Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Design for miniaturization of wireless sensor nodes based on 3D-packaging technologies

: Niedermayer, M.; Thomasius, R.; Polityko, D.-D.; Schrank, K.; Hefer, J.; Guttowski, S.; Reichl, H.

Gessner, T. ; MESAGO Messe Frankfurt GmbH, Stuttgart:
Smart systems integration 2007 : Paris, France 27. - 28.03.2007; With CD-ROM / 1st European Conference & Exhibition on Integration Issues of Miniaturized Systems - MEMS, MOEMS, ICs and Electronic Components
Berlin: VDE-Verlag, 2007
ISBN: 3-8007-3009-X
ISBN: 978-3-8007-3009-4
European Conference & Exhibition on Integration Issues of Miniaturized Systems - MEMS, MOEMS, ICs and Electronic Components <1, 2007, Paris>
Fraunhofer IZM ()
Assisted Personal Health; wireless sensor system; design methodology; hetero system integration

Introduction The ease of integration into everyday objects and the higher mechanical robustness are the fundamental advantages miniaturized wireless sensors. The development of these embedded micro systems has to consider the interdependence of design decisions regarding network communication, wireless sensor hardware and fabrication technology. A design environment was implemented to support the functional and physical design of miniaturized wireless sensor nodes. Design for Miniaturized Wireless Sensor Nodes Conventional methodologies structure the design process into a fixed sequence of design steps. The design steps for miniaturized wireless sensor nodes include the partitioning of: - distributed / local algorithms, - analog / digital hardware, - component / package elements, - planar 2D- / spatial 3D-interconnects. These design decisions are highly correlated in the case of high miniaturization degrees. The resulting higher number of design iterations is more efficiently managed by a flexible model based design. The approach of our design methodology for miniaturization is a parallel analysis of the functional miniaturization and physical miniaturization based on geometry models for size estimations. Functional Miniaturization The target of the functional miniaturization analysis is the selection and parameterization of the sensor node architecture, resulting in a minimal volume of the total component set. The miniaturization barrier for a given architecture depends significantly on the volume elements like batteries and crystals. The dimensions of bulky elements can often be influenced by a reduction of functional component requirements. The optimal solution is application specific especially regarding the calibration and cooperation strategies. For the consideration of relevant design issues, the self-sufficient wireless sensor nodes are divided into the four subsystems communication interface, ambient interface, digital back-end, and energy supply. Physical Miniaturization The physical miniaturization analysis examines the miniaturization barrier for a given component set according to selected interconnection and substrate technologies. The components are allocated on certain vertical layers and the required wiring density is calculated from the corresponding net list. For the first iterations, a coarse partitioning is sufficient for derivation of the optimization focus. Later on, placement and routing is required on different refinement levels accomplished by signal integrity examinations and thermal simulation steps. The introduction of the concrete implementation details is preceded by design relevant packaging issues. This ranges from the selection substrate technologies and chip interconnection technologies to alternative 3D module integration processes. Implemented Prototypes Prototypes of wireless sensor nodes were implemented to verify the design approach. Several miniaturization steps with different 3D system integration technologies were realized. Starting from modules of 2.6cm edge length in conventional SMD technology, the first step in miniaturization has shrunken the wireless sensor system to 2 cm per side. At this size, the modules were realized with bare dies by flip chip mounting. Later on, prototypes of 1 cubic centimeter were developed, based on a folded flexible substrate. Finally, flip chips on both substrate sides allowed folded modules of only 6mm edge length. Bibliografie (<1000) Michael Niedermayer received the M.S. degree in Electrical Energineering from the Technical University of Berlin in 1999 and the MBA degree from the College of Technology and Economy of Berlin (FHTW) in 1999. Afterwards he worked as research engineer engineer in the field of MEMS relays for Siemens and Tyco Electronics. In 2002, he joined the Fraunhofer Institute for Reliability and Microintegration as research manager. His research interests are design methodologies, packaging technologies and miniaturized wireless sensor systems Short Description (<2000) Visions like Pervasive Computing and Ambient Intelligence expect in the future that many daily items be linked by wireless communication. Tiny self-sufficient wireless sensor nodes, also discussed under the names of Smart Dust and eGrain, will contribute to the interconnection of diverse objects. Applications like access control, catastrophe management, technical inspections, and monitoring of environmental conditions will take advantage of the small size and high mechanical robustness. While wireless sensor systems of several cubic centimeters have reached the stage of mass production e.g. for tire pressure monitoring, devices of some millimeters per side are still the subject of research projects. The development of miniaturized sensor nodes is very challenging due to the close interdependence of design decisions regarding communication protocol, hardware component selection, and integration technology choice. Hence, new efficient design methodologies are required which extend the design procedures for embedded systems with miniaturization options for packaging technologies. A design environment was implemented to support the functional and physical design of miniaturized wireless sensor nodes. The quantification of the functional component requirements as well as the derivation of appropriate packaging technologies is based on an event driven behavior simulator and a 3D placement tool. The functional and physical design decisions are synchronized by the volume estimations of the individual components and the update of the model parameters by the actualized geometry model. Depending on the volume fractions the development focus is set accordingly. Prototypes were implemented to verify the design methodology. Several wireless sensor node architectures were realized by different 3D packaging technologies. The miniaturized prototypes have reached a size of 1cm down to 0.6cm edge length.