A method of security measurement of the network data transmission

The paper puts forward a method to measure the security of network data transmission as well as determine attack category. Based on the self-similarity of network traffic and fuzzy decision making, the model is constructed to determine the attack category as follows: sample the network traffic with time intervals following Poisson distribution; divide the obtained data series of traffic into several sub series of equal size, and compute the confidence interval of Hurst parameter for each sub series with Whittle estimation method; Compare with the confidence intervals of Hurst parameter of known attack traffics, which have already been worked out with the same method, and thereby find out what the unknown attack is. On this basis, we propose an approach of security measurement of the network data transmission. According to the number of times that the routers on the transmission path are attacked, corresponding attack properties and the damage weight value of each attack, which suggests the attack´s destructivity, quantitative analysis of the security of network transmission can be accomplished. The results of the experiments show that this method not only helps make the policy of network data transmission accurately and efficiently, but also guarantees the security of the network data transmission.