In　the　packet-based　transmission　of　data　over　a　network　with　limited　bandwidth　or　temporary　sensor　failure,　measurement　data　may　be　missing.　The　H-two　filtering　problem　for　an　uncertain　discrete-time　system　with　missing　measurement　data　is　considered.　The　uncertain　parameters　in　the　system　matrices　are　assumed　to　belong　to　a　given　convex　bounded　polyhedral　domain.　The　occurrence　of　missing　measurement　data　is　random　and　is　assumed　to　be　a　Bernoulli　distributed　sequence　with　known　probability.　Sufficient　conditions　are　derived　in　terms　of　linear　matrix　inequalities　(LMIs)　for　the　existence　of　a　full-and-reduced-order　filter　ensuring　a　prescribed　H-two　performance　for　all　admissible　uncertainties　and　the　possible　data　missing.　Numerical　example　is　provided　to　demonstrate　the　feasibility　of　the　proposed　method.