Computação paralela com OPENCL aplicada à predição de perda de percurso utilizando equações parabólicas e o método das diferenças finitas

Abstract

The final project in question deals with the application of parallel computing with OpenCL for the prediction of path loss using parabolic equations and the finite differences method. OpenCL is a high-level programming language that allows the creation of applications that run in parallel on different types of computing devices, such as CPUs, GPUs, and other specialized processors. The goal of this final project is to increase the processing speed of path loss prediction using parallel computing, so that it can be performed more efficiently and quickly. To achieve this, parabolic equations and the finite differences method were implemented in OpenCL, allowing these calculations to be performed in parallel on multiple processing cores. The results show that the use of parallel computing with OpenCL allowed a significant increase in the processing speed of path loss prediction, making this task more efficient and faster. Additionally, the use of OpenCL allowed the use of different types of computing devices, expanding the possibilities of implementation of this solution in different contexts.