To facilitate readers in timely understanding of the latest accepted papers in Systems Engineering and Electronic Technology, and to accelerate the dissemination of the latest scientific research results, this journal will regularly organize information on papers that have been prioritized for publication on our website and China National Knowledge Infrastructure (CNKI). Please stay tuned!
01
Title:Blue Force Virtual Entities for LVC Training
Modeling Decision Making for Close Air Combat
Authors:Gao Ang; Dong Zhiming; Li Liang; Duan Li; Guo Qisheng
Affiliation:Army Academy of Armored Forces
Priority Publication Date:2021-02-03
Priority Publication Address:https://kns.cnki.net/kcms/detail/11.2422.TN.20210203.1037.014.html
Abstract
The Live-Virtual-Constructive (LVC) approach provides strong support for the important military topic of close air combat training. Addressing the actual decision-making modeling needs for blue force virtual entities, this paper defines the weight space and structure space of neural networks from the perspective of optimization theory, based on a comparative analysis of deep reinforcement learning and classical intelligent optimization methods. It proposes an intelligent optimization-based evolutionary neural network decision model and its solution method. By analyzing the tactical characteristics of close air combat, the aircraft flight motion model, and the actual decision-making modeling requirements, key flight states, action spaces, and fitness functions for aircraft are designed to achieve end-to-end perception and decision-making for the blue force. The proposed method is generalizable to current intelligent optimization and its improved algorithms, as well as different structured neural networks. Finally, a decision model based on classical genetic neural networks and a solution example are provided. Results indicate that the proposed method enables blue force aircraft to learn the operational characteristics of opponents through adversarial data, validating the effectiveness of the model and method.
02
Title:Three-Layer Planning Model for Offensive Electronic Countermeasure Force Requirements
Authors:Zhao Luda; Wang Bin; Zeng Wei
Affiliation:National University of Defense Technology, College of Electronic Countermeasures
Priority Publication Date:2021-02-03
Priority Publication Address:
https://kns.cnki.net/kcms/detail/11.2422.tn.20210203.1132.040.html
Abstract
To address the operational forecasting problem of electronic countermeasure force requirements, a model based on three-layer planning is proposed to estimate the number of forces required at different stages of combat, enhancing command efficiency and effectiveness. First, a detailed analysis of electronic countermeasure tactics in land offensive combat leads to the establishment of a normative engagement model for electronic countermeasures. Then, a three-layer planning model for electronic countermeasure force requirements is established based on the combat methods and stages, introducing a satisfaction function operator based on fuzzy membership to quantify model variables, and solving for optimal satisfaction at each layer using genetic algorithms to derive optimal force requirements. Finally, simulation examples yield optimal force requirements for electronic countermeasures at three different combat stages, demonstrating the applicability and efficiency of the proposed algorithm compared to three multi-layer planning solution algorithms.
03
Title:Modeling and Analysis Method of Certainty Reliability Based on Probability Measure
Authors:Li Boyuan; Kang Rui; Yu Li
Affiliation:Beijing University of Aeronautics and Astronautics
Priority Publication Date:2021-02-03
Priority Publication Address:
https://kns.cnki.net/kcms/detail/11.2422.TN.20210203.1037.012.html
Abstract
Certainty reliability theory, based on principles of reliability science, comprehensively considers product performance parameters and uncertainties during the product design phase, achieving forward and quantitative design of products. To translate certainty reliability theory into a practical reliability design process, this paper proposes a modeling and analysis method based on probability measures. The study determines key performance parameters and performance thresholds through Functional Performance Margin Analysis (FPMA), establishes performance equations, margin equations, and degradation equations through theoretical analysis, simulation modeling, and experimental modeling; quantifies parameter uncertainties, conducts certainty reliability analysis, and calculates quantitative reliability indicators. This paper provides innovative theoretical exploration and engineering practice for reliability design workflows, and to further illustrate the application process of this method, a case study of a specific electrical connector is conducted.
04
Title:Centroid Positioning Method for GNSS Interference Sources Based on Signal Propagation Correction
Authors:Liu Rui; Yang Zhiwei; Chen Qidong; Liao Guisheng; Zhen Weimin
Affiliation:National Key Laboratory of Radar Signal Processing, Xi’an University of Electronic Science and Technology
Priority Publication Date:2021-02-03
Priority Publication Address:
http://kns.cnki.net/kcms/detail/11.2422.tn.20210203.1806.070.html
Abstract
To address the widespread issue of GNSS interference, a centroid positioning method based on received signal strength (RSS) is proposed. This method faces challenges such as difficulty in extracting RSS values and significant positioning errors in complex urban environments with obstructions and multipath effects. The proposed method utilizes carrier-to-noise ratio (CNR) information output from GNSS receivers and employs neural networks to predict interference signal propagation factors, achieving GNSS interference source positioning through centroid weighting. Experimental results indicate a significant improvement in positioning accuracy with this method.
05
Title:Task-Oriented Dynamic Reconstruction Strategy for Civil Aircraft DIMA
Authors:Wang Peng; Liu Jiachen; Dong Lei; Zhao Changxiao
Affiliation:Civil Aviation University of China
Priority Publication Date:2021-02-03
Priority Publication Address:
https://kns.cnki.net/kcms/detail/11.2422.TN.20210203.0950.002.html
Abstract
To enhance the capability of distributed integrated modular avionics (DIMA) systems in civil aircraft to respond to flight mission switches and resource failures, this study analyzes the dynamic reconstruction environment and mechanisms of DIMA, forming a task-oriented dynamic reconstruction strategy. Based on this, a relationship matrix among tasks, functions, and resources is proposed, along with an evaluation system for dynamic reconstruction effectiveness, introducing a combinatorial optimization weighting method to calculate priority weights for functional applications under different task modes. Finally, case analysis compares the task-oriented dynamic reconstruction strategy with ordinary reconstruction strategies regarding the support level of system resources for system functional integrity under different task modes. Research results indicate that as the number of module failures increases, the task-oriented dynamic reconstruction strategy significantly enhances the effectiveness of dynamic reconstruction.
06
Title:Design and Performance Analysis of Double Joint Transceiver Spatial Modulation Scheme
Authors:Bu Xiangyan; Bai Zhiquan; Sun Jian; Pang Ke; Hao Xinhong; Ma Piming
Affiliation:School of Information Science and Engineering, Shandong University
Priority Publication Date:2021-02-03
Priority Publication Address:
https://kns.cnki.net/kcms/detail/11.2422.TN.20210203.1507.054.html
Abstract
This paper proposes a double joint transceiver spatial modulation (DJSM) scheme, which adds double spatial modulation (DSM) to the traditional joint transceiver spatial modulation (JSM), enhancing the information transmission rate and reliability of spatial modulation (SM) systems. The DJSM scheme performs two independent SM processes at the transmitter, mapping the input information bit sequence to two modulation symbols and their corresponding activated transmitting antenna groups, significantly improving the spectral efficiency of the system. To distinguish the two modulation symbols, optimal angle rotation is applied before transmission. Additionally, based on joint upper bound techniques, the average bit error probability (ABEP) of the DJSM scheme is provided. Finally, simulation results demonstrate that the ABEP performance of the proposed DJSM scheme significantly outperforms that of traditional JSM schemes and orthogonal joint transceiver spatial modulation (QJSM) schemes.
07
Title:MIMO Model-Based Multivariate Hybrid Modulation DCSK Scheme
Authors:Chen Xiaoting; Wang Nian; Ding Dawei; Zhang Gongquan; Lu Yu
Affiliation:Anhui University
Priority Publication Date:2021-02-03
Priority Publication Address:
https://kns.cnki.net/kcms/detail/11.2422.TN.20210203.1507.058.html
Abstract
To enhance the error performance of multi-input multi-output differential chaotic shift keying (MIMO-DCSK) systems, this paper proposes a multivariate hybrid modulation MIMO-DCSK scheme (M-ary hybrid modulation MIMO-DCSK, MHM-MIMO-DCSK). This scheme splits a frame of multivariate information, modulating it separately through chaotic spreading codes and index codes for efficient transmission. The receiver utilizes the quasi-orthogonal characteristics of chaotic sequences to eliminate multi-antenna and multi-path interference, and then performs equal gain combining (EGC) on the results of multi-antenna correlation operations, achieving full diversity gain of the multipath MIMO system without requiring channel state information (CSI). Simulation results indicate that the proposed MHM-MIMO-DCSK outperforms existing multivariate MIMO-DCSK schemes, especially showing significant performance improvements under high-order modulation conditions.
08
Title:Intrusion Detection Based on Feature Selection and Tree-Structured Parzen Estimation
Authors:Jin Zhigang; Wu Tong
Affiliation:Tianjin University
Priority Publication Date:2021-02-03
Priority Publication Address:
https://kns.cnki.net/kcms/detail/11.2422.TN.20210203.1507.056.html
Abstract
In response to the rapidly changing risks and challenges in the current cybersecurity landscape, this paper proposes an intrusion detection method based on correlation analysis for feature selection and tree-structured Parzen estimation. This method compresses data dimensions through a correlation analysis-based feature selection approach, screening the original dataset to generate a new subset of features; it optimizes the random forest algorithm using the TPE algorithm from sequential model optimization. Experimental results show that the proposed method achieves higher detection efficiency while enhancing overall performance compared to other intrusion detection methods that apply machine learning algorithms, effectively improving the practicality of intrusion detection technology.
09
Title:Instantaneous Coastline Automatic Extraction Algorithm Based on Improved Deep Learning Network for SAR Images
Authors:Wang Bin; Wang Guoyu
Affiliation:School of Information Science and Engineering, Ocean University of China
Priority Publication Date:2021-02-03
Priority Publication Address:
https://kns.cnki.net/kcms/detail/11.2422.TN.20210203.0958.004.html
Abstract
To address the issues of accuracy and automation in the extraction of large-scale instantaneous coastlines during synthetic aperture radar (SAR) image interpretation, a new instantaneous coastline automatic extraction algorithm based on deep learning networks is proposed. First, SAR images are enhanced using Lee filtering to suppress coherent speckle. Next, an upgraded residual network is used as the backbone network to extract features of seawater targets at four levels. Then, the four-level features are fully extracted through global convolution networks, dense connection networks, and decoder networks, producing segmentation results for seawater. Finally, the Sobel operator is used to separate the coastline and fuse it with the original SAR image for clearer results. Comparisons with experimental results from fully convolutional networks and refinement networks show that the proposed algorithm achieves more accurate coastline extraction, reducing false alarms and missed detections, and demonstrating better performance.
WeChat ID｜Systems Engineering and Electronic Technology
WeChat ID｜JSEE