An Ultra-Low Power RSSI Amplifier for EEG Feature Extraction to Detect Seizures

This brief presents an ultra-low power received signal strength indicator (RSSI) amplifier circuit that can be used to detect seizures through feature extraction from electroencephalography (EEG) signals. The RSSI-based feature extraction method provides a low-power area-efficient solution for analog computing based seizure detection hardware. A 6-stage RSSI amplifier circuit was designed in 65nm CMOS technology to cover the dynamic range of EEG signals with an area of ${266\mu m \times 531 \mu m}$ . Measurement results of the RSSI circuit show that it has a power consumption of 31.6nW, covers a 45dB dynamic range, and has a linearity error of less than ±1dB.