原始舰船通信信号包含高维时频特征和冗余噪声，直接处理会导致计算复杂度指数级增长，且噪声可能淹没关键特征，降低分类准确性。首先，通过改进的最大相关最小冗余算法结合信噪比动态调节因子，实现高维时频特征的有效降维，剔除冗余噪声并保留关键干扰特征；其次，构建信号观测矩阵，提取时域矩偏度、峰度及包络起伏度特征，为分类提供基础数据；最后，设计双分支TextCNN结构，结合全局平均池化（GAP）与Softmax分类器，实现干扰信号的高效分类。实验结果表明，所提方法降维后特征子集规模为100～200，有效去除冗余。F-measure均值达0.94，FLOPS较低，能够为复杂海洋环境下的干扰信号分类提供高效、可靠的解决方案。
The original ship communication signals contain high-dimensional time-frequency features and redundant noise. Direct processing can lead to exponential growth in computational complexity, and noise may overwhelm key features, reducing classification accuracy. Firstly, by combining the improved maximum correlation minimum redundancy algorithm with the dynamic adjustment factor of signal-to-noise ratio, effective dimensionality reduction of high-dimensional time-frequency features is achieved, eliminating redundant noise and preserving key interference features; Secondly, construct a signal observation matrix to extract time-domain moment skewness, kurtosis, and envelope fluctuation features, providing basic data for classification; Finally, a dual branch TextCNN structure is designed, combined with Global Average Pooling (GAP) and Softmax classifier, to achieve efficient classification of interference signals. The experimental results show that the proposed method reduces the dimensionality of the feature subset to a size between 100-200, effectively removing redundancy. The F-measure average is 0.94, with low FLOPS, providing an efficient and reliable solution for classifying interference signals in complex marine environments.
2026,48(1): 188-193 收稿日期：2025-8-25
DOI：10.3404/j.issn.1672-7649.2026.01.027
分类号：U665；TP391
作者简介：李金梅(1986-),女,博士,高级工程师,研究方向为网络信息安全
参考文献：
[1] 邢小东, 侯飞. 舰船多信道无线网络抗干扰算法设计[J]. 舰船科学技术, 2023, 45(16): 141-144.
XING X D, HOU F. Design of anti-jamming algorithm for ship multi-channel wireless network[J]. Ship Science and Technology, 2023, 45(16): 141-144.
[2] 刘冰, 郝新红, 蔡鑫. 基于功率谱熵的无线电引信目标与干扰信号分类方法[J]. 北京航空航天大学学报, 2024, 50(3): 913-919.
LIU B, HAO X H, CAI X. Classification method of radio fuze target and interference signal based on power spectrum entropy[J]. Journal of Beijing University of Aeronautics and Astronautics, 2024, 50(3): 913-919.
[3] 刘春雨, 袁俊, 孙晓萌, 等. 基于去噪扩散概率模型的离线真实无线干扰信号分类[J]. 电波科学学报, 2025, 40(3): 546-551.
LIU C Y, YUAN J, SUN X M, et al. Offline real wireless interference signal classification based on denoising diffusion probability model[J]. Chinese Journal of Radio Science, 2025, 40(3): 546-551.
[4] CORONADO V M L, MANUEL M R, HERNÁNDEZ E J, et al. Bending classification from interference signals of a fiber optic sensor using shallow learning and convolutional neural networks[J]. Pattern Recognition Letters, 2024, 186: 354-360.
[5] 刘冰, 郝新红, 秦高林, 等. 引信目标与干扰信号稀疏分类识别方法[J]. 北京航空航天大学学报, 2025, 51(2): 498-506.
LIU B, HAO X H, QIN G L, et al. Sparse classification and recognition method of fuzed targets and jamming signals[J]. Journal of Beijing University of Aeronautics and Astronautics, 2025, 51(2): 498-506.
[6] 李程, 陈明虎, 施育鑫, 等. 基于DCGAN-CNN的小样本通信干扰信号识别[J]. 无线电通信技术, 2025, 51(1): 70-79.
LI C, CHEN M H, SHI Y X, et al. Small sample communication jamming signal recognition based on DCGAN-CNN[J]. Radio Communications Technology, 2025, 51(1): 70-79.
[7] 张晓玲, 李玮. 基于小波变换的舰船主干网络干扰信号防御算法[J]. 舰船科学技术, 2021(18): 163-165.
ZHANG X L, LI W. The algorithm of jamming signal defense in the main network of ships based on wavelet transform[J]. Ship Science and Technology, 2021(18): 163-165.
[8] 吴琼, 王强. 应用傅里叶变换的舰船动力控制平台干扰信号防御算法[J]. 舰船科学技术, 2021(4): 88-90.
WU Q, WANG Q. Algorithm of interference signal defense for ship power control platform based on fourier transform[J]. Ship Science and Technology, 2021(4): 88-90.
[9] 于隆, 尹娜. 复杂电磁环境干扰下舰船网络信息安全传输方法[J]. 舰船科学技术, 2025, 47(10): 161-165.
YU L, YIN N. Safe transmission method of ship network information under complex electromagnetic environment interference[J]. Ship Science and Technology, 2025, 47(10): 161-165.
[10] 周雪芳, 高长全, 刘阳. 面向大数据网络的舰船辐射干扰信号检测方法[J]. 舰船科学技术, 2022, 44(3): 147-150.
ZHOU X F, GAO C Q, LIU Y. Detection method of ship radiated interference signal oriented to big data network[J]. Ship Science and Technology, 2022, 44(3): 147-150.
[11] 吴铮. 舰船网络化电气传动干扰信号实时监测方法分析[J]. 舰船科学技术, 2020(20): 106-108.
WU Z. Analysis of real-time monitoring method for interference signal of ship’s networked electric drive[J]. Ship Science and Technology, 2020(20): 106-108.
[12] 赵小霞. 舰船防污染电气控制平台干扰信号控制算法研究[J]. 舰船科学技术, 2020(14): 217-219
ZHAO X X. Research on interference signal control algorithm of ship anti-pollution electrical control platform[J]. Ship Science and Technology, 2020(14): 217-219
