Joint Sensing and Processing Resource Allocation in Vehicular Ad-Hoc Networks

The performance of smart vehicle (SV) applications like autonomous driving and in-vehicle augmented reality based traffic information system depends on the Field of View (FoV) and the timely processing of the SVs sensor data. Vehicular networking (VN) technology can enhance the performance of these applications by enabling a SV to access the sensing and processing capabilities of other neighbouring SVs. The processing and storage capacity of a SV is limited compared to cloud servers and the communication link between two SVs is unreliable due to their mobility and the nature of wireless channels. Hence, developing efficient processing and sensing schemes for SVs and VNs can help in optimizing the performance of SV applications. In this paper, the authors propose Contextual Bandits (CB), Markov decision process (MDP) and deep Q-network (DQN) based sensing and processing schemes for VNs. Simulation results show that the proposed schemes outperform the baseline schemes in a variety of scenarios.

• Record URL:
  • https://doi.org/10.1109/TIV.2021.3124208
• Availability:
  • Find a library where document is available. Order URL: http://worldcat.org/issn/23798858
• Supplemental Notes:
  • Copyright © 2023, IEEE.
• Authors:
  • Chattopadhyay, Rajarshi
  • Tham, Chen-Khong
• Publication Date: 2023-1

Language

• English

Media Info

• Media Type: Web
• Features: References;
• Pagination: pp 616-627
• Serial:
  • IEEE Transactions on Intelligent Vehicles
  • Volume: 8
  • Issue Number: 1
  • Publisher: Institute of Electrical and Electronics Engineers (IEEE)
  • ISSN: 2379-8858
  • Serial URL: http://ieeexplore.ieee.org/xpl/RecentIssue.jsp?punumber=7274857

Subject/Index Terms

• TRT Terms: Autonomous vehicles; Field of vision; Information processing; Level 5 driving automation; Mobile communication systems; Optimization; Sensors; Vehicular ad hoc networks
• Subject Areas: Data and Information Technology; Highways; Vehicles and Equipment;

Filing Info

• Accession Number: 01875453
• Record Type: Publication
• Files: TRIS
• Created Date: Mar 13 2023 10:23AM