回声状态网络混沌跳频码预测方法
发布时间:2018-12-12 23:54
【摘要】:针对现有跳频码预测方法存在的缺乏记忆能力、运算量大、训练过程复杂等问题,提出了基于回声状态网络的混沌跳频码预测方法。该方法在跳频码相空间重构的基础上,利用回声状态网络内部动态储备池的循环记忆功能,通过调整各权值矩阵的数值大小达到记忆数据的目的,解决了跳频码预测的问题。仿真实验表明该方法对Logistic-Kent映射、Lorenz系统和Mackey-Glass系统三种混沌跳频码都有较好的预测效果,并与其他方法的实验结果进行了比较,证明回声状态网络在混沌跳频码预测方面的可行性及优越性。
[Abstract]:In order to solve the problems existing in the existing frequency hopping code prediction methods, such as lack of memory ability, large amount of computation and complex training process, a chaotic frequency hopping code prediction method based on echo state network is proposed. Based on the phase space reconstruction of frequency hopping codes and the cyclic memory function of the dynamic storage pool in the echo state network, the method achieves the purpose of memorizing the data by adjusting the numerical size of each weight matrix, and solves the problem of frequency hopping code prediction. The simulation results show that the proposed method can predict the chaotic frequency hopping codes of Logistic-Kent map, Lorenz system and Mackey-Glass system, and is compared with the experimental results of other methods. The feasibility and superiority of echo state network in chaotic frequency hopping code prediction are proved.
【作者单位】: 解放军电子工程学院;安徽省电子制约技术重点实验室;
【分类号】:TN914.41
[Abstract]:In order to solve the problems existing in the existing frequency hopping code prediction methods, such as lack of memory ability, large amount of computation and complex training process, a chaotic frequency hopping code prediction method based on echo state network is proposed. Based on the phase space reconstruction of frequency hopping codes and the cyclic memory function of the dynamic storage pool in the echo state network, the method achieves the purpose of memorizing the data by adjusting the numerical size of each weight matrix, and solves the problem of frequency hopping code prediction. The simulation results show that the proposed method can predict the chaotic frequency hopping codes of Logistic-Kent map, Lorenz system and Mackey-Glass system, and is compared with the experimental results of other methods. The feasibility and superiority of echo state network in chaotic frequency hopping code prediction are proved.
【作者单位】: 解放军电子工程学院;安徽省电子制约技术重点实验室;
【分类号】:TN914.41
【相似文献】
相关期刊论文 前10条
1 甘良才,包永强,杨桂文,茹国宝;一类新的宽间隔跳频码的实现[J];武汉大学学报(自然科学版);1999年01期
2 甘良才,包永强;一类新的宽间隔跳频码的实现[J];电波科学学报;1999年02期
3 李赞,金力军,常义林;一种快速的跳频码序列产生方法[J];西安电子科技大学学报;2001年02期
4 张申如;跳频码序列动态特性和抗预测设计要求[J];应用科学学报;2004年01期
5 何维苗,冯冈;构造宽间隔跳频码序列的两种算法之比较[J];解放军理工大学学报(自然科学版);2004年04期
6 蔡善法;刘刚;汤宏伟;;长周期跳频码序列性能统计检测[J];电讯技术;2011年05期
7 范越非;卢宏涛;;跳频码序列混沌预测的性能研究[J];计算机应用与软件;2011年08期
8 李斌;林建宗;;多址通信的跳频码序列的研究[J];军事通信技术;1990年02期
9 扈新林,姚富强;一种实用的跳频码序列产生方法[J];军事通信技术;1995年01期
10 程郁凡,洪福明;跳频码序列复杂度与随机性分析[J];电子科技大学学报;1996年S3期
相关会议论文 前1条
1 项飞;甘良才;;基于G函数的跳频码发生器的实现[A];武汉市首届学术年会通信学会2004年学术年会论文集[C];2004年
相关博士学位论文 前1条
1 王q,
本文编号:2375475
本文链接:https://www.wllwen.com/kejilunwen/wltx/2375475.html