Yu Cheng
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Yu Cheng has a total of 15 co-authors. Here are the top co-authors by number of shared publications:
- Yuchun Chang
Dalian University of Technology
3 shared publications Guoqiang Zhong
School of Integrated Circuits, Dalian University of Technology, Dalian, China
2 shared publications- Shanshan Lou
School of Integrated Circuits, Dalian University of Technology, Dalian, China
2 shared publications - Botao Xiong
School of Integrated Circuits, Dalian University of Technology, Dalian, China
1 shared publication - Hongwei Liang
Dalian University of Technology
1 shared publication - Hongyu Hou
School of Integrated Circuits, Dalian University of Technology, Dalian 116000, China
1 shared publication
Latest Publications
- Open Access
A PFM-Based Calibration Method for Low-Power High-Linearity Digital Pixel
The nonlinearity problem of digital pixels restricts the reduction in power consumption at the pixel-level circuit. The main cause of nonlinearity is discussed in this article and low power consumption is attained by reducing the static current in capacitive transimpedance amplifiers (CTIAs) and comparators. Linearity was successfully improved through the use of an off-chip calibration method. A 64 × 64 array prototype digital readout integrated circuit (DROIC) was fabricated using a 0.18 μm 1P6M CMOS process. Experimental results indicated that the post-calibration linearity reached 99.6% with an input current of up to 1.5 μA. The static power consumption per digital pixel was 6 μW.
Open AccessA PFM-Based Calibration Method for Low-Power High-Linearity Digital Pixel
The nonlinearity problem of digital pixels restricts the reduction in power consumption at the pixel-level circuit. The main cause of nonlinearity is discussed in this article and low power consumption is attained by reducing the static current in capacitive transimpedance amplifiers (CTIAs) and comparators. Linearity was successfully improved through the use of an off-chip calibration method. A 64 × 64 array prototype digital readout integrated circuit (DROIC) was fabricated using a 0.18 μm 1P6M CMOS process. Experimental results indicated that the post-calibration linearity reached 99.6% with an input current of up to 1.5 μA. The static power consumption per digital pixel was 6 μW.
- 3 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Electron Devices
- Vol. 71 (10) , 6178-6184
Noise Improvement and High Dynamic Range in a CMOS Image Sensor With Shift Biasing Structure
In this article, a noise improvement and dynamic range (DR) enhanced CMOS image sensor (CIS), using a shift biasing structure, is proposed. The shift biasing pixel is constructed from a conventional 4T pixel and three additional transistors, including a bias transistor (BT) and two switch transistors. The BT works as a switch to connect the source and drain to accumulate electrons in the source follower (SF) in shift biasing mode and provides an extra capacitance to enlarge the capacity of floating diffusion (FD) in low gain (LG) mode. Electron distribution and noise performance of SF are simulated to prove the function of shift biasing pixels. A prototype sensor, with a 512 $\times$ 128-pixel array, is fabricated with a 180 nm CIS process. Measurement shows a temporal noise of 2 e $^{{-}}$ is achieved, which is a 2 e $^{{-}}$ noise improvement compared with the reference 4T readout mode of shift biasing pixel. Additionally, an over 92 dB DR is realized by combining shift biasing mode and LG mode without any specific process or complex control circuits.
- 3 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Electron Devices
- Vol. 71 (10) , 6178-6184
Noise Improvement and High Dynamic Range in a CMOS Image Sensor With Shift Biasing Structure
In this article, a noise improvement and dynamic range (DR) enhanced CMOS image sensor (CIS), using a shift biasing structure, is proposed. The shift biasing pixel is constructed from a conventional 4T pixel and three additional transistors, including a bias transistor (BT) and two switch transistors. The BT works as a switch to connect the source and drain to accumulate electrons in the source follower (SF) in shift biasing mode and provides an extra capacitance to enlarge the capacity of floating diffusion (FD) in low gain (LG) mode. Electron distribution and noise performance of SF are simulated to prove the function of shift biasing pixels. A prototype sensor, with a 512 $\times$ 128-pixel array, is fabricated with a 180 nm CIS process. Measurement shows a temporal noise of 2 e $^{{-}}$ is achieved, which is a 2 e $^{{-}}$ noise improvement compared with the reference 4T readout mode of shift biasing pixel. Additionally, an over 92 dB DR is realized by combining shift biasing mode and LG mode without any specific process or complex control circuits.
- 10 August 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Electron Devices
- Vol. 70 (9) , 4719-4724
An Over 140 dB Dynamic Range CMOS Image Sensor Combined DCG and Logarithmic Response
A dynamic range enhancement pixel architecture based on a standard CMOS image sensor (CIS) process with a pinned photodiode (PPD) is proposed, combining dual conversion gain (DCG) and logarithmic response. Due to a unique charge compensation mechanism of the proposed pixel, in a single exposure, three images are captured with the same integration time, related to different incident light levels. As a result, an over 140 dB dynamic range image is able to be obtained by merging these images. Photo response measurements show the image sensor exhibits over 92 dB linear dynamic range and more than 51 dB logarithmic dynamic range, without signal-to-noise ratio (SNR) performance degradation under low light conditions.
- 10 August 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Electron Devices
- Vol. 70 (9) , 4719-4724
An Over 140 dB Dynamic Range CMOS Image Sensor Combined DCG and Logarithmic Response
A dynamic range enhancement pixel architecture based on a standard CMOS image sensor (CIS) process with a pinned photodiode (PPD) is proposed, combining dual conversion gain (DCG) and logarithmic response. Due to a unique charge compensation mechanism of the proposed pixel, in a single exposure, three images are captured with the same integration time, related to different incident light levels. As a result, an over 140 dB dynamic range image is able to be obtained by merging these images. Photo response measurements show the image sensor exhibits over 92 dB linear dynamic range and more than 51 dB logarithmic dynamic range, without signal-to-noise ratio (SNR) performance degradation under low light conditions.