1.Information Quantum Technology Laboratory, School of Information Science and Technology, Southwest Jiaotong University, Chengdu 610031, China 2.Quantum Optoelectronics Laboratory, School of Physical Science and Technology, Southwest Jiaotong University, Chengdu 610031, China 3.Photonics Laboratory, School of Science, Donghua University, Shanghai 201620, China
Fund Project:Project supported by the National Natural Science Foundation of China (Grant Nos. 11974290, 61871333)
Received Date:26 October 2020
Accepted Date:24 February 2021
Available Online:07 June 2021
Published Online:20 July 2021
Abstract:Noise is one of the main factors affecting the performance index of weak signal detection devices, and the optimal filtering algorithm is an effective method to adaptively extract various useful weak signals from the white noise background. In order to improve the performance of single photon detector (especially the photon number resolution ability), one mainly focuses on the optimization of detector hardware such as the optimization of photosensitive materials and the technology of device fabrication. However, in this paper the performance of microwave kinetic Inductance detector (MKID) in the way of data processing is improved. Considering the fact that the template of light pulse signal in the optimal filtering algorithm is obtained by taking the average, we replace the noise model in the original optimal filtering algorithm with the white noise model and the whitening noise model. Then we process the photon response data that are detected by the MKID in an extremely low temperature environment. The results show that the energy resolution (one of the main performance indexes of single photon detector) of MKID is improved by about 15%, and we achieve an infrared single photon energy resolution of 0.26 eV. In this paper, the application and development trends of superconducting single photon detector are briefed. Then, how the MKID responds to weak coherent optical signal in low temperature environment, and the process of signal conversion, acquisition and output are explained in detail. According to the optimal filtering algorithm, we use different noise models to analyze the results of the signals detected by MKID. After that, we count the optimal amplitude multiple, perform the Gaussian fitting analysis on the statistical graph, and compare the energy resolution with the photon number resolution of the optimal filtering algorithm under different noise models. As a result, we find that under the white noise model, the optimal filtering algorithm is used to obtain the best result for MKID processing, and high energy resolution can be achieved. Keywords:optimal filtering/ microwave kinetic inductance detector/ energy resolution/ noise model
全文HTML
--> --> --> -->
2.1.实验测量系统
图1为实验所采用的测量系统简单示意图. 其中弱光脉冲由1550 nm激光二极管输出, 经衰减器后由光纤导入对准10 mK环境中由1/4波长超导谐振器构成的探测器芯片, 照射到探测器芯片上的光子能量拆散超导库珀对成为准粒子, 导致谐振器频率可检测的变化. 测试信号由微波信号产生器输入后一路作为本地信号, 另一路作为探测器的探测信号; 在输出端的IQ混频器实现本地信号和探测信号的混频, 经低通滤波器滤波、AD模数转换后由数据采集卡进行收集. 图 1 实验测量系统示意图 Figure1. Schematic diagram of the experimental system for single-photon detection.
图3简单表示了一个数据的采集时间操作序列. 其中, 蓝色为IQ信号输出; 红色代表200 ns宽的光脉冲; 黄色为信号产生器的120 Hz触发信号并同步至模数转换AD采集卡, 其采样率为2.5 MHz. 每次实验在触发模式下采集20000个脉冲, 并对脉冲前后各5000个(2 ms), 共10000个(4 ms)数据点进行记录分析. 在数据处理过程中, 取前5000个数据点用作噪声分析, 取2501—7500这5000个数据点进行幅值估计. 图 3 触发模式记录时间序列信号的示意图 Figure3. Schematic diagram of the signal triggers and records in time domain.
其中, $ {a}_{n} $表示第$ n $个高斯峰的幅度, $ {u}_{n} $表示第$ n $个高斯峰的中心点, ${{\sigma }}_{n}^{2}$表示第$ n $个高斯峰分布的方差. 图5中, 蓝色手指峰部分为多次脉冲幅值倍数统计图, 红色曲线为高斯相加拟合的图像. 图 5 光学衰减17 dB (a)和20 dB (b)下脉冲信号幅值统计的高斯拟合结果 Figure5. Gaussian fitting results of pulse signal amplitudes under optical attenuation of 17 dB (a) and 20 dB (b), respectively, for the experimental noises.
这表明, 在白噪声情况下最优滤波算法本质上是一个利用最小二乘法进行最小方差拟合的过程. 这时, 最优幅值倍数$ \tilde {A} $可计算为未知脉冲和模板脉冲的互相关函数除以模板脉冲的自相关函数, 或者简单地说等于未知脉冲向量除以模板脉冲向量. 由此, 模板获取的方式大为简便, 从而极大减少算法的运算量. 图8给出了基于白噪声模型、光学衰减17 dB和20 dB下弱相关光脉冲响应的光子数峰及其高斯拟合结果. 对比于之前图5所示的数据处理结果, 光子数峰的分辨能力有所提升. 基于此拟合结果, 应用(4)式和(5)式可计算出相应的能量分辨, 如表2和表3所示. 图 8 采用白噪声模型后, 光学衰减17 dB (a)和20 dB (b)下的脉冲幅值统计分布以及拟合图像 Figure8. Gaussian fitting results of pulse signal amplitudes under optical attenuation of 17 dB (a) and 20 dB (b), respectively, for the ideal white noises.
能量分辨
$ \Delta E_{{0}}{/{\rm{eV}}} $
$ \Delta E_{{1}}{/{\rm{eV}}} $
$ \Delta E_{2}{/{\rm{eV}}} $
$ \Delta E_{3}{/{\rm{eV}}} $
$ \Delta E_{4}{/{\rm{eV}}} $
$ \Delta E_{5}{/{\rm{eV}}} $
实测噪声
0.1015
0.3526
0.4360
0.4691
0.6140
—
白噪声模型
0.1489
0.2992
0.3772
0.4382
0.4448
0.5113
提高
–31.83%
+17.85%
+15.59%
+7.05%
+38.04%
—
表2光学衰减17 dB下使用实测噪声和白噪声模型处理后探测器能量分辨对比 Table2.Comparison of detector energy resolutions after processing with measured noise and white noise model under optical attenuation of 17 dB.
表3光学衰减20 dB下使用实测噪声和白噪声模型处理后探测器能量分辨对比 Table3.Comparison of detector energy resolutions after processing with measured noise and white noise model under optical attenuation of 20 dB.
23.2.噪声白化模型 -->
3.2.噪声白化模型
现实情况下严格的白噪声是不存在的, 所以以上采用白噪声模型来实现最优滤波算法的数据处理过于理想化. 为此, 对实测噪声特性进行白化处理[24], 使之尽量接近最优滤波算法中约定的白噪声模型. 实测噪声的白化处理主要步骤是: 首先, 对噪声数组求自相关得到噪声的自相关函数$ {R}_{xx} $; 然后, 再对$ {R}_{xx} $进行toeplitz变换得到噪声的协方差矩阵$ {R} $; 其次, 将得到的矩阵进行Cholesky分解得到下三角矩阵$ {{{L}}} $; 最后, 将下三角矩阵求逆矩阵便可得到白化后噪声的自相关矩阵$ {{{W}}}={{L}}^{{-1}} $. 图9为实测噪声白化后的特性. 图 9 白化后的探测器噪声 (a)自相关函数; (b)功率谱密度$ J{'}\left(f\right) $ Figure9. Autocorrelation function (a) and Power spectral density (b) of the whitened noise.
其中S代表脉冲模板向量, d代表未知脉冲向量, W是白化后噪声自相关矩阵. 图10给出了白化噪声后的弱相干光脉冲的探测器光子数响应峰值拟合分布. 由此, 可计算出各光子数峰的能量分辨, 如表4和表5所示. 图 10 噪声白化后, 光学衰减17 dB (a)和20 dB (b)下探测器的光子数响应拟合 Figure10. Gaussian fitting results of pulse signal amplitudes under optical attenuation of 17 dB (a) and 20 dB (b), respectively, for the whitened noises.
能量分辨
$ \Delta {{E}}_{{0}}{/{\rm{eV}}} $
$ \Delta {{E}}_{{1}}{/{\rm{eV}}} $
$ \Delta {{E}}_{2}{/{\rm{eV}}} $
$ \Delta {{E}}_{3}{/{\rm{eV}}} $
$ \Delta {{E}}_{4}{/{\rm{eV}}} $
$ \Delta {{E}}_{5}{/{\rm{eV}}} $
实测噪声
0.1015
0.3526
0.4360
0.4691
0.6140
—
噪声白化
0.1469
0.3274
0.4263
0.4897
0.5009
0.6216
提高/%
–44.73
+7.15
+2.22
–4.39
+18.42
—
表4光学衰减17 dB下实测噪声和噪声白化后处理得到的探测器能量分辨对比 Table4.Comparison of energy resolutions for the experimental noises and the whitening ones, where the optical pulse is attenuated 17 dB.
能量分辨
$ \Delta {{E}}_{{0}}{/{\rm{eV}}} $
$ \Delta {{E}}_{{1}}{/{\rm{eV}}} $
$ \Delta {{E}}_{2}{/{\rm{eV}}} $
$ \Delta {{E}}_{3}{/{\rm{eV}}} $
实测噪声
0.0955
0.3200
0.4199
0.4758
噪声白化
0.1932
0.2855
0.39649
0.4196
提高/%
–100.02
+10.78
+5.60
+11.81
表5光学衰减20 dB下实测噪声和噪声白化后处理得到的探测器能量分辨对比 Table5.Comparison of energy resolutions for the experimental noises and the whitening ones, where the optical pulse is attenuated 20 dB.
表6光学衰减20 dB下原始滤波和改进噪声模型后探测器能量分辨对比(括号中是改进后相对于实测噪声处理所得到的能量分辨的提高百分比) Table6.Comparison of energy resolutions of the detector for the experimental noise, white noise and withened noise, respectively. The improvemence is relative to the those for the experimental noise. Here, the optical pulse is attenuated 20 dB.