Center for Biomedical Technology Christian Doppler Laboratory for Innovative Therapy Approaches in Sepsis Center for Regenerative Medicine and Orthopedics Center for Integrated Sensor Systems Micro- and Nano Sensors Distributed Systems and Sensor Networks Modeling and Simulation Water and environmental sensors Publications Team Scientific Events Awards Cooperations Infrastructure Contact Center for Management in Healthcare Center for Traditional Chinese Medicine and Complementary Medicine Center for Medical Specialisations Center for Interdisciplinary Dentistry News Courses Research and Consulting Co-operations Advisory Board Team

Integrated Sensor Systems

Noticeable or not, sensors are part of our daily life. They are the basis for innovative solutions in such diverse application fields as environmental monitoring, medical technology, or industrial automation, but also household appliances or cars. As different as these application domains are the sensor principles employed. And still there is the need for novel sensors tailored to new applications.

The traditional way to develop a new sensor is to start with the analysis and development of a new measurement principle. In concurrent steps boundary conditions are considered, and appropriate analogue and digital processing of sensor signals is attached. Finally, after all these steps have been finished, advanced topics such as data processing, signal correction, and communication technology will be tackled to achieve optimal measurement and integration in the overall system. Usually this post-processing is done by external electronic devices.

Although this bottom-up approach is quite common for research and is valuable for proof-of-concept prototyping it is in general not feasible to achieve optimal results for a particular target application. For instance, despite outstanding measurement principles a sensor might not be applicable since it might not fulfil boundary conditions of the application or is incompatible with the communication and signal processing facilities of the superordinate system.

Focusing research and development activities on the pure sensor frontend or transducer thus is no longer adequate for modern applications. Converting the actual measurand into an electrical signal is only one aspect. The particular challenges today are also the embedding of the sensor into a wider context and the networking with other sensors, in brief: the integration into a complete system. This in turn means that no longer a sensor per se is in the focus of interest, but a sensor system. The development of such systems not only requires know how in the classical field of sensor technology, but equally demands expertise in microelectronics and integrated circuit design, algorithm design and software engineering, embedded systems as well as communications and network technology.

Integration aspects in integrated sensor systems can be manifold, and all of them must be equally considered. Most obvious, integration is related to circuit design, with respect to miniaturization of the sensor or the components relevant for signal processing. On the other hand, integration has a functional facet, comprising mechanisms to increase fault tolerance, but also to combine various sensor principles in order to improve accuracy, measurement range, or robustness. Last but not least, integration has to be seen in a system-related context – in terms of horizontal and vertical interconnection of sensors and higher-level systems. This integrative operational approach is based on two premises:
A modern sensor-actuator system comprises a transducer capturing the measurand, an intelligent controller responsible for signal conditioning, conversion, and processing, as well as communication facilities. These modules form a functional unit, but need not necessarily be monolithically integrated.
Independent optimization of the three modules does not necessarily lead to an optimal complete system. Rather, mutual design matching is indispensable. Often feedback or active approaches are to be preferred.
Successful design of sensor systems therefore includes much more than optimised transformation of the physical measurand into a value that can be processed more easily. Signal buffering, continuous recalibration, self diagnosis, signalisation of faults, but also alarm functions or reporting are functions that are required for state-of-the-art sensor systems. Energy efficiency is another important topic especially for wireless sensor systems. Only the integration of all these extended aspects as well as an overall optimization will transform a sensor into a successful integrated sensor system.
back to mission statement
back to main page