C1.3 - Statistical Analysis of Compensating Properties of Reconfigurable Analog Circuits for Generic Self-X Sensor Interface

SENSOR+TEST Conferences 2009
2009-05-26 - 2009-05-28
Congress Center Nürnberg
Proceedings SENSOR 2009, Volume I
C1 - Sensor Electronics I
S. Lakshmanan, A. Koenig - Technische Universität Kaiserslautern, Kaiserslautern, Germany
263 - 268


Nowadays, sensor signal conditioning is very crucial and challenging, as the type and variety of sensors coming into the market are rapidly increasing. Moreover, developing signal conditioning IC’s are function of area, cost and robustness to maintain signal integrity. Field programmable analog approaches and the most recent evolvable hardware approaches offers solution for certain extent. Flexibility, and reliability are key issues in designing and operating such systems. However, issues related to fault tolerance and preserving of sensor system performance, quality of registration, and signal processing are not covered by the discussed approaches. The large variety of different sensor principles demands for a large range of sensor electronics interfacing to the sensing elements to perform signal conditioning before conversion to the digital domain. Predominantly, hard-wired solutions with off-the-shelf components are employed, which imply low flexibility with regard to drifts and changes. More recent approaches try to use reconfigurable analog arrays to achieve rapid-prototyping and sufficient flexibility for the sensor electronics. One industrial example is the Vortex family of analog array chips provided by Anadigm Inc., which consist of amplifier and comparator cells and digitally configurable capacitors, that are used as pseudo-resistors and capacitors along with CMOS-switches in the feed-back of the amplifier cells in a Switched-Capacitor (SC) approach. PGA 309 from Texas Instruments is another type of sensor signal conditioning chip available in the market. Moreover, in the field of evolutionary electronics/evolvable hardware, circuits synthesis are carried out by learning procedures on a flexible transistor level granular hardware structure called Field Programmable Transistor Arrays (FPTA). The rate at which novel and new sensor products coming to market and usage are tremendously increasing. Hence development of sensor interface electronics is still a challenging task, which needs special know-how and experience in a multi-disciplinary field of electronics, physics, mechanics etc. However, for the front end communication from sensors to the interface electronics, no standards/general interface electronics for all sensors are available. Therefore optimization of the sensor electronics for the numerous types of electrical signals and sensor characteristics e.g., V, I, C, R, L based-inputs, requires dedicated designs for each of the particular sensor elements. As a result, in spite of the huge industrial interest, the development of sensor systems progresses rather slowly. For example, QuantumX is a discrete product available in the automation market which uses several dedicated types of conditioning for the various sensor type signals, basically working with “one sensor- one Asic” approach. Therefore the main aim of our research work is to address and provide solutions to these problems by combining the sensor industrial demand, and programmable analog arrays with the evolutionary concepts to built a generic, flexible, self-x sensorinterface chips, which are easy to use and can be applied for a wide variety of sensor related products.