Multi-Modal Big Data Analyzing Architecture for Industrial Internet of Things

Опубликована: Окт. 24, 2024

Язык: Английский

---

Federated Learning‐Based Mobile Traffic Prediction in Satellite‐Terrestrial Integrated Networks

Software Practice and Experience, Год журнала: 2024, Номер unknown

Опубликована: Ноя. 19, 2024

ABSTRACT Introduction With the development and integration of satellite terrestrial networks, mobile traffic prediction has become more important than before, which is basis for service provision resource scheduling when supporting various vertical applications. However, existing methods, especially deep learning‐based require massive data model training. Due to privacy concerns, are not easily shared among different parties, making it difficult obtain a precise model. Methods To mitigate leakage risk, federated learning framework proposed in this study satellite‐terrestrial integrated networks achieve tradeoff between accuracy. In framework, local models trained base stations on ground, global aggregated edge server space. Results A with an adaptive graph convolutional network (AGCN) long short‐term memory (LSTM) modules validated numerical experiments, achieves lowest error real‐world dataset compared other neural (GNN) variants setting. Conclusion Numerical experiments demonstrate effectiveness approach, outperforms GNN lower errors.

Язык: Английский

---

Guest Editorial of the Special Section on Multimodal Data-Driven Decision-Making for Next-Generation Consumer Electronics

IEEE Transactions on Consumer Electronics, Год журнала: 2024, Номер 70(3), С. 5979 - 5982

Опубликована: Авг. 1, 2024

Язык: Английский

---

Intercity round-trip multi-region demand prediction based on multi-task fusion recurrent graph attention network

Neural Computing and Applications, Год журнала: 2024, Номер unknown

Опубликована: Дек. 22, 2024

Язык: Английский

---