Performance comparison of DWT scheduling alternatives on programmable platforms

The Discrete Wavelet Transformations (DWT) are data intensive algorithms. Energy dissipation and execution time of such algorithms heavily depends on data memory hierarchy performance when programmable platforms are considered. Existing filtering operations scheduling alternatives for the 1D-DWT, employ different levels of data access locality. However locality of data references, usually comes at the expense of complex control and addressing. In this paper, the two main scheduling alternatives for the 1D-DWT are compared in terms of energy and speed. Additionally, we describe and evaluate the effect of an in-place mapping scheme, which minimizes memory requirements and improves locality of data reference, for the 1D-DWT. As execution platform, two commercially available general purpose processors are used