In　order　to　provide　the　users　with　reliable　wireless　communication　service　in　high-speed　mobility　scenarios,　we　need　to　obtain　channel　state　information　by　channel　estimation.　However,　the　wireless　channel　presents　a　characteristic　of　dynamic　change　in　high-speed　mobility　environments.　It　will　bring　great　challenge　to　channel　estimation.　Aiming　at　the　high-speed　railway　communication　system,　the　issue　of　the　fast　time-varying　channel　estimation　is　investigated　in　this　paper.　The　impulse　response　of　the　fast　time-varying　channel　is　modeled　as　the　product　of　the　basis　functions　and　the　coefficients　by　introducing　a　basis　expansion　model　(BEM).　Meanwhile,　the　comb　pilot　clusters　are　inserted　in　frequency　domain.　The　coefficients　of　basis　functions　are　derived　by　the　least-square　estimation　criterion　so　as　to　realize　the　estimation　of　the　channel　impulse　response.　Simulation　results　show　that　optimization　generalized　complex　exponential　BEM　(OGCE-BEM)　has　the　smallest　normalized　mean　square　error　among　the　various　types　of　BEMs.　©　2018,　ICST　Institute　for　Computer　Sciences,　Social　Informatics　and　Telecommunications　Engineering.