Teaching
"EE150: Signal and System (Spring, 2023; Spring, 2024)"
"CS287: Network Intelligence (Spring, 2022; Autumn, 2022; Autumn, 2023)"
Basic Course Information
Course Level: Graduate
Credit/Contact Hour: 3/48
Major: CS/EE/IE
Prerequisite: EE150 Introduction to Communication Systems, CS182 Introduction to Machine Learning
School/Institute: SIST
Instructor: Yijie (Lina) Mao
Course Introduction
Recent advances in artificial intelligence (AI) and machine learning (ML) has opened a new era of “intelligent wireless networks”, which pushes forward the use of AI technologies to resolve the challenges such as network planning, operation, management and troubleshooting in wireless networks. The development of intelligent wireless networks has been growing explosively and is becoming one of the biggest trends in related academic, research, and industry communities.
This course is designed for postgraduate students who are working on the frontiers of wireless communication networks. This course aims at delivering a comprehensive overview of ML, and its application to deal with various problems in the physical (PHY) and medium access control (MAC) layers of communication networks such as transceiver design, channel estimation, prediction, and compression, resource allocation, semantic communications, distributed and federated learning and communications, etc.
Learning Goal
There are three main learning goals of this course. The first goal is to provide a comprehensive overview of convex optimization and ML technologies. The second goal is to disclose timely research challenges in wireless communication networks. The third goal is to provide a state-of-the-art review of how to apply AI and ML technologies to deal with various problems in the lower layers of communication networks such as transceiver design, channel estimation, prediction, and compression, resource allocation, semantic communications, distributed and federated learning and communications, etc. This course plays an important role in understanding the frontiers of wireless communication networks and cultivating students' ability to master and apply interdisciplinary knowledge of AI and wireless communication.
Instructional Pedagogy
The course adopts a lecture-based and seminar-assisted teaching method. This course will also organize panel discussion for students to read, discuss and present articles of intelligent wireless networks in the mid-term, and guide students to conduct course projects on the cutting-edge topics of intelligent wireless networks at the end of the term. Ultimately, students are capable of answering the following critical questions:
What are the major research challenges in wireless communication and what are the limitations of existing approaches to address those challenges?
What are the major benefits of empowering AI in wireless communications networks and how to utilize AI for efficient data acquisition, knowledge discovery, network planning, operation and management of wireless networks?
How to effectively redesign the PHY or MAC layers so as to incorporate AI?
What are the research challenges of AI-empowered wireless networks?
Course Content and Schedule
Chapter I: Introduction to Machine Learning (Week 1,2)
Overview of AI and Wireless Communications
Basic Principles of Machine Learning
Linear Regression
Linear Classification
Feedforward Neural Network
Convolutional Neural Network
Recurrent Neural Network
Chapter II: Introduction to MIMO Wireless Networks (Week 3,4)
Point-to-Point MIMO Channels
Multi-user MIMO
Multiple Access Techniques
Chapter III: Machine Learning for Wireless Networks (Week 5,6,8)
Machine Learning for Symbol Detection
Machine Learning for Resource Allocation
Machine Learning for Channel Estimation
Semantic Communications
Chapter IV: Wireless Networks for Machine Learning (Week 9,10)
Introduction to Edge AI
Introduction to Federated Learning
Wireless Techniques for Edge Training
Edge Inference Systems
Mid-term Paper Presentation (Week 7)
Choose one paper from the paper pool (or others I agree)
In depth review (novelty system model algorithm results)
Time: 15 minute presentation 5 minute Q&A
Form: Oral presentation with slides (English recommended)
Final Project Presentation (Week 11, 12)
Check the Course Final Project for detailed requirements.
Grading Policy
(1) Assessment method: "homework + technical report + defense presentation" form of assessment.
(2) Grade evaluation: "30% homework, 30% technical report, 30% defense presentation, 10% attendance".
Recommended Readings
Students are recommended to read the Best Readings summarized in:
Course Level: Graduate
Credit/Contact Hour: 3/48
Major: CS/EE/IE
Prerequisite: EE150 Introduction to Communication Systems, CS182 Introduction to Machine Learning
School/Institute: SIST
Instructor: Yijie (Lina) Mao
What are the major research challenges in wireless communication and what are the limitations of existing approaches to address those challenges?
What are the major benefits of empowering AI in wireless communications networks and how to utilize AI for efficient data acquisition, knowledge discovery, network planning, operation and management of wireless networks?
How to effectively redesign the PHY or MAC layers so as to incorporate AI?
What are the research challenges of AI-empowered wireless networks?
Chapter I: Introduction to Machine Learning (Week 1,2)
Overview of AI and Wireless Communications
Basic Principles of Machine Learning
Linear Regression
Linear Classification
Feedforward Neural Network
Convolutional Neural Network
Recurrent Neural Network
Chapter II: Introduction to MIMO Wireless Networks (Week 3,4)
Point-to-Point MIMO Channels
Multi-user MIMO
Multiple Access Techniques
Chapter III: Machine Learning for Wireless Networks (Week 5,6,8)
Machine Learning for Symbol Detection
Machine Learning for Resource Allocation
Machine Learning for Channel Estimation
Semantic Communications
Chapter IV: Wireless Networks for Machine Learning (Week 9,10)
Introduction to Edge AI
Introduction to Federated Learning
Wireless Techniques for Edge Training
Edge Inference Systems
Mid-term Paper Presentation (Week 7)
Choose one paper from the paper pool (or others I agree)
In depth review (novelty system model algorithm results)
Time: 15 minute presentation 5 minute Q&A
Form: Oral presentation with slides (English recommended)
Final Project Presentation (Week 11, 12)
Check the Course Final Project for detailed requirements.
(1) Assessment method: "homework + technical report + defense presentation" form of assessment.
(2) Grade evaluation: "30% homework, 30% technical report, 30% defense presentation, 10% attendance".