“…Consider the observations and , where the is the sample of Gaussian-distributed stochastic process with variance σ and mean µ . Based on the observations, the GP predicts the value of for next ; this correlation function [ 25 , 26 , 27 , 28 ] is written as where the dimension of is represented by , and and are the hyper parameters. The liklihood function is calculated as where is an vector, and is covariance matrix.…”
Section: Theoretical Investigations For Antenna Design and Proposed A...mentioning
The performance of wireless networks is related to the optimized structure of the antenna. Therefore, in this paper, a Machine Learning (ML)-assisted new methodology named Self-Adaptive Bayesian Neural Network (SABNN) is proposed, aiming to optimize the antenna pattern for next-generation wireless networks. In addition, the statistical analysis for the presented SABNN is evaluated in this paper and compared with the current Gaussian Process (GP). The training cost and convergence speed are also discussed in this paper. In the final stage, the proposed model’s measured results are demonstrated, showing that the system has optimized outcomes with less calculation time.
“…Consider the observations and , where the is the sample of Gaussian-distributed stochastic process with variance σ and mean µ . Based on the observations, the GP predicts the value of for next ; this correlation function [ 25 , 26 , 27 , 28 ] is written as where the dimension of is represented by , and and are the hyper parameters. The liklihood function is calculated as where is an vector, and is covariance matrix.…”
Section: Theoretical Investigations For Antenna Design and Proposed A...mentioning
The performance of wireless networks is related to the optimized structure of the antenna. Therefore, in this paper, a Machine Learning (ML)-assisted new methodology named Self-Adaptive Bayesian Neural Network (SABNN) is proposed, aiming to optimize the antenna pattern for next-generation wireless networks. In addition, the statistical analysis for the presented SABNN is evaluated in this paper and compared with the current Gaussian Process (GP). The training cost and convergence speed are also discussed in this paper. In the final stage, the proposed model’s measured results are demonstrated, showing that the system has optimized outcomes with less calculation time.
“…Multiple-input multiple-output (MIMO) systems can also play a critical role in 5G networks, as they can improve channel capacity and communication reliability. However, meeting the stringent requirements of 5G technology may necessitate the use of larger MIMO antenna arrays [7][8][9]. Antennas are designed for good performance in terms of efficiency, peak gain, and radiation patterns supporting multiple wireless standards [10].…”
The paper focuses on the design, measurement, and performance analysis of a high-gain cross-orthogonal series fed two dipole antenna (STDA) arrays with side-wall reflectors. The antenna is specifically designed for 4G Long Term Evolution (LTE) and sub-6 GHz 5G band applications. The designed antenna is capable of operating at multiple frequencies aiming to support 4G LTE and the sub-6 GHz 5G application bands. To improve the radiation characteristics and prevent coupling effects in the presence of side-wall reflectors, parasitic strip pair directors are included in the antenna design. Furthermore, the performance of the designed STDA is evaluated by forming different array configurations, such as 2 × 1, 2 × 2, and 2 × 3 arrays. The various array configurations are proposed to investigate the effect of the projected array arrangements on the radiation pattern, impedance bandwidth, and gain characteristics. The results of the measurements show that the radiation characteristics of the antenna have been improved significantly. The proposed antenna operates at six distinct frequencies for S11 ≤ 10 dB. The operating frequencies at 1.8, 2.35, and 2.6 GHz can be utilized for LTE and 3.2, 4.2, and 5.2 GHz which can support sub-6 GHz 5G bands. The antenna is characterized by its compact size, measuring around 89 mm × 71 mm, while still achieving high gain of 12.3 dB for single STDA element with parasites and with reflector. These results emphasize the importance of the proposed design, which incorporates parasitic strip pair directors and side-wall reflectors. This design methodology plays a crucial role in enhancing the performance of the prescribed STDA array for both 4G LTE and sub-6 GHz 5G applications.
“…The higher the number of arrays, the higher the antenna gain value. The authors in [13] designed a microstrip Yagi antenna appropriate for wireless fidelity (Wi-Fi) applications. The authors focus on developing the antenna by enhancing the gain using the array method with two branch elements.…”
<span lang="EN-US">This paper aims to investigate and design a Yagi disc antenna with a variable number of director elements for Band 3 in <a name="_Hlk139294022"></a>fourth-generation long term evolution (4G LTE) mobile applications. The array technique was introduced by increasing the number of director elements to achieve superior results and better performance, such as higher gain and lower return loss. Initially, the simulated results of return loss and gain with one director element were -19.02 dB and 8.51 dBi, respectively. Then, by increasing the number of directors to three and five elements, the antenna’s performance improved significantly from -32.44 to -42.68 dB for return loss and from 8.51 to 11.17 dBi for gain, respectively. The simulated circular Yagi disc antenna provided a response in the range of 1.78 to 1.82 GHz. Therefore, a model was fabricated and tested to validate the antenna design. The measured results matched well with the simulated ones. By increasing the number of director elements, the measurement results of gain and return loss at a frequency of 1.8 GHz also showed improvement from 7.70 to 11.09 dBi and from -27.31 to -32.91 dB, respectively. Meanwhile, the measured antenna provided a wider bandwidth in the range of 1.72-1.82 GHz.</span>
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.