As　an　effective　supplement　to　terrestrial　communications,　satellite　communications　play　an　important　role　in　global　communications.　However,　in　satellite　communication　networks,　communication　resources　are　limited,　data　transmission　links　are　unstable,　and　satellite　communication　has　signal　attenuation　and　transmission　delay.　How　to　enable　stable　and　efficient　data　transmission　is　a　major　challenge　in　satellite　communication　networks.　Data　coding　plays　an　important　role　in　satellite　communication　since　it　avoids　the　need　of　frequent　feedbacks　in　the　transmission　process,　ensures　reliable　communication,　and　reduces　receiving　delays.　In　a　practical　satellite　network,　when　the　transmission　channel　condition　is　unstable,　using　fixed　encoding　packet　size　may　lead　to　prolonged　delay　and　reduce　the　effective　data　transmission　rate.　Therefore,　it　is　particularly　important　to　adaptively　adjust　the　size　of　encoded　packets　according　to　the　file　size　and　network　environment.　In　this　paper,　we　propose　a　satellite　network　data　transmission　algorithm　based　on　adaptive　fountain　codes.　The　data　is　transmitted　in　LT　codes.　We　propose　two　algorithms,　including　a　two-side-search　dynamic　iterative　algorithm　and　adaptive　packet　size　encoding　transmission　algorithm　to　find　the　optimal　encoding　packet　size　and　grouping　to　minimize　the　total　delay　and　improve　the　effective　data　transmission　rate.　The　simulation　results　show　that　the　proposed　algorithm　has　a　smaller　delay　and　a　higher　effective　data　transmission　rate　than　those　using　the　fixed　packet　size　algorithm.　When　the　bit　error　rate　is　high,　it　can　also　maintain　a　low　delay　and　a　high　effective　data　transmission　rate.　©　2021　IEEE.