Dynamic Resource Allocation for SDN and Edge Computing based 5G Network

The exponential increase in network traffic leads to considerable stress in 5G communication. The ultra-high reliability and low latency communication in 5G provides an insight on application of artificial intelligence with big volume of data. To meet the constraints of reduced delay, edge computing can be deployed. Edge computing is a promising solution for reduction in latency of computation-intensive tasks. And also, for reliable transmission, software defined network controller can be employed in the centralized infrastructure. So, this proposed work focuses on the forecasting of traffic flow in the network and dynamic allocation of the resource by the centralized infrastructure. The traffic prediction is modeled with Long Short-Term Memory neural network based on long-term time series traffic flow in the network. The effectiveness of the proposed model is analyzed with the real-time captured data which exhibit improved prediction accuracy.