A Review on Neural Amplifier Design for Brain–Machine Interface

Brain–machine interface (BMI)-based neural signal recording is a new approach that is frequently used to track brain activity. Neural signals are recorded using microelectrode implanted in the brain. The collected signal from the electrode has a low amplitude and a high noise level. It requires a high-performance neural amplifier system that has highly immune to noise and with high gain for recording neural local field potential (LFP) and action potential (AP) signals. Integrated circuit (IC) technology can be used to develop and construct neural amplifier-based systems internally, leading to ultra-large-scale integration (ULSI) or application-specific integrated circuit (ASIC) solutions. Due to its small form factor and lightweight with low power consumption, the IC-based neural amplifiers are now deployed in portable easy to carry neural recording system. In this paper, the review will focus on various neural amplifiers design topology and operational transconductance amplifier (OTA) architecture and their pros and cons based on their performance in terms of noise efficiency and power consumption.