Universiti Teknologi Malaysia Institutional Repository

Contextual multi-armed bandit based beam allocation in MMWAVE V2X communication under blockage.

Cassillas, Arturo Medina and Kose, Abdulkadir and Lee, Haeyoung and Foh, Chuan Heng and Chee, Yen Leow (2023) Contextual multi-armed bandit based beam allocation in MMWAVE V2X communication under blockage. In: 97th IEEE Vehicular Technology Conference, VTC 2023-Spring, 20 June 2023 - 23 June 2023, Florence, Italy.

Full text not available from this repository.

Official URL: http://dx.doi.org/10.1109/VTC2023-Spring57618.2023...

Abstract

Due to its low latency and high data rates support, mmWave communication has been an important player for vehicular communication. However, this carries some disadvantages such as lower transmission distances and inability to transmit through obstacles. This work presents a Contextual Multi-Armed Bandit Algorithm based beam selection to improve connection stability in next generation communications for vehicular networks. The algorithm, through machine learning (ML), learns about the mobility contexts of the vehicles (location and route) and helps the base station make decisions on which of its beam sectors will provide connection to a vehicle. In addition, the proposed algorithm also smartly extends, via relay vehicles, beam coverage to outage vehicles which are either in NLOS condition due to blockages or not served any available beam. Through a set of experiments on the city map, the effectiveness of the algorithm is demonstrated, and the best possible solution is presented.

Item Type:Conference or Workshop Item (Paper)
Uncontrolled Keywords:beam allocation; mmWave networks; V2X.
Subjects:T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK6570 Mobile Communication System
Divisions:Electrical Engineering
ID Code:107924
Deposited By: Muhamad Idham Sulong
Deposited On:16 Oct 2024 06:16
Last Modified:16 Oct 2024 06:16

Repository Staff Only: item control page