【正文】 碩士學(xué)位論文基于信號先驗信息的 DOA 估計算法研究RESEARCH OF DOA ESTIMATION ALGORITHMS BASED ON PRIORI KNOWLEDGE OF SIGNALS國內(nèi)圖書分類號：TN974 學(xué)校代碼：10213國際圖書分類號： 密級：公開 工學(xué)碩士學(xué)位論文基于信號先驗信息的 DOA 估計算法研究碩士研究生 ：導(dǎo) 師 ：申請學(xué)位 ： 工學(xué)碩士學(xué)科 ： 信息與通信工程所 在 單 位 ： 信息與電氣工程學(xué)院答 辯 日 期 ：授予學(xué)位單位 ：Classified Index: TN974 : Dissertation for the Master Degree in EngineeringRESEARCH OF DOA ESTIMATION ALGORITHMS BASED ON PRIORI KNOWLEDGE OF SIGNALSCandidate： Yang LeiSupervisor： XingpengAcademic Degree Applied for： Master of EngineeringSpeciality： Information and Communication EngineeringAffiliation： School of Information and Electrical EngineeringDate of Defence： June, 2022DegreeConferringInstitution： Harbin Institute of Technology哈爾濱工業(yè)大學(xué)工學(xué)碩士學(xué)位論文 I 摘 要隨著通信技術(shù)的發(fā)展，作為陣列信號處理領(lǐng)域的一個重要分支，DOA 估計技術(shù)的研究已經(jīng)取得了巨大的進步。傳統(tǒng)的 DOA 估計技術(shù)在對來波信號完全未知的情況下就可以實現(xiàn)波達角度的估計檢測。近些年來，DOA 估計技術(shù)的應(yīng)用不僅局限在導(dǎo)航定位、電子偵查與對抗等軍事領(lǐng)域。在許多民用應(yīng)用領(lǐng)域，如機電測量、生物醫(yī)學(xué)等方面也取得了廣泛的應(yīng)用。在這些應(yīng)用中，來波信號的先驗信息通常是完全已知的，這便推動了基于信號先驗信息的 DOA 估計算法研究的興起。本文針對先驗信息已知信號的 DOA 估計算法進行了探討研究，主要分析了兩大類算法，一類是以最大似然準則為基礎(chǔ)的最大似然估計類算法，如 DEML 算法、CDEML 算法、WDEML 算法等。最大似然估計類算法具有極高的測角精度與分辨力，相比于傳統(tǒng)的 DOA 估計算法在性能上有著極大的提升，但其存在著運算量大的問題。另一類是利用信號先驗信息進行迭代運算的 DOA 估計算法，如 MSWFMUSIC 算法、線陣 PASI 算法等。相比于傳統(tǒng)的 DOA 估計算法，利用迭代技術(shù)的算法通常有著突出的性能，如能夠適應(yīng)低信噪比、少快拍的惡劣環(huán)境，能夠提高測角精度或者是能降低算法的運算復(fù)雜度等。對于最大似然估計類算法，本文主要分析了影響其算法性能的各個因素條件，并與 MUSIC 算法進行了對比，總結(jié)得到了算法的優(yōu)勢與不足。同時分析了接收端所用參考信號先驗信息的變化對算法性能的影響，得到了相應(yīng)的結(jié)論與規(guī)律。對基于迭代技術(shù)的 DOA 估計算法，本文主要研究了其所擁有的一些特點，分析了迭代對于算法性能和運算復(fù)雜度所帶來的優(yōu)勢。針對線陣 PASI 算法應(yīng)用的局限性問題，本文提出了一種基于陣列平移的圓陣 PASI 算法，解決了 PASI 算法在其他陣列形式中因難以構(gòu)造合適的子陣而無法應(yīng)用的問題。該算法利用陣列平移的方式構(gòu)造平滑子陣，再通過平移形成的子陣接收來波信號。由此獲得的各個子陣的接收數(shù)據(jù)便可以應(yīng)用 PASI 算法進行 DOA估計。對所提出算法進行了性能的仿真分析，并與其它基于信號先驗信息的 DOA估計算法進行了比較。關(guān)鍵詞：DOA 估計；最大似然；先驗信息；迭代；陣列平移哈爾濱工業(yè)大學(xué)工學(xué)碩士學(xué)位論文 II AbstractWith the development of munication technology, the DOA estimation technology has achieved great progress as an important branch of array signal processing. Traditional DOA estimation technique can estimate the direction of arriving signals in the case that the signals are pletely unknown for the receiver. In recent years, DOA estimation technology is not only confined to the military fileds, such as the navigation and positioning, the electronic surveilance and bat, etc. but also applied to many civilian fileds, such as mechanical and electrical measurements, biomedicine, etc. In these applications, the priori knowledge of the signals is usually pletely known for the receiver. This has promoted the research of DOA estimation algorithm based on priori knowledge of signals.The paper makes discussions and researches to DOA estimation algorithms based on priori knowledge of signals. The paper mainly analyzes two types of algorithm. One type is the maximum likelihood estimation algorithms based on the maximum likelihood criterion, such as DEML algorithm, CDEML algorithm, WDEML algorithm, etc. The maximum likelihood estimation algorithm has a very high accuracy and resolution. Compared to the traditional DOA estimation algorithm, it has made great improvement in performance, but it has the problem that the putation is large. The other make DOA estimation through the iterative technology by using the priori knowledge of signals, such as MSWFMUSIC algorithm, PASI algorithm in linear array, etc. Compared to the traditional DOA estimation algorithms, the algorithms using the iterative technology usually has outstanding performance, for example it can adapt to low SNR and low snapshops, it can improve the accuracy and reduce the putational plexity, etc.For maximum likelihood estimation algorithms, this paper mainly analyzes various factors which influence the performance of algorithms and makes a parison to the MUSIC algorithm, then sums to obtain the advantages and disadvantages of the algorithm. We also analyzes the influence of changes of signals’ priori knowledge used by receiver on this algorithm and gets the corresponding conclusions and laws. For the DOA estimation algorithm using iterative technique, the paper mainly studies some of the characterstics that it has and analyzes the advantages of algorithms’ performance and putational plexity which iterative technique brings.For the limitations problem of the application of the PASI algorithm in the linear array, this paper proposes a PASI algorithm based on array translation in the circular array, which solves the problem that the PASI algorithm can’t be used in other forms of array because of the difficulty to construct appropriate subarray. This algorithm 哈爾濱工業(yè)大學(xué)工學(xué)碩士學(xué)位論文 III constructs subarrays by using the way of array translation and receives signals through these subarrays. Then the signals receiving through this way can be used in PASI algorithm to estimate the direction of signals. We make the simulation analysis of performance for the proposed algorithm and make parisons with other DOA estimation algorithms based on the priori knowledge of signals.Keywords: DOA estimation, maximum likelihood, priori knowledge, iteration, array translation哈爾濱工業(yè)大學(xué)工學(xué)碩士學(xué)位論文 III 目 錄摘 要 ................................................................................................................................IABSTRACT......................................................................................................................II第 1 章 緒 論 .................................................................................................................1 課題背景及研究的目的和意義 .............................................................................1 相關(guān)理論的研究歷史與現(xiàn)狀分析 .........................................................................2 傳統(tǒng)的 DOA 估計理論 ..................................................................................2 先驗信息已知的信號 DOA 估計理論 ..........................................................3 本文的主要研究內(nèi)容及結(jié)構(gòu) .............