Resource Efficient Pre-processor for Drift Removal in Neurochemical Signals

Abstract

A necessary requirement for chemometric platforms is pre-processing of the acquired chemical signals to remove baseline drift in the signal. The drift could originate from sensor characteristics or from background chemical activity in the surrounding environment. A recent emerging field is neurochemical monitoring to detect and quantify neural activity. In this paper, a resource efficient pre-processing system is presented to remove drift from the acquired neurochemical signal. The drift removal technique is based on baseline manipulation without requiring window based processing. The target application, for demonstration purposes, is the recording of vagal pH signals to enable closed-loop Vagus Nerve Stimulation (VNS). The final design is multiplier-free and results in an Application Specific Integrated Circuit (ASIC) that is 640 μm by 625 μm in area.