Optimum Array Processing Detection Estimation

and Modulation Theory Part IV 1st Edition Harry

L. Van Trees pdf download

https://ebookgate.com/product/optimum-array-processing-detection-
estimation-and-modulation-theory-part-iv-1st-edition-harry-l-van-
trees/

Get Instant Ebook Downloads – Browse at https://ebookgate.com

Instant digital products (PDF, ePub, MOBI) available
Download now and explore formats that suit you...

Optimum Array Processing Part IV of Detection Estimation

and Modulation Theory 1st Edition Harry L. Van Trees

https://ebookgate.com/product/optimum-array-processing-part-iv-of-
detection-estimation-and-modulation-theory-1st-edition-harry-l-van-
trees/
ebookgate.com

Modulation Detection and Coding 1st Edition Tommy Öberg

https://ebookgate.com/product/modulation-detection-and-coding-1st-
edition-tommy-oberg/

ebookgate.com

Handbook on Array Processing and Sensor Networks 1.

Auflage Edition Simon Haykin

https://ebookgate.com/product/handbook-on-array-processing-and-sensor-
networks-1-auflage-edition-simon-haykin/

ebookgate.com

Advances in Multiuser Detection Wiley Series in

Telecommunications and Signal Processing 1st Edition
Michael L. Honig
https://ebookgate.com/product/advances-in-multiuser-detection-wiley-
series-in-telecommunications-and-signal-processing-1st-edition-
michael-l-honig/
ebookgate.com
Enzyme Kinetics and Mechanism Part F Detection and
Characterization of Enzyme Reaction Intermediates 1st
Edition Daniel L. Purich
https://ebookgate.com/product/enzyme-kinetics-and-mechanism-part-f-
detection-and-characterization-of-enzyme-reaction-intermediates-1st-
edition-daniel-l-purich/
ebookgate.com

Modeling Estimation and Optimal Filtering in Signal

Processing 1st Edition Mohamed Najim

https://ebookgate.com/product/modeling-estimation-and-optimal-
filtering-in-signal-processing-1st-edition-mohamed-najim/

ebookgate.com

Female Labor Supply Theory and Estimation James P. Smith

https://ebookgate.com/product/female-labor-supply-theory-and-
estimation-james-p-smith/

ebookgate.com

Periodically Correlated Random Sequences Spectral Theory

and Practice Wiley Series in Probability and Statistics
1st Edition Harry L. Hurd
https://ebookgate.com/product/periodically-correlated-random-
sequences-spectral-theory-and-practice-wiley-series-in-probability-
and-statistics-1st-edition-harry-l-hurd/
ebookgate.com

Probability Concepts and Theory for Engineers 1st Edition

Harry Schwarzlander

https://ebookgate.com/product/probability-concepts-and-theory-for-
engineers-1st-edition-harry-schwarzlander/

ebookgate.com
Optimum Array Processing
Optimum Array Processing
Part IV of Detection, Estimation,
and Modulation Theory

Harry L. Van Trees

WILEY-
INTERSCIENCE
A JOHN WILEY & SONS, INC., PUBLICATION
Designations used by companies to distinguish their products are often
claimed as trademarks. In all instances where John Wiley & Sons, Inc., is
aware of a claim, the product names appear in initial capital or ALL
CAPITAL LETTERS. Readers, however, should contact the appropriate
companies for more complete information regarding trademarks and
registration.

Copyright  2002 by John Wiley & Sons, Inc. All rights reserved.

No part of this publication may be reproduced, stored in a retrieval system

or transmitted in any form or by any means, electronic or mechanical,
including uploading, downloading, printing, decompiling, recording or
otherwise, except as permitted under Sections 107 or 108 of the 1976
United States Copyright Act, without the prior written permission of the
Publisher. Requests to the Publisher for permission should be addressed to
the Permissions Department, John Wiley & Sons, Inc., 605 Third Avenue,
New York, NY 10158-0012, (212) 850-6011, fax (212) 850-6008,
E-Mail: [email protected].

This publication is designed to provide accurate and authoritative

information in regard to the subject matter covered. It is sold with the
understanding that the publisher is not engaged in rendering professional
services. If professional advice or other expert assistance is required, the
services of a competent professional person should be sought.

ISBN 0-471-22110-4

This title is also available in print as ISBN 0-471-09390-4.

For more information about Wiley products, visit our web site at
www.Wiley.com.
To Diane

For her continuing support and

encouragement during the many years
that this book was discussed, researched,
and finally written. More importantly,
for her loyalty, love, and
understanding during a sequence of
challenging periods,

and to

Professor Wilbur Davenport, whose

book introduced me to random
processes and who was a mentor, friend,
and supporter during my career at
Massachusetts Institute of
Technology.
Contents

Preface xix

1 Introduction 1
1.1 Array Processing ......................... 2
1.2 Applications ............................ 6
1.2.1 Radar ........................... 6
1.2.2 Radio Astronomy ..................... 7
1.2.3 Sonar ........................... 8
1.2.4 Communications ..................... 9
1.2.5 Direction Finding ..................... 10
1.2.6 Seismology ........................ 10
1.2.7 Tomography ....................... 11
1.2.8 Array Processing Literature ............... 12
1.3 Organization of the Book .................... 12
1.4 Interactive Study ......................... 14

2 Arrays and Spatial Filters 17

2.1 Introduction ............................ 17
2.2 Frequency-wavenumber Response and Beam Patterns ..... 23
2.3 Uniform Linear Arrays ...................... 37
2.4 Uniformly Weighted Linear Arrays ............... 42
2.4.1 Beam Pattern Parameters ................ 46
2.5 Array Steering .......................... 51
2.6 Array Performance Measures .................. 59
2.6.1 Directivity ........................ 60
2.6.2 Array Gain vs. Spatially White Noise (A,) ...... 63
2.6.3 Sensitivity and the Tolerance Factor .......... 66
2.6.4 Summary ......................... 70
2.7 Linear Apertures ......................... 71
Viii Contents

2.7.1 Frequency-wavenumber Response ............ 71

2.7.2 Aperture Sampling ..................... 74
2.8 Non-isotropic Element Patterns ................ 75
2.9 Summary ............................. 78
2.10 Problems ............................. 79

Synthesis of Linear Arrays and Apertures 90

3.1 Spectral Weighting ........................ 95
3.2 Array Polynomials and the z-Transform ............ 109
3.2.1 z-Transform ........................ 109
3.2.2 Real Array Weights ................... 110
3.2.3 Properties of the Beam Pattern Near a Zero ...... 114
3.3 Pattern Sampling in Wavenumber Space ............ 118
3.3.1 Continuous Aperture ................... 118
3.3.2 Linear Arrays ....................... 120
3.3.3 Discrete Fourier Transform ............... 122
3.3.4 Norms ........................... 126
3.3.5 Summary ......................... 128
3.4 Minimum Beamwidth for Specified Sidelobe Level ....... 128
3.4.1 Introduction ....................... 128
3.4.2 Dolph-Chebychev Arrays ................ 130
3.4.3 Taylor Distribution .................... 143
3.4.4 Villeneuve fi Distribution ................ 147
3.5 Least Squares Error Pattern Synthesis ............. 149
3.6 Minimax Design ......................... 156
3.6.1 Alternation Theorem ...... Y. ........... 159
3.6.2 Parks-McClellan-Rabiner Algorithm .......... 160
3.6.3 Summary ......................... 163
3.7. Null Steering ........................... 165
3.7.1 Null Constraints ..................... 165
3.7.2 Least Squares Error Pattern Synthesis with Nulls ... 166
,3.8 Asymmetric Beams ........................ 173
3.9 Spatially Non-uniform Linear Arrays .............. 178
3.9.1 Introduction ....................... 178
3.9.2 Minimum Redundancy Arrays ............. 179
39.3 Beam Pattern Design Algorithm ............ 183
3.10 Beamspace Processing ...................... 192
3.10.1 Full-dimension Beamspace ................ 192
3.10.2 Reduced-dimension Beamspace ............. 193
3.10.3 Multiple Beam Antennas ................ 200
Contents ix

3.10.4 Summary ......................... 200

3.11 Broadband Arrays ........................ 200
3.12 Summary ............................. 204
3.13 Problems ............................. 207

4 Planar Arrays and Apertures 231

4.1 Rectangular Arrays ........................ 233
4.1.1 Uniform Rectangular Arrays .............. 233
4.1.2 Array Manifold Vector .................. 249
4.1.3 Separable Spectral Weightings ............. 251
4.1.4 2-D z-Transforms ..................... 251
4.1.5 Least Squares Synthesis ................. 253
4.1.6 Circularly Symmetric Weighting and Windows .... 259
4.1.7 Wavenumber Sampling and 2-D DFT ......... 260
4.1.8 Transformations from One Dimension to Two Dimen-
sions ............................ 264
4.1.9 Null Steering ....................... 269
4.1.10 Related Topics ...................... 272
4.2 Circular Arrays .......................... 274
4.2.1 Continuous Circular Arrays (Ring Apertures) ..... 275
4.2.2 Circular Arrays ...................... 280
4.2.3 Phase Mode Excitation Beamformers .......... 284
4.3 Circular Apertures ........................ 289
4.3.1 Separable Weightings .................. 290
4.3.2 Taylor Synthesis for Circular Apertures ........ 294
4.3.3 Sampling the Continuous Distribution ......... 298
4.3.4 Difference Beams ..................... 299
4.3.5 Summary ......................... 304
4.4 Hexagonal Arrays ......................... 305
4.4.1 Introduction ....................... 305
4.4.2 Beam Pattern Design .................. 307
4.4.3 Hexagonal Grid to Rectangular Grid Transformation . 314
4.4.4 Summary ......................... 316
4.5 Nonplanar Arrays ......................... 316
4.5.1 Cylindrical Arrays .................... 317
4.5.2 Spherical Arrays ..................... 320
4.6 Summary ............................. 321
4.7 Problems ............................. 322
X Contents

5 Characterization of Space-time Processes . 332

5.1 Introduction ............................ 332
5.2 Snapshot Models ......................... 333
5.2.1 Frequency-domain Snapshot Models .......... 334
5.2.2 Narrowband Time-domain Snapshot Models ...... 349
5.2.3 Summary ......................... 352
5.3 Space-time Random Processes .................. 353
5.3.1 Second-moment Characterization ............ 353
5.3.2 Gaussian Space-time Processes ............. 359
5.3.3 Plane Waves Propagating in Three Dimensions .... 361
5.3.4 1-D and 2-D Projections ................. 365
5.4 Arrays and Apertures ...................... 369
5.4.1 Arrays ........................... 369
5.4.2 Apertures ......................... 374
5.5 Orthogonal Expansions ..................... 375
5.5.1 Plane-wave Signals .................... 377
5.5.2 Spatially Spread Signals ................. 385
5.5.3 Frequency-spread Signals ................ 390
5.5.4 Closely Spaced Signals .................. 393
5.5.5 Beamspace Processors .................. 393
5.5.6 Subspaces for Spatially Spread Signals ......... 394
5.6 Parametric Wavenumber Models ................ 394
5.6.1 Rational Transfer Function Models ........... 395
5.6.2 Model Relationships ................... 407
5.6.3 Observation Noise ........ .......... 408
2 .

5.6.4 Summary ......................... 414

5.7 Summary ............................. 414
5.8 Problems ............................. 415

6 Optimum Waveform Estimation 428

6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . 428
6.2 Optimum Beamformers . . . . . . . . . . . . . . . . . . . . . 439
6.2.1 Minimum Variance Distortionless Response (MVDR)
Beamformers . . . . . . . . . . . . . . . . . . . . . . . 440
6.2.2 Minimum Mean-Square Error (MMSE) Estimators . . 446
6.2.3 Maximum Signal-to-Noise Ratio (SNR) . . . . . . . . 449
6.2.4 Minimum Power Distortionless Response (MPDR) Beam-
formers . . . . . . . . . . . . . . . . . . . . . . . . . . 451
6.2.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . 452
6.3 Discrete Interference . . . . . . . . . . . . . . . . . . . . . . . 452
Contents Xi

6.3.1 Single Plane-wave Interfering Signal .......... 453

6 3.2 Multiple Plane-wave Interferers 465
63.3 Summary: Discrete Interference .......................... 471
64. Spatially Spread Interference .................. 473
6.4.1 Physical Noise Models .................. 473
6.4.2 ARMA Models ...................... 474
65. Multiple Plane-wave Signals ................... 477
6.5.1 MVDR Beamformer ................... 477
6.5.2 MMSE Processors .................... 485
66. Mismatched MVDR and MPDR Beamformers ......... 488
6.6.1 Introduction ....................... 488
6.6.2 DOA Mismatch ...................... 490
6.6.3 Array Perturbations ................... 501
6.6.4 Diagonal Loading ..................... 505
6.6.5 Summary ......................... 510
67. LCMV and LCMP Beamformers ................ 513
6.7.1 Typical Constraints ................... 514
6.7.2 Optimum LCMV and LCMP Beamformers ...... 526
6.7.3 Generalized Sidelobe Cancellers ............. 528
6.7.4 Performance of LCMV and LCMP Beamformers ... 532
6.7.5 Quiescent Pattern (QP) Constraints .......... 547
6.7.6 Covariance Augmentation ................ 554
6.7.7 Summary ......................... 555
68. Eigenvector Beamformers .................... 556
6.8.1 Principal-component (PC) Beamformers ........ 560
6.8.2 Cross-spectral Eigenspace Beamformers ........ 567
6.8.3 Dominant-mode Rejection Beamformers ........ 569
6.8.4 Summary ......................... 573
69. Beamspace Beamformers ..................... 575
6.9.1 Beamspace MPDR .................... 576
6.9.2 Beamspace LCMP .................... 583
6.9.3 Summary: Beamspace Optimum Processors ...... 585
6.10 Quadratically Constrained Beamformers . . . . . . . . . . . . 585
6.11 Soft-constraint Beamformers . . . . . . . . . . . . . . . . . . 593
6.12 Beamforming for Correlated Signal and Interferences . . . . . 599
6.12.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . 599
6.12.2 MPDR Beamformer: Correlated Signals and Interference600
6.12.3 MMSE Beamformer: Correlated Signals and Interference603
6.12.4 Spatial Smoothing and Forward-Backward Averaging 605
6.12.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . 620
Xii Contents

6.13 Broadband Bearnformers ..................... 621

6.13.1 Introduction ....................... 621
6.13.2 DFT Beamformers .................... 627
6.13.3 Finite impulse response (FIR) Beamformers ...... 647
6.13.4 Summary: Broadband Processing ............ 664
6.14 Summary ............................. 666
6.15 Problems ............................. 669

7 Adaptive Beamformers 710

7.1 Introduction ............................ 710
7.2 Estimation of Spatial Spectral Matrices ............ 712
7.2.1 Sample Spectral Matrices ................ 714
7.2.2 Asymptotic Behavior ................... 717
7.2.3 Forward-Backward Averaging .............. 718
7.2.4 Structured Spectral Matrix Estimation ......... 726
7.2.5 Parametric Spatial Spectral Matrix Estimation .... 726
7.2.6 Singular Value Decomposition .............. 727
7.2.7 Summary ......................... 727
7.3 Sample Matrix Inversion (SMI) ................. 728
7.3.1 SINRsmi Behavior: MVDR and MPDR ........ 731
7.3.2 LCMV and LCMP Beamformers ............ 739
7.3.3 Fixed Diagonal Loading ................. 739
7.3.4 Toeplitz Estimators ................... 751
7.3.5 Summary ......................... 751
7.4 Recursive Least Squares (RLS) ................. 752
7.4.1 Least Squares Formulation ............... 752
7.4.2 Recursive Implement ation ................ 756
7.4.3 Recursive Implementation of LSE Beamformer .... 763
7.4.4 Generalized Sidelobe Canceller ............. 766
7.4.5 Quadratically Constrained RLS ............. 768
7.4.6 Conjugate Symmetric Beamformers .......... 773
7.4.7 Summary ......................... 777
7.5 Efficient Recursive Implementation Algorithms ........ 778
7.5.1 Introduction ....................... 778
7.5.2 QR Decomposition (QRD) ............... 779
7.6 Gradient Algorithms ....................... 789
7.6.1 Introduction ....................... 789
7.6.2 Steepest Descent: MMSE Beamformers ........ 791
7.6.3 Steepest Decent: LCMP Beamformer ......... 799
7.6.4 Summary ......................... 805
Contents Xii

7.7 LMS Algorithms ......................... 805

7.7.1 Derivation of the LMS Algorithms ........... 806
7.7.2 Performance of the LMS Algorithms .......... 813
7.7.3 LMS Algorithm Behavior ................ 817
7.7.4 Quadratic Constraints .................. 822
7.7.5 Summary: LMS algorithms ............... 826
7.8 Detection of Signal Subspace Dimension ............ 82'7
7.8.1 Detection Algorithms .................. 828
7.8.2 Eigenvector Detection Tests ............... 841
7.9 Eigenspace and DMR Beamformers ............... 845
7.9.1 Performance of SMI Eigenspace Beamformers ..... 846
7.9.2 Eigenspace and DMR Beamformers: Detection of Sub-
space Dimension ...................... 850
7.9.3 Subspace tracking .................... 860
7.9.4 Summary ......................... 863
7.10 Beamspace Beamformers ..................... 864
7.10.1 Beamspace SMI ...................... 865
7.10.2 Beamspace RLS ..................... 869
7.10.3 Beamspace LMS ..................... 872
7.10.4 Summary: Adaptive Beamspace Processing ...... 873
7.11 Broadband Beamformers ..................... 874
7.11.1 SMI Implementation ................... 875
7.11.2 LMS Implementation ................... 878
7.11.3 GSC: Multichannel Lattice Filters ........... 884
7.11.4 Summary ......................... 885
7.12 Summary ............................. 885
7.13 Problems ............................. 887

8 Parameter Estimation I: Maximum Likelihood 917

81. Introduction. ........................... 917
82. Maximum Likelihood and Maximum a posteriori Estimators . 920
8.2.1 Maximum Likelihood (ML) Estimator ......... 922
8.2.2 Maximum a posteriori (MAP) Estimator ....... 924
8.2.3 Cramer-Rao Bounds ................... 925
83. Parameter Estimation Model .................. 933
8.3.1 Multiple Plane Waves .................. 933
8.3.2 Model Perturbations ................... 936
8.3.3 Parametric Spatially Spread Signals .......... 938
8.3.4 Summary ......................... 938
84. Cramer-Rao Bounds ....................... 938
xiv Contents

8.4.1 Gaussian Model: Unknown Signal Spectrum ...... 939

8.4.2 Gaussian Model: Uncorrelated Signals with Unknown
Power ........................... 958
8.4.3 Gaussian Model: Known Signal Spectrum ....... 967
8.4.4 Nonrandom (Conditional) Signal Model ........ 971
8.4.5 Known Signal Waveforms ................ 978
8.4.6 Summary ......................... 980
8.5 Maximum Likelihood Estimation ................ 984
8.5.1 Maximum Likelihood Estimation ............ 984
8.5.2 Conditional Maximum Likelihood Estimators ..... 1004
8.5.3 Weighted Subspace Fitting ............... 1009
8.5.4 Asymptotic Performance ................. 1014
8.5.5 Wideband Signals .................... 1015
8.5.6 Summary ......................... 1018
8.6 Computational Algorithms ................... 1018
8.6.1 Optimization Techniques ................ 1018
8.6.2 Alternating Maximization Algorithms ......... 1025
8.6.3 Expectation Maximization Algorithm ......... 1031
8.6.4 Summary ......................... 1037
8.7 Polynomial Parameterization .................. 1037
8.7.1 Polynomial Parameterization .............. 1038
8.7.2 Iterative Quadratic Maximum Likelihood (IQML) . . 1039
8.7.3 Polynomial WSF (MODE) ............... 1045
8.7.4 Summary ......................... 1053
8.8 Detection of Number of Signals ................. 1054
8.9 Spatially Spread Signals ..................... 1055
8.9.1 Parameterized S(&+) .................. 1055
8.9.2 Spatial ARMA Process ................. 1062
8.9.3 Summary ......................... 1062
8.10 Beamspace algorithms ...................... 1062
8.10.1 Introduction ....................... 1062
8.10.2 Beamspace Matrices ................... 1065
8.10.3 Beamspace Cramer-Rao Bound ............. 1073
8.10.4 Beamspace Maximum Likelihood ............ 1081
8.10.5 Summary ......................... 1088
8.11 Sensitivity, Robustness, and Calibration ............ 1088
8.11.1 Model Perturbations .................. 1089
8.11.2 Cram&-Rao Bounds ................... 1090
8.11.3 Sensitivity of ML Estimators .............. 1098
8.11.4 MAP Joint Estimation .................. 1099
Contents xv

8.11.5 Self-Calibration Algorithms ............... 1101

8.11.6 Summary ......................... 1102
8.12 Summary ............................. 1102
8.12.1 Major Results ....................... 1102
8.12.2 Related Topics ...................... 1105
8.12.3 Algorithm complexity .................. 1108
8.13 Problems ............................. 1109

9 Parameter Estimation II 1139

9.1 Introduction ........................... 1139
9.2 Quadratic Algorithms ...................... 1140
9.2.1 Introduction ....................... 1140
9.2.2 Beamscan Algorithms .................. 1142
9.2.3 MVDR (Capon) Algorithm ............... 1144
9.2.4 Root Versions of Quadratic Algorithms ........ 1147
9.2.5 Performance of MVDR Algorithms ........... 1148
9.2.6 Summary ......................... 1149
9.3 Subspace Algorithms ....................... 1155
9.3.1 Introduction ....................... 1155
9.3.2 MUSIC .......................... 1158
9.3.3 Minimum-Norm Algorithm ............... 1163
9.3.4 ESPRIT .......................... 1170
9.3.5 Algorithm Comparison .................. 1189
9.3.6 Summary ......................... 1190
9.4 Linear Prediction ......................... 1194
9.5 Asymptotic Performance ..................... 1195
9.5.1 Error Behavior ...................... 1195
9.5.2 Resolution of MUSIC and Min-Norm .......... 1203
9.5.3 Small Error Behavior of Algorithms .......... 1211
9.5.4 Summary ......................... 1233
9.6 Correlated and Coherent Signals ................ 1233
9.6.1 Introduction ....................... 1233
9.6.2 Forward-Backward Spatial Smoothing ......... 1235
9.6.3 Summary. ........................ 1241
9.7 Beamspace Algorithms ...................... 1243
9.7.1 Beamspace MUSIC .................... 1243
9.7.2 Beamspace Unitary ESPRIT .............. 1247
9.7.3 Beamspace Summary .................. 1251
9.8 Sensitivity and Robustness ................... 1251
9.9 Planar Arrays ........................... 1255
xvi Contents

9.9.1 Standard Rectangular Arrays .............. 1255

9.9.2 Hexagonal Arrays .................... 1272
9.9.3 Summary: Planar Arrays ................ 1279
9.10 Summary ............................. 1279
9.10.1 Major Results ....................... 1279
9.10.2 Related Topics ...................... 1282
9.10.3 Discussion ......................... 1285
9.11 Problems ............................. 1285

10 Detection and Other Topics 1318

10.1 Optimum Detection ....................... 1318
10.1.1 Classic Binary Detection ................. 1319
10.1.2 Matched Subspace Detector ............... 1320
10.1.3 Spatially Spread Gaussian Signal Processes ...... 1321
10.1.4 Adaptive Detection .................... 1323
10.2 Related Topics .......................... 1327
10.3 Epilogue .............................. 1329
10.4 Problems ............................. 1329

A Matrix Operations 1340

A.1 Introduction ............................ 1340
A.2 Basic Definitions and Properties ................ 1341
A.2.1 Basic Definitions ..................... 1341
A.2.2 Matrix Inverses ...................... 1347
A.2.3 Quadratic Forms ..................... 1348
A.2.4 Partitioned Matrices ................... 1349
A.2.5 Matrix products ..................... 1351
A.2.6 Matrix Inequalities .................... 1356
A.3 Special Vectors and Matrices .................. 1356
A.3.1 Elementary Vectors and Matrices ............ 1356
A.3.2 The vet(A) matrix .................... 1358
A.3.3 Diagonal Matrices .................... 1359
A.3.4 Exchange Matrix and Conjugate Symmetric Vectors . 1361
A.3.5 Persymmetric and Centrohermitian Matrices ..... 1362
A.3.6 Toeplitz and Hankel Matrices .............. 1364
A.3.7 Circulant Matrices .................... 1365
A.3.8 Triangular Matrices ................... 1366
A.39 Unitary and Orthogonal Matrices ............ 1367
A.3.10 Vandermonde Matrices .................. 1368
A.3.11 Projection Matrices ................... 1369
Contents xvii

A.3.12 Generalized Inverse .................... 1370

A.4 Eigensystems ........................... 1372
A.4.1 Eigendecomposition ................... 1372
A.4.2 Special Matrices ..................... 1376
A.5 Singular Value Decomposition .................. 1381
A.6 QR Decomposition ........................ 1387
A.6.1 Introduction ....................... 1387
A.6.2 QR Decomposition .................... 1388
A.6.3 Givens Rotation ..................... 1390
A.6.4 Householder Transformation ............... 1394
A.7 Derivative Operations ...................... 1397
A.7.1 Derivative of Scalar with Respect to Vector ...... 1397
A.7.2 Derivative of Scalar with Respect to Matrix ...... 1399
A.7.3 Derivatives with Respect to Parameter ......... 1401
A.7.4 Complex Gradients .................... 1402

B Array Processing Literature 1407

B.1 Journals .............................. 1407
B.2 Books ............................... 1408
B.3 Duality .............................. 1409

C Notation 1414
C.l Conventions ............................ 1414
C.2 Acronyms ............................. 1415
C.3 Mathematical Symbols ...................... 1418
C.4 Symbols .............................. 1419

Index 1434
Preface

Array processing has played an important role in many diverse application

areas. Most modern radar and sonar systems rely on antenna arrays or
hydrophone arrays as an essential component of the system. Many commu-
nication systems utilize
phased arrays or multiple beam antennas to achieve
their performance objectives. Seismic arrays are widely used for oil explo-
ration and detection of underground nuclear tests. Various medical diagnosis
and treatment techniques exploit arrays. Radio astronomy utilizes very large
antenna arrays to achieve resolution goals. It appears that the third genera-
tion of wireless systems will utilize adaptive array processing to achieve the
desired system capacity. We discuss various applications in Chapter 1.
My interest in optimum array processing started in 1963 when I was
an Assistant Professor at M.I.T. and consulting with Arthur D. Little on
a sonar project for the U.S. Navy. I derived the optimum processor for
detecting Gaussian plane-wave signals in Gaussian noise [VT66a], [VT66b].
It turned out that Bryn [Bry62] had published this result previously (see also
Vanderkulk [Van63]). My work in array processing decreased as I spent more
time in the general area of detection, estimation, and modulation theory.
In 1968, Part I of Detection, Estimation, and Modulation Theory [VT681
was published. It turned out to be a reasonably successful book that has been
widely used by several generations of engineers. Parts II and III ([VT7la],
[VT7lb]) were published in 1971 and focused on specific application areas
such as analog modulation, Gaussian signals and noise, and the radar-sonar
problem. Part II had a short life span due to the shift from analog modu-
lation to digital modulation. Part III is still widely used as a reference and
as a supplementary text. In a moment of youthful optimism, I indicated in
the Preface to Part III and in Chapter III-14 that a short monograph on
optimum array processing would be published in 1971. The bibliography
lists it as a reference, (Optimum Array Processing, Wiley, 1971), which has
been subsequently cited by several authors. Unpublished class notes [VT691
contained much of the planned material. In a very loose sense, this text is

xix
xx Preface

the extrapolation of that monograph.

Throughout the text, there are references to Parts I and III of Detection,
Estimation, and Modulation Theory. The referenced material is available in
several other books, but I am most familiar with my own work. Wiley has
republished Parts I and III [VTOla], [VTOlb] in paperback in conjunction
with the publication of this book so the material will be readily available.
A few comments on my career may help explain the thirty-year delay. In
1972, M.I.T. loaned me to the Defense Communications Agency in Washing-
ton, D.C., where I spent three years as the Chief Scientist and the Associate
Director for Technology. At the end of this tour, I decided for personal
reasons to stay in the Washington, D.C., area. I spent three years as an
Assistant Vice-President at COMSAT where my group did the advanced
planning for the INTELSAT satellites. In 1978, I became the Chief Scientist
of the United States Air Force. In 1979, Dr.Gerald Dinneen, the former
director of Lincoln Laboratories, was serving as Assistant Secretary of De-
fense for C31. He asked me to become his Principal Deputy and I spent two
years in that position. In 1981, I joined M/A-COM Linkabit. Linkabit is the
company that Irwin
Jacobs and Andrew Viterbi started in 1969 and sold to
M/A-COM in 1979. I started an Eastern operations, which grew to about
200 people in three years. After Irwin and Andy left M/A-COM and started
Qualcomm, I was responsible for the government operations in San Diego
as well as Washington, D.C. In 1988, M/A-COM sold the division. At that
point I decided to return to the academic world.
I joined George Mason University in September of 1988. One of my
priorities was to finish the book on optimum array processing. However, I
found that I needed to build up a research center in order to attract young
research-oriented faculty and doctoral students. This process took about six
years. The C31 Center of Excellence in Command, Control, Communica-
tions, and Intelligence has been very successful and has generated over $30
million in research funding during its existence. During this growth period,
I spent some time on array processing, but a concentrated effort was not
possible.
The basic problem in writing a text on optimum array processing is that,
in the past three decades, enormous progress had been made in the array pro-
cessing area by a number of outstanding researchers. In addition, increased
computational power had resulted in many practical applications of opti-
mum algorithms. Professor Arthur Baggeroer of M.I.T. is one of the leading
contributors to array processing in the sonar area. I convinced Arthur, who
had done his doctoral thesis with me in 1969, to co-author the optimum
array processing book with me. We jointly developed a comprehensive out-
Preface xxi

line. After several years it became apparent that the geographical distance
and Arthur’s significant other commitments would make a joint authorship
difficult and we agreed that I would proceed by myself. Although the final
outline has about a 0.25 correlation with the original outline, Arthur’s col-
laboration in structuring the original outline and commenting on the results
have played an important role in the process.
In 1995, I took a sabbatical leave and spent the year writing the first
draft. I taught a one-year graduate course using the first draft in the 1996-
1997 academic year. A second draft was used in the 1997-1998 academic
year. A third draft was used by Professor Kristine Bell in the 19984999
academic year. Unlike the M.I.T. environment where I typically had 40-
50 graduate students in my detection and estimation classes, our typical
enrollment has been 8-10 students per class. However, many of these stu-
dents were actively working in the array processing area and have offered
constructive suggestions.
The book is designed to provide a comprehensive introduction to opti-
mum array processing for students and practicing engineers. It will prepare
the students to do research in the array processing area or to implement
actual array processing systems. The book should also be useful to people
doing current research in the field. We assume a background in probability
theory and random processes. We assume that the reader is familiar with
Part I of Detection, Estimation, and Modulation Theory [VT68], [VTOla]
and parts of Part III [VT7lb], [VTOlb]. The first use of [VT68], [VTOla] is
in Chapter 5, so that a detection theory course could be taken at the same
time. We also assume some background in matrix theory and linear alge-
bra. The book emphasizes the ability to work problems, and competency in
MATLAB @ is essential.
The final product has grown from a short monograph to a lengthy text.
Our experience is that, if the students have the correct background and
motivation, we can cover the book in two fifteen-week semesters.
In order to make the book more useful, Professor Kristine Bell has de-
veloped a Web site:
http://ite.gmu.edu/DetectionandEstimationTheory/
that contains material related to all four parts of the Detection, Estimation,
and Modulation Theory series.
The Optimum Array Processing portion of the site contains:
(i) MATLAB@ scripts for most of the figures in the book. These scripts
enable the reader to explore different signal and interference environ-
ments and are helpful in solving the problems. The disadvantage is
xxii Preface

that a student can use them without trying to solve the problem inde-
pendently. We hope that serious students will resist this temptation.

(ii) Several demos that allow the reader to see the effect of parameter
changes on beam patterns and other algorithm outputs. Some of the
demos for later chapters allow the reader to view the adaptive behavior
of the system dynamically. The development of demos is an ongoing
process.
(iii) An erratum and supplementary comments regarding the text will be
updated periodically on the Web site. Errors and comments can be
sent to either [email protected] or kbellegmuedu.
(iv) Solutions, including MATLAB@ scripts where appropriate, to many of
the problems and some of the exams we have used. This part is pass-
word protected and is only available to instructors. To obtain a pass-
word, send an e-mail request to either [email protected] or [email protected].

In order to teach the course, we created a separate LATEX file con-

taining only the equations. By using Ghostview, viewgraphs containing the
equations can be generated. A CD-rom with the file is available to instruc-
tors who have adopted the text for a course by sending me an e-mail at
hlvQgmu.edu.
The book has relied heavily on the results of a number of researchers.
We have tried to acknowledge their contributions. The end-of-chapter bibli-
ographies contain over 2,000 references. Certainly the book would not have
been possible without this sequence of excellent research results.
A number of people have contributed in many ways and it is a pleasure to
acknowledge them. Andrew Sage, founding dean of the School of Information
Technology and Engineering at George Mason University, provided continual
encouragement in my writing efforts and extensive support in developing the
C”1 Center. The current dean, Lloyd Griffiths, has also been supportive of
my work.
A number of the students taking my course have offered constructive
criticism and corrected errors in the various drafts. The following deserve
explicit recognition: Amin Jazaeri, Hung Lai, Brian Flanagan, Joseph Her-
man, John Uber, Richard Bliss, Mike Butler, Nirmal Warke, Robert Zar-
nich, Xiaolan Xu, and Zhi Tian suffered through the first draft that con-
tained what were euphemistically referred to as typos. Geoff Street, Stan
Pawlukiewicz, Newell Stacey, Norman Evans, Terry Antler, and Xiaomin
Lu encountered the second draft, which was significantly expanded. Roy
Bethel, Paul Techau, Jamie Bergin, Hao Cheng, and Xin Zhang critiqued
Preface xxiii

the third draft. The final draft was used in my Optimum Array Processing
course during the 2000-2001 academic year. John Hiemstra, Russ Jeffers,
Simon Wood, Daniel Bray, Ben Shapo, and Michael Hunter offered useful
comments and corrections. In spite of this evolution and revision, there are
probably still errors. Please send corrections to me at [email protected] and
they will he posted on the Web site.
Two Visiting Research Professors, Shulin Yang and Chen-yang Yang also
listened to the course and offered comments. Drs. Shulin Yang, Chen-yang
Yang, and Ms. Xin Zhang composed the book in LATEX and provided im-
portant editorial advice. Aynur Abdurazik and Muhammad Abdulla did the
final LATEX version. Their competence and patience have been extraordi-
nary. Joshua Kennedy and Xiaomin Lu drew many of the figures. Four
of my graduate research assistants, Miss
I Zhi Tian, Miss Xiaolan Xu, Mr.
Xiaomin Lu, and Miss Xin Zhang worked most of the examples in various
chapters. Their help has been invaluable in improving the book.
A separate acknowledgment is needed for Professor Kristine Bell. She
did her doctoral dissertation in the array processing area for Professor Yariv
Ephraim and me, and she has continued to work with me on the text for
several years. She has offered numerous insights into the material and into
new developments in many areas. She also taught the two-semester course
in 1998--1999 and developed many aspects of the material. Her development
of the Web site adds to the pedagogical value of the book.
Several colleagues agreed to review the manuscript and offer criticisms.
The group included many of the outstanding researchers in the array pro-
cessing area. Dan Fuhrmann, Norman Owsley, Mats Viberg, and Mos Kaveh
reviewed the entire book and offered numerous corrections and suggestions.
In addition, they pointed out a number of useful references that I had missed.
Petre Stoica provided excellent comments on Chapters 7-10, and two of his
students, Erik Larsson and Richard Abrhamsson,additionalprovided
com-
ments. Louis Scharf, Ben Friedlander, Mati Wax, and John Buck
provided
constructive comments on various sections of the book. Don Tufts provided
a large amount of historical material that was very useful. I appreciate
the time that all of these colleagues took from their busy schedules. Their
comments have improved the book.

Harry L. Van Trees

January 2002
xxiv Bibliography

Bibliography
[Bry62] F. Bryn. Optimum signal processing of three-dimensional array operating on
Gaussian signals and noise. J. Acoust. Sot. Amer., 34(3):289-297, March 1962.
[Van631 V. Vanderkulk. Optimum processing for acoustic arrays. J. Bit. IRE, 26(4):286-
292, October 1963.
[VT66a] H. L Van Trees. Optimum processing for passive sonar arrays. Proc. IEEE Ocean
Electronics Symp., pages41-65, Honolulu, Hawaii, 1966.
[VT66b] H. L. Van Trees. A unified theory for optimum array processing. Technical
R,eport 4160866, Dept. of the Navy Naval Ship Systems Command, Arthur D.
Little, Inc., Cambridge,MA, Aug. 1966.
[VT681 H. L. V an Trees. Detection, Estimation, and Modulation Theory, Part I. Wiley,
New York, 1968.

[VTOla] H. L. Van Tr ees. Detection, Estimation, and Modulation Theory, Part I. Wiley
Interscience, New York, 2001.

[VT691 H. L. Van Tr ees. Multi-Dimensional and Multi- Variable Processes. unpublished

class notes, M.I.T, 1969.

[VT7la] H. L. Van Tr ees. Detection, Estimation, and Modulation Theory, Part II. Wiley,
New York, 1971.

[VT7lb] H. L. Van Trees. Detection, Estimation, and Modulation Theory, Part III. Wiley,
New York, 1971.
[VTOlb] H. L. V an Trees. Detection, Estimation, and Modulation Theory, Part III. Wiley
Interscience, New York, 2001.
Optimum Array Processing
Chapter 1

Introduction

In Parts I, II, and III of Detection, Estimation, and Modulation Theory

(DEMT) [VT68], [VTOla], [VT7la], [VT7lb], [VTOlb], we provide a rea-
sonably complete discussion of several areas:

(i) Detection theory

In this case, we were concerned with detecting signals in the presence
of Gaussian noise. The class of signals included known signals, signals
with unknown parameters, and signals that are sample functions from
Gaussian random processes. This problem was covered in Chapter I-4
and Chapters III-1 through 111-5.

(ii) Estimation theory

In this case, we were concerned with estimating the parameters of
signals in the presence of Gaussian noise. This problem was covered
in Chapter I-4 and Chapters III-6 and 111-7.

(iii) Modulation theory

In this case, we were concerned with estimating a continuous waveform
(or the sampled version of it). If the signal has the waveform in it
in a linear manner, then we have a linear estimation problem and
obtain the Wiener filter or the Kalman-Bucy filter as the optimum
estimator. This problem was covered in Chapter I-6. The case of
nonlinear modulation is covered in Chapter I-5 and Volume II.

All of the results in the first three volumes consider signals and noises
that could be characterized in the time domain (or equivalently, the fre-
quency domain). In this book, we consider the case in which the signals and

1
Chapter 2

Arrays and Spatial Filters

2.1 Introduction

We assume that we have a signal or multiple signals that are located in some
region of a space-time field. We also have noise and/or interference that is
located in some region of a space-time field. In the applications of interest
these regions have some overlap.
An array is used to filter signals in a space-time field by exploiting their
spatial characteristics. This filtering may be expressed in terms of a de-
pendence upon angle or wavenumber. Viewed in the frequency domain this
filtering is done by combining the outputs of the array sensors with complex
gains that enhance or reject signals according to their spatial dependence.
Usually, we want to spatially filter the field such that a signal from a partic-
ular angle, or set of angles, is enhanced by a constructive combination and
noise from other angles is rejected by destructive interference.
The design of arrays to achieve certain performance criteria involves
trade-offs among the array geometry, the number of sensors, signal-to-noise,
and signal-to-interference ratios, as well as a number of other factors.
There are two aspects of array design that determine their performance as
spatial filters. First, their geometry establishes basic constraints upon their
operation. Line arrays can resolve only one angular component. This leads
to a cone of uncertainty and right/left ambiguities. Circular arrays have
different patterns than crossed or planar arrays. Frequently the geometry
is established by physical constraints and the designer may have limited
freedom in specifying the array geometry.
The second aspect is the design of the complex weightings of the dat,a at
each sensor output. The choice of these weightings determines the spatial

17
18 2. I Introduction

filtering characteristics of the array for a given geometry.

In this chapter we introduce the basic definitions and relationships that
are used to analyze and synthesize arrays. Our approach is to introduce the
concepts for an arbitrary array . geometry. We then specialize the result to
a uniform linear array and then further specialize the result to a uniform
weighting. In Chapter 3, we return to linear arrays and provide a detailed
discussion of the analysis and synthesis of linear arrays. In Chapter 4, we
study the analysis and synthesis of planar and volume arrays.
This chapter is organized in the following manner. In Section 2.2, we
introduce the frequency-wavenumber response function and beam pattern
of an array. We employ wavenumber variables with dimensions of inverse
length for a number of reasons. First, array coordinates and wavenumbers
are conjugate Fourier variables, so Fourier transform operations are much
simpler. Second, all the powerful properties of harmonic analysis as extended
to homogenous processes can be used directly and the concept of an array as
a spatial filter is most applicable. Third, angle variables specify array filter
responses over a very restricted region of wavenumber space. While it does
describe the response over the region for all real, propagating signals, that
is, those space-time processes that implicitly satisfy a wave equation when
one assigns a propagation speed and direction, there are a lot of advantages
to considering the entire wavenumber space. The so-called virtual space, or
wavenumber realm where real signals cannot propagate is very useful in the
analysis of array performance.
In Section 2.3, we specialize these results to a uniform linear array and
study the characteristics of the beam pattern. In Section 2.4, we further
specialize these results to the case of a uni formly weighted linear array. This
leads to a beam pattern that we refer to as the conventional beam pattern. It
will play a fundamental role in many of our subsequent studies. In Section
2.5, we discuss array steering and show how it affects the beam pattern
in wavenumber space and in angle space. In Section 2.6, we define three
irnportant performance measures:

(i) Directivity
(ii) Array gain
(iii) Tolerance function

These performance measures are utilized throughout our discussion.

The discussion in the first six sections assumes that the sensors are
isotropic (i.e., their response is independent of the direction of arrival of the
signal). In Section 2.7, we introduce the concept of pattern multiplication
to accommodate non-isotropic sensors. In Section 2.8, we consider the case
Other documents randomly have
different content
 The Correct Representation of the Pelvic Organs.

Normal Movements of the Uterus.

Strictly, the uterus can have no absolutely normal position or location,

because it has a certain normal range of movements which depend to
some extent upon respiration, intra-abdominal forces, and locomotion, but
more especially upon the varying quantity of material in the rectum and
bladder. Its normal position, then, varies within the limits of its normal
movements. If the body of the uterus rest upon the bladder, it must rise as
the bladder becomes distended, and, conversely, if the urine be drawn
through a catheter while the woman is lying on her back, the uterus,
notwithstanding the opposing influence of its own weight, immediately
follows the receding wall of the bladder and returns through an angle of
45°, or possibly even 90°, to its accustomed position. The dotted lines in
Fig. 2 indicate the degree of version and flexion consequent upon the
varying quantity of fluid in the bladder.

The full rectum forces the uterus in the opposite direction, toward the
symphysis, and thereby counteracts the influence of the bladder. This
anterior movement is, however, somewhat limited, and is confined to the
cervical portion, except when the body has been forced back into close
proximity with the rectum by the over-distended bladder.

Normal Supports of the Uterus.

The uterus is maintained in its normal position and location by the

following agents:

a. The uterine ligaments;

b. The pelvic floor.3

3 For a description of the female pelvic floor see Hart's Atlas.

a. Physiologically, these ligaments are relaxed; the state of tension would

be pathological; they do not fix the uterus; they only tend to limit its
movements to their normal range. Backward displacement of the body is
resisted by the round ligaments, backward displacement of the cervix by
the utero-vesical ligaments and by the vesico-vaginal wall. Forward and
downward displacements are resisted by the utero-sacral ligaments, and
excessive lateral motion by the broad ligaments. This restraining power is
doubtless greater in the utero-sacral than in any of the other ligaments.

b. The pelvic floor, which is the chief support of the uterus, is divided into
two segments, the pubic and the sacral. The pubic segment4 is composed
of bladder, urethra, anterior vaginal wall, and bladder peritoneum. It is
attached in front to the symphysis pubis and laterally to the anterior bony
walls of the pelvis. The sacral segment5 is composed of rectum, perineum,
posterior vaginal wall, and strong tendinous and muscular tissue. It is
attached to the coccyx, to the sacrum, and to the posterior wall of the
bony pelvis.
4 Hart and Barbour's Manual of Gynecology.

5 Ibid.

Permeating the pelvic floor in all directions, entering into the composition
of its single parts, binding them together, and sending its processes to the
bony pelvis, is the pelvic connective tissue, upon the integrity of which
depends the integrity of the pelvic floor as a uterine support. Its pernicious
influence as a pathological factor will be considered hereafter. The old idea
that the uterus is supported by the vaginal walls or by the perineum or by
the uterine ligaments is obsolete; they are important parts of the pubic and
sacral segments, and as such contribute their share, but the pelvic floor as
a whole supports the uterus. The various uterine supports are to a great
extent the seat of motor influence. They consequently not only resist
excessive movement, but also serve to return the organ from its
physiological migrations.

DEFINITION AND NOMENCLATURE OF DISPLACEMENTS.—In the foregoing pages the

normal location, position, movements, and supports of the uterus have
been defined. Those conditions are pathological which induce changes to
positions or locations beyond the defined limits, or which so fix the organ
that its normal movements are prevented. The displacements are divided
into mal-locations and malpositions.

The mal-locations in which the entire uterus occupies a place outside its
normal limits are as follows: ascent, retro-location, ante-location, lateral
location, descent.

The malpositions are determined by excessive change in the inclination of

the uterine axis. They are further divided into flexions, in which the organ
is bent upon itself in an abnormal degree, manner, or direction; and
versions, in which the axis of the unflexed uterus inclines in an abnormal
degree or direction. The malpositions are retroversion, retroflexion, lateral
version, lateral flexion, anteversion, anteflexion.
SYMPTOMS AND DIAGNOSIS IN GENERAL.—Each variety of displacement may be
indicated by its own group of symptoms and physical signs. These will be
presented in the study of the special lesions. To avoid repetition, those
symptoms and signs which pertain to no special displacement, but which
belong to all alike, will be mentioned at once. They may arise either from
the displacement itself or from its possible complications, of which the
following are examples: Metritis, ovaritis, salpingitis, atresia and stenosis,
cystitis, vesical catarrh, rectitis, rectal catarrh, peri-uterine cellulitis and
peritonitis, uterine catarrh, tumors, cicatrices, etc.

Uterine displacement may be a cause or an effect of associated

complications, or together with them it may be a concurrent result of some
common cause, or it may have had primarily no pathological connection
with them. The symptoms of displacement refer to the pelvic organs or to
the nervous system. Among the symptoms which refer to the pelvic organs
are—difficulty in walking and standing; pelvic pain, more or less constant;
dysmenorrhoea, menorrhagia, sterility, frequent abortion, constipation,
painful or difficult defecation, dysuria, polyuria, tenesmus, etc. Among the
symptoms which refer to the nervous system are—neuralgia in various
parts, paralysis, hysteria, nervous dyspepsia, anæmia, chlorosis, spinal
irritation, etc.

The final diagnosis must always depend upon direct examination of the
uterus itself. The first division of the above group of symptoms is not likely
to escape notice as indicative of displacement, but the nervous symptoms
are constantly disregarded or treated without reference to their possible
pelvic origin. The frequent dependence of these nervous phenomena upon
displacement is proved by their persistence in many cases after ordinary
treatment, by their prompt disappearance upon permanent replacement
and retention of the uterus by mechanical means, and by their equally
prompt recurrence upon removal of the support. The presence, therefore,
of the second division of the group or any part thereof, even though the
first be absent, will justify, may even necessitate, a careful investigation
into the state of the pelvic organs.

That examination which results only in giving the name to a special variety
of displacement, and does not include the complicating lesions, would not
furnish a sufficient guide to the therapeutic indications, and is therefore
inadequate. The successful treatment, for instance, of an anteflexion
dependent upon inflammation of the utero-sacral ligaments must include
the removal of the inflammation.

An important prerequisite to examination is the absence of material in the

rectum and bladder. The full rectum distorts the vaginal walls, deprives the
examiner of the space necessary for the introduction of the speculum, and
throws the uterus out of its accustomed position. Much more troublesome
is the presence of even a small quantity of urine in the bladder, because it
causes the patient to render the abdominal muscles tense when the hand
is placed over the lower portion of the abdomen for bimanual palpation,
and makes it impossible to engage the uterus between the hand and the
examining finger. The distended bladder by pushing the uterus upward and
backward makes bimanual palpation almost useless. It is not surprising
that conflicting opinions are common, when one day the patient is
examined with rectum and bladder full, another day empty; one day in the
dorsal, another in Sims's or the knee-chest position; one day with the
cylindrical or bivalve speculum, another day with Sims's or Simon's.

For digital examination the dorsal position is preferred: the patient should
be drawn close to the edge of a bed, or preferably a table, the thighs being
flexed, the feet about fifteen inches apart, and the knees widely separated.
The examiner should stand facing the patient, never at the side. The index
finger of the left6 hand, lubricated with vaseline or oil, then slowly
advances over the perineum into the vagina, noting the condition of the
perineum, the presence or absence of cicatrices or of sub-involution of the
vagina or perineum, the capacity of the vagina, the condition, size, and
direction of the cervix, its distance from the sacrum and vulva, its mobility
or fixation. Now, for the first time, the right hand is pressed well down
behind the pubes, and the uterus is engaged between it and the examining
finger. (See Figs. 16 and 17.) In this way the examiner may determine
more accurately the position, location, and size of the entire organ; may
detect the possible presence of complicating tumors, both inflammatory
and non-inflammatory; may also note, if possible, the location and
condition of the ovaries, which, especially in the posterior displacements,
are liable to be prolapsed and excessively sensitive, and to constitute,
therefore, a most intractable complication. The index finger sweeps around
the cervix in search of tender places which may be the result of former
cellulitis or the expression of some neurosis. Above all, the digital
examination requires a light, gentle, delicate touch.
 6 The left-hand method of examination is incomparably superior to the right. The palmar
surface of the index finger is more easily directed toward the left side of the pelvis, which is
especially subject to disease. Its tactile sense is more acute and more easily educated. The
stronger right hand should be free to palpate the surface of the abdomen in conjoined
manipulation.

In exploring the uterine cavity to learn its position the fine silver-wire probe
of Emmet—not the sound—should be used. The uterus, if freely movable,
is liable to be thrown out of its accustomed position by the heavier,
unyielding sound. The sound also causes much more pain and exposes the
patient to great danger of cellulitis. The frequent lighting and relighting of
pelvic inflammation by injudicious slight manipulations of the uterus
doubtless led Emmet to the utterance of a prophecy which ought to
become classical: "A great advance in the treatment of the diseases of
women will be made whenever practitioners become so impressed with the
significance of cellulitis as to apprehend its existence in every case. The
successful operator in this branch of surgery will always be on the lookout
for the existence of cellulitis, and take measures to guard against its
occurrence."

When the probe or the sound is used without the speculum, the patient
should be on the back and the index finger of the left hand should be used
as a guide. The bivalve and cylindrical specula are almost useless in
explorations of the interior of the uterus. The exploration is most
effectually and gently made with Sims's speculum, the patient being in the
left latero-prone position. In some cases the probe cannot be passed by
any other method.

Ascent of the Uterus.

This mal-location may result from traction above or from pressure below.
The organ may be drawn upward and backward by shortening of the
utero-sacral ligaments, which results from inflammation and which usually
induces a troublesome form of anteflexion. The enlarged pregnant uterus
sometimes becomes attached by adhesive inflammation to a portion of the
peritoneum in one of the higher zones of the pelvis or in the abdomen, and
the organ may consequently remain fixed in its elevated position after
involution. A tumor connected with the uterus or its appendages which has
grown too large to be retained in the pelvis may, upon rising into the
abdomen, drag the uterus with it. Pressure below may come from
excessive distension of the rectum or bladder, or from a large accumulation
of menstrual fluid in the vagina, or from a tumor originating in any portion
of the pelvis below the level of the uterus. In diagnosis, prognosis, and
treatment this displacement is wholly subordinate to the more significant
lesions of which it is only the incidental result.

Retro-location of the Uterus.

The uterus may be forced back into a post-normal location by the presence
of a tumor in front or by the distended bladder, or it may be drawn back
and fixed by peritoneal adhesions. Retro-location is liable to induce vesical
irritation by putting the vesico-vaginal wall on the stretch and thereby
dragging on the neck of the bladder. This intractable symptom is
sometimes relieved by Emmet's buttonhole operation of urethrotomy, for
an account of which see section on Anteflexion. This operation would
obviously be applicable also for the relief of the same symptom when
caused by ascent of the uterus.

Ante-location of the Uterus.

The causes of this displacement are similar to those which produce retro-
location; they are—distension of the rectum, post-uterine hæmatocele,
post-uterine tumors, and peritoneal adhesions. Ante-location often causes
vesical irritation, consequent upon the invasion by the uterus of that space
which belongs to the bladder.

Lateral Locations of the Uterus.

The entire uterus is often displaced to the right or the left by a tumor or by
an inflammatory exudate. The latter occurs as a product of cellulitis,
usually in the left broad ligament, and crowds the organ toward the
opposite side of the pelvis. After resolution the ligament, shortened by
inflammatory contraction, draws the uterus to the affected side and fixes it
there. Lateral displacement from this cause often accompanies laceration
of the cervix, the cellulitis having occurred on the side corresponding to the
laceration.

Descent or Prolapse of the Uterus.

The nature of this displacement is clearly indicated by its name. It is

convenient to distinguish three degrees of descent: In the first the organ is
displaced downward and forward until sufficient space has been gained
between the cervix and the sacrum to permit the body to turn back into
extreme retroversion; in the second the cervix descends to the vulva; in
the third the uterus protrudes partially or wholly through the vulva,
constituting a condition sometimes called procidentia.

ETIOLOGY AND CLINICAL HISTORY.—Descent may be the result of any or all of

the following causes: I. Pressure from above; II. Weakening of the
supports; III. Increased weight of the uterus; IV. Traction from below.
Either of the above conditions being the primary cause, the others singly or
combined may result.

I. Pressure from above may depend upon the presence of a pelvic or

abdominal tumor, ascites, fecal accumulations, tight or heavy clothing, etc.

II. The uterine supports may be weakened and relaxed in consequence of

subinvolution, senile atrophy, abnormally large pelvis, increased weight of
the uterus, pressure from above, traction from below, etc.

III. Increased weight of the uterus may be caused by congestion,

subinvolution, hypertrophy, hyperplasia, pregnancy, fluid in the
endometrium, uterine tumors, etc.

IV. Traction from below may be due to vaginal cicatrices, abnormally short
vagina, falling of the pelvic floor, etc.

Obviously, descent of the vesico- and recto-vaginal walls, or, more

comprehensively, the sacral and pubic segments of the pelvic floor, involves
also concurrent descent of the uterus. Descent of the vagina, therefore,
must be studied in connection with the descent of the uterus. Excessive
descent of the vaginal walls usually originates with parturition.
In labor the anterior wall of the vagina is so depressed, stretched, and
shortened by the advancing head that during and after the second stage
the anterior lip of the cervix may be seen behind the urethra. If the
puerperium progress favorably, with prompt involution of the uterus,
vagina, perineum, and peritoneum, the relaxation of the vesico-vaginal wall
and of the utero-sacral supports disappears and the uterus resumes its
normal multiparous location and position.7 But if the enlarged uterus
remain in the long axis of the vagina, with its fundus incarcerated in the
hollow of the sacrum between the utero-sacral ligaments, and with its
sacral supports so stretched that they cannot recover their contractile
power, and with involution of all the pelvic organs arrested, the descent
may not only persist, but may even progress with constantly increasing
cystocele to the third degree of prolapse. The downward influence of the
above conditions may be materially increased by rupture of the perineum,
and consequent prolapse of the recto-vaginal wall into a pouch called
rectocele.
7 The anteflexion of the multiparous uterus is less than that of the virgin.

In the great majority of cases of complete prolapse the posterior vaginal

wall in its descent is peeled off from the rectum, leaving the latter in its
normal position. In rare instances the lower portion of the rectum is also
found to have extruded in extreme rectocele, making a pouch below and in
front of the anus, where fecal matter may accumulate and remain in hard
scybalæ.

Obviously, complete prolapse of the uterus is only an incident to the

prolapse of the pelvic floor. The whole mechanism is in all respects
analogous to that of hernia. The extruded mass drags after it a peritoneal
sac, which, hernia-like, contains small intestine. This sac forces its way to
the pelvic outlet and extrudes through the vulva, having the inverted
vagina for its covering.

FIG. 3.
First Degree of Prolapse of the Post-partum Uterus. The posterior vaginal wall has been
changed from its normal forward direction to a vertical direction by perineal rupture and
anterior displacement of the cervix; the vesico-vaginal wall descends in cystocele, becomes
hypertrophied, and drags the heavy uterus after it. The descending uterus carries with it a
reduplication of the vaginal walls.
In descent of the first degree the location of the uterus is either changed
to a lower level, the position remaining normal, or, as is more common, the
cervix having moved nearer to the symphysis and the organ turns back into
retroversion. In a given case suppose the vaginal walls from some cause to
have become relaxed and to have settled to a lower level in the pelvis. As
an associated fact the uterus to which these walls are attached must then
also occupy a place correspondingly nearer to the vulva—i.e. the location
of the uterus has changed, so that space enough intervenes between it
and the hollow of the sacrum for the former to turn back into the position
of retroversion or retroflexion. If, on the contrary, the descending uterus
still maintains its normal anteversion and anteflexion, it must occupy space
which belongs to the bladder. The vesical irritation consequent upon this
mal-location has generally been ascribed to the anteversion and
anteflexion, which are therefore oftentimes wrongly pronounced
pathological. The prompt relief which follows permanent replacement of
the organ in the normal location, even though in so doing its anteposition
be exaggerated, proves that the symptoms depend upon the mal-location,
not upon the anteposition. The importance of a clear distinction, therefore,
between location and position becomes apparent. Vesical irritation,
moreover, is sometimes caused by the dragging of the uterus upon the
neck of the bladder. This traction occurs not only in ascent, but also when
the organ descends below a certain level.

FIG. 4.
 Showing Extreme Descent of the Uterus and of the Pelvic Floor, and the Hernial
Character of the Lesion.

In the foregoing paragraphs traction due to the falling pelvic floor has been
discussed as a cause of descent. The impairment of the uterine supports
may, however, be such that instead of falling and dragging the uterus after
them, they simply permit it to descend along the vaginal canal by the force
of its own weight, and to carry with it the reduplicated vaginal walls. This
influence is generally enforced by the increased weight of the diseased
organ. The vagina more readily becomes a track for the descending uterus
when from any cause the normal forward direction of the vaginal canal
changes toward the vertical: this change may occur either as the result of
a forward displacement of its upper extremity, involving anteposition of the
cervix, or of a retro-displacement of its lower extremity in consequence of
rupture or subinvolution of the perineum. (See Fig. 3.) Descent in the track
of the vagina is obviously combined with some degree of retroversion,
because the axes of the uterus and vagina then correspond.

The PATHOLOGICAL ANATOMY may involve all the displaced organs. The
circulation throughout the pelvis is impeded by traction upon the vessels,
and the entire pelvic contents therefore become the subject of venous
congestion, with consequences disastrous to local innervation and
nutrition.

The ovaries may suffer concurrent displacement, with resulting

inflammatory and cystic enlargement. The peritoneum which enters into
the formation of the uterine ligaments and of the pelvic floor is dragged
along with the uterus.

The vagina is hypertrophied and swollen. Its mucous membrane becomes

the seat of acute vaginitis and chronic catarrh. In the third degree of
descent the exposed vagina, no longer lubricated by the normal secretions
of the uterus, becomes dry, parchment-like, oedematous, eroded, and
ulcerated. Sometimes the cul-de-sac of Douglas is distended by downward
pressure of the intestines, by a small tumor, or by ascitic fluid, and a
consequent hernial sac may protrude into the vagina through some portion
of the posterior vaginal fornix. The anterior fornix is subject to a similar
accident. These conditions are designated enterocele vaginalis, anterior
and posterior.

The rectum and bladder are subject to inflammation and chronic catarrh,
and the bladder especially to concurrent descent. The uterus may be
enlarged from any one or all of a variety of causes—congestion,
subinvolution, hypertrophy, and hyperplasia. Its cervix is often the seat of
extreme erosion or so-called ulceration. The endometrium, in order to
relieve the organ of its surplus blood, gives forth an excessive secretion of
mucus, which upon being increased in quantity becomes vitiated in quality.
This is termed uterine catarrh. The enlargement of the uterus often
pertains more to the cervix than to the body, especially in prolapse of the
second and third degrees. An explanation of this may be found in Figs. 5
and 6.
 FIG. 5. FIG. 6.

Descent of the Virgin Uterus into the Vaginal Descent of the Uterus, showing Excessive Circular
Canal, showing the Reduplicated Vaginal Walls. Enlargement of the Lacerated Cervix, consequent
The utero-vaginal attachment, points X and Z, upon Reduplication of the Vaginal Walls and Out-
appears to be at X' and Z'. The apparent increase rolling of Intracervical Tissues. The divided
of length in the vaginal portion of the cervix due fragments of the os externum are at a and b. The
to the reduplication is measured by the distance curved lines forming the angles 1, 2, 3, 4, and 5
from X and Z to X' and Z'. indicate the gradual process of the eversion. The
angle of the laceration at point 1 has been forced
by the swelling and out-rolling of the mucous and
submucous tissues of the cervix to point 5. The
apparent os externum is at point 5. The utero-
vaginal attachment X and Z seems to be at X'
and Z'. The vaginal portion of the cervix therefore
appears much larger and longer than it actually
is.
Apparent elongation and disproportionate circular enlargement of the
cervix are conditions which almost every standard author wrongly calls
hypertrophic elongation and circular hypertrophy. The question of
elongation is easily settled by placing the patient in the knee-chest
position. Then the uterus by its own weight falls toward the diaphragm, the
vagina unfolds, and the apparent utero-vaginal attachment X' Z' (Figs. 5
and 6) disappears, disclosing the actual attachment, X Z. Further, the point
of the sound, passed into the bladder while the cervix is exposed by Sims's
speculum, may be placed against the anterior wall of the cervix at Z, which
would be impossible if the attachment were at Z'.

The comparatively small amount of hypertrophy in disproportionate circular

enlargement is proved by the operation of trachelorraphy or by bringing
the points a and b (Fig. 6) together with uterine tenacula, the organ being
exposed by Sims's speculum. Then the out-rolled intracervical mucous
tissues are rolled back, the proper diameter of the cervix is restored, and a
laceration on one or both sides, extending past the vaginal attachment,
becomes apparent.

Hypertrophy or hyperplasia usually causes a nearly symmetrical

enlargement of the entire organ. At any rate, those cases in which the
reduplication of the vaginal walls does not almost entirely explain the great
elongation so called, or in which great disproportionate circular
enlargement has not been caused by laceration of the cervix, are the rare
exceptions. The great merit of having secured general assent to the
foregoing proposition, and of having given to the subject a new and right
direction, must be accorded to Emmet. The cervix now is seldom
amputated except for malignant disease.

Congestion of the uterus consequent upon obstruction in the stretched and

displaced veins is often so extreme as to induce a state analogous to
erection. Measurements by the probe just before and a few minutes after
replacement generally show an appreciable decrease in the length of the
uterine canal. If the prolapse has been of the third degree, the difference
may amount to one or even two inches. It is important not to confound the
enlargement of congestion with increase in the solid constituents of the
organ.
SYMPTOMS AND COURSE.—A dragging sensation and pelvic and abdominal pain
are generally present. Rectocele and cystocele and rectal and vesical
catarrh often cause painful and severe functional disturbances of the
rectum and bladder. In descent of the third degree excoriations of the
exposed vagina and cervix sometimes cause extreme suffering. The course
is ordinarily chronic, but attacks of acute vaginitis and pelvic peritonitis are
not uncommon. The peritonitis sometimes effects a spontaneous cure by
peritoneal adhesions which fasten the uterus in an elevated position and
hold it permanently. The symptoms of descent may be so severe as to
necessitate absolute rest in bed. In other cases they are often attended
with very little discomfort.

DIAGNOSIS is by inspection, palpation, and exploration. The prolapsed uterus

may be distinguished from cystocele, rectocele, inverted uterus, and fibroid
tumor by the presence of the os externum. The sound may be passed
through the urethra into the cystocele, and the finger through the anus
into the rectocele. The length of the uterus may be determined by the
sound, the size, shape, position, extent of descent, and difficulty of
replacement by conjoined manipulation.

PROPHYLAXIS.—This requires such measures during labor as may be

necessary to prevent long and powerful pressure upon the pelvic floor.
After labor any injury to the perineum should be promptly repaired. The
vagina should be kept clean by irrigations. The urine, if necessary, should
be regularly drawn and the bowels moved daily without straining. If
conditions be present likely to induce subinvolution—such, for example, as
pelvic inflammation or laceration of the cervix—they should receive
treatment at the proper time. Undue relaxation of the pelvic floor
necessitates a more prolonged rest in bed, the use of astringent douches,
and the application of a pessary when the patient resumes the upright
position.

TREATMENT.—The first indication is replacement, which in the first and

second degree of descent is not difficult unless the uterus be held down by
cicatrices or by a tumor. Complicating pelvic cellulitis and peritonitis may
render replacement dangerous or impossible, and may for a time
contraindicate all direct treatment. Replacement of the organs from the
third degree of prolapse is accomplished in the inverse order of their
descent: first, the posterior vaginal wall, then the uterus, and last the
anterior vaginal wall. Not infrequently the completely prolapsed uterus and
pelvic floor, hernia-like, become strangulated. Then taxis will usually suffice
if supplemented by hot applications, elastic pressure, anodynes, and the
knee-chest position. Should these fail anæsthesia may be required.

Undue pressure from above should if possible be removed. The clothing

should be loose, and the weight of the skirts supported from the shoulders
either by straps or preferably by buttoning them upon a waist made for the
purpose. This waist is a good substitute for the corset, which under all
circumstances and in all its forms is injurious. Increased uterine weight
from subinvolution or congestion is to be overcome by appropriate means.
Enlargement of the uterus when due to hypertrophy or hyperplasia is
generally incurable. Amputation of the cervix for what was formerly
considered circular hypertrophy and hypertrophic elongation is now seldom
or never required for the purpose of decreasing uterine weight. Amputation
except for malignant disease has given place to the operation of
trachelorraphy. Tumors exerting pressure above or traction below should if
possible be removed. Regulation of the bowels and general tonics are
usually necessary. The knee-chest position assumed several times a day
causes the uterus to gravitate toward the diaphragm, and thereby gives
temporary rest to the overburdened supports. While in this position the
patient should separate the labia, so that the air may rush in and the
vagina become expanded. The measures enumerated above, together with
rigid care of the diet and of such other hygienic requirements as the
individual case may demand, are essential as adjuvants to the more special
treatment which almost every case requires.

In exceptional cases of sudden descent, even to the third degree,

replacement alone is sometimes followed by permanent relief; but if the
descent has been gradual it always recurs immediately after replacement.
Measures are therefore required for the maintenance of the uterus in its
normal location and position. This indication is fulfilled by pessaries and by
operations.

Pessaries.—The function of the pessary is not only to maintain the uterus

on the health level in its normal location, but also, if possible, in its normal
position, which requires the cervix to be about one inch from the sacrum.
The cervix being thus placed, the organ cannot turn back into retroversion,
because in so doing the fundus would encounter the sacrum. The direction
of least resistance would then be forward into the normal anterior position.
The application of the pessary is then based upon the general proposition
that if the cervix be normally placed the body of the uterus will in the
absence of complications take care of itself. Since the vagina at its upper
extremity is attached to the cervix, displacement of the latter is clearly
impossible if the upper extremity of the vagina be sustained in its normal
location. The pessary restores and maintains the relations of the relaxed
vaginal walls by crowding the posterior vaginal cul-de-sac backward into
the hollow of the sacrum. It thereby also holds the attached cervix within a
proper distance of the sacrum. The Hodge pessary or some modifications
thereof fulfils this purpose in ordinary cases more satisfactorily than any
other.

FIG. 7.

The Emmet Curves.

FIG. 8.

The Albert Smith Curves.

The curves of the pessary demand careful attention in its application.
When the uterus is below the normal level, the broad ligaments are
necessarily rendered more tense than natural, and the blood-vessels, more
especially the veins, which are looped one upon the other, and which
traverse these ligaments to and from the uterus, are made to collapse. This
causes venous congestion and consequent increase in weight of the uterus
—a condition favorable to malposition, uterine catarrh, and pathological
changes in structure. A pessary which will raise the uterus to the health
level clearly fulfils an indication. A pessary which raises it above the health
level renders the broad ligaments tense and reproduces a condition which
it was designed to relieve. Maintenance of the uterus upon the health level
depends largely upon the curves of the pessary. The accompanying cuts
illustrate the shape and curve of the Hodge pessary as modified by Emmet
and Albert Smith. Fig. 7 represents the curve of Emmet, and Fig. 8 that of
Albert Smith. For convenience let us characterize that curve which rests in
the posterior vaginal cul-de-sac as the uterine curve, and that which
occupies that part of the vagina adjacent to the pubis the pubic curve. The
acuteness and length of the uterine curve determine the height to which
the pessary will lift the uterus. The longer and more acute the curve, the
higher the uterus will be lifted, and vice versâ. The smaller curve of the
Emmet modification will answer the average indication more nearly than
the sharper curve of the Albert Smith modification, which may lift the
uterus too high. The pubic should generally be proportioned to the uterine
curve; that is, the greater the uterine, the greater the pubic curve. A
pessary properly adjusted in all other respects may, by pressure upon the
urethra and neck of the bladder, create vesical tenesmus and urethral
irritation. This calls for increase in the pubic curve. The pubic curve may,
however, be so great that the lower part of the pessary occupies the centre
of the vulva, where it may create irritation. For this condition lessening of
the pubic curve is the remedy. The pessary should not be so wide as to
distend the vagina. Its length should be measured by the distance from the
lower extremity of the symphysis pubis to the posterior vaginal cul-de-sac,
less the thickness of the finger. If properly adjusted it should sustain the
pelvic floor in its normal relations and the uterus in stable equilibrium.

The uterus in the first and second degrees of descent is usually either
retroverted or retroflexed. The reader is therefore referred to the remarks
on the application of pessaries in the treatment of these displacements.
In advance prolapse dependent upon extensive injuries to the perineum
and other parts of the pelvic floor, and usually associated with extreme
subinvolution of all the pelvic organs, the axis of the vagina is often
changed from its forward oblique to the vertical direction. (See Fig. 3.) The
downward traction of the prolapsing cystocele and rectocele upon the
fornix of the vagina may then be so great that the pessary is inadequate to
maintain in place the upper extremity of the vagina. The cervix then moves
forward, the corpus turns back, and the whole uterus easily descends in a
vertical direction along the prolapsing walls of the vagina to the second or
third degree of prolapse. In this condition pessaries which disappear within
the vagina are liable to be forced out with the prolapsing pelvic floor, or if
retained seldom maintain the uterus in position. In such cases the various
cup pessaries which are supplied with external attachments and abdominal
belts are often used, but they are inadequate, because they either so fix
the uterus as to prevent its normal movements, or they hold it in such
unstable equilibrium that it may assume any one of the various
malpositions, anterior, posterior, or lateral; and they are open to the further
serious objection of constantly reminding the patient of their presence. As
an expedient the uterus may sometimes be held within the pelvis by means
of a large Albert Smith pessary with extreme uterine and pubic curves. The
rational treatment, however, requires first an operation on the anterior
vaginal wall to restore the fornix of the vagina to its normal place in the
hollow of the sacrum, and with it the attached cervix; and second, an
operation at the vaginal outlet to bring the posterior wall in contact with
the anterior, and thereby to restore the lower extremity of the vagina to its
normal place under the pubis.

ANTERIOR ELYTRORRHAPHY.—Numerous operations on the vaginal walls have

been devised for the purpose of narrowing the vagina, and thus preventing
descent along the vaginal canal, but they are temporary in their results,
because, as long as the direction of the vagina remains vertical, its walls
again become dilated by the prolapsing uterus and the former condition is
re-established. The operation to be effective is performed as follows: A
Sims's speculum of long blade, perforated at its extreme end, to which the
cervix has been attached by a piece of silver wire, passing through the
perforation and the posterior lip, is introduced, the patient being in Sims's
position. The cervix is thereby drawn by the point of the speculum far back
into the hollow of the sacrum. The author finds this preferable to the
method described by Emmet, who has the cervix held back by a sponge
probang in the hand of an assistant. The space in the anterior part of the
pelvis is now so increased that the uterus readily falls forward into decided
anteversion. While the uterus is thus held in position by its attachment to
the blade of the speculum, the operator with two uterine tenacula finds in
the loose vaginal tissue on either side of the cervix two points which can
be brought together in front of the cervix. Then at each of the two lateral
points a surface is denuded with the curved scissors about one-half inch
square, and in front of the cervix a surface an inch long by half an inch
wide across the anterior vaginal wall close to the uterine attachment. A No.
26 silver-wire suture is then passed, as shown in Fig. 9, and twisted as
shown in Fig. 10, so as to secure the lateral denuded surfaces in contact
with the larger surface in front of the cervix.

FIG. 9.

The First Suture before Twisting in Emmet's Operation for Procidentia (Emmet).

FIG. 10.
 Folds on the Anterior Vaginal Wall formed after
Twisting the First Suture (Emmet).

Inasmuch as the operation often fails at the point of the first suture, the
author has usually introduced two or three of this kind instead of one. Two
longitudinal folds are now formed on the anterior vaginal wall, which serve
as guides for denuding and turning in the remaining redundant tissue by a
line of sutures, which should extend forward along the centre of the
vesico-vaginal wall until the folds are lost in the vaginal surface near the
neck of the bladder. Sometimes the redundant tissue about the urethra
cannot be disposed of by turning it in from side to side. Then it is desirable
to make a crescentic denudation across the lower portion of the vagina, its
concavity being on the uterine side, and to unite the margins below to
those above by means of a curved line of sutures. The completed
operation is shown in Fig. 11.

FIG. 11.
 Emmet's Operation for Procidentia and Urethrocele completed. Sims's Speculum, Left Latero-prone
Position (Emmet).

The after-treatment requires the self-retaining Sims's sigmoid catheter in

the urethra for a week or frequent catheterization, absolute rest in bed,
hot-water vaginal douches, regulation of the bowels, and the removal of
the sutures on the twelfth day. After the completion of the operation the
cervix is maintained near the hollow of the sacrum, and the organ remains
normally anteverted and anteflexed, making an acute angle with the
vesico-vaginal wall, which has now been restored to its normal direction
and length. Unfortunately, it is not unusual to abandon the patient after
this operation, in the vain hope that the uterus and anterior vaginal wall
will maintain their normal relations without the support of the perineum
and posterior vaginal wall. This is a great mistake, because the cystocele
and procidentia almost always completely reappear within a few months.
Anterior elytrorrhaphy, therefore, is simply one of the steps in the
treatment.
PERINEORRHAPHY.—This is the name usually applied to the repair of the
ruptured perineum, but the scope of the operation has been extended to
include also the surgical treatment of rectocele and relaxation of the
posterior vaginal wall. The most scientific operation yet devised is the one
proposed by Emmet,8 which is performed as follows: The patient being
etherized and in the lithotomy position, the operator seizes with a
tenaculum the crest of the rectocele or posterior vaginal wall at a point
which can be drawn forward without undue traction—point a. With another
tenaculum the lowest caruncle or vestige of the hymen (point b), and with
another the posterior commissure of the vulva (point c), are hooked up.
The triangle included between these points defines one-half of the surface
to be denuded. The three tenacula are now placed in the hands of
assistants, the sides of the triangle are made tense by traction, and the
included surface denuded. The tenaculum at c is then removed, and the
middle point of the line a b is caught and drawn toward the interior of the
vagina in the direction of the vaginal sulcus on that side, and the sutures
are introduced, as in Fig. 13. The same thing is then repeated on the other
side, and the sutures are all tightened, forming a line of union running
back into each sulcus, as shown in Fig. 14.
8 Trans. Am. Gynæcological Society, 1883; Principles and Practice of Gynecology, 3d ed.

FIG. 12.
a is at the crest of the rectocele; b at the caruncle just within the labium; and c at the posterior
commissure. The cut represents that half of the surface to be denuded which is on the operator's
right. The dotted lines represent the other half, on the left.

FIG. 13.
The Sutures in Place. When secured they will unite a d with b d, and
lift the perineum up in contact with the anterior vaginal wall.

FIG. 14.
 All the Vaginal Sutures Twisted. One suture, including the crest of the
rectocele and the labium majus on either side, and three superficial
external sutures, are yet to be secured. The lines a d and d b, Fig. 13,
have been brought into coincidence by means of the sutures, and now
form the line of union d b. The tissues between the lines a c and c b,
Fig. 13, have been so lifted up and are so held under the line of union
d b that the line c b, Fig. 13, has been reduced to c b, Fig. 14, which
makes the external portion of the wound insignificant in extent.

The essential part of the operation inside the vagina almost always
succeeds, but the external part of the rupture at the posterior commissure
often fails to unite; furthermore, the operation as described by Emmet
does not overcome the patulous condition of the introitus vaginæ in case
of great relaxation of the vagina. The author has sought to obviate the first
of these difficulties by the use of deep silver sutures instead of the
superficial ones described by Emmet. They should be introduced before
tightening the vaginal sutures, and should be passed far around in the
posterior vaginal wall, their points of entrance and exit being the same as
for the three lower unsecured superficial external sutures in Fig. 14. The
second difficulty may be overcome by further denuding a triangular surface
in the vaginal sulcus on each side, the base of the triangle corresponding
to the line a b, Fig. 12, and its apex being in the vaginal sulcus at a
distance corresponding to the degree of relaxation. This increases the
length of the lines of union running into the sulci represented by d b and e
f, Fig. 14. In the vaginal portion of the wound silk or catgut is preferable to
silver, the latter being difficult to remove.

Emmet is entitled to great credit for having given to the profession an

operation which brings the posterior vaginal walls up against the anterior
more perfectly than any other, and which, being mostly inside of the
vagina, is therefore followed by very little of the pain during convalescence
which formerly rendered perineorrhaphy one of the most trying operations
in gynecology. The operation furthermore has demonstrated the former
teachings relative to the direction of perineal rupture9 and the tissues
involved to be incorrect, or at least inadequate.
9 At the meeting of the American Medical Association in June, 1883, the author presented a
paper describing the transverse laceration of the perineum and its operative treatment, which
was published with illustrations in the transactions by the journal of the Association, Dec. 22,
1883. This communication referred only to the recent rupture and the immediate operation.

Retroversion.

Retroversion is that position of the uterus in which the fundus is posterior

to the axis of the pelvic inlet. If the cervix be in its normal place near the
sacrum, retroversion is scarcely possible, because it is prevented by the
proximity of the over-arching sacrum. (See Fig. 2.) The first degree of
prolapse must therefore precede any considerable backward turning of the
uterus. When the cervix has been displaced downward and forward so far
that its distance from the sacrum is equal to or greater than the length of
the uterus, retroversion to any extent becomes possible. (See Figs. 3 and
16.)

ETIOLOGY AND HISTORY.—From the above it follows that the causes of

commencing retroversion must be identical with the causes of the first
degree of prolapse. After the puerperium the relaxation of the supports
and the weight of the organ may persist, and spontaneous replacement
may be prevented by the pressure and weight of the intestines upon the
anterior surface. Every act of defecation forces the cervix forward and
downward, and the uterus, being in the axis of the vagina, and having
therefore little support below, must depend upon the subinvoluted
peritoneal suspensory ligaments and pelvic fascia, which are inadequate.
This condition is very often induced by abortions, with resulting increased
weight and relaxation of the vaginal walls. Local peritonitis and cellulitis
may permanently fix the corpus in its retroverted position by cicatricial
bands and adhesions.

SYMPTOMS AND COURSE.—The displacement and its complications usually

cause bearing-down sensations, a feeling of heaviness in the pelvis,
exhaustion upon walking and standing, especially the latter, and
constipation. After the puerperium the extreme engorgement of the pelvic
organs often produces uterine hemorrhage, which should not be
confounded with the returning menstruation. Especially after abortion the
hemorrhage often persists for a long time unless cured by treatment.
Gradual or sudden replacement may occur spontaneously, or the causes
may continue active, and even be enforced by cystocele and rectocele. The
displacement may also be complicated by disease and displacement of the
ovaries. Organic disease of the uterine walls may induce a superadded
retroflexion. The heavy organ may descend along the relaxed subinvoluted
vaginal walls even to complete procidentia.

DIAGNOSIS AND PROGNOSIS.—The symptoms outlined in the preceding

paragraph indicate the probability of displacement, but the diagnosis
depends upon direct examination of the uterus. Conjoined manipulation
and the probe will usually show the retroverted organ with the cervix
displaced toward the pubes and with the corpus in the hollow of the
sacrum. The introduction of the probe is contraindicated by cellulitis and
peritonitis. In certain cases of anteflexion, as represented in Fig. 23, the
cervix is bent forward in the vaginal axis as in retroversion. The condition is
in reality one of retroversion of the cervix with high anteflexion of the
corpus, which may usually be detected by careful conjoined examination.
The prognosis with treatment is generally favorable both for speedy relief
and ultimate recovery.
TREATMENT.—As in descent, the treatment consists in removing cellulitis,
peritonitis, and other complications, in the use of pessaries, and in
operations on the anterior and posterior vaginal walls if needed. Inasmuch
as the treatment corresponds to that of retroflexion, it will be presented
under that subject.

Retroflexion.

ETIOLOGY AND PATHOLOGY.—Retroflexion is that displacement in which the

organ is bent backward upon itself. It usually results from, and is
associated with, retroversion, but for convenience the double displacement
will be termed retroflexion. It may be caused by the great weight of the
corpus, the soft flexible state of the uterine walls during and after
involution, intra-abdominal forces, downward pressure during defecation,
tight clothing, and not commonly by the obstetric bandage.

The ovaries, unless fixed elsewhere by adhesions, are displaced with, and
held down on either side of, the corpus, sometimes enlarged from
inflammation, often adherent, and always extremely sensitive. Chronic
metritis, cellulitis, and peritonitis, with adhesions more or less firm, are
usually present, and not infrequently as the result of gonorrhoea, abortion,
or injudicious treatment. Peritoneal adhesions between the corpus and the
cul-de-sac of Douglas sometimes make replacement impossible. In rare
cases the displacement is congenital.

FIG. 15.
 Extreme Retroflexion, with Hypertrophy of the Corpus, which impinges upon the rectum and
compresses the recto-vaginal wall.

SYMPTOMS AND COURSE.—Among the most pronounced symptoms are profuse

uterine catarrh, menstrual disorders, sterility, abortion, weakness, pain in
the back, painful defecation, rectal tenesmus, the symptoms of pelvic
inflammation, neurasthenia, and other nervous symptoms. The uterine
catarrh is due to an effort on the part of the engorged pelvic organs to
relieve themselves by an exaggerated secretion of mucus from the uterus,
which upon being increased in quantity becomes vitiated in quality, and
therefore pathological. Menorrhagia and abortion may also result from
congestion. Dysmenorrhoea and sterility result from the general anæmic
condition and from the inflammatory complications, and from the
obstruction in the uterine canal or in the blood-vessels at the angle of
flexure. (See Pathology of Anteflexion.) The rectal symptoms are caused by
the pressure of the corpus uteri upon the rectum, which gives the
sensation to the patient of an overloaded bowel.

Should pregnancy occur, the rapid growth of the uterus may induce
spontaneous reposition at about the fourth month, when the fundus rises
out of the pelvis, but if the corpus be incarcerated under the sacral
promontory from adhesions or from any other cause, the uterus will, unless
manually replaced, relieve itself by abortion.

Abdominal pains, nervous dyspepsia, and neuralgia in distant parts of the

body are often present; indeed, the nervous symptoms may be of the most
exaggerated character, and may comprise all that is implied by the word
hysteria in its most comprehensive signification.

DIAGNOSIS.—Digital touch discloses the cervix low in the pelvis, and the
fundus uteri is felt through the posterior vaginal wall in the cul-de-sac of
Douglas. Conjoined manipulation with the index finger of the left hand,
first in the vagina and then in the rectum, and the right hand over the
hypogastric region, will show the size, form, consistency, and location of
the uterus, the degree of the flexure, and the difficulty of replacement. An
inflammatory exudate or hæmatocele, posterior to the uterus, or a fibroid
in the posterior uterine wall, may be mistaken for the retroflexed corpus.
The probe will always verify the diagnosis, but if there be great tenderness
with fixation in the cul-de-sac of Douglas, treatment should be directed
against the inflamed condition, and the final diagnosis made by repeated
examinations or after the disappearance of the inflammation. Great and
lasting injury is often done in the attempt to complete the diagnosis at the
first examination. The presence of a fibroid in the posterior uterine wall
with post-uterine inflammation is a serious complication both in diagnosis
and treatment. If the rectum be overloaded with fecal matter, the diagnosis
should be deferred. The displacement is distinguished from the presence of
an ovary or small ovarian tumor in the pouch of Douglas by careful
bimanual examination and by the probe.

TREATMENT OF RETROVERSION AND RETROFLEXION.—The objects of treatment are

replacement and retention of the uterus. The obstacles to replacement are
cellulitis, peritonitis, and fixation of the uterus, and these complications
often require weeks, and in severe cases months, of treatment preparatory
to replacement. Some of the general therapeutic suggestions under the
subject of descent are also applicable to the retro-positions. Rest,
massage, careful regulation of the bowels, feeding, and general tonics are
essential. For the inflammation small blisters over the inguinal regions
frequently repeated, and the daily application of the cotton and glycerin
plug to the cervix, and dry cupping over the sacrum, are most efficacious.
The glycerin may be combined with alum, tannin, chloral hydrate, or
iodoform. Thymoline in small quantities partially destroys the disagreeable
iodoform odor. The most useful and essential topical application is the hot-
water vaginal douche, but its use will be followed by failure and
disappointment if it be applied in the ordinary way. The following is quoted
from a paper by the author which was published in the Chicago Medical
Gazette, Jan. 1, 1880:

"Ordinary Method of Application. "Proper Method of Application.

"I. Ordinarily, the douche is applied with the "I. It should invariably be given with the patient lying
patient in the sitting posture, so that the on the back, with the shoulders low, the knees drawn
injected water cannot fill the vagina and up, and the hips elevated on a bed-pan, so that the
bathe the cervix uteri, but, on the contrary, outlet of the vagina may be above every other part of
returns along the tube of the syringe as fast it. Then the vagina will be kept continually
as it flows in. overflowing while the douche is being given.
"II. The patient is seldom impressed with "II. It should be given at least twice every day,
the importance of regularity in its morning and evening, and generally the length of
administration. each application should not be less than twenty
minutes.
"III. The temperature is ordinarily not "III. The temperature should be as high as the
specified or heeded. patient can endure without distress. It may be
increased from day to day, from 100° or 105° to 115°
or 120° Fahr.
"IV. Ordinarily, the patient abandons its use "IV. Its use, in the majority of cases, should be
after a short time." continued for months at least, and sometimes for two
or three years. Perseverance is of prime importance."

"A satisfactory substitute for the bed-pan may be made as follows: Place
two chairs at the side of an ordinary bed with space enough between them
to admit a bucket; place a large pillow at the extreme side of the bed
nearest the chairs; spread an ordinary rubber sheet over the pillow, so that
one end of the sheet may fall into the bucket below in the form of a
trough. The douche may then be given with the patient's hips drawn well
out over the edge of the bed and resting on the pillow, and with one foot
on each chair; the water will then find its way along the rubber trough into
the bucket below." The Davidson syringe, which has an interrupted
current, is preferable to any of the fountain syringes.

As the tenderness disappears the cotton plugs may be increased in

quantity, and thereby made to serve as temporary support for the uterus
until a more permanent pessary can be substituted. The sluggish
circulation in the pelvis and torpid condition of the bowels may be much
relieved by the daily application of the wet pack. A small flannel sheet
folded lengthwise to the width of two feet, dipped in very hot water, and
dried by passing it through a wringer, is wound about the hips and covered
by another dry one. At the end of a half hour, during which time the
patient maintains the recumbent position, the sheets are removed. When
the tenderness has been sufficiently reduced, gentle attempts at
replacement may be made every day or two by conjoined manipulation.
The patient's tolerance of manipulation may thus be observed and the way
prepared for complete replacement and permanent retention after the
subsidence of the inflammation.

In retroversion and retroflexion always replace the uterus before adjusting

the pessary, otherwise the instrument will press upon the sensitive uterus,
when one of three unfortunate results must occur: (1) The pessary may
not be tolerated on account of pain; (2) the pessary may be forced down
by pressure from above so near to the vulva that it will fail to do the least
good; (3) the uterus, finding it impossible to hold its position against the
pessary, instead of taking its proper position will often be bent over it in
exaggerated retroflexion, with the cervix between the pessary and the
pubes and the body between the pessary and the sacrum, or the whole
organ may slip off to one side of the instrument into a malposition more
serious than the one for which relief is sought. The safest and most
effective method of replacement is by conjoined manipulation, as
represented in Figs. 16 and 17. The dotted lines in the former indicate the
gradual elevation of the corpus out of the hollow of the sacrum to the
pelvic brim, where it may be anteverted by the fingers of the right hand
pressed well down behind its posterior wall. During the process of
anteversion the index finger of the left hand in the anterior fornix of the
vagina presses the cervix back to its place in the hollow of the sacrum, as
in Fig. 17. Efficient reposition of the uterus is very often impossible without
anæsthesia.

FIG. 16.
Welcome to Our Bookstore - The Ultimate Destination for Book Lovers
Are you passionate about books and eager to explore new worlds of
knowledge? At our website, we offer a vast collection of books that
cater to every interest and age group. From classic literature to
specialized publications, self-help books, and children’s stories, we
have it all! Each book is a gateway to new adventures, helping you
expand your knowledge and nourish your soul
Experience Convenient and Enjoyable Book Shopping Our website is more
than just an online bookstore—it’s a bridge connecting readers to the
timeless values of culture and wisdom. With a sleek and user-friendly
interface and a smart search system, you can find your favorite books
quickly and easily. Enjoy special promotions, fast home delivery, and
a seamless shopping experience that saves you time and enhances your
love for reading.
Let us accompany you on the journey of exploring knowledge and
personal growth!

ebookgate.com