本文選題：ISAR　切入點：微多普勒　出處：《哈爾濱工業(yè)大學》2017年碩士論文　論文類型：學位論文

【摘要】：對逆合成孔徑雷達(Inverse Synthetic Aperture Radar,ISAR)成像而言,目標上的微多普勒效應不僅會造成參數(shù)提取和運動誤差補償困難,而且會造成圖像散焦,形成方位向干擾帶,嚴重影響成像質量。因此,研究微多普勒抑制問題對ISAR的高精度成像有重要意義,同時在現(xiàn)代化戰(zhàn)爭的高精度目標識別對抗中,有重要應用價值。本文針對實際應用中的ISAR成像微多普勒抑制問題,建立合理有效的數(shù)學模型,并基于矩陣填充的理論,實現(xiàn)ISAR成像中微多普勒抑制研究。論文的主要內容及創(chuàng)新點如下:首先,確立了基于矩陣填充理論的微多普勒抑制研究方案。在詳細分析了微多普勒抑制研究的應用背景和主要影響的基礎上,結合微多普勒抑制研究的主要方法;分析得到單幅ISAR圖像中,無論在實數(shù)域還是復數(shù)域微多普勒圖像都具有低秩的特性,并且提出了基于低秩矩陣分解的單幅實數(shù)域ISAR像微多普勒抑制和復數(shù)域ISAR像微多普勒抑制的研究方案。此外,本文又從另一個角度去解決微多普勒抑制問題:首先生成一序列ISAR圖像,將其重新排列成新的觀測矩陣,根據(jù)差分閾值法進行低秩和稀疏的檢測,并進行稀疏和低秩性分解,將目標和微多普勒進行分離,達到微多普勒抑制的效果。兩類研究方案為全文的研究內容奠定了基礎。其次,分析了基于轉臺模型RD成像振動和旋轉微多普勒效應。針對早期“點目標”模型不足以反應目標微動信息的現(xiàn)狀,以飛機轉臺為例,建立了ISAR成像的轉臺模型;結合經典RD成像算法,詳細分析了振動和旋轉兩種典型微動造成的微多普勒效應;實現(xiàn)了微動散射點的回波分析和微多普勒效應干擾仿真,并給出了基于實測數(shù)據(jù)的微多普勒效應結果。再次,提出了基于低秩矩陣分解的ISAR微多普勒抑制方法。詳細分析了飛機螺旋槳單散射點復數(shù)微多普勒圖像和模值微多普勒圖像的低秩性;針對ISAR圖像稀疏和微多普勒圖像低秩的特點,分析了矩陣的有效秩,并對含有微多普勒ISAR像進行SVD低秩擬合,保證微多普勒的低秩性;結合魯棒性主分量分析(RPCA)理論和加速近鄰梯度(APG)算法,實現(xiàn)了對ISAR像的模值RPCA和復數(shù)RPCA處理,并且在復數(shù)域中對APG算法進行了修正,重新構建了復數(shù)域的收縮算子。然后,分析對比了仿真中和實測數(shù)據(jù)中模值RPCA和復數(shù)RPCA處理效果。通過仿真和實測數(shù)據(jù)處理,驗證了基于低秩矩陣分解方法實現(xiàn)ISAR微多普勒抑制的有效性;結合對圖像信噪比和相干系數(shù)的分析,詳細說明了本文所用方法對ISAR成像微多普勒效應的抑制效果良好。最后,在一定小轉角內,連續(xù)成像一序列ISAR圖像,保證目標基本沒有轉動,而微多普勒在這個期間內產生了大的變動。因此,將一序列圖像中的每一幅圖像轉化成一個一維向量,重新組成新的觀測矩陣,通過差分閾值法進行檢測,得到目標為低秩矩陣,微多普勒為稀疏矩陣,并對其進行基于IALM算法和APG算法的低秩矩陣分解,并對實驗結果進行分析對比。
[Abstract]:The inverse synthetic aperture radar (Inverse Synthetic Aperture Radar, ISAR) imaging, micro Doppler effect on the target will not only cause the parameter extraction and motion compensation is difficult, but also cause the defocus, form azimuth disturbance, seriously affect the image quality. Therefore, it has important significance to study the inhibition of high precision imaging for micro Doppler the problem of ISAR, and high precision of target recognition in the Modern War confrontation, has important application value. In this paper, ISAR imaging for the practical application of micro Doppler suppression, to establish a reasonable mathematical model based on matrix theory, the realization of ISAR imaging in micro Doppler suppression research. The main contents and innovation points are as follows: firstly, establish the matrix filling theory of micro Doppler suppression scheme based on research. In a detailed analysis of the micro Doppler inhibition studies should be used and the main background Based on the influence, combining with the main methods of micro Doppler inhibition; analysis of single ISAR image, has the characteristics of low rank in the real domain or the complex field of micro Doppler image, and put forward a real domain ISAR low rank matrix decomposition based on micro Doppler like inhibition and complex ISAR like micro Doppler suppression the research program. In addition, this paper also from another point of view to solve the problem of micro Doppler suppression: first generate a sequence of ISAR images, which can be rearranged into a new observation matrix, based on the detection of the differential threshold method of low rank and sparse, and sparse and low rank decomposition of the target and the micro Doppler to achieve separation of micro Doppler inhibitory effect. Two kinds of schemes for the study content of this thesis lays a foundation. Secondly, analyzes the vibration and rotation RD imaging turntable model based on micro Doppler effect for. Status quo of early "target" model do not reflect the target micro motion information, the aircraft simulator as example, established a model of ISAR imaging turntable; combined with the classic RD imaging algorithm, micro Doppler effect of vibration and rotation of two kinds of typical fretting caused by the detailed analysis of the implementation of the micro; scatter echo analysis and Simulation of micro Doppler effect the interference, and gives the micro Doppler effect results based on measured data. Thirdly, the low rank matrix decomposition ISAR suppression method based on micro Doppler. A detailed analysis of aircraft propeller single scattering point complex micro Doppler image and the modulus of low rank Doppler micro image; according to the characteristics of ISAR image sparse and micro Doppler image analysis of the effective rank, rank of matrix, and the micro Doppler ISAR like SVD containing low rank fitting, ensuring low rank of micro Doppler; combined with the analysis of robust principal component (RPCA) and The nearest neighbor and accelerating gradient (APG) algorithm, the realization of the ISAR like mode value RPCA and complex RPCA processing, which is modified and the APG algorithm in the complex domain, constructs contraction operator complex domain. Then, analysis and comparison of the RPCA and the complex number RPCA treatment effect simulation and measured data model value through simulation and experimental data processing, to verify the low rank matrix decomposition method to realize the effectiveness of ISAR suppression based on micro Doppler; combined with the analysis of the image signal-to-noise ratio and coherent coefficient, a detailed description of the method used in this paper the inhibitory effect on ISAR imaging and micro Doppler effect is good. Finally, in a small corner, continuous imaging of a sequence of ISAR images, which basically does not rotate, and the micro Doppler in this period had a big change. Therefore, will be transformed into a one-dimensional vector of each image in a sequence of images, again. The observation matrix is detected by differential threshold method, and the target is low rank matrix. Micro Doppler is sparse matrix. And the low rank matrix factorization based on IALM algorithm and APG algorithm is carried out, and the experimental results are analyzed and compared.

【學位授予單位】：哈爾濱工業(yè)大學
【學位級別】：碩士
【學位授予年份】：2017
【分類號】：TN957.52

【參考文獻】

相關期刊論文 前10條

1 汪東芳;鄭睿;;高分辨率ISAR成像技術研究[J];艦船科學技術;2016年16期

2 靳繼紅;李麗;;艦船目標ISAR成像研究[J];艦船科學技術;2016年08期

3 張榆紅;邢孟道;徐剛;;基于稀疏孔徑的聯(lián)合稀疏約束干涉ISAR機動目標三維成像[J];電子與信息學報;2015年09期

4 秦曉東;張仕元;喻榮梅;;基于時間分段數(shù)據(jù)的彈道目標微多普勒時頻分析方法[J];空軍預警學院學報;2015年04期

5 王兆云;張興敢;柏業(yè)超;;基于微多普勒的錐體目標進動和結構參數(shù)估計[J];南京大學學報(自然科學);2014年02期

6 袁斌;徐世友;劉洋;陳曾平;;基于復數(shù)經驗模式分解的非旋轉對稱空間進動目標回波分離及成像研究[J];電子與信息學報;2013年01期

7 張群;羅迎;何勁;;雷達目標微多普勒效應研究概述[J];空軍工程大學學報(自然科學版);2011年02期

8 羅迎;池龍;張群;郭英;;用慢時間域積分法實現(xiàn)雷達目標微多普勒信息的提取[J];電子與信息學報;2008年09期

9 陳行勇;劉永祥;黎湘;郭桂蓉;;雷達目標微多普勒特征提取[J];信號處理;2007年02期

10 陳行勇;劉永祥;黎湘;郭桂蓉;;微多普勒分析和參數(shù)估計[J];紅外與毫米波學報;2006年05期

相關博士學位論文 前2條

1 王寶帥;基于微多普勒效應的空中飛機目標分類研究[D];西安電子科技大學;2015年

2 汪玲;逆合成孔徑雷達成像關鍵技術研究[D];南京航空航天大學;2006年

相關碩士學位論文 前2條

1 付夢;基于壓縮感知的ISAR抗干擾成像方法研究[D];哈爾濱工業(yè)大學;2016年

2 劉陽;ISAR成像中JEM干擾的抑制[D];哈爾濱工業(yè)大學;2008年

，

本文編號：1600515