子任務卸載方法及成本考量下資源配置之研究

以作者查詢圖書館館藏

、以作者查詢臺灣博碩士

、以作者查詢全國書目

、勘誤回報

、線上人數：28

、訪客IP：3.147.104.120

姓名

胡宗德(Tsung-Te Hu) 查詢紙本館藏

畢業系所

通訊工程學系

論文名稱

子任務卸載方法及成本考量下資源配置之研究
(Study for Resource Allocation of Subtask Offloading Method and Cost Consideration)

相關論文

★ 應用MSPP至DWDM都會光纖網路的設計	★ 光網路與WiMAX整合架構研究及其簡化雛型實驗
★ 以Linux系統為基礎之NAT效能優化研究及其實作	★ 光波長劃分多工網路之路徑保護機制研究
★ 標籤交換網路下具有服務品質路由安排之研究	★ 以訊務相關性為基礎的整合性服務可調整QoS排程器之研究
★ 以群體播送支援IPv6環境下移動式網路連結更新之研究	★ 無線區域網路資源動態分配之效能研究
★ 在微觀移動環境下有效資源保留之路徑管理研究	★ 無線網路交握程序之預先認證方法分析與比較
★ 無線區域網路虛擬允入控制之研究	★ IPv6環境下移動網路之連結更新程序及其效能之研究
★ 具有限數量波長轉換節點的分波多工網路之群播波長分配與容量計算研究	★ 階層化行動式IPv6移動錨點選擇機制研究
★ 具高能量移動節點之叢集式感測網路效能研究	★ 預先註冊之快速換手階層化行動式IPv6研究

檔案

[Endnote RIS 格式]

[Bibtex 格式]

[相關文章]

[文章引用]

[完整記錄]

[館藏目錄]

至系統瀏覽論文 (2025-1-1以後開放)

摘要(中)

科技水平在近幾十年快速的進步，行動裝置的問世大大改變人們的生活方式，在資訊爆炸的現代，行動裝置需處理各種大小不同的任務需求，當本地裝置無法負荷任務需求時，就會將任務透過網路傳送至運算處理能力更加強大的雲端伺服器(Cloud Server)處理，完成之後再回傳至本地裝置。雖然雲端伺服器的運算能力非常強大，但隨著物聯網以及智慧型裝置的普及，大量的任務都必須透過骨幹網路進行傳輸，由於裝置與雲端伺服器的距離較遠，導致網路壅塞以及產生過長的傳輸延遲，已無法滿足現今許多低延遲任務的需求，而行動邊緣運算(Mobile Edge Computing)的出現，用戶可以透過行動網路將任務卸載(Offload)至每個區域所屬的行動邊緣運算伺服器上處理之後再將其回傳，而行動邊緣運算伺服器因為較靠近用戶端，可以有效減少任務的傳輸時間，更加容易滿足用戶低延遲的需求。
本篇論文提出的Subtasks pre-allocation strategy integrated delay time and cost consideration將要卸載任務切割成數個獨立的子任務，之後計算每一個子任務的花費成本，運用0/1背包問題在任務最大容忍時間內，挑選出讓子任務分配至Local UE、D2D UE、MEC Server所得到花費成本最低的組合，並以平行處理的方式將子任務進行卸載。

摘要(英)

In recent years, technology has advanced very rapidly. Mobile devices have changed people′s lifestyles. Mobile devices need to handle various tasks of different sizes. When the local device cannot handle the task requirements, the task will be transmitted to the computing processing through the network. The more powerful cloud server (Cloud Server) processes, and then returns to the local device after completion. Although the computing power of the cloud server is very powerful, with the popularization of the Internet of Things and smart devices, a large number of tasks must be transmitted through the backbone network. Due to the long distance between the device and the cloud server, resulting in network congestion and excessive transmission delay, it has been unable to meet the needs of many low-latency tasks. Mobile Edge Computing (Mobile Edge Computing) allows users to offload tasks through the mobile network to the Mobile Edge Computing server in each region for processing and then send them back. Being close to the user terminal can effectively reduce the transmission time of tasks and make it easier to meet the user′s low-latency requirements.
The Subtasks pre-allocation strategy integrated delay time and cost consideration proposed in this paper divides the task to be offloaded into several independent subtasks, and its purpose is to process the subtasks in parallel. Then calculate the cost of each subtask, and use the 0/1 knapsack problem to select the combination with the lowest cost for assigning subtasks to Local UE, D2D UE, and MEC Server within the maximum tolerance time of the task.

關鍵字(中)

★ 行動邊緣運算
★ 0/1背包問題

關鍵字(英)

★ Mobile Edge Computing
★ 0/1 Knapsack Problem

論文目次

目錄
摘要 I
Abstract II
致謝 III
目錄 IV
圖目錄 VI
表目錄 VIII
2. 第一章緒論 1
2.1. 研究背景 1
2.2. 研究動機與目的 1
2.3. 章節概要 2
3. 第二章相關研究背景 3
3.1. LTE架構基本介紹 3
3.2. 邊緣運算(Edge Computing) 4
3.3. 0/1背包問題(0/1 Knapsack Problem) 7
3.4. 相關文獻 10
4. 第三章研究方法 12
4.1. 系統架構 12
4.2. 系統流程 21
4.2.1. 系統參數 21
4.2.2. 調整子任務花費成本 23
4.2.3. 綜合考慮延遲時間和成本之子任務預先分配策略運行流程 24
5. 第四章模擬結果與討論 27
5.1. 模擬環境 27
5.2. 模擬參數 28
5.3. 模擬結果比較 28
5.3.1. 不同最大容忍時間與D2D UE數量之任務接受率 29
5.3.2. 不同最大容忍時間與D2D UE數量之子任務接受比例 36
5.3.3. 不同最大容忍時間與D2D UE數量之任務平均花費成本比例 39
5.3.4. 不同最大容忍時間與D2D UE數量之任務平均等待時間 41
6. 第五章結論 45
7. 參考文獻 46

參考文獻

[1] [Online]. Available: https://www.796t.com/content/1546095068.html [Accessed 11 24 2022].
[2] [Online].Available:https://www.newton.com.tw/wiki/eNB%E5%8A%9F%E8%83%BD [Accessed 11 24 2022].
[3] [Online].Available: https://vocal.media/01/challenges-in-edge-computing[Accessed 11 24 2022].
[4] M. De Donno, K. Tange and N. Dragoni, "Foundations and Evolution of Modern Computing Paradigms: Cloud, IoT, Edge, and Fog," in IEEE Access, 2019
[5] K. Dolui and S. K. Datta, "Comparison of edge computing implementations: Fog computing, cloudlet and mobile edge computing," in Global Internet of Things Summit (GIoTS), 2017.
[6] L. Tang, Shibo He,"Multi-User Computation Offloading in Mobile Edge Computing: A Behavioral Perspective" in Computer Science IEEE Network,2018
[7] [Online]. Available: https://zh.wikipedia.org/zh-hk/%E8%83%8C%E5%8C%85%E9%97%AE%E9%A2%98[Accessed 11 26 2022].
[8] [Online]. Available : https://web.ntnu.edu.tw/~algo/KnapsackProblem.html#3 [Accessed 11 28 2022].
[9] Si Thu Thant Sin,"The Parallel Processing Approach to the Dynamic Programming Algorithm of Knapsack Problem" in 2021 IEEE Conference of Russian Young Researchers in Electrical and Electronic Engineering ,2021
[10] Chuan Sun; L. Hui; Xiuhua Li; Junhao We; Qingyu Xiongl; Xiaofei Wang; Victor C.M. Leun,"Task Offloading for End-Edge-Cloud Orchestrated
Computing in Mobile Networks",in 2020 IEEE Wireless Communications and Networking Conference (WCNC), 2020
[11] Saibal Ghosh, Dharma P Agrawal," A high performance hierarchical caching framework for mobile edge computing environments", in 2021 IEEE Wireless Communications and Networking Conference (WCNC),2021
[12] Fanfan Wu; Xiuhua Li; Hui Li; Qilin Fan; Linquan Zhu," Energy-Time Efficient Task Offloading for Mobile Edge Computing in Hot-Spot Scenarios", in ICC 2021 - IEEE International Conference on Communications,2021

指導教授

陳彥文(Yen-Wen Chen)

審核日期

2023-1-3

推文