A Low-Power, Area-Efficient ADC-Direct Artifact-Tolerant Neural Recording System Using Digital Block Sharing

This article presents a front-end block for recording neural signals of the direct Analog-to-Digital Converter(ADC) type based on the continuous-time Delta-Sigma Modulator (CT-DSM) to reduce power consumption and occupied space. The system works as a CT-DSM with a single-bit quantizer in the artifact-free state. When the artifact is present, the DSM is saturated, which is detected by the digital block, and the second feedback path estimates the amplitude of the artifact. By reducing the amplitude of the artifact from the input signal, the system can convert the neural signal when the artifact is present. The process of designing and implementing the proposed circuit is based on three general ideas of improved circuit design of the block with the highest power consumption, using a 7-bit counter to detect the saturation of the DSM and sharing the digital part. The implementation of the transistor level in CMOS technology is TSMC 0.18um with a channel area of 0.013mm2 and power consumption of 4.6uW.