Neural Chip SAND in online data processing of extensive air showers Original Research Article

The neural chip SAND (Simple Applicable Neural Device) was designed to accelerate computations of neural networks at a very low cost basis, due to the fact that only few peripheral chips are necessary to use the neural network chip in applications. Four SAND-chips were implemented on one PCI-board. The board is highly usable for hardware triggers in particle physics. The performance of a SAND-PCI-board is 800 Mega Connections per Second due to four neuro-chips, each with four parallel 16 bit multipliers and 40 bit adders. SAND is able to implement feedforward neural networks with a maximum of 512 input neurons and three hidden layers. Kohonen feature maps and radial basis function networks may be also calculated. The application of the SAND-PCI-board is proposed for cosmic ray physics to allow online analysis of extensive air showers.