Spectrum assignment allocation based on artificial swarm intelligence using a bioinspired hybrid metaheuristic algorithm

Abstract

The exponential growth of the new mobile and wireless services, inherent to the needs of a hyperconnected society, demands more and more the almost immediate and permanent access of different radio resources, which implies attending in a careful way, aspects such as the expected shortage of the electromagnetic spectrum and its considerable underutilization in current radiocommunications, thus motivating the investigation of emerging models for dynamic spectrum allocation. In this way, Dynamic Spectrum Access is presented as an efficient solution for the resilient reuse of wireless communication channels in a shared scheme of frequency bands. The main challenge of dynamic spectrum allocation is to be able to guarantee to all the users of a network, protection against the interference that could be generated during the simultaneous occupation of a communication channel. Therefore, it is considered pertinent to restrict access to an overloaded number of users, in order to achieve peaceful coexistence in a given coverage area, taking into account that the simultaneous use of a channel by one or more users will be possible as long as they do not exceed the interference threshold imposed by the system. Consequently, this work uses Artificial Intelligence based on a bio-inspired hybrid metaheuristic algorithm, called Optimization by Swarm of Socio-Cognitive Particles, in order to solve the problem of Dynamic Spectrum Access in a Heterogeneous Network, having as a multi-objective utility function the metrics associated with user assignment and data rate maximization, thus mitigating the harmful impact of interference and successfully assigning spectrum to an average of 80 % of the users considered in the analysis scenario.