A fully differential chopper-stabilized LNA for EEG acquisition applications
Samaneh Sedighi Maragheh, Marzieh Moradi, Massoud DoustiPurpose
This study proposes a fully differential, low-power, low-noise amplifier (LNA) with a 2nd-order Gm-C low-pass filter (LPF) for portable electroencephalogram (EEG) systems. The purpose of this study is to achieve an optimal noise-power tradeoff simultaneously minimizing input-referred noise (IRN) and power consumption. The study examines the impact of the proposed circuit design on the key parameters of EEG acquisition systems, particularly with regard to noise reduction and the overall system performance of an EEG system.
Design/methodology/approach
The research methodology combines both theoretical analysis and postlayout simulations using 0.18-µm CMOS technology to design and evaluate the proposed LNA architecture. The LNA incorporates a fully differential chopper-stabilized recycling double-folded cascode (FDCRDFC) amplifier and a 2nd-order Gm-C LPF. Its performance is benchmarked against conventional fully differential folded cascode amplifiers to highlight gains in noise reduction and power consumption.
Findings
The proposed LNA achieves a significantly reduced IRN of 0.62 µVrms over 0.1–100 Hz while consuming only 0.327 µW. The architecture has a mid-band gain of 64 dB and a bandwidth of 0.1–168 Hz, making it well-suited for portable EEG applications. The effective size of the proposed amplifier, including pads, is 560 × 560 µm2.
Research limitations/implications
This study focuses on the amplifier’s design and its impact on noise reduction and power consumption. Further studies could explore the integration of multiple channels in high-density EEG arrays or active ripple suppression techniques using a chopper or tunable filtering for multimodal bio-signal acquisition.
Practical implications
The significantly lower noise floor achieved by the proposed LNA enhances the signal quality in EEG acquisition systems, enabling more reliable real-time monitoring and improved accuracy in brain-health diagnostics.
Originality/value
The principal originality of this study compared to traditional approaches, lies in the co-optimized integration of the three key techniques: a recycling double-folded cascode (RDFC) amplifier for enhanced gain-efficiency, an NMOS-based chopper modulator for 1/f noise cancelation and a weak-inversion Gm-C filter for ripple suppression. This synergistic approach achieves an exceptional noise-power tradeoff, which is the critical benchmark for next-generation portable EEG acquisition.