To deal with the emitter identification problem in Multi-Function Radar (MFR) based on Stochastic Context-Free Grammar (SCFG) model, a MFR emitter identification method based on Stochastic Infinite State Automata (SISA) was proposed on the basis of syntactic modeling. The grammar production rules in "Mercury" MFR control module and the characteristic production rules in "Mercury" MFR system were used in this method to reconstruct an SCFG, which was further used to construct an SISA for identification subsequently. Theoretical analysis and simulation results show that the proposed method can realize MFR emitter identification. Within a certain range, the average recognition rate can be improved by adding the amount of grammar production rules, and the identification performance is superior to Stochastic Push-Down Automata (SPDA) constructed by SCFG. The experimental results validate the reliability and effectiveness of the proposed method.