Research On Denial Of Service Attack Detection Algorithm Based On S-Kohonen Neural Network

Denial of Service(DoS)is a simple but effective network attacking way.It is an attempt to make a computer or other network resource unavailable to its intended users,such as to temporarily or indefinitely interrupt or suspend services of a host connected to the Internet.As people are increasingly dependent on the network,millions of users become the victim of this kind of attack.This paper studies the DoS attack detection algorithm based on Kohonen neural network.In the existing DoS attack detection method,the algorithm based on artificial neural network is a more effective algorithm.The advantage is that the parallel processing ability is strong,and the robustness and fault tolerance of the noise data,and can be approximated to the complex Non-linear relationship,with associative memory function.However,this kind of algorithm also has the problem that the network convergence is slow and can not gradually concentrate the same neurons.To overcome the above problems,this paper has done the following research work:(1)This paper improves the convergence speed of the Kohonen network by adjusting the learning rate and learning radius of the network and increasing the output layer into a supervised learning network.(2)This paper will add a layer of network weight,by adjusting the weight,winning neurons and peripheral neurons closer the input sample,overcome the problem to a certain extent that the neuron can not be gradually concentrated,and proposed a DoS detection algorithm based on S-Kohonen network.(3)In order to make the DoS detection algorithm based on S-Kohonen network more perfect,the LVQ algorithm can be combined with the S-Kohonen network so that it can classify the input vector pattern.Self-learning,reduce the network structure,improve the detection rate of the algorithm and reduce the computing time.This paper proposed a DoS detection algorithm based on S-Kohonen network.The improved Kohonen network is introduced into the DoS detection algorithm.The algorithm proposed in this paper is further optimized by the LVQ algorithm.The simulation results show that the proposed algorithm is effective.