QQ客服 官方微博 反馈留言

题名	浅海环境下主动声纳抗混响、抗多途研究与系统实现
作者	李宇
学位类别	博士
答辩日期	2005
授予单位	中国科学院声学研究所
授予地点	中国科学院声学研究所
关键词	浅海 混响抑制 多途宽容处理 波形设计 主动声纳 DSP系统
中文摘要	随着海军战略从深海转向浅海以及安静型潜艇的出现，在复杂浅海环境下的主动声纳信号处理及应用技术，越来越引起人们的重视。在这个背景下，本论文系统地研究了浅海环境下的主动声纳，特别是低频主动声纳的混响抑制处理和多途宽容处理两大问题。同时，结合近凡年的项目经验，研究了基于DSP的声纳信号处理软件系统的设计与实现问题。本论文的基本内容和主要创新点有如下几点：1）结合大量海试数据，通过主动声纳波形设计理论分析，揭示了现代主动声纳波形设计的趋势是使用复杂多样的组合信号。提出了具有良好混响抑制性能的新型梳状波形。引入并分析了阶梯调频LFM脉冲串在混响环境下的距离多普勒分辨性能。2）研究了基于混响统计模型的混响抑制方法。通过海试数据，分析了基于高阶累积量的从预白处理方法，在非高斯混响背景下的混响抑制性能。同时，利用海试数据，证实了大孔径拖曳线列阵接收混响的非瑞利性，并分析了Pagetest等检测器在非瑞利混响背景下的检测性能。3）研究了基于特征检测的混响抑制方法。引入WVD一Radon变换域检测方法，并分析了它在真实混响背景下的微弱目标检测能力。4）研究了空时处理抑制混响的方法。分析了SPOC等三种先时后空的自适应处理算法在实际混响背景下混响抑制效果。提出了将Pcl方法与快速的奇异值分析的算法结合的基于Schur-typePCI的棍响抑制快速算法，并通过实际海试数据的分析说明了这种算法在大大的减少计算量的同时，又不会降低PCI的抑制性能，为浅海环境下强混响的抑制提供了一个实际可应用的方法。5）研究了多途起伏统计模型下的宽容处理方法。引入了TFRC算法，并通过海试数据分析，证实了其在多途起伏下目标检测的宽容检测性能。提出了基于复合声纳波形设计的多子带多副本宽容处理方法，并利用实际环境加仿真复杂信号数据，验证了这种算法对低信噪比信号的检钡”能力。6）研究了基于物理模型的宽容处理方法。利用海试数据，比较了不同物理模型下模基匹配滤波的宽容检测性能。7）结合近几年来的项目经验，研究了基于DsP的声纳信号处理软件系统的设计思想，探讨了设计实时、可靠声纳信号处理系统所必须考虑的有关软件体系结构、模块化设计，以及程序设计等问题，为声纳软件系统的实现提供了一些切实可行的方法和思路。8）在算法研究基础上提出了一些新颖的系统实现方案，其中一些方案己经得到实际海试的验证。
英文摘要	In recent years, along with the change in strategy of the Navy for underwater defense from the deep sea to the shallow water and emergence of the "quiet" submarine, the active sonar and its relative signal processing technique became remarkable. Under these backgrounds, this dissertation presents some novel signal processing methods and analyzes their performance for active sonar in the shallow-water reverberation or multi-path environment in the first part. Meanwhile, this dissertation discusses the design and implementation of DSP-based sonar software system in the second part. The creative and innovative works in this dissertation are listed as follow: 1. This dissertation first focused on the waveform design for active sonar. Re- vealed the trend of waveform design for active sonar in shallow water, that is, one has to use the pulse train of time frequency coded combined signal to suppress the reverber- ation or improve the range resolution, especially for low frequency active sonar. And designed two novel active sonar waveforms: the new Comb waveform and Stepped Frequency Modulation Pulse Train of LFM which has good range-doppler resolution in reverberation environment. 2. This dissertation analyzed different reverberation suppression algorithms based on statistic characteristic of the reverberation in shallow water. First, an AR pre-whitening filter based on third-order statistic parameter is introduced, and it was proved that it's more effective than the other whitening or non-whitening algorithms in non-Gauss reverberation environment by analysis of the real sea trial data. Then, by real data the detection performance of four algorithms like Page Test and so on are com- pared in non-Rayleigh reverberation environment. 3. This dissertation discussed the feature-based detection algorithm for reverbera-tion suppression, and gave WVD-Radon transform and its analytic results for real weak target detection in the shall-water. 4. This dissertation paid attention to the space-time processing methods in active sonar for reverberation rejection. First, Tried to apply four temporal-before-spatial processing methods. The real sea trial data processing indicated that the weak signal detection capability can be improved in the reverberation environment through match-filter before DOA(Direct Of Arrive). Then, a quick subspace separation and estimation method named the Schur-type method was used to resolve the problem of reverberation suppression. Meanwhile, the real data analysis showed that two kinds of the Schur-type based PCI methods have similar effect on estimating and deleting the strong bottom reverberation to the SVD-based PCI method. But these Schur-type methods are more efficiently on computation and more easily to be implemented in real-time parallel DSP system. 5. This dissertation presented some novel robust adaptive statistic signal process- ing for the weak signal detection in the multi-path shallow water, first, this disserta- tion took time frequency replica correlation algorithm(TFRC), used in communication usually, to do sonar robust adaptive processing. And the real data processing results confirmed that the TFRC is more robust and effective than others. Then, this disser- tation originally presented the multi-subband and multi-replica algorithm based on the design of combined active sonar signal, and it showed us the performance improvement in multi-path environment by simulated target data in real shallow water. 6. This dissertation discussed the matched or mismatched performance of physics model based matched filter in real data by using different physical acoustic models. 7. This dissertation summarized the architecture design of DSP-based sonar soft- ware system according to the author's developing experiences on this work in recent years. 8. This dissertation gave three system implementation schemes based on above study, one of them has been used in the real sea experiment and its capability and reliability has been validated. Key Words: shallow water, reverberation suppression, robust adaptive processing in multi-path environment, waveform design, active sonar, DSP system.
语种	中文
公开日期	2011-05-07
页码	131
内容类型	学位论文
源URL	[http://159.226.59.140/handle/311008/924]
专题	声学研究所_声学所博硕士学位论文_1981-2009博硕士学位论文
推荐引用方式 GB/T 7714	李宇. 浅海环境下主动声纳抗混响、抗多途研究与系统实现[D]. 中国科学院声学研究所. 中国科学院声学研究所. 2005.

个性服务

查看访问统计

相关权益政策

暂无数据

收藏/分享

除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。