单通道星载AIS信号分离与检测研究

发布时间：2018-02-24 21:53

本文关键词： 船舶自动识别系统单通道分离改进EMD算法冗余小波变换信道跟踪　出处：《天津理工大学》2017年硕士论文　论文类型：学位论文

【摘要】：船载自动识别系统(Automatic Identification System,AIS)作为一种新型的数字导航系统,不仅在防避碰和自动识别方面取得了成功,而且在加强海上生命安全、提高航行的效率、以及保护海洋环境等方面有着十分重要的作用。针对星载AIS系统中存在的信噪比较低、多普勒频移较大、信道时变较快和多用户干扰严重等问题,本文主要工作如下:(1)提出了一种基于改进经验模态分解(Empirical Mode Decomposition,EMD)的单通道信号分离算法。该算法首先利用极值点对称延拓算法对接收信号进行延拓,该过程改善了传统EMD算法中的端点效应。然后采用传统EMD算法将观测信号分解成一系列频率不同的本征模态函数(Intrinsic mode component,IMF);然后根据KL散度(Kullback-Leible divergence)的值,提取IMF分量中有用部分;最后利用独立分量分析(Independent component analysis,ICA)算法实现部分AIS信号分离。仿真结果表明,该算法可以实现低信噪比下多路信号混合时的单通道分离,并且混合的源信号之间的相对频偏越大,分离效果越好。(2)提出了一种基于冗余小波变换(Redundant discrete wavelet transform,RDWT)的单通道星载AIS信号分离算法。该方法首先利用RDWT算法将观测信号分解成频率不同的近似信号和细节信号,该过程完成由单通道分离向多通道分离的转化;然后利用ICA算法实现AIS源信号分离。仿真结果表明,该算法实现了多路混合信号的分离,并且分离效果几乎不受时延影响。(3)提出了一种信道和相位联合跟踪下的信号检测算法。该算法将信道和相位实时跟踪结合起来,实现了参数与序列的联合估计。仿真结果表明,该算法性能优于差分检测、维特比译码和基于逐幸存路径处理(per-survivor processing,PSP)的信号检测算法,并且抗频偏能力较强。
[Abstract]:As a new type of digital navigation system, automatic Identification system (AIS) has not only been successful in preventing collision and automatic recognition, but also has enhanced the safety of life and the efficiency of navigation. In view of the problems of low SNR, large Doppler frequency shift, fast time-varying channel and serious multi-user interference in spaceborne AIS system, it is very important to protect the marine environment. The main work of this paper is as follows: (1) A single channel signal separation algorithm based on improved empirical Mode decomposition (EMD) is proposed. This process improves the endpoint effect in the traditional EMD algorithm, and then decomposes the observed signal into a series of intrinsic mode component functions with different frequencies by using the traditional EMD algorithm, and then extracts the useful parts of the IMF component according to the KL divergence and Kullback-Leible divergence. Finally, the independent component analysis (ICA) algorithm is used to realize partial AIS signal separation. The simulation results show that the algorithm can realize the single-channel separation of multi-channel signals mixed with low SNR, and the greater the relative frequency deviation between the mixed source signals is. The better the separation effect is, a single channel space-borne AIS signal separation algorithm based on redundant wavelet transform (redundant discrete wavelet transform RDWT) is proposed. Firstly, the observed signal is decomposed into approximate signal and detail signal with different frequency by using RDWT algorithm. The process completes the transformation from single-channel separation to multi-channel separation, and then realizes the AIS source signal separation using ICA algorithm. The simulation results show that the algorithm realizes the separation of multi-channel mixed signals. A signal detection algorithm based on channel and phase joint tracking is proposed, which combines channel and phase tracking in real time, and realizes the joint estimation of parameters and sequences. The performance of this algorithm is superior to that of differential detection, Viterbi decoding and signal detection algorithm based on per-survivor processing PSPs.
【学位授予单位】：天津理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：U675.7;TN911.7

【相似文献】