User-centric Resource Configuration Schemes In Dense Small Cell Networks

Dense small cell network with wireless backhaul is an emerging way to provide high data rate and network capacity for 5G networks.Caching popular files at the base stations is viewed as a promising approach to reduce backhaul bandwidth consumption in dense small cell networks.To improve the quality of service(QoS),it needs to take full advantage of caching.In this thesis,we propose a spectrum resource configuration algorithm and a cacheaware algorithm of user-centric CoMP clustering for caching-based small cell networks.Firstly,we study the spectrum resource configuration scheme in dense small cell networks with caching.To divide the spectrum resources into backhaul and access resource,we formulate our problem as a multi-dimensional Markov chain model,which facilitates us to calculate the blocking probability as a function of cache capacity and spectrum resource.In order to minimize blocking probability,we propose a spectrum resource configuration algorithm with limited caching capacity,which can significantly reduce blocking probability and improve the QoS of system.Then,we propose a cache-aware algorithm of user-centric CoMP clustering to reduce the interference in small cell networks.Each user is associated with a user-centric CoMP cluster which is designed with the consideration of caching at each base station.Considering the limited backhaul bandwidth and maximum size of CoMP cluster,we perform the cacheaware algorithm of user-centric CoMP clustering in distributed way to design the CoMP cluster and precoder for each user in order to obtain the maximum weighted sum-rate.Moreover,an uplink pilot calculated by each user is applied to avoid the global channel estimation.The gains of CoMP transmission can be guaranteed by the proposed algorithm.Finally,we conduct numerous simulation for both the proposed algorithm.The simulation results show significant gains of both algorithms,comparing with the existing works.