Loading [a11y]/accessibility-menu.js
Node-Oriented Slice Reconfiguration Based on Spatial and Temporal Traffic Prediction in Metro Optical Networks | IEEE Journals & Magazine | IEEE Xplore
Scheduled Maintenance: On Monday, 27 January, the IEEE Xplore Author Profile management portal will undergo scheduled maintenance from 9:00-11:00 AM ET (1400-1600 UTC). During this time, access to the portal will be unavailable. We apologize for any inconvenience.

Node-Oriented Slice Reconfiguration Based on Spatial and Temporal Traffic Prediction in Metro Optical Networks


Abstract:

Given the spring-up of diverse new applications with different requirements in metro optical networks, network slicing provides a virtual end-to-end resource connection w...Show More

Abstract:

Given the spring-up of diverse new applications with different requirements in metro optical networks, network slicing provides a virtual end-to-end resource connection with customized service provision. To improve the quality-of-service (QoS) of slices with long-term operation in networks, it is beneficial to reconfigure the slice adaptively, referring to the future traffic state. Considering the busy-hour Internet traffic with daily human mobility, the tidal pattern of traffic flow occurs in metro optical networks, expressing both temporal and spatial features. To achieve high QoS of slices, this paper proposes a node-oriented slice reconfiguration (NoSR) scheme to reduce the penalty of slices, where a gradient-based priority strategy is designed to reduce the penalties of slices overall penalties in reconfiguration. Besides, given that a precise traffic prediction model is essential for efficient slice reconfiguration with future traffic state, this paper presents the model combining the graph convolutional network (GCN) and gated recurrent unit (GRU) to extract the traffic features in space and time dimensions. Simulation results show that the presented GCN-GRU traffic prediction model achieves a high forecasting accuracy, and the proposed NoSR scheme efficiently reduces the penalty of slices to guarantee a high QoS in metro optical networks.
Published in: IEEE Transactions on Network and Service Management ( Volume: 21, Issue: 6, December 2024)
Page(s): 6731 - 6743
Date of Publication: 02 September 2024

ISSN Information:

Funding Agency:


References

References is not available for this document.