In the massive machine type communication (mMTC) scenario of 5G

in order to solve the problem of insufficient spectrum resources and imperfect resource allocation

this paper proposes a cognitive SCMA system model based on the best virtual gateway. This model divides machine type communication devices (MTCDs) into independent clusters based on QoS requirements and geographic location. In each cluster

cognitive radio technology is used to perceive LTE (Long Time Evolution) users who may provide idle spectrum resources

which are defined as virtual gateways. Then the best virtual gateway is selected according to the rule of maximum satisfaction. The spectrum provided by the best virtual gateway is divided into multiple subcarrier groups

and design an optimal matching algorithm between MTCDs and subcarrier groups. MTCDs access the network through the SCMA mode. Simulation results show that the proposed scheme can effectively improve the throughput of the system.