Title :
New DCT computation algorithm for VLIW architecture
Author :
Ji, Li ; Ming, Li Xue
Author_Institution :
Sch. of Inf. Eng., Beijing Univ. of Posts & Telecommun., China
Abstract :
A new DCT computation algorithm is presented, which is based on an existing fast DCT algorithm and has fully utilized the parallelism of the VLIW (very long instruction word) architecture of DSP to achieve high speed performance. Simulation results show that this method can reduce computing cycles by more than 70% and 20%, compared with normal and fast DCT computation algorithms, respectively.
Keywords :
computational complexity; discrete cosine transforms; parallel architectures; video coding; DCT computation algorithm; VLIW architecture; discrete cosine transform; fast DCT; very long instruction word architecture; video coding; Computational modeling; Computer aided instruction; Computer architecture; Concurrent computing; Digital signal processing; Discrete cosine transforms; High performance computing; Parallel processing; VLIW; Video coding;
Conference_Titel :
Signal Processing, 2002 6th International Conference on
Print_ISBN :
0-7803-7488-6
DOI :
10.1109/ICOSP.2002.1180978
