High-accuracy neural-network-based array synthesis including element coupling

Ayestaran, R.G.; Las-Heras, Fernando; Herrán, L.F.

doi:10.1109/lawp.2006.870366

Cited by 15 publications

(13 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…But these criteria are not accurate enough. In [27][28][29], MC is defined. The coupling between two transmitter antennas is defined in [24] as:…”

Section: Dipole Array Design For Bearable Coupling Between Adjacent Ementioning

confidence: 99%

Nonuniformly Spaced Linear Array Design for the Specified Beamwidth/Sidelobe Level or Specified Directivity/Sidelobe Level With Coupling Consideration

Oraizi¹,

Fallahpour²

2008

PIER M

View full text Add to dashboard Cite

Abstract-In this paper, we investigate nonuniformly spaced linear arrays (NUSLA) rigorously. Several important problems in NUSLA design are solved with the combination of the Genetic Algorithm and Conjugate Gradient method (GA-CG). The pattern synthesis for the specified beamwidth and minimum achievable sidelobe level (SLL) are performed and for the first time, the graphs which show the relation between the beamwidth, sidelobe level and number of elements for NUSLA are derived. Also, the NUSLA's pattern for the specified directivity and sidelobe level is synthesized. The graphs showing the behavior of NUSLA relative to the increase of its length for a fixed number of elements are derived. These graphs show the relations between the directivity and sidelobe level of NUSLA with its length. As a practical design, an array of parallel dipoles is designed for specified beamwidth/sidelobe level or specified directivity/sidelobe level. Furthermore, a novel Neural Network based model for the NUSLA is presented for the rapid and accurate computation of Sparameters. The computed S-parameters are used for the computation of coupling among elements. Then the GA-CG method can adjust these values in the synthesis process to achieve desired pattern and bearable coupling among elements.

show abstract

“…But these criteria are not accurate enough. In [27][28][29], MC is defined. The coupling between two transmitter antennas is defined in [24] as:…”

Section: Dipole Array Design For Bearable Coupling Between Adjacent Ementioning

confidence: 99%

Nonuniformly Spaced Linear Array Design for the Specified Beamwidth/Sidelobe Level or Specified Directivity/Sidelobe Level With Coupling Consideration

Oraizi¹,

Fallahpour²

2008

PIER M

View full text Add to dashboard Cite

show abstract

“…Employing machine learning (ML) in advanced computational electromagnetics and relevant applications were initiated long time ago [13][14][15][16]. Artificial neural networks (ANN) have been proposed for array synthesis [17], source reconstruction [18], NF to FF transformation [19], etc. Due to the recent blooming learning technologies, convolutional neural networks (ConvNets) [20,21] have become one of the most important new methods in deep learning applications.…”

Section: Introductionmentioning

confidence: 99%

“…The advantages of the proposed method are: (1) the proposed ConvNet model allows calculation using much fewer field samples than the conventional SRM; (2) without inverse solving, the proposed method avoids handling a singular numerical system; (3) the proposed ConvNet approach can make further field information to do reconstruction; (4) the proposed method has satisfactory accuracy and superior performance over traditional neural networks [17][18][19]. Compared to traditional neural networks (NNs) [17][18][19]24], ConvNet can more efficiently map the relations between inputs and outputs mainly by convolutional layer and activation layer [20,21]. It does not need very large number of neural units to handle problems such as source reconstruction.…”

Section: Introductionmentioning

confidence: 99%

Applying Convolutional Neural Networks for the Source Reconstruction

Yao¹,

Sha²,

Jiang³

2018

PIER M

View full text Add to dashboard Cite

This paper proposes a novel source reconstruction method (SRM) based on the convolutional neural network algorithm. The conventional SRM method usually requires the scattered field data oversampled compared to that of target object grids. To achieve higher accuracy, the conventional SRM numerical system is highly singular. To overcome these difficulties, we model the equivalent source reconstruction process using the machine learning. The equivalent sources of the target are constructed by a convolutional neural networks (ConvNets). It allows us to employ less scattered field samples or radar cross section (RCS) data. And the ill-conditioned numerical system is effectively avoided. Numerical examples are provided to demonstrate the validity and accuracy of the proposed approach. Comparison with the traditional NN is also benchmarked. We further expand the proposed method into the direction of arrival (DOA) estimation to demonstrate the generality of the proposed procedure.

show abstract

“…Two major data‐learning approaches have been proposed: the empirical risk minimization (ERM) principle and the structural risk minimization (SRM) principle . The best‐known example of the ERM principle is the artificial neural networks . When the ERM is applied, the learning method intends to find a function that relates perfectly the available inputs with their corresponding outputs.…”

Section: Introductionmentioning

confidence: 99%

Support vector regression for near‐field multifocused antenna arrays considering mutual coupling

Muñiz

Ayestaran

Laviada

et al. 2015

Int J Numerical Modelling

View full text Add to dashboard Cite

Support vector regression is proposed to obtain the coupling matrix that accounts for the real behavior of an antenna array in the near-field multifocusing framework. The obtained model can then be easily included in this synthesis method in order to obtain the feeding weights which satisfy the radiation pattern requirements under real conditions. An example of the performance of support vector regression using information from method of moments is also provided to demonstrate the performance and accuracy of the technique. SUPPORT VECTOR REGRESSION, NEAR-FIELD MULTIFOCUSING, MUTUAL COUPLING NEAR-FIELD MULTIFOCUSING FRAMEWORK UNDER REAL CONDITIONSThe aim of the NF-MF framework is obtaining the excitations to apply to the array elements so that the entire antenna focuses on the desired points simultaneously. Method explained in Reference [1] is used for this purpose, where an MP optimization or a PO synthesis of the feeding weights can be obtained according to the preferences of the application. Thus, excitations ! in (7) can be obtained using the NF-MF technique, with the following modifications to fulfill mutual coupling conditions.The commented synthesis technique in Reference [1] is based on a proper cost function defined by the maximum and minimum limits allowed for the field at each position of space, and its minimization is solved using an equivalent least squares problem and the Levenberg-Marquardt method [2]. A Jacobian matrix is then required, where some partial derivatives of the radiated field according to the

show abstract

High-accuracy neural-network-based array synthesis including element coupling

Cited by 15 publications

References 4 publications

Nonuniformly Spaced Linear Array Design for the Specified Beamwidth/Sidelobe Level or Specified Directivity/Sidelobe Level With Coupling Consideration

Nonuniformly Spaced Linear Array Design for the Specified Beamwidth/Sidelobe Level or Specified Directivity/Sidelobe Level With Coupling Consideration

Applying Convolutional Neural Networks for the Source Reconstruction

Support vector regression for near‐field multifocused antenna arrays considering mutual coupling

Contact Info

Product

Resources

About