We have recently submitted a paper to an analytical chemistry (AC) journal. The topic is a combined baseline/background estimation and denoising, and it is applied to chromatograms. As it was evaluated (on synthetic signals) on Gaussian noise, one of the reviewers suggested that in recent detectors, the noise follows more generally Poisson distributions, and would be related to the square root of the intensity of the data (including siganl and background).
I did not find revelant references: how quantitative, on which type of AC signals? Any suggestions?
Using a mass spectrometer as a detector the signal does follow a Poission distribution due to the events where zero, one, or more than two ions collide with the photomultiplier tube, that is "how many times did zero ions arrive at your detector during a time interval, how many times did one ion arrive during a time interval, how many times did tow ion arrive, etc."
Maybe these will be helpful if I'm answering your question correctly:
http://bioinformatics.oxfordjournals.org/content/24/8/1070.full
http://openi.nlm.nih.gov/detailedresult.php?img=2577660_1471-2105-9-423-5&req=4
Good luck in your search
@David Jensen and @M. Farooq Wahab : thank you very much for the sources that provide me with interesting starting points
Your question has been aptly and rightly answered By Dr. M. Farooq Wahab, The reviewer of you paper has given the right suggestion. Try to search a little more painstakingly and you will get many references on this. In fact it is applied to analytical techniques like FTIR spectroscopy.
- Badges
- Science method
- Similar topics
- Chemistry
- General Biochemistry
- Biochemical Techniques
- Analytical Chemistry