Analysis on Password Attack Model and Password Generation

Identity authentication is an important line of defense of network security, password has long been the mainstream way of identity authentication, password attack plays an important role in password security research. Probabilistic Context Free Grammar (PCFG) and the Markov chain-based model are the two most effective password attack methods. They can effectively model and describe the password from the substructure composition level and the character dependence level, respectively. This paper first analyzes the various encryption algorithms for cryptographic database, which are currently mainstream, and describes the comparison and strength of these algorithms. Moreover, this paper analyzes the current cipher structure within the password database. At last, this paper briefly describes the current method of password attacks and focuses on three most important methods, including Markov chain, OMEN algorithm and context-independent, to crack the password. This paper discusses the development of these three algorithms in current academic field, including the Recursive-OMEN algorithm based on structure partition and string reorganization, Adam algorithm based on dictionary attack, and genetic algorithm.