sayaf,
Thanks for including the image -- it's a big help. Now I'm not the greatest expert at filtering, but to my eye you're getting a pretty reasonable approximation. I don't know how much better you can get when you only have about 8 PWM cycles for defining a half-cycle of the sine wave you want to see. At least, it looks pretty good if you have to handle a very general case of modulation.
Can the modulation always be considered a pure sine wave with a constant frequency throughout your sampling time? If so, then you can use that knowledge to your advantage -- for example, pass a few cycles of your filtered signal into one of the tone-extraction vi's that ship with LabVIEW. I'm sure there are other good (probably better) methods as well.
-Kevin P.
CAUTION! New LabVIEW adopters -- it's too late for me, but you *can* save yourself. The new subscription policy for LabVIEW puts NI's hand in your wallet for the rest of your working life. Are you sure you're *that* dedicated to LabVIEW? (Summary of my reasons in this post, part of a voluminous thread of mostly complaints starting here).