Meteorological Prediction Implemented on Field-Programmable Gate Array

Abstract

In this work, a temperature predictor has been designed. The prediction is made by a multilayer perceptron neural network. Initially, the floating-point algorithm was evaluated. Afterward, the fixed-point algorithm was designed on a field-programmable gate array. The architecture was fully parallelized, and a maximum delay of 74 ns was obtained. The design tool used is a Xilinx system generator.