Multisensors Signal Processing Using Microcontroller and Neural Networks Identification

An approach of multisensor signals processing at microprocessor or 8 bit microcontrollers is showed in this paper. This approach is based on multisensor individual conversion function or individual characteristic curve identification method with the use of neural networks. It allows to reduce an amount of calibration points and increase the precision of identification with comparatively to well known methods. The realization of this approach was made on developed system for ultraviolet level measurement based on 8-bit microcontrollers.