带乘性噪声模型的水声信道与字符估计算法研究

发布时间：2018-08-30 18:07

【摘要】：带乘性噪声系统是传统的卡尔曼滤波线性系统的推广，可以解决复杂通信环境下的信号自适应估计问题。海洋环境由于存在很多随机不均匀性，如海面具有随机时变、空变的波浪，海水中有冷热不均的随机水团，海底的地貌及声学特性也是随机不均匀的，这些不均匀性都会对声信号的传播造成影响。因此，可以说作为水声信道的海洋介质是一个高噪声、强混响、信道带宽窄、多途效应的干扰严重的信道，又是一个时变、空变、频变的信道。其主要特征表现为：传播损失、多径效应、频散效应。本文基于水声信道的特性并结合乘性噪声的作用机理，用带乘性噪声系统模型来刻画随机时变的水声信道。针对单发多收及多发多收水声通信系统，，主要完成了以下工作：一、针对SIMO水声通信系统，建立了复杂多通道带乘性噪声系统模型，该模型刻画了一种更复杂的通道特性，并且更具有一般性。由于水声信道存在着明显的衰落特性及随机时变因素，因此，用带乘性噪声系统模型描述水声信道具有一定的合理性。并且在该模型的基础上，基于正交投影定理和线性最小方差准则，给出了状态最优滤波算法及固定域平滑算法。二、在SIMO系统的基础上，将模型推广到MIMO水声通信系统，建立了带乘性噪声系统的状态空间模型，并且推导了基于线性最小方差准则的状态最优滤波算法。三、对SIMO水声通信系统的最优滤波算法、固定域平滑算法及MIMO系统的最优滤波算法进行了计算机仿真，验证了方法的有效性。并且本文还采用多径瑞利衰落信道得到的第三方接收数据，将SIMO系统的最优滤波算法与相关文献中的自适应均衡算法进行仿真对比，验证了本文算法的有效性。
[Abstract]:The system with multiplicative noise is a generalization of the traditional Kalman filter linear system, which can solve the problem of signal adaptive estimation in complex communication environment. Due to the existence of many random inhomogeneity in marine environment, such as random time-varying waves, spatially varying waves, random water masses with uneven heat and cold in sea water, the geomorphological and acoustic characteristics of the sea floor are also random and uneven. These inhomogeneity will affect the sound signal propagation. Therefore, it can be said that the oceanic medium as a underwater acoustic channel is a channel with high noise, strong reverberation, narrow channel band and serious interference of multipath effect, and is also a time-varying, spatially variable and frequency-varying channel. Its main characteristics are: propagation loss, multipath effect and dispersion effect. Based on the characteristics of underwater acoustic channel and the action mechanism of multiplicative noise, a stochastic time-varying underwater acoustic channel is described by using a system model with multiplicative noise. The main work of this paper is as follows: firstly, for SIMO underwater acoustic communication system, a complex multi-channel model with multiplicative noise is established. This model characterizes a more complex channel characteristic and is more general. Because of the obvious fading characteristics and random time-varying factors in underwater acoustic channel, it is reasonable to describe the underwater acoustic channel with multiplicative noise system model. On the basis of the model, the state optimal filtering algorithm and the fixed domain smoothing algorithm are presented based on the orthogonal projection theorem and the linear minimum variance criterion. Secondly, based on the SIMO system, the model is extended to the MIMO underwater acoustic communication system, the state space model of the system with multiplicative noise is established, and the state optimal filtering algorithm based on the linear minimum variance criterion is derived. Thirdly, the computer simulation of the optimal filtering algorithm for SIMO underwater acoustic communication system, the fixed-domain smoothing algorithm and the optimal filtering algorithm for MIMO system are carried out, and the validity of the method is verified. Moreover, the third party received data from multipath Rayleigh fading channel is used to simulate and compare the optimal filtering algorithm of SIMO system with the adaptive equalization algorithm in related literatures, which verifies the effectiveness of the proposed algorithm.
【学位授予单位】：中国海洋大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TB56

【参考文献】