基于双路径特征融合网络的背景信号抑制算法

doi:10.12305/j.issn.1001-506X.2025.07.33

系统工程与电子技术 ›› 2025, Vol. 47 ›› Issue (7): 2406-2413.doi: 10.12305/j.issn.1001-506X.2025.07.33

• 通信与网络 • 上一篇

基于双路径特征融合网络的背景信号抑制算法

张琬滢¹^,², 高由兵¹^,², 李泽一¹^,², 李鹏飞¹^,², 张伟²^,³^,*

1. 中国电子科技集团公司第二十九研究所, 四川成都 610036
2. 电磁空间安全全国重点实验室, 四川成都 610036
3. 电子科技大学信息与通信工程学院, 四川成都 611731

收稿日期:2024-06-26 出版日期:2025-07-16 发布日期:2025-07-22
通讯作者: 张伟
作者简介:张琬滢 (2000—), 女, 硕士研究生, 主要研究方向为电子对抗、信号与信息处理
高由兵 (1982—), 男, 高级工程师, 硕士, 主要研究方向为电子对抗
李泽一 (1994—), 男, 工程师, 博士, 主要研究方向为通信对抗
李鹏飞 (1993—), 男, 高级工程师, 博士, 主要研究方向为电子对抗
张伟 (1985—), 男, 研究员级高级工程师, 博士, 主要研究方向为电子对抗
基金资助:
国家自然科学基金(U20B2070);国家自然科学基金(U23B2013)

Background signal suppression algorithm based on dual-path feature fusion net

Wanying ZHANG¹^,², Youbing GAO¹^,², Zeyi LI¹^,², Pengfei LI¹^,², Wei ZHANG²^,³^,*

1. The 29th Research Institute of China Electronics Technology Group Corporation, Chengdu 610036, China
2. National Key Laboratory of Electromagnetic Space Security, Chengdu 610036, China
3. School of Information and Communication Engineering, University of Electronic Science and Technology of China, Chengdu 611731, China

Received:2024-06-26 Online:2025-07-16 Published:2025-07-22
Contact: Wei ZHANG

摘要/Abstract

摘要：

有效抑制接收机中混叠的背景信号, 对关键重要信号的检测至关重要。现有方法缺乏对信号上下文信息的有效利用, 难以解决噪声环境下单通道时频交叠的背景信号抑制问题。对此，从中频时域信号出发, 提出一种基于双路径特征融合的掩码抑制(dual-path feature fusion mask-based suppression, DPFF-MS)算法。该算法利用神经网络拟合掩码抑制模型, 实现背景信号抑制。通过一组卷积编解码器实现信号的高维变换和逆变换, 降低噪声影响; 设计一种双路径特征融合(dual-path feature fusion, DPFF)模块, 利用不同路径的长短时记忆(long short-term memory, LSTM)网络交替提取局部特征和全局上下文信息, 并采用迭代注意特征融合(iterative attention feature fusion, iAFF)优化不同尺度特征融合过程, 以充分利用信号脉内和脉间信息, 解决时频交叠情况下的抑制问题。实验结果表明, 相较其他信号处理方法和神经网络模型, 所提模型对尺度不变信号源噪比(scale-invariant source-to-noise ratio, SI-SNR)和背景脉冲抑制率两个指标都有显著提升; 且极大地减少了模型参数, 使其易于部署, 具备较高应用价值。

关键词: 电子侦察, 背景信号抑制, 深度学习, 多尺度特征融合

Abstract:

It is vital for the detection of critical and important signals to suppress background signals at the end of receiver effectively. Current methods inadequately utilize signal context information, rendering it challenging to address the issue of background signal suppression of single channel in noisy environments and time-frequency overlaps. To regard this, a dual-path feature fusion mask-based suppression (DPFF-MS) algorithm is proposed, utilizing mid-frequency time-domain signals. The algorithm utilizes a neural network to fit the mask suppression model, effectively suppressing background signals. The high-dimensional transformation and inverse transformation of signals can be enabled by the employment of a suite of convolutional encoder-decoder networks, diminishing the adverse effects of noise. A dual-path feature fusion (DPFF) module is developed, which leverages long short-term memory (LSTM) networks of varying paths to alternately extract both local features and global contextual information. An iterative attention feature fusion (iAFF) is used to optimize the process of fusing features of different scales, fully exploiting the intra-pulse and inter-pulse information to address the suppression issue in time-frequency overlapping environments. The experimental results indicate that, in comparison to other signal processing approaches and neural network models, the proposed model shows significant enhancements in terms of scale-invariant source-to-noise ratio (SI-SNR) and background pulse suppression rate. Furthermore, it significantly reduces the number of model parameters, making it straightforward to deploy and possess high application value.

Key words: electronic reconnaissance, suppression of background signal, deep learning, multi-scale feature fusion

中图分类号:

TN971

张琬滢, 高由兵, 李泽一, 李鹏飞, 张伟. 基于双路径特征融合网络的背景信号抑制算法[J]. 系统工程与电子技术, 2025, 47(7): 2406-2413.

Wanying ZHANG, Youbing GAO, Zeyi LI, Pengfei LI, Wei ZHANG. Background signal suppression algorithm based on dual-path feature fusion net[J]. Systems Engineering and Electronics, 2025, 47(7): 2406-2413.

图/表 8

图1

图2

图3

表1

表2

图4

图5

图6

参考文献 31

1	RIM J W , KOH I S . Survivability simulation of airborne platform with expendable active decoy countering RF missile[J]. IEEE Trans.on Aerospace and Electronic Systems, 2020, 56 (1): 197- 207.
2	孙国敏, 张伟, 邵怀宗, 等. 基于低秩张量完备的电磁大数据标注补全算法[J]. 系统工程与电子技术, 2024, 46 (2): 381- 390. doi: 10.12305/j.issn.1001-506X.2024.02.02
	SUN G M , ZHANG W , SHAO H Z , et al. A low-rank tensor completion based method for electromagnetic big data annotation recovery[J]. Systems Engineering and Electronics, 2024, 46 (2): 381- 390. doi: 10.12305/j.issn.1001-506X.2024.02.02
3	WEI G H , LI M , ZHAO H Z . The general evaluation model for radio-frequency blocking effects of frequency equipment in complex electromagnetic environment[J]. IEEE Trans.on Antennas and Propagation, 2023, 71 (9): 7418- 7425.
4	CHEN T , YANG B Y , GUO L M . Radar pulse stream clustering based on MaskRCNN instance segmentation network[J]. IEEE Signal Processing Letters, 2023, 30 (1): 1022- 1026.
5	HASANI H , KHORAVI M R . Pulse deinterleaving based on fusing PDWs and PRI extraction process for radar-assisted edge devices considering computational costs[J]. EURASIP Journal on Wireless Communications and Networking, 2021, 2021, 98.
6	SU Y H , CHEN Z , GONG L F , et al. An improved adaptive radar signal sorting algorithm based on DBSCAN by a novel CVI[J]. IEEE Access, 2024, 12, 43139- 43154.
7	XIANG H R , SHEN F R , ZHAO J . Deep ToA mask-based recursive radar pulse deinterleaving[J]. IEEE Trans.on Aerospace and Electronic Systems, 2023, 59 (2): 989- 1006.
8	CHEN T, GUO W, SHEN B. A new radar emitter recognition method based on pulse sample figure[C]//Proc. of the 8th International Conference on Fuzzy Systems and Knowledge Discovery, 2011: 1902-1905.
9	ZHU R Q, WANG C Y, JIN S P, et al. Radar signal pre-sorting based on Mean- Shift algorithm[C]//Proc. of the 7th International Conference on Intelligent Computing and Signal Processing, 2022: 1766-1769.
10	LI X Q , LIU Z M , HUANG Z T . Deinterleaving of pulse streams with denoising autoencoders[J]. IEEE Trans.on Aerospace and Electronic Systems, 2020, 56 (6): 4767- 4778.
11	GUO R H , DONG Y Y , ZHANG L D , et al. Radar reconnaissance pulse-splitting modeling and detection method[J]. Remote Sensing, 2024, 16 (3): 521.
12	高霞, 全英汇, 李亚超, 等. 基于BSS的FDA-MIMO雷达主瓣欺骗式干扰抑制方法[J]. 系统工程与电子技术, 2020, 42 (9): 1927- 1934. doi: 10.3969/j.issn.1001-506X.2020.09.07
	GAO X , QUAN Y H , LI Y C , et al. Main-lobe deceptive jamming suppression with FDA-MIMO radar based on BSS[J]. Systems Engineering and Electronics, 2020, 42 (9): 1927- 1934. doi: 10.3969/j.issn.1001-506X.2020.09.07
13	CHEN S , LIN Y , YUAN Y , et al. Suppressive interference suppression for airborne SAR using BSS for singular value and eigenvalue decomposition based on information entropy[J]. IEEE Trans.on Geoscience and Remote Sensing, 2023, 61, 5205611.
14	ZHANG H L , ZHANG G , XUE B , et al. Mainlobe jamming suppression via improved BSS method for rotated array radar[J]. Journal of Systems Engineering and Electronics, 2022, 33 (6): 1151- 1158.
15	DAVIES M E , JAMES C J . Source separation using single channel ICA[J]. Signal Processing, 2007, 87 (8): 1819- 1832.
16	KOPSINIS Y , MCLAUGHLIN S . Development of EMD-based denoising methods inspired by wavelet thresholding[J]. IEEE Trans.on Signal Processing, 2009, 57 (4): 1351- 1362.
17	MOSTAFA M H , CHAMAANI S , SACHS J . Singular spectrum analysis-based algorithm for vitality monitoring using M-sequence UWB sensor[J]. IEEE Sensors Journal, 2020, 20 (9): 4787- 4802.
18	张伟, 王沙飞, 林静然, 等. 基于孪生网络的电磁目标跨模式识别算法[J]. 电子学报, 2022, 50 (6): 1281- 1290.
	ZHANG W , WANG S F , LIN J R , et al. Cross-modal recognition algorithm of electromagnetic targets via siamese network[J]. Acta Electronica Sinica, 2022, 50 (6): 1281- 1290.
19	曾令华, 高由兵, 梁超, 等. 基于深度学习的同型雷达的脉冲中频信号分离技术[J]. 电子信息对抗技术, 2023, 38 (6): 53- 60.
	ZENG L H , GAO Y B , LIANG C , et al. Intermediate frequency pulse signal separation technology of homogeneous radar based on deep learning[J]. Electronic Information Warfare Technology, 2023, 38 (6): 53- 60.
20	HE Y H, SANG X, LI Y G. A radarsignal sorting algorithm based on improved FastICA[C]//Proc. of the IEEE 3rd International Conference on Information Technology, Big Data and Artificial Intelligence, 2023: 1625-1628.
21	LIU S , WANG B M , ZHANG L Y . A method for blind source separation of multichannel electromagnetic radiation in the field[J]. IEEE Access, 2020, 8, 191341- 191354.
22	WU Z W , YUAN D , ZHANG F G , et al. Improving accuracy of low-cost vehicle attitude estimation by de-noising MIMU based on SSA and ICA[J]. IEEE Trans.on Instrumentation and Measurement, 2021, 70, 9507213.
23	DUTT R, MONDAL S, ACHARYYA A. Single channel blind source separation using dual extended kalman filter[C]//Proc. of the IEEE International Symposium on Circuits and Systems, 2021.
24	LUO Y, MESGARANI N. TasNet: time-domain audio separation network for real-time, single-channel speech separation[C]//Proc. of the IEEE International Conference on Acoustics, Speech and Signal Processing Single-Channel Speech Separation, 2018: 696-700.
25	LUO Y , MESGARANI N . Conv-TasNet: surpassing ideal time-frequency magnitude masking for speech separation[J]. IEEE/ACM Trans.on Audio, Speech, and Language Processing, 2019, 27 (8): 1256- 1266.
26	LUO Y, CHEN Z, YOSHIOKA T. Dual-path RNN: efficient long sequence modeling for time-domain single-channel speech separation[C]//Proc. of the ICASSP 2020-2020 IEEE International Conference on Acoustics, Speech and Signal Processing, 2020: 46-50.
27	DAI Y M, GIESEKE F, OEHMCKE S, et al. Attentional feature fusion[C]//Proc. of the IEEE Winter Conference on Applications of Computer Vision, 2021: 3559-3568.
28	LI A D , YU G C , ZHENG C S , et al. A general unfolding speech enhancement method motivated by Taylor's theorem[J]. IEEE/ACM Trans.on Audio, Speech, and Language Processing, 2023, 31, 3629- 3646.
29	ZHANG Z W , ZHU M T , LI Y J , et al. JDMR-Net: joint detection and modulation recognition networks for LPI radar signals[J]. IEEE Trans.on Aerospace and Electronic Systems, 2023, 59 (6): 7575- 7589.
30	CUI G L , LI H B , MURALIDHAR R , et al. MIMO radar waveform design with constant modulus and similarity constraints[J]. IEEE Trans.on Signal Processing, 2014, 62 (2): 343- 353.
31	FENG H C , TANG B , WAN T . Radar pulse repetition interval modulation recognition with combined net and domain-adaptive few-shot learning[J]. Digital Signal Processing, 2022, 127, 103562.

辐射源类型	脉宽/μs	载频/MHz	脉冲重复周期/μs	波形	脉宽调制	载频调制	脉冲重复周期调制
关重雷达	1~2	5~45	10~60	线性调频	脉组内固定	每5 MHz一个频点脉组间捷变	脉组内固定
背景雷达	0.1~2	5~45	500~2 000	单载频	固定	固定频点	抖动

网络	参数量/MB
Conv-TasNet	16.738
DPRNN- TasNet	14.06
DPFF-MS	0.873

[1]	张新征, 闫梦可, 朱晓林. 噪声伪标签容忍的半监督SAR目标识别[J]. 系统工程与电子技术, 2025, 47(6): 1796-1805.
[2]	姜智杰, 宋恒, 胡楠, 段兰茜, 曹平. 隧道环境毫米波雷达目标识别与分类算法[J]. 系统工程与电子技术, 2025, 47(5): 1453-1460.
[3]	聂千祁, 沙明辉, 朱应申, 王崇宇, 崔念强. 基于盲源分离结合奇异谱分析的雷达多分量信号识别方法[J]. 系统工程与电子技术, 2025, 47(4): 1168-1175.
[4]	袁磊, 杨艳娟, 郭毅, 戴鹏. 相关噪声下基于深度学习的LDPC码码率半盲识别算法[J]. 系统工程与电子技术, 2025, 47(4): 1335-1345.
[5]	沈夏闰, 李若楠, 张昊田. 基于CVAE-LSTM的服务器KPI异常检测[J]. 系统工程与电子技术, 2025, 47(3): 1019-1027.
[6]	刘晓琳, 郭梦娇, 李卓. Dueling DQN优化下的航班延误自适应图卷积循环网络预测方法[J]. 系统工程与电子技术, 2025, 47(2): 568-579.
[7]	付卫红, 张鑫钰, 刘乃安. 基于多尺度融合神经网络的同频同调制单通道盲源分离算法[J]. 系统工程与电子技术, 2025, 47(2): 641-649.
[8]	蔡伟, 王鑫, 蒋昕昊, 杨志勇, 陈栋. 基于解耦的小样本目标检测方法研究[J]. 系统工程与电子技术, 2024, 46(9): 2941-2950.
[9]	陈晓萱, 徐书文, 胡绍海, 马晓乐. 基于卷积与自注意力的红外与可见光图像融合[J]. 系统工程与电子技术, 2024, 46(8): 2641-2649.
[10]	陈柯宇, 杨健, 张伟, 孙国敏, 邵怀宗. 基于极大似然估计的非理想场景雷达信号分选算法[J]. 系统工程与电子技术, 2024, 46(7): 2276-2284.
[11]	汪强龙, 高晓光, 吴必聪, 胡子剑, 万开方. 受限玻尔兹曼机及其变体研究综述[J]. 系统工程与电子技术, 2024, 46(7): 2323-2345.
[12]	孙先涛, 江汪洋, 陈文杰, 陈伟海, 智亚丽. 基于感兴趣区域的物体抓取位姿检测[J]. 系统工程与电子技术, 2024, 46(6): 1867-1877.
[13]	陈雪梅, 刘志恒, 周绥平, 余航, 刘彦明. 基于HRNet的高分辨率遥感影像道路提取方法[J]. 系统工程与电子技术, 2024, 46(4): 1167-1173.
[14]	张天文, 张晓玲, 邵子康, 曾天娇. 基于掩模注意型交互的SAR舰船实例分割[J]. 系统工程与电子技术, 2024, 46(3): 831-838.
[15]	龚峻扬, 付卫红, 刘乃安. SAR图像目标轮廓增强预处理模块设计[J]. 系统工程与电子技术, 2024, 46(12): 4010-4017.

基于双路径特征融合网络的背景信号抑制算法

Background signal suppression algorithm based on dual-path feature fusion net

RichHTML

可视化

摘要/Abstract

引用本文

使用本文

图/表 8

参考文献 31

相关文章 15

编辑推荐

Metrics

本文评价