Identifying Radar Targets using the GMDH Deep Neural Network

Radar is an electromagnetic device used to detect and determine the position of targets. The most basic task of radar is to extract information about the target by measuring the electromagnetic field characteristics of the return waves from the target. The radar environment of each country is one of the security and strategic areas of each country. Maintaining the security of this environment and identifying its goals can be one of the important requirements. Challenges and problems such as inaccuracy and inaccuracy of detection and high error are raised in the detection of radar targets. Various methods have been proposed so far, such as techniques based on target natural intensification frequencies, reversible signal polarization, machine learning methods, etc., to detect radar targets. Despite the many uses of these methods, they have not yet been able to meet the challenges of radar.  Therefore, in this paper, we have identified radar targets using the GMDH Deep Learning Algorithm. By simulating the proposed method and comparing it with other methods such as RIN, SAE, SCAE, SDAE, CNN, LSVM, K-SVD, the average has improved by 5%.