滤波多音调制下的水声信道估计

发布时间：2018-07-09 20:41

本文选题：水声通信 + 滤波多音调制　；参考：《浙江大学》2017年硕士论文

【摘要】：水声通信作为主要的水下远距离无线通信方式,在很多领域有着重要的应用。然而水声通信的研究面临着很多方面的困难,如:水声信号随着信号频率的增加而衰减;水声信号在传播过程中存在着时变的多径效应;水声传播速度较低等。本文针对水声通信面临的困难,结合多载波调制技术的特点,描述了一种在无线通信中应用比较广泛的基于滤波器组的多载波调制技术,即滤波多音调制技术(Filtered-Multitone Modulation,FMT),并且在该调制技术下,结合水声信道的特性进行水声信道的估计和均衡。本文首先从理论上推导了 FMT调制技术在水声通信中的可行性,通过仿真分析FMT调制技术的特点,其相对于正交频分复用调制技术(Orthogonal Frequency Division Multiplexing,OFDM)技术的特点是其载波之间不是严格的“正交性”,然而其子载波之间有着间隔,子载波之间的干扰几乎可以忽略,这样对于抗Doppler有着一定的优势。研究水声信道的特性是设计水声通信系统的基础。本文从水声信道的特性出发,介绍了水声信道中典型的时延Doppler双扩展特性和稀疏特性,并在此基础上对水声信道进行建模分析。通过将几条主要的到达径分离来近似表达水声信道,此外信道的幅度时变和时延时变通过多项式进行拟合的方法来进行表示。本文在此基础上对水声信道进行参数化建模,并将此与FMT调制技术进行结合,进行后续的信道估计。在得到水声信道的参数化模型之后,本文通过该模型下的输入输出关系,将高维的信道估计问题转化为有限维度的信道参数估计问题,并且采用压缩感知的算法进行简便求解。本文通过信道的幅度时变和时延时变的特点,提出一种分步估计的方法,该方法结合信号的前后导频来估计相应时刻的信道,然后结合信道建模特性拟合信道的相应参数来进行信道估计。通过仿真进行对上述方法的验证。水声通信中时延扩展引起严重的符号间干扰(Inter-SymbolInterference,ISI),特别是在FMT调制技术背景下,ISI是不可忽略的,是限制浅海水声通信性能的一个重要原因。针对这个问题,FMT调制需要在接收端进行相应的均衡处理。本文主要描述了在FMT调制技术下的基于递归最小二乘(Recursive Least Squares,RLS)算法的自适应判决反馈(Decision Feedback Equalization,DFE)均衡,并在此基础上结合分数间隔采样以及基于锁相环的Doppler补偿方法进行优化。此外,本文还提出了一种基于信道估计的均衡,根据理论推导FMT调制技术载波间干扰可忽略的特点,将每一个子载波上符号进行独立均衡。本文将以上两种方法进行了仿真分析,验证其性能的可靠性。最后论文通过海试数据进行了 FMT调制系统的可靠性分析。
[Abstract]:Underwater acoustic communication, as the main underwater long-range wireless communication, has important applications in many fields. However, the research of underwater acoustic communication is faced with many difficulties, such as the attenuation of underwater acoustic signal with the increase of signal frequency, the existence of time-varying multipath effect in the process of underwater acoustic signal propagation, the low speed of underwater acoustic propagation, and so on. In view of the difficulties faced by underwater acoustic communication and the characteristics of multicarrier modulation technology, this paper describes a multi-carrier modulation technology based on filter banks, which is widely used in wireless communication. Filtered-Multitone Modulation / FMT (Filtered-Multitone Modulation / FMT) technique is used to estimate and equalize underwater acoustic channels in combination with the characteristics of underwater acoustic channels. In this paper, the feasibility of FMT modulation technology in underwater acoustic communication is deduced theoretically, and the characteristics of FMT modulation technology are analyzed by simulation. Compared with orthogonal Frequency Division Multiplexing OFDM (OFDM), its characteristic is that its carrier is not strictly orthogonal. However, there is a gap between its subcarriers, and the interference between subcarriers can be almost ignored. In this way, it has a certain advantage in anti-Doppler. Studying the characteristics of underwater acoustic channel is the basis of designing underwater acoustic communication system. Based on the characteristics of underwater acoustic channel, this paper introduces the typical delay Doppler double spread and sparse characteristics in underwater acoustic channel. Based on this, the underwater acoustic channel is modeled and analyzed. Several major paths of arrival are separated to approximate the underwater acoustic channel, in addition, the channel amplitude time-varying and time-delay time-varying are represented by polynomial fitting method. Based on this, the parametric modeling of underwater acoustic channel is carried out, which is combined with FMT modulation technology to carry out subsequent channel estimation. After the parameterized model of underwater acoustic channel is obtained, the high-dimensional channel estimation problem is transformed into a finite dimensional channel parameter estimation problem through the input and output relationship under the model, and the compressed sensing algorithm is used to solve the problem. Based on the characteristics of time-varying amplitude and time-delay, a step by step method is proposed to estimate the channel at the corresponding time by combining the pilot frequency of the signal. Then the channel estimation is carried out by fitting the corresponding parameters of the channel with the channel modeling characteristics. The method is verified by simulation. Inter-Symbol-Interference- ISI (Inter-Symbol-Interference- ISI) caused by delay spread in underwater acoustic communication, especially in the context of FMT modulation technology, can not be ignored and is an important reason for limiting the performance of shallow water acoustic communication. To solve this problem, FMT modulation needs to be equalized at the receiver. This paper mainly describes the adaptive decision feedback (DFE) equalization based on the recursive least squares (RLS) algorithm under FMT modulation, and optimizes the algorithm based on the fractional interval sampling and Doppler compensation based on phase-locked loop (PLL). In addition, an equalization based on channel estimation is proposed. Based on the theoretical derivation of the negligible characteristics of inter-carrier interference (ISI) in FMT modulation technology, the symbols on each subcarrier are equalized independently. In this paper, the two methods are simulated and analyzed to verify the reliability of the performance. Finally, the reliability of FMT modulation system is analyzed by sea test data.
【学位授予单位】：浙江大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN929.3

【参考文献】