Energy-Aware Multi-Server Mobile Edge Computing: A Deep Reinforcement Learning Approach

We investigate the problem of computation offloading in a mobile edge computing architecture, where multiple energy-constrained users compete to offload their computational tasks to multiple servers through a shared wireless medium. We propose a multi-agent deep reinforcement learning algorithm, where each server is equipped with an agent, observing the status of its associated users and selecting the best user for offloading at each step. We consider computation time (i.e., task completion time) and system lifetime as two key performance indicators, and we numerically demonstrate that our approach outperforms baseline algorithms in terms of the trade-off between computation time and system lifetime.

[1]  Zhao Chen,et al.  Decentralized computation offloading for multi-user mobile edge computing: a deep reinforcement learning approach , 2018, EURASIP Journal on Wireless Communications and Networking.

[2]  Adlen Ksentini,et al.  On Using Edge Computing for Computation Offloading in Mobile Network , 2017, GLOBECOM 2017 - 2017 IEEE Global Communications Conference.

[3]  Jie Zhang,et al.  Mobile-Edge Computation Offloading for Ultradense IoT Networks , 2018, IEEE Internet of Things Journal.

[4]  Yong Zhao,et al.  Communication-Constrained Mobile Edge Computing Systems for Wireless Virtual Reality: Scheduling and Tradeoff , 2018, IEEE Access.

[5]  Ness B. Shroff,et al.  Efficient Beam Alignment in Millimeter Wave Systems Using Contextual Bandits , 2017, IEEE INFOCOM 2018 - IEEE Conference on Computer Communications.

[6]  Navid Naderializadeh,et al.  Feedback-Based Interference Management in Ultra-Dense Networks via Parallel Dynamic Cell Selection and Link Scheduling , 2018, 2018 IEEE International Conference on Communications (ICC).

[7]  Frank H. P. Fitzek,et al.  Demonstration of VR / AR offloading to Mobile Edge Cloud for low latency 5G gaming application , 2018, 2018 15th IEEE Annual Consumer Communications & Networking Conference (CCNC).

[8]  Navid Naderializadeh,et al.  Ultra-Dense Networks in 5G: Interference Management via Non-Orthogonal Multiple Access and Treating Interference as Noise , 2017, 2017 IEEE 86th Vehicular Technology Conference (VTC-Fall).

[9]  Kaibin Huang,et al.  Multiuser Resource Allocation for Mobile-Edge Computation Offloading , 2016, 2016 IEEE Global Communications Conference (GLOBECOM).

[10]  Anja Klein,et al.  Efficient resource allocation in mobile-edge computation offloading: Completion time minimization , 2017, 2017 IEEE International Symposium on Information Theory (ISIT).

[11]  Alex Graves,et al.  Playing Atari with Deep Reinforcement Learning , 2013, ArXiv.

[12]  Lingyang Song,et al.  EdgeFlow: Open-Source Multi-layer Data Flow Processing in Edge Computing for 5G and Beyond , 2018, IEEE Network.

[13]  Zhi Chen,et al.  An Incentive-Aware Job Offloading Control Framework for Mobile Edge Computing , 2018, ArXiv.

[14]  Anja Klein,et al.  A Distributed Algorithm for Multi-Stage Computation Offloading , 2018, 2018 IEEE 7th International Conference on Cloud Networking (CloudNet).

[15]  Ness B. Shroff,et al.  Out-of-Band Millimeter Wave Beamforming and Communications to Achieve Low Latency and High Energy Efficiency in 5G Systems , 2018, IEEE Transactions on Communications.