A5086187883- IoT and Edge/Fog Computing
- Age of Information Optimization
- Caching and Content Delivery
- Energy Harvesting in Wireless Networks
- IoT Networks and Protocols
- Energy Efficient Wireless Sensor Networks
- Advanced Wireless Communication Technologies
- Blockchain Technology Applications and Security
- Privacy-Preserving Technologies in Data
- Opportunistic and Delay-Tolerant Networks
- Advanced Neural Network Applications
Guangdong University of Technology
2019-2024
Most of existing Multi-access edge computing (MEC) studies consider the remote cloud server as a special server, opportunity edge-cloud collaboration has not been well exploited. We propose dependency-aware offloading scheme in MEC with cooperation under task dependency constraints. Each mobile device limited budget and to determine which sub-task should be computed locally or sent cloud. To address this issue, we divide problem into two application finishing time minimization sub-problems...
In vehicular edge computing, efficient strategies for model deployment and task offloading offer tremendous potential to reduce response time machine learning inference. However, existing works do not pay much attention that there are shared structures among different types of inference tasks. This limits the improvement in time. paper aims fill this gap by investigating a share-aware joint problem multi-task computing. We formulate with an objective minimize total all requests, under...
In vehicular networks, task scheduling at the microarchitecture-level and network-level offers tremendous potential to improve quality of computing services for deep neural network (DNN) inference. However, existing works only focus on either one two levels, which results in inefficient utilization resources. This paper aims fill this gap by formulating a two-level problem DNN inference tasks network, with an objective minimizing total weighted sum response time energy consumption all under...
Multi-access edge computing (MEC) is emerging to improve the quality of experience mobile devices including internet things sensors by offloading intensive tasks MEC servers. Existing MEC-enabled cooperative computation works focus on optimization total energy consumption but fail exploit multi-relay diversity and min-max fairness participated sensors. We explore a typical wireless sensor network with multi-source, multi-relay, one server, where relay nodes can provide both communication...
As the extension of cloud computing, multi-access edge computing (MEC) can better support applications to accomplish larger tasks. Existing works on energy optimization MEC systems fail utilize multi-relay diversity, which plays a vital role reduce power consumption mobile devices by leveraging different offloading modes. In this paper, we propose novel architecture with multi-source, multi-relay, and single server in an orthogonal frequency division multiplexing access (OFDMA) wireless...
Most existing works on edge service caching and request routing fail to consider the influence of loading time. Meanwhile, requests generated by end devices will change dynamically, which means that strategy should adapt accordingly. In this paper, we investigate cost-aware joint model with cooperative computing, considering both time dynamic user requests. A system throughput maximization problem is formulated, proved be NP-hard. Then, a randomized rounding-based online algorithm M/(M − 2...
Abstract Existing works on caching in multi-access edge computing focus service and request routing. However, loading cost execution time influenced by resource sharing have not been well exploited. To fill this gap, we investigate the joint optimization problem over deep neural network (DNN) model DNN routing with collaboration edge-enabled wireless sensor networks. A is formulated, objective of maximizing throughput, under constraints budget, accuracy latency etc. The proof NP-hardness for...
In mobile edge computing, the energy harvesting enables sustainable work of battery-powered devices. However, most existing works in computing with do not consider task dependency and dynamic harvesting. this paper, a model by considering is proposed. The problem completion time minimization formulated, which NP-hard. A strategy based on wireless power transfer proposed to solve allocating slots dynamically for Besides, we propose greedy algorithm giving priority offload sub-tasks place...