A Novel Forwarding Policy under Cloud Radio Access Network with Mobile Edge Computing Architecture

Nowadays, dozens of low-latency required application are emerging, traditional mobile network architecture would not be able to support such applications anymore in the future. Cloud radio access network (C-RAN) combined with Multi-access/mobile edge computing (MEC) seems to be one of the most feasible new RAN architectures to fulfill the requirement. With the assistance of MEC, the computing resource could be allocated more efficiently. In this paper, firstly the advantage of generalized-processor-sharing model (GPS) compared with first-in-first-out (FIFO) and processor-sharing (PS) are discussed in order to figure out the practical queueing behavior in MEC system. Next, the relationship between theoretical traffic intensity factor and realistic system CPU utilization condition is correlated. Finally, based on the discussion, a two threshold forwarding policy (TTFP) algorithm is proposed to dynamically arrange the data traffic according to current system traffic states. The result of the simulation articulates that TTFP algorithm could efficiently fulfill the applications who requires low entire waiting time as possible in high intensity traffic condition.