Based on our research work in the last ten years, based on the processing of radar signals and the implementation of various radar systems for observing the Earth's surface, we will deepen our knowledge in this research project from radar design in Tera Hertz (THz). Waves in the THz field are very suitable for observing the surface and structures of objects that do not contain organic impurities, so it is of great importance in practice, as we can ensure safety, determine the quality of products and materials that can not be detected with the naked eye. a better quality of life through the implementation of modern technologies. With modern technologies, we can observe micro-cracks on materials and detect the presence of materials behind walls and under clothing. This area is very interesting and is developing very fast but the equipment needed to implement THz radar is very unaffordable. In cooperation with the Laboratory for Electro-Optical Systems at UM-FERI, we also have the opportunity to produce THz radar from standard commercial telecommunication electronic components. The idea is to transfer the knowledge from generating and focusing SAR data to the THz domain, research it and also implement such a radar. We will examine devices for generating and receiving THz signals. Here we see a contribution to THz radar design by generating THz waves using microfiber fiber processing and laser diode guidance. THz radar will be used for three-dimensional (3D) tomography. Our researched work can later be directed to applications for detecting the properties of observed objects. To this end, we will expand the field of t.i. deep learning based on neural networks. A novelty in this project will be the use of deep neural networks with complex data.
The goals of the project are two. The first is to explore the most appropriate technology with which THz radar can be made today. Here we will focus on innovative approaches to generating THz waves. The second goal of the research project is to analyze the interaction of THz waves with different objects and to lay the foundations for the processing of THz radio waves that we will be able to use in future real-world applications.
In this research project, we will first make a laboratory prototype of THz radar. The radar will be based on a similar principle of interaction of the optical signal with the photoconductive antenna. This can be achieved through the micro-processing of optical fibers and the guidance of laser diodes. The current state of the literature shows that it is not possible to generate any signal using THz waves. In the research project we will propose a special way of controlling the laser diode with which it will be possible to change the frequency of the THz signal and thus influence the shape of the output signal needed to implement THz radar with an artificial aperture. In this part, we will investigate the ways of generating a chirp signal using a laser and a laser diode and observe the shape of the received signal on the receiving antennas of photoconductive antennas. Here we will propose an antenna design by creating an antenna array in the THz domain, for signal generation purposes. In this way, we will be able to direct the generated signals and deal with radar signal generation and reception techniques such as Spotlight, Stripmap and Stearing modes of radar operation. The antenna field is also important for signal reception, as we do not want to move the radar along the track, but we will create an antenna field and thus simulate the movement of the platform.
In the last part of the project, we will use a deep neural network, where the operations of all neural network operations will be written mathematically with complex numbers. In internal research, we found that complex neural network notation can increase pattern recognition efficiency.
English 
Slovenian