Muutke küpsiste eelistusi

E-raamat: Non-Gaussian Statistical Communication Theory

(West Midlands Probation Service, UK)
  • Formaat - PDF+DRM
  • Hind: 161,72 €*
  • * hind on lõplik, st. muud allahindlused enam ei rakendu
  • Lisa ostukorvi
  • Lisa soovinimekirja
  • See e-raamat on mõeldud ainult isiklikuks kasutamiseks. E-raamatuid ei saa tagastada.
  • Raamatukogudele

DRM piirangud

  • Kopeerimine (copy/paste):

    ei ole lubatud

  • Printimine:

    ei ole lubatud

  • Kasutamine:

    Digitaalõiguste kaitse (DRM)
    Kirjastus on väljastanud selle e-raamatu krüpteeritud kujul, mis tähendab, et selle lugemiseks peate installeerima spetsiaalse tarkvara. Samuti peate looma endale  Adobe ID Rohkem infot siin. E-raamatut saab lugeda 1 kasutaja ning alla laadida kuni 6'de seadmesse (kõik autoriseeritud sama Adobe ID-ga).

    Vajalik tarkvara
    Mobiilsetes seadmetes (telefon või tahvelarvuti) lugemiseks peate installeerima selle tasuta rakenduse: PocketBook Reader (iOS / Android)

    PC või Mac seadmes lugemiseks peate installima Adobe Digital Editionsi (Seeon tasuta rakendus spetsiaalselt e-raamatute lugemiseks. Seda ei tohi segamini ajada Adober Reader'iga, mis tõenäoliselt on juba teie arvutisse installeeritud )

    Seda e-raamatut ei saa lugeda Amazon Kindle's. 

The book is based on the observation that communication is the central operation of discovery in all the sciences. In its "active mode" we use it to "interrogate" the physical world, sending appropriate "signals" and receiving nature's "reply". In the "passive mode" we receive nature's signals directly. Since we never know a prioriwhat particular return signal will be forthcoming, we must necessarily adopt a probabilistic model of communication. This has developed over the approximately seventy years since it's beginning, into a Statistical Communication Theory (or SCT). Here it is the set or ensemble of possible results which is meaningful. From this ensemble we attempt to construct in the appropriate model format, based on our understanding of the observed physical data and on the associated statistical mechanism, analytically represented by suitable probability measures. Since its inception in the late '30's of the last century, and in particular subsequent to World War II, SCT has grown into a major field of study. As we have noted above, SCT is applicable to all branches of science. The latter itself is inherently and ultimately probabilistic at all levels. Moreover, in the natural world there is always a random background "noise" as well as an inherent a priori uncertainty in the presentation of deterministic observations, i.e. those which are specifically obtained, a posteriori.

The purpose of the book is to introduce Non-Gaussian statistical communication theory and demonstrate how the theory improves probabilistic model. The book was originally planed to include 24 chapters as seen in the table of preface. Dr. Middleton completed first 10 chapters prior to his passing in 2008. Bibliography which represents remaining chapters are put together by the author's close colleagues; Drs. Vincent Poor, Leon Cohen and John Anderson.

email pressbooks@ieee.org to request Ch.10

 
Foreword xv
Visualizing the Invisible xvii
Acknowledgments xxi
About the Author xxiii
Editor's Note xxv
Introduction 1(14)
1 Reception as a Statistical Decision Problem
15(62)
1.1 Signal Detection and Estimation
15(2)
1.2 Signal Detection and Estimation
17(5)
1.2.1 Detection
17(3)
1.2.2 Types of Extraction
20(1)
1.2.3 Other Reception Problems
21(1)
1.3 The Reception Situation in General Terms
22(5)
1.3.1 Assumptions: Space-Time Sampling
22(3)
1.3.2 The Decision Rule
25(1)
1.3.3 The Decision Problem
26(1)
1.3.4 The Generic Similarity of Detection and Extraction
27(1)
1.4 System Evaluation
27(8)
1.4.1 Evaluation Functions
27(3)
1.4.2 System Comparisons and Error Probabilities
30(1)
1.4.3 Optimization: Bayes Systems
31(1)
1.4.4 Optimization: Minimax Systems
32(3)
1.5 A Summary of Basic Definitions and Principal Theorems
35(5)
1.5.1 Some General Properties of Optimum Decision Rules
35(1)
1.5.2 Definitions
36(1)
1.5.3 Principal Theorems
37(1)
1.5.4 Remarks: Prior Probabilities, Cost Assignments, and System Invariants
38(2)
1.6 Preliminaries: Binary Bayes Detection
40(6)
1.6.1 Formulation I: Binary On-Off Signal Detection
42(1)
1.6.2 The Average Risk
43(1)
1.6.3 Cost Assignments
43(2)
1.6.4 Error Probabilities
45(1)
1.7 Optimum Detection: On-Off Optimum Processing Algorithms
46(4)
1.7.1 The Logarithmic GLRT
48(1)
1.7.2 Remarks on the Bayes Optimality of the GLR
48(2)
1.8 Special On-Off Optimum Binary Systems
50(7)
1.8.1 Neyman-Pearson Detection Theory
50(1)
1.8.2 The Ideal Observer Detection System
51(1)
1.8.3 Minimax Detectors
52(1)
1.8.4 Maximum Aposteriori (MAP) Detectors from a Bayesian Viewpoint
53(4)
1.8.5 Bayesian Sequential Detectors
57(1)
1.9 Optimum Detection: On-Off Performance Measures and System Comparisons
57(12)
1.9.1 Error Probabilities: Optimum Systems
58(7)
1.9.2 Error Probabilities: Suboptimum Systems
65(1)
1.9.3 Decision Curves and System Comparisons
66(3)
1.10 Binary Two-Signal Detection: Disjoint and Overlapping Hypothesis Classes
69(4)
1.10.1 Disjoint Signal Classes
69(1)
1.10.2 Overlapping Hypothesis Classes
70(3)
1.11 Concluding Remarks
73(4)
References
74(3)
2 Space-Time Covariances and Wave Number Frequency Spectra: I. Noise and Signals with Continuous and Discrete Sampling
77(64)
2.1 Inhomogeneous and Nonstationary Signal and Noise Fields I: Waveforms, Beam Theory, Covariances, and Intensity Spectra
78(13)
2.1.1 Signal Normalization
79(1)
2.1.2 Inhomogeneous Nonstationary (Non-WS-HS) Noise Covariances
80(3)
2.1.3 Narrowband Fields
83(5)
2.1.4 Noise and Signal Field Covariances: Narrowband Cases
88(3)
2.2 Continuous Space-Time Wiener-Khintchine Relations
91(11)
2.2.1 Directly Sampled Approximation of the W-Kh Relations (Hom-Stat Examples)
93(2)
2.2.2 Extended Wiener-Khintchine Theorems: Continuous Inhomogeneous and Nonstationary Random (Scalar) Fields
95(5)
2.2.3 The Important Special Case of Homogeneous---Stationary Fields---Finite and Infinite Samples
100(2)
2.3 The W-Kh Relations for Discrete Samples in the Non-Hom-Stat Situation
102(6)
2.3.1 The Amplitude Spectrum for Discrete Samples
102(5)
2.3.2 Periodic Sampling
107(1)
2.4 The Wiener-Khintchine Relations for Discretely Sampled Random Fields
108(7)
2.4.1 Discrete Hom-Stat Wiener-Khintchine Theorem: Periodic Sampling and Finite and Infinite Samples
110(2)
2.4.2 Comments
112(3)
2.5 Aperture and Arrays---I: An Introduction
115(23)
2.5.1 Transmission: Apertures and Their Fourier Equivalents
116(4)
2.5.2 Transmission: The Propagating Field and Its Source Function
120(6)
2.5.3 Point Arrays: Discrete Spatial Sampling
126(3)
2.5.4 Reception
129(5)
2.5.5 Narrowband Signals and Fields
134(3)
2.5.6 Some General Observations
137(1)
2.6 Concluding Remarks
138(3)
References
139(2)
3 Optimum Detection, Space-Time Matched Filters, and Beam Forming in Gaussian Noise Fields
141(98)
3.1 Optimum Detection I: Selected Gaussian Prototypes---Coherent Reception
142(12)
3.1.1 Optimum Coherent Detection. Completely Known Deterministic Signals in Gauss Noise
142(4)
3.1.2 Performance
146(4)
3.1.3 Array Processing II: Beam Forming with Linear Arrays
150(4)
3.2 Optimum Detection II: Selected Gaussian Prototypes---Incoherent Reception
154(22)
3.2.1 Incoherent Detection: I. Narrowband Deterministic Signals
154(15)
3.2.2 Incoherent Detection II. Deterministic Narrowband Signals with Slow Rayleigh Fading
169(3)
3.2.3 Incoherent Detection III: Narrowband Equivalent Envelope Inputs---Representations
172(4)
3.3 Optimal Detection III: Slowly Fluctuating Noise Backgrounds
176(12)
3.3.1 Coherent Detection
176(4)
3.3.2 Narrowband Incoherent Detection Algorithms
180(3)
3.3.3 Incoherent Detection of Broadband Signals in Normal Noise
183(5)
3.4 Bayes Matched Filters and Their Associated Bilinear and Quadratic Forms, I
188(31)
3.4.1 Coherent Reception: Causal Matched Filters (Type 1)
190(2)
3.4.2 Incoherent Reception: Causal Matched Filters (Type 1)
192(3)
3.4.3 Incoherent Reception-Realizable Matched Filters; Type 2
195(3)
3.4.4 Wiener-Kolmogoroff Filters
198(2)
3.4.5 Extensions: Clutter, Reverberation, and Ambient Noise
200(2)
3.4.6 Matched Filters and Their Separation in Space and Time I
202(5)
3.4.7 Solutions of the Discrete Integral Equations
207(7)
3.4.8 Summary Remarks
214(1)
3.4.9 Signal-to-Noise Ratios, Processing Gains, and Minimum Detectable Signals. I
214(5)
3.5 Bayes Matched Filters in the Wave Number-Frequency Domain
219(16)
3.5.1 Fourier Transforms of Discrete Series
219(11)
3.5.2 Independent Beam Forming and Temporal Processing
230(5)
3.6 Concluding Remarks
235(4)
References
235(4)
4 Multiple Alternative Detection
239(32)
4.1 Multiple-Alternative Detection: The Disjoint Cases
239(15)
4.1.1 Detection
240(2)
4.1.2 Minimization of the Average Risk
242(2)
4.1.3 Geometric Interpretation
244(1)
4.1.4 Examples
245(5)
4.1.5 Error Probabilities, Average Risk, and System Evaluation
250(3)
4.1.6 An Example
253(1)
4.2 Overlapping Hypothesis Classes
254(8)
4.2.1 Reformulation
255(2)
4.2.2 Minimization of the Average Risk for Overlapping Hypothesis Classes
257(2)
4.2.3 Simple (K + 1) - ary Detection
259(1)
4.2.4 Error Probabilities, Average and Bayes Risk, and System Evaluations
260(2)
4.3 Detection with Decisions Rejection: Nonoverlapping Signal Classes
262(9)
4.3.1 Optimum (K + 1) - ary Decisions with Rejection
264(1)
4.3.2 Optimum (K + 1) - ary Decision with Rejection
265(1)
4.3.3 A Simple Cost Assignment
266(1)
4.3.4 Remarks
267(3)
References
270(1)
5 Bayes Extraction Systems: Signal Estimation and Analysis, p(H1) = 1
271(36)
5.1 Decision Theory Formulation
272(15)
5.1.1 Nonrandomized Decision Rules and Average Risk
272(2)
5.1.2 Bayes Extraction With a Simple Cost Function
274(4)
5.1.3 Bayes Extraction With a Quadratic Cost Function
278(3)
5.1.4 Further Properties
281(2)
5.1.5 Other Cost Functions
283(4)
5.2 Coherent Estimation of Amplitude (Deterministic Signals and Normal Noise, p(H1) = 1)
287(7)
5.2.1 Coherent Estimation of Signal Amplitude Quadratic Cost Function
287(3)
5.2.2 Coherent Estimation of Signal Amplitude (Simple Cost Functions)
290(1)
5.2.3 Estimations by (Real) θ Filters
291(2)
5.2.4 Biased and Unbiased Estimates
293(1)
5.3 Incoherent Estimation of Signal Amplitude (Deterministic Signals and Normal Noise, p(H1) = 1)
294(6)
5.3.1 Quadratic Cost Function
294(4)
5.3.2 "Simple" Cost Functions SCF1 (Incoherent Estimation)
298(2)
5.4 Waveform Estimation (Random Fields)
300(4)
5.4.1 Normal Noise Signals in Normal Noise Fields (Quadratic Cost Function)
300(1)
5.4.2 Normal Noise Signals in Normal Noise Fields ("Simple" Cost Functions)
301(3)
5.5 Summary Remarks
304(3)
References
305(2)
6 Joint Detection and Estimation, p(H1) ≤ 1: I. Foundations
307(74)
6.1 Joint Detection and Estimation under Prior Uncertainty [ p(H1) ≤ 1]: Formulation
309(6)
6.1.1 Case 1: No Coupling
312(2)
6.1.2 Case 2: Coupling
314(1)
6.2 Optimal Estimation [ p(H1) ≤ 1]: No Coupling
315(11)
6.2.1 Quadratic Cost Function: MMSE and Bayes Risk
316(3)
6.2.2 Simple Cost Functions: UMLE and Bayes Risk
319(7)
6.3 Simultaneous Joint Detection and Estimation: General Theory
326(24)
6.3.1 The General Case: Strong Coupling
326(5)
6.3.2 Special Cases I: Bayes Detection and Estimation With Weak Coupling
331(2)
6.3.3 Special Cases II: Further Discussion of γp<1|QCF for Weak or No Coupling
333(3)
6.3.4 Estimator Bias (p ≤ 1)
336(2)
6.3.5 Remarks on Interval Estimation, p(H1) ≤ 1
338(1)
6.3.6 Detection Probabilities
339(2)
6.3.7 Waveform Estimation (p ≤ 1): Coupled and Uncoupled D and E
341(1)
6.3.8 Extensions and Modifications
342(3)
6.3.9 Summary Remarks
345(5)
6.4 Joint D and E: Examples-Estimation of Signal Amplitudes [ p(H1) ≤ 1]
350(28)
6.4.1 Amplitude Estimation, p(H1) = 1
352(3)
6.4.2 Bayes Estimators and Bayes Error, p(H1) ≤ 1
355(3)
6.4.3 Performance Degradation, p < 1
358(9)
6.4.4 Acceptance or Rejection of the Estimator: Detection Probabilities
367(4)
6.4.5 Remarks on the Estimation of Signal Intensity l0 ≡ a20
371(7)
6.5 Summary Remarks, p(H)1 ≤ 1: I---Foundations
378(3)
References
379(2)
7 Joint Detection and Estimation under Uncertainty, pk (H1) <
1. II. Multiple Hypotheses and Sequential Observations
381(54)
7.1 Jointly Optimum Detection and Estimation under Multiple Hypotheses, p(H1) ≤ 1
382(18)
7.1.1 Formulation
383(6)
7.1.2 Specific Cost Functions
389(7)
7.1.3 Special Cases: Binary Detection and Estimation
396(4)
7.2 Uncoupled Optimum Detection and Estimation, Multiple Hypotheses, and Overlapping Parameter Spaces
400(7)
7.2.1 A Generalized Cost Function for K-Signals with Overlapping Parameter Values
402(1)
7.2.2 QCF: Overlapping Hypothesis Classes
403(3)
7.2.3 Simple Cost Functions (SCF1,2): Joint D + E with Overlapping Hypotheses Classes
406(1)
7.3 Simultaneous Detection and Estimation: Sequences of Observations and Decisions
407(21)
7.3.1 Sequential Observations and Unsupervised Learning: I. Binary Systems with Joint Uncoupled D + E
407(7)
7.3.2 Sequential Observations and Unsupervised Learning: II. Joint D + E for Binary Systems with Strong and Weak Coupling
414(3)
7.3.3 Sequential Observations and Unsupervised Learning: III. Joint D + E Under Multiple Hypotheses with Strong and Weak Coupling and Overlapping Hypotheses Classes
417(6)
7.3.4 Sequential Observations and Overlapping Multiple Hypothesis Classes: Joint D + E with No Coupling
423(2)
7.3.5 Supervised Learning (Self-Taught Mode): An Introduction
425(3)
7.4 Concluding Remarks
428(7)
References
432(3)
8 The Canonical Channel I: Scalar Field Propagation in a Deterministic Medium
435(104)
8.1 The Generic Deterministic Channel: Homogeneous Unbounded Media
437(28)
8.1.1 Components of the Generic Channel: Coupling
438(1)
8.1.2 Propagation in An Ideal Medium
439(2)
8.1.3 Green's Function for the Ideal Medium
441(7)
8.1.4 Causality, Regularity, and Reciprocity of the Green's Function
448(1)
8.1.5 Selected Green's Functions
449(4)
8.1.6 A Generalized Huygens Principle: Solution for the Homogeneous Field αH
453(10)
8.1.7 The Explicit Role of the Aperture or Array
463(2)
8.2 The Engineering Approach: I---The Medium and Channel as Time-Varying Linear Filters (Deterministic Media)
465(8)
8.2.1 Equivalent Temporal Filters
466(5)
8.2.2 Causality: Extensions of the Paley-Wiener Criterion
471(2)
8.3 Inhomogeneous Media and Channels---Deterministic Scatter and Operational Solutions
473(21)
8.3.1 Deterministic Volume and Surface Scatter: The Green's Function and Associated Field α(Q)
475(3)
8.3.2 The Associated Field and Equivalent Solutions for Volumes and Surfaces
478(2)
8.3.3 Inhomogeneous Reciprocity
480(4)
8.3.4 The GHP for Inhomogeneous Deterministic Media including Backscatter
484(9)
8.3.5 Generalizations and Remarks
493(1)
8.4 The Deterministic Scattered Field in Wave Number-Frequency Space: Innovations
494(5)
8.4.1 Transform Operator Solutions
496(2)
8.4.2 Commutation and Convolution
498(1)
8.5 Extensions and Innovations, Multimedia Interactions
499(10)
8.5.1 The η-Form: Multimedia Interactions
500(3)
8.5.2 The Feedback Operational Representation and Solution
503(4)
8.5.3 An Estimation Procedure for the Deterministic Mass Operators Q and η
507(1)
8.5.4 The Engineering Approach II: Inhomogeneous Deterministic Media
508(1)
8.6 Energy Considerations
509(26)
8.6.1 Outline of the Variation Method
510(2)
8.6.2 Preliminary Remarks
512(2)
8.6.3 Energy Density and Density Flux: Direct Models---A Brief Introduction
514(2)
8.6.4 Equal Nonviscous Elastic Media
516(6)
8.6.5 Energy Densities and Flux Densities in the Dissipative Media
522(5)
8.6.6 Extensions: Arrays and Finite Duration Sources and Summary Remarks
527(8)
8.7 Summary: Results and Conclusions
535(4)
References
536(3)
9 The Canonical Channel II: Scattering in Random Media; "Classical" Operator Solutions
539(60)
9.1 Random Media: Operational Solutions---First- and Second-Order Moments
541(24)
9.1.1 Operator Forms: Moment Solutions and Dyson's Equation
543(8)
9.1.2 Dyson's Equation in Statistically Homogeneous and Stationary Media
551(9)
9.1.3 Example: The Statistical Structure of the Mass Operator Q(d)1, with (Q) = 0
560(4)
9.1.4 Remarks
564(1)
9.2 Higher Order Moments Operational Solutions for The Langevin Equation
565(15)
9.2.1 The Second-Order Moments: Analysis of the Bethe-Salpeter Equation (BSE)
565(3)
9.2.2 The Structure of Q(d)12
568(2)
9.2.3 Higher-Order Moment Solutions (m ≥ 3) and Related Topics
570(2)
9.2.4 Transport Equations
572(2)
9.2.5 The Gaussian Case
574(1)
9.2.6 Very Strong Scatter: Saturation ||η|| ~ 1
575(4)
9.2.7 Remarks
579(1)
9.3 Equivalent Representations: Elementary Feynman Diagrams
580(18)
9.3.1 Diagram Vocabulary
581(5)
9.3.2 Diagram Approximations
586(8)
9.3.3 A Characterization of the Random Channel: First- and Second-Order-Moments I
594(2)
9.3.4 Elementary Statistics of the Received Field
596(2)
9.4 Summary Remarks
598(1)
References 599(2)
Appendix A1 601(16)
Index 617
David Middleton, PhD, graduated from Harvard University where he began his career at the institution's Radio Research Laboratoryworking on radar countermeasures as well as passive and active jamming during World War IIbefore teaching there. A recipient of numerous prizes and awards related to his work on communication theory, Dr. Middleton was a fellow of the IEEE, the American Physical Society, the Acoustical Society of America, and the American Association for the Advancement of Science.