雷达目标检测深层自编码器自适应优化算法
|
摘要
研究了现阶段雷达低小慢目标探测技术的难点与方法。分析了深层自编码器基本模型与算法,通过引入自适应学习理论,提出了基于Rumelhart函数的深层自编码器自适应算法(RDAAA),并证明了算法的收敛性。优化算法避免了网络训练过程中出现惩罚过度的现象,克服了学习速率过高导致网络振荡发散,或学习速率过小降低网络收敛速度等缺陷。利用两种数据集对RDAAA、基于交叉熵函数的深层自编码器学习算法(CDAA)与误差反向传播算法(BPA)进行模式识别能力分析,结果表明在确定限定误差与选取最佳学习速率的情况下,RDAAA相对于CDAA与BPA收敛速度最快,正确识别率更高。围绕雷达目标检测与深度学习理论,分析了低小慢目标特性,将目标检测问题转化为模式分类问题,利用上述三种算法进行目标检测仿真实验,结果表明RDAAA与CDAA的性能明显优于BPA,且RDAAA的检测率更高,特别是处于低信噪比阶段,仍可保持较高的发现概率。
The difficulties and methods of radar low-small-slow target detection technology are studied. It analyzes the basic model and algorithm of deep autoencoder. By introducing adaptive learning theory, Rumelhart function-Deep Atuoecoder Adaptive learning Algorithm(RDAAA)is proposed, and the convergence of the algorithm is proved. The optimization algorithm avoids the phenomenon of excessive punishment in the network training process, overcomes the disadvantages of excessively high learning rate leading to oscillating and divergent in network or low learning rate leading to reduce network convergence speed. Two types of data sets are used to analyze the pattern recognition ability of RDAAA, Cross-entropy function-Deep Autoencoder learning Algorithm(CDAA)and error Back Propagation Algorithm(BPA). In the case of determining the limit error and selecting the optimal learning rate, the results show that RDAAA has the fastest convergence rate and higher correct recognition rate than CDAA and BPA. Focusing on radar target detection and deep learning theory, the characteristics of low-small-slow target are analyzed, and the target detection problem is transformed into a problem of pattern classification. Using the above three algorithms for target detection simulationexperiments, the results show that the performance of RDAAA and CDAA is significantly better than that of BPA, and the detection rate of RDAAA is higher, especially in the low signal-to-noise ratio stage, and the high probability of discovery can still be maintained.
The difficulties and methods of radar low-small-slow target detection technology are studied. It analyzes the basic model and algorithm of deep autoencoder. By introducing adaptive learning theory, Rumelhart function-Deep Atuoecoder Adaptive learning Algorithm(RDAAA)is proposed, and the convergence of the algorithm is proved. The optimization algorithm avoids the phenomenon of excessive punishment in the network training process, overcomes the disadvantages of excessively high learning rate leading to oscillating and divergent in network or low learning rate leading to reduce network convergence speed. Two types of data sets are used to analyze the pattern recognition ability of RDAAA, Cross-entropy function-Deep Autoencoder learning Algorithm(CDAA)and error Back Propagation Algorithm(BPA). In the case of determining the limit error and selecting the optimal learning rate, the results show that RDAAA has the fastest convergence rate and higher correct recognition rate than CDAA and BPA. Focusing on radar target detection and deep learning theory, the characteristics of low-small-slow target are analyzed, and the target detection problem is transformed into a problem of pattern classification. Using the above three algorithms for target detection simulationexperiments, the results show that the performance of RDAAA and CDAA is significantly better than that of BPA, and the detection rate of RDAAA is higher, especially in the low signal-to-noise ratio stage, and the high probability of discovery can still be maintained.
引文
[1] Skolnik M I.雷达手册[M].3版.北京:电子工业出版社,2015:1-2.
[2] Skolnik M I.雷达系统导论[M].北京:电子工业出版社,2014:1-3.
[3]陈小龙,关键,黄勇,等.雷达低可观测目标探测技术[J].科技导报,2017,35(11):30-38.
[4]郭建明,谭怀英.雷达技术发展综述及第5代雷达初探[J].现代雷达,2012,34(2):1-3.
[5]袁赛柏,金胜,朱天林.认知雷达技术与发展[J].现代雷达,2016,38(1):1-3.
[6]江涛,孙俊.量子雷达探测目标的基本原理与进展[J].中国电子科学研究院学报,2014,9(1):10-16.
[7]叶利华,王磊,赵利平.低小慢无人机降落野外场景识别方法[J].计算机应用,2017,37(7):2008-2013.
[8]戚春,张聃,蔡云泽.基于交互式多模型的红外/雷达低小慢目标跟踪算法研究[J].上海航天,2012,29(6):37-41.
[9]陈小龙,关键,何友,等.高分辨稀疏表示及其在雷达动目标检测中的应用[J].雷达学报,2017,6(3):239-251.
[10]刘凯,林基明,郑霖,等.基于深度自编码网络的慢速移动目标检测[J].计算机工程,2018,44(2):129-134.
[11] Haykin S,Bhattacharya T K.Modular learning strategy for signal detection in a nonstationary environment[J].IEEE Transactions on Signal Processing,1997,45(6):1619-1637.
[12] Zhai S,Jiang T.Target detection and classification by measuring and processing bistatic UWB radar signal[J].Measurement,2014,47(1):547-557.
[13] He X,Jiang T.Target identification in foliage environment using UWB radar with hybrid wavelet-ICA and SVM method[J].Physical Communication,2014,13:197-204.
[14] Zhai S,Jiang T.Application of Ultra-Wide band radar for sense-through-foliage target detection and recognition[C]//Proceedings of the Third International Conference on Communications,Signal Processing and Systems,2015:479-487.
[15]焦李成,杨淑媛,刘芳,等.神经网络七十年:回顾与展望[J].计算机学报,2016,39(8):1697-1716.
[16]邓力,俞栋.深度学习方法及应用[M].谢磊,译.北京:机械工业出版社,2016:39-46.
[17]吴岸成.神经网络与深度学习[M].北京:电子工业出版社,2016:66-86.
[18]李玉鑑,张婷.深度学习导论及案例分析[M].北京:机械工业出版社,2016:48-56.
[19]颜丹,蒋加伏.基于栈式去噪自动编码器的边际Fisher分析算法[J].计算机工程与应用,2017,53(5):134-139.
[20]赵瑞娟,官金安,谢国栋.稀疏降噪自编码器在IR-BCI的应用研究[J].计算机工程与应用,2017,53(11):167-171.
[21]阮晓刚.神经计算科学-在细胞的水平上模拟脑功能[M].北京:国防工业出版社,2006:356-399.
[22] Hou X,Wan S S,Liu Rui.Quantum gate circuit neural network optimization algorithm based on performance function[C]//Proceedings of 2018 International Conference on Communication,Network and Artificial Intelligence(CNAI 2018),Beijing,China,2018:321-326.
[23]李楠,侯旋.量子自适应前向对传算法研究[J].电子与信息学报,2013,35(11):2778-2783.
[24]田玉芳,尹志盈,姬光荣,等.基于SVM的海面弱目标检测[J].中国海洋大学学报,2013,43(7):104-109.
[2] Skolnik M I.雷达系统导论[M].北京:电子工业出版社,2014:1-3.
[3]陈小龙,关键,黄勇,等.雷达低可观测目标探测技术[J].科技导报,2017,35(11):30-38.
[4]郭建明,谭怀英.雷达技术发展综述及第5代雷达初探[J].现代雷达,2012,34(2):1-3.
[5]袁赛柏,金胜,朱天林.认知雷达技术与发展[J].现代雷达,2016,38(1):1-3.
[6]江涛,孙俊.量子雷达探测目标的基本原理与进展[J].中国电子科学研究院学报,2014,9(1):10-16.
[7]叶利华,王磊,赵利平.低小慢无人机降落野外场景识别方法[J].计算机应用,2017,37(7):2008-2013.
[8]戚春,张聃,蔡云泽.基于交互式多模型的红外/雷达低小慢目标跟踪算法研究[J].上海航天,2012,29(6):37-41.
[9]陈小龙,关键,何友,等.高分辨稀疏表示及其在雷达动目标检测中的应用[J].雷达学报,2017,6(3):239-251.
[10]刘凯,林基明,郑霖,等.基于深度自编码网络的慢速移动目标检测[J].计算机工程,2018,44(2):129-134.
[11] Haykin S,Bhattacharya T K.Modular learning strategy for signal detection in a nonstationary environment[J].IEEE Transactions on Signal Processing,1997,45(6):1619-1637.
[12] Zhai S,Jiang T.Target detection and classification by measuring and processing bistatic UWB radar signal[J].Measurement,2014,47(1):547-557.
[13] He X,Jiang T.Target identification in foliage environment using UWB radar with hybrid wavelet-ICA and SVM method[J].Physical Communication,2014,13:197-204.
[14] Zhai S,Jiang T.Application of Ultra-Wide band radar for sense-through-foliage target detection and recognition[C]//Proceedings of the Third International Conference on Communications,Signal Processing and Systems,2015:479-487.
[15]焦李成,杨淑媛,刘芳,等.神经网络七十年:回顾与展望[J].计算机学报,2016,39(8):1697-1716.
[16]邓力,俞栋.深度学习方法及应用[M].谢磊,译.北京:机械工业出版社,2016:39-46.
[17]吴岸成.神经网络与深度学习[M].北京:电子工业出版社,2016:66-86.
[18]李玉鑑,张婷.深度学习导论及案例分析[M].北京:机械工业出版社,2016:48-56.
[19]颜丹,蒋加伏.基于栈式去噪自动编码器的边际Fisher分析算法[J].计算机工程与应用,2017,53(5):134-139.
[20]赵瑞娟,官金安,谢国栋.稀疏降噪自编码器在IR-BCI的应用研究[J].计算机工程与应用,2017,53(11):167-171.
[21]阮晓刚.神经计算科学-在细胞的水平上模拟脑功能[M].北京:国防工业出版社,2006:356-399.
[22] Hou X,Wan S S,Liu Rui.Quantum gate circuit neural network optimization algorithm based on performance function[C]//Proceedings of 2018 International Conference on Communication,Network and Artificial Intelligence(CNAI 2018),Beijing,China,2018:321-326.
[23]李楠,侯旋.量子自适应前向对传算法研究[J].电子与信息学报,2013,35(11):2778-2783.
[24]田玉芳,尹志盈,姬光荣,等.基于SVM的海面弱目标检测[J].中国海洋大学学报,2013,43(7):104-109.