基于压缩感知的水下目标回波信号处理技术研究

发布时间：2018-06-30 21:38

本文选题：水下回波信号 + 先验信息　；参考：《杭州电子科技大学》2017年硕士论文

【摘要】：水下目标回波信号是主动声呐探测和识别的基础。水下环境复杂多变,存在各种噪声和干扰,尤其随着隐身技术的发展和各种小目标的出现,使得目标回波信号越来越弱,因此,弱回波信号检测和处理问题是水声信号处理领域研究的热点和难点问题。另外,为了提高系统的探测精度,增加抗干扰能力,提高目标检测概率等,需要不断增加信号带宽。然而带宽的增加使得数据量急剧增大,给信号的采集、存储、传输和处理等带来极大的负担。在这种背景下,本文对压缩感知在水下回波信号处理中的应用展开研究,拟解决目前水下回波信号处理中存在的弱信号检测和采样数据量过大等难题,主要内容有:首先,阐述了本论文的研究背景和意义,并对压缩感知理论以及其在水下信号处理中的应用进行了国内外研究现状综述,给出了本论文的研究思路和结构安排。其次,以压缩感知理论为基础,重点研究了几种常见的稀疏基、测量矩阵、匹配追踪的重构算法以及相应的改进算法,给出了信号重构效果测衡量标准,并从水下回波信号基本理论出发,采用亮点模型实现了水下回波信号的仿真。在此基础上,通过仿真实验,比较了不同稀疏基、不同测量矩阵及不同重构算法对水下回波信号处理结果的影响,提出基于离散余弦稀疏基、高斯随机矩阵和分段正交匹配追踪算法的压缩感知处理框架。再次,充分分析水下回波信号的形成原理和结构特性,研究回波信号与入射信号的关系、回波信号的分块特性等,将入射信号和块稀疏特性作为先验信息,融入稀疏分解和重构过程,提出融入先验信息的压缩感知处理方法。并进一步应用到水下回波信号处理中,采用信噪比的提高量、匹配度、相对误差等指标衡量了该方法的处理效果。仿真实验结果充分显示了该方法在提高信噪比和减少数据量方面的优势。最后,总结了本文的主要工作和创新,并对下一步应展开的研究进行了展望。
[Abstract]:Underwater target echo signal is the basis of active sonar detection and recognition. The underwater environment is complex and changeable, there are various noises and disturbances, especially with the development of stealth technology and the appearance of various small targets, the echo signal of the target becomes weaker and weaker. Weak echo signal detection and processing is a hot and difficult problem in the field of underwater acoustic signal processing. In addition, in order to improve the detection accuracy of the system, increase the ability of anti-jamming and improve the probability of target detection, it is necessary to continuously increase the signal bandwidth. However, the increase of bandwidth makes the amount of data increase rapidly, which brings great burden to signal acquisition, storage, transmission and processing. In this context, the application of compression sensing in underwater echo signal processing is studied in this paper, and the problems of weak signal detection and excessive sampling data in underwater echo signal processing are solved. The main contents are as follows: first of all, The research background and significance of this paper are expounded, and the theory of compression sensing and its application in underwater signal processing are summarized at home and abroad, and the research ideas and structure of this paper are given. Secondly, based on the theory of compression perception, several common sparse bases, measurement matrices, matching tracking reconstruction algorithms and corresponding improved algorithms are studied, and the measurement criteria for signal reconstruction effect are given. Based on the basic theory of underwater echo signal, the simulation of underwater echo signal is realized by using bright spot model. On this basis, the effects of different sparse bases, different measurement matrices and different reconstruction algorithms on the underwater echo signal processing results are compared through simulation experiments, and a discrete cosine sparse basis is proposed. Gao Si random matrix and piecewise orthogonal matching tracking algorithm compression perception processing framework. Thirdly, the formation principle and structure characteristic of underwater echo signal are analyzed fully, the relation between echo signal and incident signal, the block characteristic of echo signal are studied, and the incident signal and block sparse characteristic are taken as prior information. In the process of sparse decomposition and reconstruction, this paper proposes a method of processing compression perception with prior information. It is further applied to underwater echo signal processing. The improvement of signal-to-noise ratio, matching degree and relative error are used to evaluate the processing effect of the method. The simulation results show the advantages of this method in improving signal-to-noise ratio and reducing the amount of data. Finally, the main work and innovation of this paper are summarized, and the future research is prospected.
【学位授予单位】：杭州电子科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TB56;TN911.7

【参考文献】