With the extensively study on moving target defense, optimal strategy generation has become one of the key problems in current research. A novel of incomplete information Markov game theoretic approach to strategy generation for moving target defense is proposed to solve the existing problems. On the one hand, traditional matrix game structure and complete information assumption often fail to describe moving target defense confrontation accurately. To deal with this inaccuracy, moving target defense game model based on incomplete information Markov game theory is constructed by introducing moving attack surface and moving exploration surface concept, extending optimal strategy selection of moving target defense with incomplete information from mono-state or mono-phase to multi-stated and multi-phased. On the other hand, traditional models care little about defense cost in the process of optimal strategy generation. After comprehensively analyzing the impact of defensive cost and defensive benefit on strategy generation, an optimal strategy generation algorithm is designed to prevent the deviation of the selected strategies from actual network conditions, thus ensuring the correctness of optimal strategy generation. Finally, simulation and deduction experiments have been performed in a case study so as to confirm the feasibility and the effectiveness of the proposed approach.