FPGA implementation of a dedicated processor for temperature prediction

In this paper a dedicated processor based on Artificial Neural Networks (ANN) for predicting the daily maximum temperature at a particular location has been implemented. The network architecture and weights are determined with the help of software tools using the daily maximum temperature, pressure and humidity at the specified location as inputs to the network. The model is trained using the data of one year duration of Guwahati, India. The model is then implemented on Xilinx Virtex-6 FPGA and tested with the data. This approach provides results with a good accuracy.