High Speed Neural Network Chip on PCI-Board
نویسندگان
چکیده
A novel neural chip SAND (Simple Applicable Neural Device) is described. It is highly usable for massive parallel neural network computations. The chip is optimized for a high input data rate (50 MHz, 16 bit data) at a very low cost basis. The performance of a single SAND chip is 200 MOPS due to four parallel 16 bit multipliers and 40 bit adders working in one clock cycle. The chip 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. Up to four chips are implemented on a PCI-board for industrial applications and on a VME board for trigger analysis in
منابع مشابه
Implementing very high-speed hierarchical MLP-based classification systems in real-time industrial environments
Using a proper combination of special HW and SW, it is possible to exploit the capabilities of the MultiLayer-Perceptron-based “tree architecture” even in very high-speed industrial classification problems. In particular, the paper shows as, adopting a specially developed 128 MCPS ASIC Neural coprocessor (able to synthesize a whole 64 input – 128 hidden – 64 output MLP on-chip), it has been pos...
متن کاملNeural Network Chips for Trigger Purposes in High Energy Physics
Two novel neural chips SAND (Simple Applicable Neural Device) and SIOP (Serial Input Operating Parallel) are described. Both are highly usable for hardware triggers in particle physics. The chips are optimized for a high input data rate at a very low cost basis. The performance of a single SAND chip is 200 MOPS due to four parallel 16 bit multipliers and 40 bit adders working in one clock cycle...
متن کاملA multi-chip pulse-based neuromorphic infrastructure and its application to a cortical model of orientation selectivity
The growing interest in pulse-mode processing by neural networks is encouraging the development of hardware implementations of massively parallel networks of Integrate-and-Fire (I&F) neurons distributed over multiple chips. Address-event representation (AER) has long been considered a convenient transmission protocol for spike based neuromorphic devices. One missing, long-needed feature of AER-...
متن کاملA Neurocomputer Board Based on the ANNA Neural Network Chip
A board is described that contains the ANN A neural-network chip, and a DSP32C digital signal processor. The ANNA (Analog Neural Network Arithmetic unit) chip performs mixed analog/digital processing. The combination of ANNA with the DSP allows high-speed, end-to-end execution of numerous signal-processing applications, including the preprocessing, the neural-net calculations, and the postproce...
متن کاملOptimization of Material Removal Rate in Electrical Discharge Machining Alloy on DIN1.2080 with the Neural Network and Genetic Algorithm
Electrical discharge machining process is one of the most Applicable methods in Non-traditional machining for Machining chip in Conduct electricity Piece that reaching to the Pieces that have good quality and high rate of machining chip is very important. Due to the rapid and widespread use of alloy DIN1.2080 in different industry such as Molding, lathe tools, reamer, broaching, cutting guillot...
متن کامل