Computation Offloading in Multi-Access Edge Computing Networks: A Multi-Task Learning Approach

@article{Yang2019ComputationOI,
  title={Computation Offloading in Multi-Access Edge Computing Networks: A Multi-Task Learning Approach},
  author={Bofu Yang and Xuelin Cao and Joshua Bassey and Xiangfang Li and Timothy S. Kroecker and Lijun Qian},
  journal={ICC 2019 - 2019 IEEE International Conference on Communications (ICC)},
  year={2019},
  pages={1-6}
}
  • Bofu Yang, Xuelin Cao, L. Qian
  • Published 20 May 2019
  • Computer Science
  • ICC 2019 - 2019 IEEE International Conference on Communications (ICC)
Multi-access edge computing (MEC) has already shown the potential in enabling mobile devices to bear the computation-intensive applications by offloading some tasks to a nearby access point (AP) integrated with a MEC server (MES). However, due to the varying network conditions and limited computation resources of the MES, the offloading decisions taken by a mobile device and the computational resources allocated by the MES may not be efficiently achieved with the lowest cost. In this paper, we… 

Figures and Tables from this paper

Learning-Based Computation Offloading for Edge Networks with Heterogeneous Resources

TLDR
A model-free offloading algorithm based on reinforcement learning is proposed, which can make offloading decisions based on learning experience and can decrease the system utility by up to 50% compared with two widely used algorithms.

Joint Computation Offloading and Resource Allocation for NOMA-Enabled Industrial Internet of Things

TLDR
Simulation results show that compared with orthogonal multiple access (OMA) based method, the proposed iterative algorithm outperforms in total completion delay and network transmission rate.

Joint User Association and Value-Aware Computation Offloading for MEC-Enabled Networks

TLDR
An optimization problem is formulated, which aims to improve the potential benefits from the operator's perspective under the constraints of users' quality of service (QoS), and a value-aware scheme to classify tasks so that the operator can make reasonable pricing for different performance requirements of tasks is presented.

Mobile-Edge-Computing-Based Hierarchical Machine Learning Tasks Distribution for IIoT

TLDR
In this article, a novel framework of mobile edge computing (MEC)-based hierarchical machine learning (ML) tasks distribution for the Industrial Internet of Things is proposed and an optimal offloading strategy selection algorithm is proposed.

Intelligent Task Dispatching and Scheduling Using a Deep Q-Network in a Cluster Edge Computing System

TLDR
The proposed intelligent task dispatches model is designed based on a deep Q-network (DQN) algorithm to update the task dispatching policy and has a better convergence performanc in terms of the average completion time of all offloaded tasks.

Fair Computation Efficiency Scheduling in NOMA-Aided Mobile Edge Computing

TLDR
A heuristic search algorithm is adopted to obtain the optimal policy, including uplink transmit power allocation policy and local computing resource allocation policy, and simulation results show the superiority of the proposed scheme by contrast with several baseline schemes.

A Computation Task Offloading Scheme based on Mobile-Cloud and Edge Computing for WBANs

TLDR
A Computation Task Offloading Scheme based on Differential Evolution algorithm, called CTOS-DE, which can provide a best computation task offloading decision in terms of the total cost and load balancing for WBANs is investigated.

Mobility-Aware QoS Promotion and Load Balancing in MEC-Based Vehicular Networks: A Deep Learning Approach

TLDR
This paper forms the joint offloading decision and resource allocation problem as a Mixed Integer NonLinear Programming (MINLP) problem, and proposes a novel framework to effectively address the problem, where PSO accelerates the training by providing high quality labeled data to DNN.

Identification of the Key Parameters for Computational Offloading in Multi-Access Edge Computing

TLDR
It is demonstrated that a certain threshold of link speed is required for shorter completion times by offload, and the MD CPU workload determines it, and MD energy usage can be reduced considerably by offloading for varying complexity applications provided a sufficiently link speed was available to the MEC network.

References

SHOWING 1-10 OF 24 REFERENCES

Computation offloading for mobile edge computing: A deep learning approach

  • Shuai YuXin WangR. Langar
  • Computer Science
    2017 IEEE 28th Annual International Symposium on Personal, Indoor, and Mobile Radio Communications (PIMRC)
  • 2017
TLDR
A dynamic offloading framework for mobile users is developed, considering the local overhead in the mobile terminal side, as well as the limited communication and computation resources in the network side, and the Deep Supervised Learning method is developed to minimize the computation and offloading overhead.

Efficient Computation Offloading for Multi-Access Edge Computing in 5G HetNets

TLDR
Numerical results corroborate that the collaborative computation offloading scheme for multiple MEC servers can not only reduce the overall computation overhead efficiently, but also achieve a Nash equilibrium in a finite number of steps.

Joint offloading and computing optimization in wireless powered mobile-edge computing systems

TLDR
A wireless powered multiuser MEC system, where a multi-antenna access point (AP) (integrated with an MEC server) broadcasts wireless power to charge multiple users and each user node relies on the harvested energy to execute latency-sensitive computation tasks, is considered.

Hybrid computation offloading in fog and cloud networks with non-orthogonal multiple access

TLDR
To reduce offloading transmission latency and release the constraint of the limitedRadio resource, non-orthogonal multiple access (NOMA), which enables the multiple users to transmit data to the same FN on the same radio resource, is introduced into fog and cloud networks.

Optimal radio resource allocation for mobile task offloading in cellular networks

TLDR
This work poses the optimal radio resource allocation problem for the mobile task offloading in cellular networks, and solves the problem, and extensive numerical results show the performance comparison of the optimalRadio resource allocation plan and two baseline plans.

Offloading in Mobile Cloudlet Systems with Intermittent Connectivity

TLDR
An optimal offloading algorithm for the mobile user in such an intermittently connected cloudlet system, considering the users' local load and availability of cloudlets is developed, and it is proved that the optimal policy of the MDP has a threshold structure.

Machine Learning-Based Runtime Scheduler for Mobile Offloading Framework

TLDR
This study considers 19 different machine learning algorithms and four workloads, with a dataset obtained through the deployment of an Android-based remote offloading framework prototype on actual mobile and cloud resources, and uses Instance-Based Learning to evaluate an online adaptive scheduler for mobile offloading.

On Multi-Access Edge Computing: A Survey of the Emerging 5G Network Edge Cloud Architecture and Orchestration

TLDR
This paper analyzes the MEC reference architecture and main deployment scenarios, which offer multi-tenancy support for application developers, content providers, and third parties, and elaborates further on open research challenges.

A Scalable MAC Framework for Internet of Things Assisted by Machine Learning

TLDR
A scalable MAC framework assisted by Machine Learning, called MML, which can dynamically adjust the CAP length based on the knowledge of the number of active devices and a stable throughput can be achieved, and the analytical and simulation results demonstrate the superiority of the proposed MML.

Energy-Efficient Resource Allocation in Fog Computing Supported IoT with Min-Max Fairness Guarantees

TLDR
This paper proposes to optimise the partial computation offloading in an OFDMA based FC IoT system, while ensuring fairness among IoT links with respect to their energy consumption, and proposes an iterative algorithm for resource allocation.