本文選題：窄帶雷達　切入點：輪式和履帶式車輛　出處：《西安電子科技大學(xué)》2014年碩士論文　論文類型：學(xué)位論文

【摘要】：隨著雷達技術(shù)的發(fā)展,雷達自動目標(biāo)識別已經(jīng)成為了未來雷達發(fā)展的方向。窄帶雷達的探測距離遠,并且在我國裝備的數(shù)量較多,所以對窄帶雷達目標(biāo)分類具有重要意義。另一方面,窄帶雷達的分辨率較低,目標(biāo)回波包含的信息較少,對目標(biāo)進行識別的難度大。而在地面運動的輪式車輛和履帶車輛目標(biāo),由于它們的驅(qū)動方式不同,車身相對旋轉(zhuǎn)部件不一樣,導(dǎo)致二者的微多普勒回波有較大差異,可以用此特征進行分類。本文即圍繞窄帶雷達條件下的運動車輛目標(biāo)分類問題進行了研究,主要內(nèi)容可以概括為以下三方面:1.對運動車輛目標(biāo)的微多普勒效應(yīng)進行介紹,給出了車輛目標(biāo)的微多普勒模型。首先對于旋轉(zhuǎn)單散射點的微運動進行了介紹,給出了其微多普勒信號模型的數(shù)學(xué)表達式。在此基礎(chǔ)上,給出了旋轉(zhuǎn)體的微運動模型,并且依據(jù)模型的數(shù)學(xué)表達式指出了影響目標(biāo)的微多普勒信號的相關(guān)變量。針對車輛目標(biāo)的運動特點,進一步給出了車輪與履帶的微多普勒模型,從二者模型的數(shù)學(xué)表達式上推導(dǎo)出微多普勒信號的差異之處。通過對實測的車輛目標(biāo)雷達回波信號進行分析,指出了輪式車輛和履帶式車輛回波的差異。2.針對運動車輛目標(biāo)的分類識別流程,對一些通用的信號處理方法以及分類方法進行了研究。在目標(biāo)回波的雜波抑制方面,分別介紹了基于脈沖對消MTI、基于CLEAN算法和基于廣義匹配濾波器的雜波抑制方法,通過對比各種方法的雜波抑制效果,介紹了各方法的優(yōu)缺點。在目標(biāo)識別特征的提取方面,介紹了基于多普勒分布特征和能量分布特征的提取方法,通過對真實目標(biāo)進行分類,給出了各種特征的分類效果。在分類算法的選擇方面,介紹了LDC、KNN、SVM三種分類算法,通過對各算法的分析比較,給出了各種不同分類算法的適用條件。3.對于窄帶雷達運動車輛目標(biāo)的分類識別,在DSP上進行了硬件工程的實現(xiàn)。在上述分析的目標(biāo)分類識別的各種實現(xiàn)方法的基礎(chǔ)上,綜合考慮目標(biāo)分類效果以及工程的實時性需求,選用合適的方法進行目標(biāo)分類系統(tǒng)的設(shè)計,然后分析了DSP上硬件工程各模塊的設(shè)計。最后,通過分析硬件分類系統(tǒng)的運算精度、運算時間和分類效果,驗證了該窄帶雷達運動車輛目標(biāo)分類系統(tǒng)的可行性與可靠性。
[Abstract]:With the development of radar technology, radar automatic target recognition has become the direction of radar development in the future. On the other hand, the resolution of narrowband radar is low, the target echo contains less information, and it is difficult to recognize the target. On the other hand, the target of wheeled vehicle and tracked vehicle moving on the ground, Because of their different driving modes, the relative rotating parts of the body are different, which leads to a great difference in the micro-Doppler echo between them. This paper focuses on the classification of moving vehicle targets under the condition of narrowband radar. The main contents can be summarized as follows: 1. The micro-Doppler effect of moving vehicle targets is introduced. In this paper, the micro-Doppler model of vehicle target is given. Firstly, the micro-motion of rotating single scattering point is introduced, and the mathematical expression of the micro-Doppler signal model is given. On this basis, the micro-motion model of rotating object is given. According to the mathematical expression of the model, the related variables of the micro-Doppler signal affecting the target are pointed out. According to the motion characteristics of the vehicle target, the micro-Doppler model of the wheel and track is further given. From the mathematical expressions of the two models, the differences of the micro-Doppler signals are deduced. The difference of echo between wheeled vehicle and tracked vehicle is pointed out. Aiming at the classification and recognition flow of moving vehicle target, some general signal processing methods and classification methods are studied. In the aspect of clutter suppression of target echo, The methods of clutter suppression based on pulse cancellation MTI, CLEAN algorithm and generalized matched filter are introduced, and the advantages and disadvantages of each method are compared. The extraction method based on Doppler distribution feature and energy distribution feature is introduced. By classifying real target, the classification effect of various features is given. In the selection of classification algorithm, three classification algorithms of LDC/ KNN- SVM are introduced. Through the analysis and comparison of each algorithm, the suitable conditions of different classification algorithms are given. 3. For the classification and recognition of moving vehicle target of narrowband radar, The realization of hardware engineering on DSP is carried out. On the basis of all kinds of realization methods of target classification and recognition mentioned above, the effect of target classification and the real time requirement of engineering are considered synthetically. A suitable method is selected to design the target classification system, and then the design of each module of hardware engineering on DSP is analyzed. Finally, the operation precision, operation time and classification effect of the hardware classification system are analyzed. The feasibility and reliability of the moving vehicle target classification system for narrowband radar are verified.
【學(xué)位授予單位】：西安電子科技大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2014
【分類號】：TN957.51

【參考文獻】

相關(guān)博士學(xué)位論文 前1條

1 陳渤;基于核方法的雷達高分辨距離像目標(biāo)識別技術(shù)研究[D];西安電子科技大學(xué);2008年

相關(guān)碩士學(xué)位論文 前1條

1 符婷;基于微多普勒特征的目標(biāo)分類方法研究[D];西安電子科技大學(xué);2011年

，

本文編號：1579428