Dynamic spectrum access with goe-location database: A utility-based distributed learning approach

This article investigates the problem of efficient distributed dynamic spectrum access in self-organizing networks with goe-location database, using a game theoretic approach. In particular, we propose a novel utility function, which aims to complete the data transmission of users and improve the satisfactory level of network by considering the efficient use of TV White-space Spectrum (TVWS) resource. Moreover, we formulate the problem as a distributed TVWS access game model. Compared with the traditional utility functions, the proposed function indicates a higher network satisfactory level than some others. To ensure that the network operates in a steady state, a learning-automata based uncoupled algorithm is used. Simulation results validate that the proposed utility function improves the network satisfactory level greatly.