A Low-Power ECoG/EEG Processing IC With Integrated Multiband Energy Extractor

摘要

Electrocorticography (ECoG) implants have recently demonstrated promising results towards potential use in brain-computer interfaces (BCIs). Spectral changes in ECoG signals can provide insight on functional mapping of sensorimotor cortex. We present a 6.4 $mu$ W electrocorticography (ECoG)/electroencephalography (EEG) processing integrated circuit (EPIC) with 0.46 $mu V_{rm rms}$ noise floor intended for emerging brain-computer interface (BCI) applications. This chip conditions the signal and simultaneously extracts energy in four fully programmable frequency bands. Functionality is demonstrated by tuning the four bands to important frequency bands used by ECoG/EEG applications: $alpha$ (8–12 Hz), $beta$ (18–26 Hz), low-$gamma$ (30–50 Hz), and $gamma$ (70–100 Hz). Measured results from in vivo ECoG recording from the primary motor cortex of an awake monkey are presented.