A systematic analysis of security and performance in the internet of drones: re-evaluation and enhancement of the PMAP protocol

Abstract

This dissertation investigates authentication protocols applied to the context of the Internet of Drones (IoD), an environment characterized by high mobility, computational constraints, and strict security requirements. The work is divided into two complementary parts. In the rst part, a systematic literature review is conducted, following established methodological guidelines, with the objective of identifying, classifying, and analyzing recent authentication and key agreement protocols for IoD. Aspects such as communication topologies, threat models, employed cryptographic primitives, security properties analyzed both formally and informally, and performance metrics are examined. In the second part, a technical reassessment of the PMAP protocol is carried out, encompassing the Drone-to-ZSP (D2Z) and Drone-to-Drone (D2D) scenarios. The protocol is analyzed from di erent perspectives, including the formal Real-or-Random (RoR) model, automated veri cation using tools such as AVISPA, Scyther, and ProVerif, an informal analysis of resistance to attacks, and an evaluation of communication and computational costs. The results indicate that, under the adopted models, no violations of the speci ed security properties were identi ed in the formal veri cation procedures. However, the analysis also reveals limitations related to the absence of Perfect Forward Secrecy, exposure to privileged insider attacks, the lack of speci c mitigation mechanisms against denial-of-service attacks, and the computational cost of Hénon map based operations on constrained platforms. From a performance perspective, the protocol proved to be executable in an embedded environment, but with signi cant latency, especially in the D2D scenario. Based on the obtained results, it is observed that the advancement of the eld depends not only on the proposal of new cryptographic mechanisms, but also on the adoption of more standardized and robust evaluation methodologies, combining formal models, informal analyses, and compatible metrics that enable comparison across di erent studies. In this way, this dissertation contributes to a more structured understanding of the properties, limitations, and evaluation challenges of authentication protocols applied to the Internet of Drones.