Reconfigurations Selection in Cognitive, Beyond 3G, Radio Infrastructures

B3G (beyond the 3^rd generation) wireless infrastructures can be efficiently realized by exploiting cognitive networking concepts. Cognitive, wireless access, infrastructures can dynamically configure their transceivers with the appropriate radio access technologies (RATs) and spectrum, in order to, reactively or proactively, adapt to the environment requirements and conditions. Reconfiguration decisions call for advanced management functionality. This paper provides such management functionality by addressing a pertinent problem, called "RAT and spectrum selection, QoS assignment and traffic distribution" (RSQT). Our work contributes in four main areas. First, we formally define and solve a fully distributed problem version, which is very important for the management of a particular reconfigurable element. Second, we propose robust learning and adaptation, strategies for estimating (discovering) the performance potentials of alternate reconfigurations. Third, we give a computationally efficient solution to the problem of exploiting the performance potentials of reconfigurations and thus selecting the best ones. Finally, we present results that expose the behaviour and efficiency of our schemes