Artificial Intelligence-based Cayley Menger Securex-means Clustering and Hyper Gradient Reliable Data Transmission in Wsn

Abstract

WSN is a self-configured wireless network with no infrastructure which is made up of a huge number of dispersed sensors that are installed randomly. In a WSN, the sensor nodes connect wirelessly and are utilized for directing and observing the system and physical elements in a specified region. Wireless sensor nodes are interrelated to the base station which operates as WSN System's processing unit. The sensed data is shared by the WSN base station via the internet.WSN is utilized in the area of examination, processing, storage, and mining industries. Moreover, direct sensed data delivery to the sink node is a requirement in WSNs as far as large-scale networks are concerned. The utilization of clustering can permit the direct broadcast of information to the corresponding sink node, bringing about substantial energy reductions. In addition, the selection of a significant cluster head would also result in the saving of energy considerably. This proposed method develops an Artificial Intelligence-based CayleyMenger secure X-means cluster formation and Hyper Gradient DescentTrusted Neighbor Establishment (CMX- HGDTNE) for reliable data transmission. Secure cluster formation and chosen of cluster head performed by employing the Cayley–MengerSecure X-means Cluster Formation algorithm. Second Hyper Gradient Descent and Determinist Trusted Neighbor Establishment for reliable data transmission is applied to the clusters being formed via trusted neighbor identification within the clusters based on data traffic via Pearson Correlation between sensor nodes in the cluster. Experimental evaluation is carried out by several performance parameters, like, energy consumption, cluster formation time, PDR and packet drop rate. CMX-HGDTNE technique for secure cluster formation and reliable data transmission improves performances of PDR by 15%, and minimizes energy consumption by 40%, clustered time by 17% as than the conventional methods.