A cellular neural network and utility-based radio resource scheduler for multimedia CDMA communication systems

The paper proposes a cellular neural network and utility (CNNU)-based radio resource scheduler for multimedia CDMA communication systems supporting differentiated quality-of-service (QoS). Here, we define a relevant utility function for each connection, which is its radio resource function weighted by a QoS requirement deviation function and a fairness compensation function. We also propose cellular neural networks (CNN) to design the utility-based radio resource scheduler according to the Lyapunov method to solve the constrained optimization problem. The CNN is powerful for complicated optimization problems and has been proved that it can rapidly converge to a desired equilibrium; the utility-based scheduling algorithm can efficiently utilize the radio resource for system, keep the QoS requirements of connections guaranteed, and provide the weighted fairness for connections. Therefore, the CNNU-based scheduler, which determines a radio resource assignment vector for all connections by maximizing an overall system utility, can achieve high system throughput and keep the performance measures of all connections to meet their QoS requirements. Simulation results show that the CNNU-based scheduler attains the average system throughput greater than the EXP and the HOLPRO scheduling schemes by an amount of 23% and 33%, respectively, in the QoS guaranteed region.