← 返回 JSSC 论文列表JSSC 2010第4期RF & Wireless0.13μmCMOS Image Sensor
A 05 V Sub-Microwatt CMOS Image Sensor With Pulse-Width Modulation Read-Out Scot
研究了一种0.5V超低功耗CMOS图像传感器,采用脉宽调制读出技术,适用于视网膜假体等应用。
0.5V, 140nJ/帧, 700nW, 23.4dB SNR
CMOS图像传感器超低功耗脉宽调制视网膜假体无线传感器
▸0.5V超低电压工作:采用激进电压缩放技术(aggressive voltage scaling),在0.5V甚至0.45V超低电压下实现CMOS图像传感器功能,突破传统3T有源像素传感器的电压限制,实测在0.5V时功耗仅700nW(0.5fps),能效比提升显著。
▸脉宽调制读出技术:提出基于脉冲宽度调制(PWM)的新型读出电路架构,替代传统模拟电压域信号处理,在0.5V低压环境下有效抑制噪声干扰,实测信噪比达23.4dB,解决了低压操作下的信号完整性难题。
▸超低功耗系统设计:通过协同优化像素阵列(128×128)与数字接口,实现140nJ/帧的超低能耗(8.5fps@0.5V),相比传统图像传感器能耗降低1-2个数量级,满足毫米级无线传感器的严苛功耗需求。
▸噪声量化分析方法:创新性地建立低压操作下的噪声量化模型,通过测试芯片验证0.5V工作时的噪声特性,为超低压图像传感器设计提供关键理论支撑。
Abstract
Sylvester
Abstract—Energy minimization is a critical goal in size-con-
strained wireless sensors. Sensing elements are traditionally power
hungry and require special attention in low energy systems. In
this work, we study ultra-low power image sensors. In particular,
we explore the use of aggressive voltage scaling in CMOS image
sensors for applications ranging from retinal prostheses to battle-
field monitoring and surveillance. We begin with a discussion of
the challenges faced by a traditiona