If you want to do better than that, I think you're going to have to zero each individual pulse to its own (trough-to-trough) baseline, interpolating the zero for each individual data point between troughs. This won't give you real-time data, you'll always be one pulse behind, unless you want to collect all the data and process the whole trace at once. It sounds like a lot of work, but it wouldn't take the computer much time, you already have in your favor the facts that (1) you have a very regular waveform, with a well-regulated period, and (2) the troughs are well separated, with a clearly identified (read: huge) peak in between.
So you can do this in two ways.
One peak at a time: If you identify the minima between two peaks (the peaks don't even have to be the exact maxima, just pretty close), calculate the straight line between them, and zero each data point to the corresponding point on the calculated line. Continue this for the rest of the trace, you've got it. The absolute peak heights may be off a couple of percent or so, I don't know if this will be significant or not.
The whole trace at once:
Find all of the minima for the whole trace, as above. You could do a "general polynomial fit" with a high enough order that you hit each of the minima (within whatever tolerance you decide) and use that equation to zero each data point.
It looks like you've got enough short-term noise in your trace that either of these should give you a good baseline-subtracted data set to work further from.
Cameron
To err is human, but to really foul it up requires a computer.
The optimist believes we are in the best of all possible worlds - the pessimist fears this is true.
Profanity is the one language all programmers know best.
An expert is someone who has made all the possible mistakes.
To learn something about LabVIEW at no extra cost, work the online LabVIEW tutorial(s):
LabVIEW Unit 1 - Getting Started</ a>
Learn to Use LabVIEW with MyDAQ</ a>
