本文選題：DOA跟蹤　切入點(diǎn)：稀疏重構(gòu)　出處：《電子科技大學(xué)》2014年碩士論文　論文類型：學(xué)位論文

【摘要】：在移動(dòng)通信中,為了進(jìn)一步提高系統(tǒng)容量和鏈路質(zhì)量,基于空分復(fù)用的高精度的DOA跟蹤成為研究熱點(diǎn)。高精度的DOA跟蹤在個(gè)人定位業(yè)務(wù)、戰(zhàn)場移動(dòng)通信、雷達(dá)、聲納、地震以及語音信號(hào)處理等領(lǐng)域中都得到廣泛的關(guān)注。因此,高精度的DOA跟蹤有非常重要的意義和廣闊的應(yīng)用前景。目前的高精度DOA跟蹤在國內(nèi)外已經(jīng)出現(xiàn)了大量理論和技術(shù)研究,而傳統(tǒng)的DOA跟蹤方法仍存在較多限制,所以進(jìn)一步提高DOA跟蹤精度具有非常重要的意義。本文圍繞高精度DOA跟蹤算法展開研究,主要工作如下:1、首先對(duì)傳統(tǒng)的MUSIC方法的DOA估計(jì)進(jìn)行改進(jìn),使之應(yīng)用于DOA跟蹤,再與子空間DOA跟蹤的另一種基于URV-ESPRIT方法的算法進(jìn)行仿真對(duì)比分析。分析結(jié)果表明:基于MUSIC的DOA跟蹤算法跟蹤精度較高,但是不能跟蹤相干源,且運(yùn)算量大;基于URV-ESPRIT方法運(yùn)算量相對(duì)較小,但同樣不適用于處理相干源和運(yùn)動(dòng)軌跡接近的情況;2、給出了基于PASTd的DOA跟蹤方法和基于FDPM方法的DOA跟蹤方法的仿真對(duì)比分析。分析結(jié)果表明:基于PASTd的DOA跟蹤方法跟蹤精度較高、計(jì)算量小,但當(dāng)目標(biāo)運(yùn)動(dòng)軌跡較近時(shí)性能明顯惡化;基于FDPM方法在處理運(yùn)動(dòng)軌跡較近的情況時(shí)跟蹤性能會(huì)隨著接近度下降;兩種算法在處理信號(hào)源相干的情況時(shí)跟蹤性能有所下降;3、在上述研究基礎(chǔ)上,將壓縮感知理論引入DOA跟蹤中,給出了一種基于壓縮感知的DOA跟蹤方法,該方法對(duì)數(shù)據(jù)進(jìn)行加窗和SVD處理后,在稀疏理論框架下,對(duì)DOA的變化量利用偏離網(wǎng)格的貝葉斯稀疏重構(gòu)方法進(jìn)行估計(jì)。仿真結(jié)果表明該方法比傳統(tǒng)的基于子空間的DOA跟蹤方法的精度更高、分辨率更高,且在信號(hào)源相干或者運(yùn)動(dòng)軌跡較近的情況下也具有良好的跟蹤性能。本文在高精度DOA跟蹤算法的基礎(chǔ)上,給出傳統(tǒng)子空間方法和基于子空間的快速跟蹤算法的仿真對(duì)比分析,討論各類方法的優(yōu)勢和劣勢。同時(shí)給出了一種基于壓縮感知的DOA跟蹤方法,并通過仿真實(shí)驗(yàn)對(duì)比分析證明這些高精度的DOA跟蹤算法研究有著重要的意義和廣闊的前景。
[Abstract]:In mobile communication, in order to further improve system capacity and link quality, high precision DOA tracking based on space division multiplexing has become a research hotspot. High precision DOA tracking in personal positioning services, battlefield mobile communications, radar, sonar, Earthquake and speech signal processing have received wide attention. High precision DOA tracking has very important significance and broad application prospect. At present, there have been a lot of theoretical and technical research on high precision DOA tracking at home and abroad, but there are still many limitations in traditional DOA tracking methods. Therefore, it is very important to improve the tracking accuracy of DOA. This paper focuses on the high-precision DOA tracking algorithm. The main work is as follows: 1. Firstly, the DOA estimation of the traditional MUSIC method is improved to be applied to DOA tracking. The simulation results show that the tracking accuracy of DOA tracking algorithm based on MUSIC is high, but it can not track coherent source, and the computation is heavy. Based on the URV-ESPRIT method, the computation cost is relatively small. However, it is also not suitable to deal with the case where coherent source and motion trajectory are close to each other. The simulation analysis of DOA tracking method based on PASTd and DOA tracking method based on FDPM method is presented. The results show that the DOA tracking method based on PASTd is based on PASTd. The tracking accuracy of the method is high, The computational complexity is small, but the performance of the target is obviously deteriorated when the moving track is close to the target, and the tracking performance will decrease with the approaching degree when the FDPM method is used to deal with the near motion trajectory. The tracking performance of the two algorithms in the case of coherent signal source is degraded. On the basis of the above research, the compression sensing theory is introduced into DOA tracking, and a DOA tracking method based on compressed sensing is proposed. After windowing and SVD processing, the method is based on sparse theory. The variation of DOA is estimated by Bayesian sparse reconstruction method which deviates from the grid. The simulation results show that the proposed method has higher accuracy and higher resolution than the traditional subspace-based DOA tracking method. In this paper, based on the high precision DOA tracking algorithm, the traditional subspace method and the fast tracking algorithm based on subspace are compared with each other. This paper discusses the advantages and disadvantages of various methods, and presents a DOA tracking method based on compression sensing. The simulation results show that these high-precision DOA tracking algorithms have important significance and broad prospects.
【學(xué)位授予單位】：電子科技大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2014
【分類號(hào)】：TN911.7

【參考文獻(xiàn)】

相關(guān)期刊論文 前4條

1 周云鐘,陳天麒,黃香馥;高分辨率信號(hào)DOA估計(jì)的跟蹤算法[J];電子科學(xué)學(xué)刊;1997年03期

2 葉中付,吳濤;基于相控陣單通道接收機(jī)高分辨DOA估計(jì)的跟蹤算法[J];電子科學(xué)學(xué)刊;2000年02期

3 淦華東,李志舜,李樂,蘇，

本文編號(hào)：1588040