Mapping the integrated emission from redshifted neutral hydrogen is an emerging and completely independent way of optical surveys to measure Baryonic Acoustic Oscillations (BAOs) at low redshifts. The use of BAOs as a standard ruler is now a well established technique to probe dark energy. The observations of the HI signal will be contaminated by instrumental noise, atmospheric and Galactic foregrounds, which are expected to be at least four orders of magnitude brighter than the HI signal we wish to detect. We will present these potential contaminants and systematic effects. Then, we will introduce how to simulate the kind of data which would be produced in a single dish intensity mapping experiment like the BINGO experiment. With a model of the observations of this instrument, we will discuss its ability to separate the HI signal from foregrounds, instrumental and atmospheric 1/f noise.