Map　matching　is　the　process　of　matching　GPS　points　to　roads.　Local　methods　and　incremental　methods　usually　have　faster-running　speed,　however,　the　performance　of　these　methods　may　be　insufficient　for　complex　road　networks　and　data　with　significant　error.　Global　methods　can　get　better　results　in　complex　situations,　while　most　of　them　handle　high-frequency　GPS　data　slowly.　In　order　to　solve　these　problems,　a　weight-based　algorithm(FWMM)　is　developed　in　this　paper.　We　introduce　a　new　method　of　weight　fusion　to　avoid　the　influence　of　dimension　and　adopt　Particle　Swarm　Optimization(PSO)　for　parameter　estimation.　In　order　to　accelerate　the　matching　process　of　data　with　a　high　sampling　rate,　we　designed　a　mechanism　which　can　adaptively　accelerate　high-frequency　data　piece.　Thus　this　method　can　be　more　accurate　than　existing　methods　mentioned　above.　©　2019　Association　for　Computing　Machinery.