跳到主要內容

臺灣博碩士論文加值系統

(44.211.239.1) 您好!臺灣時間:2023/01/31 04:57
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果 :::

詳目顯示

我願授權國圖
: 
twitterline
研究生:張鍀淅
研究生(外文):De-Si Chang
論文名稱:在5G異質網路環境中物聯網的移動邊緣計算分流
論文名稱(外文):Mobile Edge Computation Offloading for Internet of Things in 5G Heterogeneous Networks
指導教授:高勝助高勝助引用關係
指導教授(外文):Shang-Juh Kao
口試委員:廖宜恩張阜民
口試委員(外文):I-En LiaoFu-Min Chang
口試日期:2020-07-01
學位類別:碩士
校院名稱:國立中興大學
系所名稱:資訊科學與工程學系所
學門:工程學門
學類:電資工程學類
論文種類:學術論文
論文出版年:2020
畢業學年度:108
語文別:英文
論文頁數:30
中文關鍵詞:移動邊緣計算物聯網5G異構網絡最小成本最大流量
外文關鍵詞:Mobile Edge Computing (MEC)Internet of Things (IoT)5G Heterogeneous NetworkMinimum Cost Maximum Flow (MCMF)
相關次數:
  • 被引用被引用:0
  • 點閱點閱:161
  • 評分評分:
  • 下載下載:0
  • 收藏至我的研究室書目清單書目收藏:0
由於物聯網 (IoT)迅速發展。大量物聯網設備出現在我們的生活中,例如智慧型手機、傳感器、相機、可穿戴設備。然而隨著技術的發展,一些計算密集型的應用程式也出現在物聯網中,例如交互式遊戲、擴增實境 (AR)、虛擬實境 (VR)、自然語言處理 (NLP)、人臉辨識。但是物聯網設備的資源對於這些應用程式是不夠的 (如CPU、內存和電池)。針對這問題,移動邊緣計算 (MEC)是一種很有前途的技術,它將計算任務從設備上傳到MEC伺服器進行運算。有了MEC技術的幫助下,我們可以提高設備的電池壽命以及提高用戶的體驗質量 (QoE)。因為每個設備所需的計算資源和延遲要求不同,所以如何為設備選擇合適的MEC伺服器是一個重要的課題。如果計算任務卸載到不合適的MEC伺服器上,所選的MEC伺服器的資源將會巨大的浪費。另外在5G異構網絡 (HetNets)中,包含一個傳統的大型基地台(MBS)覆蓋著幾個小型基地台 (SBSs),每個基地台藉由光纖連接一個MEC伺服器。因為覆蓋範圍的關係,因此任何設備都可以向MBS請求伺服,而SBS只能為訪問它的用戶進行伺服,意味著每個物聯網設備可以訪問多個MEC伺服器。之前的卸載機制並不適合5G HetNets環境中。
本論文提出了一種考慮MEC伺服器選擇和能量消耗的5G HetNets計算任務卸載機制,以最大限度地提高成功卸載任務數量和減少總能量開銷為目標。我們將提出的機制建立為多目標問題,並採用最小成本最大流量 (MCMF)方法來解決這個問題。我們將總能量開銷和成功卸載任務的數量分別視為成本和網路流的流量,其中總能量包含傳輸能耗和邊緣伺服器計算能耗。
為了驗證該機制的性能,我們使用Java程式語言實作了4000*4000平方公尺的實驗環境。實驗環境中包含1個MBS和14個SBSs。我們假設MBS-MEC伺服器的計算能力優於與SBS-MEC伺服器。在任務卸載次數和總能耗方面,將我們的方法與貪婪和隨機卸載進行比較。實驗結果表明,我們方法的平均任務所消耗的能量分別降低了16.83%和19.53%且任務卸載數量分別增加了1.38%和10%。
The evolution of the Internet of Things (IoT) has happened swiftly. Large of IoT devices appear around our lives, such as mobiles, sensors, cameras and wearable devices. With technological development, several computation-intensive applications appear in IoT networks, such as interactive gaming, augmented reality, virtual reality, face recognition and natural language processing. However, the resources of IoT devices, such as CPU, memory, and battery are insufficient for advanced applications. Regarding this issue, Mobile Edge Computing (MEC) is a promising technique by offloading computation tasks from IoT devices to MEC servers. With the help of MEC techniques, the battery lifetime of IoT device can be increased and the quality of experience (QoE) for computation-intensive applications can be enhanced. In addition, in 5G heterogeneous networks (HetNets), there is one traditional macro base station (MBS) overlaid with several small cell base stations (SBSs). Each base station connects to a MEC server by the optical fiber. All IoT devices in the macro-cell coverage could be served by MBS directly and SBSs only serve the users who access to them. It means that each IoT device could access more than one MEC server. Because the computing resources of MEC server and latency requirement of IoT device are varied, how to select an appropriate MEC server for each IoT device is an important topic. If the computation task offloads to inappropriate MEC server, the resources of selected MEC server could be enormous wasted.
This thesis proposes a computation task offloading mechanism for 5G HetNets by taking MEC server selection and energy consumption into account simultaneously. We aim to maximize the number of successful offloading tasks and minimize the overall energy overhead. To do that, we formulate the offload problem as two-objective problem, and solve the problem by transferring this problem to the minimum cost maximum flow (MCMF) problem. We treat the overall energy overhead and the number of successful offloading tasks as cost and flow respectively, where the overall energy overhead includes transmission energy consumption and edge server computation energy consumption.
To verify the performance of the proposed mechanism, we implement a 4000*4000 square meter experimental environment by using Java programming language. One MBS and fourteen SBSs are included in the experimental environment. We assume that the computing capacity of MEC server associated with MBS is better than the MEC server associated with SBS. We compare the proposed approach to greedy energy scheme and random offloading scheme in terms of the number of task offloading and overall energy consumption. The simulation results reveal that the proposed scheme can lower than 16.83% and 19.53% average of task energy consumption over greedy energy scheme and random offloading scheme, respectively. Moreover, the number of tasks offloading obtained using the proposed scheme is 1.38% and 10% higher than that obtained using the greedy energy scheme and random offloading scheme, respectively.
Contents
中文摘要 i
Abstract ii
Contents iv
List of Figures v
List of Tables vi
Chapter 1. Introduction 1
1.1 Research Motivation 1
1.2 Thesis Contributions and Structure 4
Chapter 2. Related Work 5
2.1 Related Studies 6
Chapter 3. System Model and Problem Formulation 9
3.1 System Architecture 9
3.2 Communication Model 12
3.3 Mobile Edge Execution Model 13
3.4 Problem Formulation 14
3.5 Minimum Cost Maximum Flow (MCMF) 17
Chapter 4. Simulation and Performance Evaluation 24
4.1 Experimental Setting 24
4.2 Performance Evaluation 26
Chapter 5. Conclusions 28
References 29
[1]BRIEF IL. Internet of things. 2019.
[2]Sheng J, Hu J, Teng X, Wang B, Pan X. Computation Offloading Strategy in Mobile Edge Computing. Information. 2019;10(6):191.
[3]Mao Y, You C, Zhang J, Huang K, Letaief KB. A survey on mobile edge computing: The communication perspective. IEEE Communications Surveys & Tutorials. 2017;19(4):2322-2358.
[4]Yang L, Zhang H, Li M, Guo J, Ji H. Mobile edge computing empowered energy efficient task offloading in 5G. IEEE Transactions on Vehicular Technology. 2018;67(7):6398-6409.
[5]Guo H, Liu J, Zhang J. Efficient computation offloading for multi-access edge computing in 5G HetNets. Paper presented at: 2018 IEEE International Conference on Communications (ICC), 2018.
[6]Li S, Tao Y, Qin X, Liu L, Zhang Z, Zhang P. Energy-aware mobile edge computation offloading for IoT over heterogenous networks. IEEE Access. 2019;7:13092-13105.
[7]Sabella D, Vaillant A, Kuure P, Rauschenbach U, Giust F. Mobile-edge computing architecture: The role of MEC in the Internet of Things. IEEE Consumer Electronics Magazine. 2016;5(4):84-91.
[8]Imran MA, Sambo YA, Abbasi QH. Enabling 5G Communication Systems to Support Vertical Industries. Wiley Online Library; 2019.
[9]Mach P, Becvar Z. Mobile edge computing: A survey on architecture and computation offloading. IEEE Communications Surveys & Tutorials. 2017;19(3):1628-1656.
[10]Chen Z, Cheng S. Computation Offloading Algorithms in Mobile Edge Computing System: A Survey. Paper presented at: International Conference of Pioneering Computer Scientists, Engineers and Educators, 2019.
[11]Zhang W, Wen Y, Guan K, Kilper D, Luo H, Wu DO. Energy-optimal mobile cloud computing under stochastic wireless channel. IEEE Transactions on Wireless Communications. 2013;12(9):4569-4581.
[12]Mao Y, Zhang J, Letaief KB. Joint task offloading scheduling and transmit power allocation for mobile-edge computing systems. Paper presented at: 2017 IEEE wireless communications and networking conference (WCNC), 2017.
[13]Guo J, Song Z, Cui Y, Liu Z, Ji Y. Energy-efficient resource allocation for multi-user mobile edge computing. Paper presented at: GLOBECOM 2017-2017 IEEE Global Communications Conference, 2017.
[14]Yang L, Cao J, Cheng H, Ji Y. Multi-user computation partitioning for latency sensitive mobile cloud applications. IEEE Transactions on Computers. 2014;64(8):2253-2266.
[15]Pham Q-V, Leanh T, Tran NH, Park BJ, Hong CS. Decentralized computation offloading and resource allocation for mobile-edge computing: A matching game approach. IEEE Access. 2018;6:75868-75885.
[16]Huynh LN, Pham Q-V, Nguyen QD, Pham X-Q, Nguyen V, Huh E-N. Energy-Efficient Computation Offloading with Multi-MEC Servers in 5G Two-Tier Heterogeneous Networks. Paper presented at: International Conference on Ubiquitous Information Management and Communication, 2019.
[17]Lemieux N, Zhao M. Small Cells, Big Impact: Designing Power Soutions for 5G Applications.http://www.ti.com/lit/wp/slyy166/slyy166.pdf?ts=1588842702480. Published 2019. Accessed 26-Jun, 2019.
[18]Zhang J, Guo H, Liu J. Energy-Aware Task Offloading for Ultra-Dense Edge Computing. Paper presented at: 2018 IEEE International Conference on Internet of Things (iThings) and IEEE Green Computing and Communications (GreenCom) and IEEE Cyber, Physical and Social Computing (CPSCom) and IEEE Smart Data (SmartData), 2018.
[19] X. Chen, L. Jiao, W. Li, and X. Fu, "Efficient multi-user computation offloading for mobile-edge cloud computing," IEEE/ACM Transactions on Networking, vol. 24, no. 5, pp. 2795-2808, 2015.
電子全文 電子全文(網際網路公開日期:20230728)
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top