5.5 A 2µW 40mVpp linear-input-range chopper- stabilized bio-signal amplifier with boosted input impedance of 300MΩ and electrode-offset filtering

Modern neuromodulation requires closed-loop functionality, where neural recordings are used to adapt stimulation patterns in real time. A closed-loop system requires the neural sensing front-end to record small neural signals in the presence of large stimulation artifacts. The amplitude of artifacts can be a few 10s of mV, and their power is usually in the same frequency band as the signals of interest, requiring non-traditional adaptive filtering to attenuate the artifacts. This requires a sensing front-end that can handle large signals while maintaining the signal integrity of the accompanying small neural signals. State-of-the-art front-ends saturate beyond an input of ~5mV and have limited linearity, making them incapable of handling large artifacts. This work presents a front-end that can tolerate up to ±20mV artifacts in the signal band of 1Hz to 5kHz. To digitize a 1mV neural signal to 8 bits in the presence of a 20mV artifact, the front-end requires a 12b linearity.