With the dynamic description method for self and antigen, and the concept of dynamic immune tolerance for lymphocytes in network-security domain presented in this paper, a new immune based dynamic intrusion detection ...With the dynamic description method for self and antigen, and the concept of dynamic immune tolerance for lymphocytes in network-security domain presented in this paper, a new immune based dynamic intrusion detection model (Idid) is proposed. In Idid, the dynamic models and the corresponding recursive equations of the lifecycle of ma- ture lymphocytes, and the immune memory are built. Therefore, the problem of the dynamic description of self and nonself in computer immune systems is solved, and the defect of the low efficiency of mature lymphocyte generating in traditional computer immune systems is overcome. Simu- lations of this model are performed, and the comparison experiment results show that the proposed dynamic intrusion detection model has a better adaptability than the traditional methods.展开更多
Inspired by biological immune system, a new dynamic detection model for computer virus based on immune system is proposed. The quantitative description of the model is given. The problem of dynamic description for sel...Inspired by biological immune system, a new dynamic detection model for computer virus based on immune system is proposed. The quantitative description of the model is given. The problem of dynamic description for self and nonself in a computer virus immune system is solved, which reduces the size of self set. The new concept of dynamic tolerance, as well as the new mechanisms of gene evolution and gene coding for immature detectors is presented, improving the generating efficiency of mature detectors, reducing the false-negative and false-positive rates. Therefore, the difficult problem, in which the detector training cost is exponentially related to the size of self-set in a traditional computer immune system, is thus overcome. The theory analysis and experimental results show that the proposed model has better time efficiency and detecting ability than the classic model ARTIS.展开更多
基金This work was supported by the National Natural Science Foundation of China under Grant No.60373 110the Specialized Research Fund for the Doctoral Progrant of Higher Education of China uinder Grant No. 200306 10003. the New Century Excellent Expert Pro-gram of Ministry of Education of China under Grant No. NCET-04-0870the Inmovation Foundation of Sichuan University under Grant No.2004CF10.
文摘With the dynamic description method for self and antigen, and the concept of dynamic immune tolerance for lymphocytes in network-security domain presented in this paper, a new immune based dynamic intrusion detection model (Idid) is proposed. In Idid, the dynamic models and the corresponding recursive equations of the lifecycle of ma- ture lymphocytes, and the immune memory are built. Therefore, the problem of the dynamic description of self and nonself in computer immune systems is solved, and the defect of the low efficiency of mature lymphocyte generating in traditional computer immune systems is overcome. Simu- lations of this model are performed, and the comparison experiment results show that the proposed dynamic intrusion detection model has a better adaptability than the traditional methods.
基金the National Natural Science Foundation of China (Grant No.60573130)the 863 Project of China (Grant No.2006AA01Z435)
文摘Inspired by biological immune system, a new dynamic detection model for computer virus based on immune system is proposed. The quantitative description of the model is given. The problem of dynamic description for self and nonself in a computer virus immune system is solved, which reduces the size of self set. The new concept of dynamic tolerance, as well as the new mechanisms of gene evolution and gene coding for immature detectors is presented, improving the generating efficiency of mature detectors, reducing the false-negative and false-positive rates. Therefore, the difficult problem, in which the detector training cost is exponentially related to the size of self-set in a traditional computer immune system, is thus overcome. The theory analysis and experimental results show that the proposed model has better time efficiency and detecting ability than the classic model ARTIS.
