• JSTS:Journal of Semiconductor Technology and Science
• /
• v.13 no.2
• /
• pp.108-113
• /
• 2013

This paper describes a low power asynchronous successive approximation register (SAR) type 12b analog-to-digital converter (ADC) for biomedical applications in a 0.35 ${\mu}m$ CMOS technology. The digital-to-analog converter (DAC) uses a capacitive split-arrays consisting of 6-b main array, an attenuation capacitor C and a 5-b sub array for low power consumption and small die area. Moreover, splitting the MSB capacitor into sub-capacitors and an asynchronous SAR reduce power consumption. The measurement results show that the proposed ADC achieved the SNDR of 68.32 dB, the SFDR of 79 dB, and the ENOB (effective number of bits) of 11.05 bits. The measured INL and DNL were 1.9LSB and 1.5LSB, respectively. The power consumption including all the digital circuits is 6.7 ${\mu}W$ at the sampling frequency of 100 KHz under 3.3 V supply voltage and the FoM (figure of merit) is 49 fJ/conversion-step.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.11
• /
• pp.1627-1634
• /
• 2021

This paper introduces a low-power compact ADC circuit for analog Convolutional filter for low-power neural network accelerator SOC. While convolutional neural network accelerators can speed up the learning and inference process, they have drawback of consuming excessive power and occupying large chip area due to large number of multiply-and-accumulate operators when implemented in complex digital circuits. To overcome these drawbacks, we implemented an analog convolutional filter that consists of an analog multiply-and-accumulate arithmetic circuit along with an ADC. This paper is focused on the design optimization of a low-power 8bit SAR ADC for the analog convolutional filter accelerator We demonstrate how to minimize the capacitor-array DAC, an important component of SAR ADC, which is three times smaller than the conventional circuit. The proposed ADC has been fabricated in CMOS 65nm process. It achieves an overall size of 1355.7㎛², power consumption of 2.6㎼ at a frequency of 100MHz, SNDR of 44.19 dB, and ENOB of 7.04bit.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.59 no.2
• /
• pp.355-358
• /
• 2010

As CMOS technology continues to scale down, signal processing is favorably done in the digital domain, which requires Analog-to-Digital (A/D) Converter to be integrated on-chip. This paper presents a design methodology of 12-bit 1-MS/s Rail-to-Rail fully differential SAR ADC using Deep N-well Switch based on binary search algorithm. Proposed A/D Converter has the following architecture and techniques. Firstly, chip size and power consumption is reduced due to split capacitor array architecture and charge recycling method. Secondly, fully differential architecture is used to reduce noise between the digital part and converters. Finally, to reduce the mismatch effect and noise error, the circuit is designed to be available for Rail-to-Rail input range using simple Deep N-well switch. The A/D Converter fabricated in a TSMC 0.18um 1P6M CMOS technology and has a Signal-to-Noise-and-Distortion-Ratio(SNDR) of 69 dB and Free-Dynamic-Range (SFDR) of 73 dB. The occupied active area is $0.6mm^2$.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.54 no.2
• /
• pp.47-52
• /
• 2017

In this paper, CMOS SAR A/D converter 1.8V supply for the design of an A/D converter having an middle speed for the biological signal processing was designed. This paper proposes design of a 10-bit SAR Analog to Digital Converter improving linearity driven by MSB node of C-DAC array divided into 4 equal parts. It enhances linearity property, by retaining the analog input signal charging time at MSB node. Because MSB node samples analog input, it enhances resolution through getting initial input signal precisely. By using split capacitor on C-DAC array, it reduced chip size and power dissipation. The Proposed SAR A/D Converter is fabricated in 0.18um CMOS and measured 7.5 bits of ENOB at sampling frequency 4MS/s and power supply of 1.8V. It occupies a core area of $850{\times}650um^2$ and consumes 123.105uW. Therefore it results in 170.016fJ/step of FOM(Figure of Merit).