Cognitive Radios: A Case for Artificial Collective Intelligence in Mobile Communications Networks
time: 2018-04-03

Artificial intelligence, widely known as AI, is improving our life on a daily basis and will continue to transform every inch of our life, from as simple as a washing machine with learning ability to a massive AI-empowered network of driverless automobiles. Turning our attention to the area of mobile communications networks, there is a need for spectrum regulators to find ways to more intelligently and flexibly utilise the precious spectrum, and AI presence is therefore expected. Indeed, the concept of cognitive radio which was first introduced by Joseph Mitola III in 1998, aimed to do exactly that. Over the past decade since his work, scientists and engineers attempted to adopt a game-theoretic approach to tackle the problem of autonomous spectrum sharing by cognitive radios. The results are unfortunately mostly negative reconfirming the fact that competition among selfish individuals is bound to result in the tragedy of the commons, with “cognitive” radios still excessively interfering with each other without a proper reconciliation mechanism. In this talk, we consider the autonomous resource allocation problem for the orthogonal frequency-division multiple-access (OFDMA) interference channel where each user (i.e., cognitive mobile radio) can freely occupy any of the subcarriers in the network, and is required to decide its own subcarrier allocation, with only local channel state information. We use this network as an example, to show that collective intelligence for a group of self-organising cognitive radios is possible, and that cognitive radios can be empowered with forward-looking ability to negotiate with each other for the benefits of not only individuals but all. Simulation results demonstrate that in the example of autonomous OFDMA, the proposed approach can achieve the network sum-rate that is extremely close to the optimal centralised solution. The talk is concluded by a discussion of extending the use of such approach to a hierarchical primary-secondary spectrum sharing network.

 

Kai-Kit Wong received the BEng, the MPhil, and the PhD degrees, all in Electrical and Electronic Engineering, from the Hong Kong University of Science and Technology, Hong Kong, in 1996, 1998, and 2001, respectively. After graduation, he took up academic and research positions at the University of Hong Kong, Lucent Technologies, Bell-Labs, Holmdel, the Smart Antennas Research Group of Stanford University, and the University of Hull, UK. He is Chair in Wireless Communications at the Department of Electronic and Electrical Engineering, University College London, UK.

 

His current research centers around 5G and beyond mobile communications, including topics such as massive MIMO, full-duplex communications, millimetre-wave communications, edge caching and fog networking, physical layer security, wireless power transfer and mobile computing, V2X communications, and of course cognitive radios. There are also a few other unconventional research topics that he has set his heart on, including for example, fluid antenna communications systems, remote ECG detection and etc. He is a co-recipient of the 2013 IEEE Signal Processing Letters Best Paper Award and the 2000 IEEE VTS Japan Chapter Award at the IEEE Vehicular Technology Conference in Japan in 2000, and a few other international best paper awards.

 

He is Fellow of IEEE and IET and is also on the editorial board of several international journals. He has served as Senior Editor for IEEE Communications Letters since 2012 and also for IEEE Wireless Communications Letters since 2016. He had also previously served as Associate Editor for IEEE Signal Processing Letters from 2009 to 2012 and Editor for IEEE Transactions on Wireless Communications from 2005 to 2011. He was also Guest Editor for IEEE JSAC SI on virtual MIMO in 2013 and currently the Guest Editor for IEEE JSAC SI on physical layer security for 5G.