無線網路環境下阻斷攻擊偵測之研究

Abstract

近年來隨著無線網路技術成為新興科技的發展重點，而無線網路設備價格的降低，使得越來越多使用者開始利用無線網路作為他們連結網際網路的媒介。儘管無線網路的普及帶來了許多便利，相對的也衍生出許多網路安全方面的隱憂。這些問題將隨著無線網路的使用率的增加而日益嚴重。 \n 有鑑於此，本研究探討了目前無線網路環境下的網路入侵和攻擊事件的各類方式，並針對無線網路環境下發生的阻斷攻擊事件，提出一個以無線網路基台的SNMP 流量資料為基礎，並利用貝氏分類法做為預測判斷的自動偵測模式，實際以校園無線網路環境進行驗證此一偵測模式，結果達到95% 的準確率，僅有3%的遺漏率和7%的誤報率。此一偵測模式的結果顯示可有效的驗證出無線網路環境下的阻斷式攻擊事件的發生。

have changed significantly in the past 2 years that brought forth a whole new regimentation to the Internet users. Wireless LANs provide high-speed, cable-free access and inexpensive facilities that makes it more convenient and attractive nowadays. While the wireless network has accumulated a large demand of wireless network services, the security concern is gradually important in the recent years. \n In this investigation, an auto-detective model has proposed to forecast the wireless network from DoS attacks. Based on the Naive Bayesian Classification and the SNMP traffic data, the results show that the proposed detective model has 95% accuracy, with 3% missing rate and 7% false alarm rate, which can effectively forecast the DoS attack. Based on the research findings, implications and limitations are discussed.