|
Preface |
9 |
|
|
Acknowledgements |
13 |
|
|
Contents |
15 |
|
|
PART I. INTRODUCTION |
23 |
|
|
1. Adaptive Filtering |
25 |
|
|
1.1 Linear Adaptive Filters |
27 |
|
|
1.1.1 Linear Adaptive Filter Algorithms |
29 |
|
|
1.2 Nonlinear Adaptive Filters |
31 |
|
|
1.2.1 Adaptive Volterra Filters |
31 |
|
|
1.3 Nonclassical Adaptive Systems |
32 |
|
|
1.3.1 Artificial Neural Networks |
32 |
|
|
1.3.2 Fuzzy Logic |
33 |
|
|
1.3.3 Genetic Algorithms |
33 |
|
|
1.4 A Brief History and Overview of Classical Theories |
34 |
|
|
1.4.1 Linear Estimation Theory |
34 |
|
|
1.4.2 Linear Adaptive Filters |
35 |
|
|
1.4.3 Adaptive Signal Processing Applications |
36 |
|
|
1.4.4 Adaptive Control |
38 |
|
|
1.5 A Brief History and Overview of Nonclassical Theories |
39 |
|
|
1.5.1 Artificial Neural Networks |
39 |
|
|
1.5.2 Fuzzy Logic |
40 |
|
|
1.5.3 Genetic Algorithms |
40 |
|
|
1.6 Fundamentals of Adaptive Networks |
41 |
|
|
1.7 Choice of Adaptive Filter Algorithm |
45 |
|
|
2. Linear Systems and Stochastic Processes |
47 |
|
|
2.1 Basic Concepts of Linear Systems |
49 |
|
|
2.2 Discrete-time Signals and Systems |
51 |
|
|
2.3 The Discrete Fourier Transform (DFT) |
53 |
|
|
2.3.1 Discrete Linear Convolution Using the DFT |
54 |
|
|
2.3.2 Digital Sampling Theory |
55 |
|
|
2.4 The Fast Fourier Transform (FFT) |
59 |
|
|
2.5 The z-Transform |
62 |
|
|
2.5.1 Relationship between Laplace Transform and z-Transform |
62 |
|
|
2.5.2 General Properties of the DFT and z-Transform |
66 |
|
|
2.6 Summary of Discrete-time LSI Systems |
68 |
|
|
2.7.1 Phase Response from Frequency Magnitude Response |
72 |
|
|
2.8 Linear Algebra Summary |
73 |
|
|
2.8.1 Vectors |
73 |
|
|
2.8.2 Linear Independence, Vector spaces, and Basis Vectors |
74 |
|
|
2.8.3 Matrices |
75 |
|
|
2.8.4 Linear Equations |
77 |
|
|
2.8.5 Special Matrices |
78 |
|
|
2.8.6 Quadratic and Hermitian Forms |
81 |
|
|
2.8.7 Eigenvalues and Eigenvectors |
81 |
|
|
2.9 Introduction to Stochastic Processes |
83 |
|
|
2.10 Random Signals |
85 |
|
|
2.11 Basic Descriptive Properties of Random Signals |
86 |
|
|
2.11.1 The Mean Square Value and Variance |
86 |
|
|
2.11.2 The Probability Density Function |
87 |
|
|
2.11.3 Jointly Distributed Random Variables |
90 |
|
|
2.11.4 The Expectation Operator |
90 |
|
|
2.11.5 The Autocorrelation and Related Functions |
91 |
|
|
2.11.6 Power Spectral Density Functions |
94 |
|
|
2.11.7 Coherence Function |
95 |
|
|
2.11.8 Discrete Ergodic Random Signal Statistics |
96 |
|
|
2.11.9 Autocovariance and Autocorrelation Matrices |
97 |
|
|
2.11.10 Spectrum of a Random Process |
98 |
|
|
2.11.11 Filtering of Random Processes |
100 |
|
|
2.11.12 Important Examples of Random Processes |
102 |
|
|
2.12 Exercises |
104 |
|
|
2.12.1 Problems |
104 |
|
|
PART II. MODELLING |
109 |
|
|
3. Optimisation and Least Squares Estimation |
111 |
|
|
3.1 Optimisation Theory |
111 |
|
|
3.2 Optimisation Methods in Digital Filter Design |
113 |
|
|
3.3 Least Squares Estimation |
117 |
|
|
3.4 Least Squares Maximum Likelihood Estimator |
119 |
|
|
3.5 Linear Regression - Fitting Data to a Line |
120 |
|
|
3.6 General Linear Least Squares |
121 |
|
|
3.7 A Ship Positioning Example of LSE |
122 |
|
|
3.9 Measure of LSE Precision |
130 |
|
|
3.10 Measure of LSE Reliability |
131 |
|
|
3.11 Limitations of LSE |
132 |
|
|
3.12 Advantages of LSE |
132 |
|
|
3.13 The Singular Value Decomposition |
133 |
|
|
3.13.1 The Pseudoinverse |
134 |
|
|
3.13.2 Computation of the SVD |
134 |
|
|
3.14 Exercises |
138 |
|
|
3.14.1 Problems |
138 |
|
|
4. Parametric Signal and System Modelling |
141 |
|
|
4.1 The Estimation Problem |
142 |
|
|
4.2 Deterministic Signal and SystemModelling |
143 |
|
|
4.2.1 The Least Squares Method |
144 |
|
|
4.2.2 The Padé Approximation Method |
146 |
|
|
4.2.3 Prony’s Method |
149 |
|
|
4.2.4 Autocorrelation and Covariance Methods |
155 |
|
|
4.3 Stochastic Signal Modelling |
159 |
|
|
4.3.1 Autoregressive Moving Average Models |
159 |
|
|
4.3.2 Autoregressive Models |
161 |
|
|
4.3.3 Moving Average Models |
162 |
|
|
4.4 The Levinson-Durbin Recursion and Lattice Filters |
163 |
|
|
4.4.1 The Levinson-Durbin Recursion Development |
164 |
|
|
4.4.2 The Lattice Filter |
168 |
|
|
4.4.3 The Cholesky Decomposition |
171 |
|
|
4.4.4 The Levinson Recursion |
173 |
|
|
4.5 Exercises |
176 |
|
|
4.5.1 Problems |
176 |
|
|
PART III. CLASSICAL FILTERS and SPECTRALANALYSIS |
179 |
|
|
5. OptimumWiener Filter |
181 |
|
|
5.1 Derivation of the Ideal Continuous-time Wiener Filter |
182 |
|
|
5.2 The Ideal Discrete-time FIR Wiener Filter |
184 |
|
|
5.2.1 General Noise FIR Wiener Filtering |
186 |
|
|
5.2.2 FIRWiener Linear Prediction |
187 |
|
|
5.3 Discrete-time Causal IIR Wiener Filter |
189 |
|
|
5.3.1 Causal IIR Wiener Filtering |
191 |
|
|
5.3.2 Wiener Deconvolution |
192 |
|
|
5.4 Exercises |
193 |
|
|
5.4.1 Problems |
193 |
|
|
6. Optimum Kalman Filter |
195 |
|
|
6.1 Background to The Kalman Filter |
195 |
|
|
6.2 The Kalman Filter |
196 |
|
|
6.2.1 Kalman Filter Examples |
203 |
|
|
6.3 Kalman Filter for Ship Motion |
207 |
|
|
6.3.1 Kalman Tracking Filter Proper |
208 |
|
|
6.3.2 Simple Example of a Dynamic Ship Model |
211 |
|
|
6.3.3 Stochastic Models |
214 |
|
|
6.3.4 Alternate Solution Methods |
214 |
|
|
6.3.5 Advantages of Kalman Filtering |
215 |
|
|
6.3.6 Disadvantage of Kalman Filtering |
215 |
|
|
6.4 Extended Kalman Filter |
216 |
|
|
6.5 Exercises |
216 |
|
|
6.5.1 Problems |
216 |
|
|
7. Power Spectral Density Analysis |
219 |
|
|
7.1 Power Spectral Density Estimation Techniques |
220 |
|
|
7.2 Nonparametric Spectral Density Estimation |
221 |
|
|
7.2.1 Periodogram Power Spectral Density Estimation |
221 |
|
|
7.2.2 Modified Periodogram - Data Windowing |
225 |
|
|
7.2.3 Bartlett’s Method - Periodogram Averaging |
227 |
|
|
7.2.4 Welch’s Method |
228 |
|
|
7.2.5 Blackman-Tukey Method |
230 |
|
|
7.2.6 Performance Comparisons of Nonparametric Methods |
231 |
|
|
7.2.7 Minimum Variance Method |
231 |
|
|
7.2.8 Maximum Entropy (All Poles) Method |
234 |
|
|
7.3 Parametric Spectral Density Estimation |
237 |
|
|
7.3.1 Autoregressive Methods |
237 |
|
|
7.3.2 Moving Average Method |
240 |
|
|
7.3.3 Autoregressive Moving Average Method |
241 |
|
|
7.3.4 Harmonic Methods |
241 |
|
|
7.4 Exercises |
245 |
|
|
7.4.1 Problems |
245 |
|
|
PART IV. ADAPTIVE FILTER THEORY |
247 |
|
|
8. Adaptive Finite Impulse Response Filters |
249 |
|
|
8.1 Adaptive Interference Cancelling |
250 |
|
|
8.2 Least Mean Squares Adaptation |
252 |
|
|
8.2.1 Optimum Wiener Solution |
253 |
|
|
8.2.2 The Method of Steepest Gradient Descent Solution |
255 |
|
|
8.2.3 The LMS Algorithm Solution |
257 |
|
|
8.2.4 Stability of the LMS Algorithm |
259 |
|
|
8.2.5 The Normalised LMS Algorithm |
261 |
|
|
8.3 Recursive Least Squares Estimation |
261 |
|
|
8.3.1 The Exponentially Weighted Recursive Least Squares Algorithm |
262 |
|
|
8.3.2 Recursive Least Squares Algorithm Convergence |
265 |
|
|
8.3.3 The RLS Algorithm as a Kalman Filter |
266 |
|
|
8.4 Exercises |
267 |
|
|
8.4.1 Problems |
267 |
|
|
9. Frequency Domain Adaptive Filters |
269 |
|
|
9.1 Frequency Domain Processing |
269 |
|
|
9.1.1 Time Domain Block Adaptive Filtering |
270 |
|
|
9.1.2 Frequency Domain Adaptive Filtering |
271 |
|
|
9.2 Exercises |
278 |
|
|
9.2.1 Problems |
278 |
|
|
10. Adaptive Volterra Filters |
279 |
|
|
10.1 Nonlinear Filters |
279 |
|
|
10.2 The Volterra Series Expansion |
281 |
|
|
10.3 A LMS Adaptive Second-order Volterra Filter |
281 |
|
|
10.4 A LMS Adaptive Quadratic Filter |
283 |
|
|
10.5 A RLS Adaptive Quadratic Filter |
284 |
|
|
10.6 Exercises |
286 |
|
|
10.6.1 Problems |
286 |
|
|
11. Adaptive Control Systems |
289 |
|
|
11.1 Main Theoretical Issues |
290 |
|
|
11.2 Introduction to Model-reference Adaptive Systems |
292 |
|
|
11.2.1 The Gradient Approach |
293 |
|
|
11.2.2 Least Squares Estimation |
295 |
|
|
11.2.3 A General Single-input-single-output MRAS |
296 |
|
|
11.2.4 Lyapunov’s Stability Theory |
299 |
|
|
11.3 Introduction to Self-tuning Regulators |
302 |
|
|
11.3.1 Indirect Self-tuning Regulators |
304 |
|
|
11.3.2 Direct Self-tuning Regulators |
305 |
|
|
11.4 Relations between MRAS and STR |
306 |
|
|
11.5 Applications |
307 |
|
|
PART V. NONCLASSICALADAPTIVE SYSTEMS |
309 |
|
|
12. Introduction to Neural Networks |
311 |
|
|
12.1 Artificial Neural Networks |
311 |
|
|
12.1.1 Definitions |
312 |
|
|
12.1.2 Three Main Types |
312 |
|
|
12.1.3 Specific Artificial Neural Network Paradigms |
314 |
|
|
12.1.4 Artificial Neural Networks as Black Boxes |
315 |
|
|
12.1.5 Implementation of Artificial Neural Networks |
316 |
|
|
12.1.6 When to Use an Artificial Neural Network |
317 |
|
|
12.1.7 How to Use an Artificial Neural Network |
317 |
|
|
12.1.8 Artificial Neural Network General Applications |
318 |
|
|
12.1.9 Simple Application Examples |
319 |
|
|
12.2 A Three-layer Multi-layer Perceptron Model |
322 |
|
|
12.2.1 MLP Backpropagation-of-error Learning |
324 |
|
|
12.2.2 Derivation of Backpropagation-of-error Learning |
325 |
|
|
12.2.3 Notes on Classification and Function Mapping |
330 |
|
|
12.2.4 MLP Application and Training Issues |
330 |
|
|
12.3 Exercises |
332 |
|
|
12.3.1 Problems |
332 |
|
|
13. Introduction to Fuzzy Logic Systems |
335 |
|
|
13.1 Basic Fuzzy Logic |
335 |
|
|
13.1.1 Fuzzy Logic Membership Functions |
336 |
|
|
13.1.2 Fuzzy Logic Operations |
337 |
|
|
13.1.3 Fuzzy Logic Rules |
338 |
|
|
13.1.4 Fuzzy Logic Defuzzification |
339 |
|
|
13.2 Fuzzy Logic Control Design |
340 |
|
|
13.2.1 Fuzzy Logic Controllers |
341 |
|
|
13.3 Fuzzy Artificial Neural Networks |
344 |
|
|
13.4 Fuzzy Applications |
345 |
|
|
14. Introduction to Genetic Algorithms |
347 |
|
|
14.1 A General Genetic Algorithm |
348 |
|
|
14.2 The Common Hypothesis Representation |
349 |
|
|
14.3 Genetic Algorithm Operators |
351 |
|
|
14.4 Fitness Functions |
352 |
|
|
14.5 Hypothesis Searching |
352 |
|
|
14.6 Genetic Programming |
353 |
|
|
14.7 Applications of Genetic Programming |
354 |
|
|
14.7.1 Filter Circuit Design Application of GAs and GP |
355 |
|
|
14.7.2 Tic-tac-to Game Playing Application of GAs |
356 |
|
|
PART VI. ADAPTIVE FILTER APPLICATION |
359 |
|
|
15. Applications of Adaptive Signal Processing |
361 |
|
|
15.1 Adaptive Prediction |
362 |
|
|
15.2 Adaptive Modelling |
364 |
|
|
15.3 Adaptive Telephone Echo Cancelling |
365 |
|
|
15.4 Adaptive Equalisation of Communication Channels |
366 |
|
|
15.5 Adaptive Self-tuning Filters |
368 |
|
|
15.6 Adaptive Noise Cancelling |
368 |
|
|
15.7 Focused Time Delay Estimation for Ranging |
370 |
|
|
15.7.1 Adaptive Array Processing |
371 |
|
|
15.8 Other Adaptive Filter Applications |
372 |
|
|
15.8.1 Adaptive 3-D Sound Systems |
372 |
|
|
15.8.2 Microphone arrays |
373 |
|
|
15.8.3 Network and Acoustic Echo Cancellation |
374 |
|
|
15.8.4 Real-world Adaptive Filtering Applications |
375 |
|
|
16. Generic Adaptive Filter Structures |
377 |
|
|
16.1 Sub-band Adaptive Filters |
377 |
|
|
16.2 Sub-space Adaptive Filters |
380 |
|
|
16.2.1 MPNN Model |
382 |
|
|
16.2.2 Approximately Piecewise Linear Regression Model |
384 |
|
|
16.2.3 The Sub-space Adaptive Filter Model |
386 |
|
|
16.2.4 Example Applications of the SSAF Model |
388 |
|
|
16.3 Discussion and Overview of the SSAF |
392 |
|
|
References |
395 |
|
|
Index |
403 |
|
|
More eBook at www.ciando.com |
0 |
|