User Scheduling and Task Offloading in Multi-Tier Computing 6G Vehicular Network

Zhang, Haijun; Feng, Lizhe; Liu, Xiangnan; Long, Keping; Karagiannidis, George K.

doi:10.1109/jsac.2022.3227097

Cited by 21 publications

(5 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Fortunately, deep reinforcement learning in artificial intelligence can solve such high-dimensional time-varying feature problems with limited and inaccurate information [27,28]. Deep reinforcement learning algorithms for task offloading management are used in some of the literature [29][30][31][32]. Based on this, the computation tasks of TaVs are offloaded to edge vehicles and cloud networks to acquire more computation resources [29].…”

Section: Related Workmentioning

confidence: 99%

“…Deep reinforcement learning algorithms for task offloading management are used in some of the literature [29][30][31][32]. Based on this, the computation tasks of TaVs are offloaded to edge vehicles and cloud networks to acquire more computation resources [29]. The problem of computation offloading and resource allocation for tasks offloading to VECS through V2I links is addressed in [30].…”

Section: Related Workmentioning

confidence: 99%

“…As seen in Table 1, this literature is based on the three execution modes, M 1 , M 2 , and M 3 , which all involve the local execution mode M 0 . Most of the research in the M 1 and M 2 modes adopts the partial offloading mode [9,10,[14][15][16]25,29], and most of the research in the M 3 mode adopts the 0-1 offloading mode [19,20,23,31,32]. Furthermore, we have studied the use of V2V and V2I links to extend the system's computing resources [33,34].…”

Section: Related Workmentioning

confidence: 99%

See 2 more Smart Citations

Multi-User Computation Offloading and Resource Allocation Algorithm in a Vehicular Edge Network

Liu,

Zheng,

Zhang

et al. 2024

Sensors

View full text Add to dashboard Cite

In Vehicular Edge Computing Network (VECN) scenarios, the mobility of vehicles causes the uncertainty of channel state information, which makes it difficult to guarantee the Quality of Service (QoS) in the process of computation offloading and the resource allocation of a Vehicular Edge Computing Server (VECS). A multi-user computation offloading and resource allocation optimization model and a computation offloading and resource allocation algorithm based on the Deep Deterministic Policy Gradient (DDPG) are proposed to address this problem. Firstly, the problem is modeled as a Mixed Integer Nonlinear Programming (MINLP) problem according to the optimization objective of minimizing the total system delay. Then, in response to the large state space and the coexistence of discrete and continuous variables in the action space, a reinforcement learning algorithm based on DDPG is proposed. Finally, the proposed method is used to solve the problem and compared with the other three benchmark schemes. Compared with the baseline algorithms, the proposed scheme can effectively select the task offloading mode and reasonably allocate VECS computing resources, ensure the QoS of task execution, and have a certain stability and scalability. Simulation results show that the total completion time of the proposed scheme can be reduced by 24–29% compared with the existing state-of-the-art techniques.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Multi-User Computation Offloading and Resource Allocation Algorithm in a Vehicular Edge Network

Liu,

Zheng,

Zhang

et al. 2024

Sensors

View full text Add to dashboard Cite

show abstract

“…Reduce the sample complexity by exploiting permutation properties Sample complexity [90] Beamforming for interference networks, power allocation and cooperative beamforming for multi-cell MU-MISO systems, Generalize to problem sizes Sample complexity, generalizability to problem sizes (numbers of users and BSs) and environment parameters (cell size, noise power and transmit power) Spatial-based GCN [91] Cooperative beamforming and RIS association for multi-RIS aided MU-MIMO system Improve scalability and generalizability by capturing the underlying topology Scalability [93] User association and power allocation for terahertz wireless networks Integrate wireless network topology, improve size generalizability Generalizability to the number of transmitters [94] Power control for interference networks…”

Section: Aided Miso Systemsmentioning

confidence: 99%

Meta-Learning for Wireless Communications: A Survey and a Comparison to GNNs

Zhao,

Wu,

et al. 2024

IEEE Open J. Commun. Soc.

View full text Add to dashboard Cite

Deep learning has been used for optimizing a multitude of wireless problems. Yet most existing works assume that training and test samples are drawn from the same distribution, which is not true in dynamic environments. As a result, a deep neural network (DNN) may require retraining with newly collected samples to adapt to a new scenario. The retraining complexity can be reduced by introducing inductive biases, which can either be learned automatically or be embedded in DNNs manually. For example, inductive biases can be learned from related tasks by meta-learning in the form of good initializations, reusable modules and feature extractors, or be introduced by designing structural DNNs such as graph neural networks (GNNs) with appropriate permutation equivariance (PE) properties. All previous works on meta-learning overlooked the prior-known PE properties, which widely exist in wireless policies and can be leveraged to improve sample efficiency. This article reviews meta-learning for wireless communications and compares it with GNNs trained for a single task. We first introduce meta-learning by comparing it with conventional learning, analyze its inductive biases, and review its applications for radio transmission. We then compare meta-learning with GNNs trained for a single task by taking beamforming problem as an example to show their pros and cons in solving the mismatch issues. Simulation results show that GNNs are sample efficient while meta-learning can reduce the time complexity for adaptation. Both the considered GNNs and meta-learning methods are inefficient for adapting to environments with time-varying problem scales.

show abstract

“…Computational offloading is a challenging problem, where the users frequently relocate, resulting in constant variations in communication links, channel quality, and signal strength [5]. This poses a major challenge in wireless networks as it becomes essential to optimally and dynamically adapt computational offloading and resource allocation decisions to the time-varying wireless channel conditions and resources available on the MEC servers in real-time [6].…”

Section: Introductionmentioning

confidence: 99%

Artificial Intelligence-Defined Wireless Networking for Computational Offloading and Resource Allocation in Edge Computing Networks

Shah,

Gregory,

Bouhafs

et al. 2024

IEEE Open J. Commun. Soc.

View full text Add to dashboard Cite

The advent of the Internet of Everything and new Ultra-Reliable Low-Latency Communication (URLLC) services has resulted in an exponential growth in data demands at the network's edge. To meet the stringent performance requirements of evolving 5G (and beyond) applications, deploying dedicated resources closer to mobile users is essential. Multi-Access Edge Computing (MEC) is a promising technology for bringing computational resources closer to users. However, the distributed and limited MEC resources must be effectively optimized to maximize the number of mobile users benefiting from low-latency MEC services at each time slot in highly congested, large-scale, and dynamic wireless network scenarios. In this research, we propose and evaluate a novel Artificial Intelligence-Defined Wireless Networking (AIDWN) approach that builds on conventional Software-Defined Networking (SDN), implementing a new AI-defined application plane for computational offloading and resource allocation in MEC-enabled wireless networks. The AIDWN approach implements a deep reinforcement learning framework and deep neural networks that dynamically adapt optimal computational offloading and wireless resource allocation decisions while considering the handover, mobility, and coordinated resource allocation challenges in highly dynamic and mobile multi-MEC server environments. Compared to recent state-of-theart proposals, the proposed AIDWN demonstrates a substantial performance improvement, utilizing more than 90% of MEC resources per time slot across all MEC servers. It also accommodates significantly more mobile users in highly congested wireless network scenarios. We identified various future research directions highlighting the potential of the AIDWN approach in simplifying the management of nextgeneration wireless networks.

show abstract

User Scheduling and Task Offloading in Multi-Tier Computing 6G Vehicular Network

Cited by 21 publications

References 39 publications

Multi-User Computation Offloading and Resource Allocation Algorithm in a Vehicular Edge Network

Multi-User Computation Offloading and Resource Allocation Algorithm in a Vehicular Edge Network

Meta-Learning for Wireless Communications: A Survey and a Comparison to GNNs

Artificial Intelligence-Defined Wireless Networking for Computational Offloading and Resource Allocation in Edge Computing Networks

Contact Info

Product

Resources

About