CBA: multi source fusion model for fast and intelligent target intention identification

doi:10.23919/JSEE.2024.000023

Journal of Systems Engineering and Electronics ›› 2024, Vol. 35 ›› Issue (2): 406-416.doi: 10.23919/JSEE.2024.000023

• SYSTEMS ENGINEERING • Previous Articles

CBA: multi source fusion model for fast and intelligent target intention identification

Shichang WAN¹^,²(), Qingshan LI¹^,*(), Xuhua WANG¹(), Nanhua LU¹()

¹ School of Computer Science and Technology, Xidian University, Xi’an 710071, China
² School of Mathematics and Computer Application, Shangluo University, Shangluo 726000, China

Received:2023-03-30 Online:2024-04-18 Published:2024-04-18
Contact: Qingshan LI E-mail:wanshichang@126.com;qshli@mail.xidian.edu.cn;daleiwxh@163.com;lunanhua@qq.com
About author:
WAN Shichang was born in 1984. He received his M.S. degree from Shaanxi Normal University, Xi’an, China, in 2010. He is currently pursuing his Ph.D. degree in the School of Computer Science and Technology, Xidian University. His research interests include UAV swarm combat application, multi-source data fusion, and intelligent data analysis. E-mail: wanshichang@126.com

LI Qingshan was born in 1973. He received his B.S., M.S. and Ph.D. degrees from Xidian University, Xi’an, China, in 1995, 1999 and 2003, respectively. He is a professor in Xidian University. His research interests include agent-oriented software engineering, intelligent data analysis, and decision support system. E-mail: qshli@mail.xidian.edu.cn

WANG Xuhua was born in 1984. He received his B.S., M.S. and Ph.D. degrees from Air Force Engineering University, Xi’an, China, in 2007, 2009 and 2013, respectively. He is a lecturer in Xidian University. His research interests include radar network modeling, UAV swarm combat application, and anti-UAV technology. E-mail: daleiwxh@163.com

LU Nanhua was born in 1999. He received his B.S. degree from Wuhan University of Technology, Wuhan, China, in 2021. He is studying for his M.S. degree at Xidian University. His research interests include multi-UAV mission planning, radar network modeling, and 3D scene simulation. E-mail: lunanhua@qq.com
Supported by:
This work was supported by the National Natural Science Foundation of China (61502523).

Abstract

Abstract:

How to mine valuable information from massive multi-source heterogeneous data and identify the intention of aerial targets is a major research focus at present. Aiming at the long-term dependence of air target intention recognition, this paper deeply explores the potential attribute features from the spatiotemporal sequence data of the target. First, we build an intelligent dynamic intention recognition framework, including a series of specific processes such as data source, data preprocessing, target space-time, convolutional neural networks-bidirectional gated recurrent unit-atteneion (CBA) model and intention recognition. Then, we analyze and reason the designed CBA model in detail. Finally, through comparison and analysis with other recognition model experiments, our proposed method can effectively improve the accuracy of air target intention recognition, and is of significance to the commanders’ operational command and situation prediction.

Key words: intention, massive data, deep network, artificial intelligence

Shichang WAN, Qingshan LI, Xuhua WANG, Nanhua LU. CBA: multi source fusion model for fast and intelligent target intention identification[J]. Journal of Systems Engineering and Electronics, 2024, 35(2): 406-416.

Figures/Tables 21

Fig 1

Fig 2

Table 1

Table 2

Table 3

Fig 3

Fig 4

Fig 5

Fig 6

Table 4

Table 5

Table 6

Table 7

Table 8

Table 9

Table 10

Fig 7

Fig 8

Table 11

Table 12

Table 13

References 32

1	BEN-BASSAT M, FREEDY A Knowledge requirements and management in expert decision support systems for (military) situation assessment. IEEE Trans. on Systems, Man, and Cybernetics, 1982, 12 (4): 479- 490. doi: 10.1109/TSMC.1982.4308852
2	ZHOU W W, ZHANG J Y, GU N N, et al. Recognition of combat intention with insufficient expert knowledge. Proc. of the 3rd International Conference on Computational Modeling, Simulation and Applied Mathematics, 2018: 316−321.
3	GAO X G, YANG Y Threat assessment for warship air defense based on Bayesian network. Tactical Missile Technology, 2020, (4): 47- 57.
4	HOSSEINI S, IVANOV D Bayesian networks for supply chain risk, resilience and ripple effect analysis: a literature review. Expert Systems with Applications, 2020, 161, 113649. doi: 10.1016/j.eswa.2020.113649
5	YANG A W, LI Z W, XU A, et al Threat level assessment of the air combat target based on weighted cloudy dynamic Bayesian networks. Flight Dynamics, 2020, 38 (4): 87- 94.
6	GE S. Research on tactical intention recognition based on rule discovery and Bayesian reasoning. Harbin: Harbin Engineering University, 2015. (in Chinese)
7	GE S, XIA X Z Study of optimal decision model based on BN. Applied Mechanics and Materials, 2014, 687, 1552- 1556.
8	YANG J, GAO W Y, LIU J Threat assessment method based on Bayesian network. Journal of PLA University of Science and Technology, 2010, 11 (1): 43- 48.
9	YAO Q K, LIU S J, HE X Y, et al Research and prospect of battlefield target operational intention recognition. Journal of Command and Control, 2017, 3 (2): 127- 131.
10	LI A, ZHENG B Y, LI L Intelligent transportation application and analysis for multi-sensor information fusion of Internet of Things. IEEE Sensors Journal, 2020, 21, 25035- 25042.
11	GONG Y M, MA Z Y, WANG M J, et al. A new multi-sensor fusion target recognition method based on complementarity analysis and neutrosophic set. Symmetry, 2020, 12(9): 1435−1435.
12	LI B N, PEI W, LI J Q, et al. ISAR target recognition system design based on artificial intelligence. Proc. of the 11th International Symposium on Multispectral Image Processing and Pattern Recognition, 2020: 149−155.
13	XUE J J, ZHU J, XIAO J Y, et al Panoramic convolutional long short-term memory networks for combat intension recognition of aerial targets. IEEE Access, 2020, 8, 183312- 183323. doi: 10.1109/ACCESS.2020.3025926
14	LI Z W, LI S Q, PENG M Y, et al Air combat intention recognition method of target based on LSTM improved by attention mechanism. Electronics Optics & Control, 2023, 30 (3): 1- 9.
15	QIAO D F, LIANG Y, MA C X, et al Recognition and prediction of group target intention in multi-domain operations. Systems Engineering and Electronics, 2022, 44 (11): 3403- 3412.
16	TENG F, LIU S, SONG Y F BiLSTM-Attention: an air target tactical intention recognition model. Aero Weaponry, 2021, 28 (5): 24- 32.
17	DAI F H. Research on intention recognition technology in battlefield based on improved Markov model. Changsha: National University of Defense Technology, 2019. (in Chinese)
18	WANG X H, LIN Z K, HU Y H, et al Learning embedding features based on multisense-scaled attention architecture to improve the predictive performance of air combat intention recognition. IEEE Access, 2022, 10, 104923- 104933. doi: 10.1109/ACCESS.2022.3204706
19	FAN H Air target intention estimation for airborne situation awareness system. Ordnance Industry Automation, 2022, 41 (4): 14- 18.
20	ZHANG J T, ZHOU W N, YAN X F, et al Intelligent prediction for behavioral intent of marine targets based on LSTM and fuzzy reasoning. Journal of China Academy of Electronics and Information Technology, 2022, 17 (9): 897- 904.
21	YANG Y T, YANG J, LI J G Research on air target tactical intention recognition based on EMEBN. Fire Control & Command Control, 2022, 47 (5): 163- 170.
22	HU Z Y, LIU H L, GONG S J, et al Target intention recognition based on random forest. Modern Electronics Technique, 2022, 45 (19): 1- 8.
23	CAI Q Y. Research on power behavior analysis method based on load data preprocessing. Nanjing: Nanjing University of Posts and Telecommunications, 2022. (in Chinese)
24	LI Z L, XU K, XIE J F, et al Deep multiple instance convolutional neural networks for learning robust scene representations. IEEE Trans. on Geoscience and Remote Sensing, 2020, 58 (5): 3685- 3702. doi: 10.1109/TGRS.2019.2960889
25	LIU C, TANG X, MA J J, et al. Remote sensing images feature learning based on multibranch networks. Proc. of the IEEE International Geoscience and Remote Sensing Symposium, 2020: 2057−2060.
26	WU J X Introduction to convolutional neural networks. National Key Lab for Novel Software Technology, 2017, 5 (23): 495- 495.
27	ZHANG T, LIU X L, GAO Y P, et al Sentiment analysis of attention mechanism based on convolutional neural network and bidirectional gated recurrent unit network. Science Technology and Engineering, 2021, 21 (1): 269- 274.
28	VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need. Proc. of the 31st Conference on Neural Information Processing System, 2017: 6000−6010.
29	BELLO I, ZOPH B, VASWANI A, et al. Attention augmented convolutional networks. Proc. of the IEEE/CVF International Conference on Computer Vision, 2019: 3286−3295.
30	JIANG Q, HUANG R M, HUANG Y C, et al Application of BP neural network based on genetic algorithm optimization in evaluation of power grid investment risk. IEEE Access, 2019, 7, 154827- 154835. doi: 10.1109/ACCESS.2019.2944609
31	ZHONG Y W, WU X L Effects of cost-benefit analysis under back propagation neural network on financial benefit evaluation of investment projects. PloS ONE, 2020, 15 (3): e0229739. doi: 10.1371/journal.pone.0229739
32	XUE J K, SHEN B A novel swarm intelligence optimization approach: sparrow search algorithm. Systems Science & Control Engineering, 2020, 8 (1): 22- 34.

Environment	Label
Sun	1
Rain	2
Thunderstorm	3
Fog	4
Gale	5

Category	Description	Label
Radar status	Air-to-air radar power-on status	−2
	Ground radar power-on status	−1
	Air-to-surface radar power-on status	0
	Silence	1
Motivation type	8-type	0
	Low jump	1
	High speed shake	2
	S-type	3
	Turn around	4
Friend-or-foe attribute	Hostile plane	0
	Friendly aircraft	1
	Unidentified aircraft	2
Target size	Big target	0
	Middle target	1
	Small target	2

Category	Range
Velocity/(km/h)	600−850
	750−950
	735−1470
Height/m	50−1000
	1000−6000
	15000
	10000−11000
RCS/dBsm	Radar reflection cross-sectional area
Position change	−0.5
	−0.4
	−0.3
	−0.2
	−0.1
	0.1
	0.2
	0.3
	0.4
	0.5
Formation	Cuneiform (0)
	Trapezoid (1)
	Herringbone (2)
	Diamond (3)
	Longitudinal shape (4)
Azimuth/mil	0−6400

Name	Version/configuration
Operating system	Windows10
Computer memory	8GB
CPU	i7-9700 CPU
Programming language	Python 3.8.0
Compilation environment	Anaconda 3
Deep learning network	TensorFlow 2.0.0，keras2.8.0
Library	Os,numpy,pandas, etc.

Confusion matrix		Result of intention recognition		Total
Confusion matrix		1	0	Total
Real label	1	TP	FN	Actual positive (TP+FN)
Real label	0	FP	TN	Actual negative(FP+TN)
Total		Positive recognition (TP+FP)	Negative recognition (FN+TN)	TP+FN+ FP+TN

CBA: multi source fusion model for fast and intelligent target intention identification

RichHTML

PDF (PC)

Knowledge

Abstract

Cite this article

Share this article

Figures/Tables 21

References 32

Related Articles 4

Recommended Articles

Metrics

Comments

Filter number	Loss	MAE
8	0.1523	0.3684
16	0.1038	0.3219
32	0.0667	0.2580
64	0.0532	0.2316
128	0.0268	0.1632

Kernel size	MAE	R²
1	0.1573	1.246 9e+06
2	0.1605	1.299 4e+06
3	0.1988	1.990 3e+06
4	0.1725	1.539 7e+06

Dropout	Loss	R²
0.1	0.0525	2.625 9e+06
0.2	0.0305	1.523 1e+06
0.3	0.0249	1.220 4e+06

Parameter	Value
Filter	128
CNN activation	Relu
CNN dropout	0.3
Dense activation	Sigmoid
Time steps	15
Loss	MSE
Optimizer	Adam
Metrics	MAE
Epoch	50
Batch size	32
Dropout	0.3

Layer	Value
Convolution	Filers=64, kernel-size=2, activation=relu
Dropout	0.3
BiGRU	Unit=32
Dropout	0.3
Attention	Attention_3d_block2, flatten
Dense	Activation=sigmoid

Name	Loss	MAE	R²	Epoch/s	Step/s
GRU	0.1757	0.4173	−8.786 6e+06	39	39
LSTM	0.1708	0.4048	−8.52 4e+06	31	31
CBA	0.0303	0.1738	−1.514 4e+06	56	56
BiLSTM	0.0591	0.2429	−2.956 4e+06	66	66

Intention	Accuracy	Recall	F1-score
Attack	93.8	92.5	93.2
Transport	95.3	86.3	90.4
Electronic jamming	90.8	97.4	94.0
Early warning detection	89.2	97.5	93.2
Scout	96.3	97.5	96.7
Retreat	94.7	100	97.3

[1]	Jun TANG, Gang LIU, Qingtao PAN. Review on artificial intelligence techniques for improving representative air traffic management capability [J]. Journal of Systems Engineering and Electronics, 2022, 33(5): 1123-1134.
[2]	Liting SUN, Xiang WANG, Zhitao HUANG. Unintentional modulation microstructure enlargement [J]. Journal of Systems Engineering and Electronics, 2022, 33(3): 522-533.
[3]	Wantian WANG, Ziyue TANG, Yichang CHEN, Yongjian SUN. Parity recognition of blade number and manoeuvre intention classification algorithm of rotor target based on micro-Doppler features using CNN [J]. Journal of Systems Engineering and Electronics, 2020, 31(5): 884-889.
[4]	Yuanyuan Zhang and Wenhai Wu. Integrated parallel forecasting model based on modified fuzzy time series and SVM [J]. Journal of Systems Engineering and Electronics, 2017, 28(4): 776-.

Parameter	LSTM	CBA	BiLSTM	GRU
Max	0.280811012	0.273370922	0.280811012	0.276434325
Min	0.055925872	0.052450593	0.055925872	0.073835136
Average	0.180188963	0.160373669	0.180188963	0.175134731