量子计算实用化面临的关键瓶颈是如何实现量子计算系统的规模化及小型化，能够在 4–77 K 乃至 1 K 以下可靠运行的超低温芯片（Cryo-IC）快速发展，已成为支撑量子计算规模化、小型化的关键基础。工作在稀释制冷机内部的超导量子比特对温度极其敏感，温度变化会直接影响量子比特的相干时间、频率漂移、线宽与偏置稳定性。基于低温CMOS温度传感器实现量子计算系统冷端节点的温度实时监测对于量子比特操控系统的参数校准与超低温操控芯片热噪声预算至关重要。本文系统回顾了跨越室温至超低温的CMOS温度传感器技术谱系。首先概述室温范围内温度传感器的主流实现形式，包括基于BJT/ MOS组合的 PTAT/CTAT 混合方案、基于多种 TCR 电阻的阻性方案，以及环振读出架构，并比较其不确定度、能耗、面积与标定成本。随后梳理不同工艺中BJT、MOS和SiGe器件的超低温特性。进而对近年面向4 K等超低温应用的超低温温度传感器进行综述与横向对比。

The practical deployment of quantum computing is bottlenecked by the need to scale and miniaturize quantum systems. Rapid advances in cryogenic integrated circuits (Cryo-ICs) that operate reliably at 4–77 K and even below 1 K have become a key foundation for this scaling and miniaturization. Superconducting qubits inside dilution refrigerators are extremely temperature-sensitive: fluctuations directly impact coherence time, frequency drift, linewidth, and bias stability. Consequently, real-time temperature monitoring at cold-end nodes using cryogenic CMOS temperature sensors is essential for calibrating qubit-control parameters and budgeting the thermal noise of ultra-low-temperature control chips. This paper presents a systematic review of CMOS temperature-sensor technologies spanning room temperature to cryogenic operation. We first overview mainstream room-temperature implementations—including hybrid BJT/MOS PTAT/CTAT schemes, resistor-based sensors leveraging various TCR resistors, and ring-oscillator readout architectures—and compare their uncertainty, energy consumption, area, and calibration cost. We then summarize the cryogenic characteristics of BJT, MOS, and SiGe devices across different processes. Finally, we survey and cross-compare recent cryogenic temperature sensors targeting 4 K and related ultra-low-temperature applications.