Shuxian Sun
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Shuxian Sun has a total of 9 co-authors. Here are the top co-authors by number of shared publications:
- Cong Wang
Jilin University
2 shared publications - Fengye Hu
Jilin University
2 shared publications Chaoying Wang
College of Communication Engineering, Jilin University, Changchun, China
1 shared publication- Ke Xue
Changchun Institute of Optics, Fine Mechanics and Physics, Chinese Academy of Sciences, Changchun 130033, China
1 shared publication Maiou Liu
College of Communication Engineering, Jilin University, Changchun, China
1 shared publication- Wenxi Sun
College of Communication Engineering, Jilin University, Changchun, China
1 shared publication
Latest Publications
- 10 April 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 11 (13) , 23532-23543
Task Scheduling and Power Allocation in Multiuser Multiserver Vehicular Networks by NOMA and Deep Reinforcement Learning
In the pursuit of achieving optimal functionality for Internet of Vehicles (IoV), the integration of multiaccess edge computing (MEC) emerges as a solution, offering high bandwidth, low latency, robust security, and reliability services. In this article, we consider a multiuser multiserver vehicular network scenario, where the nonorthogonal multiple access (NOMA) technology in 5G is used to optimize spectrum resource utilization. We first formulate the problem using mixed integer nonlinear programming (MINLP) and propose a task scheduling scheme based on deep reinforcement learning (DRL) to handle high-dimensional state and action spaces and to approximate the optimal solution. We then proposed solutions to the NOMA clustering and power allocation problems in order to further reducing system latency in the uplink transmission stage. Simulation results underscore the efficacy of our proposed algorithm in systems with unevenly distributed computing resources, showcasing superior performance compared to alternative algorithms.
- 10 April 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 11 (13) , 23532-23543
Task Scheduling and Power Allocation in Multiuser Multiserver Vehicular Networks by NOMA and Deep Reinforcement Learning
In the pursuit of achieving optimal functionality for Internet of Vehicles (IoV), the integration of multiaccess edge computing (MEC) emerges as a solution, offering high bandwidth, low latency, robust security, and reliability services. In this article, we consider a multiuser multiserver vehicular network scenario, where the nonorthogonal multiple access (NOMA) technology in 5G is used to optimize spectrum resource utilization. We first formulate the problem using mixed integer nonlinear programming (MINLP) and propose a task scheduling scheme based on deep reinforcement learning (DRL) to handle high-dimensional state and action spaces and to approximate the optimal solution. We then proposed solutions to the NOMA clustering and power allocation problems in order to further reducing system latency in the uplink transmission stage. Simulation results underscore the efficacy of our proposed algorithm in systems with unevenly distributed computing resources, showcasing superior performance compared to alternative algorithms.
- 26 June 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Vehicular Technology
- Vol. 72 (12) , 16369-16381
Latency-Energy Joint Optimization for Task Offloading and Resource Allocation in MEC-Assisted Vehicular Networks
In this paper, we study the task offloading problem on mobile edge in vehicular networks. Specifically, we take computational resource constraints into consideration, and aim to simultaneously reduce latency and energy consumption. For this purpose, we establish an offloading model that consists of local edge computing resources, edge server resources of both macro and subsidiary base stations, as well as cloud computing server resources. Each task can be offloaded through one of five strategies, and is evaluated via a loss function determined by its latency and energy consumption. Based on this model, our goal is to solve a mixed-integer non-linear optimization problem (MINLP) whose objective function is the weighted sum of the task-specific loss functions. To address this optimization problem, we split it into two sub-problems, referred to as resource allocation and offloading strategy. We develop a method based on Block Coordinate Descent technique combining convex optimization and Gray Wolf algorithm (BCD-CONGW) that alternatively solves the two sub-problems, until convergence. The former sub-problem is convex and can be solved in polynomial time, whereas the latter is non-convex and hence NP-hard. For the latter, we relax discrete variables and employ Gray Wolf algorithm with elite strategy to approximate its optimal point. By numerical evaluations, we show that our method outperforms existent methods in terms of latency and energy consumption.
- 26 June 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Vehicular Technology
- Vol. 72 (12) , 16369-16381
Latency-Energy Joint Optimization for Task Offloading and Resource Allocation in MEC-Assisted Vehicular Networks
In this paper, we study the task offloading problem on mobile edge in vehicular networks. Specifically, we take computational resource constraints into consideration, and aim to simultaneously reduce latency and energy consumption. For this purpose, we establish an offloading model that consists of local edge computing resources, edge server resources of both macro and subsidiary base stations, as well as cloud computing server resources. Each task can be offloaded through one of five strategies, and is evaluated via a loss function determined by its latency and energy consumption. Based on this model, our goal is to solve a mixed-integer non-linear optimization problem (MINLP) whose objective function is the weighted sum of the task-specific loss functions. To address this optimization problem, we split it into two sub-problems, referred to as resource allocation and offloading strategy. We develop a method based on Block Coordinate Descent technique combining convex optimization and Gray Wolf algorithm (BCD-CONGW) that alternatively solves the two sub-problems, until convergence. The former sub-problem is convex and can be solved in polynomial time, whereas the latter is non-convex and hence NP-hard. For the latter, we relax discrete variables and employ Gray Wolf algorithm with elite strategy to approximate its optimal point. By numerical evaluations, we show that our method outperforms existent methods in terms of latency and energy consumption.