評估無線網路環境中的攻擊數目及應用

Estimating the Number of Attacks in Wireless Networks and its Applications

賴韋嘉(Wei-Chia Lai)

指導教授 ： 方士豪

元智大學/電機通訊學院/電機工程學系/碩士(2012年)

摘要

近年來無線基地台快速的建設，無線網路已經成為我們生活中的一部分。然而無線基地台的傳播媒介是開放性的，容易受到惡意的網路攻擊。大部分的網路攻擊可藉由加密技術保護，以保障使用者的隱私及安全。但是有些物理攻擊是傳統的加密機制無法抵禦的，這些攻擊是非常難被偵測。

本研究提出兩種估測無線網路中的攻擊數目的方法，最大聯集概率比(Maximum union-based likelihood ratio)和最大聯集概率差(Maximum union-based likelihood difference)。這些估測攻擊數目的方法是利用"邏輯或"與"邏輯與"的組合性的演算法。這方法首先要在不同的參考位置測量沒有被攻擊的訊號特徵並建立機率模型。這演算法可以透過不同的組合方式，觀察接收到的訊號特徵在參考位置上的機率變化，並評估出最有可能的攻擊數目。

我們建立一個模擬的無線網路環境，分析最大聯集概率比和最大聯集概率差以及兩個檢測攻擊的演算法，簡易離群演算法 和隨機抽樣共識演算法。
本實驗討論在不同的訊號攻擊程度下，估測攻擊數目的能力，並分成低估、正確的估測和高估攻擊數目。模擬顯示在較多的攻擊數目下，我們提出的方法比其他的偵測方法能更有效地估測攻擊數目。我們也用真實量測的無線基地台訊號進行實驗。實驗的結果再次證明我們的方法比簡易離群演算法和隨機抽樣共識演算法能更有效的正確估測攻擊數目。
我們提出的方法能夠正確的評估攻擊個數，也能夠提升針對攻擊設計的定位系統的性能。分類式演算法和感測器選擇方法的定位系統若加入攻擊數目的估測，能在一步的提升強健性。實驗結果顯示了這兩種定位方法的性能被提升4.15% 和0.42%的50%誤差圓徑(circular error probable)以及2.01% 和 8.34%的67%誤差圓徑。

關鍵字

惡意攻擊 無線網路 強健定位

並列摘要

Received signal strength (RSS) is commonly employed in network services.
However, it's sensitive to malicious attacks due to the nature of its open medium.
Traditional cryptographic technique can provide personal privacy. However, it can't defend the physical attacks.
This paper proposes two methods to estimate number of attacks, called the Maximum Union-based likelihood Ratio (MULR) and Maximum Union-based likelihood Difference (MULD).
These methods adopt "union" and "intersection" operators to combine all possible combinations, which is capable of estimating correct number of attacks under more attacks.
In simulation, we evaluate capability to estimate number of attacks. The results demonstrated these approaches better than A Simple Outlier (ASO) and RANdom SAmple Consensus (RANSAC).
The experimental results demonstrated capability in an actual Wi-Fi network again.
In application, we try to use estimating results to improve cluster-based and sensor selection methods in positioning performance. The results show that the performance is improved by our proposed method.

並列關鍵字

malicious attack wireless robust localization

參考文獻

[1] D. Anzai and S. Hara,“A simple outlier data rejection algorithm for an RSSIbased ML location estimation in wireless sensor networks,”in IEEE Vehicular Technology Conference, pp. 1–5, 2008.

[3] J. Yang, Y. Chen, V. Lawrence, and V. Swaminathan, “Robust wireless localization to attacks on access points,”IEEE Sarnoff Symposium, pp. 1 –5, 2009.

[4] A. Kushki, K. Plataniotis, and A. Venetsanopoulos, “Sensor selection for mitigation of RSS-based attacks in wireless local area network positioning,”in IEEE International Conference on Acoustics, Speech and Signal Processing, pp. 2065–2068, 2008.

[5] S.-H. Fang, W.-J. Lai, and Y.-C. Liang,“An encryption-based approach for protecting privacy in network-based location systems,” International Conference on Machine Learning and Cybernetics, vol. 1, pp. 377 –380, 2011.

[6] S. Misra and S. Myneni, “On identifying power control performing sybil nodes in wireless sensor networks using RSSI,” Global Telecommunications Conference, pp. 1–5, 2010.