<![CDATA[This paper describes the idea of designing and implementing an field programmable gate array (FPGA) based system on chip (SoC) for forest fires prediction (FFP). The FFP in the proposed system is based on the decision tree (DT) algorithm implemented as an intellectual property (IP) in the FFP Zynq-SoC architecture that constitutes the processing part of a smart sensor node. This latter processes the collected meteorological data and takes decision locally at sensor node level; without having to send massive data to the base station for decision. The performance of the decision-tree software classifier in terms of accuracy and recall are about 75% and 0.88, respectively. The hardware implementation results of the DT -based forest fires prediction SoC show that the developed DT IP core is area and power efficient.]]>
Copyrights © 2026
