Resource Allocation For Multicast Services Over Multi-cell Networks

With recent explosively growth in data traffic,mobile network operators seek way to more efficiently deliver that data.The radio access network,as the last hop,often constrains the network capacity,being the bottleneck because of the limited radio resource.How to improve the network spectral efficiency has been a critical problem for mobile network operators.Multicast is an effective approach for point-to-multipoint content delivery over wireless networks,especially when content requests have significant geographical and temporal similarities.Multicast enables multiple users subscribing a same video content to be served at the same time and share the same radio resources for downlink transmission,which provides an open and new business proposition for mobile network operators.Moreover,since the 3GPP standard specifies Evolved Multimedia Broadcast Multicast Service(eMBMS)for multicast over LTE networks,also known as LTE multicast,it is gaining increasing attention as a method to deliver large amount of multimedia data to massive users.Moreover,similar to unicast,multicast scheduling and resource allocation in OFDMA(Orthogonal Frequency Division Multiple Access)systems is also one of the most challenging issues currently being addressed.LTE multicast enables multiple users subscribing a same video content to form a multicast group and share the same radio resources for downlink transmission.Due to the intrinsic heterogeneity of user channel conditions,the transmission rate of a multicast service is usually constrained by the user with the worst channel condition in the allocated resources.Therefore,it is crucial to perform dynamic multicast user grouping or content splitting in order to make the best use of the scare radio resources.In this paper,we investigate the resource allocation problem in OFDMA-based multi-cell networks for multicast services.To eliminate the effect of the multicast rate constraint,the idea of multicast regrouping is adopted to further improve the system performance.Multi-cell networks,different from the traditional single-layer cellular networks,can improve the system performance based on multi-cell centralized scheduling.Two kinds of network architecture are studied in this paper,single frequency network and heterogeneous network.Simulation results demonstrate the effectiveness of the proposed resource allocation algorithms.