An application of the stochastic A/D conversion to multi-bit delta-sigma modulators is considered, and a novel correction technique for D/A converter (DAC) error is proposed. The stochastic A/D conversion can reduce the area of the quantizer and allows large mismatches. The proposed calibration technique corrects DAC errors using a programmable quantizer. The programmable quantizer has a non-linear characteristic that cancels DAC errors. Using this technique, we can decrease the influence of DAC errors without using conventional dynamic element matching. This A/D converter has a non-linear quantization characteristic, so output digital code must be corrected using a programmable encoder. This code correction and setting of the quantization levels are carried out based on calibration data obtained using genetic algorithm.

In the present paper, we propose a novel high-resolution analog-to-digital converter (ADC) for low-power biomedical analog frontends, which we call the successive stochastic approximation ADC. The proposed ADC uses a stochastic flash ADC (SF-ADC) to realize a digitally controlled variable-threshold comparator in a successive-approximationregister ADC (SAR-ADC), which can correct errors originating from the internal digital-to-analog converter in the SAR-ADC. For the residual error after SAR-ADC operation, which can be smaller than thermal noise, the SF-ADC uses the statistical characteristics of noise to achieve high resolution. The SF-ADC output for the residual signal is combined with the SAR-ADC output to obtain high-precision output data using the supervised machine learning method.

In this paper, we discuss a high-precision and low-power analog-to-digital converter (ADC) which is required for wearable biomedical measurement sensors driven by a battery. In particular, we focus on the successive approximation register ADC (SAR-ADC), and propose its calibration algorithm using the machine learning. We derive a calibration function for the outputs of the SAR-ADC by taking into account its characteristics, and show the least squares method of determining the parameter values of the function to minimize the residual errors. Furthermore, from the practical viewpoint, we propose an incremental learning for the calibration, where additional data sets are selected on the basis of the Bayesian predictive distributions which are obtained at each additional learning step. Through numerical experiments, we observed that the mean residual errors obtained by the proposed method are less than 1 LSB, and the method needs a small amount of training data.

This paper presents a high-precision biomedical sensor system with a novel analog-frontend (AFE) IC and error correction by machine learning. The AFE IC embeds an analog-to-digital converter (ADC) architecture called successive stochastic approximation ADC. The proposed ADC integrates a stochastic flash ADC (SF-ADC) into a successive approximation register ADC (SAR-ADC) to enhance its resolution. The SF-ADC is also used as a digitally controlled variable threshold comparator to provide error correction of the SAR-ADC. The proposed system also calibrates the ADC error using the machine learning algorithm on an external PC without additional power dissipation at a sensor node. Due to the flexibility of the system, the design complexity of an AFE IC can be relaxed by using these techniques. The target resolution is 18 bits, and the target bandwidth (without digital low-pass filter) is about 5 kHz to deal with several types of biopotential signals. The design is fabricated in a 130-nm CMOS process and operates at 1.2-V supply. The fabricated ADC achieves the SNDR of 88 dB at a sampling frequency of 250 kHz by using the proposed calibration techniques. Due to the high-resolution ADC, the input-referred noise is 2.52 µV rms with a gain of 28.5 dB.

No abstract