基于稀疏反演的相參捷變頻雷達(dá)信號(hào)處理
發(fā)布時(shí)間：2018-08-06 12:15
【摘要】：相參捷變頻雷達(dá)具有優(yōu)異的低截獲、抗干擾和電磁兼容性能，研究其信號(hào)處理問(wèn)題具有重要的應(yīng)用價(jià)值。論文主要考慮捷變頻雷達(dá)中多目標(biāo)距離速度聯(lián)合估計(jì)問(wèn)題，圍繞在信號(hào)處理中如何利用觀測(cè)場(chǎng)景中稀疏信息的問(wèn)題展開(kāi)研究，取得了以下成果。 推導(dǎo)了捷變頻雷達(dá)脈沖回波的信號(hào)模型，將距離-速度聯(lián)合估計(jì)問(wèn)題建模為線(xiàn)性方程組求解的問(wèn)題，并說(shuō)明該線(xiàn)性方程組是欠定的。采用傳統(tǒng)的匹配濾波處理欠定方程組，會(huì)出現(xiàn)嚴(yán)重的旁瓣平臺(tái)，造成虛警、小目標(biāo)被旁瓣掩蓋等問(wèn)題。由于在同一個(gè)粗分辨單元內(nèi)目標(biāo)個(gè)數(shù)往往較少，捷變頻雷達(dá)觀測(cè)場(chǎng)景呈現(xiàn)出了明顯的稀疏性。壓縮感知算法通過(guò)挖掘場(chǎng)景的稀疏性這一先驗(yàn)信息，能夠有效地求解該線(xiàn)性方程組，從而抑制旁瓣、準(zhǔn)確重建場(chǎng)景。 論證了壓縮感知算法在捷變頻雷達(dá)中的適用性。通過(guò)嚴(yán)格分析捷變頻雷達(dá)中觀測(cè)矩陣的性質(zhì)，定量地給出了關(guān)于場(chǎng)景稀疏度和雷達(dá)參數(shù)的充分條件。當(dāng)滿(mǎn)足該條件時(shí)，利用壓縮感知算法可在無(wú)噪聲的情況下精確重建稀疏場(chǎng)景，，或在有噪聲的情況下穩(wěn)健恢復(fù)場(chǎng)景。仿真和實(shí)測(cè)數(shù)據(jù)也驗(yàn)證了壓縮感知算法在捷變頻雷達(dá)中距離-速度聯(lián)合估計(jì)中的有效性。 解決了壓縮感知算法中的模型失配問(wèn)題。實(shí)際觀測(cè)場(chǎng)景通常是在距離速度連續(xù)二維空間上稀疏。將連續(xù)空間離散化時(shí)，場(chǎng)景中目標(biāo)可能并不在格點(diǎn)上，從而出現(xiàn)模型失配問(wèn)題，導(dǎo)致傳統(tǒng)壓縮感知算法性能下降。提出了基于總體最小二乘的自適應(yīng)匹配追蹤算法來(lái)解決該失配問(wèn)題。算法將格點(diǎn)誤差建模成未知參數(shù)，采用約束總體最小二乘算法自適應(yīng)地估計(jì)格點(diǎn)誤差，并利用估計(jì)結(jié)果相應(yīng)地調(diào)整觀測(cè)矩陣，降低模型失配的影響，從而提高場(chǎng)景估計(jì)的魯棒性。 將認(rèn)知機(jī)制引入捷變頻雷達(dá)中，利用已經(jīng)獲得的場(chǎng)景先驗(yàn)信息，進(jìn)一步提高壓縮感知算法重建場(chǎng)景的精度。提出了最小化重建誤差的Cramer-Rao界(CRB)作為雷達(dá)脈沖載頻的優(yōu)化設(shè)計(jì)準(zhǔn)則，通過(guò)降低CRB使得雷達(dá)觀測(cè)系統(tǒng)能夠提供更多信息，降低不同目標(biāo)回波之間的相互干擾。論文還根據(jù)捷變頻雷達(dá)觀測(cè)矩陣的結(jié)構(gòu)特點(diǎn)，提出了最小化CRB的一種近似準(zhǔn)則，能夠有效地降低計(jì)算量。根據(jù)雷達(dá)不同的潛在需求，提出了序貫和批處理兩種優(yōu)化載頻的工作模式。
[Abstract]:Coherent frequency agile radar has excellent performance of low interception, anti-jamming and electromagnetic compatibility, so it has important application value to study the problem of signal processing. In this paper, the problem of joint range velocity estimation of multi-target in frequency-agility radar is considered. The problem of how to use sparse information in observation scene in signal processing is studied, and the following results are achieved. The signal model of pulse echo of agile frequency conversion radar is derived. The range velocity joint estimation problem is modeled as a linear equation system, and it is shown that the linear equation group is under determined. Using traditional matched filtering to deal with underdetermined equations will lead to serious sidelobe platform resulting in false alarm and small target being covered by sidelobe. Because of the small number of targets in the same coarse resolution unit, the observational scene of frequency agile radar shows obvious sparseness. By mining the prior information of scene sparsity, the compressed perceptual algorithm can solve the linear equations effectively, which can suppress the sidelobe and reconstruct the scene accurately. The applicability of compression sensing algorithm in frequency agility radar is demonstrated. By strictly analyzing the properties of observation matrix in frequency agility radar, the sufficient conditions about scene sparsity and radar parameters are given quantitatively. When this condition is satisfied the sparse scene can be accurately reconstructed without noise or the scene can be recovered stably in the presence of noise by using the compression sensing algorithm. The simulation and measured data also verify the effectiveness of the compression sensing algorithm in the range velocity joint estimation of frequency agility radar. The problem of model mismatch in compressed sensing algorithm is solved. The actual observation scene is usually sparse in the range velocity continuous two-dimensional space. When the continuous space is discretized, the target in the scene may not be on the lattice point, which leads to the model mismatch problem, which leads to the performance degradation of the traditional compression sensing algorithm. An adaptive matching tracking algorithm based on population least squares is proposed to solve the mismatch problem. The lattice error is modeled as an unknown parameter, and the constrained population least square algorithm is used to estimate the lattice error adaptively, and the observation matrix is adjusted accordingly using the estimation results to reduce the effect of model mismatch. In order to improve the robustness of scenario estimation. The cognitive mechanism is introduced into the frequency agility radar and the scene prior information obtained is used to further improve the accuracy of the compressed perceptual algorithm to reconstruct the scene. Cramer-Rao bound (CRB), which minimizes the reconstruction error, is proposed as the optimal design criterion for radar pulse carrier frequency. By reducing the CRB, the radar observation system can provide more information and reduce the mutual interference between the echoes of different targets. According to the structural characteristics of the observational matrix of frequency agile radar, an approximate criterion for minimizing CRB is proposed, which can effectively reduce the computational complexity. According to the different potential requirements of radar, two optimal carrier frequency operation modes, sequential and batch processing, are proposed.
【學(xué)位授予單位】：清華大學(xué)
【學(xué)位級(jí)別】：博士
【學(xué)位授予年份】：2014
【分類(lèi)號(hào)】：TN957.51
本文編號(hào)：2167700
[Abstract]:Coherent frequency agile radar has excellent performance of low interception, anti-jamming and electromagnetic compatibility, so it has important application value to study the problem of signal processing. In this paper, the problem of joint range velocity estimation of multi-target in frequency-agility radar is considered. The problem of how to use sparse information in observation scene in signal processing is studied, and the following results are achieved. The signal model of pulse echo of agile frequency conversion radar is derived. The range velocity joint estimation problem is modeled as a linear equation system, and it is shown that the linear equation group is under determined. Using traditional matched filtering to deal with underdetermined equations will lead to serious sidelobe platform resulting in false alarm and small target being covered by sidelobe. Because of the small number of targets in the same coarse resolution unit, the observational scene of frequency agile radar shows obvious sparseness. By mining the prior information of scene sparsity, the compressed perceptual algorithm can solve the linear equations effectively, which can suppress the sidelobe and reconstruct the scene accurately. The applicability of compression sensing algorithm in frequency agility radar is demonstrated. By strictly analyzing the properties of observation matrix in frequency agility radar, the sufficient conditions about scene sparsity and radar parameters are given quantitatively. When this condition is satisfied the sparse scene can be accurately reconstructed without noise or the scene can be recovered stably in the presence of noise by using the compression sensing algorithm. The simulation and measured data also verify the effectiveness of the compression sensing algorithm in the range velocity joint estimation of frequency agility radar. The problem of model mismatch in compressed sensing algorithm is solved. The actual observation scene is usually sparse in the range velocity continuous two-dimensional space. When the continuous space is discretized, the target in the scene may not be on the lattice point, which leads to the model mismatch problem, which leads to the performance degradation of the traditional compression sensing algorithm. An adaptive matching tracking algorithm based on population least squares is proposed to solve the mismatch problem. The lattice error is modeled as an unknown parameter, and the constrained population least square algorithm is used to estimate the lattice error adaptively, and the observation matrix is adjusted accordingly using the estimation results to reduce the effect of model mismatch. In order to improve the robustness of scenario estimation. The cognitive mechanism is introduced into the frequency agility radar and the scene prior information obtained is used to further improve the accuracy of the compressed perceptual algorithm to reconstruct the scene. Cramer-Rao bound (CRB), which minimizes the reconstruction error, is proposed as the optimal design criterion for radar pulse carrier frequency. By reducing the CRB, the radar observation system can provide more information and reduce the mutual interference between the echoes of different targets. According to the structural characteristics of the observational matrix of frequency agile radar, an approximate criterion for minimizing CRB is proposed, which can effectively reduce the computational complexity. According to the different potential requirements of radar, two optimal carrier frequency operation modes, sequential and batch processing, are proposed.
【學(xué)位授予單位】：清華大學(xué)
【學(xué)位級(jí)別】：博士
【學(xué)位授予年份】：2014
【分類(lèi)號(hào)】：TN957.51
【參考文獻(xiàn)】
相關(guān)期刊論文 前8條
1 劉崢,張守宏;跳頻脈沖雷達(dá)目標(biāo)的運(yùn)動(dòng)參數(shù)估計(jì)[J];電子科學(xué)學(xué)刊;2000年04期
2 周穎;陳永光;王國(guó)玉;汪連棟;;信息化訓(xùn)練場(chǎng)中雷達(dá)電磁兼容問(wèn)題[J];火力與指揮控制;2009年08期
3 ZHAO Yao;FENG Jing;ZHANG BingChen;HONG Wen;WU YiRong;;Current progress in sparse signal processing applied to radar imaging[J];Science China(Technological Sciences);2013年12期
4 ;Improved FOCUSS method for reconstruction of cluster structured sparse signals in radar imaging[J];Science China(Information Sciences);2012年08期
5 劉崢,張守宏;跳頻脈沖ISAR成像運(yùn)動(dòng)補(bǔ)償?shù)淖钚〔ㄐ戊胤椒╗J];西安電子科技大學(xué)學(xué)報(bào);2000年06期
6 劉崢,張守宏;頻率編碼脈沖信號(hào)的模糊函數(shù)與編碼優(yōu)化[J];系統(tǒng)工程與電子技術(shù);1999年11期
7 劉崢,張守宏;跳頻脈沖雷達(dá)目標(biāo)的運(yùn)動(dòng)速度參數(shù)估計(jì)[J];信號(hào)處理;2000年02期
8 劉振;魏璽章;黎湘;;Low sidelobe robust imaging in random frequency-hopping wideband radar based on compressed sensing[J];Journal of Central South University;2013年03期
本文編號(hào)：2167700
本文鏈接：http://sikaile.net/kejilunwen/wltx/2167700.html
最近更新
教材專(zhuān)著
