Optimization of delay performance in multicast CPC scheduling

Recently, cognitive pilot channel (CPC) has been proposed to assist spectrum awareness and network selection in the heterogeneous network and dynamic spectrum access mechanism. The CPC channel can provide the necessary wireless environment information for the terminals and avoid long time- and energy-consumption for spectrum scanning. In previous researches, two different CPC information delivery modes have been proposed, i.e., broadcast CPC and on-demand CPC. In order to overcome the defects of these two approaches, Feng et al. proposed a multicast CPC method to optimize the delay performance and information scheduling as an evolution of the on-demand mode. However, in their mode, the waiting time of each mesh is fixed never mind the user distribution is high or not. In this paper, we analyze the relationship between waiting time and user distribution in different scenarios. And according to the difference among their user distribution, we propose setting an optimal waiting time for each mesh. While in actual systems, the mesh number usually is very large. Due to the computational complexity, it is difficult to get the optimal waiting time of each mesh. So we propose a clustering scheme as a suboptimal solution. This suboptimal solution especially fits for the case of centralized distribution scenario. Simulation results will be shown to demonstrate the performance of the optimal and suboptimal solutions.