Contents
Contributors
Preface
1. Artificial
Intelligence Based Approach for Robust Radar Detection in
Maritime Scenarios
1.1. Introduction
1.2. Case Study
1.2.1. Target Model
1.2.2. Interference Model
1.3. Reference CGLR Detectors
1.4. Intelligent Detectors
1.4.1. Intelligent Agent Architectures
1.4.2. Error Function of the Training Process
1.5. Experiments and Results
1.5.1. Detectors Design
1.5.2. Real Scenario
1.6. Conclusions
Acknowledgements
References
2. Robust Hierarchical Bayesian Modeling for
Automatic Model Order Determination
2.1. Introduction
2.1.1. Non-Negative Data Representation
2.1.2. Sparseness in Data Representation
2.1.3. Non-Negative Matrix Factorization and Model Order
Determination
2.1.3.1. Non-Negative Matrix Factorization
2.1.3.2. Model Order Determination
2.2. Hierarchical Bayesian Modeling
2.3. Data Model
2.3.1. Gaussian Distribution
2.3.2. Poisson Distribution
2.3.3. Gamma Distribution
2.4. Parameter Model 47
2.4.1. Half Normal Distribution
2.4.2. Exponential Distribution
2.4.3. Tweedie Distribution
2.5. Hyper-parameter Model
2.5.1. Informative Prior
2.5.1.1. Gamma Distribution
2.5.1.2. Tweedie Distribution
2.5.1.3. Variational Bayesian Distribution
2.5.2. Non-Informative Prior
2.5.2.1. Stirling’s Algorithm
2.5.2.2. Expectation Maximization (EM) Algorithm
2.6. Result from Different Methods
2.6.1. Fence Dataset
2.6.2. Swimmer Dataset
2.6.3. Face Dataset
2.6.4. Music Dataset
2.7. Conclusion
References
3. Complex Wavelet Additive Synthesis of Audio
Signals: Introduction and Applications
3.1. Introduction
3.2. Audio, Time-Frequency Distributions and Wavelets
3.2.1. Basic Definitions
3.2.2. The Complex Continuous Wavelet Transform
3.2.2.1. General Considerations
3.2.2.2. Extracting Information from the Complex Continuous
Wavelet Transform
3.3. Gaussian Complex Filtering
3.3.1. The Morlet Wavelet Revised
3.3.2. Mathematical Considerations
3.3.2.1. The Simplest Case: A Pure Cosine with Constant
Amplitude
3.3.2.2. Pure Cosine with Variable Amplitude
3.3.2.3. The Sum of n Pure Cosines
3.3.2.4. The Quadratic Phase FM Signal
3.3.3. A General Monocomponent Signal: The Linear Chirp
Approximation
3.3.4. The Additive Synthesis Method
3.4. CWAS: Discretization
3.4.1. Constant Resolution Filter Bank
3.4.2. Variable Resolution Filter Bank
3.5. The Complex Wavelet Additive Synthesis algorithm
3.5.1. Calculation of the Wavelet Coefficients
3.5.2. Wavelet Spectrogram and Scalogram
3.5.3. Partials and Additive Synthesis Reconstruction
3.5.4. Partials and Tracking of Partials
3.6. Resynthesis and Extraction of Time-Frequency
Characteristics of the Audio Signal
3.6.1. Audio Model: Time-Frequency Characteristics
3.6.2. Plotting the Characteristics of the Signal
3.6.2.1. 2D Representation
3.6.2.2. 3D Representation
3.7. CWAS Applications
3.7.1. F0 Estimation
3.7.1.1. General Considerations
3.7.1.2. NHRF0
3.7.1.3. Polyphony Inference: Accumulated Energy
3.7.1.4. HRF0
3.7.1.5. Discrete Cepstrum
3.7.1.6. Estimation Results
3.7.2. Blind Audio Source Separation
3.7.2.1. General Considerations
3.7.2.2. BASS using CWAS
3.8. Conclusions and Future Work
References
4. A Review of the Recent Cepstrum-based Methods for
Acoustic Feedback Cancellation
4.1. Introduction
4.2. Cesptral Analysis of the SR System
4.3. Cesptral Analysis of the AFC System
4.3.1. Cesptral Analysis of the Microphone Signal
4.3.2. Cesptral Analysis of the Error Signal
4.4. AFC Methods Based on Cepstral Analysis
4.4.1. AFC-CM
4.4.2. AFC-CE
4.5. Simulation Configuration
4.5.1. Simulated Environment
4.5.1.1. Feedback Path
4.5.1.2. Forward Path
4.5.1.3. Delay Filter
4.5.2. Evaluation Metrics
4.5.2.1. Maximum Stable Gain
4.5.2.2. Misalignment
4.5.2.3. Frequency-weighted Log-spectral Signal Distortion
4.5.2.4. W-PESQ
4.5.3. Speech Database
4.6. Simulation Results
4.6.1. Results for Long Acoustic Feedback Path
4.6.2. Results for Short Acoustic Feedback Path
4.7. Conclusions
References
5. Covariance Analysis of Periodically Correlated
Random Processes for Unknown Non-stationarity Period
5.1. Introduction
5.2. Period Estimation as a Problem of Searching for Hidden
Periodicities
5.3. PCRP-model of Hidden Periodicities
5.4. Coherent Covariance Analysis
5.4.1. The Estimation of the Mean Function Period
5.4.2. Estimation of Mean Function and Its Fourier
Coefficients
5.4.3. Estimation of Covariance and Correlation Function
Period
5.4.4. Estimation of Covariance Function and Its Fourier
Coefficients
5.4.5. The Verification of the Developed Approach
5.4.5.1. The Analysis of Simulated Data
5.4.5.2. The Covariance Analysis of Vibration of the Fast
Moving Spindle
5.5. Component Covariance Analysis
5.5.1. Period Estimation of Deterministic Component
5.5.1.1. Cosine and Sine Transforms of Realization
5.5.1.2. Component Period Estimation
5.5.2. Covariance Functionals
5.5.2.1. Cosine and Sine Covariance Functionals
5.5.2.2. Component Covariance Functional
5.5.3. Correlation Functionals
5.5.3.1. Cosine and Sine Correlation Functionals
5.5.3.2. Component Correlation Functional
5.5.4. Moment Functions Estimators
5.5.4.1. Mean Function Estimator
5.5.4.2. The Covariance Function Estimator
5.5.5. Period Estimation for Simulated and Real Data
5.5.5.1. Simulated Data Analysis
5.5.5.2. Application to Vibration Signal Analysis
5.6. Comparison with Other Results
5.7. Conclusions
References
6. Learning Colours from Textures by Effective
Representation of Images
6.1. Introduction
6.2. Early Learning-based Colourisation Methods
6.3. Learning Colours from Texture on Non-linear Manifolds
6.3.1. Colourisation Problem
6.3.2. Manifold of Patch Populations
6.3.3. Sparsity Constraints
6.3.4. Semantic Groups
6.3.5. Prediction Algorithm
6.3.6. Implementation Details
6.3.7. Experiment Results
6.4. Transferring Colours by Deep Neural Networks
6.4.1. Fully Convolutional and Diluting Neural Networks
6.4.2. Colouration by FCN
6.5. End Notes
References
7. The Physical Layer of Low Power Wide Area Networks:
Strategies, Information Theory’s Limit and Existing Solutions
7.1. Introduction
7.2. Constraints of the Physical Layer
7.2.1. Power Amplifier and Constant Envelope Modulations
7.2.2. Achieving Long Range
7.2.3. Sensitivity of the Receiver
7.3. Strategies
7.3.1. Reduce the Bandwidth
7.3.2. Reduce the Required Signal-to-Noise Ratio
7.3.2.1. Reduce the Spectral Efficiency
7.3.2.2. Reduce the Spectral Efficiency Along with the
Required E_b/N_0
7.3.2.3. Reduce the Level of Required E_b/N_0
7.3.3. Comparison to the Information Theory’s Limit
7.4. Existing Solutions
7.4.1. Proprietary Solutions
7.4.1.1. Sigfox
7.4.1.2. LoRa
7.4.2. Standardized Technologies
7.4.2.1. 802.15.4k
7.4.2.2. Narrow-Band IoT
7.4.3. Turbo-FSK
7.4.3.1. Transmitter
7.4.3.2. Receiver
7.4.4. Performance
7.5. Conclusion
References
8. Deep Belief Networks and Its Applications in
Evaluation of Vehicle Interior Noise
8.1. Introduction
8.2. Related Works
8.2.1. Psychoacoustics-based Interior Sound Quality
Prediction
8.2.2. Machine-learning-based Interior Sound Quality
Prediction
8.3. Deep Belief Networks
8.3.1. Restricted Boltzmann Machine Architecture
8.3.2. Deep Belief Network Architecture
8.3.3. Development of Deep Belief Networks to Model
Continuous Data
8.4. Recording and Evaluating the Vehicle Interior Noise
8.4.1. Development of the Vehicle Interior Noise Database
8.4.2. Subjective Evaluation of the Vehicle Interior Noise
8.5. Interior Noise Feature Extraction
8.5.1. Psychoacoustic-based Feature Extraction
8.5.2. Energy-based Feature Extraction
8.5.3. Feature Comparison and Selection
8.6. Development of the CRBM-DBN-based Sound Quality
Prediction Model
8.6.1. Performance Measurements
8.6.2. Designing the Architecture of CRBM-DBN
8.6.3. Model Verification and Comparison
8.7. Conclusion
Acknowledgments
References
9. Selection of Diagnostic Symptoms for
Technical Condition Assessment and Prognosis
9.1. Introduction
9.2. Singular Value Decomposition (SVD) Method
9.2.1. Theoretical Outline
9.2.2. Examples
9.3. Information Content Measure (ICM) Method
9.3.1. Background
9.3.2. The Question of Stationarity
9.3.3. The Question of Abrupt Changes
9.3.4. Representativeness Factor
9.4. Example
9.5. Conclusions
Acknowledgements
References
10. Nonlinear Acoustic Echo Cancellation Using
a Dynamic Micro-speaker Model
10.1. Introduction
10.2. Loudspeaker Nonlinearities
10.3. Quasi-linear Model of the Micro-speaker
10.4. Model Identification of a System Characterised by
Nonlinear Damping
10.5. Experimental Identification of a Micro-speaker
10.6. Conclusions
Acknowledgements
References
11. A New Damage Detection Method for Bridge Condition
Assessment in Structural Health Monitoring (SHM)
11.1. Introduction
11.1.1. An Idea of the System State Representation
Methodology (SRM)
11.1.2. Advances of SRM Compare with the Traditional Methods
and Research Scopes
11.1.3. Data Management
11.1.4. Nonlinear Simulation and Feature Extracting
11.1.5. Condition Assessment
11.2. SRM Concept and Algorithms
11.2.1. Definition of the State of a System
11.2.2. System State Defining
11.2.3. System State Approximating
11.2.4. Optimal State Representation Based on a Kernel
Function
11.3. SRM Assessment Method
11.3.1. Probability-based Assessment Methods
11.3.1.1. Statistic Test Methods
11.3.1.2. System State Indices
11.3.1.3. Hypothesis Probability with Two Divided-levels
11.3.1.4. Approximate Probability with Two Divided-levels
11.3.1.5. Approximate Probability with Multi-Levels
11.3.1.6. Assessment Based on State Probability
11.3.1.7. Assessment Based on Information Entropy
11.3.2. A Simple Simulation Experimental Result
11.3.3. Summary
11.4. SRM Tool 1: Compatible Gradient Algorithm for Large Scale
Linear Problem
11.4.1. A Simple Gradient Algorithm for Linear Problem
11.4.2. Compatible Gradient Algorithm
11.4.3. A Simple Example
11.4.4. Summary
11.4.5. Monotonic Descent Conjugates Gradient (MDCG) Method
11.5. SRM Tool 2: Frequency Slice Wavelet Transform Theory
and Methods
11.5.1. Introduction
11.5.2. Frequency Slice Wavelet Transform Analysis
11.5.2.1. Frequency Slice Wavelet Transform
11.5.2.2. Bandwidth-to-Frequency Ratio Property
11.5.2.3. Scale Selection
11.5.2.4. Reconstructive Procedure Independency Property
11.5.2.5. FSF Design Property
11.5.2.6. Filter Property
11.5.2.7. Discrete FSF
11.5.2.8. Discrete FSWT
11.5.3. FSWT Signal Process Flow
11.5.4. Comparing with Classical Transforms
11.5.5. Applications
11.5.5.1. Signal Filter in Time and Frequency Domain
11.5.5.2. Separation of Close Modals
11.5.5.3. Resolution and Symmetry
11.5.5.4. Energy Leakage Problem
11.5.5.5. Scale Optimization
11.5.5.6. Dynamic Scale
11.5.5.7. A Real Life Signal
11.5.6. Summary
11.6. SRM Feature Extracting Methods
and Applications
11.6.1. Basic Idea of System Feature
11.6.2. Approximate Feature Vectors Problem
11.6.3. Extracting Damping Feature
11.6.4. Feature Extract Algorithm
11.6.5. Overall Logics of SRM Method
11.6.6. Application to a Laboratory Bridge Monitoring System
and Discussions
11.6.6.1. Outline of SRM-based Damage Detection
11.6.6.2. Overview of Bridge Model Girder Experiment
11.6.6.3. Experimental Results and Discussions
11.7. Concluding Remarks
References
12. On Line Diagnosis in Induction Motors and Load
12.1. Introduction
12.2. Internal Fault Detection in Induction Motors
12.2.1. Motor Current Signature Analysis
12.2.2. Extended Park’s Vector Approach
12.2.3. Towards an Integrated Fault Diagnostic System
12.2.4. Experimental Results
12.2.4.1. The Experimental Prototype
12.2.4.2. Case Study 1
12.2.4.3. Case Study 2
12.2.4.4. Case Study 3
12.2.4.5. General Comments
12.3. Mechanical Faults Detection beyond the Motor
12.3.1. General Comments
12.3.2. Misaligned Drives Model
12.3.3. Fault Detection Method
12.3.4. Experimental Results
12.4. Gear Box Fault Detection
12.5. Conclusions
Acknowledgements
References
13. Vocal Folds Dynamics by Means of Optical Flow
Techniques: A Review of the Methods
13.1. Introduction
13.1.1. Voice Assessment and Endoscopy
13.1.2. Scientific and Clinical Applications of Laryngeal
High-Speed Videoendoscopy
13.1.3. From a Big Amount of Data to a Synthesized View
13.2. Optical-Flow Computation
13.2.1. Theoretical Background
13.2.2. Optical Flow Evaluation
13.2.3. Optical Flow in LHSV
13.3. Application of OF Computation to LHSV Image Analysis
13.3.1. Database
13.3.2. Image Processing Implementation
13.3.3. Optical Flow Facilitative Playbacks
13.3.3.1. Optical Flow Kymogram (OFKG)
13.3.3.2. Optical Flow Glottovibrogram (OFGVG)
13.3.3.3. Glottal Optical Flow Waveform (GOFW)
13.3.3.4. Vocal Folds Displacements Trajectories
13.3.4. Reliability Assessment of Optical Flow Playbacks
13.4. Assessment of OF-Based Playbacks
13.4.1. Displacements Trajectories Comparison
13.4.2. Comparison between OFGVG, GVG and |dxGVG|
13.4.3. Glottal Velocity: Derivative of Glottal Area Waveform
and Glottal Optical FlowWaveform
13.4.4. Optical Flow Using Local Dynamics along One Line:
Digital Kymogram and Optical Flow Kymogram
13.5. Conclusions
Acknowledgements
References
Index |