A Game theory based Intrusion detection system for low constrained IoT devices.

Through shady networks, heterogeneous devices are connected to build the Internet of Things (IoT). However, protecting an IoT network is of the utmost significance. The best option for protecting IoT networks is cryptography. Encrypting every piece of data is not desired, though, as it places a heavy computational burden on IoT devices with limited processing power. As a viable alternative, anomaly-based intrusion detection systems offer a way to secure low constraint devices. The majority of recent research has suggested employing energy- and computationally-intensive machine learning algorithms for IDS. The lack of a defined framework for recognising the attack is another challenge for machine learning-based intrusion detection systems (IDS), which must repeatedly train for each new attempt. This research suggests an energy-saving and false-positive rate-lowering IDS that is based on game theory. The suggested IDS offers a mathematical framework for attack detection. Furthermore, as the proposed IDS is only triggered in response to the appearance of a new attack signature, the system is appropriate for devices with low constraints.