【摘要】：星載船舶自動識別系統(tǒng)（Automatic Identification System，AIS）是各國競相發(fā)展的船舶動態(tài)監(jiān)控系統(tǒng)，在軍事、經(jīng)濟和政治等領域均具有重要的意義。多用戶分離是星載AIS中的一個核心技術，直接影響到信號的接收效果和檢測性能，該課題的研究對其它衛(wèi)星通信系統(tǒng)同樣具有重要的價值。本文以星載AIS為應用背景，對時隙碰撞概率、信號分離算法和接收機的多用戶分離等方面進行研究。 建立了星載AIS接收信號時隙碰撞概率分析模型。該模型以泊松分布為基礎，為了使概率分析更加貼近實際情況，模型建立的過程中考慮到了：A、B類船臺的消息發(fā)送間隔是不同的；不同航行狀態(tài)的船舶消息發(fā)射間隔是不同的；信號混疊程度對信息完整性的影響是不同的。 提出了變步長的復值快速獨立分量分析（Fast Independent Component Analysis,FastICA）分離算法。以非高斯性為基礎建立了表示混合信號獨立性的目標函數(shù)，并采用具有較高收斂速度的牛頓迭代法對目標函數(shù)進行優(yōu)化。為使目標函數(shù)始終朝著最優(yōu)解的方向收斂，，迭代過程采用可變步長的牛頓迭代法，同時，為了保持算法收斂速度，基于Wolfe-Powell準則確定迭代步長的搜索范圍。仿真實驗對混合的信號進行了提取，并對不同的FastICA算法進行對比，證明基于Wolfe-Powell準則的FastICA，僅增加很少的計算量就解決了收斂性的問題。 針對AIS信號的空間傳輸特點，提出了一種衛(wèi)星接收機的多用戶分離方法。首先采用三交叉的二元線陣對不同極化方向的信號進行接收，基于極化分集減小時隙碰撞的概率；然后利用數(shù)字波束形成，將衛(wèi)星的視場劃分為多個帶狀區(qū)域，基于空間分集減小時隙碰撞的概率；最后結合變步長的復值FastICA分離算法對時隙碰撞的AIS信號進行逐個提取。仿真實驗模擬了接收機接收前端的主要信號處理流程，證明該方法是切實可行的。
[Abstract]:Spaceborne ship automatic Identification system (Automatic Identification System,AIS) is a dynamic ship monitoring system developed by many countries. It is of great significance in military, economic and political fields. Multi-user separation is a core technology in spaceborne AIS, which directly affects the signal receiving effect and detection performance. The research of this topic is also of great value to other satellite communication systems. Taking spaceborne AIS as the application background, this paper studies the slot collision probability, signal separation algorithm and multi-user separation of receiver. A time-slot collision probability analysis model for spaceborne AIS received signals is established. The model is based on Poisson distribution. In order to make the probability analysis more close to the actual situation, it is considered that the message sending interval of class B ship is different in the process of establishing the model. The effect of signal aliasing on information integrity is different. A variable step fast independent component analysis (Fast Independent Component Analysis,FastICA) separation algorithm is proposed. On the basis of non-Gao Si, the objective function representing the independence of mixed signals is established, and the Newton iteration method with high convergence rate is used to optimize the objective function. In order to keep the objective function converging towards the optimal solution, the variable step size Newton iteration method is used in the iterative process. Meanwhile, in order to maintain the convergence speed of the algorithm, the search range of the iteration step size is determined based on the Wolfe-Powell criterion. Simulation results show that FastICA, based on Wolfe-Powell criterion can solve the problem of convergence with only a small amount of computation. According to the spatial transmission characteristics of AIS signal, a multi-user separation method for satellite receiver is proposed. In order to reduce the probability of slot collision based on polarization diversity, a binary linear array with three intersections is used to receive signals in different polarization directions, and then the satellite's field of view is divided into several zonal regions by digital beamforming. The probability of slot collision is reduced based on spatial diversity. Finally, the AIS signal of slot collision is extracted one by one with the variable step size complex value FastICA separation algorithm. The simulation results show that the method is feasible and feasible.
【學位授予單位】：天津理工大學
【學位級別】：碩士
【學位授予年份】：2014
【分類號】：U675.7

【參考文獻】

相關期刊論文 前2條

1 汪軍，何振亞;瞬時混疊信號盲分離[J];電子學報;1997年04期

2 付衛(wèi)紅;史凡;劉乃安;;適用于時變信道環(huán)境的盲源分離算法[J];電子科技大學學報;2012年04期

本文編號：2201846