Software Implementation of Spectral Correlation Density Analyzer with RTL2832U SDR and Qt Framework

Shevgunov, Timofey; Efimov, Evgeniy

doi:10.1007/978-3-030-19813-8_18

Cited by 11 publications

(3 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Since a high resolution in both distance (denoted by ρ) and angle (denoted by φ) for the observed radar target is assumed, the signal could be described with Equation 1where stands for complex-valued space-time radar signal echoed from p-th individual scatterer, is the slice of the radar image across the distance (in fact, the form of this slice will correspond to the form of probe pulse), is a term for squared antenna pattern, φ p determines the main beam direction of the antenna. The term n(t) describes the additive noise which exhibits stationary or cyclostationary properties (Shevgunov et al, 2018b;Shevgunov and Efimov, 2019b) in the frequency band width enough for efficient functioning of the radar system. The schematic structure of the reflected signal is shown in Figure 1, where one can see red thin solid line as the three-dimensional plot of the wave in the time domain.…”

Section: Methodsmentioning

confidence: 99%

Scattering Target Identification Based on Radial Basis Function Artificial Neural Networks in the Presence of Non-Stationary Noise

Shevgunov¹,

Efimov²,

Kirdyashkin³

2019

PTQ

View full text Add to dashboard Cite

The paper deals with the radar target discrimination problem performed on complex radar images. The approach based on radial basis function (RBF) artificial neural network (ANN) is proposed for the identification of point scatterers placed within a radar image. The renewed concept of simple adaptive units as the foundation for network assembling allows one to design an ANN-based feature extraction scheme for the two-dimensional signal processing. It was shown that ANN implementing RBF neural processing units could be applied for the identification of radar targets described by the set of separated scatterers, even in cases where the relative distance between the scatterers is comparable to or less than the effective width of each scatterer. The obtained results indicate a high accuracy estimation of separate scatterer centers in the presence of noise which is not limited to the stationary case but supposed to be cyclostationary. It was also shown that the parameters describing the coordinates of scattering centers could be successfully extracted from the trained ANN after about one hundred epochs spent on ANN training process, which is carried out by means of modified gradient descent method. The main result is to demonstrate the possibility of using neural networks to automatically analyze radar images, which is an integral part of a set of tasks that form the target recognition problem. The proposed algorithm implements an approach of identification systems made using a neural network training procedures.

show abstract

Section: Methodsmentioning

confidence: 99%

Scattering Target Identification Based on Radial Basis Function Artificial Neural Networks in the Presence of Non-Stationary Noise

Shevgunov¹,

Efimov²,

Kirdyashkin³

2019

PTQ

View full text Add to dashboard Cite

show abstract

“…And also, choosing the desired C++ version, it is possible to add other items or bind external libraries. One of the best features of Qt is the signal, and slot [8,9] mechanism, which connects two or more objects or functions regardless of their state and dependency. This allows us to easily talk about the graphical design part from the C++ side.…”

Section: Overview Qt Frameworkmentioning

confidence: 99%

Industrial user interface software design for visual python AI applications using embedded linux based systems

Dikme¹

2021

J. appl. phys. sci.

View full text Add to dashboard Cite

Python software is one of the most popular languages for arti􀅫icial intelligence applications, especially in the academic area, because of its easy syntax and ready-to-use libraries. But for industrial usage, due to the nature of the declarative languages, python is not the preferred language when it is needed to add more functionality to the application, like user interactions or other software abstractions, which needs more system resources and stability, especially in embedded systems. If there is not enough resource to build a new model for AI application for desired software language, it will be perfect to have the advantage of the ability of python in the AI 􀅫ield. Instead of creating one python application or more than one python layer in the system, it is ef􀅫icient to abstract the AI application, which is written in python language, and handle all other activities with more ef􀅫icient languages or frameworks. In this paper, we will see how we can use a visual python AI application, which communicates with another software layer written by C++ using the Qt framework for a user interface in an ef􀅫icient way, running in the backend to handle only AI-related processes. In the example, the python application detects faces in the backend and sends related visual data to the frontend application using interprocess communication. The frontend application will be ef􀅫icient from a memory usage perspective and 􀅫lexible for customer usage in an industrial way. The whole working demo, consisting of a python face detection application and a C++ program, is available in the given GitHub link [1] and is explained in a detailed way for software design and the user interface, which will be written in QML language.

show abstract

“…The analysis above shows that it is enough to take 5 most sensitive DCT coefficients while the ANN contains 8 neurons in the hidden layer. The results of the numerical simulation indicate that MSE generally decreases as the number of DCT coefficients in the RDCTC sets or the number of the neurons in the hidden layer increases (Haykin, 1994;Shevgunov and Efimov, 2019b). It is important to note that the drawback of the increase of either size of hidden layer or the size of RDCTC sets is the increase of the calculation cost.…”

Section: Numerical Simulationmentioning

confidence: 99%

Using Artificial Neural Networks for Time Difference of Arrival Target Localization Based on Reduced Discrete Cosine Transform

Shevgunov¹

2019

PTQ

View full text Add to dashboard Cite

The angular coordinates of the pulse source are determined by comparing the signals received simultaneously on several channels. To solve this issue, the application of neural networks is highly important. In this article, the application of the artificial neural network (ANN) approach to the task of target localization is discussed. The research was performed on the basis of a feature extraction technique executed by a discrete cosine transform, which allowed to obtain a compact representation of the signal energy subjected to digital processing. The author defines the angle-of-arrival estimation scheme based on time difference of arrival estimators and the particular problem of estimating constant delays as informative parameters embedded into received signals that are noisy and damped copies of the reference signal. The adaptive element framework is used for synthesis of the feedforward ANN which is fed with the reduced set of the most sensitive discrete cosine transform (DCT) coefficients, which provide a concise representation of the first-order cyclostationary random process. The investigation on the delay estimation accuracy has been carried out to evaluate the performance of the ANNs with different size of their hidden layer and various numbers of the DCT coefficients at their input. It has been proved that five DCT coefficients are enough for the discrimination of the phase shift in the whole range. In turn, it results in the reliable delay estimation produced by the trained ANN that contains eight neurons in its hidden layer.

show abstract

Software Implementation of Spectral Correlation Density Analyzer with RTL2832U SDR and Qt Framework

Cited by 11 publications

References 11 publications

Scattering Target Identification Based on Radial Basis Function Artificial Neural Networks in the Presence of Non-Stationary Noise

Scattering Target Identification Based on Radial Basis Function Artificial Neural Networks in the Presence of Non-Stationary Noise

Industrial user interface software design for visual python AI applications using embedded linux based systems

Using Artificial Neural Networks for Time Difference of Arrival Target Localization Based on Reduced Discrete Cosine Transform

Contact Info

Product

Resources

About