本文選題：微動(dòng)特性 + 正弦調(diào)頻信號�。� 參考：《國防科學(xué)技術(shù)大學(xué)》2014年博士論文

【摘要】：微動(dòng)是目標(biāo)或目標(biāo)部件除了質(zhì)心平動(dòng)之外的一種具有小幅、往復(fù)特點(diǎn)的運(yùn)動(dòng)分量,反映了目標(biāo)的精細(xì)特征,在雷達(dá)目標(biāo)識別和成像技術(shù)領(lǐng)域占有獨(dú)特的重要作用。當(dāng)前,微動(dòng)參數(shù)估計(jì)面臨一些技術(shù)瓶頸,傳統(tǒng)時(shí)頻分析方法估計(jì)精度低,基于模型的方法對不同微動(dòng)形式的適應(yīng)性差,給以上述兩類方法為主體的微動(dòng)參數(shù)估計(jì)理論的實(shí)用性帶來了嚴(yán)峻挑戰(zhàn)。本文圍繞雷達(dá)目標(biāo)微動(dòng)特性反演問題展開,提出了正弦調(diào)頻傅立葉變換等一系列微動(dòng)參數(shù)估計(jì)方法,大幅度提高了微動(dòng)參數(shù)估計(jì)精度和微動(dòng)模型的適用范圍,研究成果對目標(biāo)微動(dòng)特性反演、雷達(dá)成像等問題具有指導(dǎo)意義。第一章為緒論,闡述了論文的研究背景及意義,綜述了微動(dòng)特性反演領(lǐng)域的發(fā)展概況和研究現(xiàn)狀,并介紹了論文的主要工作。第二章從微動(dòng)模型出發(fā),將微動(dòng)分為簡諧微動(dòng)與復(fù)雜微動(dòng)、普通微動(dòng)與小幅微動(dòng),基于多正弦調(diào)頻信號建立了囊括上述微動(dòng)形式的微動(dòng)目標(biāo)雷達(dá)回波模型。研究了時(shí)頻分析方法與基于模型的方法的局限性,以偽魏格納分布為例,分析了基于時(shí)頻分布估計(jì)正弦調(diào)頻信號瞬時(shí)頻率的偏差和隨機(jī)誤差,從理論上揭示了時(shí)頻分析方法在正弦調(diào)頻信號分析中面臨的技術(shù)瓶頸。最后,指出了時(shí)頻分析方法在微動(dòng)參數(shù)估計(jì)中的適用條件。第三章建立了正弦調(diào)頻信號空間,提出了正弦調(diào)頻傅立葉變換。與時(shí)頻分析采用短時(shí)窗做瞬時(shí)近似不同,正弦調(diào)頻傅立葉變換通過微多普勒調(diào)制信息的長時(shí)間積累,顯著提高了微動(dòng)參數(shù)估計(jì)精度,并從模型角度,根本上解決了多正弦調(diào)頻信號參數(shù)多、估計(jì)算法復(fù)雜的難題。研究了正弦調(diào)頻傅里葉變換估計(jì)精度與可估計(jì)的微動(dòng)幅度范圍,提出了微多普勒噪聲比的概念,分析了正弦調(diào)頻傅里葉變換中的相位纏繞問題與相位解纏方法,并研究了多分量多正弦調(diào)頻信號正弦調(diào)頻傅里葉變換的特點(diǎn),得到了新方法在估計(jì)多正弦調(diào)頻信號參數(shù)時(shí)的歸一化均方根誤差經(jīng)驗(yàn)公式以及多分量信號變換結(jié)果的經(jīng)驗(yàn)公式。接下來,提出了兩種相應(yīng)的解決具體問題的算法:基于正弦調(diào)頻傅立葉變換的瞬時(shí)頻率估計(jì)方法以及車輛振動(dòng)譜估計(jì)方法。理論分析和仿真實(shí)驗(yàn)結(jié)果表明,正弦調(diào)頻傅立葉變換能夠精確估計(jì)調(diào)制微弱的多正弦調(diào)頻信號參數(shù),較現(xiàn)有方法大幅提高了估計(jì)精度和抗噪性能,降低了可估計(jì)的微動(dòng)幅度下限,能夠估計(jì)未知微動(dòng)形式、多微動(dòng)頻率成分的目標(biāo)微動(dòng)特性。第四章提出了正弦調(diào)頻稀疏恢復(fù)算法,在正弦調(diào)頻信號空間建立傅立葉調(diào)頻字典,并引入稀疏恢復(fù)方法,借用正弦調(diào)頻信號空間分析正弦調(diào)頻信號的優(yōu)勢,以及稀疏貝葉斯學(xué)習(xí)對非均勻欠采樣的適應(yīng)能力,能夠分析低數(shù)據(jù)率、非均勻采樣的正弦調(diào)頻信號,實(shí)現(xiàn)微動(dòng)參數(shù)估計(jì)。較現(xiàn)有基于稀疏恢復(fù)的正弦調(diào)頻信號參數(shù)估計(jì)方法,正弦調(diào)頻稀疏恢復(fù)算法將參數(shù)維數(shù)降至1維,具有更高的穩(wěn)健性、估計(jì)精度和算法效率。分析了正弦調(diào)頻稀疏恢復(fù)算法的微動(dòng)譜模糊問題,分析了傅立葉調(diào)頻字典的相干性和網(wǎng)格劃分問題。研究了遠(yuǎn)程低頻段雷達(dá)中段目標(biāo)進(jìn)動(dòng)參數(shù)估計(jì)問題,針對遠(yuǎn)程低頻段雷達(dá)估計(jì)中段目標(biāo)進(jìn)動(dòng)參數(shù)受限于數(shù)據(jù)率低、采樣非均勻、目標(biāo)微多普勒效應(yīng)弱等難題,基于正弦調(diào)頻空間稀疏恢復(fù)算法,采用美國鋪路爪雷達(dá)參數(shù)仿真,通過約5分鐘的跟蹤數(shù)據(jù)積累,可實(shí)現(xiàn)進(jìn)動(dòng)頻率的精確估計(jì)。第五章提出了可實(shí)現(xiàn)大轉(zhuǎn)角下不同散射中心類型的清晰成像的微動(dòng)目標(biāo)非理想散射中心成像方法。將正弦調(diào)頻傅立葉變換與HRRP序列結(jié)合,先提取不同散射中心的微動(dòng)信息,實(shí)現(xiàn)其位置的粗估計(jì)。建立了散射中心復(fù)合模型,進(jìn)而,通過散射中心類型判別、基于微動(dòng)譜的滑動(dòng)型散射中心參數(shù)估計(jì)等一系列處理方法,實(shí)現(xiàn)散射中心位置的精估計(jì),并反演相應(yīng)的目標(biāo)結(jié)構(gòu)特征。目標(biāo)非理想散射中心成像方法能夠反映目標(biāo)不同類型散射中心的空間分布,給出目標(biāo)主要結(jié)構(gòu)的尺寸角度信息,相比傳統(tǒng)成像方法,更加直觀、全面。第六章總結(jié)全文,并指出需要進(jìn)一步研究的問題。本文從信號模型與信號處理方法角度在傳統(tǒng)微動(dòng)特性反演技術(shù)方面取得了新的突破,在正弦調(diào)頻信號空間中提出了一系列微動(dòng)目標(biāo)參數(shù)估計(jì)、特性反演、非線性調(diào)頻信號分析的理論與方法,實(shí)現(xiàn)了目標(biāo)微動(dòng)特性的精細(xì)反演,拓展了微動(dòng)幅度的可估計(jì)范圍。取得的成果對于微動(dòng)參數(shù)估計(jì)、雷達(dá)成像、目標(biāo)識別和電子對抗具有一定的參考價(jià)值。
[Abstract]:As a target or target component, a motion component with small amplitude and reciprocating features, in addition to the translational motion of the center of mass, reflects the fine characteristics of the target. It plays a unique and important role in the field of radar target recognition and imaging technology. At present, the estimation of the microdynamic parameters is faced with some technical bottlenecks, and the traditional time frequency analysis method has a low estimation accuracy. The model method has brought a severe challenge to the practicability of the theory of microdynamic parameter estimation with the two methods as the main body, which is poor in the adaptability of different microdynamic forms. In this paper, a series of parameter estimation methods such as sinusoidal FM Fu Liye transform, such as sinusoidal FM transformation, are proposed in this paper. The accuracy of the estimation of dynamic parameters and the scope of the application of the microdynamic model, the research results have the guiding significance for the inversion of the target fretting characteristics and the radar imaging. The first chapter is the introduction, expounds the research background and significance of the paper, summarizes the development and research status of the field of microdynamic inversion, and introduces the main work of the paper. Second chapters From the microdynamic model, the micromotion is divided into simple harmonic fretting and complex micromovement, ordinary micromotion and small amplitude micromotion. Based on the multi sinusoidal FM signal, the microdynamic target radar echo model including the above micromotion is established. The limitation of the time frequency analysis method and the model based method is studied. The pseudo Wegener distribution is taken as an example to analyze the time based time. The frequency distribution is used to estimate the deviation and random error of the instantaneous frequency of sinusoidal frequency modulation signal. In theory, it reveals the technical bottleneck of the time-frequency analysis method in the analysis of sinusoidal frequency modulation signal. Finally, it points out the application conditions of the time frequency analysis method in the estimation of the microdynamic parameters. The third chapter establishes the sinusoidal frequency modulation signal space and puts forward the sinusoidal frequency modulation. Fu Liye transform is different from the instantaneous approximation of time frequency analysis using short time window. Sinusoidal FM Fu Liye transform can significantly improve the precision of the estimation of microdynamic parameters by long time accumulation of micro Doppler modulation information. From the model angle, the multi sinusoidal frequency modulation signal is more complex and the algorithm is complex. The concept of micro Doppler noise ratio is proposed and the phase entanglement and phase unwrapping method in the sinusoidal FM Fu Liye transform is analyzed. The characteristics of the sinusoidal FM Fu Liye transform of multicomponent multi sinusoidal FM signal are studied, and the new method is obtained in the estimation of the multiple Fu Liye transform. The normalized mean square root error empirical formula and the empirical formula for the result of multicomponent signal transformation are given in the parameters of chirp signal. Then, two corresponding algorithms are proposed: the instantaneous frequency estimation method based on sinusoidal FM Fu Liye transform and the method of vehicle vibration spectrum estimation. Theoretical analysis and simulation experiment junction The results show that the sinusoidal FM Fu Liye transform can accurately estimate the parameters of the weak modulated multi sinusoidal frequency modulation signal. Compared with the existing methods, the estimation precision and the anti noise performance are greatly improved. The lower limit of the estimated fretting amplitude is reduced and the target fretting characteristics of the unknown fretting form and the multi fretting frequency are estimated. The fourth chapter puts forward the sinusoidal modulation. The frequency sparse recovery algorithm is used to establish the Fu Liye frequency modulation dictionary in the sinusoidal frequency modulation signal space, and the sparse recovery method is introduced, the sinusoidal FM signal is used to analyze the advantages of the sinusoidal frequency modulation signal and the sparse Bayesian learning is adapted to the non-uniform undersampling, and the sine frequency modulation signal with low data rate and non-uniform sampling can be analyzed. Compared with the existing parameter estimation method of the sinusoidal frequency modulation signal based on sparse recovery, the sinusoidal FM sparse recovery algorithm reduces the parameter dimension to 1 dimension, has higher robustness, the estimation precision and the algorithm efficiency. This paper analyzes the micro motion spectrum fuzzy problem of the sinusoidal FM sparse recovery algorithm, and analyzes the phase of the Fu Liye frequency modulation dictionary. The parameter estimation of middle target precession in long range low frequency radar is studied. Aiming at the problem of low rate of data, inhomogeneous sampling and weak Doppler effect in the middle segment of the long range low frequency radar, the US paving claw radar is used on the basis of the sine FM space sparse recovery algorithm. Parameter simulation, through the accumulation of about 5 minutes of tracking data, can achieve accurate estimation of the precession frequency. The fifth chapter proposes a non ideal scattering center imaging method which can realize clear imaging of different scattering center types at large rotation angle. The sinusoidal FM Fu Liye transform is combined with the HRRP sequence to extract the different scattering centers. The scattering center complex model is established by moving information. Then a series of processing methods, such as the type discrimination of the scattering center type and the parameter estimation of the sliding scattering center based on the microdynamic spectrum, can be used to realize the precise estimation of the position of the scattering center and inversion of the corresponding target structure characteristics. The imaging method of the target non ideal scattering center is used. It can reflect the spatial distribution of different types of scattering centers of the target, and give the dimension and angle information of the main structure of the target. Compared with the traditional imaging method, it is more intuitive and comprehensive. The sixth chapter summarizes the full text, and points out the problems that need further study. This paper from the angle of signal model and signal processing method in the traditional microdynamic characteristics inversion technology A new breakthrough has been made. In the sinusoidal frequency modulation signal space, a series of theory and methods of parameter estimation, characteristic inversion and Nonlinear FM signal analysis are proposed. The fine inversion of the microdynamic characteristics of the target is realized and the estimated range of the microdynamic amplitude is extended. The results obtained are for the estimation of the microdynamic parameters, the radar imaging, and the target recognition. Do not have a certain reference value with electronic countermeasures.

【學(xué)位授予單位】：國防科學(xué)技術(shù)大學(xué)
【學(xué)位級別】：博士
【學(xué)位授予年份】：2014
【分類號】：TN957.51

【參考文獻(xiàn)】

相關(guān)期刊論文 前2條

1 黃孟俊;趙宏鐘;付強(qiáng);馮國瑜;;一種基于微多普勒特征的海面角反射器干擾鑒別方法[J];宇航學(xué)報(bào);2012年10期

2 李東文;熊曉燕;李博;;振動(dòng)加速度信號處理探討[J];機(jī)電工程技術(shù);2008年09期

，

本文編號：1861670