Multi channel low-noise low-power amplifier for neural signal acquisition

Yohei Yasuda, Nobuhiko Nakano

Research output: Contribution to journalArticle

Abstract

This paper describes a low-noise and low-power spike neural signal amplifier design that has cutoff frequency compensation between the channels and chips variations. The variation of the frequency characteristics of amplifiers should be minimized among the channels and chips. That is a requirement to do the statistical correlation analysis from a neuroscience point of view. Our design includes the adjustable cutoff frequency using 4 bit variable capacitance. After the compensation the variation of the cutoff frequency was reduced to -0.4 kHz to +0.3 kHz from -1.1 kHz to +3.6 kHz that is the value of before trimming under the condition of the target cutoff frequency is 10 kHz. We designed a multi neural signal amplifier using ROHM 0.18 μm CMOS process. The designed neural amplifier has the capacitive coupled differential input to reject large dc offsets generated at the electrode-tissue interface and to avoid the large common mode noise. To achieve the high energy efficiency with low noise to observe the few tens of μV order spike signal, the MOS transistors in OTA are operated at subthreshold region and combined with low pass filter that consumes less than a hundred nW. The amplifier yielded a midband gain of 37.9 dB and the input-referred noise was measured to be 3.76 μVrms while consuming 4.30 μW with a ±0.9 V power supply. These results corresponding to Noise Efficiency Factor (NEF)=2.23 that are close to the value of the limit using a single differential OTA by CMOS process.

Original languageEnglish
Pages (from-to)2-11
Number of pages10
JournalIEEJ Transactions on Electronics, Information and Systems
Volume135
Issue number1
DOIs
Publication statusPublished - 2015

Fingerprint

Cutoff frequency
Power amplifiers
Trimming
Low pass filters
MOSFET devices
Energy efficiency
Capacitance
Tissue
Electrodes
Compensation and Redress

Keywords

  • Bio-signal
  • CMOS LSI
  • Cutoff frequency variation
  • Low-noise low-power design
  • Neural signal amplifier

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Cite this

Multi channel low-noise low-power amplifier for neural signal acquisition. / Yasuda, Yohei; Nakano, Nobuhiko.

In: IEEJ Transactions on Electronics, Information and Systems, Vol. 135, No. 1, 2015, p. 2-11.

Research output: Contribution to journalArticle

@article{2bcb2888ece8464f88342c829b9cd40f,
title = "Multi channel low-noise low-power amplifier for neural signal acquisition",
abstract = "This paper describes a low-noise and low-power spike neural signal amplifier design that has cutoff frequency compensation between the channels and chips variations. The variation of the frequency characteristics of amplifiers should be minimized among the channels and chips. That is a requirement to do the statistical correlation analysis from a neuroscience point of view. Our design includes the adjustable cutoff frequency using 4 bit variable capacitance. After the compensation the variation of the cutoff frequency was reduced to -0.4 kHz to +0.3 kHz from -1.1 kHz to +3.6 kHz that is the value of before trimming under the condition of the target cutoff frequency is 10 kHz. We designed a multi neural signal amplifier using ROHM 0.18 μm CMOS process. The designed neural amplifier has the capacitive coupled differential input to reject large dc offsets generated at the electrode-tissue interface and to avoid the large common mode noise. To achieve the high energy efficiency with low noise to observe the few tens of μV order spike signal, the MOS transistors in OTA are operated at subthreshold region and combined with low pass filter that consumes less than a hundred nW. The amplifier yielded a midband gain of 37.9 dB and the input-referred noise was measured to be 3.76 μVrms while consuming 4.30 μW with a ±0.9 V power supply. These results corresponding to Noise Efficiency Factor (NEF)=2.23 that are close to the value of the limit using a single differential OTA by CMOS process.",
keywords = "Bio-signal, CMOS LSI, Cutoff frequency variation, Low-noise low-power design, Neural signal amplifier",
author = "Yohei Yasuda and Nobuhiko Nakano",
year = "2015",
doi = "10.1541/ieejeiss.135.2",
language = "English",
volume = "135",
pages = "2--11",
journal = "IEEJ Transactions on Electronics, Information and Systems",
issn = "0385-4221",
publisher = "The Institute of Electrical Engineers of Japan",
number = "1",

}

TY - JOUR

T1 - Multi channel low-noise low-power amplifier for neural signal acquisition

AU - Yasuda, Yohei

AU - Nakano, Nobuhiko

PY - 2015

Y1 - 2015

N2 - This paper describes a low-noise and low-power spike neural signal amplifier design that has cutoff frequency compensation between the channels and chips variations. The variation of the frequency characteristics of amplifiers should be minimized among the channels and chips. That is a requirement to do the statistical correlation analysis from a neuroscience point of view. Our design includes the adjustable cutoff frequency using 4 bit variable capacitance. After the compensation the variation of the cutoff frequency was reduced to -0.4 kHz to +0.3 kHz from -1.1 kHz to +3.6 kHz that is the value of before trimming under the condition of the target cutoff frequency is 10 kHz. We designed a multi neural signal amplifier using ROHM 0.18 μm CMOS process. The designed neural amplifier has the capacitive coupled differential input to reject large dc offsets generated at the electrode-tissue interface and to avoid the large common mode noise. To achieve the high energy efficiency with low noise to observe the few tens of μV order spike signal, the MOS transistors in OTA are operated at subthreshold region and combined with low pass filter that consumes less than a hundred nW. The amplifier yielded a midband gain of 37.9 dB and the input-referred noise was measured to be 3.76 μVrms while consuming 4.30 μW with a ±0.9 V power supply. These results corresponding to Noise Efficiency Factor (NEF)=2.23 that are close to the value of the limit using a single differential OTA by CMOS process.

AB - This paper describes a low-noise and low-power spike neural signal amplifier design that has cutoff frequency compensation between the channels and chips variations. The variation of the frequency characteristics of amplifiers should be minimized among the channels and chips. That is a requirement to do the statistical correlation analysis from a neuroscience point of view. Our design includes the adjustable cutoff frequency using 4 bit variable capacitance. After the compensation the variation of the cutoff frequency was reduced to -0.4 kHz to +0.3 kHz from -1.1 kHz to +3.6 kHz that is the value of before trimming under the condition of the target cutoff frequency is 10 kHz. We designed a multi neural signal amplifier using ROHM 0.18 μm CMOS process. The designed neural amplifier has the capacitive coupled differential input to reject large dc offsets generated at the electrode-tissue interface and to avoid the large common mode noise. To achieve the high energy efficiency with low noise to observe the few tens of μV order spike signal, the MOS transistors in OTA are operated at subthreshold region and combined with low pass filter that consumes less than a hundred nW. The amplifier yielded a midband gain of 37.9 dB and the input-referred noise was measured to be 3.76 μVrms while consuming 4.30 μW with a ±0.9 V power supply. These results corresponding to Noise Efficiency Factor (NEF)=2.23 that are close to the value of the limit using a single differential OTA by CMOS process.

KW - Bio-signal

KW - CMOS LSI

KW - Cutoff frequency variation

KW - Low-noise low-power design

KW - Neural signal amplifier

UR - http://www.scopus.com/inward/record.url?scp=84920414712&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84920414712&partnerID=8YFLogxK

U2 - 10.1541/ieejeiss.135.2

DO - 10.1541/ieejeiss.135.2

M3 - Article

AN - SCOPUS:84920414712

VL - 135

SP - 2

EP - 11

JO - IEEJ Transactions on Electronics, Information and Systems

JF - IEEJ Transactions on Electronics, Information and Systems

SN - 0385-4221

IS - 1

ER -