
Research Article
Bandwidth Allocation for eMBB and mMTC Slices Based on AI-Aided Traffic Prediction
@INPROCEEDINGS{10.1007/978-3-031-37139-4_11, author={Xiaoli Zhang and Kai Liang and Jen-Jee Chen and Jiaxin Liu}, title={Bandwidth Allocation for eMBB and mMTC Slices Based on AI-Aided Traffic Prediction}, proceedings={IoT as a Service. 8th EAI International Conference, IoTaaS 2022, Virtual Event, November 17-18, 2022, Proceedings}, proceedings_a={IOTAAS}, year={2023}, month={7}, keywords={Network slicing Traffic prediction ConvLSTM GRU mMTC eMBB}, doi={10.1007/978-3-031-37139-4_11} }
- Xiaoli Zhang
Kai Liang
Jen-Jee Chen
Jiaxin Liu
Year: 2023
Bandwidth Allocation for eMBB and mMTC Slices Based on AI-Aided Traffic Prediction
IOTAAS
Springer
DOI: 10.1007/978-3-031-37139-4_11
Abstract
Network slicing has emerged as a key enabler to address the diverse requirements of the fast-growing Internet of Things (IoT), such as enhanced Mobile Broad Band (eMBB) and massive Machine Type Communication (mMTC). However, the dynamic nature of service requirements propels challenges for resource allocation of network slicing. This paper proposes a bandwidth allocation method serving eMBB and mMTC slices with the aid of deep learning-based traffic prediction. The problem is formulated as maximizing the number of users served by the mMTC slice while satisfying the rate requirement of users served by the eMBB slice. We adopt traffic prediction by deep learning methods (i.e., ConvLSTM and GRU) to facilitate dynamic allocation of the bandwidth resource, which can be obtained by solving the integer programming problem. Numerical results show that the proposed algorithm outperforms the traditional method in terms of the average number of served mMTC users.