A 32-Channel Time-Multiplexed Artifact-Aware Neural Recording System.

2021 
This paper presents a high dynamic range, lowpower, low-noise mixed-signal front-end for the recording of local field potentials or electroencephalographic signals with invasive neural implants. It features time-multiplexing of 32 channels at the electrode interface for area saving and offers the ability to spatially delta encode signals to take advantage of the large correlations between nearby channels. The circuit also implements a mixed-signal voltage-triggered auto-ranging algorithm which allows to attenuate large interferers in digital domain while preserving neural information, thus effectively increasing the dynamic range of the system while avoiding the onset of saturation. A prototype, fabricated in a standard 180 nm CMOS process, has been experimentally verified in-vitro and shows an integrated input-referred noise in the 0.5200 Hz band of 1.4 Vrms for a spot noise of about 85 nV/Hz. The system draws 1.5 W per channel from 1.2 V supply and obtains 71 dB + 26 dB (with artifact compression) dynamic range, without penalising other critical specifications such as crosstalk between channels or common-mode and power supply rejection ratios.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    0
    References
    0
    Citations
    NaN
    KQI
    []