Scholarlink

The rapid growth in technology has increased the applications of wireless sensor networks (WSN) in a wide range. The sensor nodes gather the surrounding information and further transmit central storage where the data are secured. However, the security of the data is decreased in the transmission process due to jamming attacks in the network. The existing model tried to generate the jamming attack detection mechanism to clear the disturbances in the data transmitting path, but failed to reach their outcome due to loss of data packets during the transmission. The gated deep reinforcement learning with sea lion optimization (GDRL with SLO) is proposed to minimize the loss of data packets during the transmission process. The gated deep reinforcement learning (GDRL) interpreted the details of the data related to jamming attacks in the network and the sea lion optimization (SLO) model enhanced the global searching ability of the GDRL that exhibited the appropriate jamming attack detection in the WSN. The developed GDRL with SLO has attained higher throughput and packet delivery ratio of 94 Mbps and 97% compared to the existing reinforcement learning-based gradient monitored (RLGM) model.

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

Gated Deep Reinforcement Learning with Sea Lion Optimization for Detecting Jamming Attacks in Wireless Sensor Networks

  • 章节

  • 发表日期：

    2026年

查看全文
  • S. Prabhu,
  • Hima Bindu Gogineni,
  • Boddepalli Prameela,
  • R. Rana Veer Samara Sihman Bharattej,
  • D. Navaneetha

摘要

The rapid growth in technology has increased the applications of wireless sensor networks (WSN) in a wide range. The sensor nodes gather the surrounding information and further transmit central storage where the data are secured. However, the security of the data is decreased in the transmission process due to jamming attacks in the network. The existing model tried to generate the jamming attack detection mechanism to clear the disturbances in the data transmitting path, but failed to reach their outcome due to loss of data packets during the transmission. The gated deep reinforcement learning with sea lion optimization (GDRL with SLO) is proposed to minimize the loss of data packets during the transmission process. The gated deep reinforcement learning (GDRL) interpreted the details of the data related to jamming attacks in the network and the sea lion optimization (SLO) model enhanced the global searching ability of the GDRL that exhibited the appropriate jamming attack detection in the WSN. The developed GDRL with SLO has attained higher throughput and packet delivery ratio of 94 Mbps and 97% compared to the existing reinforcement learning-based gradient monitored (RLGM) model.