Feasibility of cognitive machine-to-machine communication using cellular bands

We evaluate the feasibility of cognitive machine-to-machine communication on cellular bands from engineering and business perspectives. We propose a hierarchical network structure for cognitive M2M communication, where cluster headers gather M2M traffic using cognitive radio and forward it to the cellular networks. This structure can resolve the congestion problem that arises in conventional M2M systems. We obtain the optimal network parameters that minimize congestion at the radio access network. In addition, we investigate the business value of cognitive M2M on cellular bands. Taking into account the network usage fee, service fee, and hardware production costs, we model the profit structure of M2M services and derive the condition under which the CR type of M2M communication is superior to conventional M2M communication. We find that the optimal network design parameters (i.e., the number of cluster headers and cluster size) for business value would be different from the parameters expected from an engineering point of view. We believe that cognitive M2M communication can be a good solution for dealing with excessive M2M traffic in cellular networks, in terms of technical feasibility and business opportunity.