Improved PSO-based adversarial model for WSN

WSNs generally assume the presence of a centralised entity that acts as a centralised data collection point. Sometimes the operating systems may become hostile in an unattended environment, which may result in a real danger of node and data compromise. Thus, in the presence of a powerful adversary, securing data stored on unattended sensors presents interesting challenges and opens an exciting new line of research. Thus, it becomes very important to study and analyse the different adversarial models and defence techniques. In our proposed work, we have designed a robust adversarial model by considering special features of the hostile WSN environment. An improved particle swarm optimisation (PSO) algorithm is presented that uses a multi-objective function to compromise the WSN with maximum node contribution and minimum resource expenditure. It is shown that the 'improved-PSO' required less iterations, resource expenditure, and time to compromise the network as compared with existing models.