Reconfigurable smart sensors for dependable systems

The realisation of dependable systems requires improved reliability and confidence assessment of the data obtained from sensors within the system. This may be achieved at the system level by sensor validation techniques through the systematic analysis of redundant sensor data using analytic or spatial redundancy. In automotive applications where cost is a major factor, the use of redundant sub-systems needs to be avoided wherever possible, so some form of self validation becomes necessary. In this paper, an integrated diagnostic reconfiguration (IDR) approach is presented using the application example of an interface ASIC for a Wheatstone bridge pressure sensor. This approach provides improved reliability and confidence assessment within the smart sensor by the integration of fault tolerant reconfiguration and internal self validation. Capabilities and limitations of the approach are discussed along with an overview of the design methodology employed. An example of the use of IDR to detect and isolate a functional fault is included with a description of fault tolerance through the use of exclusion configurations