基于端到端优化的可编程微波光子滤波器

林蔚; 陈慧彬; 郭红英

doi:10.37188/CO.2026-0008

基于端到端优化的可编程微波光子滤波器

doi: 10.37188/CO.2026-0008

cstr: 32171.14.CO.2026-0008

林蔚^1,,
陈慧彬^{2, 3, ,},
郭红英¹

1.
漳州职业技术学院电子信息学院, 福建漳州 363000
2.
泉州师范学院光电工程系, 福建泉州 362000
3.
福建省先进微纳光子技术与器件重点实验室, 福建泉州 362000

基金项目: 国家自然科学基金（No. 61705119）

详细信息

作者简介:
林　蔚（1973—），女，福建漳州人，2010年于四川大学获得四川大学电子与通信工程领域工程硕士学位，现任漳州职业技术学院正高，主要从事微波光子学、电子信息、嵌入式物联网应用的研究。E-mail：200551000057@jmu.edu.cn

陈慧彬（1985—），女，福建宁德人，2016年于中国科学院大学获得凝聚态物理专业理学博士学位，现任泉州师范学院副教授，硕士生导师。主要从事固体激光器、半导体激光器、光神经网络的研究。E-mail：chenhuibin@qztc.edu.cn

中图分类号: TP394.1;TH691.9
计量
- 文章访问数: 6
- HTML全文浏览量: 5
- PDF下载量: 0
- 被引次数: 0
出版历程
- 网络出版日期: 2026-04-22

Programmable microwave photonic filter based on end-to-end optimization

LIN Wei^1
,,
CHEN Hui-bin^{2, 3
, ,},
GUO Hong-ying¹

1.
School of Electronic Information, Zhangzhou Institute of Technology, Zhangzhou, 363000, China
2.
College of Physics and Information Engineering, Quanzhou Normal University, Quanzhou 362000, China
3.
Fujian Provincial Key Laboratory for Advanced Micro-nano Photonics Technology and Devices, Quanzhou 362000, China

Funds: Supported by the National Natural Science Foundation of China (No. 61705119)

More Information

Corresponding author: chenhuibin@qztc.edu.cn

摘要

摘要:
基于加权时延结构的微波光子滤波器同时利用了光学器件和射频器件的优势，具有可重构、低成本、大带宽的特性，在微波频段提供了灵活高效的信号处理能力。然而，由于加权时延结构的复杂性，系统中分立的光电器件均会对不同波长上的权重抽头产生干扰，如：光频率梳的包络和增益竞争、EDFA的增益不均匀性和非线性、调制器受限的滤波带宽。这些因素均会导致抽头的权重偏离理想值，从而导致微波滤波器的畸变。本文提出了一种端到端优化的方式，通过将微波光子滤波器作为一个黑盒系统，通过实时监控系统最后输出的光谱形状（即每个tap的权重值），计算并反馈波长的抽头分布与理想抽头之间的差值，用于实时调整波长整型器的滤波系数，使得输出的光谱权重始终保持在设计的状态。通过这种端到端优化的方法，我们实现了0.05 dB的光谱重构精度，并且完成了带外抑制比高达47 dB的射频低通滤波器。
- 光子射频滤波器 /
- 微波光子学 /
- 锁模激光器
Abstract:
The microwave photonic filter based on weighted delay structure simultaneously leverages the advantages of photonic and radio-frequency components, featuring reconfigurability, low cost, and wide bandwidth, providing flexible and efficient signal processing capabilities in the microwave band. However, due to the complexity of the weighted delay structure, discrete optoelectronic components in the system can interfere with the weighted taps at different wavelengths—such as the envelope and gain competition of optical frequency combs, the gain non-uniformity and nonlinearity of EDFA, and the limited filtering bandwidth of modulators. These factors cause deviations in the weighted taps from their designed values, leading to distortion in the microwave filter. This paper proposes an end-to-end optimization approach by treating the microwave photonic filter as a black-box system. By monitoring the spectral shape (i.e., the weight values of each tap) of the final output in real time, the difference between the wavelength taps and the ideal taps is calculated and feedbacked to adjust the filtering coefficients of the waveshaper in real time, ensuring the output spectral weights remain in the designed state. Through this end-to-end optimization approach, we achieved a spectral reconstruction accuracy of 0.05dB and completed an RF low-pass filter with an out-of-band rejection ratio of up to 47dB.
- photonic RF filters /
- microwave photonics /
- mode-locked laser.

HTML全文

图 1 (a) 半导体锁模激光器；(b) 锁模激光器的频率梳

Figure 1. (a) Customized semiconductor mode-locked laser; (b) The frequency comb of the mode-locked laser

下载: 全尺寸图片幻灯片

图 2 光子射频滤波器的实验装置。EDFA：掺铒光纤放大器。EOM：电光调制器。SMF：单模光纤。WSS：波长选择开关PD：光电探测器。VNA：矢量网络分析仪。OSA：光谱分析仪。

Figure 2. Experimental setup of the photonic RF filter. EDFA: erbium-doped fiber amplifier. EOM: Electro-optic modulator. SMF: single-mode fiber. WSS: wavelength selective switch PD: photodetector. VNA: vector network analyzer. OSA: optical spectrum analyzer.

下载: 全尺寸图片幻灯片

图 3 (a) 经过一次迭代优化的光谱；(2)经过四次迭代优化的光谱

Figure 3. (a) Spectrum optimized through one iteration; (2) Spectrum optimized through four iterations

下载: 全尺寸图片幻灯片

图 4 对应于不同抽头数的形状梳的光谱。（a）10个抽头；（b）16个抽头；（c）36个抽头；（d）72个抽头；

Figure 4. Optical spectra of the shaped comb corresponding to different tap numbers. (a) 10 taps; (b) 16 taps; (c) 36 taps; (d) 72 taps;

下载: 全尺寸图片幻灯片

图 5 具有不同抽头数的sinc型光子射频滤波器

Figure 5. Sinc-style photonic RF filters with different tap numbers.

下载: 全尺寸图片幻灯片

图 6 高斯切趾sinc滤波器对应的光谱. (a) σ=5, (a) σ=15, (a) σ=25, (a) σ=35.

Figure 6. Optical spectra of the Gaussian-apodized sinc filter. (a) σ=5, (a) σ=15, (a) σ=25, (a) σ=35.

下载: 全尺寸图片幻灯片

图 7 具有不同系数的高斯切趾型光子射频滤波器

Figure 7. Gaussian-apodized photonic RF filters with different coefficient.

下载: 全尺寸图片幻灯片

参考文献(25)

[1]	CAPMANY J, ORTEGA B, PASTOR D. A tutorial on microwave photonic filters[J]. Journal of Lightwave Technology, 2006, 24(1): 201-229. doi: 10.1109/JLT.2005.860478
[2]	YAO J P. Microwave photonics[J]. Journal of Lightwave Technology, 2009, 27(3): 314-335.
[3]	CAPMANY J, NOVAK D. Microwave photonics combines two worlds[J]. Nature Photonics, 2007, 1(6): 319-330.
[4]	SEEDS A J, WILLIAMS K J. Microwave photonics[J]. Journal of Lightwave Technology, 2006, 24(12): 4628-4641.
[5]	MARPAUNG D, ROELOFFZEN C, HEIDEMAN R, et al. Integrated microwave photonics[J]. Laser & Photonics Reviews, 2013, 7(4): 506-538.
[6]	MA C, CHEN H, YE X W, et al. Ultra-high resolution microwave photonic radar with post-bandwidth synthesis[J]. Chinese Optics Letters, 2020, 18(7): 072501. doi: 10.3788/COL202018.072501
[7]	XU X Y, WU J Y, NGUYEN T G, et al. Photonic microwave true time delays for phased array antennas using a 49 GHz FSR integrated optical micro-comb source [Invited][J]. Photonics Research, 2018, 6(5): B30-B36.
[8]	SHU H W, CHANG L, TAO Y SH, et al. Microcomb-driven silicon photonic systems[J]. Nature, 2022, 605(7910): 457-463.
[9]	KIM J, PARK M J, PERROTT M H, et al. Photonic subsampling analog-to-digital conversion of microwave signals at 40-GHz with higher than 7-ENOB resolution[J]. Optics Express, 2008, 16(21): 16509-16515.
[10]	YAO J P. Photonic generation of microwave arbitrary waveforms[J]. Optics Communications, 2011, 284: 3723-3736.
[11]	. ZHANG W P, LEDERMAN J C, DE LIMA T F, et al. A system-on-chip microwave photonic processor solves dynamic RF interference in real time with picosecond latency[J]. Light: Science & Applications, 2024, 13: 14.
[12]	. XU SH F, ZOU X T, MA B W, et al.. Deep-learning-powered photonic analog-to-digital conversion[J]. Light: Science & Applications, 2019, 8(1): 66.
[13]	PELUSI M, LUAN F, VO T D, et al. Photonic-chip-based radio-frequency spectrum analyser with terahertz bandwidth[J]. Nature Photonics, 2009, 3: 139-143.
[14]	ZOU X H, ZOU F, CAO Z ZH, et al. A multifunctional photonic integrated circuit for diverse microwave signal generation, transmission, and processing[J]. Laser & Photonics Reviews, 2019, 13(6): 1800240.
[15]	HERVÁS J, RICCHIUTI A L, LI W, et al. Microwave photonics for optical sensors[J]. IEEE Journal of Selected Topics in Quantum Electronics, 2017, 23(2): 5602013.
[16]	XU X Y, WU J Y, NGUYEN T G, et al. Advanced RF and microwave functions based on an integrated optical frequency comb source[J]. Optics Express, 2018, 26(3): 2569-2583.
[17]	ZHU ZH J, CHI H, JIN T, et al. All-positive-coefficient microwave photonic filter with rectangular response[J]. Optics Letters, 2017, 42(15): 3012-3015.
[18]	POLO V, VIDAL B, CORRAL J L, et al. Novel tunable photonic microwave filter based on laser arrays and N/spl times/N AWG-based delay lines[J]. IEEE Photonics Technology Letters, 2003, 15(4): 584-586.
[19]	LIU D, NGO N Q, TJIN S C. Microwave photonic bandpass filter using a multiwavelength semiconductor-optical-amplifier ring laser[J]. Optical Engineering, 2007, 46(5): 054401.
[20]	XU X Y, TAN M X, WU J Y, et al. Advanced adaptive photonic RF filters with 80 taps based on an integrated optical micro-comb source[J]. Journal of Lightwave Technology, 2019, 37(4): 1288-1295.
[21]	XU X Y, TAN M X, WU J Y, et al. Microcomb-based photonic RF signal processing[J]. IEEE Photonics Technology Letters, 2019, 31(23): 1854-1857.
[22]	WU J Y, XU X Y, NGUYEN T G, et al. RF photonics: an optical Microcombs’ perspective[J]. IEEE Journal of Selected Topics in Quantum Electronics, 2018, 24(4): 6101020.
[23]	LI M, LI W ZH, YAO J P. Tunable optoelectronic oscillator incorporating a high-Q spectrum-sliced photonic microwave transversal filter[J]. IEEE Photonics Technology Letters, 2012, 24(14): 1251-1253.
[24]	WU J CH, XIE Y W, HONG SH H, et al. Programmable multichannel-parallel microwave photonic processor on silicon[J]. Laser & Photonics Reviews, 2026, 20(1): e00378.
[25]	XU X Y, TAN M X, CORCORAN B, et al. 11 TOPS photonic convolutional accelerator for optical neural networks[J]. Nature, 2021, 589(7840): 44-51.