The Design of a Low-Noise, High-Speed Readout-Integrated Circuit for Infrared Focal Plane Arrays

Abstract

This paper describes the design of a low-noise, high-speed readout-integrated circuit for use in InGaAs infrared focal plane arrays, and analyzes the working principle and noise index of the pixel circuit in detail. The design fully considers the dynamic range, noise, and power consumption of the pixel circuit in which a capacitance transimpedance amplifier structure is adopted as the input stage circuit, and chip fabrication via an XFAB 0.18 µm CMOS process is successfully realized. The ROIC adopts monolithic integration and implements various functions, such as windowing, subsampling, and different integration and readout modes. The ROIC reached an array scale of 32 × 32, a frame rate of 100 Hz, and a readout rate of 20 Mbps with an analog power consumption of less than 52 mW. The measurement results show that the input reference noise can be reduced to 143 e- via the CDS, and the fully customized scheme has certain advantages in the research of high-performance ROICs.

Fast Prediction Method for Scattering Parameters of Rigid-Flex PCBs Based on ANN

“…The BP neural network architecture employed in this study comprised an input layer, an output layer, and two hidden layers. The dimension of the input layer was [1,6], corresponding to the optimization variables for the grid ground plane. The dimension of the output layer was [1,201], matching the number of frequency points.…”

“…The dimension of the input layer was [1,6], corresponding to the optimization variables for the grid ground plane. The dimension of the output layer was [1,201], matching the number of frequency points. The S-parameters in the frequency domain were represented by a complex-valued function depending on frequency.…”

“…In recent decades, short-wave infrared InGaAs focal plane array detectors [1] have been extensively used in aerospace remote sensing. With the continuous improvement in remote sensing payload indicators, the total number of remote sensing data has increased rapidly [2].…”