High-throughput comparative genomic hybridization arrays have recently been developed to detect DNA copy number (DCN) aberrations. The DCN data from these arrays is often very noisy and thus calls for appropriate denoising methods. It has been recognized that a) the physical distances between adjacent probes are not uniform and that b) there are often outliers in the data. Previously proposed denoising methods for DCN data did not consider these two issues at the same time. In this paper, we address the two issues simultaneously in a new wavelet denoising scheme, called rSWTi (robust Stationary Wavelet Transform based denoising with interpolation), which extends the traditional stationary wavelet denoising approach. Empirical results on synthetic data showed that our method outperformed other methods by large margins as measured in the root mean squared error.