فهرست مطالب
Preface
Organization
Contents
Hardware for High-Performance Computing and Signal Processing
Design of a Pipeline Computing Module as Part of a Specialized VLSI
1 Introduction
2 Flow Control in a Pipeline Computing Structure
2.1 Development of Components of a Pipeline Computer
2.2 Integration of a Pipeline Computer into a Computer System (VLSI)
3 Selecting the Pipeline Depth Based on Topological Basis
4 Characteristics of a Pipeline Accelerator in FPGA CAD
5 Conclusions
References
Speech Enhancement Based on Two-Stage Neural Network with Structured State Space for Sequence Transformation
1 Introduction
2 Neural Network Model Used for Speech Enhancement
2.1 Speech Enhancement Applying Complex Ideal Ratio Masks
2.2 Proposed Model
2.3 Structured State Space for a Sequence Layer
2.4 Full-Band and Sub-band Transform Blocks
2.5 Loss Functions
3 Experiments and Results
3.1 Datasets
3.2 Implementation Details
3.3 Performance Evaluation Metrics
3.4 Loss Function Choice
3.5 Optimal Number of S4 Layers
3.6 Comparison of the Proposed Model Having Alternatives
3.7 Discussion
4 Conclusion
References
Designing a Graphics Accelerator with Heterogeneous Architecture
1 Introduction
2 Architecture of a Specialized Graphics Accelerator
3 Architectural Solutions for Graphics Accelerator
3.1 Programmable Task Distribution
3.2 Software Memory Management
3.3 Distribution of Tasks Depending on Data
3.4 Redistribution of Resources within the Computing Cluster
3.5 Thread Management According to the SIMD Approach
3.6 Combining Raster and Cache Memory for General-Purpose Computing
4 Automation of Design of Specialized Nodes
5 Conclusions
References
Spectrophotometer for Field Studies
1 Introduction
2 Proposed Approach
2.1 AS7341 Integral Spectrometer
2.2 Wireless Microcontroller CC2640
2.3 Structure of the Spectrophotometer
2.4 Software Structure
3 Experiments and Results
3.1 Dataset Description
3.2 Discussion
4 Conclusion
References
Comparative Study of Practical Implementation of Time Delay Estimation Methods on Single Board Computer
1 Introduction
2 Overview of TDE Methods
2.1 Time-Domain Methods
2.2 Frequency-Domain Methods
2.3 Signal Processing in Practical TDE
2.4 Variants of Fourier Transform Implementation
3 Computational Study
3.1 Raspberry Pi 4B Hardware
3.2 Testing Software
3.3 Estimation of Computation Time
3.4 Estimation of Memory Usage
4 Discussion
5 Conclusion
References
Information Technologies and Computer Simulation of Physical Phenomena
Study of the Functional Characteristics of TiNi Coatings by the Computer-Aided Simulation Using Parallel Computing
1 Introduction
2 Brief Description of the Program Complex for Coating Simulation
3 Implementation of Program System for Parallel Running of SIMD-Tasks for Coating Simulation
3.1 Functionality of the Program System
3.2 Computational Tests for Determination of Optimal Loading Mode of Computing Nodes
4 Using a Program System with an Architecture for Parallel Running of Similar SIMD-Tasks for Coating Simulation
5 Conclusions
References
On Hierarchical Convergence of the Heterogeneous Multiscale Finite Element Method Using Polyhedral Supports
1 Introduction
2 Problem Statement and Solution Method
2.1 Discretization of Computational Domain
2.2 Heterogeneous Multiscale Finite Element Method
3 Numerical Estimation of FE-HMM Accuracy
3.1 Homogeneous Computational Domain
3.2 Computational Domain with Inclusions
4 Parallelization of Computing Schemes
5 Conclusions
References
Parallelization of Finite-Volume Numerical Methods of Computational Fluid Dynamics by Means of Shared Memory Computing Systems
1 Introduction
2 Problem Statement
3 Parallelization of Computing Flows of Conserved Values
3.1 Reduction of the Parallelization Problem
3.2 Application of Graph Coloring in Parallelization
3.3 Optimized Greedy Coloring Method
3.4 Verification of the Algorithm
4 Discussion
References
Computing Technologies in Data Analysis and Decision Making
Development and Comparative Study of Data Compression Methods Used in Technical Monitoring Systems
1 Introduction
2 Problem Description
3 Methods Based on Data Normalization and Structuring
4 Compression Methods Based on Difference Schemes and RLE-Algorithms
5 Lossy Compression Methods
6 Conclusions
References
Object Recognition Based on Three-Dimensional Computer Graphics
1 Introduction
2 Problem Statement
3 Related Work
4 Proposed Solution
5 Proof of Work
6 Conclusions
References
Adaptive Methods for the Structural Optimization of Neural Networks and Their Ensemble for Data Analysis
1 Introduction
2 Materials and Methods
2.1 Designing the Structure of Neural Networks Using the Probabilistic Approach
2.2 Providing Diversity for Individual Neural Network Solvers
2.3 Ensemble Design Using Modified Genetic Programming
3 Numerical Experiments
3.1 Simulated Datasets
3.2 Test Problems
3.3 Pool of Methods
3.4 Raw Results Processing
3.5 Results and Discussion
4 Conclusion
References
Self-adaptation Method for Evolutionary Algorithms Based on the Selection Operator
1 Introduction
2 Methods of Self-adaptation for Genetic Algorithm
2.1 Genetic Algorithm
2.2 Fitness-Based Adaptation
2.3 Population-Level Dynamic Probabilities
2.4 Self-configuring Genetic Algorithm
2.5 SHAGA
3 Proposed Approach
4 Experimental Setup and Results
4.1 Parameters of the Genetic Algorithm
4.2 Results of Solving Binary Optimization Problems
4.3 Results of Solving Real-Valued Optimization Problems
5 Conclusion
References
Application of U-Net Architecture Neural Network for Segmentation of Brain Cell Images Stained with Trypan Blue
1 Introduction
2 Materials and Methods
3 Results
4 Discussion
5 Conclusions
References
Language Model Architecture Based on the Syntactic Graph of Analyzed Text
1 Introduction
2 Materials and Methods
2.1 Graph Core and Function Levels
2.2 Graphlet-Based Methods
2.3 Metric of Similarity of Neighbors
2.4 Model-Parametric Representation
3 Results and Discussion
4 Conclusions
References
Information and Computing Technologies in Automation and Control Science
Classic and Modern Methods of Automatic Parking Control of Self-driving Cars
1 Introduction
2 Peculiarities of the Parking Control Problem
2.1 Equations of SDC Parking Dynamics
2.2 Mechanical Limitations of SDC Parking
2.3 Space Limitations of SDC Parking
2.4 Boundary Conditions of SDC Parking
2.5 Setting the SDC Parking Control Problem
3 Parking Based on Dubins and Reeds-Shepp Motion Models
4 Parking Based on Reinforcement Learning
5 Conclusion
References
Application of Smoothing Technique to Model Predictive Traffic Signal Control
1 Introduction
2 Transport Network Model
2.1 Traffic Controllers
2.2 Macroscopic Traffic Flow Model
2.3 Numerical Scheme
2.4 Intersection Model
3 Model Predictive Control
3.1 Target Function
3.2 Zero-Order Smoothing Optimization Technique
4 Numerical Experiments
5 Conclusion
References
Design Features of the Frequency-Controlled Electric Drive for Positioning Mechanisms
1 Introduction
2 Problems of Positioning Mechanism Automation
2.1 Control Panel Functions
2.2 Design Constraints
2.3 Principle of Operation
3 Method of Adjusting the Speed of a Positioning Mechanism
3.1 Control Logic
3.2 Results
3.3 Discussion
4 Conclusion
References
Development of a Mathematical Model to Study the Energy Indicators of Electric Drives Using the DFC-IM System
1 Introduction
2 Principles ForPositioning Accuracy
2.1 Control Panel Functions
2.2 Generation of Mathematical Model Equations
3 Block Diagram of Calculation Software in the DFC-IM System
3.1 Control Logic
3.2 Results
3.3 Discussion
4 Conclusion
References
Intelligent Data Analysis for Materials Obtained Using Selective Laser Melting Technology
1 Introduction
2 Materials and Methods
3 Results and Discussion
4 Conclusions
References
Computing Technologies in Information Security Applications
Fourier Chromagrams for Fingerprinting, Verification and Authentication of Digital Audio Recordings
1 Introduction
2 Proposed Approach
3 Dataset for Testing
4 Numerical Experiments and Discussion
4.1 Influence of Changes in Tempo
4.2 Noise Robustness
4.3 Influence of Playing Instruments and Music Chords
5 Conclusion
References
Methodology of Expert-Agent Cognitive Modeling for Preventing Impact on Critical Information Infrastructure
1 Introduction
2 Analysis of Existing Models
3 Expert-Agent Cognitive Modeling
3.1 Model for Preventing Impact on Critical Information Infrastructure
3.2 Machine Learning
3.3 Neural Network
4 Comparative Analysis of Alarming Models
5 Conclusion
References
Author Index
Organization
Contents
Hardware for High-Performance Computing and Signal Processing
Design of a Pipeline Computing Module as Part of a Specialized VLSI
1 Introduction
2 Flow Control in a Pipeline Computing Structure
2.1 Development of Components of a Pipeline Computer
2.2 Integration of a Pipeline Computer into a Computer System (VLSI)
3 Selecting the Pipeline Depth Based on Topological Basis
4 Characteristics of a Pipeline Accelerator in FPGA CAD
5 Conclusions
References
Speech Enhancement Based on Two-Stage Neural Network with Structured State Space for Sequence Transformation
1 Introduction
2 Neural Network Model Used for Speech Enhancement
2.1 Speech Enhancement Applying Complex Ideal Ratio Masks
2.2 Proposed Model
2.3 Structured State Space for a Sequence Layer
2.4 Full-Band and Sub-band Transform Blocks
2.5 Loss Functions
3 Experiments and Results
3.1 Datasets
3.2 Implementation Details
3.3 Performance Evaluation Metrics
3.4 Loss Function Choice
3.5 Optimal Number of S4 Layers
3.6 Comparison of the Proposed Model Having Alternatives
3.7 Discussion
4 Conclusion
References
Designing a Graphics Accelerator with Heterogeneous Architecture
1 Introduction
2 Architecture of a Specialized Graphics Accelerator
3 Architectural Solutions for Graphics Accelerator
3.1 Programmable Task Distribution
3.2 Software Memory Management
3.3 Distribution of Tasks Depending on Data
3.4 Redistribution of Resources within the Computing Cluster
3.5 Thread Management According to the SIMD Approach
3.6 Combining Raster and Cache Memory for General-Purpose Computing
4 Automation of Design of Specialized Nodes
5 Conclusions
References
Spectrophotometer for Field Studies
1 Introduction
2 Proposed Approach
2.1 AS7341 Integral Spectrometer
2.2 Wireless Microcontroller CC2640
2.3 Structure of the Spectrophotometer
2.4 Software Structure
3 Experiments and Results
3.1 Dataset Description
3.2 Discussion
4 Conclusion
References
Comparative Study of Practical Implementation of Time Delay Estimation Methods on Single Board Computer
1 Introduction
2 Overview of TDE Methods
2.1 Time-Domain Methods
2.2 Frequency-Domain Methods
2.3 Signal Processing in Practical TDE
2.4 Variants of Fourier Transform Implementation
3 Computational Study
3.1 Raspberry Pi 4B Hardware
3.2 Testing Software
3.3 Estimation of Computation Time
3.4 Estimation of Memory Usage
4 Discussion
5 Conclusion
References
Information Technologies and Computer Simulation of Physical Phenomena
Study of the Functional Characteristics of TiNi Coatings by the Computer-Aided Simulation Using Parallel Computing
1 Introduction
2 Brief Description of the Program Complex for Coating Simulation
3 Implementation of Program System for Parallel Running of SIMD-Tasks for Coating Simulation
3.1 Functionality of the Program System
3.2 Computational Tests for Determination of Optimal Loading Mode of Computing Nodes
4 Using a Program System with an Architecture for Parallel Running of Similar SIMD-Tasks for Coating Simulation
5 Conclusions
References
On Hierarchical Convergence of the Heterogeneous Multiscale Finite Element Method Using Polyhedral Supports
1 Introduction
2 Problem Statement and Solution Method
2.1 Discretization of Computational Domain
2.2 Heterogeneous Multiscale Finite Element Method
3 Numerical Estimation of FE-HMM Accuracy
3.1 Homogeneous Computational Domain
3.2 Computational Domain with Inclusions
4 Parallelization of Computing Schemes
5 Conclusions
References
Parallelization of Finite-Volume Numerical Methods of Computational Fluid Dynamics by Means of Shared Memory Computing Systems
1 Introduction
2 Problem Statement
3 Parallelization of Computing Flows of Conserved Values
3.1 Reduction of the Parallelization Problem
3.2 Application of Graph Coloring in Parallelization
3.3 Optimized Greedy Coloring Method
3.4 Verification of the Algorithm
4 Discussion
References
Computing Technologies in Data Analysis and Decision Making
Development and Comparative Study of Data Compression Methods Used in Technical Monitoring Systems
1 Introduction
2 Problem Description
3 Methods Based on Data Normalization and Structuring
4 Compression Methods Based on Difference Schemes and RLE-Algorithms
5 Lossy Compression Methods
6 Conclusions
References
Object Recognition Based on Three-Dimensional Computer Graphics
1 Introduction
2 Problem Statement
3 Related Work
4 Proposed Solution
5 Proof of Work
6 Conclusions
References
Adaptive Methods for the Structural Optimization of Neural Networks and Their Ensemble for Data Analysis
1 Introduction
2 Materials and Methods
2.1 Designing the Structure of Neural Networks Using the Probabilistic Approach
2.2 Providing Diversity for Individual Neural Network Solvers
2.3 Ensemble Design Using Modified Genetic Programming
3 Numerical Experiments
3.1 Simulated Datasets
3.2 Test Problems
3.3 Pool of Methods
3.4 Raw Results Processing
3.5 Results and Discussion
4 Conclusion
References
Self-adaptation Method for Evolutionary Algorithms Based on the Selection Operator
1 Introduction
2 Methods of Self-adaptation for Genetic Algorithm
2.1 Genetic Algorithm
2.2 Fitness-Based Adaptation
2.3 Population-Level Dynamic Probabilities
2.4 Self-configuring Genetic Algorithm
2.5 SHAGA
3 Proposed Approach
4 Experimental Setup and Results
4.1 Parameters of the Genetic Algorithm
4.2 Results of Solving Binary Optimization Problems
4.3 Results of Solving Real-Valued Optimization Problems
5 Conclusion
References
Application of U-Net Architecture Neural Network for Segmentation of Brain Cell Images Stained with Trypan Blue
1 Introduction
2 Materials and Methods
3 Results
4 Discussion
5 Conclusions
References
Language Model Architecture Based on the Syntactic Graph of Analyzed Text
1 Introduction
2 Materials and Methods
2.1 Graph Core and Function Levels
2.2 Graphlet-Based Methods
2.3 Metric of Similarity of Neighbors
2.4 Model-Parametric Representation
3 Results and Discussion
4 Conclusions
References
Information and Computing Technologies in Automation and Control Science
Classic and Modern Methods of Automatic Parking Control of Self-driving Cars
1 Introduction
2 Peculiarities of the Parking Control Problem
2.1 Equations of SDC Parking Dynamics
2.2 Mechanical Limitations of SDC Parking
2.3 Space Limitations of SDC Parking
2.4 Boundary Conditions of SDC Parking
2.5 Setting the SDC Parking Control Problem
3 Parking Based on Dubins and Reeds-Shepp Motion Models
4 Parking Based on Reinforcement Learning
5 Conclusion
References
Application of Smoothing Technique to Model Predictive Traffic Signal Control
1 Introduction
2 Transport Network Model
2.1 Traffic Controllers
2.2 Macroscopic Traffic Flow Model
2.3 Numerical Scheme
2.4 Intersection Model
3 Model Predictive Control
3.1 Target Function
3.2 Zero-Order Smoothing Optimization Technique
4 Numerical Experiments
5 Conclusion
References
Design Features of the Frequency-Controlled Electric Drive for Positioning Mechanisms
1 Introduction
2 Problems of Positioning Mechanism Automation
2.1 Control Panel Functions
2.2 Design Constraints
2.3 Principle of Operation
3 Method of Adjusting the Speed of a Positioning Mechanism
3.1 Control Logic
3.2 Results
3.3 Discussion
4 Conclusion
References
Development of a Mathematical Model to Study the Energy Indicators of Electric Drives Using the DFC-IM System
1 Introduction
2 Principles ForPositioning Accuracy
2.1 Control Panel Functions
2.2 Generation of Mathematical Model Equations
3 Block Diagram of Calculation Software in the DFC-IM System
3.1 Control Logic
3.2 Results
3.3 Discussion
4 Conclusion
References
Intelligent Data Analysis for Materials Obtained Using Selective Laser Melting Technology
1 Introduction
2 Materials and Methods
3 Results and Discussion
4 Conclusions
References
Computing Technologies in Information Security Applications
Fourier Chromagrams for Fingerprinting, Verification and Authentication of Digital Audio Recordings
1 Introduction
2 Proposed Approach
3 Dataset for Testing
4 Numerical Experiments and Discussion
4.1 Influence of Changes in Tempo
4.2 Noise Robustness
4.3 Influence of Playing Instruments and Music Chords
5 Conclusion
References
Methodology of Expert-Agent Cognitive Modeling for Preventing Impact on Critical Information Infrastructure
1 Introduction
2 Analysis of Existing Models
3 Expert-Agent Cognitive Modeling
3.1 Model for Preventing Impact on Critical Information Infrastructure
3.2 Machine Learning
3.3 Neural Network
4 Comparative Analysis of Alarming Models
5 Conclusion
References
Author Index
