Publication

Abstract : The tremendous increase in network requirements over multiple scenarios has urged the need to design an efficient framework to ensure Quality of service (QoS) in the given network Thus, the given paper proposes a novel QoS framework based on SDN (software defined networking) and gated recurrent unit-feed forward neural network (GRU-FFNN). SDN acts a centralized controller for effective network management which tracks network links while GRU-FFNN creates relationship between the packet nodes in the network. In order to analyze trade-off between latency, bandwidth, and packet loss ratio, an adaptive path selection algorithm is being used. The results show superior performance of the proposed work by achieving the highest accuracy (94.56%), lowest latency (25 ms) and lowest packet loss ratio (10.6%), thus surpassing the existing works.