A data obfuscation based on state transition graph of mealy automata

Abstract

Mealy automata model can be used to obfuscate constants and strings in programs, as to the obfuscation, the structure of state transition graph of mealy machine is simple and easy to test. To solve this problem, a data obfuscation based on state transition graph of mealy machine is proposed. With iteration of state transition graph of mealy machine, redundant states, transition functions and output functions based on probability are added into the graph, and then constants and strings in programs are obfuscated by the mealy machine. Analysis and experiment validated that redundant states and transition functions can increase the complexity of the structure of state transition graph. Output functions based on probability can increase the randomization of output obfuscated data. Obfuscation can be effective to improve the performance of mealy machine to resist static and dynamic reverse analysis. © 2014 Springer International Publishing Switzerland.