Computationally Efficient Surrogate-Assisted Design of Pyramidal-Shaped 3-D Reflectarray Antennas

Mahoutı, Peyman; Belen, Mehmet A.; Çalık, Nurullah; Koziel, Slawomir

doi:10.1109/tap.2022.3191131

Cited by 21 publications

(26 citation statements)

References 65 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [ 40 ], the authors construct a DNN-based end-to-end system optimization model to reduce the data at the transmitter end while improving the decoding accuracy. For the joint optimization of different blocks, the DL approach can utilize a data-driven model based on expert knowledge and a big data system [ 41 , 42 ]. Furthermore, the authors [ 43 , 44 ] provide model-based optimization approaches in the physical layer.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Deep Reinforcement Learning-Assisted Optimization for Resource Allocation in Downlink OFDMA Cooperative Systems

Tefera

Zhang

Jin

2023

Entropy

View full text Add to dashboard Cite

This paper considers a downlink resource-allocation problem in distributed interference orthogonal frequency-division multiple access (OFDMA) systems under maximal power constraints. As the upcoming fifth-generation (5G) wireless networks are increasingly complex and heterogeneous, it is challenging for resource allocation tasks to optimize the system performance metrics and guarantee user service requests simultaneously. Because of the non-convex optimization problems, using existing approaches to find the optimal resource allocation is computationally expensive. Recently, model-free reinforcement learning (RL) techniques have become alternative approaches in wireless networks to solve non-convex and NP-hard optimization problems. In this paper, we study a deep Q-learning (DQL)-based approach to address the optimization of transmit power control for users in multi-cell interference networks. In particular, we have applied a DQL algorithm for resource allocation to maximize the overall system throughput subject to the maximum power and SINR constraints in a flat frequency channel. We first formulate the optimization problem as a non-cooperative game model, where the multiple BSs compete for spectral efficiencies by improving their achievable utility functions while ensuring the quality of service (QoS) requirements to the corresponding receivers. Then, we develop a DRL-based resource allocation model to maximize the system throughput while satisfying the power and spectral efficiency requirements. In this setting, we define the state-action spaces and the reward function to explore the possible actions and learning outcomes. The numerical simulations demonstrate that the proposed DQL-based scheme outperforms the traditional model-based solution.

show abstract

Section: Related Workmentioning

confidence: 99%

“…We next define the training parameters for our DQN model. Determination of these hyperparameter values becomes more challenging in DL-based resource allocation [ 42 ]. In this work, we do not over-parameterize the structure of a neural network.…”

Section: Numerical Simulationsmentioning

confidence: 99%

Deep Reinforcement Learning-Assisted Optimization for Resource Allocation in Downlink OFDMA Cooperative Systems

Tefera

Zhang

Jin

2023

Entropy

View full text Add to dashboard Cite

show abstract

“…Data-driven surrogate modeling is one of the efficient solution methods for overcoming the mentioned drawback of high cost and computationally inefficient fullwave EM modeling. Data-driven surrogate modeling has proved its usage in the design procedure of highfrequency devices as a low-cost surrogate of the various electrical and field responses of high-frequency stages such as scattering parameters [S], 14,15 reflection phase characteristics in reflect-arrays, 16 characteristic impedance, 17 and prediction resonant frequency of antenna designs. [18][19][20] In each of the mentioned works, different types of Artificial Intelligence (AI) regression methods such as polynomial, 21,22 kriging, [23][24][25] Support Vector Regression (SVR), [26][27][28][29] Artificial Neural Networks (ANNs), [30][31][32][33][34] and Deep Learning (DL) [35][36][37][38][39] had been used to create an accurate, stable mapping between the given input space of the problem and the targeted characteristic as the output of the model.…”

Section: Introductionmentioning

confidence: 99%

Data driven surrogate modeling of horn antennas for optimal determination of radiation pattern and size using deep learning

Piltan

Kīzīlay

Belen

et al. 2023

Micro & Optical Tech Letters

View full text Add to dashboard Cite

Horn antenna designs are favored in many applications where ultra‐wide‐band operation range alongside of a high‐performance radiation pattern characteristics are requested. Scattering‐parameter characteristics of antennas is an important design metric, where inefficiency in the input would drastically lower the realized gain. However, satisfying the requirement for scattering parameters are not enough for having an antenna with high‐performance results, where the radiation characteristic of the design can be changed independently than the scattering parameters behavior. A design might have a high‐efficiency performance, but the radiation characteristics might not be acceptable. Furthermore, there are other design considerations such as size and volume of the design alongside of these conflicting characteristics, which directly affect the manufacturing cost and limits the possible applications. In this work, by using data‐driven surrogate modeling, it is aimed to achieve a computationally efficient design optimization process for horn antennas with high radiation performance alongside of being small in or within the limits of the desired application limits. Here, the geometrical design variables, operation frequency, and radiation direction of the design will be taken as the input, while the realized gain of the design is taken as the output of the surrogate model. Series of powerful and commonly used artificial intelligence algorithms, including Deep Learning had been used to create a data‐driven surrogate model representation for the handled problem, and 80% computational cost reduction had been obtained via proposed approach. As for the verification of the studied optimization problem, an optimally designed antenna is prototyped via the use of three‐dimensional printer and the experimental results ware compared with the results of surrogate model.

show abstract

“…The optimization was done by reinforcement learning. Classical methods like Gaussian process-based Bayesian optimization [17], [24], [25], [35] and evolutionary optimization [42], [50] also could be applied. Some methods [46], [51] use performance predictors together with Bayesian uncertainty estimation.…”

Section: Introductionmentioning

confidence: 99%

Multi-Fidelity Neural Architecture Search With Knowledge Distillation

et al. 2023

View full text Add to dashboard Cite

Neural architecture search (NAS) targets at finding the optimal architecture of a neural network for a problem or a family of problems. Evaluations of neural architectures are very time-consuming. One of the possible ways to mitigate this issue is to use low-fidelity evaluations, namely training on a part of a dataset, fewer epochs, with fewer channels, etc. In this paper, we propose a Bayesian multi-fidelity method for neural architecture search: MF-KD. The method relies on a new approach to low-fidelity evaluations of neural architectures by training for a few epochs using a knowledge distillation. Knowledge distillation adds to a loss function a term forcing a network to mimic some teacher network. We carry out experiments on CIFAR-10, CIFAR-100, and ImageNet-16-120. We show that training for a few epochs with such a modified loss function leads to a better selection of neural architectures than training for a few epochs with a logistic loss. The proposed method outperforms several state-of-the-art baselines.

show abstract

Computationally Efficient Surrogate-Assisted Design of Pyramidal-Shaped 3-D Reflectarray Antennas

Cited by 21 publications

References 65 publications

Deep Reinforcement Learning-Assisted Optimization for Resource Allocation in Downlink OFDMA Cooperative Systems

Deep Reinforcement Learning-Assisted Optimization for Resource Allocation in Downlink OFDMA Cooperative Systems

Data driven surrogate modeling of horn antennas for optimal determination of radiation pattern and size using deep learning

Multi-Fidelity Neural Architecture Search With Knowledge Distillation

Contact Info

Product

Resources

About