Scholarlink

This paper presents a new trust management model that combines logistic functions, Bayesian inference, Dempster-Shafer theory, and fuzzy logic to improve the dependability and effectiveness of wireless sensor networks (WSNs). Our strategy entails conducting thorough trust assessments, evaluating energy trust, and combining the PEGASIS protocol with PEDEL-Net, a DeepWalk-LSTM-based chain leader selection algorithm. The model demonstrated tremendous improvements in network durability, with the number of active nodes increasing to 70 on the 1000th iteration and inactive nodes decreasing to approximately 30. The network demonstrated a strong and consistent flow of communication, with up to 18,000 packets sent to the base station and a throughput of 70,000 units maintained. Furthermore, there was a clear energy control implementation, as the average remaining energy dropped uniformly from 50 units at the start to nearly zero units near the end of simulation. When compared to existing methods, this method outperformed them in terms of Packet Delivery Ratio. The proposed method achieved the lowest PDR of 0.979 for 100 nodes and the highest of 0.992 for 300 nodes, indicating that it has very high efficiency in delivering packets as network size increases. This result is significantly larger than other techniques such as IBOLSR, PSOGA, GS-EERA, or EHR, which had inconsistent improvements despite being superior to traditional methods such as GS-EERA or HER. The fact that the proposed method is robust demonstrates its dependability in ensuring successful transmission and overall performance improvement of WSNs via trust-based energy management approaches.

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 Novel Approach to Trust Management in WSNs with Logistic Functions and Bayesian Inference: The PEDEL-Net Model

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Bojja Nagabhushana Babu,
  • M. Gunasekaran

摘要

This paper presents a new trust management model that combines logistic functions, Bayesian inference, Dempster-Shafer theory, and fuzzy logic to improve the dependability and effectiveness of wireless sensor networks (WSNs). Our strategy entails conducting thorough trust assessments, evaluating energy trust, and combining the PEGASIS protocol with PEDEL-Net, a DeepWalk-LSTM-based chain leader selection algorithm. The model demonstrated tremendous improvements in network durability, with the number of active nodes increasing to 70 on the 1000th iteration and inactive nodes decreasing to approximately 30. The network demonstrated a strong and consistent flow of communication, with up to 18,000 packets sent to the base station and a throughput of 70,000 units maintained. Furthermore, there was a clear energy control implementation, as the average remaining energy dropped uniformly from 50 units at the start to nearly zero units near the end of simulation. When compared to existing methods, this method outperformed them in terms of Packet Delivery Ratio. The proposed method achieved the lowest PDR of 0.979 for 100 nodes and the highest of 0.992 for 300 nodes, indicating that it has very high efficiency in delivering packets as network size increases. This result is significantly larger than other techniques such as IBOLSR, PSOGA, GS-EERA, or EHR, which had inconsistent improvements despite being superior to traditional methods such as GS-EERA or HER. The fact that the proposed method is robust demonstrates its dependability in ensuring successful transmission and overall performance improvement of WSNs via trust-based energy management approaches.