DCT processor architecture based on computation sharing

Soonkeon Kwon1, Jongsun Park1, K. Roy1
1School of Electrical and Computer Engineering, Purdue University, West Lafayette, IN, USA

Tóm tắt

In this paper, we present a new discrete cosine transform (DCT) processor architecture using computation sharing multiplication (CSHM). We introduce a computation sharing multiplier based DCT architecture to achieve image quality and hardware complexity trade-off and analyze the performance. Comparison of the performance, area and power consumption with a DA (distributed arithmetic) based DCT architecture is performed. The result shows that the proposed architecture improves power consumption by 14% and area by 41% with acceptable image quality degradation.

Từ khóa

#Discrete cosine transforms #Computer architecture #Image quality #Signal processing algorithms #Energy consumption #Image coding #Matrix decomposition #Power engineering computing #Computational Intelligence Society #Hardware

Tài liệu tham khảo

bhaskaran, 1996, Image Video Compression Standards Algorithms and Architectures muhammad, 1999, Algorithmic and architectural techniques for low power digital signal processing blair, 1998, Design for the discrete cosine transform in VLSL, IEE Proc Comput Digit Tech, 145, 127, 10.1049/ip-cdt:19981911 10.1109/53.29648 park, 2000, Non adaptive and adaptive filter implementation based on sharing multiplication, IEEE International Conference on Acoustics Speech and Signal Processing(ICASSP) chen, 1977, A Fast Computational Algorithm for the Discrete Cosine Transform, IEEE Trans on Communications, com 25