An Efficient Hardware Realization of Distributed Arithmetic Based Discrete Cosine Transform

Abstract

Objective: This paper presents an efficient design for the implementation of Discrete Cosine Transform (DCT) which is widely used in still and motion picture compression. In this work cyclic convolution and Grouped Distributed Arithmetic (GDA) technique are used. Method: In GDA, Read Only Memory (ROM) size could be reduced compared to conventional DA. For example, 1-D DCT of transform length N=13, the proposed design requires 31 Look-Up Tables (LUTs) while conventional DA requires 384 LUTs. Findings: From the synthesis results, it is found that proposed design has less area requirement and low power consumption compared to the conventional DA of same length. Conclusion: This paper proposes an efficient design for implementation of cyclic convolution based on modified GDA technique.