OBS network blocking probability prediction using ensemble technique

In present time, artificial intelligence techniques are widely used in network field to improve the performance and security of optical burst switching networks. In our work we have used different predictor models and feature reduction algorithms to predict the blocking probability of upcoming traffic in a OBS network which will help us improve the performance of the network drastically. In this work, ensemble of predictors like Linear Support Vector Machine (SVM), Linear Regression (LR) and Random Forest (RF) are used to train and predict burst contention or burst blocking probability. From the result, we can conclude that this approach works better and gives accurate as well as faster result than other approaches. We observe the proposed model competently and predict the blocking probability with higher accuracy, along with lowering the number of burst loss. Thus, it will help future network designers to have an advanced idea about the performance of the network under specific framework.