|
NeuroMatrix® NM6405 DSP
NM6405 is a high performance DSP oriented RISC processor designed for real time data flow processing. The architecture is based on the advanced VLIW/SIMD NMC3 core, and consists of a 32/64-bit RISC processor and a 64-bit VECTOR co-processor. The co-processor supports vector/matrix operations with elements of variable bit length (US Pat. 6539368 B1).
Documentation
Features:
- 32/64-bit RISC processor core;
- 64-bit vector coprocessor (VECTOR COPROCESSOR);
- Four dual port internal SRAM banks (8ëx64 bit each);
- 1ëÈ64bit of instruction cache memory (ICACHE);
- Address generator unit (AGU);
- Two 64-bit external memory interfaces (LMI É GMI), that operate at speed up to 130 MHz;
- Two Byte communication ports with throughput each up to 150 MB per second;
- Eight general purpose IO ports (GPIO);
- JTAG port for debug and test purposes;
- Two independent DMA controllers;
- Internal/External interrupts controller (INTERRUPT);
- Two universal 32-bit timers (TIMER0 É TIMER1).
Main characteristics
- CMOS technology - 0.25µ;
- package - 576 BGA
- Clock frequency - 150 MHz;
- Power supply - 2,5 V (core) 3,3 ÷ (I/O buffers);
- Power consumption - less than 3,8 W;
- Ambient temperature: -55°C ... +85°C.
RISC processor
- Data width - 32 bit;
- Instruction width - 32 É 64 bit;
- Address space - 4GÈ32 bit;
- 3 scalar instruction per clock cycle (ALU operation, address modification input/output operation);
- Performance - 150 MIPS (450 MOPS).
Vector coprocessor
- Programmable data length from 2 to 64 bits (64bit length data words packed);
- Basic operation is integer data matrix multiplication by integer data matrix;
- Concurrent execution of 2 saturation operation with input data flow;
- Performance (íáó - Multiplication and Accumulation per clock cycle)
-
- 2 MAC for 32-bit data;
- 4 MAC for 16- bit data;
- 24 MAC for 8- bit data;
- 80 MAC for 4-Ò bit data;
- 224 MAC for 2- bit data.
Applications
- hydro- and radiolocation;
- IR and video processing;
- Artificial Neural net emulation;
- Navigation;
- CDMA É TDMA base stations;
- Vector and matrix computations.
The NM6405 processor is oriented for massive dataflow processing with short bit length signal samples.
The software design kit NM-SDK Version 3.0 includes an optimizing C++ compiler (ISO/IEC 14882:1998 standard), assembler, disassembler, linker, debugger and real-time DSP and NeuroMatrixR Processing Library (NMPL). The compiler adheres to the C++ standard, including templates, and uses the enhanced optimizing algorithms that allow increasing program execution speed and decreasing code size. The assembly language has an intuitive syntax and is close to high-level languages so it can simplify the development and understanding of source code for math-intensive real-time algorithms.
Single DSP Evaluation Board with PCI interface can be used for software design. |