基于时延神经网络模型的舰船辐射噪声目标识别*

水声目标被动识别是水声信号处理领域的研究热点之一。海洋环境中存在的不规则噪声干扰,使得基于传统方法的水声目标被动识别技术在实际的应用场景中效果不佳。本文采用一种基于时延网络(Time Delay Neural Network,TDNN)模型的舰船辐射噪声目标识别方法,该方法利用目标的短时平稳特性和长时关联特性对目标的声纹特征进行建模,使用梅尔谱图提取目标信号的初级特征,再通过融合注意力机制和时延神经网络的深度学习模型实现高级特性提取,最后再利用余弦相似度实现不同目标的类别划分。该方法在ShipsEar数据集和自行采集的数据进行测试验证,目标识别准确率分别达到79.2%和73.9%,可证明本文方法的有效性。     

Hyperchaos of four state autonomous system with three positive Lyapunov exponents

This Letter gives the results of numerical simulations of Quantum Cellular Neural Network (Quantum-CNN) autonomous system with four state variables. Three positive Lyapunov exponents confirm hyperchaotic nature of its dynamics.     

Hyperchaos of two coupled Bose--Einstein condensates with a three-body interaction

We investigate the dynamics of two tunnel-coupled Bose--Einstein condensates (BECs) in a double-well potential. The effects of the three-body recombination loss and the feeding of the condensates from the thermal cloud are studied in the case of attractive interatomic interaction. An imaginary three-body interaction term is considered and a two-mode approximation is used to derive three coupled equations which describe the total atomic numbers of the two condensates, the relative population and relative phase respectively. Theoretical analyses and numerical calculations demonstrate the existence of chaotic and hyperchaotic behaviour by using a periodically time-varying scattering length.     

Spatiotemporal chaos and noise

Low-dimensional chaotic dynamical systems can exhibit many characteristic properties of stochastic systems, such as broad Fourier spectra. They are distinguishable from stochastic processes through finite values for their dimension, Lyapunov exponents, and Kolmogorov-Sinai entropy. We discuss how these characteristic observables are modified in spatiotemporal chaotic systems like. coupled map lattices. We analyze with the help of Lyapunov concepts how the stochastic limit is approached and how these properties can be observed directly through local dimension measurements from reconstructed time series. Finally, we discuss the interaction of spatiotemporal attractors with external noise and possible connections to problems of pattern selection and stability.     

舰船辐射噪声的分形布朗运动模型

通过对分形布朗运动的研究,建立了舰船辐射噪声的分形布朗运动模型。在此模型基础上,提取了舰船辐射噪声的分数维参数,并构造了基于分形布朗随机场的分形特征矢量,在水声目标识别系统中提供了一种新的特征分析手段、通过分类对比实验,表明分形特征矢量可以成为水声目标分类的重要依据,能够有效地补充其他特征提取方法的不足.     

船舶辐射噪声信号仿真及评价

船舶辐射噪声信号仿真应尽可能多地仿真特定船舶辐射噪声的信号特征,达到具体应用对信号仿真度的要求。对船舶辐射噪声仿真信号进行评价,是调整仿真参数和判定仿真信号是否达到要求的必要步骤。该文给出了一套对实测船舶辐射噪声信号进行重构仿真及评价的方法,仿真过程考虑了信号的听觉、连续谱、调制谱和线谱特征,评价方法对各特征进行指标提取和量化,并通过实测与仿真信号之间的相似度计算评价仿真信号的仿真度。对实测船舶辐射噪声信号进行重构仿真的试验结果显示,仿真信号具有较高的特征仿真度,与计算的仿真度评价数值一致。     

船舶水下辐射噪声信号理论模型及仿真

分析典型船舶水下辐射噪声实验数据,获取船舶水下辐射噪声功率谱特征,阐明了准周期随机声脉冲序列信号模型作为船舶水下辐射噪声模型的合理性和普适性,在此基础上,引入爆炸衰减型余弦脉冲信号作为该模型的基本组成单元,给予理论分析,据以数值构建船舶辐射噪声,并与海上试验获取的商船辐射噪声数据进行对比,结果明显一致,证明了该方法的有效性和实用性.     

A novel hyperchaos system only with one equilibrium

This Letter presents a new hyperchaotic system by introducing an additional state feedback into a three-dimensional quadratic chaotic system. The system only has one equilibrium, but it can evolve into periodic, quasi-periodic, chaotic and hyperchaotic dynamical behaviors. Basic bifurcation analysis of the new system is investigated by means of Lyapunov exponent spectrum and bifurcation diagrams. We find that the new hyperchaotic system possesses two big positive Lyapunov exponents within a large range of parameters. Therefore, the new hyperchaotic system may have good application prospects.     

激光大气闪烁的分形分析

 采用分形理论分析了激光大气闪烁的统计特征。研究结果表明：在弱起伏条件下，激光大气闪烁的分形维和奇异性随光强起伏的增强而增大，而其长期相关性则减小；不同Fresnel尺度下具有相同闪烁指数的激光大气闪烁的分形特征存在着明显的差别；在强起伏条件下，有限的数据中尚未发现分形维有饱和现象，因此可以用来描述激光大气闪烁的强度。     

基于听觉特性的舰船辐射噪声信号仿真

针对工程应用中对舰船辐射噪声信号仿真结果在听觉感受上模拟度欠佳的问题,建立了一种基于音色参量谱质心的辐射噪声仿真方法。采用基于实测谱的连续谱仿真模拟,并且通过整体对比实测信号的听觉特性来搜索符合特征样本音色的线谱族幅值参量,在一定程度上改善了仿真信号的听觉特征。通过仿真最优匹配度趋势来寻求最优搜索次数,用以缩短仿真时间,得到了符合工程应用的结果。