簡易檢索 / 詳目顯示

研究生: 張佳銘
Chia-ming Chang
論文名稱: 針對空間關聯資料之異常偵測
Outlier Detection for Spatial Correlated Data
指導教授: 陳郁堂
Yie-tarng Chen
口試委員: 呂永和
Yung-ho Lu
方文賢
Wen-hsien Fang
陳省隆
Hsing-lung Chen
林銘波
Ming-bo Lin
學位類別: 碩士
Master
系所名稱: 電資學院 - 電子工程系
Department of Electronic and Computer Engineering
論文出版年: 2007
畢業學年度: 95
語文別: 英文
論文頁數: 44
中文關鍵詞: 空間關聯感測網路異常偵測最長子字串
外文關鍵詞: spatial correlated, sensor networks, abnormal detection, Longest common subsequence
相關次數: 點閱:169下載:2
分享至:
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報
  • 散佈於廣大無線感測網路裡的感測節點個體是很容易招受實體的攻擊與入侵,敵方透過入侵節點來進而破壞整個資料關聯的網路,因為這些被入侵的節點被利用來散播錯誤的資料,如果這些被入侵的節點未能即時偵測出來,則這些錯誤資料將會傳回資料集合點(Sink)。將導致不只發生錯誤的警報也會發生能源耗損的問題。因此發展一個有效的偵測方法是必要的,因為這樣的攻擊並無法被基本金鑰管理偵測出來。
    在這篇研究裡,我們提出了Motif-finding演算法,這個演算法可以正確及有效的判斷是否二個節點為相似或不相似。因此根據Motif-finding演算法我們進而提出二個偵測方法,其中一種是根據循序假設檢定的分散式的投票機制,另一個方法是根據事前的訓練得到一個關鏈性的圖形,利用這個圖型我們去偵測出異常節點。實驗結果顯示了分散式投票機制可以利用較短的時間快速偵測出異常節點,且此兩個偵測方法可以正確的偵測出異常節點


    Individual sensor nodes in a large-scale wireless sensor network are subject to physically security compromises. Adversary destroys wireless sensor networks through these compromised sensors, because these compromised sensors can be used to inject false sensing data. If these compromised Sensors are not detected, these false data reports are forwarded to sink (data collection point) It will result in not only false alarms but also consumption of the battery energy in each sensor. For these cases, it is necessary to develop a mechanism to detect them, because compromised nodes can not be detected by key management.
    In this research, we present a Motif-finding approach to examine if the two sensors are similar. Based on the Motifs finding approach, we propose two false data detection scheme. One is a distributed collective detection based on sequential hypothesis test, and the other is correlation graph detection through the result of advanced training. The results of simulation show the distributed collective detection can effectively detect the abnormal sensor node with short detection time, and both distributed collective detection and correlation graph detection have high detection rate.

    Contents CHAPTER 1 INTRODUCTION 1 1.1. Motivation and Problem Description 2 CHAPTER 2 RELATED WORK 4 2.1. A key-Management Scheme for Distributed Sensor Networks 4 2.2. The Longest Common Subsequence Problem 6 2.3. Clustering with minimum spanning tree 6 2.4. Statistical En-route Filtering of injected dales Data in Sensor Networks 6 2.5. Motifs 7 2.6. Random Key Redistribution Schemes for sensor Networks 7 2.7. Random Key Redistribution Schemes for sensor Networks 7 CHAPTER 3 ARCHITECTURE OF PROPOSED OUTLIER DETECTION SCHEME FOR SPATIAL CORRELATED DATA 8 3.1. Overview 8 3.2. Motif- Finding in Time Series Data 10 3.2.1. Piecewise Aggregate Approximation Transformation 11 3.2.2. Word Transformation 12 3.2.3. Longest Common Subsequence Finding 13 3.2.4. Motif-discovery 16 3.3. System time 17 3.4. Distributed Monitoring 18 3.4.1. Distributed hashed-based voting mechanism 18 3.4.2. Sequential Hypothesis Test 19 3.5. Correlation graph detection 21 3.5.1. Advanced training 21 1. Min-spanning tree clustering 22 2. Correlation graph modification 23 3.5.2. Outlier detection 24 CHAPTER 4 PERFORMANCE EVALUATION 25 4.1. Performance Evaluation Criteria 27 4.2. Simulation Result 27 CHAPTER 5 CONCLUSION 31 REFERENCES 32

    References
    [1] F. Ye, H. Luo, S. Lu, and L. Zhang. Statistical en-route detection and filtering of injected false data in sensor networks. In IEEE INFOCOM 2004, 2004
    [2] W. Zhang and G. Cao, “Group rekeying for filtering false data in sensor networks: A predistribution and local collaboration based approach,” IEEE INFOCOM’05, 2005.
    [3] Laurent Eschenauer and Virgil D. Gligor. A key management scheme for distributed sensor networks. In Proceedings of the 9th ACM Conference on Computer and Communication Security, pages 41–47, November 2002.
    [4] H. Chan, A. Perrig, and D. Song. Random key predistribution schemes for sensor networks. In IEEE Symposium on Security and Privacy, May 2003.
    [5] Lin, J., Keogh, E., Patel, P. & Lonardi, S. (2002). “Finding motifs in time series”, 2nd Workshop on Temporal Data Mining at the 8th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. July 23 - 26. Edmonton, Alberta, Canada.
    [6] R.A. Wangner and M. J Fischer ”The string-to-string correction problem,” Journal of the ACM. Vol 21,No. 1,pp. 168-173, 1974
    [7] Christos Siaterlis and Vasilis Maglaris, “Detecting incoming and outgoing DDoS attacks at the edge using a single set of network characteristics”, Computers and Communications, 2005. ISCC 2005. Proceedings. 10th IEEE Symposium on 27-30, Page(s):469 – 475, June 2005
    [8] A. Wald. Sequential Analysis. New York: Wiley, 1947.
    [9] J. Jung, V. Paxson, A. W. Berger, and H. Balakrishnan. Fast portscan detection using sequential hypothesis testing. In Proceedings IEEE Symposium Security and Privacy, pages 211-225, May 2004.
    [10] H. Chan, A. Perrig, and D. Song. Random key predistribution schemes for sensor networks. In IEEE Symposium on Research in Security and Privacy, 2003.
    [11] Keogh, E., & Pazzani, M. (2000). A simple dimensionality reduction technique for fast similarity search in large time series databases. In 4th Pacific- Asia Conference on Knowledge Discovery and Data Mining. Kyoto, Japan
    [12] Thomas H. Cormen and Charles E.Leiserson, Introduction to Algorithms, 2thed., McGraw-Hill Bok Co., 2001

    QR CODE