Scholarlink

<p>Due to the proliferation of Internet of Things (IoT)-based Wireless Sensor Networks (WSN) technology in different phases of various sectors, comprehensive solutions related to energy consumption and security are indispensable. However, in traditional methods for energy optimization and anomaly detection in WSNs there are certain issues: An inability to process real-time data, unavailability to extend the adaptation to behave and respond with dynamic network conditions, and poorly suited for continuously evolving anomalies. These limitations contribute to the decrease of the system’s overall performance and dependability in actual big and complicated sociotechnical networks. To overcome these challenges, this research presents LEGO-WSN (Long Short-Term Memory (LSTM) with Attention Mechanism and Genetic Algorithm (GA) Optimization for WSNs), an intelligent solution for improving energy efficiency and real-time faulty node identification of WSNs. The proposed LEGO-WSN combines GA and LSTM to enhance energy optimization and improve the detection of anomalies. The work discussed here introduces a novel approach called LEGO-WSN, which incorporates LSTM with the help of an attention layer along with a Genetic Algorithm for the operation of fault diagnosis in anomaly detection. The GA improves the network transmission parameters and plans the sensor’s operations, while the LSTM structure, complemented by attention mechanisms, identify the features of time series to encode blackhole attacks. The data set for the purpose of this study can be obtained from Kaggle and is a real life WSN data with variety of different environmental and network conditions. The impact of the proposed GA-LSTM model is measured in terms of energy consumption and real-time anomaly detection while also aims at flexibility in terms of the network environment in which it is implemented. The results show substantial enhancements in energy efficiency, with a 20% reduction in energy consumption, and high accuracy in anomaly detection, achieving 99% accuracy, 98% precision, and 99% recall. LEGO-WSN demonstrates a novel, scalable, and reliable solution for optimizing WSN performance while enhancing security and energy efficiency.</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

Next generation AI powered framework for autonomous energy optimization and real time anomaly detection in IoT driven wireless sensor networks

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年11月20日

查看全文
  • M. Parameswari,
  • Nancy P,
  • R. Jeya Malar

摘要

Due to the proliferation of Internet of Things (IoT)-based Wireless Sensor Networks (WSN) technology in different phases of various sectors, comprehensive solutions related to energy consumption and security are indispensable. However, in traditional methods for energy optimization and anomaly detection in WSNs there are certain issues: An inability to process real-time data, unavailability to extend the adaptation to behave and respond with dynamic network conditions, and poorly suited for continuously evolving anomalies. These limitations contribute to the decrease of the system’s overall performance and dependability in actual big and complicated sociotechnical networks. To overcome these challenges, this research presents LEGO-WSN (Long Short-Term Memory (LSTM) with Attention Mechanism and Genetic Algorithm (GA) Optimization for WSNs), an intelligent solution for improving energy efficiency and real-time faulty node identification of WSNs. The proposed LEGO-WSN combines GA and LSTM to enhance energy optimization and improve the detection of anomalies. The work discussed here introduces a novel approach called LEGO-WSN, which incorporates LSTM with the help of an attention layer along with a Genetic Algorithm for the operation of fault diagnosis in anomaly detection. The GA improves the network transmission parameters and plans the sensor’s operations, while the LSTM structure, complemented by attention mechanisms, identify the features of time series to encode blackhole attacks. The data set for the purpose of this study can be obtained from Kaggle and is a real life WSN data with variety of different environmental and network conditions. The impact of the proposed GA-LSTM model is measured in terms of energy consumption and real-time anomaly detection while also aims at flexibility in terms of the network environment in which it is implemented. The results show substantial enhancements in energy efficiency, with a 20% reduction in energy consumption, and high accuracy in anomaly detection, achieving 99% accuracy, 98% precision, and 99% recall. LEGO-WSN demonstrates a novel, scalable, and reliable solution for optimizing WSN performance while enhancing security and energy efficiency.