Article
Computer Science, Information Systems
Mumraiz Khan Kasi, Sarah Abu Ghazalah, Raja Naeem Akram, Damien Sauveron
Summary: The study proposes a multi-agent reinforcement learning solution for optimal placement of mobile edge servers to minimize network latency and balance server loads. Sharing information among RL agents enhances overall network performance but also raises security risks. Further analysis is conducted to address potential security attacks and countermeasures.
Article
Computer Science, Information Systems
Bahareh Bahrami, Mohammad Reza Khayyambashi, Seyedali Mirjalili
Summary: MEC is a promising communication paradigm that utilizes edge servers near end users to enable IoT and 5G scenarios. These servers provide virtualized resources and host various MEC applications, allowing user equipment and IoT devices to offload tasks. Optimizing edge server placement can greatly enhance the performance of mobile applications.
CLUSTER COMPUTING-THE JOURNAL OF NETWORKS SOFTWARE TOOLS AND APPLICATIONS
(2023)
Article
Computer Science, Information Systems
Xingbing Zhao, Yu Zeng, Hongwei Ding, Bo Li, Zhijun Yang
Summary: In this study, the placement of edge servers in smart cities was investigated, and a multi-objective optimization algorithm was utilized to enhance system performance and efficiency. Experimental results demonstrate that MNSGA-II has a significant advantage in reducing system overhead compared to traditional methods.
PEER-TO-PEER NETWORKING AND APPLICATIONS
(2021)
Article
Computer Science, Information Systems
Yuanyi Chen, Dezhi Wang, Nailong Wu, Zhengzhe Xiang
Summary: The dynamic movement of users significantly affects the workload balance of edge servers. This study proposes a mobility-aware edge server placement method, including a fast heuristic algorithm in the offline phase and a deployment strategy adjustment mechanism based on cooperative game theory in the online phase.
COMPUTER COMMUNICATIONS
(2023)
Article
Computer Science, Information Systems
Xinglin Zhang, Zhenjiang Li, Chang Lai, Junna Zhang
Summary: This article proposes a complete process that combines edge server deployment and service placement, aiming to maximize the overall profit of edge servers by considering the current edge server placement structure and different service request rates and prices. The formulated problem is solved using a two-step method and extensive evaluations based on real-world data demonstrate the superiority of the proposed algorithm over baseline methods.
IEEE INTERNET OF THINGS JOURNAL
(2022)
Article
Computer Science, Information Systems
Jinfeng Dou, Fangzheng Yuan, Jiabao Cao, Xuejia Meng, Xiaoguang Ma, Zhongwen Guo
Summary: With the goal of reducing user request delay and balancing resources, this paper proposes a placement combination approach for heterogeneous services and servers in mobile edge computing (MEC). Two solution algorithms are introduced to optimize the combination placement, demonstrating significant effectiveness in reducing user request delay through extensive simulations.
JOURNAL OF GRID COMPUTING
(2023)
Article
Computer Science, Information Systems
Shahrukh Khan Kasi, Mumraiz Khan Kasi, Kamran Ali, Mohsin Raza, Hifza Afzal, Aboubaker Lasebae, Bushra Naeem, Saif Ul Islam, Joel J. P. C. Rodrigues
Summary: Rapid advancements in Industry 4.0, machine learning, and digital twins have imposed new constraints on latency, reliability, and processing in IIoT and mobile devices, necessitating the emergence of mobile-edge computing. The problem of selecting optimal locations for edge servers in a large network infrastructure to balance workload and minimize access delay is addressed using genetic algorithm and local search algorithms to efficiently find the best solution. The experimental results demonstrate the effectiveness of genetic algorithm in quickly searching through a large solution space to minimize the cost function in edge server placement.
IEEE INTERNET OF THINGS JOURNAL
(2021)
Article
Computer Science, Artificial Intelligence
Feiyan Guo, Bing Tang, Jiaming Zhang
Summary: The paper addresses the issue of edge server placement by proposing an optimization model and an improved heuristic algorithm to achieve multi-objective optimization, ensuring service quality and enhancing performance.
JOURNAL OF INTELLIGENT & FUZZY SYSTEMS
(2021)
Article
Computer Science, Information Systems
Xiaohan Jiang, Peng Hou, Hongbin Zhu, Bo Li, Zongshan Wang, Hongwei Ding
Summary: This paper studies the efficient and intelligent dynamic edge server placement problem considering time-varying network states and placement costs. Two deep reinforcement learning-based algorithms are proposed to achieve intelligent decision-making and performance improvement. Experimental results show that the proposed algorithms outperform comparison algorithms by 13.20% to 61.84% and 23.09% to 66.32%, respectively.
Article
Computer Science, Information Systems
Sheuli Chakraborty, Kaushik Mazumdar
Summary: This study proposes a scheme that dynamically selects edge cloud for offloading tasks and checks task dependencies, achieving good performance in sensor mobile edge computing.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2022)
Article
Computer Science, Information Systems
Xinglin Zhang, Jinyi Zhang, Chaoqun Peng, Xiumin Wang
Summary: Mobile edge computing (MEC) deploys computing and storage resources close to mobile devices, allowing resource demanding applications to run on mobile devices with low latency. This study investigates the multimodal optimization problem of edge server placement (MESP) and proposes a heuristic algorithm that combines particle swarm optimization and niching technology to minimize system response time.
ACM TRANSACTIONS ON SENSOR NETWORKS
(2023)
Article
Automation & Control Systems
Kun Cao, Liying Li, Yangguang Cui, Tongquan Wei, Shiyan Hu
Summary: This article focuses on deploying heterogeneous edge servers in mobile edge-cloud computing systems to optimize response time. By using a two-stage approach, consisting of offline planning and online dynamic adjustment, the system performance can be significantly improved and the fairness of base stations can be enhanced.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2021)
Article
Computer Science, Information Systems
Haojun Teng, Zhetao Li, Kun Cao, Saiqin Long, Song Guo, Anfeng Liu
Summary: In this paper, a novel task offloading architecture called Flex-MEC is proposed for efficient task allocation and scheduling (TAS) between MEC servers. By adding metadata before task data, the offloading process in Flex-MEC is redesigned to enable TAS planning without task data receiving. A multi-server multi-task allocation and scheduling (MMAS) problem is formulated to maximize the MEC system profit, and both a distributed scheme and a centralized scheme are proposed to solve the NP-complete MMAS problem with low complexity. Experimental results demonstrate that these two schemes achieve better performance than compared schemes.
IEEE TRANSACTIONS ON MOBILE COMPUTING
(2023)
Article
Computer Science, Information Systems
Peng Hou, Bo Li, Zongshan Wang, Hongwei Ding
Summary: The article proposes a C-V2X intelligent converged network based on MEC to meet the high requirements of IoVs. By modeling the hierarchical placement and configuration of ESs and proposing heuristic algorithms, the efficiency of the solution has been effectively improved.
Article
Computer Science, Information Systems
Yuanzhe Li, Ao Zhou, Xiao Ma, Shangguang Wang
Summary: This article studies how to properly place edge servers in a 5G network in order to guarantee access delay and maximize profit. It proposes a profit model that considers both access delay and energy consumption, and uses a particle swarm optimization-based algorithm to optimize profit. The algorithm introduces a weight value and a service-level agreement to balance the tradeoff between access delay and energy consumption. Experimental results show that the algorithm excels in achieving the highest profit.
IEEE INTERNET OF THINGS JOURNAL
(2022)
Article
Computer Science, Information Systems
Xiaoheng Deng, Jian Yin, Peiyuan Guan, Neal N. Xiong, Lan Zhang, Shahid Mumtaz
Summary: The development of Industrial Internet of Things (IIoT) and Industry 4.0 has transformed the traditional manufacturing industry. With the mobile-edge computing (MEC) system, computation-intensive tasks can be offloaded from resource-constrained IIoT devices to nearby MEC servers, resulting in lower delay and energy consumption for better Quality of Service (QoS).
IEEE INTERNET OF THINGS JOURNAL
(2023)
Article
Mathematics
Shenli Zhu, Xiaoheng Deng, Wendong Zhang, Congxu Zhu
Summary: In this paper, a new one-dimensional fractional chaotic map is proposed and an image encryption scheme based on parallel DNA coding is designed. The new chaotic system has a larger range of chaotic parameters and better chaotic characteristics compared to traditional one-dimensional chaotic systems, making it more suitable for information encryption applications. Additionally, a parallel DNA coding-based image encryption algorithm is proposed, overcoming the shortcomings of common DNA coding-based image encryption algorithms. Simulation experiments and security analysis results demonstrate the good encryption performance, less time overhead, and robustness to attacks, indicating the potential of the proposed image encryption scheme in secure communication applications.
Article
Computer Science, Information Systems
Yixiang Hu, Xiaoheng Deng, Congxu Zhu, Xuechen Chen, Laixin Chi
Summary: This article focuses on integrating wireless power transfer with mobile edge computing (MEC) in industrial Internet of Things (IIOT) systems. By using deep reinforcement learning, an online resource allocation and computation offloading approach is proposed to manage energy harvesting and select appropriate computing modes, thus maximizing the computing rate and task execution success rate for heterogeneous tasks.
ACM TRANSACTIONS ON MANAGEMENT INFORMATION SYSTEMS
(2023)
Article
Computer Science, Information Systems
Xiaoheng Deng, Jincai Zhu, Xinjun Pei, Lan Zhang, Zhen Ling, Kaiping Xue
Summary: This paper proposes a Flow Topology based Graph Convolutional Network (FT-GCN) approach for label-limited IoT network intrusion detection. By leveraging flow traffic patterns and flow topological structure, FT-GCN is deployed at edge servers in IoT networks to detect intrusions. It constructs an interval-constrained traffic graph (ICTG) considering the time correlation of traffic flows, and enhances key statistical features of traffic flows using a Node-Level Spatial (NLS) attention mechanism. Intrusion identification in IoT networks is achieved by learning the combined representation of statistical flow features and flow topological structure with the cost-effective Topology Adaptive Graph Convolutional Networks (TAGCN).
IEEE TRANSACTIONS ON NETWORK AND SERVICE MANAGEMENT
(2023)
Article
Automation & Control Systems
Xiaoheng Deng, Xinjun Pei, Shengwei Tian, Lan Zhang
Summary: The advent of 5G has brought new opportunities for Industrial Internet of Things (IoT) to leapfrog beyond current capabilities. However, the growing IoT has also attracted adversaries who develop new malware attacks on IoT applications. Deep-learning-based methods are expected to combat these sophisticated malwares, but they are not feasible for battery-powered end devices like Android smartphones. Edge computing enables near-real-time analysis of IoT data by shifting computation-intensive tasks to nearby edge servers. However, coordinating the task offloading among multiple users is challenging due to varying channel conditions and latency requirements. To address these challenges, we propose a hierarchical security framework for IoT malware detection that leverages the computation capacity and proximity benefits of edge computing. We also provide a delay-aware computational offloading strategy and construct a coordinated representation learning model, called Two-Stream Attention-Caps, to capture evolving malware attack patterns. Experimental results demonstrate superior detection performance compared to state-of-the-art systems on four benchmark datasets.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2023)
Article
Engineering, Civil
Leilei Wang, Xiaoheng Deng, Jinsong Gui, Xuechen Chen, Shaohua Wan
Summary: The integration of Mobile Edge Computing (MEC) and microservice architecture enables the implementation of sustainable Internet of Vehicles (IoV). With MEC, microservices can be dynamically placed on Edge Service Providers (ESPs) to reduce latency and resource consumption. However, the growth of IoV leads to high computation and resource overheads, calling for judicious service placement. In this paper, we propose a Microservice-oriented Service Placement (MOSP) mechanism to address these issues and achieve improved resource savings, latency reduction, and service speed.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2023)
Review
Computer Science, Hardware & Architecture
Leilei Wang, Xiaoheng Deng, Jinsong Gui, Ping Jiang, Feng Zeng, Shaohua Wan
Summary: Ground transportation systems are struggling with the increasing urbanization and vehicular numbers worldwide. The solution lies in pushing urban mobility into the sky with Urban Air Mobility (UAM), which offers faster, safer, cleaner, and more comprehensive transportation. UAM, with its autonomous maneuverability, convenience, mobility and communication capabilities, is an indispensable component of the Intelligent Transportation System (ITS). This work explores the definition, structure, feasibility, and key features of UAM, and provides guidance for its development, while also analyzing the challenges and non-technical issues that may hinder its research and deployment.
JOURNAL OF SYSTEMS ARCHITECTURE
(2023)
Article
Green & Sustainable Science & Technology
Husnain Mushtaq, Xiaoheng Deng, Mubashir Ali, Babur Hayat, Hafiz Husnain Raza Sherazi
Summary: Autonomous vehicles are crucial for improving urban mobility in a smarter and more connected urban environment. However, existing 3D LiDAR object detection systems have limitations, leading to subpar performance. In this study, a dynamic feature abstraction with self-attention (DFA-SAT) is proposed to address these limitations and enhance object detection performance.
Article
Chemistry, Multidisciplinary
Yuezhong Wu, Falong Xiao, Fumin Liu, Yuxuan Sun, Xiaoheng Deng, Lixin Lin, Congxu Zhu
Summary: This paper proposes an improved YOLOv5-based visual fault detection algorithm for substation equipment. By introducing a deformable convolution module and a simplified BiFPN structure, the algorithm enhances the recognition of small targets in complex substation scenarios, achieving promising experimental results.
APPLIED SCIENCES-BASEL
(2023)
Article
Computer Science, Information Systems
Xiaoheng Deng, Jiahao Zhao, Zhufang Kuang, Xuechen Chen, Qi Guo, Fengxiao Tang
Summary: This paper discusses a multi-UAV-enabled MEC system, which optimizes multiple factors to maximize the computation efficiency of the terminal system, and proposes an algorithm for computation efficiency maximization.
IEEE TRANSACTIONS ON EMERGING TOPICS IN COMPUTING
(2023)
Article
Mathematics
Shenli Zhu, Xiaoheng Deng, Wendong Zhang, Congxu Zhu
Summary: This paper proposes a new two-dimensional discrete hyperchaotic system and uses it to design a pseudo-random number generator and an efficient color image encryption algorithm. The system has very complex dynamic properties and can generate highly random chaotic sequences. The proposed PRNG can generate highly random bit sequences that meet all NIST testing standards. The color image encryption algorithm achieves pixel permutation and diffusion in parallel, improving both encryption speed and security level.
Article
Mathematics
Jiaqi Liu, Hucheng Xu, Xiaoheng Deng, Hui Liu, Deng Li
Summary: This paper proposes an incentive mechanism based on ensemble learning and prospect theory to address the problems with existing incentive mechanisms. It predicts the participants' duration and uses different incentive mechanisms for short-term and long-term participants to improve the coverage of crowdsensing tasks.
Article
Computer Science, Theory & Methods
Xiaoheng Deng, Haowen Tang, Xinjun Pei, Deng Li, Kaiping Xue
Summary: In this paper, a trust hybrid user-edge evaluation-based IoT malware detection system (MDHE) is proposed. The system decomposes a complex deep learning model into two parts deployed on edge servers and end devices. Trusted devices are selected using a trust evaluation mechanism for model training. A private feature generation method is developed to extract subgraph features and perturbed using differential privacy technology to protect user privacy. The perturbed features are reconstructed on an edge server, and a Capsule Network is used to identify malware.
IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY
(2023)
Article
Engineering, Civil
Ping Jiang, Xiaoheng Deng, Shaohua Wan, Huamei Qi, Shichao Zhang
Summary: This paper proposes a mutual learning framework, CE-MGN, to improve the robustness of object recognition in adverse weather conditions through continuous interaction between different tasks. Experimental results show that the framework performs stably under different weather conditions.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2023)
Article
Engineering, Civil
Weishang Wu, Xiaoheng Deng, Ping Jiang, Shaohua Wan, Yuanxiong Guo
Summary: This paper proposes an end-to-end multi-modal domain-enhanced framework called CrossFuser, which integrates both image and lidar modalities to generate a robust environmental representation and calculates corresponding waypoints and control commands using perception embedding.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2023)