Scholarlink

<p>It has become more vital than ever to have reliable environmental monitoring and forecasting systems able to operate in real-time. This paper presents an enhanced framework for the transmission and processing of real-time climatic data using Terahertz (THz) communication in combination with a hybrid deep learning model comprising Convolutional Neural Networks and Long Short-Term Memory (CNN-LSTM) networks. It is constructed specifically to support early warning systems and disaster management systems, the system's basic goal is to greatly increase the speed, accuracy, and dependability of the interpretation of climate data. The system that has been proposed for deployment uses THz communication to facilitate the flow of data at high rates between processing units and sensor nodes. Data pertinent to climate including temperature, humidity, wind speed, and air pressure is gathered by means of a network of dispersed sensors. While the LSTM network models temporal connections and trends, the CNN component extracts spatial information from raw data. Extensive simulations and tests based on data acquired from the actual world have shown that, in comparison to more conventional approaches to machine learning, the CNN-LSTM model greatly increases prediction accuracy. The model is completely capable of attaining a high degree of accuracy in the early identification of possible disasters like floods, cyclones, and wildfires. The technology is suitable for very important real-time applications as THz communication ensures very low latency and quite data loss. The results of this study underline, in the framework of the evolution of resilient and smart disaster response systems, the synergy that exists between intelligent data analytics and wireless communication capabilities. The proposed method gradually improves the data transmission rate by 98.75, prediction accuracy by 94%, latency by 32%, sensor network coverage by 96%, temporal modelling capability by 98.2%, spatial feature extraction by 96.1%, and disaster detection scope by 99%.</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

Real-Time Climate Data Transmission and Processing Using CNN-LSTM in Terahertz Communication for Disaster Management and Early Warning Systems

  • 期刊论文

  • 发表日期：

    2026年6月19日

查看全文
  • Kamaraja Pandian P,
  • Usha Rani B.,
  • V. Priya,
  • S. P. Manikandan,
  • Sivaram Murugan,
  • V. Porkodi,
  • Mohammed Ali Shaik

摘要

It has become more vital than ever to have reliable environmental monitoring and forecasting systems able to operate in real-time. This paper presents an enhanced framework for the transmission and processing of real-time climatic data using Terahertz (THz) communication in combination with a hybrid deep learning model comprising Convolutional Neural Networks and Long Short-Term Memory (CNN-LSTM) networks. It is constructed specifically to support early warning systems and disaster management systems, the system's basic goal is to greatly increase the speed, accuracy, and dependability of the interpretation of climate data. The system that has been proposed for deployment uses THz communication to facilitate the flow of data at high rates between processing units and sensor nodes. Data pertinent to climate including temperature, humidity, wind speed, and air pressure is gathered by means of a network of dispersed sensors. While the LSTM network models temporal connections and trends, the CNN component extracts spatial information from raw data. Extensive simulations and tests based on data acquired from the actual world have shown that, in comparison to more conventional approaches to machine learning, the CNN-LSTM model greatly increases prediction accuracy. The model is completely capable of attaining a high degree of accuracy in the early identification of possible disasters like floods, cyclones, and wildfires. The technology is suitable for very important real-time applications as THz communication ensures very low latency and quite data loss. The results of this study underline, in the framework of the evolution of resilient and smart disaster response systems, the synergy that exists between intelligent data analytics and wireless communication capabilities. The proposed method gradually improves the data transmission rate by 98.75, prediction accuracy by 94%, latency by 32%, sensor network coverage by 96%, temporal modelling capability by 98.2%, spatial feature extraction by 96.1%, and disaster detection scope by 99%.