The　problem　of　robust　passive　filtering　for　a　class　of　uncertain　networked　system　with　time　varying　delays　and　missing　measurements　is　studied.　The　random　delays　in　the　system　are　assumed　to　be　bounded　but　unknown.　The　missing　measurements　are　stochastic　and　described　by　a　binary　switching　sequence　satisfying　Bernoulli　distribution.　A　sufficient　condition　for　existence　of　a　full-and　reduced-order　filter,　such　that　the　filter　error　system　is　exponentially　mean-square　stable　and　a　prescribed　passive　performance　is　guaranteed,　is　presented　via　linear　matrix　inequalities　(LMIs).　Finally　the　effectiveness　of　the　proposed　approach　is　illustrated　by　a　numeric　example.