Abstract
This paper proposes a three-point nonlinear calibration scheme for an ultra-low-power, high-precision temperature sensor to address the issue where the temperature error of a 0.8 μW sensor exceeds ±1 °C in RFID (Radio-Frequency Identification) temperature measurement systems. The proposed calibration scheme introduces a temperature-dependent nonlinearity coefficient to the traditional linear calibration, effectively compensating for the sensor's nonlinear output characteristics. To minimize calibration costs, a scheme embedding the calibration algorithm into the reader is proposed, along with a dichotomy-based approach for efficient temperature calibration. The experimental results demonstrate that, within the temperature range of -30 °C to 90 °C, the temperature error of five sensor samples can be reduced from ±8 °C to between -1 °C and 0.6 °C. This solution has been successfully implemented in mass production.