Principles of Space- Time Adaptive Processing 3rd Edition By Richard Klemm The Institution of Engineering and Technology
Contents Biography Preface to the first edition Preface to the second edition Preface to the third edition xv xvii xxi xxiii 1 Introduction 1 1.1 Preliminary remarks 1 1.1.1 Basics ofmtiradar 3 1.1.2 One-dimensional clutter cancellation 4 1.1.3 Aspects of air-and spaceborne radar 5 1.1.4 Impact of platform motion 6 1.1.5 Some notes on phased array radar 11 1.1.6 Systems and experiments 12 1.1.7 Validity of modeis 14 1.1.8 Historical overview 15 1.2 Radar signal processing tools 17 1.2.1 The optimum processor 17 1.2.2 Orthogonal projection 25 1.2.3 Linear subspace transforms 27 1.2.4 Clutter suppression with digital filters 34 1.2.5 Examples 39 1.2.6 Angle or frequency domain processing 41 1.3 Spectral estimation 44 1.3.1 Signal match (SM) 45 1.3.2 Minimum variance estimator, MVE 46 1.3.3 Maximum entropy method, MEM 47 1.3.4 Orthogonal projection, MUSIC 48 1.3.5 Comparison of spectral estimators 49 1.4 Summary 50
viii Contents 2 Signal and interference modeis 51 2.1 Transmit and receive process 51 2.2 The Doppler effect 52 2.3 Space-time Signals 53 2.3.1 The spatial dimension: array geometry 53 2.3.2 The temporal dimension: pulse trains 55 2.4 Interference 57 2.4.1 Ground clutter 57 2.4.2 Moving clutter 59 2.4.3 Jamming 59 2.4.4 Noise 60 2.5 Decorrelation effects 60 2.5.1 Temporal decorrelation 61 2.5.2 Spatial decorrelation: effect of System bandwidth 63 2.5.3 Doppler spread within ränge gate 64 2.5.4 System Doppler spread 67 2.5.5 Total correlation model 67 2.6 The Standard parameter set 68 2.6.1 Multiple-time around clutter 68 2.6.2 Remark on image quality 69 2.7 Summary 69 3 Properties of airborne clutter 71 3.1 Space-Doppler characteristics 71 3.1.1 Isodops 71 3.1.2 Doppler-azimuth clutter trajectories 73 3.2 The space-time covariance matrix 79 3.2.1 The components 82 3.2.2 The displaced phase centre antenna (DPCA) principle... 86 3.2.3 Eigenspectra 98 3.3 Power spectra 102 3.3.1 Fourier spectra 103 3.3.2 High-resolution spectra 105 3.4 Effect of radar parameters on interference spectra 106 3.4.1 Array orientation 106 3.4.2 Temporal and spatial sampling 107 3.4.3 Decorrelation effects 109 3.4.4 Clutter and jammer spectra 114 3.5 Aspects of adaptive space-time clutter rejection 114 3.5.1 Illustration of the principle 114 3.5.2 Some conclusions 115 3.6 Summary 118
Contents ix 4 Fully adaptive space-time processors 121 4.1 Introduction 121 4.2 General description 122 4.2.1 The Optimum adaptive processor (OAP) 122 4.2.2 The orthogonal projection processor (OPP) 127 4.3 Optimum processing and motion compensation 129 4.3.1 Principle of RF motion compensation 129 4.3.2 Correction patterns 131 4.3.3 Interrelation with the Optimum processor 133 4.4 Influence of radar parameters 139 4.4.1 Transmit beamwidth 139 4.4.2 Array and sample size 141 4.4.3 Sampling effects 142 4.4.4 Influence of the CNR 145 4.4.5 Bandwidth effects 146 4.4.6 Moving clutter 153 4.5 Range-Doppler IF matrix 153 4.6 Summary 155 5 Space-time subspace techniques 159 5.1 Principle of space-time subspace transforms 160 5.2 The auxiliary eigenvector processor (AEP) 163 5.2.1 Comparison with the optimum adaptive processor (OAP)... 165 5.2.2 Reduction of the number of Channels 166 5.2.3 Bandwidth effects 167 5.3 Auxiliary Channel processor (ACP) 170 5.3.1 Comparison with optimum processor 172 5.3.2 Reduction of the number of Channels 172 5.3.3 Bandwidth effects 172 5.4 Other space-time transforms 173 5.4.1 Single auxiliary elements and echo samples transform... 173 5.4.2 Space-time sample subgroups 173 5.4.3 Space-time blocking matrices 174 5.4.4 TheJDL-GLR 174 5.5 Aspects of implementation 175 5.5.1 General properties 175 5.5.2 Auxiliary eigenvector processor 176 5.5.3 Auxiliary Channel processor 177 5.6 Summary 177 6 Spatial transforms for linear arrays 179 6.1 Subarrays 180 6.1.1 Overlapping uniform subarrays (OUS) 181 6.1.2 Effect of subarray displacement 183 6.1.3 Non-uniform subarrays 187 6.2 Auxiliary sensor techniques 191
x Contents 6.2.1 Symmetrie auxiliary sensor configuration (SAS) 191 6.2.2 Bandwidth effects 199 6.2.3 Asymmetrie auxiliary sensor configuration 200 6.2.4 Optimum planar antennas 202 6.3 Other techniques 204 6.3.1 Spatial blocking matrix transform 204 6.3.2 S-A-processing 206 6.3.3 CPCT Processing 209 6.4 Summary 211 7 Adaptive space-time digital filters 213 7.1 Least Squares FIR filters 215 7.1.1 Principle of space-time least Squares FIR filters 215 7.1.2 Füll antennaarray 224 7.1.3 Spatial transforms and FIR nltering 226 7.2 Impact of radar parameters 230 7.2.1 Sample size 230 7.2.2 Decorrelation effects 233 7.2.3 Depth of the clutter noteh 236 7.2.4 Computation of the filter coefficients 236 7.3 Other filter techniques 236 7.3.1 FIR filters for spatial and temporal dimension 236 7.3.2 The projeetion technique 237 7.3.3 Space-time HR filters 238 7.3.4 Adaptive DPCA (ADPCA) 238 7.4 Summary 239 8 Antenna related aspects 241 8.1 Introduction 241 8.2 Non-linear array configurations 242 8.2.1 Circular planar arrays 242 8.2.2 Randomly spaced arrays 257 8.2.3 Conformal arrays 262 8.3 Array coneepts with omnidirectional coverage 264 8.3.1 Four linear arrays 265 8.3.2 Circular ring arrays 265 8.3.3 Horizontal planar arrays 267 8.4 STAP and conventional MTI processing 272 8.4.1 Introduction 272 8.4.2 Linear arrays 273 8.4.3 Circular planar array 275 8.4.4 Volume array 280 8.5 Other antenna related aspects 280 8.5.1 Sparse arrays for spacebased radar 280 8.5.2 Polarisation-space-time processing 284 8.5.3 Radome effects 286
Contents xi 8.5.4 Alternating transmit approach 288 8.6 Summary 288 9 Space-frequency processing 297 9.1 Introduction 297 9.2 The auxiliary space-time Channel processor (ACP) 300 9.3 The Symmetrie auxiliary sensor/echo processor 301 9.3.1 Computing the inverses of the spectral covariance matrices.. 303 9.4 Frequency domain FIR filier (FDFF) 304 9.5 Frequency-dependent spatial processing (FDSP) 306 9.5.1 Spatial blocking matrices 310 9.6 Comparison of processors 310 9.7 Angle-Doppler subgroups 311 9.7.1 General description 311 9.7.2 Comparison angle-doppler subgroup architectures with other techniques 314 9.7.3 Other post-doppler techniques 315 9.8 Summary 315 10 Radar ambiguities 317 10.1 Range ambiguities 318 10.1.1 Multiple-time-around clutter, linear arrays 318 10.1.2 Multiple-time-around clutter, circular planar arrays 326 10.2 Doppler ambiguities 329 10.2.1 Preliminaries 329 10.2.2 Clutter and target modeis 331 10.2.3 Pseudorandom staggering 332 10.2.4 Quadratic staggering 334 10.2.5 Impact of platform acceleration 335 10.2.6 Space-time FIR filter processing 336 10.3 Summary 338 11 STAP under jamming conditions 343 11.1 Introduction 343 11.2 Simultaneous Jammer and clutter cancellation 344 11.2.1 Optimum adaptive processing (OAP) 345 11.2.2 Coherent repeaterjammers 347 11.2.3 Space-time auxiliary Channel processors 347 11.2.4 Spatial auxiliary Channel processors 349 11.3 Circular arrays with subarray processing 352 11.3.1 Adaptive space-time processing versus temporal clutter filtering 352 11.3.2 Two-dimensional arrays in multi-jammer scenarios 353 11.4 Separate jammer and clutter cancellation 356 11.4.1 Optimum jammer cancellation and auxiliary Channel clutter filter357 11.4.2 Jammer and clutter auxiliary Channel Alters cascaded 361 11.5 Jamming in the range-doppler IF matrix 369
xii Contents 11.6 Terrain scatteredjamming 370 11.6.1 Transmit waveform 370 11.6.2 Adaptive multipath cancellation 371 11.7 Summary 373 12 Space-time processing for bistatic radar 377 12.1 Effect of bistatic radar on STAP processing 378 12.1.1 Discussion of the bistatic clutter Doppler 378 12.1.2 Numerical examples 380 12.2 Realistic bistatic geometries (tandem coniiguration) 385 12.2.1 Two aircraft with aligned flight paths 385 12.2.2 Two aircraft with parallel flight paths (horizontal across-track) 386 12.2.3 Two aircraft, transmitter above receiver (vertical across-track) 387 12.2.4 A note on ränge dependence 388 12.3 Ambiguities in bistatic STAP radar 389 12.3.1 Range ambiguities 389 12.3.2 Range and Doppler ambiguities 391 12.4 Use of sparse arrays in bistatic spaceborne GMTI radar 393 12.4.1 Introduction 393 12.4.2 DPCA in bistatic configurations 394 12.4.3 Some numerical examples 395 12.4.4 Comparison with fully filled array 396 12.5 Summary 396 13 Interrelated problems in SAR and ISAR 403 13.1 Clutter rejection for multichannel ISAR 405 13.1.1 Models 406 13.1.2 Space-time FIR fütering 410 13.1.3 Effect of clutter cancellation on ISAR resolution 411 13.2 Jammer nulling for multichannel radar/sar 413 13.2.1 Models 413 13.2.2 Comparison of modeis 416 13.2.3 MV spectra of Jammers and noise 418 13.2.4 Space-TIME FIR filter approach 419 13.2.5 Effect of broadband Jammer cancellation on SAR resolution. 420 13.3 Summary 422 14 Target parameter estimation 425 14.1 CRB for space-time ML estimation 425 14.1.1 Theprinciple 426 14.1.2 The Cramer-Rao bound 428 14.1.3 Some properties of the Cramer-Rao bound 430 14.2 Impact of radar parameters on the CRB 436 14.2.1 Environmental effects 437 14.2.2 Impact of System parameters 439 14.3 Order reducing transform processors 442
Contents xiii 14.3.1 Conventional processing 443 14.3.2 Space-time transforms 444 14.3.3 Spatial transforms 446 14.3.4 Auxiliary sensor/echo processing (ASEP) 447 14.4 Space-time monopulse processing 450 14.4.1 Nickel 's approach 450 14.4.2 Numerical examples 455 14.4.3 Ground target tracking with monopulse radar 459 14.5 Summary 465 15 Influence of the radar equation 477 15.1 Fundamentals 477 15.1.1 From notional radar concepts to realistic Operation 477 15.1.2 The radar equation 478 15.1.3 SNIR and probability of detection 479 15.1.4 Mapping SNIR onto probability of detection 480 15.2 Numerical examples 481 15.2.1 Optimum space-time processing at subarray level 481 15.2.2 Suboptimum space-time processors 485 15.2.3 One-dimensional processing 486 15.3 Summary 486 16 Special aspects of airborne MTI radar 491 16.1 Antenna array errors 491 16.1.1 Tolerances of sensor positions 492 16.1.2 Array Channel errors 495 16.1.3 Channel equalisation 500 16.2 Range dependence of clutter Doppler 505 16.2.1 Impact on adaptation and filtering 505 16.2.2 Doppler compensation 506 16.3 Aspects of implementation 512 16.3.1 Comparison of techniques in terms of computational complexity 512 16.3.2 Comparison of pre-and post-doppler architectures 516 16.3.3 Effect of short-time data processing 518 16.3.4 Inclusion of signal in adaptation 518 16.3.5 Homogeneity of clutter background 521 16.3.6 Non-adaptive space-time filtering 524 16.3.7 Further limitations 525 16.4 Adaptive algorithms 529 16.4.1 Approximations of the optimum processor 529 16.4.2 QR-decomposition 532 16.4.3 Orthogonal projection algorithms 534 16.5 Alternative processor concepts 540 16.5.1 Least Squares predictive transform 540 16.5.2 Direct datadomain (D 3 ) approaches 540 16.5.3 Frequency hopping 542
xiv Contents 16.6 Summary 542 A Sonar applications 545 A.l Introduction 545 A.2 Signal processing in the modal environment 546 A.2.1 Signal modeis 547 A.2.2 Extension to space-time matched field processing 552 A.3 Active sonar application: suppression of reverberation 554 A.4 Estimation of target position and velocity 556 A.4.1 Influence of surface fluctuations 558 A.4.2 Application: a multistatic CW surveillance system 558 A.5 Summary 558 Bibliography 559 Glossary 615 Index 620