Examining spatiotemporal crowdsensing and caching for population-dynamic OTT content delivery

Area of Science:

• Computer Science
• Urban Planning
• Network Engineering

Background:

• Rapid urbanization and digitization increase urban wireless network traffic.
• Effective data traffic management is essential for dynamic urban ecosystems.
• Existing solutions struggle to cope with escalating demands in metropolitan areas.

Purpose of the Study:

• To propose a novel spatiotemporal crowdsensing and caching (SCAC) framework.
• To address surging demands and alleviate congestion in urban wireless networks.
• To provide a scalable and practical AI-based solution for metropolitan areas.

Main Methods:

• Formulating an offloading policy based on user mobility and content preferences.
• Utilizing an AI-based method at the cell level for traffic management.
• Implementing a deployment strategy starting from transportation hubs and expanding outwards.

Main Results:

• Demonstrated effectiveness in reflecting real-world urban dynamics.
• Achieved significant reductions in peak-hour traffic.
• Showcased robust performance across diverse urban settings.

Conclusions:

• The SCAC framework offers a comprehensive strategy for urban wireless traffic management.
• The study enhances understanding of spatiotemporal technology's potential in cities.
• Provides valuable insights for network operators, policymakers, and urban planners.