Research Field D: Distributed Microsystems

The communication within a network of microsystems is a capability enabling numerous applications. The investigations carried out in this research area and the methods developed in it are focused on the requirements of microsystems and take into account the issues of energy efficiency and communication robustness.

Projekte im Forschungsbereich D Verteilte Mikrosysteme:

D.1

Communication in Embedded Microsystems

D.2

Energy Efficient Automatisation in Distributed Microsystems

D.3

Modular, Decentralized, Distributed Microsystems for Instrumentation Applications

D.4

Microsystems with Wireless Communication

D.1 Communication in Embedded Microsystems

In this project we will investigate communication problems of current interest that arise in distributed microsystems. Our research will be of an algorithmic and theoretical nature. The goal is to gain insight into the principles how data transmission and exchange can be performed efficiently in modern microsystems. An important aspect of the project is that the newly designed algorithms can indeed be used in microsystems with respect to the required resources (time, space, energy consumption). In a first phase we will study energy-efficient broadcast and associated network topologies.

D.2 Energy Efficient Automatisation in Distributed Microsystems

Traditional architectures used for production automation require wired connections between sensors and actuators as well as between the components of a hierarchical control system. This structure does not provide much flexibility, e.g., if production requirements are changing, but it allows the development of reliable controllers that achieve real-time demands.

Because of their enormous flexibility and their moderate costs, autonomous micro systems will be an attractive tool for process automation in future. They built up a new distributed architecture of several nodes, which communicate with each other wireless and share different tasks of control within the network (see Figure). This new architecture will introduce additional control problems, which will be solved within this research project:

• Within a control loop the signal has to pass several nodes. On that way, cancellations, delays or constrains in signal bandwidth may disturb the signal. Methods have to be developed to cope this problems and to achieve stabile and robust control in these Networked Control Systems.
• Development tools are missing for these new architectures that provide a reliable control of the plant. Automated code generation (e.g. from Simulink) may be extended for the case of distributed systems.
• In principal, new sensor and actuator nodes may be integrated in an existing network rather easily. For this purpose, an automated identification of the plant or a self-calibration should be developed.
• If the microsystems are integrated in a mobile machine part, it must also be autarkic regarding its energy supply. The management of the precious energy stored in the micro system is a core issue. Since computing power needs also energy, this must be considered for the development of algorithms. Communication will have the same restriction. Hence, data transmission should be reduced to the minimum.

Together with other research projects of this graduate school prototypes will be developed, in order to verify the obtained results in practice.

Sensors (S), actuators (A), and controlles (C) are autonomous microsystems within an ad-hoc network that controls a production plant.

D.3 Modular, Decentralized, Distributed Microsystems for Instrumentation Applications

For a local measurement technique a complete micro laboratory with sensors, actuators and electronics plus data exchange is a necessary requirement for a continuous monitoring of physical and biological chemical parameters in medicine or process technology. Therefore micro temperature sensors and biosensors with micro fluidics and telemetric units have to be implanted (see figure 1a). The thermal sensors on the basis of amorphous germanium should make it possible to measure the thermal conductance capacity but also the convection with dynamic driven micro heating elements. In connection with already developed chemo sensors and biosensors (figure 1b) a complete local measurement laboratory has to be realized and characterized. With this, different sensor modules can be put on machine components, at home, at the body or in the clothes which control the status of the system respectively of the patient to every time. Therefore microsystems of different construction have to be assembled in a hybrid modular way (HYMOS). The aim is to produce different functional groups in the preferred technology and implement them in hybrid technology to an embedded system, which has to own an autarkic intelligence. These autarkic system modules should be connected with sensor buses and accept a wireless communication with the environment. So e.g. the temperature, the acceleration, the position, but also the bio/chemical parameters can be measured in different positions, simultaneously the measurement can be analyzed and the complete system status can be telemetrically delivered to an external station. In the case of sensors, temperature sensors and biosensors are produced with microsystem technology, common miniaturized pumps are implemented and connected to a hardware with the help of ASICs and telemetric chips. These local measurement modules are distributed arranged to do multi parameter monitoring at different positions. So, one gets a complex measurement matrix that is highly demanding for real-time data fusion. A sensor management program has to be established for the system software to implement data reduction, an adequate field bus technique and a telemetric analysis in the sensor module and in the module nets. An adequate data flow management has to be developed which adaptively takes an optimal analysis algorithm in regard to the measurement task and the incidental data flow.

Contact:
Prof. Dr. Gerald Urban
Chair of Sensors
Department of Microsystems Engineering
Georges-Köhler-Allee 103, 79110 Freiburg
Phone: +49-761-203-7260 | Fax: +49-761-203-7262 | E-mail

D.4 Microsystems with Wireless Communication

Wireless services will be more and more significant in industrial, private or for example medical applications. In many of these applications there is an increasing need to measure sensor values and to control actuators at inaccessible or rotating locations. Furthermore an interaction between the sensors and actuators is required. These applications can be established with so called autonomous microsystems. Such microsystems are capable to communicate with each other or via a base station, utilizing an RF technology. For a wide distribution of these microsystems, they must be low-cost and miniaturized. Goal of this project is to find technical solutions for the implementation of smart, wireless, autonomous microsystems. It should be investigated, how to adapt such microsystems to various sensors and boundary conditions, and how it can be itself tested and configured. Furthermore in this project it should be explored to implement novel Wireless Sensor Actuator Networks (WSANs), which could be solve many existing instrumentation and measurement problems. Both the hardware and the software of autonomous microsystems must be investigated. For example power aware approaches for the wireless communication technology or the operating system must be found, which enables a flexible and also ad-hoc communication in the network.

Contact:
Prof. Dr. Leonhard Reindl
Chair of Electrical Instrumentation
Department of Microsystems Engineering
Georges-Köhler-Allee 103, 79110 Freiburg
Phone: +49-761-203-7220 | Fax: +49-761-203-7222 | E-mail