Classification of radar clutters with Artificial Neural Network

Darzikolaei, Mohammad Akhondi; Ebrahimzade, AtaAllah; Gholami, Elahe

doi:10.1109/kbei.2015.7436109

Cited by 23 publications

(14 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A number of model based approaches have been developed to address this issue. These include amplitude distribution analysis [ 3 , 4 , 5 , 6 , 7 ], spectral analysis, auto-regressive modeling [ 8 , 9 , 10 ], and parametric and nonparametric statistical decision rules. For instance, in [ 3 ], a generalized, compound distribution model incorporating thermal noise was used to describe the high resolution radar clutter.…”

Section: Introductionmentioning

confidence: 99%

“…Among different supervised learning approaches, neural network (NN) or artificial neural network (ANN) based approaches have been shown to be very effective [ 2 , 5 , 6 , 8 , 10 , 11 ]. The common theme among all the ANN based approaches is that a feed-forward, multi-layer neural network is used to learn to classify a given clutter among multiple classes.…”

Section: Introductionmentioning

confidence: 99%

“…In particular, the proposed feature space for the sea and land clutters includes classic parameters covering the statistical characteristics of clutters, short time energy, amplitude variation, and the frequency and properties around autocorrelation. The feature space we propose here, to the best of our knowledge, is novel and different from that of [ 2 , 3 , 4 , 5 , 6 ]. In doing these, we make the following contributions: We propose a novel set of features for improving the classification performance, namely the energy, zero-crossing rate, the autocorrelation, and the shape parameter V (ECAV).…”

Section: Introductionmentioning

confidence: 99%

“…Among many different approaches for modeling clutter [ 21 ], modeling them using a statistical (or probabilistic) distribution is a robust and proven technique. Example clutter models include Rayleigh, log-normal, Weibull, and K distributions [ 6 , 22 ].…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

A Novel Method for Sea-Land Clutter Separation Using Regularized Randomized and Kernel Ridge Neural Networks

Thiyagalingam

Xue

et al. 2020

Sensors

View full text Add to dashboard Cite

Classification of clutter, especially in the context of shore based radars, plays a crucial role in several applications. However, the task of distinguishing and classifying the sea clutter from land clutter has been historically performed using clutter models and/or coastal maps. In this paper, we propose two machine learning, particularly neural network, based approaches for sea-land clutter separation, namely the regularized randomized neural network (RRNN) and the kernel ridge regression neural network (KRR). We use a number of features, such as energy variation, discrete signal amplitude change frequency, autocorrelation performance, and other statistical characteristics of the respective clutter distributions, to improve the performance of the classification. Our evaluation based on a unique mixed dataset, which is comprised of partially synthetic clutter data for land and real clutter data from sea, offers improved classification accuracy. More specifically, the RRNN and KRR methods offer 98.50% and 98.75% accuracy, outperforming the conventional support vector machine and extreme learning based solutions.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A Novel Method for Sea-Land Clutter Separation Using Regularized Randomized and Kernel Ridge Neural Networks

Thiyagalingam

Xue

et al. 2020

Sensors

View full text Add to dashboard Cite

show abstract

“…In [8], the authors describe X-band radar-based measurements of urban ground clutter, which is made from the CSIR campus in Pretoria, South-Africa. In [9], artificial neural network is used to classify four common radar clutter models. After simulating clutter models and extracting the important features, the authors make neural network with two hidden layers.…”

Section: Introductionmentioning

confidence: 99%

Generation of Weibull distribution clutter based on correlated Gaussian sequence

Wang

Xin

2017

J. Phys.: Conf. Ser.

View full text Add to dashboard Cite

The Weibull functional form for SEP event spectra M Laurenza, G Consolini, M Storini et al. -The Weibull functional form for the energetic particle spectrum at interplanetary shock waves M Laurenza, G Consolini, M Storini et al. Abstract. With the continuous development of science and technology, the electromagnetic environment becomes more complex. Accurate clutter modeling is becoming increasingly difficult, which will have adverse effects in echo analysis. In this paper, in order to overcome electromagnetic interference, we use correlated Gaussian sequence to generate Weibull distribution clutter. Simulation results show that the estimated value of the proposed method is close to the theoretical value in the aspect of probability density and power spectral density. That demonstrates the validity of our method. Finally, the conclusions are given. IntroductionAs the increasing performance desire and new proposed system of radar, the demands for radar simulation system are increasing. Echo analysis is an important part in radar simulation, and the basis is clutter modeling. However, electromagnetic environment is more and more complex, and accurate clutter modeling is becoming increasingly difficult. So clutter modeling in complex environment is becoming hot issues of research. Clutter simulation is the key link in the design of radar echo simulator, and many researchers have conducted a lot of work in related field. Simon Haykin proposes the concept of cognitive radar. Using all available knowledge, it can interrogate a propagation channel adaptively [1]. In [2], the authors add different clutter distributions into the framework of trackability, and significant differences are obtained. In [3], it is shown that when looking up or down wind local spectrum intensity is strongly linked to mean Doppler shift. The spectra will be given a temporal and spatial variability by related characteristics when combined with random fluctuations of spectrum width. In [4], different from traditional signal-dependent stochastic models, the authors propose a new approach of stochastic transfer function. In [5], the authors research on bistatic radar. The authors define bistatic clutter scenario and propose two land-clutter models. In [6], with the environment of Saudi Arabian urban, South African urban and Saudi Arabian date farms, the statistical analysis of three types of land clutter are presented. In [7], a new model is proposed, which is comprised by functions of radar frequency, polarization, sea state, and grazing angle. In [8], the authors describe X-band radar-based measurements of urban ground clutter, which is made from the CSIR campus in Pretoria, South-Africa. In [9], artificial neural network is used to classify four common radar clutter models. After simulating clutter models and extracting the important features, the authors make neural network with two hidden layers. In [10], the optimization of an information theoretic criterion for radar waveform design is described.In this paper, after describing principles of...

show abstract

K Distribution Clutter Modeling Based on Chebyshev Filter

Wang

Xin

2017

Advances in Intelligent Systems and Computing

View full text Add to dashboard Cite

Classification of radar clutters with Artificial Neural Network

Cited by 23 publications

References 10 publications

A Novel Method for Sea-Land Clutter Separation Using Regularized Randomized and Kernel Ridge Neural Networks

A Novel Method for Sea-Land Clutter Separation Using Regularized Randomized and Kernel Ridge Neural Networks

Generation of Weibull distribution clutter based on correlated Gaussian sequence

K Distribution Clutter Modeling Based on Chebyshev Filter

Contact Info

Product

Resources

About