Spatially Correlated Boltzmann Deep Learning Lamport Session Discrete Certificateless Signcryption for DoS Attack Detection and Secured WSN Communication

The fundamental properties of a Wireless Sensor Network (WSN) remain in environmental sensing and to aggregate contrasting types of data. It is utilized in numerous domains of applications, to name a few being, military surveillance, electronic healthcare monitoring, vehicular traffic monitoring, and so on. However, there are innumerable confronts and menaces, concerning the secured communication susceptibility of a WSN. This is owing to the fact that the WSN resources are said to be finite and therefore making susceptibility to numerous types of security attacks, distinctly the Denial of Service (DoS) attacks. To address these issues, in this work, a Spatially Correlated Boltzmann Lamport Session Certificateless Signcryption (SCB-LSCS) method is proposed for DoS attack detection and ensuring secured WSN communication. The SCB-LSCS method is split into two sections. In the first section, a Spatially Correlated Boltzmann Deep Learning model for DoS attack detection model is presented that with the aid of spatially correlated function optimal formation of cluster and cluster head selection is made. Followed by which the cluster head uses Boltzmann Deep Learning for DoS attack detection. In the second section, Lamport Session Discrete Certificateless Signcryption model is proposed that using Lamport Session Discrete function provides secured WSN communication. The proposed SCB-LSCS method proved to be superior, chiefly in terms of DoS attack detection accuracy, packet loss rate, routing overhead, delay and data delivery ratio, as demonstrated by carrying out comparative analysis with the existing counterparts