We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Network Traffic Prediction Using Temporal Correlation-Based LSTM Models.
- Authors
Xuan Li; Qiaoyun Zhang; Chih-Yung Chang
- Abstract
Network traffic prediction plays a pivotal role in ensuring the stability of network operations and providing consistent user experiences for network operators. This paper introduces a innovative approach known as the Network Traffic Prediction Method (NTPM), which relies on temporal correlation properties. The NTPM, as proposed, involves two key steps: the construction of a traffic-time series dataset and the training of LSTM models. Constructing the traffic-time series dataset considers factors such as throughput, impact events, and various regional-level attributes to closely approximate realworld values. The Long Short-Term Memory (LSTM) models are trained using a loss function that quantifies prediction accuracy. The implementation of NTPM leads to improved accuracy in network traffic prediction, thereby facilitating network expansion planning and enhancing the overall user experience. Simulation results affirm that the proposed NTPM exhibits robust performance.
- Subjects
COMPUTER network traffic; USER experience
- Publication
International Journal of Design, Analysis & Tools for Integrated Circuits & Systems, 2023, Vol 12, Issue 2, p15
- ISSN
2071-2987
- Publication type
Article