Po-Hsiang Chen

Prof. Po-Hsiang Chen

Tamkang University, Taipei, Taiwan
orcid

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
16 November 2022 in IEEE Transactions on Geoscience and Remote Sensing
Verified
Journal Article
Closed access
Chien-Ching ChiuChien-Ching Chiu
Chien-Ching Chiu
Po-Hsiang ChenPo-Hsiang Chen
Po-Hsiang Chen
HJHao Jiang
Hao Jiang
The electromagnetic imaging of uniaxial objects by the Artificial Intelligence (AI) technology is presented in this paper. We study the two-dimensional inverse scattering problem from uniaxial objects illuminated by the TM (Transverse Magnetic) and TE (Transverse Electric) polarized incident waves. As the uniaxial objects have different components of permittivity along different transverse directions, the problem of TE polarization will be more severe than that of TM polarization. We use the Dominant Current Scheme (DCS) and Back Propagation Scheme (BPS) to calculate the preliminary permittivity distribution. By combining with deep learning and neural networks, the permittivity distribution of those uniaxial objects can be reconstructed more accurately. U-Net is used to reconstruct the permittivity distribution, because U-Net has shared the weights and biases, which can effectively reduce the network complexity and is very suitable for solving image processing problems. In the numerical results, we added different noises to compare the reconstruction results of the DCS and BPS initial estimations through the U-Net. Numerical results show that the reconstruction permittivity for the DCS initial estimation is better than that for the BPS’s. Our diversity is that we have reconstructed the uniaxial objects by neural network successfully with less time-consuming effort and real-time imaging.

12 November 2024 in Applied Sciences
Verified
Journal Article
Open access
PDF file
Chien-Ching ChiuChien-Ching Chiu
Chien-Ching Chiu
MWMarty Wu
Marty Wu
Po-Hsiang ChenPo-Hsiang Chen
Po-Hsiang Chen
CCChen-En Chao
Chen-En Chao
ELEng Hock Lim
Eng Hock Lim
This paper explores 6G technology for indoor positioning, focusing on accuracy and reliability using convolutional neural networks (CNN) with channel state information (CSI). Indoor positioning is critical for smart applications and the Internet of Things (IoT). 6G is expected to significantly enhance positioning performance through the use of higher frequency bands, such as terahertz frequencies with wider bandwidth. Preliminary results show that 6G-based systems are expected to achieve centimeter-level positioning accuracy due to the integration of advanced artificial intelligence algorithms and terahertz frequencies. In addition, this paper also investigates the impact of self-attention (SA) and channel attention (CA) mechanisms on indoor positioning systems. The combination of these attention mechanisms with conventional CNNs has been proposed to further improve the accuracy and robustness of localization systems. CNN with SA demonstrates a 50% reduction in RMSE compared to CNN by capturing spatial dependencies more effectively.

2 June 2023 in Applied Sciences
Verified
Journal Article
Open access
PDF file
Chien-Ching ChiuChien-Ching Chiu
Chien-Ching Chiu
WCWei Chien
Wei Chien
KYKai-Xu Yu
Kai-Xu Yu
Po-Hsiang ChenPo-Hsiang Chen
Po-Hsiang Chen
ELEng Hock Lim
Eng Hock Lim
In the past, many conventional algorithms, such as self-adaptive dynamic differential evolution and asynchronous particle swarm optimization, were used to reconstruct buried objects in the frequency domain; these were unfortunately time-consuming during the iterative, repeated computing process of the scattered field. Consequently, we propose an innovative deep convolutional neural network approach to solve the electromagnetic inverse scattering problem for buried conductors in this paper. Different shapes of conductors are buried in one half-space and the electromagnetic wave from the other half-space is incident. The shape of the conductor can be reconstructed promptly by inputting the received scattered fields measured from the upper half-space into the deep convolutional neural network module, which avoids the computational complexity of Green’s function for training. Numerical results show that the root mean square error for differently shaped—circular, elliptical, arrow, peanut, four-petal, and three-petal—reconstructed images are, respectively, 2.95%, 3.11%, 17.81%, 15.10%, 14.14%, and 15.24%. Briefly speaking, not only can circular and elliptical buried conductors be reconstructed; some irregular shapes can be reconstructed well. On the contrary, the reconstruction result by U-Net for buried objects is worse since it is not able to obtain a good preliminary image by processing only the upper scattered field—that is, rather than the full space. In other words, our proposed deep convolutional neural network can efficiently solve the electromagnetic inverse scattering problem of buried conductors and provide a novel method for the microwave imaging of the buried conductors. This is the first successful attempt at using deep convolutional neural networks for buried conductors in the frequency domain, which may be useful for practical applications in various fields such as the medical, military, or industrial fields, including magnetic resonance imaging, mine detection and clearance, non-destructive testing, gas or wire pipeline detection, etc.

5 April 2024 in Sensors
Verified
Journal Article
Open access
PDF file
Chien-Ching ChiuChien-Ching Chiu
Chien-Ching Chiu
YLYang-Han Lee
Yang-Han Lee
Po-Hsiang ChenPo-Hsiang Chen
Po-Hsiang Chen
YSYing-Chen Shih
Ying-Chen Shih
HJHao Jiang
Hao Jiang
In this paper, we introduce a novel artificial intelligence technique with an attention mechanism for half-space electromagnetic imaging. A dielectric object in half-space is illuminated by TM (transverse magnetic) waves. Since measurements can only be made in the upper space, the measurement angle will be limited. As a result, we apply a back-propagation scheme (BPS) to generate an initial guessed image from the measured scattered fields for scatterer buried in the lower half-space. This process can effectively reduce the high nonlinearity of the inverse scattering problem. We further input the guessed images into the generative adversarial network (GAN) and the self-attention generative adversarial network (SAGAN), respectively, to compare the reconstruction performance. Numerical results prove that both SAGAN and GAN can reconstruct dielectric objects and the MNIST dataset under same measurement conditions. Our analysis also reveals that SAGAN is able to reconstruct electromagnetic images more accurately and efficiently than GAN.

11 December 2023 in IEEE Transactions on Microwave Theory and Techniques
Verified
Journal Article
Closed access
Chien-Ching ChiuChien-Ching Chiu
Chien-Ching Chiu
Po-Hsiang ChenPo-Hsiang Chen
Po-Hsiang Chen
YSYing-Chen Shih
Ying-Chen Shih
ELEng Hock Lim
Eng Hock Lim
This article utilizes artificial intelligence (AI) combined with a modified contrast scheme (MCS) technique to reconstruct microwave imaging of uniaxial objects. The 2-D backscattering problem of uniaxial objects is exposed to the transverse magnetic (TM) and transverse electric (TE) incident waves. The TE polarization problem is more severe than TM. The preliminary distribution of the dielectric constant is computed by using MCS and then compared with the dominant current scheme (DCS). MCS adds a local wave amplification coefficient to rectify the contrast and thus solves the inverse scattering problem (ISP) effectively. Moreover, U-Net is employed to reconstruct the distribution of the dielectric constants of the uniaxial objects. Different noises are added to compare the reconstruction results for MCS and DCS. We also verify the effectiveness of our proposed method by the measurement data provided by the Fresnel Institute. Numerical results show that the reconstruction is almost the same for small dielectric constants for both MCS and DCS. However, MCS overwhelms DCSs for dielectric constant reconstruction when the dielectric constant is large. MCS has successfully reconstructed uniaxial objects for higher dielectric constant distributions with better performance and high precision capability.

23 August 2022 in Journal of Electromagnetic Waves and Applications
Verified
Journal Article
Closed access
Chien-Ching ChiuChien-Ching Chiu
Chien-Ching Chiu
TKT. H. Kang
T. H. Kang
Po-Hsiang ChenPo-Hsiang Chen
Po-Hsiang Chen
HJH. Jiang
H. Jiang
YCY. K. Chen
Y. K. Chen
U-Net and Object-Attentional Super-Resolution Network (OASRN) neural network for electromagnetic imaging are compared and investigated in this paper. The outcome shows that though under limited training data, the regeneration capability is still highly reliable. We first transmit the electromagnetic waves to the scatterer and use the received scattered field information to calculate the estimated permittivity distribution by Green’s function, subspace method and Dominant Current Scheme (DCS). The estimation technique can effectively reduce the training process of the neural network modules. Next, we train the U-Net and OASRN modules for real-time images. Lastly, we used Root Mean Square Error (RMSE) and Structural Similarity Index Measure (SSIM) to compare and analyze the reconstructed images of the two neural networks. Numerical results show that the reconstructed image by OASRN is better than that by U-net with 5% or 20% Gaussian noise for different dielectric constant distributions.