First special issue on The
impact of Artificial Intelligence on communication networks and services
Foreword
Foreword
Editor-in-Chief Message
EDITORIAL BOARD
TABLE OF CONTENTS
LIST OF ABSTRACTS
Invited Papers
Selected Papers
RESPONSIBLE ARTIFICIAL
INTELLIGENCE: DESIGNING AI FOR HUMAN VALUES
1. INTRODUCTION
2. EXPECTATIONS ON THE IMPACT
OF AI
2.1 Literature analysis
2.2 The views of AI experts
3. RESPONSIBILITY IN AI
3.1. Responsible AI
challenges
4. CONCLUDING REMARKS
REFERENCES
RECONFIGURABLE PROCESSOR FOR
DEEP LEARNING IN AUTONOMOUS VEHICLES
1. INTRODUCTION
2. TRENDS IN AUTONOMOUS
VISION
2.1. An overview of an ADAS
system
2.2. Traditional algorithms
of autonomous vision
2.3. The rise of
convolutional neural network (CNN)
3. PROCESSORS FOR REAL-TIME
AUTONOMOUSVISION
3.1. Heterogeneous platforms
for CNN acceleration
3.2. Chances and challenges
for reconfigurable processors
3.3. Related reconfigurable
processors
4. SOFTWARE-HARDWARE
CO-DESIGN FOR A RECONFIGURABLEAUTONOMOUS VISION SYSTEM
4.1. The overall system
workflow
4.2. Compression methods
4.3. Hardware architecture
design
4.4. Performance evaluation
4.5. Tingtao: an ASIC-based
reconfigurable accelerator
5. CONCLUSION
REFERENCES
REAL-TIME MONITORING OF THE
GREAT BARRIER REEF USING INTERNET OF THINGS WITH BIG DATA ANALYTICS
1. INTRODUCTION
2. THE GREAT BARRIER REEF
MONITORING
2.1. Sensor network and
sensing elements
2.2. Securing buoys and
casing
2.3. Communication and
scheduling constraints
2.4. Scalable networking
architecture
2.5. Detecting interesting
events using AI
3. CLOUD-CENTRIC NETWORK
ARCHITECTURE FOR REAL-TIME MONITORING
3.1. Networking framework
3.2. Data framework
4. CASE STUDY: DETECTING CYCLONE
HAMISH ON HERON ISLAND OF GBR USING AI
4.1. WSN network
architecture
4.2. Cyclone Hamish
detection using AI
4.3. System of systems (SoS)
view of integrated AI
4.4. Open research
challenges
CONCLUSION
ACKNOWLEDGEMENT
REFERENCES
INCLUSION OF ARTIFICIAL
INTELLIGENCE IN COMMUNICATION NETWORKS AND SERVICES
1. INTRODUCTION
2. TRENDS IN COMMUNICATION
NETWORKS AND SERVICES
2.1. Characterized
requirements
2.2. Multimedia services
2.3. Precision management
2.4. Predictable future
2.5. Intellectualization
2.6. More attention to
security and safety
3. ADVANTAGES OF AI
3.1. Abilities of learning
3.2. Abilities of
understanding and reasoning
3.3. Ability of
collaborating
4. POSSIBILITY TO USE AI IN
COMMUNICATIONS
4.1. AI in SDN
4.2. AI in NFV
4.3. Network monitor and
control
5. AN AI-BASED NETWORK
FRAMEWORK
5.1. Intelligence plane
5.2. Agent plane
5.3. Business plane
6. A FINE EXAMPLE
7. CONCLUSION
REFERENCES
EXPLAINABLE ARTIFICIAL
INTELLIGENCE: UNDERSTANDING, VISUALIZING AND INTERPRETING DEEP LEARNING
MODELS
1. INTRODUCTION
2. WHY DO WE NEED EXPLAINABLE
AI?
3. METHODS FOR VISUALIZING,
INTERPRETING AND EXPLAINING DEEP-LEARNING MODELS
3.1. Sensitivity analysis
3.2. Layer-wise relevance
propogation
3.3. Software
4. EVALUATING THE QUALITY OF
EXPLANATIONS
5. EXPERIMENTAL EVALUATION
5.1. Image classification
5.2. Text document
classification
5.3. Human action
recognition in videos
6. CONCLUSION
REFERENCES
THE CONVERGENCE OFMACHINE
LEARNING ANDCOMMUNICATIONS
1. INTRODUCTION
2. MACHINE LEARNING IN
COMMUNICATIONS
2.1. Communication networks
2.2. Wireless communications
2.3. Security, privacy and
communications
2.4. Smart services, smart
infrastructure and IoT
2.5. Image and video
communications
3. EXEMPLAR APPLICATIONS IN
WIRELESS NETWORKING
3.1. Reconstruction of radio
maps
3.2. Deep neural networks
for sparse recovery
4. FUTURE RESEARCH TOPICS
4.1. Low complexity models
4.2. Standardized formats
for machine learning
4.3. Security and privacy
mechanisms
4.4. Radio resource and
network management
5. CONCLUSION
REFERENCES
APPLICATION OF AI TO MOBILE
NETWORK OPERATION
1. INTRODUCTION
1.1. Characteristics of
artificial intelligence (AI)
1.2. Trends of mobile
network
2. ISSUES OF MOBILE NETWORK
OPERATION
2.1. Issues of planning
process
2.2. Issues of maintenance
process
3. NETWORK OPERATION WITH AI
3.1. Approach to applying AI
to planning process
3.2. Approach to applying AI
to maintenance process
3.2.1. Application of AI to
network monitoring
3.2.1.1. Necessity of
service monitoring
3.2.1.2. Application of AI
to service monitoring
4. CONCLUSION
REFERENCES
ON ADAPTIVE NEURO-FUZZY MODEL
FOR PATH LOSS PREDICTION IN THE VHF BAND
1. INTRODUCTION
2. METHODOLOGY
2.1. Measurement Campaign
Procedure
2.2. Prediction Model
3. RESULTS AND DISCUSSION
4. CONCLUSION
ACKNOWLEDGMENT
REFERENCES
BEYOND MAD?: THE RACE FOR
ARTIFICIAL GENERAL INTELLIGENCE
1. INTRODUCTION
2. ARMS RACES AND AGI: BEYOND
MAD?
2.1. Actors in the AGI race
2.1.1. State actors
2.1.2. Corporate actors
2.1.3. Rogue actors
INTELLIGENCEBEYOND MAD?: THE
RACE FOR ARTIFICIAL GENERAL INTELLIGENCEBEYOND MAD?: THE RACE FOR
ARTIFICIAL GENERAL INTELLIGENCEBEYOND MAD?: THE RACE FOR ARTIFICIAL
GENERAL INTELLIGENCEBEYOND MAD?: THE RACE FOR ARTIFICIAL GENERAL
INTELLIGENCEBEYOND MAD?: THE RACE FOR ARTIFICIAL GENERAL
INTELLIGENCEBEYOND MAD?: THE RACE FOR ARTIFICIAL GENERAL
INTELLIGENCEBEYOND MAD?: THE RACE FOR ARTIFICIAL GENERAL
INTELLIGENCEBEYOND MAD?: THE RACE FOR ARTIFICIAL GENERAL
INTELLIGENCEBEYOND MAD?: THE RACE FOR ARTIFICIAL GENERAL INTELLIGENCE
3. AGI AND VALUE ALIGNMENT
4. SHAPING AGI RESEARCH
5. PERSPECTIVES AND SOLUTIONS
5.1. Solution 1: Global
collaboration on AGI development and safety
5.2. Solution 2: Global Task
Force on AGI to monitor, delay and enforce safety guidelines
6. CONCLUSION
ACKNOWLEDGEMENT
REFERENCES
ARTIFICIAL INTELLIGENCE FOR
PLACE-TIME CONVOLVED WIRELESS COMMUNICATION NETWORKS
1. INTRODUCTION
2. BACKDROP: OSTENTANEITY OF
AN EVENT
2.1. Unostentatious Events
2.2. Place-Time Events
2.3. Ostentatious Events
2.4. Appropriateness of APE
3. PLACE TIME COVERAGE AND
CAPACITY: NSP's DUO ORDEAL
3.1. Understanding the
network environment
3.2. The NSP's nightmare:
Ostentations network behavior
3.3. Place Time Coverage
3.4. Place Time Capacity
3.5. PTC2: Need of
unorthodox approach
4. DEALING WITH PTC2
‘ARTIFICIAL INTELLIGENTLY'
4.1. AI-Assisted
Architecture
4.2. How should AAA respond
to the PTC2?
4.2.1. Information
aggregation
4.2.2. Deep Learning
4.2.3. Disseminating
actions
4.2.4. Integrating
Alternate Solutions
REFERENCES
BAYESIAN ONLINE LEARNING-BASED
SPECTRUM OCCUPANCY PREDICTION IN COGNITIVE RADIO NETWORKS
1. INTRODUCTION
2. SYSTEM MODEL
3. ENERGY DETECTION MODEL
4. TIME-SERIES GENERATION
BASED ON ENERGY PRIMARY USER DETECTION SEQUENCE
5. TIME-SERIES PREDICTION
BASED ON BAYESIAN ONLINE LEARNING ALGORITHM
6. SIMULATION RESULTS
7. CONCLUSION
ACKNOWLEDGEMENT
REFERENCES
THE EVOLUTION OF FRAUD:
ETHICAL IMPLICATIONS IN THE AGE OF LARGE-SCALE DATA BREACHES AND
WIDESPREAD ARTIFICIAL INTELLIGENCE SOLUTIONS DEPLOYMENT
1. INTRODUCTION
2. KEY IDEAS
2.1. Data brokers
2.2 Mosaic effect
3. RECENT DATA BREACHES
3.1. Yahoo
3.2. Adult Friend Finder
3.3. eBay
3.4. Equifax
4. EVOLUTION OF FRAUD
5. OTHER ETHICAL CONSEQUENCES
6. AGGRAVATION BY AI ADVANCES
7. CHALLENGES TO BE ADDRESSED
8. CONCLUDING REMARKS
ACKNOWLEDGEMENT
REFERENCES
MACHINE INTELLIGENCE
TECHNIQUES FOR NEXT-GENERATION CONTEXT-AWARE WIRELESS NETWORKS
1. INTRODUCTION
2. DATA ACQUISITION AND
KNOWLEDGE DISCOVERY
A. Data acquisition
B. Knowledge discovery
3. NETWORK PLANNING
A. Node deployment, Energy
consumption and RF planning
B. Configuration parameter
and service planning
4. NETWORK OPERATION AND
MANAGEMENT
A. Resource allocation and
management
B. Security and privacy
protection
C. Latency optimization for
tactile applications
5. DESIGN CASE STUDIES
A. Machine learning for CIR
prediction
B. Context-aware data
transmission using NLP techniques
6. CONCLUSIONS
REFERENCES
NEW TECHNOLOGY BRINGS NEW
OPPORTUNITY FOR TELECOMMUNICATION CARRIERS: ARTIFICIAL INTELLIGENT
APPLICATIONS AND PRACTICES IN TELECOM OPERATORS
1. INTRODUCTION
2. THE UNIQUE ADVANTAGES FOR
OPERATORS TO DEVELOP AI
3. TELECOM OPERATORS'
PRACTICES IN THE FIELD OF ARTIFICIAL INTELLIGENCE
3.1. AI-based energy saving
product in data centers
3.2. AI-based public
security management platform
3.3. AI-based health
management and control
4. SUMMARY AND PROSPECT
REFERENCES
CORRELATION AND DEPENDENCE
ANALYSIS ON CYBERTHREAT ALERTS
1. INTRODUCTION
2. METHODOLOGY
2.1. Experiment set up:
network description and data mining approach