Products

Neural Networks
Dana
	Product	Article number
	DANA Board
	DANAboard IC1	PR-00300-00
	DANAboard IC1 mit Analogteil	PR-00300-05
	DANAboard IC3	PR-00300-10
	DANAboard IC3 mit Analogteil	PR-00300-15
	IC GC-Dana23 (ASIC)	PR-00301-01
Information Sheets	88000-ib-e-dana23.pdf
	88100-ib-e-danasim.pdf

Main Features:

The DANA23 (Digital Adaptive Neuro ASIC) neurochip implements a universal neurocontroller. The chip contains a complete feed-forward network with integrated weight storage. Furthermore, there is also a modified backpropagation
learning algorithm integrated into the ASIC.

Specifications (preliminary)

Technology: 0,35 μ CMOS
Chip size: 35 mm2
Supply voltage: 3,3 V
Clock frequency: 33 MHz
Power consumption: 0,5 W
Package type: CQFP 208
Dimensions: 30,6 mm x 30,6 mm x 4,1 mm
Processing speed at 33MHz: 45MCUPS1

Applications

Typical application areas for the neuro-ASIC DANA23 are:
- pattern recognition, pattern classification
-microsystem technology with optical, acoustic or movement sensors
- knowledge-based control of medical devices
- prognostic systems

The DANAsim programme is a simulation tool for neural networks developed at the GEMAC mbH and specifically adopted for the DANA23 neuro-ASIC.
With DANAsim, feed-forward networks with up to four layers (one input layer, one or two hidden layers, and one output layer) can be processed. The maximum number of neurones per layer is 64. All neurones have a quadratic activation function.
Learning and testing data for the neural network are to be supplied in the form of an appropriate pattern-file. All learning parameters for a modified form of the standard back-propagation algorithm can be adjusted easily in the program menus.

It is possible to connect the simulator to the DANA23 neuro-ASIC PCI-board, developed at GEMAC mbH. In this way, neural networks developed using the simulator can be tested on corresponding hardware.

 © 2012 GEMAC - Gesellschaft für Mikroelektronikanwendung Chemnitz mbH