A Self-Adaptive Intrusion Detection System for Securing UAV-to-UAV Communications Based on the Human Immune System in UAV Networks
Reza Fotohi,Masoud Abdan,Sanaz Ghasemi
DOI: https://doi.org/10.1007/s10723-022-09614-1
2022-07-16
Journal of Grid Computing
Abstract:Today, the use of Unmanned aerial vehicles (UAVs) technology has made significant progress due to its widespread use. UAVs are widely used in agriculture, commercial, military, civilian, and environmental applications. Therefore, for better efficiency, the UAV must be able to communicate effectively through the communication of the UAV system with the UAV and other communication of the UAV with the network infrastructure through the communication of the UAV with the infrastructure. In particular, security issues are a serious concern in such networks because the top-secret information exchanged between UAVs is prone to various attacks such as Sybil, blackhole and Flooding attacks. This paper proposes a method called SID-UAV that is resistant to malicious UAVs that threaten UAV to UAV communications. The SID-UAV method uses a self-matching system that detects the safest path between UAVs. This method includes various phases, including the decision-making phase, the path discovery and investigation phase, the destructive UAV response phase, and the information database registration phase. Also, in the SID-UAV method, three types of modules of route investigation, decision module and defense module are considered. Each of these modules is distributed in different parts of UAV networks and has sub-modules. These sub-modules have their own tasks and to perform and process information quickly, it is connected to the knowledge base to quickly record information and use the stored information. The performance of the SID-UAV method in NS-3 has been evaluated and tested. The evaluation results show the superiority of SID-UAV method over BRUIDS, SFA and SUAS-HIS methods in Packet Delivery Ratio (PDR), Packet Lost Ratio (PLR), Average False Positive (AFP), Average False Negative (AFN) and Average Detection Ratio (ADR).
computer science, information systems, theory & methods