A Reinforcement learning-based for Multi-user Task Offloading and Resource Allocation in MEC
-
- Proceedings of the Korea Information Processing Society Conference
- /
- 2022.05a
- /
- pp.45-47
- /
- 2022
Mobile edge computing (MEC), which enables mobile terminals to offload computational tasks to a server located at the user's edge, is considered an effective way to reduce the heavy computational burden and achieve efficient computational offloading. In this paper, we study a multi-user MEC system in which multiple user devices (UEs) can offload computation to the MEC server via a wireless channel. To solve the resource allocation and task offloading problem, we take the total cost of latency and energy consumption of all UEs as our optimization objective. To minimize the total cost of the considered MEC system, we propose an DRL-based method to solve the resource allocation problem in wireless MEC. Specifically, we propose a Asynchronous Advantage Actor-Critic (A3C)-based scheme. Asynchronous Advantage Actor-Critic (A3C) is applied to this framework and compared with DQN, and Double Q-Learning simulation results show that this scheme significantly reduces the total cost compared to other resource allocation schemes
A cost analysis for hospitalized patients was performed based on the RBRVS in order to determine an appropriate nursing fee schedule. The study was conducted through three phases as follows: 1) Nursing activities provided for the inpatients currently in Korea were identified and classified using a taxonomy which was developed by our research team through the Delphi process. 2) The resource-based relative points for every nursing activity according to nursing time, mental effort and judgement, technical skill, physical effort and stress were determined through a survey of 300 clinical RNs working at 5 tertiary hospitals from May 25 to July 25. 1998. 3) The nursing cost of every nursing activity for hospitalized patients was estimated based on the RBRVS. As a result, 136 nursing activities were identified and classified by nursing processes and nursing domains. However, our classification system of nursing activities should continue to be refined, and all nursing practices should be standardized. The nursing activities were given resource-based relative points ranging from 100 to 400 points, then each nursing activity was assigned a value for the RBRVS, which was determined by the exponential function of 2resource-based relative point/100. Thus, a value of 2 was calculated for 100 points, 4 for 200 points, 8 for 300 points, and 16 for 400 points. Meanwhile, the unit cost of nursing was calculated as 170 Won. The nursing cost of 136 nursing activities was estimated using the RBRVS as shown in