CMOS Ultralow Power Brain Signal Acquisition Front-Ends: Design and Human Testing

Alireza Karimi-Bidhendi, Omid Malekzadeh-Arasteh, Mao Cheng Lee, Colin M. McCrimmon, Po T. Wang, Akshay Mahajan, Charles Yu Liu, Zoran Nenadic, An H. Do, Payam Heydari

Research output: Contribution to journalArticlepeer-review

29 Scopus citations

Abstract

Two brain signal acquisition (BSA) front-ends incorporating two CMOS ultralow power, low-noise amplifier arrays and serializers operating in mosfet weak inversion region are presented. To boost the amplifier's gain for a given current budget, cross-coupled-pair active load topology is used in the first stages of these two amplifiers. These two BSA front-ends are fabricated in 130 and 180 nm CMOS processes, occupying 5.45 mm $^{2}$ and 0.352 mm $^{2}$ of die areas, respectively (excluding pad rings). The CMOS 130-nm amplifier array is comprised of 64 elements, where each amplifier element consumes 0.216 μ W from 0.4 V supply, has input-referred noise voltage (IRNoise) of 2.19 μV RMS corresponding to a power efficiency factor (PEF) of 11.7, and occupies 0.044 mm $^{2}$ of die area. The CMOS 180 nm amplifier array employs 4 elements, where each element consumes 0.69 μ W from 0.6 V supply with IRNoise of 2.3 μ V RMS (corresponding to a PEF of 31.3) and 0.051 mm 2 of die area. Noninvasive electroencephalographic and invasive electrocorticographic signals were recorded real time directly on able-bodied human subjects, showing feasibility of using these analog front-ends for future fully implantable BSA and brain- computer interface systems.

Original languageEnglish (US)
Article number7999268
Pages (from-to)1111-1122
Number of pages12
JournalIEEE Transactions on Biomedical Circuits and Systems
Volume11
Issue number5
DOIs
StatePublished - Oct 2017

Keywords

  • Analog front-end (AFE)
  • CMOS
  • electrocorticography (ECoG)
  • electroencephalogram (EEG)
  • instrumentation amplifier (InAmp)
  • noise efficiency factor (NEF)
  • operational transconductance amplifier (OTA)
  • power efficiency factor (PEF)
  • ultra-low power (ULP)
  • weak inversion (WI) region

ASJC Scopus subject areas

  • Biomedical Engineering
  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'CMOS Ultralow Power Brain Signal Acquisition Front-Ends: Design and Human Testing'. Together they form a unique fingerprint.

Cite this