The aim of the module is to introduce students to the fundamentals of machine learning and then to apply the advanced machine learning principles for the design and optimisation of wireless communications systems and mobile networks.
Recently, the research and development in wireless communications have been focused on the techniques for the fifth generation (5G) wireless systems and the potential to make these networks intelligent by adding machine learning. Therefore, this course motivates to deliver a general introduction and fundamentals of machine learning followed by the application of machine learning in the design of physical layer techniques in wireless communications and in the optimisation of mobile networks.
Exclusions: Cannot be taken with COMP3222 or COMP3223 or COMP6245 or COMP6246 or COMP6208.
Pre-requisites: ELEC3203 or ELEC3204 or ELEC3218
Aims and Objectives
Subject Specific Intellectual and Research Skills
Having successfully completed this module you will be able to:
- design and optimise intelligent mobile networks by applying the principles of machine learning
- apply the mathematical principles of probability, linear algebra and optimisation
- apply machine learning principles in the design of some physical layer techniques in wireless communications
- understand the principles of machine learning and apply the fundamental principles for regression and classification
Knowledge and Understanding
Having successfully completed this module, you will be able to demonstrate knowledge and understanding of:
- The fundamentals of Machine learning
- The application of machine learning in network design
- The application of machine learning in the design of physical layer techniques for wireless communications
General introduction and fundamentals of machine learning
- Introduction to learning and machine learning: supervised/unsupervised/reinforcement learning
- Revision of probability and statistics revision
- Revision of linear algebra
- Fundamentals of numerical optimisation
- Introduction to machine learning assisted linear regression and classification
Machine learning for wireless communications
- Machine learning for physical layer design
1- Adaptive modulation and coding (AMC): classical AMC, using support vector machines, using k-nearest
neighbours, using k-means, using reinforcement learning
2- Code Division Multiple Access (CDMA): classical design, using learning such as ACO/GA
3- Precoder design: classical design, channel prediction using deep learning
- Machine learning for mobile network design
1- User grouping/clustering in D2D, HetNets for offloading
2- Traffic prediction and interference management in HetNets
3- Clustering of small cells in Hetnet to avoid interference in CoMP
Learning and Teaching
Teaching and learning methods
The teaching and learning methods of the module include lectures, tutorials, coursework, coursework feedback, etc.
|Wider reading or practice||14|
|Preparation for scheduled sessions||18|
|Total study time||150|
Resources & Reading list
T. J. OShea, T. Erpek, and T. C. Clancy. Deep learning based MIMO communications.
S. Bi, R. Zhang, Z. Ding, and S. Cui. Wireless communications in the era of big data.
M. A. Alsheikh, S. Lin, D. Niyato and H. P. Tan. Machine Learning in Wireless Sensor Networks: Algorithms, Strategies, and Applications.
T. J. OShea, K. Karra, and T. C. Clancy. Learning approximate neural estimators for wireless channel state information.
T. J. OShea and J. Hoydis. An introduction to machine learning communications systems.
C. Jiang and H. Zhang and Y. Ren and Z. Han and K. C. Chen and L. Hanzo. Machine Learning Paradigms for Next-Generation Wireless Networks.
Christopher M. Bishop. Pattern Recognition and Machine Learning.
Jeremy Watt and Reza Borhani. Machine Learning Refined: Foundations, Algorithms, and Applications.
David J.C. Mackay. Information Theory, Inference and Learning Algorithms.
Summative assessment description
Referral assessment description
Repeat assessment description
Repeat type: Internal & External