A Micropower Chopper CBIA with DSL-Embedded Input Stage with 0.4 V EO Tolerance for Dry-Electrode Biopotential Recording
Issued Date
2023-01-01
Resource Type
ISSN
19324545
eISSN
19409990
Scopus ID
2-s2.0-85153343011
Journal Title
IEEE Transactions on Biomedical Circuits and Systems
Rights Holder(s)
SCOPUS
Bibliographic Citation
IEEE Transactions on Biomedical Circuits and Systems (2023)
Suggested Citation
Thanapitak S., Surakampontorn W., Sawigun C. A Micropower Chopper CBIA with DSL-Embedded Input Stage with 0.4 V EO Tolerance for Dry-Electrode Biopotential Recording. IEEE Transactions on Biomedical Circuits and Systems (2023). doi:10.1109/TBCAS.2023.3265273 Retrieved from: https://repository.li.mahidol.ac.th/handle/20.500.14594/81847
Title
A Micropower Chopper CBIA with DSL-Embedded Input Stage with 0.4 V EO Tolerance for Dry-Electrode Biopotential Recording
Author(s)
Author's Affiliation
Other Contributor(s)
Abstract
A chopper instrumentation amplifier (IA) dedicated for bio-potential acquisition usually requires a linearized input stage for large electrode offset voltage accommodation. This linearization leads to excessive power consumption when sufficiently low input-referred noise (IRN) is required. We present a current-balance IA (CBIA) without the need for the input stage linearization. It uses two transistors to operate as an input transconductance stage and a dc-servo loop (DSL) at the same time. An off-chip capacitor completes the DSL by ac coupling the source terminals of the input transistors via chopping switches realizing a sub-Hz high-pass cutoff frequency for dc rejection. Fabricated in a 0.35-μm CMOS process, the proposed CBIA occupies 0.41 mm2 and consumes 1.19 μW from a 3 V dc supply. Measurements show that the IA achieves an input-referred noise of 0.91 μV<sub>rms</sub> over 100 Hz bandwidth. This corresponds to a noise efficiency factor of 2.22. Typical CMRR of 102.1 dB is achieved for zero offset and degraded to 85.9 dB when a ±0.3 V input offset was applied. Gain variation of 0.5% is maintained within the range of ±0.4 V input offset. The resulting performance meets well with the requirement for ECG and EEG recording using dry electrodes. A demonstration for the use of the proposed IA on a human subject is also provided.