A Theory for Multiresolution Signal Decomposition: The Wavelet Representation
IEEE Transactions on Pattern Analysis and Machine Intelligence
Hi-index | 2.88 |
Wavelet was applied to analyze the effects of process parameters on the directional variations of etching profile nonuniformity. The profile nonuniformity was characterized by using a discrete wavelet transformation (DWT). The DWT enabled to separately evaluate parameter effects on the vertical profile nonuniformity (VPN) and the horizontal profile nonuniformity (HPN). The process parameters examined include radio frequency source power, bias power, SF6 flow rate, and substrate temperature. With increasing the bias power, the VPN increased, but the HPN decreased. This was similar for variations in the SF6 flow rate. For variations in the source power, either VPN or HPN varied inconsistently. Despite the inconsistency, the opposite effect of the bias power was maintained. It was therefore identified that the effects of all parameters but the substrate temperature on the VPN and HPN are opposite. The presented technique is general since it can be applied to examine other plasma-processed surfaces.