Compensation for Process and Temperature Dependency in a CMOS Image Sensor.

Area of Science:

• • Electrical Engineering
• • Materials Science
• • Sensor Technology

Background:

• • CMOS Image Sensors (CIS) are susceptible to performance variations caused by manufacturing processes and temperature fluctuations.
• • These variations affect critical imaging parameters such as dark current, dark signal non-uniformity (DSNU), and conversion gain (CG).
• • Accurate modeling and compensation are essential for reliable CIS performance in diverse operating conditions.

Purpose of the Study:

• • To analyze and compensate for process and temperature dependencies in CMOS image sensor arrays.
• • To develop and validate novel sensor methodologies for characterizing and mitigating these variations.
• • To enhance the stability and accuracy of CIS performance through on-chip compensation techniques.

Main Methods:

• • Proposed process sensors utilizing pixel source follower (SF) transconductance (g_m,SF) to model process variations.
• • Analyzed the thermal dependency of SF gain (A_SF).
• • Integrated bipolar junction transistor (BJT)-based temperature sensors within the CIS array for thermal information.

Main Results:

• • Process sensors effectively modeled SF gain (A_SF) and its thermal dependency.
• • On-chip BJT temperature sensors achieved an untrimmed inaccuracy within ±0.5 °C.
• • Compensated for dark signal and conversion gain thermal dependencies by at least 79% and 87%, respectively.

Conclusions:

• • The proposed methods successfully analyze and compensate for process and temperature variations in CIS.
• • On-chip temperature sensing integrated with process variation modeling significantly improves CIS stability.
• • This work provides a robust framework for enhancing the reliability and performance of CMOS image sensors.