Abstract:
Mobile traffic forecasting with high accuracy is essential for proactive network service provisioning and efficient network resource allocation in smart cities due to the rapid development of mobile cellular technologies and the rising popularity of mobile and IoT devices.
Traditional traffic forecasting uses time series prediction methods, which fail to capture mobile traffic demand’s complex dynamics and spatial relations. We propose a new deep learning framework, graph attention spatial-temporal network (GASTN), for accurate citywide mobile traffic forecasting that can capture both local geographical dependency and distant inter-region relationship when considering spatial factor.
Our spatial relation graph and structural recurrent neural networks model global near-far spatial relationships and temporal dependencies in GASTN. Two GASTN attention mechanisms integrate different effects holistically.
We also propose a collaborative global-local learning strategy for GASTN training to take advantage of the global model and local models for individual regions to improve prediction performance. Our GASTN model outperforms state-of-the-art methods in large-scale real-world mobile traffic datasets. The collaborative global-local learning strategy improves GASTN prediction performance.
Note: Please discuss with our team before submitting this abstract to the college. This Abstract or Synopsis varies based on student project requirements.
Did you like this final year project?
To download this project Code with thesis report and project training... Click Here
