prediction　of 　satellite　clock　bias　based　on　wavelet　transform　andleast　squares 　support　vector　machines

this　paper 　proposes 　a　novel　method　for 　prediction　of 　satellite 　clock 　bias scb based　onwavelet　transform　and 　least 　squares 　support 　vector　machineslssvm．the　method　first 　decomposesscb　time 　series 　into 　different 　fre quency　components 　through　the　wavelet　transform　al gorithm．thenaccording　to 　the　different 　characteristics　of 　the　decomposed　componentsthe　proper　lssvm　are 　constructed 　to　predict 　the 　components．finall ythe 　final 　prediction　value　of 　scb　is 　yielded 　by　the 　lineari ty　super position 　for 　the 　res pective　predicted 　result．the 　experiment 　shows 　that 　the　proposed　method can 　improve 　the　prediction 　accurac y　of 　scb　when　the　quadratic　pol ynomial qp modelgm11model　and　lssvm　model