school

UM E-Theses Collection (澳門大學電子學位論文庫)

Title

Biopotential readout front-end circuits using frequency-translation filtering techniques

English Abstract

Measuring bio-potential signals, such as Electroencephalogram (EEG), Electrocardiogram (ECG), and Electromyogram (EMG) against various contaminating signals in portable biomedical monitoring systems is of great challenge in minimizing the power consumption while maintaining a high performance. This thesis investigates the most critical component in the biopotential acquisition system: the readout front-end (RFE) for obtaining high quality biopotential signal. The first section is the introduction of the AC couple chopped instrumentation amplifier (ACCIA), which is improved from the conventional current feedback instrumentation amplifier by applying AC-coupling and chopper stabilization techniques. Thus, it is capable of filtering the differential DC electrode offset between the biopotential electrodes, and achieving low-noise and high-CMRR. The second section will describe two different function filter designs that are befitted to the ACCIA. The first one is the powerline notch filter for rejecting the powerline interference. Capacitor area for realizing the notch’s time constant is reduced by 1000 times compare to direct circuit method by utilizing the frequency-translation property of the chopper stabilization, which has been the technique of flicker-noise reduction, but is now extended to reject also the powerline interference fully within the ACCIA. Secondly, a chopper-stabilized response-translating lowpass filter (RT-LPF) is proposed, which can achieve an ultra-low cutoff down to 1.4-Hz. Similarly, it spreads the frequency-translation property to translate the bandpass filter characteristic to perform lowpass filtering. The implementation is further optimized by using a power-and-area-efficient Nauta cell as the operational transconductance amplifier (OTA). Simulation results show that the RT-LPF achieves 1.4-to-15-Hz cutoff tuning, –40-dB/decade stopband attenuation with low power consumption. The final section presents the entire design of the RFE. The front stage ACCIA is improved by applying the 50/60-Hz powerline notch filter and the thin- and thick-oxide transistors selection technique. Following the ACCIA there is a spike filter, which is clocked accurately to suppress the chopping spikes. In the last stage, there is a gain-BW-controllable amplifier, which further boosts the gain and enables different biopotential signal measurements through simple logic controls. Simulation results showed that the RFE can successfully eliminate the differential electrode offset up to ±50 mV, while achieving 140-dB CMRR and 51.4-nV/√Hz input-referred noise spectral density. The entire RFE consumes 35.5-µA in 3-V when the powerline notch filter is activated, which can rejects the powerline interference for more than 40-dB.

Issue date

2010.

Author

Ma, Chon Teng

Faculty

Faculty of Science and Technology

Department

Department of Electrical and Electronics Engineering

Degree

M.Sc.

Subject

Biomedical engineering

Medical electronics

Signal detection

Portable computerized instruments

Supervisor

Mak, Pui-in

Vai, Mang I

Files In This Item

TOC & Abstract

Full-text

Location
1/F Zone C
Library URL
991005550199706306