面向光声海底地形探测的声学信号建模与处理技术
发布时间：2018-12-09 11:59
【摘要】：海洋遥感是利用传感技术手段对海洋进行远距离非接触观测,以获取海洋景观和海洋要素的图像或数据资料,达到海洋环境监测、近海海洋勘察、海底地形探测、海洋动力现象观测等重要目的。目前广泛使用的海底地形测量系统有单波束回波测深仪、多波束回波测深仪以及机载激光测深仪。这些设备有各自的特点,也受制于各自的适用条件。 光声探测概念的提出旨在突破传统光学水深测量和传统声学水深测量的限制。激光作用于水体介质,在水体中激发声波,形成激光声源。利用基于水面声光耦合原理的光纤水听器作为海底回波的接收设备。将声纳的分辨率、距离优势与机载平台的速度和机动性结合起来,在若干应用场合可以作为传统地形测量的一个重要补充。 本论文针对光声海底地形探测中的声学信号建模与处理技术进行研究。光声海底地形探测使用了新型探测声源和新型信号接收方式,带来便捷的同时也带来了若干新的问题。本论文对激光声脉冲的特性及海底回波模型,海底回波达到时间估计和动态水面环境下声信号接收干扰消除等问题进行研究。 主动声纳探测体系中,发射信号的特性是主动声纳探测研究中至关重要的一部分。论文从实验采集到的激光声脉冲数据出发,对激光声脉冲信号的特性进行研究。在此基础上,结合激光声脉冲的特性、传播衰减特性、海底散射等因素,建立了一种激光声脉冲的海底回波模型,并通过仿真实验验证了模型的合理性。 光声海底地形测量问题属于含有未知参量的信号检测问题,处理的信号为海底的后向散射回波信号。论文使用特征函数相关检测的方法估计回波到达时间,此方法中特征函数为仿真产生的海底后向散射信号的包络序列,特征函数相关检测法利用了回波信号的幅度以及包络的综合特征,对海底回波的时间展宽和海底的随机起伏具有较好的拟合能力。通过仿真实验验证了特征函数相关检测算法的有效性和抗干扰能力。 论文最后从信号接收的角度,对基于水面声光耦合原理探测水下声信号的实验进行了重点研究,分析波浪可能对探测造成的干扰。在贝叶斯信号处理框架下,设计了一种基于时变AR模型的卡尔曼滤波器跟踪并抑制波浪干扰信号。通过对实验数据的滤波处理,验证了所设计滤波器较之于传统带通滤波方法的性能提升。
[Abstract]:Ocean remote sensing is the use of sensing technology to conduct remote non-contact observation of the ocean, in order to obtain images or data of ocean landscape and ocean elements, to achieve marine environmental monitoring, offshore marine exploration, seabed topography detection, Ocean dynamic phenomena observation and other important purposes. At present, there are single beam echo sounder, multi-beam echo sounder and airborne laser sounder. These equipments have their own characteristics, but also subject to their own applicable conditions. The concept of photoacoustic detection is proposed to break through the limitations of traditional optical bathymetry and acoustic bathymetry. The laser acting on the water medium excites the sound wave in the water body and forms the laser sound source. A fiber-optic hydrophone based on acousto-optic coupling principle is used to receive underwater echo. Combining sonar resolution and range advantage with the speed and maneuverability of airborne platform, it can be used as an important supplement to traditional topographic survey in several applications. In this paper, acoustic signal modeling and processing techniques in photoacoustic sea bottom topography detection are studied. The photoacoustic sea bottom topography detection uses the new detection sound source and the new signal reception method, which not only brings convenience but also brings some new problems. In this paper, the characteristics of laser acoustic pulse, the model of underwater echo, the time estimation of underwater echo and the elimination of acoustic signal receiving interference in dynamic water environment are studied. In the system of active sonar detection, the characteristics of transmitted signals are very important in the research of active sonar detection. Based on the experimental data of laser acoustic pulse, the characteristics of laser acoustic pulse signal are studied in this paper. On this basis, a submarine echo model of laser acoustic pulse is established, which is based on the characteristics of laser acoustic pulse, propagation attenuation characteristic, undersea scattering and so on, and the rationality of the model is verified by simulation experiments. The problem of photoacoustic sea bottom topography measurement belongs to the signal detection problem with unknown parameters, and the signal processed is the backscattering echo signal of the sea floor. In this paper, the method of feature function correlation detection is used to estimate the time of arrival of echo. In this method, the characteristic function is the envelope sequence of the backscattering signal generated by simulation. The characteristic function correlation method makes use of the amplitude of echo signal and the comprehensive feature of envelope, and has a good fitting ability for the time broadening of sea bottom echo and the random fluctuation of seabed. The effectiveness and anti-interference ability of the feature function correlation detection algorithm are verified by simulation experiments. In the end of this paper, the underwater acoustic signal detection based on the principle of acousto-optic coupling on water surface is studied in detail from the point of view of signal receiving, and the possible interference caused by wave is analyzed. In the framework of Bayesian signal processing, a Kalman filter based on time-varying AR model is designed to track and suppress wave interference signals. The performance of the designed filter is improved compared with the traditional bandpass filter by filtering the experimental data.
【学位授予单位】：浙江大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN912.3
[Abstract]:Ocean remote sensing is the use of sensing technology to conduct remote non-contact observation of the ocean, in order to obtain images or data of ocean landscape and ocean elements, to achieve marine environmental monitoring, offshore marine exploration, seabed topography detection, Ocean dynamic phenomena observation and other important purposes. At present, there are single beam echo sounder, multi-beam echo sounder and airborne laser sounder. These equipments have their own characteristics, but also subject to their own applicable conditions. The concept of photoacoustic detection is proposed to break through the limitations of traditional optical bathymetry and acoustic bathymetry. The laser acting on the water medium excites the sound wave in the water body and forms the laser sound source. A fiber-optic hydrophone based on acousto-optic coupling principle is used to receive underwater echo. Combining sonar resolution and range advantage with the speed and maneuverability of airborne platform, it can be used as an important supplement to traditional topographic survey in several applications. In this paper, acoustic signal modeling and processing techniques in photoacoustic sea bottom topography detection are studied. The photoacoustic sea bottom topography detection uses the new detection sound source and the new signal reception method, which not only brings convenience but also brings some new problems. In this paper, the characteristics of laser acoustic pulse, the model of underwater echo, the time estimation of underwater echo and the elimination of acoustic signal receiving interference in dynamic water environment are studied. In the system of active sonar detection, the characteristics of transmitted signals are very important in the research of active sonar detection. Based on the experimental data of laser acoustic pulse, the characteristics of laser acoustic pulse signal are studied in this paper. On this basis, a submarine echo model of laser acoustic pulse is established, which is based on the characteristics of laser acoustic pulse, propagation attenuation characteristic, undersea scattering and so on, and the rationality of the model is verified by simulation experiments. The problem of photoacoustic sea bottom topography measurement belongs to the signal detection problem with unknown parameters, and the signal processed is the backscattering echo signal of the sea floor. In this paper, the method of feature function correlation detection is used to estimate the time of arrival of echo. In this method, the characteristic function is the envelope sequence of the backscattering signal generated by simulation. The characteristic function correlation method makes use of the amplitude of echo signal and the comprehensive feature of envelope, and has a good fitting ability for the time broadening of sea bottom echo and the random fluctuation of seabed. The effectiveness and anti-interference ability of the feature function correlation detection algorithm are verified by simulation experiments. In the end of this paper, the underwater acoustic signal detection based on the principle of acousto-optic coupling on water surface is studied in detail from the point of view of signal receiving, and the possible interference caused by wave is analyzed. In the framework of Bayesian signal processing, a Kalman filter based on time-varying AR model is designed to track and suppress wave interference signals. The performance of the designed filter is improved compared with the traditional bandpass filter by filtering the experimental data.
【学位授予单位】：浙江大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN912.3
【参考文献】
相关期刊论文 前7条
1 郭熙业;苏绍t，
本文编号：2369334
本文链接：https://www.wllwen.com/kejilunwen/wltx/2369334.html
