Full Duplex HetNet FBMC relay networks Channel Estimation Interference Cancellation interference channel capacity MIMO CLI synchronization interference suppression in-band full-duplex system D-TDD 5G Multiple Antennas inter user interference OFDM mode selection WVAN channel estimation error UFMC 5G mobile communication timing misalignment ultra-dense small cell network GFDM reinforcement learning multi-access edge computing health care RSRP weighting - Computation offloading —Device-to-device (D2D) antenna arrays power uncertainty resource block management frame structure QAM Poisson arrival Short burst transmission mMTC User association Traffic Capacity OCBT Waveforms Time-division duplex self interference cancellation body area networks cellular radio quality of service amplify and forward communication Cognitive radio telecommunication traffic intercarrier interference Zigbee indoor positioning intersymbol interference spectrum sharing Cell-free Uplink SCMA system Number of training blocks Reliability Communication range Mode 3 resource allocation P-NOMA partial overlap Vehicular communication resource selection maximum likelihood method V2X user fairness cross-link interference Dynamic TDD LTE-TDD Metaheuristics QR Factorization Multi-user Receiver FS-NOMA non-orthogonal multiple access C-V2V OTDOA mMIMO User grouping packet delay estimated position updating Resource sharing Rat-dependent positioning NR positioning estimated position overlapping non-orthogonal multiple access (NOMA) overloading Power allocation Location-based distributed mode DQN spectrum partitioning massive connectivity and 5G networks. dynamic HetNet smart factory
Status : Presented 
Date : 2020-12 
Title : Deep Reinforcement Learning-based Task Offloading Decision in the Time Varying Channel 
Authors : Jinkyo Jeong, Ilmin Kim, and Daesik Hong 
Conference : ICEIC 
Abstract : This paper proposes a dynamic task offloading decision control scheme to minimize the total delay to execute computation task taking into account the time-varying channel. Specifically, we consider the practical task offloading process, where executing computation task is carried out over multiple channel coherence times. In order to make an accurate decision on the task offloading process performed over multiple channel coherence times, we utilize the model-free reinforcement learning, since environment dynamics of the system, channel transition probabilities, is challenging to estimate. We formulate a problem of minimizing the total delay of executing computation task based on a Markov decision process (MDP). In order to solve the MDP problem, we develop a model-free reinforcement learning algorithm. Simulation results show that our proposed scheme outperforms the conventional scheme. 

.

List of Articles
No.
Statussort Date
» [ICEIC] Jinkyo Jeong, Ilmin Kim, and Daesik Hong, "Deep Reinforcement Learning-based Task Offloading Decision in the Time Varying Channel" ICEIC 2021 Presented  2020-12