本文選題：空時自適應(yīng)處理　切入點：協(xié)方差矩陣估計　出處：《電子科技大學(xué)》2017年碩士論文　論文類型：學(xué)位論文

【摘要】：雜波抑制是機載雷達目標探測面臨的主要問題�？諘r自適應(yīng)處理(STAP)算法對雜波進行空域和時域的聯(lián)合處理,可顯著提高機載雷達的雜波抑制能力。實際的雜波環(huán)境往往呈現(xiàn)出非均勻、非平穩(wěn)特性,傳統(tǒng)的采用與待檢測單元相鄰的樣本對協(xié)方差矩陣進行最大似然估計的STAP算法性能受限,如何準確有效地估計待檢測單元的雜波協(xié)方差矩陣,成為進一步提升機載雷達目標探測性能的關(guān)鍵問題。本文研究非均勻雜波環(huán)境下的空時自適應(yīng)處理問題,分析了非均勻環(huán)境下進行數(shù)據(jù)樣本篩選的必要性,對現(xiàn)有樣本篩選算法進行了仿真,并利用實測數(shù)據(jù)驗證了處理算法的實際性能;在此基礎(chǔ)上,提出了一種改進樣本自適應(yīng)樣本篩選算法,在小系統(tǒng)自由度情況下,具有更優(yōu)的篩選性能。本文的具體內(nèi)容概括如下:1.介紹機載雷達雜波回波模型及STAP基本理論。從全空時自由度處理的STAP算法出發(fā),討論了降維/降秩處理的必要性,介紹了常用降維STAP算法的基本原理,并對STAP處理性能的評估參數(shù)進行說明。2.對機載雷達的非均勻雜波環(huán)境,進行信號建模與雜波仿真。對比分析了雜波的最小方差功率譜與傅里葉譜。介紹了非均勻環(huán)境下的雜波協(xié)方差矩陣估計方法及知識輔助STAP(KA-STAP)濾波器的基本原理。3.研究了兩種經(jīng)典的獨立同分布(IID)樣本篩選算法,即廣義內(nèi)積(GIP)法和基于傅里葉譜相似度(FSPS)的樣本篩選算法,討論了不同的非均勻雜波場景下,兩種算法的適用性,并進行了建模仿真和性能對比。結(jié)果表明,GIP算法對存在少量離散強雜波單元的非均勻環(huán)境有較好的篩選效果,但在存在大量非均勻樣本的環(huán)境下,算法性能較差;FSPS算法則在強非均勻性雜波環(huán)境下呈現(xiàn)出良好的穩(wěn)健性。利用兩種算法對機載雷達實測數(shù)據(jù)進行了處理,驗證了算法的有效性。4.針對在小系統(tǒng)自由度下,已有的FSPS算法在污染樣本剔除及相似樣本選擇環(huán)節(jié)都存在分辨率不足的問題,提出一種基于稀疏恢復(fù)技術(shù)的自適應(yīng)樣本篩選算法。該方法利用高精度稀疏恢復(fù)信息對參考單元樣本進行篩選。相比FSPS算法,該方法在小系統(tǒng)自由度情況下可同時提升污染樣本剔除及相似樣本選擇時的性能。通過計算機仿真驗證了該方法的有效性。
[Abstract]:Clutter suppression is the main problem in airborne radar target detection. The clutter suppression ability of airborne radar can be improved significantly. The traditional STAP algorithm using samples adjacent to the unit to estimate the covariance matrix has limited performance. How to estimate the clutter covariance matrix accurately and effectively is proposed. In this paper, the problem of space-time adaptive processing in heterogeneous clutter environment is studied, and the necessity of data sample selection in non-uniform environment is analyzed. The existing sample selection algorithm is simulated, and the actual performance of the algorithm is verified by using the measured data. On the basis of this, an improved sample adaptive sample screening algorithm is proposed, which can be used in the case of small system degrees of freedom. The detail contents of this paper are summarized as follows: 1. The clutter echo model of airborne radar and the basic theory of STAP are introduced. The necessity of dimension reduction / rank reduction processing is discussed based on the STAP algorithm of total space-time degree of freedom processing. This paper introduces the basic principle of commonly used dimensionally reduced STAP algorithm, and explains the evaluation parameters of STAP processing performance. 2. For the non-uniform clutter environment of airborne radar, The minimum variance power spectrum and Fourier spectrum of clutter are compared and analyzed. The estimation method of clutter covariance matrix and the basic principle of KAP KA-STAP filter in non-uniform environment are introduced. In this paper, two classical independent and distributed IID-based sample selection algorithms are studied. That is, the generalized inner product (GIP) method and the sample selection algorithm based on Fourier Spectrum similarity (FSPS). The applicability of the two algorithms in different heterogeneous clutter scenarios is discussed. The simulation and simulation results show that the GIP algorithm has a good screening effect on the non-uniform environment with a small number of discrete strong clutter elements, but in the presence of a large number of non-uniform samples, the GIP algorithm has a good selection effect on the non-uniform environment with a small number of discrete and strong clutter elements. The performance of the algorithm is poor and the FSPS algorithm shows good robustness in the environment of strong inhomogeneity clutter. Two algorithms are used to deal with the measured data of airborne radar, and the validity of the algorithm is verified. 4. In view of the small system degree of freedom, The existing FSPS algorithm has the problem of insufficient resolution in the selection of contaminated samples and similar samples. An adaptive sample selection algorithm based on sparse recovery technique is proposed. The method uses high precision sparse restoration information to filter reference unit samples. Compared with FSPS algorithm, the proposed algorithm is more efficient than FSPS algorithm. The proposed method can improve the performance of the proposed method for the removal of contaminated samples and the selection of similar samples at the same time in the case of small system degrees of freedom. The effectiveness of the proposed method is verified by computer simulation.
【學(xué)位授予單位】：電子科技大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2017
【分類號】：V243.2;TN957.51

【參考文獻】

相關(guān)期刊論文 前6條

1 劉漢偉;張永順;王強;吳億鋒;;基于稀疏重構(gòu)的機載雷達訓(xùn)練樣本挑選方法[J];系統(tǒng)工程與電子技術(shù);2016年07期

2 方明;劉宏偉;戴奉周;王小謨;;基于環(huán)境動態(tài)感知的空時自適應(yīng)處理[J];電子與信息學(xué)報;2015年08期

3 吳億鋒;王彤;吳建新;代保全;同亞龍;;基于道路信息的知識輔助空時自適應(yīng)處理[J];電子與信息學(xué)報;2015年03期

4 楊志偉;廖桂生;賀順;曾操;;直接數(shù)據(jù)域的自適應(yīng)匹配濾波方法[J];電子學(xué)報;2010年07期

5 周宇;張林讓;劉楠;劉昕;;空時自適應(yīng)處理中基于知識的訓(xùn)練樣本選擇策略[J];系統(tǒng)工程與電子技術(shù);2010年02期

6 吳建新;王彤;索志勇;保錚;;空時自適應(yīng)處理的滑窗遞推QR算法[J];電子與信息學(xué)報;2008年10期

相關(guān)博士學(xué)位論文 前1條

1 陽召成;基于稀疏性的空時自適應(yīng)處理理論和方法[D];國防科學(xué)技術(shù)大學(xué);2013年

，

本文編號：1639745