Deep Learning for Inverse Design of Broadband Quasi-Yagi Antenna

Zhou, Wen-Ying; Mei, Zhong-lei; Lu, Mai; Zhu, Ya-Bo

doi:https://doi.org/10.1155/2023/7819156

International Journal of RF and Microwave Computer-Aided Engineering

On this page

Abstract Introduction Discussion Conclusions Data Availability Conflicts of Interest Acknowledgments References Copyright Related Articles

Research Article | Open Access

Volume 2023 | Article ID 7819156 | https://doi.org/10.1155/2023/7819156

Deep Learning for Inverse Design of Broadband Quasi-Yagi Antenna

Wen-Ying Zhou,¹Zhong-lei Mei,²Mai Lu,¹and Ya-Bo Zhu³

Academic Editor: Amjad Omar

Received26 Oct 2022

Revised04 Dec 2022

Accepted05 Dec 2022

Published09 Feb 2023

Abstract

Deep learning (DL) approaches have been increasingly adopted to design antenna autonomously. For obtaining geometry of the broadband quasi-Yagi antenna from its physic response images directly, we propose an inverse design approach based on the optimized bidirectional symmetry GoogLeNet, which can extract the required bandwidth information to redesign the geometric parameters of antenna without changing its physical structure. It demonstrates that the bandwidth of a reference quasi-Yagi antenna is improved from 0.6 GHz to 1.15 GHz through the proposed inverse design DL approach, and the measured bandwidth value of this redesigned quasi-Yagi antenna achieves 1.16 GHz, which is improved 93% actually. The numerical and measured results indicate that the proposed DL approach could significantly improve the performance of the existed quasi-Yagi antenna and present a new attempt to apply the image processing techniques in resolving physical problem.

1. Introduction

The compact broadband Yagi-Uda antenna is required in the 5G multiple-in multiple-output (MIMO) technology. Since Yagi-Uda antenna was first proposed in 1926 [1], it has become a classical end-fire antenna with high gain. Huang and Densmore introduced a quasi-Yagi antenna array design in 1991 [2], in which its driven, director, and reflector elements were all printed on the substrate, and thus the planar layout was easier to integrate because of its low profile. However, the narrow bandwidth of quasi-Yagi antenna has seriously limited its development in 5G application.

The physical methods to enhance bandwidth of quasi-Yagi antenna include the improvement of feeding technology [3–5] and the structure modification of the driven and director elements [6–8]. In ref. [3, 4], Wu et al. proposed two modified structures of coplanar strip for wideband enhancement. In reference [5], Yang et al. proposed the combination of a wideband feeding slot-line and a modified bowtie driver to enhance the bandwidth of quasi-Yagi antenna, its band-notched characteristics was realized by a compact interdigital capacitor loaded loop resonator with a high quality factor. Compared with the simple structure change of the driven elements [6], the combined design methods have more advantages to expand bandwidth of quasi-Yagi antenna. For example, the combination of the modified transition coplanar strip-line and dual-dipole elements [7], and the dual-bowtie dipole driven was designed with the multimode resonance operation [8]. For the application of quasi-Yagi antenna in MIMO and millimeter-wave system, the features of simpler structures and easier integral feeding are required. Chaudhari and Ray proposed a compact 3-elements quasi-Yagi antenna with a bowtie-driven element and four microstrips line-fed [9]. Rehman et al. proposed a novel high gain two-port planar antenna, which adopted the inverse concentric Yagi director around the rectangular minipatch with a stepped impedance [10].

The physical design of antennas depends on the theoretical computation and empirical formula, and the accurate geometric parameters of the designed antennas are typically obtained through the electromagnetic (EM) simulation. Complex antenna design by EM simulator is prone to error and requires large amount of computational power. In order to overcome these shortcomings and improve design efficiency, deep learning (DL) can be introduced to optimize the geometric parameters of designed antennas. Through forward learning the relationship between structure geometry and its physical response, the neural network (NN) can accurately predict the chiroptical responses of the three-dimensional chiral metamaterials [11], the phase value of metasurface structure [12], the absorptivity of the design metamaterial perfect absorbers [13], the structural color of plasmonic metasurface [14], and the geometry of decoupled antenna array [15]. In order to simplify the design process, we propose an inverse DL approach to directly design the geometry of quasi-Yagi by giving physical response images.

LeNet as the first generation network of convolutional neural network (CNN), it was first proposed to recognize the document in 1998 [16]. CNN, as a successful image processing DL tool, was used to accurately identify the human face with mask by distinguishing radar signals reflection [17] and classify the 1.2 million high-resolution images in the ImageNet LSVRC-2010 contest into the 1000 different classes [18]. In addition to improving image-processing applications, the CNN can also be applied in the antenna design. Harkouss applied CNN in smart antenna design to enhance the direction of arrival (DOA) estimation performance [19]. Sahedian et al. used CNN to collect spatial information from the images for plasmonic structures design [20]. Malkiel et al. published a series of researches for inverse design of the nanophotonic structures [21–23]. By continuously improving the network, their CNN approach could generate 2D images of the target nanostructures with desired spectra. As a result, it was able to design a much wider space of geometries by arbitrary predicting image.

In this article, we intend to redesign the compact quasi-Yagi antenna with bent arms [24] by using an optimized bidirectional symmetry GoogLeNet network. We first extract the information form the desired bandwidth and radiation pattern images. Then, we use image information to inversely design the geometry of the quasi-Yagi antenna with broadband enhancement. By comparing the simulated and measured results of the redesigned quasi-Yagi antenna, we can demonstrate the consistency and effectiveness of our method. Different form the traditional physical method of antenna design, the proposed DL approach can quickly improve the performance of the existed antenna to satisfy the demand in more application fields.

2. Inverse Design DL Approach

As the rapid development of CNN, its classic network structures like AlexNet [25], VGG-Nets [26], GoogLeNet [27], ResNet [28], and DenseNet [29] have appeared successively. Although the architectures of CNN are getting more and more complex, their image processing capabilities are getting stronger and stronger. Since inception module was introduced in the GoogLeNet framework, the performance of CNN has a significant improvement. Anjum et al. proposed an approach based on the GoogLeNet and ResNet101 to accurately classify the subtypes of the brain tumor [30]; Ran et al. designed a memristive GoogLeNet neural network circuits for neuromorphic computing [31].

In this work, we propose a DL inverse design approach to predict the geometric parameters of the quasi-Yagi antenna according to the desired bandwidth and radiation pattern. Its architecture comprises two parts. The first part, i.e., the input layer, takes the S₁₁ and the radiation pattern curves of the target antenna as the input data. The second part is the specific inverse design network, which can finally produces the optimized antenna parameters at the output nodes. The whole architecture is shown in Figure 1.

2.1. Mapping Relationship

Let be the set of all supported geometry of the quasi-Yagi antenna, let be the set of one-dimensional (1D) images of all S₁₁ curves and radiation pattern curves, each group geometric parameter is associated with a valid pair of . Therefore, to be a training model that maps pair of geometric parameters associated with the radiation performance of the predictable quasi-Yagi antennas. In this study, we use the desired images of with 1D vector to be the input elements. Our training goal is to obtain the quasi-Yagi antenna with redesigned geometric parameters (model ), which have the optimal radiation performance, that is .

2.2. Obtaining the Input Images

In the optical research article like reference [32], the desired physical response of metasurface design was only the phase deviation with range of 0°~360°, and the map between input layer and output layer were set as a single channel. But for antennas, its S-parameters and radiation pattern are both important features. Since these two features were independent of each other, their values varied in an unfixed interval. Therefore, our proposed DL approach adopted two channels for training, which separately represent the S₁₁ and the radiation pattern. The designed quasi-Yagi antenna could satisfy these two physical responses at the same time.

The reference quasi-Yagi with bent arms has the simple structure similar to the classic planar quasi-Yagi [33] as shown in Figure 2(a), its images of bandwidth and radiation pattern are shown in Figures 2(b) and 2(c), respectively. One exciter and all the directors are printed on the top of the substrate, and another exciter and reflector are printed on the bottom of the substrate. All the elements material is copper with the thickness of 0.035 mm. The thickness of substrate material is 0.8 mm with the permittivity of 2.2.

(a) Layout

(b) S11

(c) Radiation pattern

As shown in Figure 2(a), five key geometric parameters decide the performance of quasi-Yagi antenna, i.e., , , , , and , which are the final optimized objects as the output elements. Their specific values are listed in Table 1. The desirable S₁₁ curve in Figure 2(b) is artificially drawn based on the simulated result of the reference antenna, and this can greatly enhance the probability for obtaining realizable designs. The desired bandwidth has thus extended from 0.6 GHz to 1.15 GHz. The radiation pattern of the reference antenna is used as the desirable image, which is shown in Figure 2(c).

2.3. Inverse Design Process

Taking the images of S₁₁ and radiation pattern as the input layer, the values of , , , , and as the output layer, we proposed an inverse DL approach based on the optimized bidirectional symmetry GoogLeNet. The network has 10 layers per channel and connected by the fully connected network (FCNN) at the last layer. The target broadband quasi-Yagi antenna can be obtained from the output layers. All activations are ReLU function.

2.3.1. Improved Inception

The main hallmark in GoogLeNet framework is embedded the inception V1 version architecture (as shown in Figure 3(a)) into CNN, which assemble multiple convolutions and pooling operations together to an independent network module. Combined the individual module, the entire GoogLeNet can be constructed. The characteristic pattern is influenced by the different convolution kernel numbers. Therefore, an inception module would provide multiple convolution kernel operations in parallel, the final convolution kernel numbers can be chosen through adjusting the parameters during the selection process. Since our images of S₁₁ and radiation pattern are both 1D pictures, we take the inception 1D module replace of inception V1 to optimize the GoogLeNet network framework. The inception 1D network structure used in this paper is shown in Figure 3(b).

(a) Inception V1

(b) Inception 1D

2.3.2. Training Process

The training data are from the simulated values obtained by EM simulation. In the frequency band from 1 GHz to 4 GHz with interval of 0.5 GHz, the varied ranges of five geometric parameters follow that: , , , , and . These geometric parameters are simulated to obtain their S₁₁ and radiation pattern images by EM simulation. Since the above optimization has far exceed the computational ability of EM simulation, we design an algorithm to randomly select 4900 simulated data with the optimized accuracy of 0.01 mm to train our proposed network, which breakthroughs the limitation of the EM simulation. The training time is about 116 minutes, and the predicted time is only 5 seconds. It should be noted that, because the variation of S₁₁ and radiation pattern values have different influences to antenna geometry, the input point number ratio between S₁₁ image and radiation pattern images is about 61 : 21 in the training network. To demonstrate the training performance of our proposed DL approach, 100 group input images are randomly selected, which is not included in the training set. Through comparing the output 5 geometric parameters (, , , , and ) with their corresponding simulated values obtained by EM simulation, the accuracy of prediction ability is verified, and all percentage errors are basically less than 5%, as shown in Figure 4.

(a)

(b)

(c)

(d)

(e)

3. Discussion and Results Analyses

Through adjusting the network structure, the LeNet and FCNN can also be used to predict the broadband quasi-Yagi. The 33 groups of antenna geometries are separately obtained by LeNet and FCNN. Their corresponding physic responses of S₁₁ and radiation pattern are simulated by EM simulation. For comparison, the predictable performances based on the LeNet, FCNN, and our DL approach are shown in Figure 5, respectively.

(a) S11 (FCNN)

(b) S11 (LeNet)

(c) S11 (this work)

(d) Radiation pattern (FCNN)

(e) Radiation pattern (LeNet)

(f) Radiation pattern (this work)

Figure 5 shows that there are some differences between predictable curves and the desirable curve in different predicted quasi-Yagi antennas obtained by three DL approaches. From Figures 5(a)–5(c), the similarity degree of S₁₁ curves is obviously less than that of radiation pattern curves (in Figures 5(d)–5(f)). Compared with FCNN and LeNet, our proposed approaches have better predictable performance.

To further quantitatively analyze the predictable performance of three DL approaches, the similarity degree of S₁₁ curves is calculated by mathematical models.

3.1. Fréchet’s Distance Analysis

The Fréchet distance [34] is a measure of similarity degree between different curves in mathematics. We use the Fréchet distance to measure the similarity degree of the S₁₁ curves () of each group of predicted antennas (33 groups in total) with the desirable S₁₁ curves () in our work. and are any point on two curves, as shown in Figure 6. The Fréchet distance error is the difference value between the predicted values and the desired values.

Then the minimum network distance between and can be calculated by the following:

The comparison results of Fréchet’s distance obtained by three DL approaches are shown in Figure 7. The abscissa shows the different error intervals of the Fréchet distance, the number “0” represents the minimum error, the number “10” represents the maximum error. The ordinate represents the occurrence frequency of the three DL approaches in different intervals. From Figure 7, among the 33 groups of antennas predicted by FCNN, the Fréchet distances error which are lower than 2 have the ratio of 3%, and the other 97% are in the interval [2, 4]. Among that antennas predicted by LeNet, the Fréchet distances error which are lower than 2 only take the proportion of 0.06%, 67% are in the interval [2, 4], and 27% are bigger than 4. Among that, antennas predicted by our proposed DL approach, the Fréchet distances which are lower than 2 take the proportion of 27% and 68% are in the interval [2, 4]. It demonstrates that the Fréchet distances of our proposed method have lower occurrence frequency in smaller error intervals.

3.2. Hausdorff’s Distance Analysis

Hausdorff’s [35] distance measures how far two subsets of a metric space are from each other in mathematics. It is the longest of all the distances from a point in one set to the closest point in the other set, as shown in Figure 8.

Assume and be two subsets of a geometric distance (, ), their Hausdorff’s distance can be defined by where represents the supremum, means the infimum, and quantifies the distance from a point to the subset . In this work, represents the predictable S₁₁ curve, which is the simulated result of each predicted antennas, and represents the desirable S₁₁ curve. The Hausdorff distance error is the difference value between the predicted values and the desired values.

We compare the sum of 33 Hausdorff distances separately obtained by three DL approaches, and the results are shown as Figure 9. It reveals that, among the 33 groups of antennas predicted by FCNN, the Hausdorff distances error which are lower than 2 have the ratio of 6%, and the other 94% are in the interval [2, 4]. Among that antennas predicted by LeNet, the Hausdorff distances error which are lower than 2 take the proportion of 45%, 12% are in the interval [2, 4], and the other 24% are bigger than 4. Among that antennas predicted by our proposed DL approach, the Hausdorff distances error which are lower than 2 take the proportion of 30% and the other 61% are in the interval [2, 4]. It demonstrates that the Hausdorff distances of our proposed method have lower occurrence frequency in smaller error intervals.

The above comparison results demonstrate that the predicted antenna which obtained by our proposed DL approach owns the highest similarity of S₁₁ curves.

3.3. Comparison of the Predictable Antenna with the Optimal Performance

Through comparing the 33 groups of geometries of antennas, obtained by the above three DL approaches, we select 3 groups optimal antennas to simulate their physical response. The obtained S₁₁ and radiation results are compared with the desirable curves, as shown in Figure 10.

(a) S11

(b) Radiation pattern

Figure 10(a) shows that, although the best bandwidth of the selected antennas, obtained by the three DL algorithms, has the same value of 1.15 GHz, the predicted quasi-Yagi antenna obtained by our proposed DL approach has the optimal S₁₁ curve with the smallest values, which cannot be obtained from FCNN and LeNet. From another perspective, the quasi-Yagi antennas predicted by our proposed DL approach cannot only get the similar S₁₁ images with the desired one but also can obtain the best one. From Figure 10(b), it can be seen that the radiation patterns obtained by the three DL algorithms are similar to the reference antenna.

4. Analysis of Experimental Results

Through our proposed DL approach training, the reference quasi-Yagi antenna geometry is redesigned to the desirable broadband quasi-Yagi antenna, and their specific values are listed in Table 1. Because effect on the antenna performance is little which has been proved in our previous study, the redesigned antenna can be further optimized by reducing from 22.75 mm to 8 mm. Its geometric parameters are shown in Figure 11(a), and the fabrication prototype is shown in Figure 11(b).

(a) Layout

(b) Photograph of the fabrication prototype

The fabrication of the redesigned quasi-Yagi antenna is measured to prove the reliability of our inverse design. The measured values of S₁₁ and radiation pattern are compared with their simulated values, and the results are shown in Figure 12.

(a) Comparison of S11

(b) Simulated value of radiation pattern

(c) Measured value of radiation pattern

Figure 12(a) shows that the experimental value of the bandwidth is 1.16 GHz. Figures 12(b) and 12(c) show that the measured results of radiation pattern are basically agreed with the simulated results. The errors of measured values are mainly caused by the machining precision and the substrate material.

The compared results between the reference and the redesigned quasi-Yagi antenna are listed in Table 1.

Table 1 shows that the total geometry of the redesigned quasi-Yagi antenna is a bit smaller than the reference antenna, but the bandwidth is actually improved 93%.

5. Conclusions

Different from the traditional inverse design method of antenna, we propose an inverse DL approach to directly redesign the reference quasi-Yagi antenna according to the desirable radiation performance images. It means that the relationship between antenna geometry and its physical response could be translated to the images learning problem. In this article, both the S₁₁ and radiation pattern images are learned at the same time to predict the geometric parameters of the required broadband quasi-Yagi antenna, which improves the design efficiency. The bandwidth of the redesigned quasi-Yagi is improved 92% with no radiation pattern deterioration. The analysis of Fréchet’s distance error and Hausdorff’s distance error confirms that our proposed bidirectional symmetry GoogLeNet can obtain the optimal quasi-Yagi antenna with the best radiation performance. The measured results of the redesigned quasi-Yagi antenna demonstrate the effectiveness of our proposed inverse design DL approach. Our work highlights the convenience of inverse design from physical response images to antenna geometry with desired performance, which can greatly shorten the design time in the complicated structure design.

Data Availability

The supporting data used to support the findings of this study are included within the article.

Conflicts of Interest

The authors declare that they have no conflicts of interest.

Acknowledgments

This work was supported by the National Natural Science Foundation of China (62161017, 61701208, and 61631007), the Natural Science Foundation of Gansu Province (21JR7RA283 and 20JR10RA604) and the Tianyou Youth Talent Lift Program of Lanzhou Jiaotong University (1520260111).

References

H. Yagi and S. Uda, “Projector of the sharpest beam of electric waves,” Proceedings of the Imperial Academy, vol. 2, no. 2, pp. 49–52, 1926.
View at: Publisher Site | Google Scholar
J. Huang and A. C. Densmore, “Microstrip Yagi array antenna for mobile satellite vehicle application,” IEEE Transactions on Antennas and Propagation, vol. 39, no. 7, pp. 1024–1030, 1991.
View at: Publisher Site | Google Scholar
J. Wu, Z. Zhao, Z. Nie, and Q.-H. Liu, “Bandwidth enhancement of a planar printed quasi-Yagi antenna with size reduction,” IEEE Transactions on Antennas and Propagation, vol. 62, no. 1, pp. 463–467, 2014.
View at: Publisher Site | Google Scholar
J. Wu, Z. Zhao, Z. Nie, and Q.-H. Liu, “Design of a wideband planar printed quasi-Yagi antenna using stepped connection structure,” IEEE Transactions on Antennas and Propagation, vol. 62, no. 6, pp. 3431–3435, 2014.
View at: Publisher Site | Google Scholar
T. Yang, D. Yang, and D. Geng, “Compact planar quasi-Yagi antenna with band-notched characteristic for WLAN and DSRC for ultra-wideband applications,” IET Microwaves, Antennas and Propagation, vol. 12, no. 7, pp. 1239–1245, 2018.
View at: Publisher Site | Google Scholar
S. S. Jehangir and M. S. Sharawi, “A single layer semi-ring slot Yagi-like MIMO antenna system with high front-to-back ratio,” IEEE Transactions on Antennas and Propagation, vol. 65, no. 2, pp. 937–942, 2017.
View at: Google Scholar
D. Yang, J. Qu, Z. Zhao, S. Liu, and Z. Nie, “Planar quasi-Yagi antenna with band rejection based on dual dipole structure for UWB,” IET Microwaves, Antennas and Propagation, vol. 10, no. 15, pp. 1708–1714, 2016.
View at: Publisher Site | Google Scholar
Z. Liang and J. Yuan, “Compact dual-wideband multi-mode printed quasi-Yagi antenna with dual-driven elements,” IET Microwaves, Antennas and Propagation, vol. 14, no. 7, pp. 662–670, 2020.
View at: Publisher Site | Google Scholar
A. D. Chaudhari and K. P. Ray, “A single-layer compact four-element quasi-Yagi MIMO antenna design for super- wideband response,” AEU - International Journal of Electronics and Communications, vol. 138, article 153878, 2021.
View at: Publisher Site | Google Scholar
R. Rehman, J. A. Sheikh, K. A. Shah, and G. Bhat, “A high gain inverse concentric Yagi director antenna for 5G millimeter-wave and satellite communication,” Progress in Electromagnetics Research B, vol. 19, pp. 127–148, 2021.
View at: Google Scholar
W. Ma, F. Cheng, and Y. Liu, “Deep-learning-enabled on-demand design of chiral metamaterials,” ACS Nano, vol. 12, no. 6, pp. 6326–6334, 2018.
View at: Publisher Site | Google Scholar
L. Jiang, X. Li, Q. Wu, L. Wang, and L. Gao, “Neural network enabled metasurface design for phase manipulation,” Optics Express, vol. 29, no. 2, pp. 2521–2528, 2021.
View at: Publisher Site | Google Scholar
C. Han, B. Zhang, H. Wang, and J. Ding, “Metamaterial perfect absorber with morphology-engineered meta-atoms using deep learning,” Optics Express, vol. 29, no. 13, pp. 19955–19963, 2021.
View at: Publisher Site | Google Scholar
N. B. Roberts and M. K. Hedayati, “A deep learning approach to the forward prediction and inverse design of plasmonic metasurface structural color,” Applied Physics Letters, vol. 119, no. 6, article 061101, 2021.
View at: Publisher Site | Google Scholar
W. Y. Zhou, Z. L. Mei, M. Lu, and Y. B. Zhu, “Improved fully-connected neural network approach for decoupling microstrip antenna array design,” Journal of Electromagnetic Waves and Applications, vol. 36, no. 14, pp. 1996–2009, 2022.
View at: Publisher Site | Google Scholar
Y. Lecun, L. Bottou, Y. Bengio, and P. Haffner, “Gradient-based learning applied to document recognition,” Proceedings of the IEEE, vol. 86, no. 11, pp. 2278–2324, 1998.
View at: Publisher Site | Google Scholar
J. Kim, J.‐. E. Lee, H.‐. S. Lim, and S. Lee, “Face identification using millimetre-wave radar sensor data,” Electronics Letters, vol. 56, no. 20, pp. 1077–1079, 2020.
View at: Google Scholar
A. Krizhevsky, I. Sutskever, and G. E. Hinton, “ImageNet classification with deep convolutional neural networks,” Communications of the ACM, vol. 60, no. 6, pp. 84–90, 2017.
View at: Publisher Site | Google Scholar
Y. Harkouss, “Direction of arrival estimation in multipath environments using deep learning,” International Journal of Communication Systems, vol. 34, no. 11, 2021.
View at: Publisher Site | Google Scholar
I. Sahedian, J. Kim, and J. Rho, “Finding the optical properties of plasmonic structures by image processing using acombination of convolutional neural networks and recurrent neural networks,” Microsystems & Nanoengineering, vol. 5, no. 1, 2019.
View at: Google Scholar
I. Malkiel, M. Mrejen, A. Nagler, U. Arieli, L. Wolf, and H. Suchowski, “Plasmonic nanostructure design and characterization via deep learning,” Light: Science & applications, vol. 7, no. 1, p. 60, 2018.
View at: Publisher Site | Google Scholar
I. Malkiel, M. Mrejen, L. Wolf, and H. Suchowski, “Machine learning for nanophotonics,” MRS Bulletin, vol. 45, no. 3, pp. 221–229, 2020.
View at: Publisher Site | Google Scholar
I. Malkiel, M. Mrejen, L. Wolf, and H. Suchowski, “Inverse design of unparametrized nanostructures by generating images from spectra,” Optics Letters, vol. 46, no. 9, pp. 2087–2090, 2021.
View at: Publisher Site | Google Scholar
W. Y. Zhou, Z. L. Mei, and M. Lu, “A compact quasi-Yagi antenna with high gain by employing the bent arms and split-ring resonators,” International Journal of Antennas and Propagation, vol. 2021, Article ID 6666657, 9 pages, 2021.
View at: Publisher Site | Google Scholar
Z. W. Yuan and J. Zhang, “Feature extraction and image retrieval based on AlexNet,” in International Conference on Digital Image Processing (ICDIP 2016), vol. 10033, pp. 65–69, Chengdu, China, 2016.
View at: Publisher Site | Google Scholar
D. Mishkin, N. Sergievskiy, and J. Matas, “Systematic evaluation of convolution neural network advances on the Imagenet,” Computer Vision and Image Understanding, vol. 161, pp. 11–19, 2017.
View at: Publisher Site | Google Scholar
B. Almas, K. Sathesh, and S. Rajasekaran, “A deep analysis of Google Net and AlexNet for lung cancer detection,” International Journal of Engineering and Advanced Technology, vol. 9, no. 2, pp. 395–399, 2019.
View at: Publisher Site | Google Scholar
M. Zhao, S. Zhong, X. Fu, B. Tang, and M. Pecht, “Deep residual shrinkage networks for fault diagnosis,” IEEE Transactions on Industrial Informatics, vol. 16, no. 7, pp. 4681–4690, 2020.
View at: Publisher Site | Google Scholar
C. Qian, Q. Jiang, Y. Shen, C. Huo, and Q. Zhang, “An intelligent fault diagnosis method for rolling bearings based on feature transfer with improved DenseNet and joint distribution adaptation,” Measurement Science and Technology, vol. 33, no. 2, article 025101, 2022.
View at: Google Scholar
S. Anjum, L. Hussain, M. Ali et al., “Detecting brain tumors using deep learning convolutional neural network with transfer learning approach,” International Journal of Imaging Systems and Technology, vol. 32, no. 1, pp. 307–323, 2021.
View at: Google Scholar
H. Ran, S. Wen, K. Shi, and T. Huang, “Stable and compact design of memristive GoogleNet neural network,” Neurocomputing, vol. 441, pp. 52–63, 2021.
View at: Publisher Site | Google Scholar
R. Zhu, T. Qiu, J. Wang et al., “Phase-to-pattern inverse design paradigm for fast realization of functional metasurfaces via transfer learning,” Nature Communications, vol. 12, no. 1, 2021.
View at: Google Scholar
G. Bulla, M. T. Le, A. A. A. Salles, and T. P. Vuong, “Miniaturized printed Yagi antenna for 2.45 GHz RFID readers,” in PIERS Proceeding, Marrakesh, Morocco, 2011.
View at: Google Scholar
H.-K. Ahn, H. Alt, M. Buchin, E. Oh, L. Scharf, and C. Wenk, “Middle curves based on discrete on Fréchet distance,” Computational Geometry, vol. 89, article 101621, 2020.
View at: Google Scholar
A. Vavpetič and E. Žagar, “On optimal polynomial geometric interpolation of circular arcs according to the Hausdorff distance,” Journal of Computational and Applied Mathematics, vol. 392, article 113491, 2021.
View at: Publisher Site | Google Scholar

Copyright

Copyright © 2023 Wen-Ying Zhou et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

PDF Download Citation

Download other formats

Order printed copies