实例介绍
【实例简介】
林肯实验室发表的技术报告，有关机载雷达空时自适应处理的，希望对大家有帮助。
MASSACHUSETTS INSTITUTE OF TECHNOLOGY LINCOLN LABORATORY SPACE-TIME ADAPTIVE PROCESSING FORAIRBORNE RADAR WARD Group 102 Accesion Fc NTIs CRA&3 TECHNICAL REPORT 1015 DTIC TAB Justification Distribution/ Availability Codes I 3 DECEMBER 1994 Avail and oi Dist Special A-1 pproved for publie release; distribution is unlimited LEXINGTON MASSACHUSETTS AbSTrACT Future airborne radars will be required to detect targets in an interference background comprised of clutter anld jamming. Space-time adaptive processing (STAP)refers to multidimensional adaptive filtering algorithms that simultaneously combine the signals from the elements of an array antenna and the multiple pulses of a coherent radar waveform, to suppress interference and provide target detec tion. STAP can improve detection of low-velocity targets obscured by mainlobe clutter, detection of targets masked by sidelobe clutter, and detection in combined clutter and jamming environments. This report analyzes a variety of approaches to STaP problem. Optimum, or fully adaptive processing is reviewed. Computa tional complexity and the need to estimate the interference from a limited amount of available data make fully adaptive STAP impractical. As a result, partially adap tive space-time processors are required. A taxonomy of reduced-dimension STAP algorithms is presented where algorithms are classified based on the type of pre processor employed. For example, beamspace algorithms use spatial preprocessing while post-doppler approaches perform temporal( doppler)filtering before adaptive processing. In some cases, the special structure of the clutter can be exploited to design preprocessors yielding minimum clutter rank. For each class, either sample matrix-inversion(SMI)or subspace-based weight computation may be employed. Simulation results are presented to illustrate various performance metrics, includ ing SNR, adapted patterns, minimum detectable velocity, and required degrees of fr eedom A CKNOWLEDGMENTS This report has been influenced by many people within Lincoln Laboratory I would like to thank Ken Senne for the technical guidance and encouragement he has given me throughout this work. Special thanks go also to Dave Martinez for providing me with time, support, and motivation to complete this report Much of my education in the area of space-time adaptive processing is a di- rect result of many enlightening discussions with Allan Steinhardt, Steve Krich, Ed Baranoski. Steve Smith, Ken Teitelbaum, and Dan Marshall. Their contributions are gratefully acknowledged. Early encouragement from Allan Steinhardt and Steve Krich was particularly appreciated. My thanks go again to Allan and ed for their careful reviews of the draft; their comments have greatly improved the presenta tion.Professor Richard Bucy of the University of Southern California also provided helpful comments on portions of this report. Finally, i would like to acknowledge the continued sponsorship of CAPT. ryan Henry of the advanced Research Projects Agency. Also, I thank Jim Hall of the Offce of Naval Research, Nancy MacMeekin of Naval Air Warfare Center, and CAPT. Dale Babin of Naval Air Systems Command for their continued interest in the area of space-time adaptive processing technology. Their support over the last two years is gratefully acknowledged TABLE OF CONTENTS abstract Acknowledgments List ofⅢ lustrations List of Tables 1. INTRODUCTION 2. AIRBORNE ARRAY RADAR SIGNAL ENVIRONMENT 2.1 Introduction 2.2 Radar System Description 7 2.3 Target 12 2.4 Noise 2.5 Jamming 18 2.6 Clutter 20 2.7 Range -Doppler View of the Interference 2.8 Summary 51 3. SPACE-TIME PROCESSING FUNDAMENTALS 53 3.1 Introduction 53 3.2 General Architecture and Assumptions 3.3 Fully Adaptive STAP 57 3.4 STAP Performance Metrics and Fully adaptive Performance 59 3.5 Sample Support and Smi 3.6 Computational Complexity 4. PARTIALLY ADAPTIVE STAP 81 4.1 Introduction 81 4.2 A Generic Architecture 81 4.3 Exploiting Low- Rank Interference 4.4 Application tO STAP 88 5. ELEMENT-SPACE STAP 95 5.1 Introduction 95 5.2 Element-Space Pre- Doppler STAP 95 5.3 Element-Space Post-Doppler TABLE OF CONTENTS (Continued 6. BEAMSPACE STAP 135 6.1 Introduction 135 6.2 Beamspace Pre-Doppler STAP 136 6.3 Beam-Space Post-Doppler STaP 153 7. ADDITIONAL PERFORMANCE RESULTS 169 8. SUMMARY AND FUTURE WORK 185 APPENDIX A-PROOF OF BRENNANS RULE 187 APPENDIX B- PROOF OF THEOREM 2 191 APPENDIX C- DISPLACED PHASE CENTER ANTENNA (DPCA) PROCESSING 195 REFERENCES 20 LIST OF ILLUSTRATIONS Fig age ace-time adaptive processing. (a) The power spectral density of the in terference(clutter and jamming) seen by an airborne radar(b) Example adapted response 3 (a) Platform geometry.(b )Top view. Processing for each array element channel. 10 4 The radar CPi datacube A ring of ground clutter for a fixed range 21 The B=1 clutter ridge. The PRF is 300 IIz.(a) Doppler frequency vS. sin p.(b) Normalized Doppler vS, spatial frequenc Clutter loci for different velocities.(a) Stationary platform, va =0.(b Doppler unambiguous, A = 0.5.(c) Doppler ambiguous, B= 1.5.(d Doppler ambiguouS, 6=2.67 27 Effective array position for successive pulses of a CPI N=4,M=3,B=1 Clutter observations are repeated by different elements on different pulses The element pattern. A -30-dB backlobe level is assumed 32 Example scenario: received CNR per column as a function of azimuth. 35 11 Illustrating Brennan's rule: clutter eigenspectra for the example radar sys tem with different platform velocities. 35 12 array geometry with velocity misalignment 13 Example clutter ridges with velocity misalignment,forβ=1.(a)φa=0° (b)φa=10°.(c)pa=45°.(d)φa=90 Example clutter ridges with velocity misalignment, for Doppler-ambiguous clutter:.(a)=26,oa=10°.(b)6=2.67,中a=60° 41 Clutter eigenspectra for different misalignment angles Clutter eigenspectra with misalignment for different backlobe levels 43 17 Clutter eigenspectra for different levels of iCm 47 8 Range-Doppler view of interference, for the case of a low -PRF waveform The clutter is ambiguous in Doppler and unambiguous in rango 50 ⅸx LIST OF ILLUSTRATIONS Continued figure Page Range-Doppler view of interference, for a high-PRF waveform. The clutter is unambiguous in Doppler but ambiguous in range 50 20 A general block diagram for a spacc-time processor 4 21 Data-domain view of space-time adaptive processing 55 Fully adaptive space-time processing 23 Example scenario: optimum fully adaptive STAP. (a)Adapted pattern.(b Principal cuts at target azimuth and Doppler 63 24 Example scenario: tapered fully adaptive STAP. (a) Adapted pattern.(b) Principal plane cuts at target azimuth and Doppler 25 SINR for the optimum and tapered fully adaptive STAPs 68 26 SINR for the optimum and tapered fully adaptive STAPs, including Doppler straddling losses 69 27 SINR loss for the fully adaptive stap 28 SINR improvement factor for the optiMum and tapered fully adaptive STAP. 29 Expected SINR loss for SMI with matched steering vector 79 30 A generic partially adaptive STAP architecturc 82 31(a)Reduced dimension adaptive processing with low-rank interference. (b) Whitening filter viewpoint 86 32 Another block diagram for reduced-dinenision space-time adaptive processing. 92 33 A taxonomy of reduced-dimension STAP algorithms, classified by the type of nonadaptive transformation applied to the CPI data 93 34 Element-space pre-Doppler STAP(a) Sub-CPI processing.(b) Full CPI pr rocessing 97 35 I=2 element-space pre -Doppler STAP.(a)Sub- CPI adapted patterN(b) Principal cuts in angle and Doppler 103 Sub-CPi weight vector decomposed into spatial beamformer components, illustrating the DPCA effect 106 LIST OF ILLUSTRATIONS (Continued) figure P age Composite pattern for Doppler bin 5.(a) Full pattern(b) Principal cuts in angle and Doppler 107 38 SINR loss for element space pre-Doppler. Zero intrinsic clutter motion 109 Block diagram of post-Doppler adaptive beamforming(factored post- Doppler). 112 Example performance for Doppler bin 6(100 Hz) with 40-dB Chebyshev Doppler filters.(a) Clutter power spectral density.(b )Adapted pattern Example performance for Doppler bin 6(100 Hz) with 80-dB Chebyshev Doppler filters.(a) Clutter power spectral density.(b) Adapted pattern 115 42 SINR loss as a function of Doppler filter sidelobe level 117 Block diagram for multiwindow post-Doppler STAP 120 44 Multiwindow post-Doppler form of Brennans rule. Example for N=4, M 3, K=2, and 6= 1. The grouping of pulses for the two filters satisfies Theorem 3 124 45 Block diagram for PRI-staggered post-Doppler STAP 126 46 Block diagram for adjacent-bin post-Doppler Stap 127 Clutter eigenspectra for multiwindow post-Doppler approaches with K 2.(a)PRI-staggered post-Doppler, covariance matrix for Doppler bin 0 (b) PRI-staggered post-Doppler, Doppler bin 3.(c)PRI-staggered post Adjacent-bin, Doppler bin 3.(f) Adjacent-bin, Doppler bin o er bin 0.(e) Doppler, Doppler bin 9.(d)adjacent-bin post-Doppler, Doppler bin 0(e 129 48 SINR loss performance for PRI-staggered and adjacent-bin post-Doppler STAP.(a) Untapered(uniform)Doppler filters(b)30-dB Chebyshev Doppler filters.(c60-dB Chebyshev Doppler filters. (d)90-dB Chebyshev Doppler filters 133 49 Beamspace pre-Doppler STAP(a) Sub-CPI processing. (b) Full CPI processing. 137 50 Beamspa.ce version of Brennan's rule for the case N=4, Kt=3, Ks=2 and b=1 142 【实例截图】
【核心代码】
林肯实验室发表的技术报告，有关机载雷达空时自适应处理的，希望对大家有帮助。
MASSACHUSETTS INSTITUTE OF TECHNOLOGY LINCOLN LABORATORY SPACE-TIME ADAPTIVE PROCESSING FORAIRBORNE RADAR WARD Group 102 Accesion Fc NTIs CRA&3 TECHNICAL REPORT 1015 DTIC TAB Justification Distribution/ Availability Codes I 3 DECEMBER 1994 Avail and oi Dist Special A-1 pproved for publie release; distribution is unlimited LEXINGTON MASSACHUSETTS AbSTrACT Future airborne radars will be required to detect targets in an interference background comprised of clutter anld jamming. Space-time adaptive processing (STAP)refers to multidimensional adaptive filtering algorithms that simultaneously combine the signals from the elements of an array antenna and the multiple pulses of a coherent radar waveform, to suppress interference and provide target detec tion. STAP can improve detection of low-velocity targets obscured by mainlobe clutter, detection of targets masked by sidelobe clutter, and detection in combined clutter and jamming environments. This report analyzes a variety of approaches to STaP problem. Optimum, or fully adaptive processing is reviewed. Computa tional complexity and the need to estimate the interference from a limited amount of available data make fully adaptive STAP impractical. As a result, partially adap tive space-time processors are required. A taxonomy of reduced-dimension STAP algorithms is presented where algorithms are classified based on the type of pre processor employed. For example, beamspace algorithms use spatial preprocessing while post-doppler approaches perform temporal( doppler)filtering before adaptive processing. In some cases, the special structure of the clutter can be exploited to design preprocessors yielding minimum clutter rank. For each class, either sample matrix-inversion(SMI)or subspace-based weight computation may be employed. Simulation results are presented to illustrate various performance metrics, includ ing SNR, adapted patterns, minimum detectable velocity, and required degrees of fr eedom A CKNOWLEDGMENTS This report has been influenced by many people within Lincoln Laboratory I would like to thank Ken Senne for the technical guidance and encouragement he has given me throughout this work. Special thanks go also to Dave Martinez for providing me with time, support, and motivation to complete this report Much of my education in the area of space-time adaptive processing is a di- rect result of many enlightening discussions with Allan Steinhardt, Steve Krich, Ed Baranoski. Steve Smith, Ken Teitelbaum, and Dan Marshall. Their contributions are gratefully acknowledged. Early encouragement from Allan Steinhardt and Steve Krich was particularly appreciated. My thanks go again to Allan and ed for their careful reviews of the draft; their comments have greatly improved the presenta tion.Professor Richard Bucy of the University of Southern California also provided helpful comments on portions of this report. Finally, i would like to acknowledge the continued sponsorship of CAPT. ryan Henry of the advanced Research Projects Agency. Also, I thank Jim Hall of the Offce of Naval Research, Nancy MacMeekin of Naval Air Warfare Center, and CAPT. Dale Babin of Naval Air Systems Command for their continued interest in the area of space-time adaptive processing technology. Their support over the last two years is gratefully acknowledged TABLE OF CONTENTS abstract Acknowledgments List ofⅢ lustrations List of Tables 1. INTRODUCTION 2. AIRBORNE ARRAY RADAR SIGNAL ENVIRONMENT 2.1 Introduction 2.2 Radar System Description 7 2.3 Target 12 2.4 Noise 2.5 Jamming 18 2.6 Clutter 20 2.7 Range -Doppler View of the Interference 2.8 Summary 51 3. SPACE-TIME PROCESSING FUNDAMENTALS 53 3.1 Introduction 53 3.2 General Architecture and Assumptions 3.3 Fully Adaptive STAP 57 3.4 STAP Performance Metrics and Fully adaptive Performance 59 3.5 Sample Support and Smi 3.6 Computational Complexity 4. PARTIALLY ADAPTIVE STAP 81 4.1 Introduction 81 4.2 A Generic Architecture 81 4.3 Exploiting Low- Rank Interference 4.4 Application tO STAP 88 5. ELEMENT-SPACE STAP 95 5.1 Introduction 95 5.2 Element-Space Pre- Doppler STAP 95 5.3 Element-Space Post-Doppler TABLE OF CONTENTS (Continued 6. BEAMSPACE STAP 135 6.1 Introduction 135 6.2 Beamspace Pre-Doppler STAP 136 6.3 Beam-Space Post-Doppler STaP 153 7. ADDITIONAL PERFORMANCE RESULTS 169 8. SUMMARY AND FUTURE WORK 185 APPENDIX A-PROOF OF BRENNANS RULE 187 APPENDIX B- PROOF OF THEOREM 2 191 APPENDIX C- DISPLACED PHASE CENTER ANTENNA (DPCA) PROCESSING 195 REFERENCES 20 LIST OF ILLUSTRATIONS Fig age ace-time adaptive processing. (a) The power spectral density of the in terference(clutter and jamming) seen by an airborne radar(b) Example adapted response 3 (a) Platform geometry.(b )Top view. Processing for each array element channel. 10 4 The radar CPi datacube A ring of ground clutter for a fixed range 21 The B=1 clutter ridge. The PRF is 300 IIz.(a) Doppler frequency vS. sin p.(b) Normalized Doppler vS, spatial frequenc Clutter loci for different velocities.(a) Stationary platform, va =0.(b Doppler unambiguous, A = 0.5.(c) Doppler ambiguous, B= 1.5.(d Doppler ambiguouS, 6=2.67 27 Effective array position for successive pulses of a CPI N=4,M=3,B=1 Clutter observations are repeated by different elements on different pulses The element pattern. A -30-dB backlobe level is assumed 32 Example scenario: received CNR per column as a function of azimuth. 35 11 Illustrating Brennan's rule: clutter eigenspectra for the example radar sys tem with different platform velocities. 35 12 array geometry with velocity misalignment 13 Example clutter ridges with velocity misalignment,forβ=1.(a)φa=0° (b)φa=10°.(c)pa=45°.(d)φa=90 Example clutter ridges with velocity misalignment, for Doppler-ambiguous clutter:.(a)=26,oa=10°.(b)6=2.67,中a=60° 41 Clutter eigenspectra for different misalignment angles Clutter eigenspectra with misalignment for different backlobe levels 43 17 Clutter eigenspectra for different levels of iCm 47 8 Range-Doppler view of interference, for the case of a low -PRF waveform The clutter is ambiguous in Doppler and unambiguous in rango 50 ⅸx LIST OF ILLUSTRATIONS Continued figure Page Range-Doppler view of interference, for a high-PRF waveform. The clutter is unambiguous in Doppler but ambiguous in range 50 20 A general block diagram for a spacc-time processor 4 21 Data-domain view of space-time adaptive processing 55 Fully adaptive space-time processing 23 Example scenario: optimum fully adaptive STAP. (a)Adapted pattern.(b Principal cuts at target azimuth and Doppler 63 24 Example scenario: tapered fully adaptive STAP. (a) Adapted pattern.(b) Principal plane cuts at target azimuth and Doppler 25 SINR for the optimum and tapered fully adaptive STAPs 68 26 SINR for the optimum and tapered fully adaptive STAPs, including Doppler straddling losses 69 27 SINR loss for the fully adaptive stap 28 SINR improvement factor for the optiMum and tapered fully adaptive STAP. 29 Expected SINR loss for SMI with matched steering vector 79 30 A generic partially adaptive STAP architecturc 82 31(a)Reduced dimension adaptive processing with low-rank interference. (b) Whitening filter viewpoint 86 32 Another block diagram for reduced-dinenision space-time adaptive processing. 92 33 A taxonomy of reduced-dimension STAP algorithms, classified by the type of nonadaptive transformation applied to the CPI data 93 34 Element-space pre-Doppler STAP(a) Sub-CPI processing.(b) Full CPI pr rocessing 97 35 I=2 element-space pre -Doppler STAP.(a)Sub- CPI adapted patterN(b) Principal cuts in angle and Doppler 103 Sub-CPi weight vector decomposed into spatial beamformer components, illustrating the DPCA effect 106 LIST OF ILLUSTRATIONS (Continued) figure P age Composite pattern for Doppler bin 5.(a) Full pattern(b) Principal cuts in angle and Doppler 107 38 SINR loss for element space pre-Doppler. Zero intrinsic clutter motion 109 Block diagram of post-Doppler adaptive beamforming(factored post- Doppler). 112 Example performance for Doppler bin 6(100 Hz) with 40-dB Chebyshev Doppler filters.(a) Clutter power spectral density.(b )Adapted pattern Example performance for Doppler bin 6(100 Hz) with 80-dB Chebyshev Doppler filters.(a) Clutter power spectral density.(b) Adapted pattern 115 42 SINR loss as a function of Doppler filter sidelobe level 117 Block diagram for multiwindow post-Doppler STAP 120 44 Multiwindow post-Doppler form of Brennans rule. Example for N=4, M 3, K=2, and 6= 1. The grouping of pulses for the two filters satisfies Theorem 3 124 45 Block diagram for PRI-staggered post-Doppler STAP 126 46 Block diagram for adjacent-bin post-Doppler Stap 127 Clutter eigenspectra for multiwindow post-Doppler approaches with K 2.(a)PRI-staggered post-Doppler, covariance matrix for Doppler bin 0 (b) PRI-staggered post-Doppler, Doppler bin 3.(c)PRI-staggered post Adjacent-bin, Doppler bin 3.(f) Adjacent-bin, Doppler bin o er bin 0.(e) Doppler, Doppler bin 9.(d)adjacent-bin post-Doppler, Doppler bin 0(e 129 48 SINR loss performance for PRI-staggered and adjacent-bin post-Doppler STAP.(a) Untapered(uniform)Doppler filters(b)30-dB Chebyshev Doppler filters.(c60-dB Chebyshev Doppler filters. (d)90-dB Chebyshev Doppler filters 133 49 Beamspace pre-Doppler STAP(a) Sub-CPI processing. (b) Full CPI processing. 137 50 Beamspa.ce version of Brennan's rule for the case N=4, Kt=3, Ks=2 and b=1 142 【实例截图】
【核心代码】
标签：
好例子网口号：伸出你的我的手 — 分享！
小贴士
感谢您为本站写下的评论，您的评论对其它用户来说具有重要的参考价值，所以请认真填写。
- 类似“顶”、“沙发”之类没有营养的文字，对勤劳贡献的楼主来说是令人沮丧的反馈信息。
- 相信您也不想看到一排文字/表情墙，所以请不要反馈意义不大的重复字符，也请尽量不要纯表情的回复。
- 提问之前请再仔细看一遍楼主的说明，或许是您遗漏了。
- 请勿到处挖坑绊人、招贴广告。既占空间让人厌烦，又没人会搭理，于人于己都无利。
关于好例子网
本站旨在为广大IT学习爱好者提供一个非营利性互相学习交流分享平台。本站所有资源都可以被免费获取学习研究。本站资源来自网友分享，对搜索内容的合法性不具有预见性、识别性、控制性，仅供学习研究，请务必在下载后24小时内给予删除，不得用于其他任何用途，否则后果自负。基于互联网的特殊性，平台无法对用户传输的作品、信息、内容的权属或合法性、安全性、合规性、真实性、科学性、完整权、有效性等进行实质审查；无论平台是否已进行审查，用户均应自行承担因其传输的作品、信息、内容而可能或已经产生的侵权或权属纠纷等法律责任。本站所有资源不代表本站的观点或立场，基于网友分享，根据中国法律《信息网络传播权保护条例》第二十二与二十三条之规定，若资源存在侵权或相关问题请联系本站客服人员，点此联系我们。关于更多版权及免责申明参见 版权及免责申明
网友评论
我要评论