Scholarlink

<p>Wearable antennas can improve signal connectivity and strength and inability to deliver real-time data. However, they face difficulties like limited bandwidth, minimized efficiency, and potential signal interference. To tackle the issue, various prevailing studies have attempted to achieve improved efficacy in the design of wearable antennas with the prediction of effectiveness, but suffer from less efficiency, accuracy, over-fitting of data, etc. The proposed system addresses these limitations by designing and calculating antenna efficacy. For this, an FR-lossy substrate-based dual-band wearable antenna with U-shaped slot has been designed using a computer simulation technology (CST) tool with dimensions of 41&#xa0;×&#xa0;44&#xa0;mm<sup>2</sup>(0.33 λ<sub>0</sub>&#xa0;×&#xa0;0.35 λ<sub>0</sub> at 2.4&#xa0;GHz). In addition, two diverse materials, aluminum nitrate and pattern gain Teflon (PTFE-lossy), have been used to analyze effectiveness of a flame-retardant (FR-lossy) substrate. It has been identified that the antenna with an FR-lossy substrate deviates between 5.8&#xa0;GHz and 2.4&#xa0;GHz, making it suitable for telemedicine applications. Essentially, the presented system employs deep bi-directional long short-term memory with random forest (D-BiLSTM-RF) to predict the effectiveness in antenna design. The D-BiLSTM captures long-term dependencies but struggles with handling irrelevant features and noisy data. To resolve this, D-BiLSTM in ensemble with RF improves the accuracy and handling irrelevant features and noisy data. The efficacy of the proposed method has been evaluated by performance metrics, and comparative analysis carried out to reveal the greater efficiency of the proposed method. Moreover, the experimental results showed that the simulated and the fabricated outcomes have similar deviations and defined that the antenna is suitable for real-world application.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A Deep Learning Approach for Classifying the Efficiency of Dual-Band Wearable Antennas in Remote Healthcare

  • 期刊论文

  • 发表日期：

    2025年11月17日

查看全文
  • R. Deepalakshmi,
  • S. Mary Praveena

摘要

Wearable antennas can improve signal connectivity and strength and inability to deliver real-time data. However, they face difficulties like limited bandwidth, minimized efficiency, and potential signal interference. To tackle the issue, various prevailing studies have attempted to achieve improved efficacy in the design of wearable antennas with the prediction of effectiveness, but suffer from less efficiency, accuracy, over-fitting of data, etc. The proposed system addresses these limitations by designing and calculating antenna efficacy. For this, an FR-lossy substrate-based dual-band wearable antenna with U-shaped slot has been designed using a computer simulation technology (CST) tool with dimensions of 41 × 44 mm2(0.33 λ0 × 0.35 λ0 at 2.4 GHz). In addition, two diverse materials, aluminum nitrate and pattern gain Teflon (PTFE-lossy), have been used to analyze effectiveness of a flame-retardant (FR-lossy) substrate. It has been identified that the antenna with an FR-lossy substrate deviates between 5.8 GHz and 2.4 GHz, making it suitable for telemedicine applications. Essentially, the presented system employs deep bi-directional long short-term memory with random forest (D-BiLSTM-RF) to predict the effectiveness in antenna design. The D-BiLSTM captures long-term dependencies but struggles with handling irrelevant features and noisy data. To resolve this, D-BiLSTM in ensemble with RF improves the accuracy and handling irrelevant features and noisy data. The efficacy of the proposed method has been evaluated by performance metrics, and comparative analysis carried out to reveal the greater efficiency of the proposed method. Moreover, the experimental results showed that the simulated and the fabricated outcomes have similar deviations and defined that the antenna is suitable for real-world application.