基于径向基神经网络的水声微弱信号检测

孟宪超; 刘纯武; 刘 晔

doi:10.11993/j.issn.1673-1948.2011.05.006

基于径向基神经网络的水声微弱信号检测

doi: 10.11993/j.issn.1673-1948.2011.05.006

国防科学技术大学机电工程与自动化学院, 湖南长沙, 410073

详细信息

作者简介:
孟宪超(1983-), 男, 在读硕士, 主要研究方向为数字化测试技术.

中图分类号: TJ630; TN911.7
计量
- 文章访问数: 1510
- HTML全文浏览量: 5
- PDF下载量: 529
- 被引次数: 0
出版历程
- 收稿日期: 2011-04-06
- 修回日期: 2011-05-14
- 刊出日期: 2011-10-31

Detection of Underwater Weak Acoustic Signal Based on Radial Basis Function Neural Network

College of Mechatronics Engineering and Automation, National University of Defense Technology, Changsha 410073, China

摘要

摘要: 为了克服传统随机信号处理手段检测水声微弱信号的缺陷, 提高水声微弱信号检测效率, 详细分析了海洋混响噪声的混沌特性, 将混沌理论中相空间重构理论与径向基神经网络相结合建立了预测模型, 从模型预测值与实际观测值的绝对误差中检测出微弱谐波信号频谱值, 通过解算获取回波信号包含目标信息, 从而提出了一种海洋混响混沌序列预测方法。相比传统随机信号处理手段检测水声微弱信号的方法, 本文建立的预测模型克服了将背景噪声简化作白噪声的低信噪比检测缺陷, 检测结果准确, 检测效率高, 适用于水声微弱信号检测。
- 信号检测 /
- 混响 /
- 混沌 /
- 相空间重构 /
- 径向基神经网络
Abstract: To enhance the detection efficiency of underwater weak acoustic signal, we analyze the chaotic characteristic of ocean reverberation, establish a prediction model by combining the theory of phase space reconstruction with the radial basis function neural network (RBF NN), and detect the frequency spectrum of the weak harmonic signal from the absolute error of the predicted value and the observed value. From the spectrum we calculate the target information, thus present a chaotic serial prediction method of ocean reverberation. Compared with the traditional random signal processing method, our prediction model overcomes the detection shortcoming of detection with low signal to noise ratio(SNR) due to simplifying background noise to white noise, and can obtain detection result accurately and efficiently. The proposed model can be suitable for weak signal detection in underwater acoustic.
- signals detection /
- reverberation /
- chaos /
- phase space reconstruction /
- radial basis function neural network(RBF NN)

HTML全文

参考文献(9)

[1]	Haykin S，Li X B. Detection of Signals in Chaos [J]. Processing of IEEE, 1995, 83(1): 83-94.
[2]	Simon H. Using Neural Networks to Dynamically Model Chaotic Events Such as Sea Clutter[J]. IEEE Sinal Proce- ssing Magazine Networks, 1998, 15(3): 66-81.
[3]	Takens F. On the Numerical Determination of the Di- mension of an Attractor[A]// Dynamical Systems and Turbulence[Z].Berlin:Springer-Verlag,1981:366-381.
[4]	Abarbanel H D I, Brown R, Sidorowich J J, et al. The Analysis of Observed Chaotic Data in Physical systems [J].Reviews of Modern Physics, 1993, 65(4): 1331-1339.
[5]	李玉霞, 吴百海, 邢志鹏. 单变量时间序列相空间重构及应用研究[J]. 组合机床与自动化加工技术, 2004, (2): 51-52. Li Yu-xia, Wu Bai-hai, Xing Zhi-peng. Phase Space Reconstruction of Single Variable Time Series and its Application[J]. Modular Machine Tool & Automatic Manufacturing Technique, 2004, (2): 51-52.
[6]	Cao L Y. Practical Method for Determining the Minimum Embedding Dimension of a Scalar Time Series[J]. Phy- sica D, 1997, 110(5): 43-50.
[7]	李彦彬, 尤凤, 黄强, 等. 基于归一化RBF网络的黄河上游径流预测[J]. 人民黄河, 2008,30 (6):36-38.
[8]	赵建利, 刘教民, 冯卫强. 基于小波包能谱熵与自组织RBF神经网络的低压断路器机械故障诊断[J], 低压电器, 2010, (4): 1-5. Zhao Jian-li, Liu Jiao-min, Feng Wei-qiang. Mechanical Fault Diagnosis for Low Voltage Circuit Breaker Based on Energy Spectrum Entropy of Wavelet Packet and Self-Organized RBF Neural Networks[J]. Genceral Low Voltage Apparatus, 2010, (4): 1-5.
[9]	周靖.基于混沌预测的水声微弱信号检测技术研究[D].长沙: 国防科技大学, 2009. Zhou Jing. Weak Signal Detection on Under Water Acoustic Based on the Prediction of Chaos[D]. Changsha: National University of Defense Technology, 2009.