为了解决降效类电磁威胁目标威胁评估面临的专家主观性不易避免、目标类型庞杂、影响因素复杂、时间约束性强等问题,建立了相适应的威胁评估指标体系,并将云模型与极限学习机结合在一起,提出了云极限学习机(cloud ELM)的威胁评估方法,构建了基于云ELM的降效类电磁威胁目标威胁评估模型。仿真结果表明,建立的指标体系与提出的评估方法能够解决降效类电磁威胁目标威胁评估面临的准确性与时效性问题。
In order to solve the problem of target threat assessment of reduced effect electromagnetic threat, such as that the subjectivity of experts was not easy to avoid, target types were numerous and jumbled, complex influencing factors, strong time constraints, the adaptive threat evaluation index system was established. A threat assessment method based on cloud ELM (Cloud) was proposed by combining cloud model with extreme learning machine, and a threat assessment model of reduced effectiveness electromagnetic threat target based on cloud ELM was constructed. The simulation results show that the index systems and assessment method proposed in this paper is able to solve the problem of accuracy and timeliness of the threat assessment of the reduced effect electromagnetic threat target
[1] WHITE F. A model for data fusion[C]//Proceedings of the First National Symposium on Sensor Fusion.[S.l.:s.n.], 1988, 12: 143-158.
[2] STEINBERG A N, BOWMAN C L, WHITE F. Revisions to the JDL data fusion model[C]//Joint NATO/IRIS Conference Proceedings.[S.l.:s.n.], 1998: 430-441.
[3] ZHANG Yinyan, LI Bicheng, CUI Jiawei. Method of targettThreat assessment based on cloudy bayesian network[J].Computer Science, 2013(10): 127-131.
[4] 陈开元, 宋元. 海面目标威胁评估指标体系分析[J].舰船电子工程, 2009, 29(6): 66-70.
[5] 冯小平, 李鹏, 杨绍全. 通信对抗原理[M].西安: 西安电子科技大学出版社, 2009: 42-63.
[6] 贺平. 雷达对抗原理[M].北京: 国防工业出版社, 2016: 7-9.
[7] 王守信, 张莉, 王帅, 等. 一种目标可满足性定性、定量表示与推理方法[J].软件学报, 2011, 22(4): 593-608.
[8] 雷英杰,王宝树, 路艳丽. 基于自适应直觉模糊推理的威胁评估方法[J].电子与信息学报, 2007, 29(12): 2805-2809.
[9] 张银燕, 李弼程, 崔家玮. 基于云贝叶斯网络的目标威胁评估方法[J].计算机科学, 2013,40(10): 127-131.
[10] HUANG Guangbin, ZHOU Hongming, DING Xiaojian, et al. Extreme learning machine for regression and multiclass classification[J].IEEE Transactions on Systems, MAN, and Cybernetics: Part B(Cybernetics), 2012, 42(2), 513-529.
[11] YANG Yimin, WU Jonathan Q M. Extreme learning machine with subnetwork hidden nodes for regression and classification[J].IEEE Transactions on Cybernetics, 2016,46(12): 1-14.
