DRL‐Based Intelligent Resource Allocation for Diverse QoS in 5G and toward 6G Vehicular Networks: A Comprehensive Survey

Abstract

The vehicular network is taking great attention from both academia and industry to enable the intelligent transportation system (ITS), autonomous driving, and smart cities. The system provides extremely dynamic features due to the fast mobile characteristics. While the number of different applications in the vehicular network is growing fast, the quality of service (QoS) in the 5G vehicular network becomes diverse. One of the most stringent requirements in the vehicular network is a safety-critical real-time system. To guarantee low-latency and other diverse QoS requirements, wireless network resources should be effectively utilized and allocated among vehicles, such as computation power in cloud, fog, and edge servers; spectrum at roadside units (RSUs); and base stations (BSs). Historically, optimization problems have mostly been investigated to formulate resource allocation and are solved by mathematical computation methods. However, the optimization problems are usually nonconvex and hard to be solved. Recently, machine learning (ML) is a powerful technique to cope with the complexity in computation and has capability to cope with big data and data analysis in the heterogeneous vehicular network. In this paper, an overview of resource allocation in the 5G vehicular network is represented with the support of traditional optimization and advanced ML approaches, especially a deep reinforcement learning (DRL) method. In addition, a federated deep reinforcement learning- (FDRL-) based vehicular communication is proposed. The challenges, open issues, and future research directions for 5G and toward 6G vehicular networks, are discussed. A multiaccess edge computing assisted by network slicing and a distributed federated learning (FL) technique is analyzed. A FDRL-based UAV-assisted vehicular communication is discussed to point out the future research directions for the networks.

Twin Delayed DDPG based Dynamic Power Allocation for Mobility in IoRT

“…5 for allocating power to maximize the average sum rate for each UE have been more realistic. The simulations have been performed for evaluating the proposed TD3 algorithm with respect to two DRL-based algorithms: traditional DQN [23] and DDPG [29], as well as two traditional algorithms: WMMSE [22] and FP [21] which are the benchmarks in order to evaluate our proposed TD3 algorithm. In the simulation, we have considered 25 RRHs with 1 Km serving ranger per RRH and the number of total UE from 25 to 125 that are equally distributed among RRHs.…”

“…Hence, the maximum travelled distance within the time slots is 14 m. As a result, UE association is considered with fixed RRH. The system parameters for simulations except the mobility model follow as [23,29] for ensuring the fair comparison, presented in Table 3.…”

“…The internet of things (IoT), which emphasizes the goal and mission of a worldwide infrastructure connecting physical items known as things and uses internet protocol to allow them to optimization is done by convex optimization. However, most of the formulated problems, for example, dynamic PA, maximization of sum rate is strongly nonconvex as well as nondeterministic polynomial-time hardness (NP-hard) [23]. As a result, it isn't easy to get the optimized result [24].…”

Performance Art Video Action Management Oriented to 6G Wireless Transmission Technology

“…Since the power consumption of the MTC terminal is mainly in data transmission, this article considers the power consumption of the MTC terminal in data transmission (Nguyen et al, 2021;Wu, Li, & Jiang, 2022;Li L et al, 2020). Therefore, when the terminal transmits bit data and transmit power, the following power consumption formula is used to evaluate the power consumption:…”

Mobility-Aware Resource Allocation in IoRT Network for Post-Disaster Communications with Parameterized Reinforcement Learning

“…In [ 24 ], power allocation optimization is conducted by convex optimization. However, most of the formulated problems, for example, dynamic PA, maximization of the coverage area, traffic offloading, traffic load balancing with user association, maximization of sum rate, etc., are strongly nonconvex as well as nondeterministic polynomial-time hardness (NP-hard) [ 25 ]. In this research, we investigate optimizing the energy efficiency and throughput of UE as well as serving cell (SC) of the MDRU-aided two-tier HetNet scenario by ensuring the QoS of mobility-aware UEs where user association and power allocation for each UE have been considered without knowing the environmental priori knowledge.…”