Time Series Forecasting of Packet Loss Rate Using Artificial Neural Network Based on Particle Swarm Optimization
Manish P Ganvir · 2015
Abstract — Packet loss severely degrades the quality of service of multimedia communication in a Wi-Fi network. In this paper, a time-series prediction model for the packet loss rate (PLR) is developed. The reason for prediction of PLR is very much useful in congestion control mechanisms. An accurate prediction method would therefore be very helpful. An artificial neural network is used as a prediction model and it is trained with Particle swarm Optimization (PSO) as a training algorithm in order to get accurate prediction of packet loss rate.