結合邏輯運算與移動平均線之量子啟發式禁忌搜尋演算法於偵測無線感測網路蟲洞攻擊

Using Quantum-inspired Tabu Search Algorithm with Logic Operation and Moving Average for Wormhole Attack Detection in WSN

何冠賢(Kuan-Hsien He)

指導教授 ： 周耀新

國立暨南國際大學/科技學院/資訊工程學系/碩士(2017年)

摘要

無線感測網路(WSN)是由許多感測器所組成，感測器一邊移動一邊蒐集資料，並透過彼此互相傳送資料將資料回收，無線感測網路可以應用在很多領域，最早是應用在軍事戰場，用來偵測敵人的位置，後來也可以應用醫療設備、生態與環境的監控等等。而在無線感測網路中，感測器彼此之間透過多點跳躍的方式傳送訊息，傳送封包的過程中容易受到各種攻擊，蟲洞攻擊就是其中的一種，蟲洞攻擊是一個很重要的安全議題，蟲洞是由一對或多對以上的惡意節點所組成，他們之間有著一條特殊的通道，不論蟲洞之間距離多遠，都能將自身的鄰居訊息透過這個通道傳送給另一端蟲洞附近的鄰居，它會使受攻擊的節點導致路由的錯誤，減少感測器的生命週期並造成整個網路拓樸的損毀，然而至今已有許多方法使用特殊的硬體設備或是消耗了許多資源來解決蟲洞攻擊問題。

在本篇研究中改善了一個只使用感測器所收集的鄰居資訊的偵測方法，不需要額外的硬體設備、複雜的計算與消耗大量的資源，藉由鄰居數計算出移動平均線(MA)，再與邏輯運算結合，並使用量子啟發式禁忌搜尋演算法(QTS)有效的找出MA的規則組合來偵測蟲洞，本研究將一個較不合理的移動模型更改為一個較常見且符合現實的移動模型，並利用此移動模型的特性抓出偵測節點進、出蟲洞的時機，提出一個使用較少資源並仍然有效的偵測蟲洞方法。最後實驗結果顯示本方法在不同的節點密度或不同的感測半徑下，都有很好的表現。

關鍵字

蟲洞攻擊 ； 無線感測網路 ； 量子啟發式禁忌搜尋演算法 ； 移動平均線 ； 邏輯運算

並列摘要

Wireless sensor networks (WSN) are composed of sensors and applied in many fields, such as military, ecological and environmental monitoring. The wormhole attack problem is one of important security issues in WSN. The wormhole is consisted of one or more pairs of malicious nodes. When a node moves to the sensing range of the wormhole, it will receive the neighbor information through the special tunnel from the other wormhole. It may cause many problems such as packets routing error, reduce the lifetime of the sensor and let the entire of network topology damage. However, several methods have been proposed that using special hardware devices or lots of resource consumption.

In this paper, we improve the method which utilizes the neighbor information collected by mobile nodes without extra hardware, complex calculation and lots of resource consumption. Using moving average (MA) indicator and combines with logic operation, AND gates and OR gates. And using Quantum-inspired Tabu Search (QTS) algorithm to effectively find the combinations to detect wormhole. However, we change an uncommon mobility model of nodes to a general one, and utilize the characteristic of mobility model to detect the node moves into the wormhole and out of the wormhole. Finally, the simulation results show that the proposed method with great performance uses less resources but still efficiently detects wormhole in different node density and transmission range.

並列關鍵字

Wormhole attacks ； Wireless Sensor Network ； Quantum-inspired Tabu Search Algorithm ； Moving Average ； Logic operation

參考文獻

[1]Y. W. Z. Zhang, and J. Wu, (2010, July). A distributed approach for hidden wormhole detection with neighborhood information. In Networking, Architecture and Storage (NAS), 2010 IEEE Fifth International Conference on(pp. 6372). IEEE.

[2]M. H. Jao, “A novel wormhole attack detection in WSN: Using Quantum inspired Tabu Search Algorithm with logic operation and moving average,” ., Puli, Taiwan, 2016.

[3]L. Hu, and D. Evans, (2004, February). Using Directional Antennas to Prevent Wormhole Attacks. In NDSS.

[4]L. Buttyan, L. Dora, and I. Vajda, (2005). Statistical wormhole detection in sensor networks. In Security and Privacy in Ad-hoc and Sensor Networks (pp. 128-141). Springer Berlin Heidelberg.

[5]G. N. Shirazi, and L. Lampe, (2010, May). Lifetime Maximization of UWB-Based Sensor Networks for Event Detection Applications. In Communications (ICC), 2010 IEEE International Conference on (pp. 1-6). IEEE.