Abstract: With the rapid development of network, information security has become more and more important.Intrusion detection method has become one of research direction in the field of information security.Intrusion detection is essentially a classification problem.It is very critical to increase the classification accuracy.Immune cloning algorithm and differential evolution algorithm are both effective methods to solve classification problems.Combined with previous good effect of intrusion detection in which immune cloning algorithm and differential evolution algorithm were used respectively.This paper proposes one intrusion detection method based on the improvement of differential evolution algorithm which includes immune cloning algorithm and differential evolution algorithm.Simulation experiments show that this method has higher detecting rate and lower error detecting rate than the method of literature, it can make the intrusion detection system recognition unknown intrusion effectively in certain extent.