Radar target identification based on feature extraction performed with RBF artificial neural networks

Efimov, Evgeny; Shevgunov, Timofey

doi:10.1109/ursigass.2014.6929360

Cited by 3 publications

(3 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In an autonomous landing system using an onboard radar [48,49], the reference trajectory and its angular deviations can be calculated based on processing signals reflected from special reflectors [50] (passive repeaters). In the most general form, the possible corner placement relative to the runway is shown in Figure 4.…”

Section: Concept Description Of Using a Radar To Ensure The Aircraft'...mentioning

confidence: 99%

Using Ground-Based Passive Reflectors for Improving UAV Landing

Yasentsev

Shevgunov

Efimov

et al. 2021

Drones

Self Cite

View full text Add to dashboard Cite

The article reviews the problem of landing on hard-to-reach and poorly developed territories, especially in the case of unmanned aerial vehicles. Various landing systems and approaches are analyzed, and their key advantages and disadvantages are summarized; afterwards, an approach with passive reflectors is considered. A formal definition is provided for the main factors relative to the accuracy analysis, and a model is presented. The way to improve the landing procedure, while simultaneously meeting various practical constraints, is analyzed; the results of numerical simulation are presented, followed by the detailed conclusion describing still remaining challenges and subjects for further research.

show abstract

Section: Concept Description Of Using a Radar To Ensure The Aircraft'...mentioning

confidence: 99%

Using Ground-Based Passive Reflectors for Improving UAV Landing

Yasentsev

Shevgunov

Efimov

et al. 2021

Drones

Self Cite

View full text Add to dashboard Cite

show abstract

“…For practical purpose, we can consider the synthesis of subsectional polynomial E-pulse that is the signal which consists of the sections of the same length where the partial signal within each section is written in the form of a polynomial [9]. The analytical expression of such sort of E-pulse is a particular case of the general model expression (6) and it can be expressed:…”

Section: The Theory Of E-pulse Conceptmentioning

confidence: 99%

“…The proposed model states that the early-time part of the target response can be expressed as superposition of many reflections received from the set of effective reflection points. Since its relative strength and position observed from the distant radar station can easily vary, in that case the aspect of the target or wave polarization change, the signal of the early-time part could not be robust for massive target discrimination, although some identification schemes supported by artificial neural networks can be applied to cope with it [5], [6]. In contrast, in the late-time interval, one can observe dumping with time natural oscillations induced by the current whose distribution on the target surface is assumed to be independent from the aspect and other factors.…”

Section: Introductionmentioning

confidence: 99%

The Development of Fast Algorithm for Continuous-Time Subsectional Polynomial E-Pulse Synthesis

Shevgunov¹

2020

TEM Journal

View full text Add to dashboard Cite

This paper discusses the development in the E-pulse technique, also known as the method of extinction pulse, which is an aspect-independent approach to ultra-wideband radar target discrimination in which each target can be characterized by the set of its natural resonances. It is shown that subsectional polynomial E-pulse can be constructed without composing a linear problem and further solution of the underlying matrix equation set. The key concept of the proposed algorithm consists of several steps, where the first one is building a skeleton E-pulse of an especial waveform, the second step is its extension, and the final step is the series of integration. The polynomial structure of the pulse allows above listed steps to be performed over the coefficients of basic functions rather than the functions themselves. As a result, the proposed solution could perform up to a thousand times faster than one based on direct matrix solution. It also provides the coefficients of the polynomial E-pulse sections without solving a linear problem associated with ill-conditioned sparse matrix in its left-hand side. The E-pulse signals synthesized by means of the fast algorithm are proven to be exactly the same as one synthesized by the direct approach. The numerical example given in the paper exposures the main features of the E-pulse technique. The discrimination scheme where two aircraft scaled model targets are involved is simulated.It was shown that the E-pulse discrimination number provides the effective tool for measuring the energy of the late-time part of the convolution as a measure of the difference of two pole sets belonging to the responses under comparison.

show abstract

Estimation of a Radiation Source Location by Difference-Energy Method

Osavchuk

Golovanov

2022

2022 Systems of Signals Generating and Processing in the Field of on Board Communications

View full text Add to dashboard Cite

Radar target identification based on feature extraction performed with RBF artificial neural networks

Cited by 3 publications

References 3 publications

Using Ground-Based Passive Reflectors for Improving UAV Landing

Using Ground-Based Passive Reflectors for Improving UAV Landing

The Development of Fast Algorithm for Continuous-Time Subsectional Polynomial E-Pulse Synthesis

Estimation of a Radiation Source Location by Difference-Energy Method

Contact Info

Product

Resources

About