Data-driven future wireless communication

Abstract

Besides supporting the traditional requirement of large-area coverage and high-data rate transmission services, IMT-2020 defined by ITU-R is also expected to supply smart and reliable interconnection among humans and things. Thus, the vision of IMT-2020 presents the convergence of wireless communication, Internet, Internet of Things (IoT) and machine-type communication (MTC), which together brings an explosive increase to traffic volume and stimulates wireless communication to the time of big data. Obviously, such vision poses big challenges to 5G and future wireless communication. This keynote speech discusses the application of computer science into future wireless communication, especially data mining techniques to accelerate the wireless research and development. Firstly, the big data tendency of wireless communication is presented and the possible ways to combine them are pointed out. In particular, a three-level structure of a wireless system is defined in order to classify the propagation environments, which will bring the complex combination into simplicity. Considering these three levels, there are different tasks like service prediction and pushing, self-organized networking, self-adapting large-scale fading modeling and so on, which can be abstracted into problems like regression, classification, clustering, etc. Since there are many powerful algorithms in the data mining domain to accomplish them, we can expect a data-driven future wireless communication to make our lives and society convenient.