本文關(guān)鍵詞： 入侵檢測 非均勻分布 變異特征 高斯分布　出處：《科技通報》2013年08期 　論文類型：期刊論文

【摘要】：網(wǎng)路入侵過程中入侵特征種類繁多,形成耦合性,很難形成較為規(guī)則的分布,傳統(tǒng)的入侵檢測方法都是假設(shè)網(wǎng)絡(luò)入侵特征呈現(xiàn)獨立高斯分布的,但是,一旦入侵特征耦合性較差,造成非高斯入侵數(shù)據(jù)擬合能力差,導(dǎo)致檢測精度不理想。為了避免上述缺陷,提出了一種基于變異特征估計算法的非均勻分布入侵檢測模型。在海量的網(wǎng)絡(luò)操作數(shù)據(jù)中,提取出變異特征,根據(jù)提取的特征能夠進行網(wǎng)絡(luò)入侵檢測。利用變異特征估計算法,能夠建立合理的非均勻分布入侵檢測模型,從而檢測出網(wǎng)絡(luò)入侵行為。實驗結(jié)果表明,在非均勻分布的環(huán)境下,利用該算法對網(wǎng)絡(luò)攻擊行為進行檢測,使非高斯數(shù)據(jù)具有更強的擬合能力,極大地降低了網(wǎng)絡(luò)入侵檢測的誤報率和漏報率,提高了入侵檢測的檢測率。
[Abstract]:In the process of network intrusion, there are many kinds of intrusion features, forming coupling, so it is difficult to form a more regular distribution. Traditional intrusion detection methods assume that the network intrusion features are distributed independently of Gao Si, but, Once the coupling of intrusion features is poor, the fitting ability of non-#china_person0# intrusion data is poor, and the detection accuracy is not ideal. In order to avoid the above defects, In this paper, a non-uniform distributed intrusion detection model based on mutation feature estimation algorithm is proposed, in which variation features can be extracted from massive network operation data, and network intrusion detection can be carried out according to extracted features. A reasonable non-uniform distributed intrusion detection model can be established to detect the network intrusion behavior. The experimental results show that the algorithm is used to detect the network attack behavior in the non-uniform distributed environment. It makes the non-#china_person0# data have stronger fitting ability, greatly reduces the false alarm rate and false alarm rate of network intrusion detection, and improves the detection rate of intrusion detection.
【作者單位】： 佛山廣播電視大學(xué)教育技術(shù)實驗中心;佛山科學(xué)技術(shù)學(xué)院信息與教育技術(shù)中心;
【基金】：廣東省教育廳、佛山市、中央電大、省電大科研項目立項 廣東省電大遠程教育開放基金項目(YJ1110)
【分類號】：TP393.08

【參考文獻】

相關(guān)期刊論文 前4條

1 汪興東,佘X,周明天,劉恒;基于BP神經(jīng)網(wǎng)絡(luò)的智能入侵檢測系統(tǒng)[J];成都信息工程學(xué)院學(xué)報;2005年01期

2 張新有;曾華q，

本文編號：1549510