本文選題：S-Kohonen網絡 + DoS�。� 參考：《小型微型計算機系統(tǒng)》2017年03期

【摘要】：拒絕服務(Denial of Service,DoS)是企圖使其預期用戶的一臺主機或其他網絡資源不可用,如臨時或無限期地中斷或暫停連接到因特網主機的服務.為了有效地阻止DoS攻擊,首先需要提高DoS攻擊檢測的準確性,提出一種基于改進Kohonen網絡的DoS攻擊檢測算法.該方法通過對DoS攻擊原始數(shù)據的預處理,為后續(xù)數(shù)據處理的方便和保證程序運行時加快收斂奠定必要的基礎,采用檢測結果的正確率作為該算法的評價指標,采用SOM學習算法是把高維空間的輸入數(shù)據映射到低維神經網絡上,并且保持原來的拓撲次序,然后建立S-Kohonen(Supervised-Kohonen)神經網絡檢測模型.實驗結果表明,與傳統(tǒng)的Kohonen方法相比,S-Kohonen網絡具有更好的檢測性能.
[Abstract]:Denial of Service (dos) is an attempt to disable a host or other network resource of its intended user, such as temporarily or indefinitely interrupting or suspending services connected to an Internet host. In order to effectively prevent DoS attacks, it is necessary to improve the accuracy of DoS attack detection. A DoS attack detection algorithm based on improved Kohonen network is proposed. By preprocessing the raw data of DoS attack, the method lays a necessary foundation for the convenience of the subsequent data processing and the guarantee of program convergence. The correct rate of the detection result is used as the evaluation index of the algorithm. The SOM learning algorithm is used to map the input data of the high-dimensional space to the low-dimensional neural network and maintain the original topological order. Then the S-Kohonenn Supervised-Kohonen neural network detection model is established. The experimental results show that the S-Kohonen network has better detection performance than the traditional Kohonen method.
【作者單位】： 河北大學電子信息工程學院;
【基金】：國家科技支撐計劃項目(2013BAK07B04)資助 國家自然基金項目(61672205)資助 河北省自然科學基金項目(F2013201170)資助 河北省高等學�？茖W技術研究重點項目(ZD2014008)資助
【分類號】：TP393.08

【相似文獻】

相關期刊論文 前10條

1 張春爐;沈建京;;基于SOM算法的文本聚類實現(xiàn)[J];計算機與現(xiàn)代化;2010年01期

2 陳志兵;黃人t，

本文編號：1887045