A Dynamic Spectrum Allocation for the Military Cognitive Radio Network using Evolutionary Algorithm

A cognitive radio network introduces the idea of spectrum sharing among primary users and secondary users in the radio network. A dynamic spectrum access is the core technology of a cognitive radio network; in which unoccupied spectrum resources that are sensed by the cognitive radios can be efficiently utilized without disturbing the primary members. Unlike commercial network, each secondary radio node in military network can be allocated with unfair spectrum resources depending on its priority. This article exploits the cooperative dynamic spectrum allocation that every secondary members in the network can cooperatively allocate the available spectrum resources based on their needs and priorities. The evolutionary algorithm, specifically the particle swarm optimization, is explored and applied to the spectrum allocation such that a secondary member can learn and adapt to the spectrum access to optimize overall reward under the constraints of spectrum resource limitation, system limitation, and quality of services (QoS) of tactical data. The result from the experiment shows that the priority-based evolutionary algorithm yields better reward than the fair-based evolutionary algorithm and the random allocation algorithm.