ML Based Light-Weight Security Protocol for IoFT Device

Forest plays a vital role in environmental change and the protection of forests is the utmost important thing. The study attempts to suggest a solution for a smart forest (Internet of Forest Things). As nowadays, the Internet of Things is emerging very vastly and a large number of researchers are working towards the field of Internet of Forest Things (loFT). The protection of smart forests is of utmost importance because the infrastructure is physically unprotected and they generate huge data about the forest. Due to scalability, smart forest controllers are small-sized and low-cost devices with less computational capabilities. Keeping it in mind, the designed algorithms should be lightweight so that they can meet the requirements. To address this issue, this paper proposes a machine learning approach to develop a lightweight security solution to address data altering and man-in-the-middle attacks. The algorithm generates a link fingerprint of the received signal strength indicator (RSSI) value of the communicating IoFT node. The packet is sent to the server unit and it is decoded and matched with the table to identify the path. A trained machine learning model at the server gets the RSSI value from the packet and successfully classify about the presence of an attack.