Optimal　power　allocation　(PA)　strategies　can　make　a　significant　rate　improvement　in　secure　spatial　modulation　(SM).　Due　to　the　lack　of　secrecy　rate　(SR)　closed-form　expression　in　secure　SM　networks,　it　is　hard　to　optimize　the　PA　factor.　In　this　paper,　two　PA　strategies　are　proposed:　gradient　descent　(GD),　and　maximum　product　of　signal-to-interference-plus-noise　ratio　(SINR)　and　artificial-noise-to-signal-plus-noise　ratio　(ANSNR)　(Max-P-SINR-ANSNR).　The　former　is　an　iterative　method　and　the　latter　is　a　closed-form　solution.　Compared　to　the　former,　the　latter　is　of　low-complexity.　Simulation　results　show　that　the　proposed　two　PA　methods　can　approximately　achieve　the　same　SR　performance　as　the　exhaustive　search　method　and　perform　far　better　than　three　fixed　PA　ones.　With　extremely　low　complexity,　the　SR　performance　of　the　proposed　Max-P-SINR-ANSNR　performs　slightly　better　and　worse　than　that　of　the　proposed　GD　in　the　low　to　medium,　and　high　signal-to-noise　ratio　regions,　respectively.