III. Problem Statement
This paper focuses on modeling and calculating trust between nodes in WSNs, based on sensed continuous data to address security issues and deal with malicious and misbehavior nodes and for assisting the decision-making process. A new trust model and a reputation system for WSNs can be proposed. The trust model establishes the continuous version of the reputation system applied to binary events and presents a new trust and reputation system for sensor Networks. This approach for mixing of second hand information from neighboring nodes with directly observed information to calculate trust between nodes in WSNs. Trust metrics can be used to evaluate the trust value of each node in the clusters. Behaviors are monitored by monitoring node (MN). Monitoring node selected at the next higher level of CH, this can also be changed dynamically along with CH. The main focus of this paper is to develop a fuzzy theory based trust and reputation model for WSNs environment.
IV. System Model
The architecture of our proposed system, consists of four major blocks namely,
i. Cluster Formation and CH selection
ii. Information Gathering
iii. Trust Evaluation and Propagation
iv. Misbehavior Detection
The detailed description about the architecture is as follows.
Fig. 2. Overall Architecture of the Proposed System
Fig.2. shows the overall architecture of the proposed work. In wireless sensor networks, the sensor nodes are densely deployed in the region. After deployment, the nodes are group together to form a clusters based on various criteria such as location and communication range. Cluster formation involves the establishment of cluster head, monitoring node and member nodes. In which the nodes are broadcast their energy and ID to base station. The node which has maximum energy that will be elected as Cluster Head (CH). Other nodes become members of clusters or local nodes. The nodes update their energy; accordingly the Cluster Head can change dynamically. The various type of trust factors are used to collect the information from the sensor nodes. The information can be gathered by direct observation and indirect observations. Trust evaluation involves the process of computation of the node’s trust value based on the trust model. The node’s trusts such as the node’s historical trust values and the node’s current trust values are computed. The node’s historical trust value is calculated by observing the node’s behavior. Based on the historical trust value, node’s capability level calculated, the node’s current trust is calculated by using the fuzzy rules prediction. This two trust value is of great importance and plays a major role in establishing trusted information to the monitoring node (MN). Based on that, trust value can be aggregated; if the threshold level is low then it will be identified as malicious node by base station. If the threshold level is high, then it will be used for communications. After that,...