光电探测器
探测器
材料科学
光电子学
分光计
宽带
光学
光电流
噪音(视频)
计算机科学
解耦(概率)
校准
光谱形状分析
重建算法
电压
光谱带
电子工程
脉冲整形
光电导性
调制(音乐)
光子学
暗电流
光电二极管
作者
Sang‐Hyeon Lee,Junho Min,Yonas Tsegaye Megra,Min‐Hye Jeong,Hwi Je Woo,Jaewon Han,Junheon Ha,Gyunghyun Jang,Young Jae Song,Hyun‐Soo Ra,Hoon Hahn Yoon,Jong‐Soo Lee
摘要
ABSTRACT Miniaturized computational spectrometers that combine spectrally tunable photodetectors with reconstruction algorithms offer a promising route toward compact, portable spectral sensing while maintaining performance comparable to traditional benchtop spectrometers. In single‐detector computational spectrometers, a reconfigurable photodetector serves as spectral‐encoding hardware, and the quality of its spectral response matrix critically determines peak‐wavelength accuracy, noise tolerance, and reconstruction fidelity. However, conventional encoding schemes based on electrically tunable single photodetectors often rely on one‐dimensional voltage sweeps, which can suffer from highly correlated response states and a practical trade‐off among electrical tunability, broadband coverage, and reconstruction robustness. Here, we address this trade‐off using a spectrally tunable floating‐base bipolar‐junction photodetector as a representative multidimensional encoder. By counter‐tuning the interfacial barrier matching between balanced and unbalanced configurations, a single detector combines excitonic‐resonance‐enhanced response, monotonic short‐wavelength absorption, and interlayer transport modulation of photoexcited carriers, thereby generating a richer and less correlated spectral response matrix. Multidimensional encoding of these complementary spectral response features enables more effective decoupling of encoded spectral datasets, even under sub‐nanoampere photocurrent operation where noise and calibration errors can critically degrade reconstruction fidelity. Our strategy provides a versatile pathway toward high‐fidelity, compact, and robust computational spectrometers for edge‐level spectral sensing and intelligent optoelectronic systems.
科研通智能强力驱动
Strongly Powered by AbleSci AI