General Theory of security and a study of hacker’s behavior in big data era

Big data brings great value as well as a lot of network security problems, which makes the hacker possess more and more attack strategies. This paper precisely describes the static form of hackers, and proposes the best dynamic hackers attack tactics under certain assumptions. When the proportion of the hacker’s resource input is its static probability distribution value, the hacker income reaches maximum. In particular, on the premise of uniform ratio of input and output, if the entropy of hacker reduces 1 bit, the hacker income will be double. Furthermore, this paper studies the optimal combination of hacker attacks and proposes a logarithmic optimal combination attack strategy that the hacker attacks several systems simultaneously. This strategy not only can maximize the hacker’s overall income, but also can maximize the income of each round attack. We find that the input-output ratio of each system will not change at the end of this round attack when hacker adopts the logarithmic optimal combination strategy, and find the growth rate of additional hacker income does not exceed the mutual information between the input-output ratio of the attacked system and the inedge information if an attacker can get some inedge information through other ways. Moreover, there is an optimum attack growth rate of hackers if time-varying attacked system is a stationary stochastic process. We can conclude that, in Big Data era, the more information the hacker gets, the more hacker income.