Novel VLSI implementation of Peano-Hilbert curve address generator

This paper presents a fast algorithm for generating Hilbert address for hardware implementation with low storage requirement. This work avoids the use of recursive functions as compared with Quinqueton´s work, and eliminates complicated bit manipulations as proposed by Butz, and does not use any look-up-tables as implemented by Kamata. Each address can be obtained in one clock cycle by one-to-one mapping using a simple incremental counter and cascading of multiplexers. The merit of our method is that it achieves very high speed when computing the Hilbert address which requires little memory storage.