Application of Bidirectional Long Short-Term Memory to Adaptive Streaming for Internet of Autonomous Vehicles

Area of Science:

• Computer Science
• Electrical Engineering
• Telecommunications

Background:

• Interconnected autonomous vehicle networks face congestion, particularly during peak traffic hours.
• Limited research addresses integrating advanced wireless technologies like 6G into connected vehicular networks.
• Autonomous vehicle passengers require robust streaming application performance, which is challenged by network congestion.

Purpose of the Study:

• To introduce a novel bandwidth management algorithm specifically designed for streaming applications in autonomous vehicles.
• To leverage 6G wireless technology for high-speed, broad-coverage vehicular networks.
• To address the challenge of providing seamless streaming services amidst network congestion.

Main Methods:

• Development of a tailored bandwidth management algorithm incorporating bandwidth relay and video clip pre-loading.
• Utilizing vehicle-to-vehicle (V2V) communication to assist in bandwidth allocation strategies.
• Simulation-based validation of the algorithm's effectiveness under various network conditions.

Main Results:

• The proposed algorithm successfully meets the bandwidth demands of diverse streaming applications for autonomous vehicle passengers.
• Demonstrated significant increase in bandwidth availability during peak user demand periods.
• Achieved an average improvement of 55% in accessible bandwidth for users.

Conclusions:

• The developed 6G-based bandwidth management algorithm is effective in enhancing streaming services for autonomous vehicles.
• The approach shows promise for efficiently managing bandwidth allocation complexities in connected autonomous vehicular networks.
• Validated the feasibility of advanced wireless technologies for improving in-vehicle entertainment and connectivity.