Enhanced support vector machine-based intelligent classification of trusted nodes in WBAN for Resilient Infrastructure
In various medical settings, ranging from hospitals to mental health care facilities and even homes, the Wireless Body Area Network (WBAN) assumes a critical role in enhancing the real-time monitoring of patients' overall health. The significance of the WBAN has grown recently due to its fundamental utility and its broad array of applications within the medical domain. As the data being transmitted across the WBAN infrastructure pertains to sensitive patient information, ensuring its security remains a matter of paramount importance. The establishment of a strong security framework holds immense necessity for any WBAN network to ensure the secure exchange of data between sensor nodes and other WBAN networks. This document introduces the Extended Support Vector Machine (ESVM) as an approach to differentiate trusted nodes within WBAN networks. This differentiation is accomplished through a classification method that reinforces the security dimensions of these networks. By employing Kernel-based Independent Component Analysis (K-ICA), relevant features are extracted from the data. The results of conducted tests unequivocally demonstrate that, when compared to various methods used previously, the proposed ESVM technique outperforms all of them in terms of its capacity to accurately classify trusted WBAN nodes in process innovation.
Volume & Issue