By "noisy" do you mean that the signal is poor (ratio of peak intensity to background intensity) or that there are many well-defined, measurable peaks (relative to the background, which should be flat) in the same area? The first one is a problem with your sample or your measurement, and the other one is just a consequence of having many similar vibrational energies within the same range. If you look through old spectroscopy papers, you can probably find assignments for each peak. Remember that it's not only stretching modes that show up in the IR, but also bending and deformation modes as well.