Dynamic Markov Model: Password Guessing Using Probability Adjustment Method

In password guessing, the Markov model is still widely used due to its simple structure and fast inference speed. However, based on random sampling generate passwords has problem of a high repetition rate, which leads low cover rate. The enumeration lower rate for high-probability passwords, it deterministic algorithm that always generates same in order, making vulnerable attack. We design dynamic distribution mechanism method. This enables probability be dynamically adjusted tend toward uniform strictly during generation process. apply propose model. Through comparative experiments RockYou dataset, we set optimal adjustment degree α. Compared with without mechanism, reduced from 75.88% 66.50% increased 37.65% 43.49%. addition, had highest passwords. Finally, avoided lack algorithm, when was run five times, reached almost as OMEN.