DocumentCode :
696598
Title :
NeuroMatrix® NM6403 DSP with Vector/Matrix engine
Author :
Fontine, Dmitri ; Tchernikov, Vladimir ; Vixne, Pavel ; Chevtchenko, Pavel
Author_Institution :
Res. Center MODULE, Moscow, Russia
fYear :
2000
fDate :
4-8 Sept. 2000
Firstpage :
1
Lastpage :
4
Abstract :
The paper describes the architecture of the NeuroMatrix® NM6403 DSP designed for image processing, signal processing and neural networks emulation [1,2]. The paper includes a brief description of the processor structure and its instruction set. The NM6403 is the first DSP based on NeuroMatrix® Core (NMC) comprises an original 32-bit VLIW RISC processor and a 64-bit SIMD Vector co-processor (VCP). In contrast to other modern general purpose DSPs and microprocessors with SIMD units such as: Texas Instruments c64xx, Intel Pentium MMX, Motorola AltiVec PowerPC G4 and Analog Devices TigerSHARC, the new DSP performs variable bit-length vector/matrix arithmetic, logic and saturation operations. The main NMC operation is matrix by vector multiplication. The NM6403 supports shared memory mode for two 64-bit external data buses. Two byte-width communication ports simplify the multiprocessor systems design. The NM6403 has been designed by RC "Module" (www.module.ru) and produced by Samsung 0.5μm CMOS technology. The peak performance - up to 14.400 MMACs (million multiplication and accumulations per second) has been achieved at a 50MHz clock rate, 3.3V operating voltage and PBGA256 package.
Keywords :
Texas Instruments computers; digital signal processing chips; instruction sets; matrix multiplication; parallel architectures; reduced instruction set computing; vector processor systems; 32-bit VLIW RISC processor; 64-bit SIMD VCP; 64-bit SIMD vector coprocessor; 64-bit external data buses; Intel Pentium MMX; MMAC; Motorola AltiVec PowerPC G4; NM6403; NMC; PBGA256 package; Samsung 0.5μm CMOS technology; Texas instrument c64xx; analog device TigerSHARC; clock rate; frequency 50 MHz; image processing; instruction set; microprocessors; million multiplication-and-accumulation-per-second; multiprocessor system design; neural network emulation; neuromatrix® NM6403 DSP; neuromatrix® core; operating voltage; processor structure; shared memory mode; signal processing; two byte-width communication port; vector multiplication; vector-matrix engine; voltage 3.3 V; Clocks; Convolution; Digital signal processing; Image processing; Macrocell networks; Reduced instruction set computing; Vectors;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Signal Processing Conference, 2000 10th European
Conference_Location :
Tampere
Print_ISBN :
978-952-1504-43-3
Type :
conf
Filename :
7075219
Link To Document :
بازگشت