Energy-efficient classification strategy for detecting interference and malicious sensor nodes in wireless body area Networks

Abstract

Wireless Body Area Networks (WBANs) play a vital role in healthcare monitoring, using wireless sensors to track physiological parameters and predict illness onset. This study proposes a novel approach for detecting interference and malicious sensor nodes in WBANs, crucial for maintaining system integrity and performance. The method combines feature-based techniques with classification strategies to accurately identify anomalies. Features are taken from WBAN nodes and used to train Support Vector Machine (SVM) classifiers, which makes interference detection work well. A neurofuzzy inference system (ANFIS) classifier is also used to train the system on trusted and untrusted nodes at the start, which makes classification easier in real-world WBAN situations. Link failures due to rogue sensor nodes can severely impact WBAN performance, emphasizing the need for efficient detection and correction mechanisms. The proposed strategy introduces a weight metric to identify broken links, enhancing system reliability. Evaluation metrics, including LFD latency and packet delivery ratio, are analyzed to assess the efficacy of the approach. By improving interference detection and addressing link failures, this study contributes to enhancing the efficiency and reliability of WBAN networks, critical for advancing healthcare monitoring technologies.