Abstract
Network traffic prediction is a fundamental prerequisite for dynamic resource provisioning in wireline and wireless networks, but is known to be challenging due to non-stationarity and due to its burstiness and self-similar nature. The prediction of network traffic at the user level is particularly challenging, because the traffic characteristics emerge from a complex interaction of user level and application protocol behavior. In this work we address the problem of predicting the network traffic at the user level over a short horizon, motivated by its applications in cellular scheduling. Motivated by recent works on robust adversarial learning, we treat the prediction problem for non-stationary traffic in an adversarial context, and propose a meta-learning scheme that consists of a set of predictors, each optimized to predict a particular kind of traffic, and of a master policy that is trained for choosing the best fit predictor dynamically based on recent prediction performance, using deep reinforcement learning. We evaluate the proposed meta-learning scheme on a variety of traffic traces consisting of video and non-video traffic. Our results show that it consistently outperforms state-of-the-art predictors, and can adapt to before unseen traffic without the need for retraining the individual predictors.
Highlights
A CCURATE network traffic prediction is becoming increasingly important for dynamic resource management in wireline and wireless networks
The inference time of the Deep Q-network (DQN) itself is determined by the network structure; for a DQN with m layers, each containing at most n neurons, the complexity is bounded by O(nm)
root mean square error (RMSE) RESULTS FOR DATA SET 7 WITH THE SEVEN PREDICTORS USED AS SUB-POLICIES, AN long short-term memory (LSTM) TRAINED ON MIXED DATA, AND EXP3.S
Summary
A CCURATE network traffic prediction is becoming increasingly important for dynamic resource management in wireline and wireless networks. For example, network traffic prediction can be used for reducing the power consumption in routers, for improving statistical multiplexing gains for quality of service provisioning, and for dynamic traffic routing in data centers [1]. Traffic prediction can be efficient in minimizing the power consumption of user equipment and of base stations through sleep scheduling, and through reduced complexity in traffic scheduling [2]–[5]. Color versions of one or more of the figures in this article are available online at http://ieeexplore.ieee.org. The time scales for prediction can vary from day ahead prediction down to the millisecond level, depending on the intended use
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
