- Opportunistic and Delay-Tolerant Networks
- Mobile Ad Hoc Networks
- Distributed systems and fault tolerance
- Caching and Content Delivery
- IoT and Edge/Fog Computing
- Cloud Computing and Resource Management
- Vehicular Ad Hoc Networks (VANETs)
- Privacy-Preserving Technologies in Data
- Blockchain Technology Applications and Security
- Cooperative Communication and Network Coding
- Metaheuristic Optimization Algorithms Research
- Advanced Multi-Objective Optimization Algorithms
- Internet Traffic Analysis and Secure E-voting
- Optimization and Search Problems
- Distributed and Parallel Computing Systems
- Mobile Agent-Based Network Management
- Mobile Crowdsensing and Crowdsourcing
- Advanced Data Storage Technologies
- Cryptography and Data Security
- Advanced Neural Network Applications
- Age of Information Optimization
- Software System Performance and Reliability
- Traffic Prediction and Management Techniques
- Wireless Networks and Protocols
- Stochastic Gradient Optimization Techniques
Sun Yat-sen University
2015-2024
Ministry of Education of the People's Republic of China
2012-2024
Gansu Provincial Maternal and Child Health Hospital
2024
Jinan University
2022
Southern University of Science and Technology
2022
University of Victoria
2021
Key Laboratory of Guangdong Province
2016-2020
China Guangzhou Analysis and Testing Center
2019
SYSU-CMU International Joint Research Institute
2014
South China University of Technology
2012
Particle swarm optimization (PSO) is predominately used to find solutions for continuous problems. As the operators of PSO are originally designed in an <i xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">n</i> -dimensional space, advancement using a discrete space at slow pace. In this paper, novel set-based (S-PSO) method some combinatorial problems (COPs) presented. The proposed S-PSO features following characteristics. First, it based on...
In this article we propose a novel D2D Crowd framework for 5G mobile edge computing, where massive crowd of devices at the network leverage network-assisted collaboration computation and communication resource sharing. A key objective is to achieve energy-efficient collaborative task executions users. Specifically, first introduce system model in detail, then formulate assignment problem by taking into account necessary constraints. We next graph-matching-based optimal policy, further...
Traffic control at intersections is a key issue and hot research topic in intelligent transportation systems. Existing approaches, including traffic light scheduling trajectory maneuver, are either inaccurate inflexible or complicated costly. More importantly, due to the dynamics of traffic, it really difficult obtain optimal solution real-time way. Inspired by emergence vehicular ad hoc network, we propose novel approach intersections. Via vehicle infrastructure communications, vehicles can...
With the fast development of Internet Vehicles (IoV), various types computation-intensive vehicular applications pose significant challenges to resource-constrained vehicles. The emerging Vehicular Edge Computing (VEC) and Intelligence (EI) can alleviate this situation by offloading computation tasks vehicles roadside edge servers. However, with many contending for communication resources at same time, how quickly efficiently make an optimal decision individual represents a fundamental...
Vehicular network has been recently used to achieve high efficient and flexible traffic scheduling at intersection roads for smart transportation systems. Different from existing works, where signal is schedule waiting vehicles each lane, we propose divide in the same lane into small groups vehicle via wireless communication rather than lights. Such direct of can reduce time improve fairness, especially when volume different lanes imbalanced. The key challenge such a design lies determining...
Fog computing is envisioned as a promising approach for supporting emerging computation-intensive applications on capacity and battery constrained mobile Internet of Things (IoT) devices. Technically speaking, massive crowd devices in close proximity can be harvested collaborate computation communication resource sharing. Hence fog enables significant potentials low-latency energy-efficient task execution. However, without an efficient incentive mechanism to stimulate sharing among devices,...
The 5G heterogeneous networks (HetNets) are capable of providing real-time computing services for autonomous vehicles (AVs) by deploying edge devices (ECDs) at macro cell base stations (MCBSs) and small (SCBSs). With the imbalanced distribution fast moving AVs contending intensely services, how to efficiently exploit cooperations among participants in HetNets improve service performance is therefore challenging. In this paper, we develop a game theoretic scheme collaborative vehicular task...
In this article we propose a novel paradigm of socially-motivated cooperative mobile edge computing, where the social tie structure among and wearable device users is leveraged for achieving effective trustworthy cooperation collaborative computation task executions. We envision that combination local networked resource sharing empowers devices with multiple flexible execution approaches, including execution, D2D offloaded direct cloud D2D-assisted execution. Specifically, system model...
This paper comes up with a SDN based On-Demand Routing Protocol, SVAO, which separates data forwarding layer and network control layer, as in SDN, to enhance the transmission efficiency within VANETs. The Roadside Service Unit plays role of Local Controller is charge selecting vehicles forward packet road segment. All state road. Correspondingly, two-level design used. Global Level distributed adopts ranked query scheme collect vehicle information determine segments along message should be...
Job scheduling in cluster is often considered as a difficult online decision-making problem, and its solution depends largely on the understanding of workload environment. People usually first propose simple heuristic algorithm, then perform repeated tedious manual tests adjustments based characteristics to gradually improve algorithm. In this work, focusing multi-cluster environments, load balancing efficient scheduling, we present RLSK, deep reinforcement learning job scheduler for...
Temporary fork is a fundamental phenomenon in many blockchains with proof of work, and the analysis temporary has recently drawn great attention. Different from existing efforts that focus on blockchain system factors such as block size, network propagation delay or generation speed, this paper we explore new key dimension computing power miners' perspective. Specifically, first propose detailed mathematical model to characterize impact competition mining pools fork. We also derive...
Workload prediction plays a crucial role in resource management of large scale cloud datacenters. Although quite number methods/algorithms have been proposed, long-term changes not explicitly identified and considered. Due to shifty user demands, workload re-locations, or other reasons, the ”resource usage pattern” workload, which is usually stable short-term view, may change dynamically range. Such dynamic cause significant accuracy degradation for algorithms. How handle such an open...
Designing protocols for solving the consensus problem faces new challenges in mobile computing environments. Among others, how we can achieve message efficiency saving resource consumption has been focus of research. In this paper, present HC protocol, a efficient protocol MANETs. We consider widely used system model where hosts fail by crashes and is equipped with Chandra-Toueg's unreliable failure detectors. Unlike existing protocols, uses two-layer hierarchy based on clusters to...
Automatic driving services have large volume, location-aware, and time-changing contents, which are suitable to be cached by the edge. However, traffic on edge will extremely high especially in area with vehicle density, if vehicles directly access contents from as they demand. A hybrid data dissemination model both vehicle-to-vehicle (V2V) vehicle-to-infrastructure (V2I) disseminations has been proposed reduce edge, (infrastructure) selectively injects leverages network disseminate data. In...
Byzantine Fault Tolerant (BFT) state machine replication protocols are used to achieve agreement among replicated servers with arbitrary faults. Most existing BFT perform well in fault-free cases, but usually suffer from serious performance degradation when faults occur. In this paper, we present DBFT, a protocol that realizes graceful faulty cases. The major novelty of DBFT lies the double-response mechanism, which lets replica nodes deterministically respond clients twice: one is after...
Federated learning (FL) enables a large number of edge devices to learn shared model without data sharing collaboratively. However, the imbalanced distribution among users poses challenges convergence performance FL. Group-based FL is novel framework improve performance, which appropriately groups and allows localized aggregations within group before global aggregation. Nevertheless, most existing methods are K-means-based approaches that need explicitly specify groups, may severely reduce...
Wireless ad hoc network is a promising networking technology to provide users with Internet access anywhere anytime. To cope resource constraints of wireless networks, data caching widely used efficiently reduce cost. In this paper, we propose an efficient algorithm which makes use the overhearing property communication improve performance. Due broadcast nature links, packet can be overheard by node within transmission range transmitter, even if not intended target. Our proposed explores...
Mobile edge cloud has been increasingly concerned by researchers due to its closer distance mobile users than the traditional on Internet. Offloading computations from devices nearby is an effective technique accelerate applications and/or save energy devices. However, usually limited computation resources and constrained access bandwidth shared multiple in proximity. Thus, allocation of among significant overall application performance. In this paper, we study network aware multi-user...
Abstract Task scheduling is a complex problem in cloud computing, and attracts many researchers’ interests. Recently, deep reinforcement learning (DRL)-based methods have been proposed to learn the policy through interacting with environment. However, most DRL focus on specific environment, which may lead weak adaptability new environments because they low sample efficiency require full retraining updated policies for environments. To overcome weakness reduce time consumption of adapting we...
The co-existence and integration of different wireless networks will significantly advance the provision ubiquitous high-performance Internet services to mobile users. In this paper, we present HAWK, a real-world implementation high performance heterogeneous for access. We have built fully featured testbed comprised networks, including mesh network (WMN), LAN (WLAN) 3G network, dual mode clients that can access any these networks. also developed protocols algorithms various system functions,...