TY - GEN
T1 - Electro-optical DSP of tera operations per second and beyond (Extended abstract)
AU - Tamir, Dan E.
AU - Shaked, Natan T.
AU - Wilson, Peter J.
AU - Dolev, Shlomi
PY - 2008/12/1
Y1 - 2008/12/1
N2 - We present a vector-by-matrix multiplier architecture incorporating the high potential of optical computing within electronics. This architecture stems from advances in optical switching technology, optical communication, and laser on silicon, and overcomes previous bottlenecks implied by the speed of transferring information to the optical vector-by-matrix multiplier. Based on this architecture, we present in detail a feasible electro-optical DSP co-processor that can obtain more than 16-Tera integer operations per second. The use of the new architecture for several principal DSP applications is detailed, showing a significant improvement of at least two orders of magnitudes, over existing DSP technologies. Examples of possible applications, including motion estimation engine, string matching, and geometry engine systems, are provided. In addition, the architecture enables an improvement of previous solutions to bounded NP-complete problems by extensively reducing the size of the solver, while preserving an efficient computation time.
AB - We present a vector-by-matrix multiplier architecture incorporating the high potential of optical computing within electronics. This architecture stems from advances in optical switching technology, optical communication, and laser on silicon, and overcomes previous bottlenecks implied by the speed of transferring information to the optical vector-by-matrix multiplier. Based on this architecture, we present in detail a feasible electro-optical DSP co-processor that can obtain more than 16-Tera integer operations per second. The use of the new architecture for several principal DSP applications is detailed, showing a significant improvement of at least two orders of magnitudes, over existing DSP technologies. Examples of possible applications, including motion estimation engine, string matching, and geometry engine systems, are provided. In addition, the architecture enables an improvement of previous solutions to bounded NP-complete problems by extensively reducing the size of the solver, while preserving an efficient computation time.
KW - Bounded NP-Complete algorithms
KW - Digital signal processing
KW - Optical computing
KW - Parallel processing
KW - Traveling sales Person problem
UR - http://www.scopus.com/inward/record.url?scp=70349863224&partnerID=8YFLogxK
U2 - 10.1007/978-3-540-85673-3_5
DO - 10.1007/978-3-540-85673-3_5
M3 - Conference contribution
AN - SCOPUS:70349863224
SN - 3540856722
SN - 9783540856726
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 56
EP - 69
BT - Optical Supercomputing - First International Workshop, OSC 2008, Proceedings
T2 - 1st International Workshop on Optical Supercomputing, OSC 2008
Y2 - 26 August 2008 through 26 August 2008
ER -