A 110-nW in-channel sigma-delta converter for large-scale neural recording implants

M Rezaei,E Maghsoudloo,M Sawan,B Gosselin
DOI: https://doi.org/10.1109/EMBC.2016.7592031
Abstract:Advancement in wireless and microsystems technology have ushered in new devices that can directly interface with the central nervous system for stimulating and/or monitoring neural circuitry. In this paper, we present an ultra low-power sigma-delta analog-to-digital converter (ADC) intended for utilization into large-scale multi-channel neural recording implants. This proposed design, which provides a resolution of 9 bits using a one-bit oversampled ADC, presents several desirable features that allow for an in-channel ADC scheme, where one sigma-delta converter is provided for each channel, enabling development of scalable systems that can interface with different types of high-density neural microprobes. The proposed circuit, which have been fabricated in a TSMC 180-nm CMOS process, employs a first order noise shaping topology with a passive integrator and a low-supply voltage of 0.6 V to achieve ultra low-power consumption and small size. The proposed ADC clearly outperforms other designs with a power consumption as low as 110 nW for a precision of 9 bits (11-fJ per conversion), a silicon area of only 82 μm × 84 μm and one of the best reported figure of merit among recently published data converters utilized in similar applications.
What problem does this paper attempt to address?