Title :
High-performance low-power inner product processor
Author :
Natter, William ; Margala, Martin ; Durdle, Nelson G.
Author_Institution :
Dept. of Electr. & Comput. Eng., Alberta Univ., Edmonton, Alta., Canada
Abstract :
This paper presents three alternative implementations of a high-performance, low-power inner product processor. Several previously published algorithms are examined, and modifications that reduce the average activity and computational time are proposed. In addition, various hardware implementations and evaluations of their power and performance are provided. The presented inner product forms an integral part of a complex shading architecture in a volume imaging system.
Keywords :
computer architecture; computer power supplies; digital arithmetic; image processing equipment; performance evaluation; power consumption; vectors; algorithm modifications; average activity; complex shading architecture; computational time; hardware implementations; high-performance low-power inner product processor; performance evaluation; power evaluation; volume imaging system; Arithmetic; Biomedical imaging; Computed tomography; Computer architecture; Energy consumption; Graphics; Hardware; Magnetic resonance imaging; Power engineering computing; Real time systems;
Conference_Titel :
Electrical and Computer Engineering, 1999 IEEE Canadian Conference on
Conference_Location :
Edmonton, Alberta, Canada
Print_ISBN :
0-7803-5579-2
DOI :
10.1109/CCECE.1999.807246
