本文選題：MIMO雷達(dá) + 長(zhǎng)時(shí)間積累��； 參考：《北京理工大學(xué)》2015年博士論文

【摘要】：多輸入多輸出雷達(dá)(MIMO radar,Multiple-Input Multiple-Output radar)利用多個(gè)發(fā)射天線發(fā)射信號(hào)和多個(gè)接收天線接收目標(biāo)回波信號(hào)進(jìn)行處理,并通過(guò)分集增益獲得了一些優(yōu)良特性。這些優(yōu)點(diǎn)使其獲得了良好的應(yīng)用前景。本文從以下幾個(gè)方面進(jìn)行了研究。其一,研究了MIMO雷達(dá)的長(zhǎng)時(shí)間積累算法。MIMO雷達(dá)進(jìn)行長(zhǎng)時(shí)間積累的優(yōu)勢(shì)在于它可以進(jìn)行幾百以至于上千個(gè)積累脈沖數(shù)的積累而不需要考慮跨波束的問(wèn)題。由于它是通過(guò)計(jì)算來(lái)形成發(fā)射波束和接收波束,發(fā)射波束為多個(gè)寬波束,甚至可以充滿整個(gè)空間以凝視方式工作,而無(wú)需波束掃描。在非相參積累方面,主要針對(duì)Hough變換進(jìn)行了改進(jìn),提出了快速方法和機(jī)動(dòng)運(yùn)動(dòng)目標(biāo)積累檢測(cè)方法;在相參積累方面對(duì)于包絡(luò)插值移位補(bǔ)償和頻域聯(lián)合補(bǔ)償方法進(jìn)行了改進(jìn),并提出了采用Zoom-FRFT等分級(jí)FRFT進(jìn)行速度和加速度局部細(xì)估的方法。在MIMO雷達(dá)實(shí)現(xiàn)中,綜合空間積累及本文中所提的慢時(shí)間積累方法來(lái)實(shí)現(xiàn)多目標(biāo)情況下的機(jī)動(dòng)弱目標(biāo)的搜索和跟蹤的性能優(yōu)于相控陣體制雷達(dá)。其二,研究了MIMO雷達(dá)基于合成寬帶的高分辨成像技術(shù)。MIMO雷達(dá)進(jìn)行合成寬帶高分辨成像的優(yōu)勢(shì)在于可同時(shí)接收多個(gè)子帶信號(hào)合成寬帶從而提高了成像效率和效果。本文提出了改進(jìn)的頻域合成方法,這種方法不僅適用于小場(chǎng)景也適用于大場(chǎng)景的應(yīng)用。本文還在參數(shù)選擇、柵瓣和旁瓣抑制、幅度和相位校正及運(yùn)動(dòng)補(bǔ)償?shù)葞讉�(gè)方面進(jìn)行了分析,提出了適合于工程應(yīng)用的實(shí)現(xiàn)途徑。在高分辨檢測(cè)方面本文結(jié)合文獻(xiàn)中提出對(duì)于低信噪比下基于神經(jīng)網(wǎng)絡(luò)的檢測(cè)方法。相比已有方法本文中的方法通用性強(qiáng)且適用于工程實(shí)現(xiàn)。其三,研究了基于VPX總線及TMS320C6678多核DSP硬件平臺(tái)的MIMO雷達(dá)信號(hào)處理軟件的高性能實(shí)現(xiàn)問(wèn)題。相比基于CPCI總線和單核DSP平臺(tái)的系統(tǒng)實(shí)現(xiàn),本文中的系統(tǒng)實(shí)現(xiàn)在并行性能、運(yùn)算能力和成本等方面都具有明顯優(yōu)勢(shì)。但是,由于新平臺(tái)自身的復(fù)雜性,且缺乏成功的借鑒經(jīng)驗(yàn),因此在初始開(kāi)發(fā)階段存在很多難點(diǎn)需要攻克。本文在介紹信號(hào)處理系統(tǒng)設(shè)計(jì)、硬件平臺(tái)和支撐軟件設(shè)計(jì)的基礎(chǔ)上對(duì)MIMO雷達(dá)信號(hào)處理軟件設(shè)計(jì)方法和關(guān)鍵問(wèn)題進(jìn)行了深入而細(xì)致的研究并給出了實(shí)現(xiàn)方案和解決方法,并在項(xiàng)目組搭建的MIMO雷達(dá)原理實(shí)驗(yàn)系統(tǒng)的基礎(chǔ)上進(jìn)行了實(shí)驗(yàn)和驗(yàn)證。文中對(duì)實(shí)驗(yàn)?zāi)康摹?shí)驗(yàn)條件、實(shí)驗(yàn)原理進(jìn)行了介紹,并給出了實(shí)驗(yàn)結(jié)果。實(shí)驗(yàn)結(jié)果表明了MIMO體制雷達(dá)的優(yōu)勢(shì)及研究算法的有效性和實(shí)用性,同時(shí)也證明了MIMO雷達(dá)信號(hào)處理軟件并行設(shè)計(jì)的可行性和高效性。
[Abstract]:Multiple-Input Multiple-Output radar is processed by transmitting signals from multiple transmit antennas and receiving echo signals from multiple receiving antennas, and some excellent characteristics are obtained by diversity gain. These advantages make it have a good application prospect. This article has carried on the research from the following several aspects. Firstly, the advantage of long time integration algorithm of MIMO radar is that it can accumulate hundreds or even thousands of integrated pulses without considering the problem of cross beam. Because it is calculated to form transmitting and receiving beams, the transmitting beams are multiple broad beams, which can even fill the entire space to work in a staring manner without beam scanning. In the aspect of non-coherent integration, the paper mainly improves the Hough transform, proposes a fast method and a mobile target accumulation detection method, and improves the envelope interpolation shift compensation and frequency domain joint compensation method in the aspect of coherent integration. A method of local fine estimation of velocity and acceleration using Zoom-FRFT and FRFT is presented. In the implementation of MIMO radar, the performance of the search and tracking of maneuvering weak targets with multiple targets is superior to that of phased array radar by integrating the space accumulation and the slow time accumulation method proposed in this paper. Secondly, the advantages of synthetic wideband high resolution imaging technology based on synthetic wideband for MIMO radar are that it can receive multiple sub-band signals to synthesize broadband at the same time, thus improving the imaging efficiency and effect. In this paper, an improved frequency domain synthesis method is proposed, which is suitable not only for small scenes but also for large scene applications. In this paper, parameters selection, gate lobe and sidelobe suppression, amplitude and phase correction and motion compensation are also analyzed, and a suitable approach for engineering applications is proposed. In the aspect of high resolution detection, a neural network-based detection method for low signal-to-noise ratio (SNR) is proposed. Compared with the existing methods, the method in this paper is universal and suitable for engineering implementation. Thirdly, the high performance realization of MIMO radar signal processing software based on VPX bus and TMS320C6678 multi-core DSP hardware platform is studied. Compared with the system implementation based on CPCI bus and single core DSP platform, the system implementation in this paper has obvious advantages in parallel performance, computing power and cost. However, due to the complexity of the new platform and the lack of successful experience, there are many difficulties in the initial development phase. On the basis of introducing the design of signal processing system, hardware platform and supporting software, the design method and key problems of MIMO radar signal processing software are studied deeply and meticulously in this paper, and the realization scheme and solution are given. Based on the experimental system of MIMO radar principle built by the project team, the experiment and verification are carried out. The purpose, conditions and principle of the experiment are introduced, and the experimental results are given. The experimental results show the advantages of the MIMO radar and the effectiveness and practicability of the algorithm. At the same time, the feasibility and efficiency of the parallel design of the MIMO radar signal processing software are also proved.
【學(xué)位授予單位】：北京理工大學(xué)
【學(xué)位級(jí)別】：博士
【學(xué)位授予年份】：2015
【分類(lèi)號(hào)】：TN957.51
，

本文編號(hào)：1886682