mixed-mode cosimulation using a standardized interface. Synchronization mechanisms are developed and used that allow an efficient concurrent cosimulation.

Estimating essential design characteristics to support project planning for ASIC design management

To enhance project planning and feasibility study for ASIC (application-specific integrated circuit) design, a chip estimation system (CES), tightly coupled with a project plan generator system (PGS) has been developed. The CES calculates chip area, speed, and power dissipation from data of a knowledge-based data acquisition system which gathers basic design characteristics, requirement specification data, and information about complexity and problems of a design. The resulting data are transferred to the PGS, which generates alternative project plans based on a design-style-specific knowledge base. Using vendor- and technology-specific cost factors, estimation of design time and design, production, and test costs is performed. The project-control system controls execution of a selected project plan.<<ETX>>

Smartphone pedestrian navigation by foot-IMU sensor fusion

Determining ones own position by means of a smartphone is an important issue for various applications in the fields of personal navigation or location-based services. Places like large airports, shopping malls or extensive underground parking lots require personal navigation but satellite signals and GPS connection cannot be obtained. Thus, alternative or complementary systems are needed. In this paper a system concept to integrate a foot-mounted inertial measurement unit (IMU) with an Android smartphone is presented. We developed a prototype to demonstrate and evaluate the implementation of pedestrian strapdown navigation on a smartphone. In addition to many other approaches we also fuse height measurements from a barometric sensor in order to stabilize height estimation over time. A very low-cost single-chip IMU is used to demonstrate applicability of the outlined system concept for potential commercial applications. In an experimental study we compare the achievable accuracy with a commercially available IMU. The evaluation shows very competitive results on the order of a few percent of traveled distance. Comparing performance, cost and size of the presented IMU the outlined approach carries an enormous potential in the field of indoor pedestrian navigation.

Failure mode and effect analysis based on electric and electronic architectures of vehicles to support the safety lifecycle ISO/DIS 26262

The draft international standard under development ISO 26262 (Road Vehicles — Functional safety —) describes a safety lifecycle for road vehicles and thereby influences all parts of development, production, operation and decommissioning. Starting from 2011, all developments of new cars should be aligned to this standard. The rapid application and adaption of the ISO 26262 is mandatory to develop safe, advanced and competitive automotive systems and systems of systems. The failure mode and effect analysis (FMEA) is a well applied engineering quality method in the automotive industry and proposed by the ISO 26262 for several analyses. The communication structure of the automotive control system are specified by the electric and electronic architecture (EEA). For a short time all this information can be processed in one tool. It can form an important contribution to the determination of input data for safety assessments. With the FMEA flow embedded in the EEA modeling, analysis can be rapidly provided with altered input data resulting from architecture modifications. This paper presents a formalized tool flow for rapid determination and accumulation of input data for failure mode and effect analysis based on an EEA model, the accomplishment of the analysis within an EEA modeling tool and the automated generation of reports, documenting the results from the FMEA according to a predefined form.

On automatic testpoint insertion in sequential circuits

A novel tool, called TIP (testpoint insertion program), that serves to improve the testability of a sequential circuit is presented. It automatically identifies locations at which test cells may be inserted is presented. For that purpose first the basic cycles and basic reconvergencies are calculated. Out of these almost all cycles are deduced. The resultant critical paths are broken at a set of locations that are selected by iteratively choosing the one which breaks the most critical paths. Heuristics are given to minimize the number of selected locations, as are ideas about further improvements. The algorithm was tested on sequential benchmarks and compared with a commercially available automatic test pattern generator. Also, the combined test-cell insertion by TIP gives accurate information regarding where to modify a given circuit to achieve a high fault coverage with optimized overhead. The proposed algorithm leads to smaller computational run times and less area overhead.<<ETX>>

On the fusion of inertial data for signal strength localization

Localization in wireless sensor networks (WSN) based on the evaluation of received signal strength (RSS) values of communication packets has received a considerable amount of research interest in the last years. In spite of the number of approaches that have been presented, the performance of most systems remains rather poor due to the unpredictable behavior of RSS values and the resulting fluctuating position estimations. In this paper, an approach to cope with these fluctuations for the purpose of ad-hoc person localization in WSN is presented. A hip-mounted inertial measurement unit (IMU), carried by the person to localize, is used to collect data on the current movement. This data is used to stabilize the RSS position estimation by means of a Kalman filter. The experimental evaluation shows that the proposed method carries the potential to improve RSS localization methods. We present results of experiments in in- and outdoor environments and quantify the improvements that can be achieved by fusing inertial data with RSS position estimation schemes for person localization. It can be concluded that fusing inertial data can improve the performance of RSS-based localization systems.