Neural network-based self-calibration/compensation of sensors operating in harsh environments [smart pressure sensor example]

A novel artificial neural network (NN)-based scheme for smart sensors operating in harsh environments is presented. The NN-based sensor model automatically calibrates and compensates with high accuracy for the nonlinear response characteristics and nonlinear dependency of the sensor characteristics on the environmental parameters. Through extensive simulated experiments, we have shown that the NN-based capacitive pressure sensor (CPS) model can provide pressure readout with a maximum full-scale error of only 1.5% over a temperature range of -50 to 200/spl deg/C for the three forms of nonlinear dependencies.