Dynamic bandwidth management based on traffic prediction using Deep Long Short Term Memory

Abstract

In dynamic bandwidth management based on traffic prediction, the traffic flows can be modeled as time-series data. State-of-the-art technique used in modeling this traffic flows is by using a linear model. In contrast, Recurrent Neural Network (RNN) has been the state-of-the-art technique in speech recognition, which data is also time-series. Therefore, we conjecture that the use of RNN can improve performance in dynamic bandwidth management based on traffic prediction. In this paper, we employ a variant of RNN called Deep Long Short Term Memory (DLSTM), which is common to be used in speech recognition. The result of this work shows that DLSTM is suitable for traffic prediction and is able to decrease packet loss ratio of a network system simulated using Network Simulator 3 (NS3).