• 1024 neurons in parallel on a chip

• Learn and recognize patterns of up to 256 bytes
• Classify patterns into up to 32,768 categories
• Automatic adaptive model generator
• Choice of Radial Basis Function (RBF) or K-Nearest Neighbor (KNN) classifiers
• Recognition time independant of the number of neurons
• Possible segmentation of the neurons into up to 128  contexts (per feature, per sensors, etc)
• Save and restore knowledge
• Simple RTL protocol with less than 2 dozen registers
• Low cost, small footprint, low power consumption
• Ability to interconnect multiple chips with a simple bus of 28 lines

This plot of the CM1K ASIC shows that it is a set of 1024 identical cells, or neurons. They are connected in parallel through a bus of 33 lines.

Download

• CM1K presentation of the classifier
• Examples of system architecture
• Technology Reference Guide
• CogniMem CM1K datasheet
• Presentation: Speed performance
• White paper: CM1K disruptive performance for DSPs