Toward Cost-Effective Mobile Video Streaming through Environment-Aware Watching State Prediction

Area of Science:

• Computer Science
• Electrical Engineering
• Human-Computer Interaction

Background:

• Mobile video streaming is popular but strains device resources (energy, bandwidth).
• Existing solutions focus on data transmission, often overlooking environmental influences on user behavior.
• Optimizing mobile video streaming requires considering the interplay between user behavior, the physical environment, and device limitations.

Purpose of the Study:

• To propose a cost-effective mobile video streaming scheme that reduces power consumption and mobile data usage.
• To incorporate environment-aware watching states into mobile video streaming optimization.
• To address the gap in research regarding the physical environment's impact on user behavior during mobile video consumption.

Main Methods:

• Predicting the user's watching state using machine learning, considering both user behavior and the physical environment within a defined time window.
• Developing a downloading algorithm that adapts to the predicted watching state, user equipment (UE) running mode in the LTE system, and the VLC player.
• Conducting experiments in a real-world environment to evaluate the proposed scheme's effectiveness.

Main Results:

• The proposed scheme effectively reduces mobile data usage compared to benchmarks (14.4% lower than energy-aware methods on average).
• Significant reduction in power consumption was observed, approximately 19% during periods of screen interaction.
• The environment-aware approach demonstrated superior performance in optimizing resource utilization for mobile video streaming.

Conclusions:

• Integrating environment-aware watching states significantly enhances the cost-effectiveness of mobile video streaming.
• The proposed scheme offers a practical solution for reducing energy and bandwidth consumption on mobile devices.
• Future research can further explore advanced machine learning models and diverse environmental factors for even greater optimization.