On Spectrum Trading for 5G Cognitive Spectrum Sharing Networks with Hybrid Access: A Game-Theoretic Approach

Abstract

Cognitive radio (CR) networks employing dynamic spectrum access have been proposed in the framework of enabling 5G wireless networks to enhance efficient use of spectrum resources. Communication requirements with regard to spectrum in 5G networks demand dynamic spectrum access giving rise to new types of spectrum markets and spectrum trading issues. Cognitive secondary users (SUs) may employ opportunistic spectrum access (OSA) to access vacant frequency bands (FBs) primarily allocated to primary users provided that they do not cause interference. Moreover, the SUs can employ exclusive spectrum access (ESA) in FBs leased at a certain price. In this work, different types of spectrum operators (SOs) are considered with regard to the type of spectrum access offered to SUs. A novel two-stage spectrum access and pricing game is proposed for two scenaria, namely: (1) a duopoly comprising an opportunistic SO and an exclusive SO and (2) a monopoly of a hybrid SO offering both ESA and OSA. In the first stage of the game, non coordinated SUs choose the access option that maximizes their utility. This symmetric N-SU spectrum access selection game has a unique mixed strategy equilibrium which depends: (1) on the prices set by the SOs in either the duopoly or the monopoly examined and (2) on the bandwidth available to the SUs under each scenario. In the second stage of the game, the bandwidth-pricing policy planned by the SOs is determined as the solution to an optimization problem maximizing the SOs revenues at the SUs equilibrium. Two algorithms are proposed to solve the optimization problems arising in each scenario examined. The numerical results obtained under the two spectrum market scenaria demonstrate how the proposed two-stage spectrum access and pricing scheme behaves for various network parameters.