|
|
| |
CM-EB1K Pattern Recognition Board for Data, Signal or Video |
| |
CM-EB1K Pattern Board for Data, Signal or Video offers developers and OEMs a comprehensive platform to evaluate the CogniMem neural network. The board features a CogniMem chip with 1024 neurons, an Actel FPGA accessible to programmers and a digital input bus for easy connectivity to external sensors. User I/Os are implemented on the board in a flexible variety of configurations, including an I2C serial bus, an RS232 bus and 8 uncommitted general purpose I/Os brought to header pins. The board can piggy-back 3 additional CogniMem expansion modules totaling 6144 neurons.
|
 |
|
| |
CM-EB1K was designed to provide you a quick-start evaluation of a CogniMem neural network, the knowledge bases built by the neurons and the speed performance of its high-speed recognition engine. It will help to shorten your design cycle and accelerate your time to market.
|
| Specifications |
| |
| |
Neural network |
|
Classify pattern vectors of up to 256 bytes |
|
Up to 32768 categories |
|
Classification status in 1 clock cycle |
|
Category readout in 36 clock cycles per |
| |
up firing neuron from smallest distance |
| |
and (equal to 3 microsec at 27 Mhz) |
|
Radial Basis Function or K-Nearest
|
| |
neighbor classifier |
|
Automatic model generator |
|
Expandable through neuron expansion |
| |
modules |
I/O buses |
|
Miniature USB Hi Speed (480 Mbps) |
|
I2C serial interface (100-400 kbit) |
|
Serial output (115,200 baud) |
|
8 parallel outputs (LVTTL 16mA) |
|
Save project to Flash memory |
|
|
High-speed recognition engine |
|
|
|
input clock signal (up to 27 Mhz) and
|
| |
vector valid signal |
|
Output category with the best match 3 us
|
| |
after receipt of the last vector data |
|
line valid input signal |
|
region of interest as small as 16x16
|
| |
pixels and as large as a full frame |
|
Built-in signature extraction |
|
Connector compatible with Micron sensor
|
| |
demo board |
FPGA Program |
|
Optional core modules (input and output |
| |
data conditioning, signature extraction, |
| |
decision logic, etc. |
Mechanical and Electrical |
|
3.3v @ <250 mA |
|
55กม66mm |
|
|
| |
Ordering information |
| |
|
CM-EB1K (include one CM-E1K module, Reference Design, Verilog example design, |
| |
CogniMem development library, documentation and Easy Trainer software) |
|
Optional additional: CM-E2K expansion module with 2048 neurons. |
| For more information, you can download some documentations from here. |
|
|
|
