AI Applications to Communications and Information Technologies The Role of Ultra Deep Neural Networks

Apply the technology of the future to networking and communications.

Artificial intelligence, which enables computers or computer-controlled systems to perform tasks which ordinarily require human-like intelligence and decision-making, has revolutionized computing and digital industries like few other developments in recent history. Tools like artificial neural networks, large language models, and deep learning have quickly become integral aspects of modern life. With research and development into AI technologies proceeding at lightning speeds, the potential applications of these new technologies are all but limitless.

AI Applications to Communications and Information Technologies offers a cutting-edge introduction to AI applications in one particular set of disciplines. Beginning with an overview of foundational concepts in AI, it then moves through numerous possible extensions of this technology into networking and telecommunications. The result is an essential introduction for researchers and for technology undergrad/grad student alike.

AI Applications to Communications and Information Technologies readers will also find:

• In-depth analysis of both current and evolving applications
• Detailed discussion of topics including generative AI, chatbots, Automatic Speech Recognition, image classification and recognition, IoT, smart buildings, network management, network security, and more
• An authorial team with immense experience in both research and industry

AI Applications to Communications and Information Technologies is ideal for researchers, industry observers, investors, and advanced students of network communications and related fields.

Daniel Minoli is Principal Consultant for DVI Communications, New York, USA, and a longtime Expert Witness and Testifying Expert in networking, wireless, video, IoT, and VoIP. In addition to working as Director of Engineering for gamut of premiere high-tech firms, he has acted as adjunct instructor at New York University and Stevens Institute of Technology, USA for twenty years. He has published extensively on networks, IP/IPv6, video, wireless communications, and related subjects.

Benedict Occhiogrosso is Co-Founder of DVI Communications, New York, USA, with extensive experience as a technology engineer, manager and executive. He is a subject matter expert in several disciplines now enhanced by artificial intelligence including telecommunications networking, speech recognition, image processing and building management systems. He has also served as a testifying expert witness and advisor on patent portfolios.

Ch 1: Overview

1.1 Introduction and Basic Concepts

1.1.1 Machine Learning

1.1.2 Deep Learning

1.1.3 Activation Functions

1.1.4 MLPs

1.1.5 RNNs

1.1.6 CNNs

1.1.7 Comparison

1.2 Learning Methods

1.3 Areas of applicability

1.4 Scope of this text

References

Glossary

Ch 2: Current and evolving applications to Natural Language Processing

2.1 Scope

2.2 Introduction

2.3 Overview of Natural Language Processing and Speech Processing

2.3.1 Feed-forward NN

2.3.2 RNNs

2.3.3 LSTM

2.3.4 Attention

2.3.5 Transformer

2.4 NLP/NLU Basics

2.4.1 Pre-training

2.4.2 NLP/NLG Architectures

2.4.3 Encoder-Decoder Methods

2.4.4 Application of Transformer

2.4.5 Other Approaches

2.5 NLG Basics

2.6 Chatbots

References

Glossary

Chapter 3:  Current and evolving applications to Speech Processing

3.1 Scope

3.2 Overview

3.2.1 Traditional Approaches

3.2.2 DNN-based feature extraction

3.3 Noise Cancellation

3.3.1 Approaches

3.3.2 Example of a system supporting noise cancellation

3.4 Training

3.5 Applications to voice interfaces used to control home devices & Digital Assistant Applications

3.6 Attention-based models

3.7 Sentiment Extraction

3.8 End-to-end learning

3.9 Speech Synthesis

3.10 Zero-shot TTS

3.11 VALL-E: Unseen speaker as an acoustic prompt

References

Glossary

Ch 4: Current and evolving applications to Video and Imaging

4.1 Overview

4.2 Convolution process

4.3 CNNs

4.3.1 Nomenclature

4.3.2 Basic Formulation of the CNN layers

4.3.3 Fully convolutional networks (FCN)

4.3.4 Convolutional Autoencoders

4.3.5 R-CNNs, Fast R-CNN, Faster R-CNN

4.4 Imaging Applications

4.4.1 Basic Image Management

4.4.2 Image segmentation and classification

4.4.3 Illustrative examples of a DNN/CNN

4.4.4 Well-known Image classification networks

4.5 Specific application Examples

4.5.1 Semantic segmentation and semantic edge detection

4.5.2 CNN Filtering Process For Video Coding

4.5.3 Virtual clothing

4.5.5 Object Detection Applications

4.5.6 Classifying video data

4.5.7 Example of Training

4.5.8 Example: Image reconstruction is used to remove artifacts

4.5.9 Example: Video Transcoding/Resolution-enhancement

4.5.10 Facial expression recognition

4.5.11 Transformer Architecture for image processing

4.5.12 Example: A GAN Approach/Synthetic Photo

4.5.13 Situational Awareness

4.6 Other models: Diffusion and Consistency Models

References

Glossary

Ch 5: Current and evolving applications to IoT and applications to Smart buildings and energy management

5.1 Introduction

5.1.1 IoT Applications

5.1.2 Smart Cities

5.2 Smart Building ML Applications

5.2.1 Basic Building Elements

5.2.2 Particle Swarm Optimization

5.2.3 Specific ML Example – Qin Model

5.2.3.1 EnergyPlus™

5.3.3.2 Modeling and Simulation

5.2.3.3 Energy Audit Stage

5.2.3.4 Optimization Stage

5.2.3.5 Model Construction

5.2.3.6 EnergyPlus Models

5.2.3.7 Real-Time Control Parameters

5.2.3.8 Neural Networks in the Qin Model (DNN, RNN, CNN)

5.2.3.9 Finding Inefficiency Measures

5.2.3.10 Particle Swarm Optimizer

5.2.3.11 Integration of Particle Swarm Optimization with Neural Networks

5.2.3.12 Deep Reinforcement Learning

5.2.3.13 Deployments

5.3 Example of a Commercial Product – BrainBox

5.3.1 Overview

5.3.2 LSTM Application - technical background

5.3.3 BrainBox Energy Optimization system

References

Glossary

Ch 6: Current and evolving applications to Network Cybersecurity

6.1 Overview

6.2 General Security Requirements

6.3 Corporate resources/intranet Security Requirements

6.3.1 Network And Endsystem Security Testing

6.3.2 Application Security Testing

6.3.3 Compliance Testing

6.4 IoT Security (IoTSec)

6.5 Blockchains

6.6 Zero Trust Environments

6.7 Areas of ML applicability

6.7.1 Example of cyberintrusion detector

6.7.2 Example of Hidden Markov Model (HMM) for intrusion detection

6.7.3 Anomaly Detection Example

6.7.4 Phishing Detection Emails Using Feature Extraction

6.7.5 Example of classifier engine to identify phishing websites

6.7.6 Example of system for data protection

6.7.7 Example of an integrated cybersecurity threat management

6.7.8 Example of a Vulnerability Lifecycle Management System

References

Glossary

Ch 7: Current and evolving applications to Network Management

7.1 Overview

7.2 Examples of Neural Network-Assisted Network Management

7.2.1 Example of NN-based Network Management system (FM)

7.2.2 Example of a model for predictions related to the operation of a telecommunication network (FM)

7.2.3 Prioritizing Network Monitoring Alerts (FM, PM)

7.2.4 System for Recognizing And Addressing Network Alarms (FM)

7.2.5 Load Control Of An Enterprise Network (PM)

7.2.6 Data Reduction To Accelerate Machine Learning For Networking (FM, PM)

7.2.7 Compressing Network Data (PM)

7.2.8 ML Predictor For A Remote Network Management Platform (FM, PM, CM, AM)

7.2.9 Cable Television (CATV) Performance Management system (PM)

References

Glossary

The New copy of this book will include any supplemental materials advertised. Please check the title of the book to determine if it should include any access cards, study guides, lab manuals, CDs, etc.

The Used, Rental and eBook copies of this book are not guaranteed to include any supplemental materials. Typically, only the book itself is included. This is true even if the title states it includes any access cards, study guides, lab manuals, CDs, etc.