A sigma-delta architecture for recording of peripheral neural signals in prosthetic applications

A recording module for peripheral neural signals is presented. The proposed device, based on a sigma-delta architecture, is made up of two main parts: an analog module as front-end stage for neural signal acquisition, pre-filtering and sigma-delta modulation and a digital unit for sigma delta decimation and system configuration. The analog module provides a gain of 200V/V in a 800Hz-8kHz frequency range, while the sigma-delta converter grants a 9bit resolution and a good noise rejection thanks to the 32 order decimator IIR filter. Behavioural and transistor-level simulation results confirm that the system is capable of recording neural signals in the order of magnitude of tens of μΥ eliminating the huge low frequency noise due to electromyographic interferences.