A Theory for Multiresolution Signal Decomposition: The Wavelet Representation
IEEE Transactions on Pattern Analysis and Machine Intelligence
Wavelets and subband coding
A scalable architecture for discrete wavelet transform
CAMP '95 Proceedings of the Computer Architectures for Machine Perception
Scalability of 2-D wavelet transform algorithms: analytical and experimental results on MPPs
IEEE Transactions on Signal Processing
IEEE Transactions on Signal Processing
Dedicated Circuits for the Generation of Windows in Image Processing Architectures
Journal of VLSI Signal Processing Systems
A Programmable Parallel VLSI Architecture for 2-D Discrete Wavelet Transform
Journal of VLSI Signal Processing Systems - Parallel VLSI architectures for image and video processing
VLSI Implementation for One-Dimensional Multilevel Lifting-Based Wavelet Transform
IEEE Transactions on Computers
Integrated Computer-Aided Engineering
A dedicated DSP architecture for discrete wavelet transform
Integrated Computer-Aided Engineering
Algorithms and architectures for 2D discrete wavelet transform
The Journal of Supercomputing
Hi-index | 0.00 |
In this paper, an efficient semi-systolic array architecture forseparable 2-D Discrete Wavelet Transform (DWT) is introduced. Thesemi-systolic array is applicable to any convolution that requires anarbitrary subsampling function. The semi-systolic array presents a betterimplementation of the convolution function of DWT. This kind ofimplementation offers a higher efficiency compared to regular systolicimplementation when applied for 2-D DWT. The architecture has an efficiencyof at least 91% which increases proportional to the number of octaveswith no change in the architecture design except for minor modifications tothe control logic and memory size. The propose architecture is scalable fordifferent size of filter and different number of octave. The communicationrouting is minimum since data transfers are limited to immediate neighboringprocessors. The components of the architecture are fairly regular andconsist of minimum number of computational units which makes it a goodcandidate for VLSI implementation.