Coordinated Multi-Point User Task Scheduling for 5-G Cloud Radio Access Networks

Serving the cell edge user has always been a challenge. Traditional approaches entail increasing the transmit power at the base station, initiate a handover process or use base station cooperation. Increasing the transmit power increases the inter cell interference and only marginally increases the throughput of cell edge user and in a dense cellular network frequent handovers will increase signaling overhead. With the advent of 4G-LTE, relays and base station cooperation techniques like Coordinated Multipoint (CoMP) were explored, however tight synchronization and large signaling overheads made it impossible to arrive at an optimal solution. Futuristic cellular networks like 5G will exist in the telecomm cloud and will comprise of heterogeneous cellular architecture consisting of numerous micro and femto cells, making the cell edge user service all the more important. The telecomm cloud will consist of virtualized Base Band Units (VBBUs), which are basically Virtual Machines (VMs) in a datacenter. This thesis proposes a CoMP user task scheduling algorithm in a Cloud Radio Access Network (C-RAN) where the QoS guarantees of each cell edge user (CoMP region user) are taken into consideration while scheduling traffic flows to cloud resources. The eNodeB selectivity of each user’s traffic flow is modelled and its implementation in the cloud is simulated. This allows better utilization of resources by minimizing cost of transmission by transmitting from only a selected few eNodeBs as opposed to all eNodeBs serving the region. The system performance is measured in terms of number of failures, throughput and percentage violation.