Search

The subject of the manuscript is the algorithms for radar imaging. This research develops imaging methods and algorithms for wideband and ultrawideband active aperture synthesis systems with antenna arrays and stochastic probing signals. The use of antenna arrays makes it possible to obtain radar images without the need to move radar or antenna system in space. The use of wideband and ultra-wideband stochastic probing signals is justified by their narrow autocorrelation functions. This increased the resolution of the obtained images. The main idea of the proposed algorithms is to filter the original wideband signal into several narrowband processes. Furthermore, only the central frequencies of each narrowband signal were processed. This approach allows us to use the classical widespread methods of aperture synthesis for the case of a wideband signal. Usually, they are applicable only for narrowband signals that satisfy the condition of a quasi-monochromatic approximation. This significantly reduces the overall computational complexity of the imaging algorithm, which simplifies its further practical implementation on the existing radioelement base. Because of the simulation, a primary radar image has been obtained and the overall performance of the proposed approach to processing wideband signals has been confirmed. An increase in the quality of the obtained image when using a multiple of frequency ranges is shown. An experimental study of the effect of processing a wideband signal only at its centre frequency instead of the entire frequency band is conducted. During the experiment, the correlation functions of the signals received by two spaced receivers were obtained. As a result, the Van Cittert-Zernike theorem has been experimentally confirmed. It allows signal processing only at its centre frequency instead of the entire frequency band. Simultaneously, the prospect of expanding the bandwidth of the probing signal is indicated. It, in the presence of a wideband element base and devices for high-speed signal processing, will further increase the imaging resolution of a radar system.

Within the maximum likelihood method an optimal algorithm for polarization target selection against the background of interfering signal reflected from the earth’s surface is synthesized. The algorithm contains joint operations of spectral interference rejection and their polarization compensation by means of certain combinations of interchannel subtraction of signals of different polarizations. The physical features of the elements of the polarization scattering matrix are investigated for the technical implementation of the synthesized algorithm.

Using the statistical theory of optimization of radio engineering systems the optimal method of coherent radar imaging of surfaces in airborne synthetic aperture radar with planar antenna arrays is developed. This method summarizes several modes of terrain observation and it is fully consistent with current trends in the development of cognitive radars with the possibilities of radiation pattern restructuring in space and adaptive reception of reflected signals. Possible modifications of the obtained optimal method for the operation of high-precision airborne radars with a wide swath are presented. The idea is to create a theoretical basis and lay the foundations for its practical application in solving a wide range of issues of statistical optimization of methods and algorithms for optimal spatiotemporal signal processing in cognitive radar systems for the formation of both high-precision and global radar images. To implement the idea, the article highlights the concept of statistical optimization of spatio-temporal processing of electromagnetic fields in on-board cognitive radar systems, which will be based on the synthesis and analysis of methods, algorithms and structures of radar devices for coherent imaging, the study of limiting errors in restoring the spatial distribution of the complex scattering coefficient, the synthesis of optimal feedback for receiver and transmitter adaptations in accordance with a priori information about the parameters of the objects of study, the area of observation and the existing sources of interference. Objective is to develop the theory and fundamentals of the technical implementation of airborne radar systems for the formation of high-precision radar images in an extended field of view from aerospace carriers. Tasks. To reach the objective it is necessary to solve following tasks: – formalize mathematical models of spatiotemporal stochastic radio signals and develop likelihood functional for observation equations in which the useful signal, receiver internal noise and interference radiation of anthropogenic objects are random processes; – to synthesize algorithms for optimal processing of spatio-temporal stochastic signals in multi-channel radar systems located on aerospace-based mobile platforms; - in accordance with the synthesized methods, to substantiate the block diagrams of their implementation; – obtain analytical expressions for the potential characteristics of the quality of radar imaging and determine the class of probing signals and space scanning methods necessary to perform various tasks of radar surveillance; ‒ to confirm some of the theoretical results by simulation methods, in which to reveal the features of the technical implementation of aerospace remote sensing radar systems.

During the crucial phases of take-off, initial climb, approach, and landing where aircraft are close to the ground, Global Navigation Satellite Systems (GNSS) signal strength may not be sufficient to guarantee safe operation, especially in the presence of potential interference, malicious or otherwise, from ground equipment. When the GNSS location is lost, aircraft typically revert to other navigation aids. The most accurate navigation aid is Distance Measuring Equipment (DME). However, whereas GNSS location is triangulated, the navigation equipment on-board aircraft can only measure two DME signals simultaneously. Therefore, location based on DME tends to be accurate only to hundreds of meters, compared to meters for GNSS. A new approach is presented for positioning using multiple DMEs. The approach is based on regression analysis for prediction of DMEs distances in time of measurement. This approach increases positioning accuracy due to availability of multiple DMEs data in the system of navigation equations. Spline functions were used in a regression model in order to achieve the most accurate prediction values. An approach was verified using real flight data and shown the decreasing of navigation system error on value depending on availability and geometry of ground stations locations.