基于改进鲸鱼算法和BiGRU的弹道目标HRRP识别方法

doi:10.12305/j.issn.1001-506X.2025.06.11

摘要/Abstract

摘要：

针对弹道中段目标高分辨距离像(high resolution range profile, HRRP)的时序特征提取和识别问题, 提出一种基于优化双向门控循环单元(bidirectional gate recurrent unit, BiGRU)的弹道目标识别方法。首先, 将HRRP数据处理为双向序列, 建立BiGRU网络并提取目标HRRP的双向时序特征。然后，使用双权重策略鲸鱼优化算法(double weight strategy whale optimization algorithm, DWSWOA)优化BiGRU模型的参数, 引入的双权重因子在鲸鱼优化算法(whale optimization algorithm, WOA)快速收敛的同时又可有效避免陷入局部最优解。基于优化BiGRU模型的目标HRRP识别实验结果表明, 所提算法相较于其他4种算法, 目标识别准确度更高, 并在噪声数据集上表现出更好的鲁棒性和可靠性。

关键词: 目标识别, 高分辨距离像, 鲸鱼优化算法, 双向门控循环单元, 置信度

Abstract:

To address the problem of time sequence feature extraction and recognition of high resolution range image (HRRP) for ballistic mid-course targets, a ballistic target recognition method based on optimized bidirectional gate recurrent unit (BiGRU) is proposed. Firstly, the HRRP data is processed as a bidirectional sequence, and a BiGRU network is established to extract the temporal features of both directions of HRRP. Secondly, the double weight strategy whale optimization algorithm (DWSWOA). Double weight factors enable higher whale optimization algorithm (WOA) converage velocity and lower probability of falling into local optimal solution, and are used to optimize the parameters of BiGRU. The experimental results of target HRRP recognition based on optimized BiGRU model show that the proposed method exhibits higher accuracy in target recognition, robustness and reliability on noisy data sets than the other four algorithms.

Key words: target recognition, high resolution range profile (HRRP), whale optimization algorithm (WOA), bidirectional gate recurrent unit (BiGRU), confidence

中图分类号:

TP183

王彩云, 贾一帆, 李晓飞, 王佳宁, 吴钇达. 基于改进鲸鱼算法和BiGRU的弹道目标HRRP识别方法[J]. 系统工程与电子技术, 2025, 47(6): 1824-1832.

Caiyun WANG, Yifan JIA, Xiaofei LI, Jianing WANG, Yida WU. Ballistic target HRRP recognition method based on improved whale algorithm and BiGRU[J]. Systems Engineering and Electronics, 2025, 47(6): 1824-1832.

图/表 15

图1

图2

图3

图4

表1

图5

图6

表2

表3

表4

表5

图7

表6

图8

图9

参考文献 31

1	XIANG Q , WANG X D , LAI J , et al. Quadruplet depth-wise separable fusion convolution neural network for ballistic target recognition with limited samples[J]. Expert Systems With App-lications, 2024, 235, 121182. doi: 10.1016/j.eswa.2023.121182
2	YANG L , ZHANG W P , JIANG W D . Recognition of ballistic targets by fusing micro-motion features with networks[J]. Remote Sensing, 2022, 14 (22): 5678. doi: 10.3390/rs14225678
3	ZHANG C, DING J J, XU Y, et al. Research on the strategy of ballistic missile midcourse recognition based on Bayesian network[C]// Proc. of the IEEE 5th Advanced Information Technology, Electronic and Automation Control Conference, 2021: 2721-2726.
4	WANG Y B, BO L, WANG F. RCS statistical feature extraction for space target recognition based on Bi-LSTM[C]//Proc. of the IEEE International Geoscience and Remote Sensing Symposium, 2023: 6049-6052.
5	PERSICO A R , CLEMENTE C , GAGLIONE D , et al. On model, algorithms, and experiment for micro-Doppler-based recognition of ballistic targets[J]. IEEE Trans.on Aerospace and Electronic Systems, 2017, 53 (3): 1088- 1108. doi: 10.1109/TAES.2017.2665258
6	SUN H X , ZHENG L , NING X . Ballistic missile warhead reco-gnition based on micro-Doppler frequency[J]. Defence Science Journal, 2008, 58 (6): 705- 709. doi: 10.14429/dsj.58.1697
7	SU W G, WANG H Q, QIN Y L, et al. The ISAR imaging of ballistic midcourse targets based on sparse Bayesian learning[C]// Proc. of the IEEE China Summit and International Conference on Signal and Information Processing, 2013: 597-601.
8	YANG H, ZHANG Y S, DING W Z. A fast recognition method for space targets in ISAR images based on local and global structural fusion features with lower dimensions[EB/OL]. [2024-05-26]. https://doi.org/10.1155/2020/3412582.
9	DU C , TIAN L , CHEN B , et al. Region-factorized recurrent attentional network with deep clustering for radar HRRP target recognition[J]. Signal Processing, 2021, 183, 108010. doi: 10.1016/j.sigpro.2021.108010
10	XIANG Q , WANG X D , LAI J , et al. Multi-scale group-fusion convolutional neural network for high-resolution range profile target recognition[J]. IET Radar, Sonar & Navigation, 2022, 16 (12): 1997- 2016.
11	PERSICO A R , ILIOUDIS C V , CLEMENTE C , et al. Novel classification algorithm for ballistic target based on HRRP frame[J]. IEEE Trans.on Aerospace and Electronic Systems, 2019, 55 (6): 3168- 3189. doi: 10.1109/TAES.2019.2905281
12	王彩云, 黄盼盼, 李晓飞, 等. 基于AEPSO-SVM算法的雷达HRRP目标识别[J]. 系统工程与电子技术, 2019, 41 (9): 1984- 1989. doi: 10.3969/j.issn.1001-506X.2019.09.10
	WANG C Y , HUANG P P , LI X F , et al. Radar HRRP target recognition based on AEPSO-SVM algorithm[J]. Systems Engineering and Electronics, 2019, 41 (9): 1984- 1989. doi: 10.3969/j.issn.1001-506X.2019.09.10
13	肖永生, 黄丽贞, 朱劼昊, 等. 最大间隔核优化的雷达目标识别新方法[J]. 信号处理, 2014, 30 (7): 783- 788.
	XIAO Y S , HUANG L Z , ZHU Z H , et al. A new radar target recognition method of the maximal margin Kernel optimization[J]. Journal of Signal Processing, 2014, 30 (7): 783- 788.
14	翟夕阳, 王晓丹, 李睿, 等. 基于二叉树直觉模糊SVM的弹道目标HRRP识别[J]. 火力与指挥控制, 2017, 42 (10): 64- 68.
	ZHAI X Y , WANG X D , LI R , et al. Ballistic target recognition of HRRP based on intuitionistic fuzzy binary tree SVM[J]. Fire Control & Command Control, 2017, 42 (10): 64- 68.
15	TU J, HUANG T, LIU X S, et al. A novel HRRP target recognition method based on LSTM and HMM decision-making[C]// Proc. of the 25th International Conference on Automation and Computing, 2019.
16	LI X Q , LIU Z M , HUANG Z T . Attention-based radar PRI modulation recognition with recurrent neural networks[J]. IEEE Access, 2020, 8, 57426- 57436. doi: 10.1109/ACCESS.2020.2982654
17	MOHSIN A , FAZEEL A , UD I D , et al. A hybrid CNN and RNN variant model for music classification[J]. Applied Sciences, 2023, 13 (3): 1476. doi: 10.3390/app13031476
18	LI K M , DAI X N , LUO Y , et al. Review of radar micro-miotion feature extraction and recognition for ballistic targets[J]. Journal of Air Force Engineering University, 2023, 24 (11): 7- 17.
19	LIU J Q, CHEN B, CHEN W C, et al. Radar HRRP target recognition with target aware two-dimensional recurrent neural network[C]//Proc. of the IEEE International Conference on Signal Processing, Communications and Computing, 2019.
20	CHU Y C, GUO Z H. Attention enhanced spatial temporal neural network for HRRP recognition[C]//Proc. of the IEEE International Conference on Acoustics, Speech and Signal Processing, 2021: 3805-3809.
21	刘家麒, 陈渤, 介茜. 基于注意力机制和双向GRU模型的雷达HRRP目标识别[J]. 雷达学报, 2019, 8 (5): 589- 597.
	LIU J L , CHEN B , JIE Q . Radar high-resolution range profile target recognition based on attention mechanism and bidirectional gated recurrent[J]. Journal of Radars, 2019, 8 (5): 589- 597.
22	WANG S M , SUN J , FU L H , et al. Identification of red jujube varieties based on hyperspectral imaging technology combined with CARS-IRIV and SSA-SVM[J]. Journal of Food Process Engineering, 2022, 45 (10): e14137. doi: 10.1111/jfpe.14137
23	JIA T B, YANG X L, MEI Y, et al. Transformer fault comprehensive diagnosis method based on improved HPO-ELM and D-S evidence theory fusion[C]//Proc. of the 3rd International Conference on Electrical Engineering and Mechatronics Technology, 2023: 30-34.
24	ZHAO H X , ZHOU Z G , ZHANG P Z . Forecasting of the short-term electricity load based on WOA-BILSTM[J]. International Journal of Pattern Recognition and Artificial Intelligence, 2023, 37 (11): 2359018. doi: 10.1142/S0218001423590188
25	PRATAP S S , SHAMIK T . A dual multimodal biometric authentication system based on WOA-ANN and SSA-DBN techniques[J]. Sci, 2023, 5 (1): 10. doi: 10.3390/sci5010010
26	LIU Z , PENG Y X . Study on denoising method of vibration signal induced by tunnel portal blasting based on WOA-VMD algorithm[J]. Applied Sciences, 2023, 13 (5): 3322. doi: 10.3390/app13053322
27	JIANG Z H , CHEN R W , LIU H , et al. A WSN clustering routing method based on an improved WOA-optimized K-means clustering algorithm[J]. Journal of Physics: Conference Series, 2023, 2670 (1): 12- 19.
28	ZHENG Z G , NITIN S , MISHRA K K , et al. Guided dynamic particle swarm optimization for optimizing digital image watermarking in industry applications[J]. Future Generation Computer Systems, 2018, 88, 92- 106. doi: 10.1016/j.future.2018.05.027
29	王童. 基于MSWOA改进Attention-BiGRU模型的电力负荷预测[J]. 软件导刊, 2023, 22 (10): 84- 89.
	WANG T . Power load forecasting based on MSWOA improved attention-BiGRU model[J]. Software Guide, 2023, 22 (10): 84- 89.
30	马泽龙. 基于改进鲸鱼算法的神经网络电力负荷预测研究[D]. 南京: 南京理工大学, 2022.
	MA Z L. Research on power load forecasting of neural network based on improved whale algorithm[D]. Nanjing: Nanjing University of Science & Technology, 2022.
31	HU L Y , ZHAO S Q . Bi-GRU model based on pooling and attention for text classification[J]. International Journal of Wireless and Mobile Computing, 2021, 21 (1): 26- 31. doi: 10.1504/IJWMC.2021.119057

参数	数值
带宽/GHz	1
中心频率/GHz	10
雷达位置	(96.174°E, 24.025°N, 0 m)
目标发射点	(87.068°E, 20.758°N, 15 m)
目标落点	(106.000°E, 25.644°N, 878 m)

学习率	准确率	置信
0.1	88.67	89.89
0.01	94.40	94.25
0.001	93.77	94.43
0.000 1	85.40	81.55
0.000 01	63.07	53.78

节点数	准确率	置信度
10	95.03	94.59
20	91.60	95.28
30	93.77	94.43
40	95.93	94.78
50	93.37	95.21

l₂正则化参数	准确率	置信度
1e-3	85.10	84.46
1e-4	86.57	93.69
1e-5	95.00	97.58
1e-6	95.90	98.48
1e-7	95.47	98.73

算法	平均准确率	平均置信度
GRU^[21]	72.58	63.02
PSO-SVM^[29]	89.17	92.28
HPO^[24]-BiGRU	93.03	93.13
WOA^[25]-BiGRU	93.17	93.84
MSWOA^[30]-BiGRU	94.60	94.39
IWOA^[31]-BiGRU	95.37	94.48
本文算法	96.87	94.98