臺灣博碩士論文加值系統

隨著微機電、無線傳輸與嵌入式處理技術的進步，讓微小的電子元件可以嵌入感測器、中央處理器及通訊元件等多樣化的功能。而這一個個的電子元件稱作感測節點，是構成無線感測網路的基本元件。無線感測網路的應用廣泛。因為它的前景十分被看好，發展至今已投入了相當的人力在相關的研究上。然而，無線感測網路的應用還是受到了一些限制，例如，有限的記憶體空間無法儲存大量的資料和應用程式，以電池為能源的感測節點無法提供源源不絕的能源。另外，當這些感測節點設置完成之後，如果需要一一收回、更新系統以符合目前環境的需要之後再重新設置感測節點，這會是一項艱難的任務。雖然我們為了達到這項目的，我們可以利用無線傳輸的方式來完成，但是在有限能源的前提之下，必須讓每一個感測節點減少不必要的動作以節省能源的消耗。因此，在本篇論文中我們討論的主題就是如何更新無線感測網路的方法。
在本篇論文中，我們提出一個動態更新無線感測網路系統的機制，讓程式設計師可以有效率的動態新增、刪除、更新整個無線感測網路中所有感測節點的程式，而且在更新的過程中不會影響到系統正常地運作。在我們的機制中可以對應用程式做模組化更新。當伺服器端有新的模組產生後，會主動告知周圍的感測器接收新模組，當這些感測器都成功接受新模組後，再將這些模組傳遞給鄰近的感測器，透過多重跳躍( Multi Hop )將新模組從伺服器端傳遍整個感測網路。從實驗結果顯示，SOS經過修改過後，節省了14%的記憶體空間以及20%的 clock cycles。

With the progress of Micro-electromechanical System (MEMS) and wireless communication technology, a tiny device can equip with microprocessor, wireless communication and sensor. Such a tiny device is called a sensor node. The wireless sensor network can be used in extensively fields and attract many researcher perform the research on this field. However, the wireless sensor network still be subjected to some restrictions. For example, the limited memory space can''t store much of data and the application programs. In addition, sensor nodes often use battery as the power source, which may have the problem of power shortage. Finally, once deployed, it is difficult to reprogram the sensor nodes.
As a result, in this thesis, we present a mechanism to wirelessly reprogram the user applications in a sensor node. By our proposed architecture, the system designer can efficiently add, remove and update the application modules. Furthermore, the reprogramming procedure is an on-line process and has no impact on the normal operating of sensor node. In addition, since the extensively energy constraint in the sensor network, we divide the function of a linking loader into two parts. A sensor node only needs to perform the loading process while the linking process is pre-processed on the server to save the energy consumption of a sensor node. From the experimental result, our scheme can save 14% memory usage compared to SOS. Furthermore, the application module execution speed of our scheme is 20 % faster than SOS.

第一章 緒論 1
1.1 簡介 1
1.2 動機 3
1.3目標 4
1.4 論文架構 5
第二章 相關研究背景及知識 6
2.1 完整映像檔方式 6
2.2 差異化修補方式 7
2.3 虛擬機器方式 10
2.4 動態更新模組方式 11
2.5 更新機制總整理 14
第三章 系統實作 16
3.1 ELF格式分析 16
3.1.1 ELF Header 17
3.1.2 Section Header Table 18
3.1.3 Symbol Table 22
3.1.4 String Table 23
3.1.5 Relocation Table 24
3.2 SOS模組更新機制分析 24
3.2.1 Jump Table、Module與Kernel Symbol關係 25
3.2.2 Module image file格式 27
3.2.3 modd_gw與sossrv 28
3.3 SOS模組更新機制修改 31
3.3.1 Linking Function 34
第四章 實驗結果與分析 41
4.1 實驗環境 41
4.2 應用程式模組結果比較 43
第五章 結論及未來工作 47
Reference 49

[1] Athanassios Boulis, Chih-Chieh Han, and Mani B. Srivastava, “Design and Implementation of a Framework for Efficient and Programmable Sensor Networks”. In Proceedings of the First International Conference on Mobile Systems, Applications, and Services(MobiSys 2003)
[2] Da-Wei Chang, Ruei-Chuan Chang, “OS Portal: An Economica Approach for Making an Embedded Kernel Extensible” Journal of Systems and Software Volume67, Issue 1, ISSN: 0164-1212(July 2003)
[3] Adam Chlipala, Jonathan Hui, Gilman Tolle, ”Deluge: Data Dissemination for Network Reprogramming at Scale”. UC Berkeley CS262/CS294-1, Fall 2003 Class Project, http://webs.cs.berkeley.edu/retreat-1-04/slides/jhui_get-deluge_paper.pdf
[4] Crossbow Technology, “Mote In Network Programming User Reference,” TinyOS document, http://webs.cs.berkeley.edu/ tos/tinyos-1.x/doc/Xnp.pdf
[5] ELF. http://www.x86.org/ftp/manuals/tools/elf.pdf
[6] Chih-Chieh Han, Ram Kumar, Roy Shea, Eddie Kohler and Mani Srivastava, “A Dynamica Operation System for Sensor Nodes”.In Proceedings of the Third International Conference on Mobile Systems, Applications, And Services (MobiSys 2005)
[7] Jaein Jeong, Sukun Kim and Alan Broad, “Network Reprogramming,” TinyOS document,
[8] Jaein Jeong and David Culler, “Incremental Network Programming for Wireless Sensors”, IEEE SECON 2004 (Oct 2004)
[9] Linux Loadable Kernel Module HOWTO, http://www.ibiblio.org/pub/Linux/docs/ HOWTO/other-formats/pdf/Module-HOWTO.pdf
[10] Yueh-Feng Lee, and Ruei-Chuan Chang, “Developing Dynamic-Reconfigurable Communication Protocol Stacks using JAVA”. Software—Practice & Experience Volume 35, Issue 6, ISSN: 0038-0644 (SP&E May 2005)
[11] Philip Levis and David Culler, "Mate: A Tiny Virtual Machine for Sensor Networks." In Proceedings of the 10th International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS X), 2002
[12] Philip Levis, Neil Patel, Scott Shenker, and David Culler, “Trickle: A Self-Regulating Algorithm for Code Propogation and maintenance in Wireless Sensor Network,” Proceedings of the First USENIX/ACM Symposium on Networked Systems Design and Implementa
[13] Philip Levis, David Gay, and David Culler, "Active Sensor Networks" In Proceedings of the 2nd USENIX/ACM Symposium on Network Systems Design and Implementation (NSDI), May 2005
[14] Ting Liu and Margaret Martonosi, “Impala: A Middleware System for Managing Autonomic, Parallel Sensor Systems,” ACM PPoPP, 2003.
[15] N. Reijers, K. Langendoen, “Efficient code distribution in wireless sensor networks”, 2nd ACM international conference on Wireless sensor networks and applications, 2003