Noise Effect Reduction on a MEMS Based AC Voltage Reference Source Using Artificial Neural Network

this paper presents a new method in order to reduce noise effect in an ac voltage reference source. the ac voltage reference source is implemented on mems technology. it uses capacitive mems technology. the reference is based on the characteristic ac current-voltage curve mems component. the multilayer neural network is used. the neural network (nn) uses the levenberg-marquardt (lm) method for training. the noise effect on an electronic circuit is investigated. the simulation results are very promising.