A Scheme for Learning User Preferences: Enabling Personalisation in Cognitive Wireless Systems

The continuous evolution of wireless systems has resulted in a number of new and powerful wireless networking standards. The concept of Beyond the Third Generation Systems (B3G) emerged in an attempt to exploit the variety of the available access standards to the benefit of end-users, operators and manufacturers. In this context, a key topic in the research area of B3G/4G networks is related to mechanisms and strategies to efficiently realize the complementary use of the diverse Radio Access Technologies (RATs), through their convergence into one composite radio environment. One of the most important features of these evolving systems is the availability of multiple access technologies, which will allow users to enjoy wireless services at any time, at any place. Evidently in order to truly enhance the experience of all users, even technology agnostic ones, functionality is required, on both the network and the user- device side, for providing the "always best connection" in a transparent manner. The focus of this paper is more on the end- user side. The target is to realize management functionality that takes into account user requirements, environment characteristics, configuration policies and experience established so as to dynamically configure the user terminal in a seamless and transparent manner, through machine learning mechanisms. This can be achieved with the help of Bayesian Networks, a technique used for encoding and learning probabilistic relationships.