Skip to Main Content
The number of processing elements (PE) can be reduces significantly using partial column structure to perform Discrete Trigonometric Transform (DTT) computation. Data reordering is required between stages (columns). A scalable interconnect network for both global and local for data reordering is presented in this paper. The network scalability is based on the transform size and the number of processing elements (PE). The structure will reorder data on the fly and eliminate the need of memory. This will help evaluating throughput vs. complexity (cost and area.). Detail analysis of hardware cost will be presented.