Non-invasive decoding of hand movements from electroencephalography based on a hierarchical linear regression model

A linear cross-coupled gate-driven quasi-floating bulk low-power wide input range transconductor

High linearity for a wider input voltage range and low-power operation of the operational transconductance amplifier (OTA) are indispensable parameters for health care applications, which require high quality and accurate signal conditioning. However, achieving low-power operation along with high linearity at low supply voltages is challenging for OTA using conventional low-power and linearization techniques. This paper proposes an OTA based on a cross-coupled gate-driven quasi-floating bulk (CGDQFB) MOSFET and source-degenerated linearization techniques, which works at a supply of ±0.5 V. The post-layout simulations of the proposed OTA are performed in the 180 nm standard CMOS process, which shows a transconductance of 0.321 μS, an output impedance of 331 MΩ, an input impedance of 897 GΩ, a DC gain of 40.54 dB, a gain-bandwidth of 0.145 MHz, a total harmonic distortion (THD) of 51.09 dB for an input voltage range of 626.7 mV at a frequency of 100.39 Hz, and a power consumption of 0.45 μW. The proposed OTA shows an input common mode range of -0.34 to 0.4 V, an output voltage swing of -0.34 to 0.34 V, a common mode rejection ratio of 91.08 dB, and a consumption area of 23 965.92 µm2. Furthermore, with 200 Monte Carlo iterations, the proposed OTA shows variability for gain and THD of 0.003 and 0.036, respectively. The proposed CGDQFB OTA is a suitable contender for conditioning bio-signals used in health care applications.

The decoder design and performance comparative analysis for closed-loop brain-machine interface system

Brain-machine interface (BMI) can convert electroencephalography signals (EEGs) into the control instructions of external devices, and the key of control performance is the accuracy and efficiency of decoder. However, the performance of different decoders obtaining control instructions from complex and variable EEG signals is very different and irregular in the different neural information transfer model. Aiming at this problem, the off-line and on-line performance of eight decoders based on the improved single-joint information transmission (SJIT) model is compared and analyzed in this paper, which can provide a theoretical guidance for decoder design. Firstly, in order to avoid the different types of neural activities in the decoding process on the decoder performance, eight decoders based on the improved SJIT model are designed. And then the off-line decoding performance of these decoders is tested and compared. Secondly, a closed-loop BMI system which combining by the designed decoder and the random forest encoder based on the improved SJIT model is constructed. Finally, based on the constructed closed-loop BMI system, the on-line decoding performance of decoders is compared and analyzed. The results show that the LSTM-based decoder has better on-line decoding performance than others in the improved SJIT model.

0.342 nW Class-AB enhanced flipped source follower low pass filter for biomedical applications

Owing to the impact of process voltage and temperature variations, the design of low-power low-pass filters (LPFs) with improved linearity is still one of the most challenging tasks for effective biological signal processing. This paper presents the design of a fourth-order Class-AB enhanced flipped source follower (EFSF) LPF circuit aimed at the detection of electroencephalography signals. The simulated results attained using complementary metal-oxide-semiconductor 180 nm technology node in Cadence Analog Design Environment demonstrate that the EFSF LPF emulates a DC-gain of -88 mdB with a bandwidth of 100 Hz and consumes 0.342 nW power from a supply voltage of 0.5 V. The calculated figure of merit for the proposed filter is 5.983 × 10-15 J with a dynamic range (DR) of 43.54 dB and input-referred noise of 91 µVrms. It consumes an area of 0.0458 mm2. To check the robustness of the proposed filter circuit, we performed Monte Carlo simulations with 200 runs. The statistical results achieved for the DC-gain, DR, and total harmonic distortion of the proposed filter show mean values of -188.09 mdB, 43.10 dB, and -41.85 dB along with standard deviation values of 285.21 mdB, 718.72 mdB, and 4.52 dB, respectively. The proposed Class-AB EFSF LPF can be used to achieve high power efficiency in future low-voltage and low-power biological systems.