發(fā)布時(shí)間：2018-06-08 07:53

  本文選題：遺傳算法 + 陣列優(yōu)化��； 參考：《哈爾濱工業(yè)大學(xué)》2014年碩士論文

【摘要】：與相控陣?yán)走_(dá)相比，多輸入多輸出(MIMO)雷達(dá)發(fā)射在采用正交波形時(shí)，雖然能夠得到波形分集增益，但同時(shí)也損失了相干處理增益。針對(duì)這一問(wèn)題，近幾年，國(guó)外一些學(xué)者試圖通過(guò)結(jié)合相控陣?yán)走_(dá)和多輸入多輸出雷達(dá)各自的優(yōu)點(diǎn)形成一種新的雷達(dá)體制，即混合MIMO-相控陣?yán)走_(dá)。 混合MIMO-相控陣?yán)走_(dá)就是把發(fā)射天線陣元按一定方式劃分形成多個(gè)子陣，每個(gè)子陣發(fā)射相互正交的波形，通過(guò)設(shè)計(jì)每個(gè)子陣的加權(quán)矢量使天線在空間形成波束，并且每個(gè)子陣可以構(gòu)成多輸入多輸出雷達(dá)模型。利用子陣內(nèi)波形的相關(guān)性和子陣間波形的正交性來(lái)同時(shí)獲取相干處理增益和波形分集增益。 本文采用發(fā)射和接收使用同一平面陣的陣列結(jié)構(gòu)，其發(fā)射端不同子陣間發(fā)射正交信號(hào)而子陣內(nèi)部發(fā)射相干信號(hào)，且在發(fā)射端和接收端均使用子陣級(jí)信號(hào)處理方法，從而實(shí)現(xiàn)相控陣和多輸入多輸出雷達(dá)的統(tǒng)一。 本文主要研究的內(nèi)容有：混合MIMO-相控陣系統(tǒng)模型的建立，推導(dǎo)出了混合MIMO-相控陣?yán)走_(dá)的發(fā)射、接收和波形分集波束方向圖；混合MIMO-相控陣?yán)走_(dá)陣列的編解碼方案和子陣劃分的約束條件；子陣優(yōu)化的目標(biāo)函數(shù)和適應(yīng)度函數(shù)；利用多目標(biāo)遺傳算法進(jìn)行子陣結(jié)構(gòu)優(yōu)化設(shè)計(jì)。 混合MIMO-相控陣系統(tǒng)模型的建立在相控陣和MIMO雷達(dá)的基礎(chǔ)之上的,我們從發(fā)射信號(hào)-目標(biāo)回波-接收信號(hào)處理這一流程分別介紹傳統(tǒng)的相控陣和MIMO雷達(dá)，然后推導(dǎo)二維子陣級(jí)MIMO-相控陣混合系統(tǒng)的信號(hào)處理流程的模型。 編解碼方案，采用染色體編碼，使平面陣列結(jié)構(gòu)編碼成一組供多目標(biāo)進(jìn)化算法處理的向量形式，并且為了使陣列結(jié)構(gòu)相對(duì)集中設(shè)置了約束條件：陣元不重疊準(zhǔn)則和滿布準(zhǔn)則等。 本文設(shè)置了多個(gè)目標(biāo)函數(shù)，它們體現(xiàn)了波形分集性能、相干處理增益性能、發(fā)射/接收方向圖的性能指標(biāo)，利用適應(yīng)度函數(shù)對(duì)目標(biāo)函數(shù)值進(jìn)行處理，，適應(yīng)度函數(shù)設(shè)計(jì)為要解決問(wèn)題的函數(shù)的負(fù)值。 多目標(biāo)遺傳算法進(jìn)行子陣結(jié)構(gòu)優(yōu)化設(shè)計(jì)，本文采用兩種多目標(biāo)遺傳算法：Pareto秩和系數(shù)加權(quán)法，分別對(duì)目標(biāo)函數(shù)進(jìn)行優(yōu)化，比較兩種方法的優(yōu)化情況，最后得到陣列子陣劃分均勻相對(duì)集中的最優(yōu)陣列結(jié)構(gòu)。
[Abstract]:Compared with phased array radar, multi-input and multi-output MIMO-radar transmit with orthogonal waveform, although it can obtain waveform diversity gain, but it also loses coherent processing gain. To solve this problem, in recent years, some foreign scholars have tried to form a new radar system by combining the advantages of phased array radar and multi-input multi-output radar. That is, hybrid MIMO-phased array radar. The hybrid MIMO-phased array radar is to divide the antenna array elements into several sub-arrays in a certain way, and each sub-array transmits orthogonal waveforms. By designing the weighted vector of each subarray, the antenna is beam-forming in space, and each subarray can form a multi-input and multi-output radar model. The coherent processing gain and waveform diversity gain are obtained simultaneously by using the correlation of waveforms in subarrays and the orthogonality of waveforms between subarrays. The transmitter transmits orthogonal signals between different subarrays and transmits coherent signals inside the subarrays, and the sub-array level signal processing method is used in both the transmitter and receiver. In order to achieve the unity of phased array and multi-input multi-output radar, the main contents of this paper are as follows: the establishment of hybrid MIMO-phased array system model, the derivation of transmission, reception and waveform diversity beam pattern of hybrid MIMO-phased array radar; The coding and decoding scheme of hybrid MIMO-phased array radar array and the constraint condition of subarray partition, the objective function and fitness function of subarray optimization; Multi-objective genetic algorithm is used to optimize the subarray structure. The hybrid MIMO-phased array system model is based on the phased array and MIMO radar. We introduce the traditional phased array and MIMO radar separately from the process of transmitting signal, target echo and receiving signal. Then we derive the model of signal processing flow of MIMO-phased array hybrid system. Using chromosome coding, the planar array structure is encoded into a set of vector forms for multi-objective evolutionary algorithms. In order to make the array structure relatively centralized, the constraint conditions are set up, such as the nonoverlapping criterion and the full spread criterion, etc. In this paper, several objective functions are set up, which embody the waveform diversity performance, coherent processing gain performance, and so on. The performance index of transmit / receive pattern is processed by fitness function. The fitness function is designed as the negative value of the function to solve the problem. The multi-objective genetic algorithm is used to optimize the subarray structure. In this paper, two kinds of multi-objective genetic algorithm: Pareto rank sum coefficient weighting method are used to optimize the objective function, and the optimization of the two methods is compared. Finally, the optimal array structure with uniform relative concentration is obtained.
【學(xué)位授予單位】：哈爾濱工業(yè)大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2014
【分類號(hào)】：TN958.92

【參考文獻(xiàn)】

相關(guān)期刊論文 前4條

1 朱學(xué)軍,陳彤,薛量,李峻;多個(gè)體參與交叉的Pareto多目標(biāo)遺傳算法[J];電子學(xué)報(bào);2001年01期

2 賴紅松,董品杰,祝國(guó)瑞;求解多目標(biāo)規(guī)劃問(wèn)題的Pareto多目標(biāo)遺傳算法[J];系統(tǒng)工程;2003年05期

3 肇瑩;劉紅星;王仲宇;王元慶;;最近鄰搜索用于分類問(wèn)題的一種改進(jìn)[J];南京大學(xué)學(xué)報(bào)(自然科學(xué)版);2009年04期

4 朱鰲鑫;遺傳算法的適應(yīng)度函數(shù)研究[J];系統(tǒng)工程與電子技術(shù);1998年11期

本文編號(hào)：1995212