A dynamic spectrum decision scheme for heterogeneous cognitive radio networks

Spectrum sensing and spectrum management are the main challenging functions that cognitive radio (CR) networks have to perform. In this paper, we focus specifically on the spectrum decision problem. This problem is worsened in the presence of users with different demands and spectrum channels with different properties in a heterogeneous network. For accurate and proper spectrum management, we propose a spectrum decision algorithm for spectrum brokers, which takes user type and spectrum channel properties into consideration. This approach increases both the number of users that can get proper spectrum bands and the throughput of the system. The algorithm also includes a "patience" option. Users that choose the "patience" option agree to wait for a predetermined amount of time until their connection is established. Meanwhile, a better spectrum may become available. This option increases the efficiency of spectrum usage too.