本文選題：雷達(dá) + 波形設(shè)計��； 參考：《哈爾濱工業(yè)大學(xué)》2014年博士論文

【摘要】：傳統(tǒng)的雷達(dá)系統(tǒng)一旦確定發(fā)射波形以后,便無法在運行時再對波形進(jìn)行更改。因此雷達(dá)波形設(shè)計算法需要考慮各種不同的性能指標(biāo),以便設(shè)計出完美的波形,使其適用于可能遭遇到的雷達(dá)場景中。然而,由于波形設(shè)計的自由度有限,當(dāng)所有指標(biāo)添加到同一個波形優(yōu)化的目標(biāo)函數(shù)之后,難免顧此失彼。隨著在線波形設(shè)計技術(shù)日趨成熟,越來越多的研究者投入到波形設(shè)計的研究中。一旦雷達(dá)可以在運行時改變發(fā)射波形,就沒有必要去設(shè)計在各個方面都“完美”的波形;波形設(shè)計算法可以針對當(dāng)前的雷達(dá)場景量身定做發(fā)射波形;此時只需要考慮當(dāng)前環(huán)境中的目標(biāo)和干擾。也就是說,在保證波形設(shè)計自由度不變的情況下減少優(yōu)化指標(biāo)之后,可以使設(shè)計波形相比傳統(tǒng)波形在某些特定的方面具有更好的性能,能夠更有效地減小當(dāng)前環(huán)境中的干擾。在線波形設(shè)計技術(shù)主要由兩個部分構(gòu)成:1)用于估計當(dāng)前雷達(dá)場景參數(shù)的估計器;2)波形設(shè)計算法。近年來,雷達(dá)的研究者們已經(jīng)提出了眾多可用于在線波形設(shè)計的優(yōu)化算法,但是這些算法都或多或少存在一些特定的問題和缺陷,使它們無法應(yīng)用于一些復(fù)雜的雷達(dá)場景中。本課題針對這些問題及缺陷,著重研究了波形設(shè)計算法。首先根據(jù)雷達(dá)場景中常見的距離旁瓣遮蔽問題,提出了迭代的功率譜逼近算法(Iterative Spectral Approximation Algorithm,ISAA)框架。在引入了脈沖壓縮的概念之后,距離旁瓣遮蔽一直是雷達(dá)設(shè)計者需要考慮的問題。匹配濾波器輸出的距離像可以建模為雷達(dá)場景與波形自相關(guān)函數(shù)的卷積;在這種模型下,波形的自相關(guān)函數(shù)類似成像模型中的點擴(kuò)散函數(shù),將一個距離單元中的回波能量散布到了其它的距離單元之中,從而導(dǎo)致了距離像的模糊。但不同于光學(xué)系統(tǒng),對于主動雷達(dá)系統(tǒng)而言,發(fā)射波形是可以掌控的要素。這意味著一旦獲取了當(dāng)前距離像的粗略信息,并了解了強(qiáng)散射體存在的距離單元,就可以通過設(shè)計在指定區(qū)間具有低自相關(guān)幅值的波形,以抑制強(qiáng)散射體距離旁瓣對特定距離區(qū)間的干擾;而ISAA框架正是在這種背景下為了設(shè)計波形而提出的。ISAA框架根據(jù)波形的恒模特性構(gòu)造約束條件,利用相關(guān)函數(shù)與譜的傅立葉變換關(guān)系在頻域構(gòu)造設(shè)計目標(biāo),通過交替投影的方法實現(xiàn)了波形優(yōu)化。本課題提出了一種新穎的動態(tài)理想自相關(guān)構(gòu)造方法(Dynamic Ideal Autocorrelation Construction,DIAC),該方法與ISAA框架結(jié)合之后,可以得到一種高效的具體算法,相比國內(nèi)外其它研究中提出的同類算法,減小了波形優(yōu)化的時間消耗。其次,提出了名為純相位非線性規(guī)劃(Phase-Only Nonlinear Programming,PONLP)的算法框架。PONLP與ISAA框架類似,也可用于波形自相關(guān)的優(yōu)化。然而與ISAA框架相比,PONLP有一個獨特的優(yōu)勢:在設(shè)計波形時,可以在其自相關(guān)序列上設(shè)置數(shù)量較多,寬度較窄的低相關(guān)幅值區(qū)間。具有這種特點的波形可以用來抑制海雜波尖峰產(chǎn)生的距離旁瓣。不同于ISAA,PONLP是一種基于梯度的優(yōu)化方法。不同于傳統(tǒng)的基于梯度的方法,PONLP使用純相位導(dǎo)數(shù)替代導(dǎo)數(shù),使用純相位一維搜索替代直線型的一維搜索。通過這種方式,算法的每一次的迭代都能保證設(shè)計波形位于可行域之中,從而提高了收斂速度。第三,針對雷達(dá)場景中可能存在的大量窄帶有源干擾,提出了秩虧傅立葉變換的概念,并將其應(yīng)用于交替投影框架之中,得到了ISAA框架的另一類具體算法。其設(shè)計波形在指定的頻率區(qū)間具有極低的能量分布,從而降低了窄帶有源干擾對信號的影響。傳統(tǒng)的雷達(dá)系統(tǒng)可以在接收端設(shè)置帶阻濾波器以抑制外部的有源干擾。然而,如果干擾和發(fā)射波形在頻率軸有重疊,那么帶阻濾波器在濾除干擾的同時,也會消除一部分來自目標(biāo)的回波能量,進(jìn)而降低信噪比。而波形設(shè)計的方法可以保證波形與干擾盡可能在頻率軸沒有重疊,使接收端的濾波器在濾除干擾的同時不會削弱回波。為了讓設(shè)計波形在具有特定的功率譜形狀的同時對波形的自相關(guān)函數(shù)形狀進(jìn)行優(yōu)化,分別提出了基于多集合交替投影的ISAA框架和基于多目標(biāo)的PONLP框架,由這兩種框架導(dǎo)出的具體算法均可完成既定的設(shè)計目標(biāo)。第四,根據(jù)多波形設(shè)計和應(yīng)用的特點,對ISAA框架進(jìn)行了擴(kuò)展。分別提出了基于ISAA的互相關(guān)優(yōu)化-內(nèi)積約束(Cross-Correlation Optimization-Inner Product Constraint,CSOIPC)算法和多維ISAA算法(Multi-Dimensional ISAA,MDISAA),基于廣義ISAA(Generalized-ISAA,GISAA)的APLOWP框架。其中CSOIPC可用于發(fā)射-接收聯(lián)合優(yōu)化,APLOWP可為瞬態(tài)極化雷達(dá)系統(tǒng)設(shè)計發(fā)射波形,而MDISAA則是一種通用的多波形設(shè)計框架,可應(yīng)用于多輸入-多輸出(Multi-Input,Multi-Output,MIMO)雷達(dá)的正交波形設(shè)計。最后,針對實際工作中遇到的各種雷達(dá)場景的特點,通過數(shù)學(xué)建模的方法構(gòu)造了若干假想的雷達(dá)場景;在這些場景中,利用計算機(jī)數(shù)值仿真對本課題提出的各種波形設(shè)計算法進(jìn)行了測試。數(shù)值仿真的結(jié)果驗證了這些算法的有效性。
[Abstract]:The traditional radar system can not change the waveform again when the waveform is confirmed, so the radar waveform design algorithm needs to consider various performance indicators so as to design a perfect waveform to apply to the possible radar scene. However, because the degree of freedom of the waveform design is limited, when the radar waveform design is limited, When the target function is added to the same waveform optimization, it is difficult to avoid it. As the online waveform design technology is mature, more and more researchers have put into the research of waveform design. Once radar can change the waveform of the launch, it is not necessary to design the "perfect" waveform in all aspects; The shape design algorithm can tailor the waveform for the current radar scene. At this time, only the target and the interference in the current environment are considered. That is to say, the design waveform can have better performance than the traditional waveform in certain aspects after reducing the optimization index under the condition that the free degree of the waveform design is kept constant. It can reduce the interference in the current environment more effectively. The online waveform design technology consists of two parts: 1) the estimator for the current radar scene parameters; 2) the waveform design algorithm. In recent years, the radar researchers have proposed a number of optimization algorithms that can be used for online wave design, but these algorithms are all or more or more. There are few specific problems and defects so that they can not be applied to some complex radar scenes. This topic focuses on the research of waveform design algorithms for these problems and defects. Firstly, according to the common distance sidelobe occlusion problem in radar scenes, an iterative work rate spectrum approximation algorithm (Iterative Spectral Approximation) is proposed. Algorithm, ISAA) frame. After the introduction of the concept of pulse compression, the distance sidelobe occlusion has been a problem that radar designers need to consider. The range image of the output of the matched filter can be modeled as the convolution of the autocorrelation function of the radar scene and the waveform; in this model, the autocorrelation function of the waveform is similar to the point diffusion in the imaging model. The function is spread the echo energy in a distance unit into other distance units, which leads to the blurring of the distance image. But unlike the optical system, the emission waveform is a controllable element for the active radar system. This means that once the rough information of the current distance image is obtained and the strong scatterer is known. In the distance unit, the waveform with low autocorrelation amplitude can be designed in the specified interval to suppress the interference of the strong scatterer's distance sidelobe to a specific distance interval; and the ISAA frame is the.ISAA frame proposed to design the waveform in this background to construct the constraint conditions based on the constant modeling of the waveform, and use the correlation function and the correlation function. The Fu Liye transform relation in the spectrum is designed in the frequency domain, and the waveform optimization is realized by alternating projection. A novel dynamic ideal autocorrelation construction method (Dynamic Ideal Autocorrelation Construction, DIAC) is proposed in this paper. After the method is combined with the ISAA framework, a kind of efficient and specific algorithm can be obtained. Compared with the similar algorithms proposed in other studies at home and abroad, the time consumption of waveform optimization is reduced. Secondly, the algorithm framework named Phase-Only Nonlinear Programming (PONLP) is similar to ISAA framework, and can also be used for waveform autocorrelation optimization. However, compared with the ISAA framework, PONLP has a unique feature. Advantage: when designing a waveform, a low correlation amplitude range is set on its autocorrelation sequence. The waveform with this characteristic can be used to suppress the distance sidelobe generated by the sea clutter peak. Unlike ISAA, PONLP is a gradient based optimization method. Different from the traditional gradient based method, PONLP makes The pure phase derivative is replaced by the derivative, and the pure phase one-dimensional search is used to replace the linear one-dimensional search. By this way, each iteration of the algorithm can ensure the design waveform is in the feasible domain, thus improving the convergence speed. Third, the rank loss Fu Li is proposed for a large number of narrow band active interference which may exist in the radar scene. The concept of leaf transformation is applied to the alternate projection frame, and another specific algorithm of the ISAA frame is obtained. The design waveform has a very low energy distribution in the specified frequency range, thus reducing the influence of the narrow band active interference on the signal. However, if the interference and emission waveforms overlap in the frequency axis, the band resistance filter also eliminates a part of the echo energy from the target, and then reduces the signal to noise ratio while filtering the interference. And the waveform design method can ensure that the waveform and interference are not overlapped at the frequency axis as much as possible, so that the receiver is filtered. It does not weaken the echo while filtering the interference. In order to optimize the shape of the autocorrelation function of the waveform at the same time that the design waveform has a specific power spectrum shape, the ISAA framework based on multi set alternating projection and the multi-objective PONLP framework are proposed respectively. The specific algorithms derived from these two frameworks can complete both. Fourth, according to the characteristics of multi waveform design and application, the ISAA framework is extended. The cross correlation optimization based on ISAA (Cross-Correlation Optimization-Inner Product Constraint, CSOIPC) and multidimensional ISAA algorithm (Multi-Dimensional ISAA, MDISAA) are respectively proposed, based on the generalized ISAA. The APLOWP framework of d-ISAA, GISAA). In which CSOIPC can be used for the joint optimization of transmit and receive, APLOWP can design the transmitting waveform for the transient polarization radar system, and MDISAA is a universal multi waveform design framework, which can be applied to the orthogonal waveform design of the multiple input multiple output (Multi-Input, Multi-Output, MIMO) radar. Finally, the actual work is done. A number of imaginary radar scenes are constructed by mathematical modeling. In these scenarios, various waveform design algorithms proposed by this subject are tested by numerical simulation in these scenes. The results of numerical simulation verify the effectiveness of these algorithms.
【學(xué)位授予單位】：哈爾濱工業(yè)大學(xué)
【學(xué)位級別】：博士
【學(xué)位授予年份】：2014
【分類號】：TN974

【參考文獻(xiàn)】

相關(guān)期刊論文 前7條

1 盧中昊;;P波段瞬態(tài)極化雷達(dá)射頻系統(tǒng)研制及其性能測試[J];電子測量與儀器學(xué)報;2009年S1期

2 李棉全;馬梁;李永禎;劉勇;王雪松;;瞬態(tài)極化雷達(dá)接收濾波器的優(yōu)化設(shè)計[J];電子學(xué)報;2010年12期

3 莊珊娜;賀亞鵬;朱曉華;;低距離旁瓣稀疏頻譜波形相位編碼設(shè)計[J];電子與信息學(xué)報;2012年05期

4 李風(fēng)從;趙宜楠;喬曉林;;抑制特定區(qū)間距離旁瓣的恒模波形設(shè)計方法[J];電子與信息學(xué)報;2013年03期

5 李風(fēng)從;趙宜楠;喬曉林;;零自相關(guān)區(qū)相位編碼波形設(shè)計[J];電子學(xué)報;2013年12期

6 趙宜楠;龐曉宇;王軍;周志權(quán);;嚴(yán)重拖尾雜波中距離擴(kuò)展目標(biāo)的自適應(yīng)極化檢測[J];哈爾濱工業(yè)大學(xué)學(xué)報;2014年04期

7 Yinan Zhao;Fengcong Li;Xiaolin Qiao;;Knowledge-based adaptive polarimetric detection in heterogeneous clutter[J];Journal of Systems Engineering and Electronics;2014年03期

，

本文編號：1977747