86 年度

1. Web System Performance Evaluation, 主持人, 工研院電通所, 86/7-87/6

2. Design of Name Service for Collaborative Web Proxy Servers, 主持人, 國科會, 86/8-87/7

3. A Prototype System of Host-based Virtual Private Network, 共同主持人, 中科院, 87/8-88/7

4. Towards the Building of a Web-Based Collaborative Computing Environment, 協同研究人員, 國科會電信國家型計畫, 87/8-88/7

 

87 年度

5. Design of Name Service for Collaborative Web Proxy Servers, 主持人, 國科會, 87/8-88/7。

6. A Prototype System of Host-based Virtual Private Network, 共同主持人, 中科院, 87/8-88/7。完成一個點對點的虛擬加密通道系統。

 

88 年度

7.政府憑證管理中心安全防護與安全強化分析研究, 88/10-89/09, 中華電信所。

8. Digital library/museum content management: rights management

 

89 年度

9. 執行中研院資訊所所內計畫『資訊網際合作實驗室之研究與開發-安全管理機制』

10 行國科會計畫「行動虛擬專線之安全程式代理系統」,NSC89-2213-E-001-044，89/08∼90/07

11 執行國際數位圖書館合作研究計畫--IDLP－智慧財產權保護機制研究－數位圖書館內涵保護及電子版權管理, NSC89-2750-P-001-022-5, 89/08∼90/07。

12. 執行中研院主題計畫『國家典藏數位化專案 – rights management』

13. 執行 TW-CERT 研究計畫 (海軍總部、國防部、研考會等單位委託計畫)

14. 與中原大學資工系田筱榮教授共同執行研究計畫（中華電信研究所、中科院等委託計畫）

 

90 年度

15. 執行所內計畫與國科會計畫『資訊網際合作實驗室之研究與開發-安全管理機制』

16執行國際數位圖書館合作研究計畫--IDLP－智慧財產權保護機制研究－數位圖書館內涵保護及電子版權管理, 90/08∼91/07。

17執行中研院主題計畫『國家典藏數位化專案 – rights management』

18 執行 TW-CERT 研究計畫 (海軍總部、國防部、研考會等單位委託計畫)

19. 與中原大學資工系田筱榮教授共同執行研究計畫（中華電信研究所、中科院等委託計畫）

中華電信研究所：整合性資訊系統之安全防禦機制研究(TL-90-8F01)

中科院：弱點資料庫設計研究(NSC 90-2623-7-033-006)

20 行經濟部九十年度科技專案國家資通安全技術服務計畫分包研究計畫『規劃我國資通安全技術研發策略』, 90/9~90/12

 

91 年度

21. 執行中研院主題計畫『國家典藏數位化專案 – rights management』，92/1∼92/12。

22. 執行 TW-CERT 研究計畫 DNS安全檢測機制 (TWNIC委託計畫), 92/1~92/12。

23. 與中原大學資工系田筱榮教授共同執行中華電信研究所計畫。

24.  執行經濟部九十一年度科技專案國家資通安全技術服務計畫分包研究計畫『規劃我國資通安全預警機制研究與規劃』, 91/5~91/12

共同主持執行 Open Source Foundry 計畫，92/1∼92/12。

 

l          目前進行中之計畫

 

1. 程式動態行為安全分析

　　程式行為與軟體安全性關係密切，失控(crash)的程式將有可能被運用而成為安全弱點所在。我們很難在程式失控後找出真正原因，因此多數研究藉助於動態程式實驗，著重於偵測錯誤與區別失控的根源。軟體廠商為了趕上市發行的時間，以致系統常伴隨著非有意的過失而產生軟體失控的狀況，有些只影響系統穩定性，有些則產生安全弱點。我們的目標是設計工具程式，進行動態安全分析，輔助找尋錯誤點，判斷是否可供運用產生系統安全缺陷。

 

　　我們首先將開發執行監控操作系統。此系統會定時進行執行狀態監測，若程式失控，可以重複原先執行步驟，嘗試觸發先前的錯誤，藉此可觀察執行程式的內部行為，例如 API 呼叫序列順序，呼叫參數與返回值關係。這些可經由包覆系統 API 呼叫的技術達成，並判斷其結果否異常。預期將完成程式狀態觀測系統，可判斷所觀測之失控點是否安全相依（是否安全可運用性）。

　　我們陸續將建立軟體應用程式與作業系統函數之間的互動介面。

　　經由包覆的技術，不但能記錄相關重要參數，也能接受測試指令進行呼叫參數與返回數值的任意替換。我們可以隨意操作程式，改變想要運作的呼叫參數，觀察其回應狀態並找出可疑的失控點。我們將設計互動介面與測試驅動系統以便控制包覆點，並提供呼叫點相關資訊以產生較細微的測試涵蓋率。預期將完成程式錯誤注入測試系統，有系統性地激發錯誤點，並產生有意的安全運用程式碼。

 

關鍵詞：軟體安全、動態分析、軟體包覆器、COTS 弱點測試、安全可運用性

 

　　Program running behavior has much to do with software security. Crashed software may be exploited to be a potential vulnerability. It is difficult to reconstruct system failures after a program has crashed. There is much research effort on detecting program errors and identifying their root causes either by static analysis or observing their running behavior through dynamic program instrument. In order to meet the time to market, software releases with unintended flaws. Some of them cause software crash, while others may introduce security vulnerabilities. Our goal is to design a tool that helps analyze program running behavior and determine if it is an exploitable vulnerability.

　　We will develop an execution instrument and interception system. This system will periodically monitor software running behavior. If the software crashes, we can roll back to the latest checkpoint and trigger the fault. We will observe the internal behavior of running programs, such as API call sequence, call parameters and return values through wrapping system call API techniques, and determine whether these things are anomalous or not. We expect to develop a dynamic instrument tool able to determine if the crash site is security exploitable.

　　We investigate the design and implementation of such a tool to instrument the interfaces between the software application and the operating system functions with an interactive software wrapper. This wrapper cannot only intercept the functions to record the parameters and the return value but also receiving testing directives to replace calling parameters and the return value with any arbitrary value. We could use this tool to easily instrument the application, change the intended OS function call parameters with testing data and observe the response of the application to find out the suspicious crash sites. We devise a GUI and testing driver to control the wrapper and provide call site information with finer-grained test coverage. We expect to complete a fault injection tool systematically generating triggers for intended exploit code.

 

Keywords: Software Security, Dynamic Analysis, Software Wrapper, COTS Vulnerability Testing, Security Exploitability

 

2. 嵌入式感測器核心惡劣環境測試與保護

　　Sensor network 所佈建的環境通常遭遇到原設計者所設想不到的狀況，例如高溫、高濕度、震動、電源不足、電磁干擾等，這是執行環境不同所導致；而所搭配的介面電路或輸出入特性也有所差異，這是人為設計的差異問題。這兩方面的異動：執行環境與設計環境，輸出入介面與設計介面，都會導致Sensor network 的不正常工作。我們研究的目標在於進行錯誤注入測試與惡劣環境模擬，希冀在佈建系統前，找出核心系統可能的弱點缺陷，以提早修正。同時設計相關保護程序，在面臨不正常外在條件激發的情況下，仍能維持最低基本需求、存活運作。因為此類型核心系統必須精簡功能，運算功能有限，如何在有限度資源情況下，進行容錯操作，是此研究具挑戰性的工作，我們預計嘗試回復導向計算(Recovery-oriented Computing)相關設計方法，在不增加多餘元件情況下，達成強健性核心設計的需求。事前的惡劣環境模擬也是設計的重點，如何產生完整、高涵蓋率的測試資料也將是本研究必須面臨的難題。我們預期改進目前已發展的sensor network 核心程式，具有自動回復運作的容錯機制，同時建立完整的惡劣環境模擬測試資料產生系統。

 

關鍵詞：嵌入式核心、探測器網路、錯誤注入測試、回復導向計算、必然失控模組。

 

　　The deployment of sensor network often encountered unexpected situations, such as high temperature, high humidity, vibration, insufficient power supply, and EM interference. They are introduced by the variation of execution environment; Furthermore, differences between interface circuit and input/output characteristic will also impact the design decisions, which are man-made variations. These two aspects of variations: execution and design environment, input/output and design interface, all attributed to malfunctions of sensor network systems. Our research objectives are to exploit fault injection tests and malicious environment simulation, wish to find potential vulnerabilities before the system deployment, and fix them in time. Similarly, we try to design related protection measures, when in front of external triggers abnormally rendered, the system can still maintain their minimum service requirement and survive to work properly. Since such micro-scale kernels must maintain their concise features, due to limited computation power, it’s a challenging work to proceed fault-tolerance operations under restricted resource.

　　We expected to use recovery-oriented computing approach to attain robustness requirement under the assumption of no redundant components. It is also an important design decision to emulate the malicious environment. How to generate complete and high coverage benchmark data will be the major obstacle of our research. We expect to improve our current sensor network kernel, with automatic recovery capability, and build a benchmark generator for the environment.

Keywords: Embedded Kernel, Sensor Network, Fault Injection, Recovery-Oriented Computing, Game Semantics, Crash-only Modules, Dependability, Micro-reboot

 

3. 開放架構的螢幕串流系統

OpenScreen: An Open Architecture for Streaming Contents for E-learning Environment and Network (or OASEC: An Open Architecture for Streaming E-learning Contents)

　　我們將發展一套開放源碼的螢幕串流系統(Screen Streaming System)。整合幾個螢幕串流幾個主要技術：螢幕畫面擷取。擷取資料壓縮。傳送畫面與顯示(rendering)。支援 MS-window 與 X-window 顯示。與多種媒體資料同步，包括視訊與語音。支援後製作、並能自動化。（製作以簡報標題之影音檢索目錄）。

支援 SMIL。支援更多媒體資料，如 lecturer gesture data、facial expression、laser pointer tracking、instant message。解決其同步與重疊顯示問題(動態 overlay)。將採用Application Sharing 相關協定應用、與Application Streaming 的泛用性應用方法。目前在 e-learning 方面技術已經相當成熟，運用的方法可分為幾類：screen capture, application preprocessing, 與application sharing (multi-point)。主要需求如下：

 

  • real-time application sharing (on-line with lecturer), sharing with multiple-users for interactions between audience and the lecturer. Most audiences are read-only sharing. We can optimize for such users. The challenge issues are how we record the interactions/sharing and automate the process for post-production. Since the interactions are in real-time, applications of the lecturer are shared by all audiences and audiences can raise any questions by sharing their actions on the screen to the lecturer and all the other audiences. (for example, type in chat room, draw the screen,…) Floor control is of concerns. Hope all interactions are recorded.
  • on-demand application sharing for pre-recorded media. Audiences can connect on their demand. Actually, teaching assistant(TA) can be of help with users interested in the pre-recorded media. TA will act like the lecturer and play with the audiences, including annotating in the streaming media.

 

4.軟體版本歷史資料發掘以驗證開放源碼外圍參與者的學習過程

Mining Version Histories for Verifying Learning Process of Legitimate Peripheral Participants

 　　Since code revisions will reflect the extent of human involvement in the development process, revision histories reveal and annotate the interaction and interface between developers and modules.

　　We therefore try to divide developers and modules into groups according to revision histories from the open source software repository, such as sourceforge.net. To describe the interactions between open source development process, Legitimate Peripheral Participation (LPP) is a representative model for partitioning developers into groups such as core term and peripheral by the evolution process of learning behavior. With the conventional module relationship, we divide modules into kernel and non-kernel (such as UI). Groups of developers and modules have been formerly partitioned in natural sense with informal criteria. In our work, we propose a developer-module link relation model to analyze these grouping structures between developers and modules. Our results discover some process cases with relative importance in the constructed graph related to project development. It reveals a certain subtle relationships of interactions between core and non-core team developers, and interfaces between kernel and non-kernel modules.

 

Keywords: Open Boundary, Open Source Software Development Process, Legitimate peripheral participants(LPP).