A process based on discrete wavelet transforms is developed for denoising and baseline correction of measured signals from Coulter counters. Given signals from a particular Coulter counting experiment, which detect passage of particles through a fluid-filled microchannel, the process uses a cross-validation procedure to pick appropriate parameters for signal denoising; these parameters include the choice of the particular wavelet, the number of levels of decomposition, the threshold value and the threshold strategy. The process is demonstrated on simulated and experimental single channel data obtained from a particular multi-channel Coulter counter processing. For these example experimental signals from 20 µm polymethacrylate and Cottonwood/Eastern Deltoid pollen particles and the simulated signals, denoising is aimed at removing Gaussian white noise, 60 Hz power line interference and low frequency baseline drift. The process can be easily adapted for other Coulter counters and other sources of noise. Overall, wavelets are presented as a tool to aid in accurate detection of particles in Coulter counters.