When Big Data Meets Machine Learning in Wireless Networks

Supervisor: Dr Yuanwei Liu

Research group(s): Networks

Recent several decades have witnessed the exponential growth in commercial data services, which lead to step in the so-called big data era. The pervasive increasing data traffic present both the imminent challenges and new opportunities to all aspects of wireless system design, such as efficient wireless caching, base station deployment and adaptive multiple access design. Machine learning, as one of the most promising artificial intelligence tools, has been invoked in many areas both in the academia and industry. Nevertheless, the application of machine learning in wireless communication scenarios is still in its infancy, which motivates to develop this project. The aim of this project is to use social media data to predict the requirements of mobile users for improving the performance of wireless networks. More particularly, a unified machine learning framework with the aid of the social media data is proposed in this project. Four stages are included in the proposed framework, which consists syntax processing, semantics analysis, data modelling and online prediction/refinement. The main benefits of the proposed framework is by utilizing the social media data which reflect the real requirements of users, to assist refining the motivation, problem formulation, and methodology of powerful machine learning algorithms in the context of wireless networks. The proposed framework is as follows.