Design of a Configurable Low-Noise 1-Channel Analog Front-End for EEG Signal Recording on 180nm CMOS Process

In this paper, the authors present a 1-channel Analog Front End (AFE) circuit for recording Electro-encephalogram (EEG) signals. The AFE is capable of supporting a variety of sensor inputs, a high input impedance, programmable gain, low noise, and wide bandwidth for EEG recording systems. The buffer or current-to-voltage converter (BCV), which can be configured to operate as a buffer or a current-to-voltage converter, is placed between the electrodes and the main amplifier stage of the AFE to achieve high input impedance and work with sensor signal types. The BCV is constructed with an Op-Amp to obtain high input impedance throughout the AFE. As the main amplifier stage of the AFE, a chopper capacitively-coupled instrumentation amplifier(CCIA) is positioned after the BCV in order to reduce input-referred noise and balance the impedance of the entire AFE design. The third stage of the AFE is a programmable gain amplifier (PGA), which allows adjustment to the overall gain of the AFE. The proposed AFE is designed and simulated on a 180nm CMOS process and operates in a wide frequency band from 0.5 Hz to 2 kHz with high input impedance greater than 2T$\Omega$. The AFE can achieve low noise efficiency with low input-referred noise of 1.8-$\mu$Vrms, and the minimum 100-dB CMRR. The AFE can be applied to acquire EEG signals, particularly in detection of epilepsy and seizures.