An Guanglei, Hutchens Chriswell, Rennaker Robert L
Annu Int Conf IEEE Eng Med Biol Soc. 2014;2014:6557-60. doi: 10.1109/EMBC.2014.6945130.
A low power, low noise implantable neural recording interface for use in a Radio-Frequency Identification (RFID) is presented in this paper. A two stage neural amplifier and 8 bit Pipelined Analog to Digital Converter (ADC) are integrated in this system. The optimized number of amplifier stages demonstrates the minimum power and area consumption; The ADC utilizes a novel offset cancellation technique robust to device leakage to reduce the input offset voltage. The neural amplifier and ADC both utilize 700mV power supply. The midband gain of neural amplifier is 58.4dB with a 3dB bandwidth from 0.71 to 8.26 kHz. Measured input-referred noise and total power consumption are 20.7μVrms and 1.90 respectively. The ADC achieves 8 bit accuracy at 16Ksps with an input voltage of ±400mV. Combined simulation and measurement results demonstrate the neural recording interface's suitability for in situ neutral activity recording.
本文介绍了一种用于射频识别(RFID)的低功耗、低噪声植入式神经记录接口。该系统集成了一个两级神经放大器和一个8位流水线模数转换器(ADC)。放大器级数的优化展示了最小的功耗和面积消耗;ADC采用了一种对器件泄漏具有鲁棒性的新型失调消除技术,以降低输入失调电压。神经放大器和ADC均采用700mV电源。神经放大器的中频增益为58.4dB,3dB带宽为0.71至8.26kHz。测得的输入参考噪声和总功耗分别为20.7μVrms和1.90。ADC在16Ksps采样率下,输入电压为±400mV时可实现8位精度。综合仿真和测量结果表明,该神经记录接口适用于原位神经活动记录。