Cristian H. Martins de Souza

University of São Paulo | USP · Department of Computer Science (IME)

Malicious software remains one of the main challenges for the security of computer systems. The growth of the Internet of Things (IoT) technological paradigm has raised several concerns regarding the security of devices connected to the Internet, especially in industrial environments driven by 4.0 technologies, where compromising or malfunctioning...

This article presents an evaluation of different machine learning algorithms for malware detection in IoT devices using the IoT-23 dataset. Models based on Random Forest, SVM, decision tree, and a convolutional neural network were implemented and compared. The results show that the Random Forest algorithm achieved the highest accuracy, while the co...

Malicious software remain one of the main challenges for the security of computer systems. The growth of the Internet of Things (IoT) technological paradigm has raised several concerns regarding the security of devices connected to the Internet, especially in industrial environments, where compromising or malfunctioning such devices can cause damag...

Malware presents a significant threat to computer systems security, especially in ARM and MIPS architectures, driven by the rise of the internet of things (IoT). This paper introduces Heimdall, a hybrid approach that integrates YARA signatures and machine learning in programmable switches for efficient malware detection in SDN-enabled IoT environme...

Over the last decade, Software-Defined Networking (SDN) has become increasingly popular in computer network infrastructures. However, due to its relatively recent implementation, protective measures still need to be fully developed. One significant security concern with SDN is its vulnerability to scanning attacks, which can escalate to more severe...

O paradigma das Redes Definidas por Software (SDN) vem sendo cada vez mais adotado nas infraestruturas de redes de computadores. No entanto, por se tratar de uma abordagem relativamente recente, poucas estratégias de segurança tem sido empregadas para sua proteção. Um dos problemas relacionados ao uso de SDN é a presença de ataques do tipo scanning...

Malware analysis is of utmost importance for information security. However, while many tools attempt to automate this process, few have a local installation option and a centralized environment for analyzing and exposing results. Furthermore, API query limitations and high license prices make these tools unattractive for newcomers to the field. Thi...

Ataques de fingerprinting constituem os primeiros passos no planejamento de ataques sofisticados em redes de computadores, como os ataques distribuídos de negação de serviço (DDoS, do inglês Distributed Denial of Service). Utilizar técnicas de alvo móvel (MTD, do inglês Moving Target Defense) em conjunto com a visão holística dos controladores das...

The expansion of the Internet has grown the possibilities for fraudulent actions. Among these possibilities, we highlight the phishing activity, created with the objective of capturing user's credentials through a false page similar to the original one. This work proposes PhishKiller, a tool capable of detecting and mitigating phishing attacks by m...

A expansão da Internet expandiu as possibilidades de ações fraudulentas. Entre essas ações, destaca-se a atividade de phishing, criada com o objetivo de capturar as credenciais do usuário através de uma página falsa similar à do serviço original. Diante disso, este trabalho propõe o PhishKiller, uma ferramenta capaz de detectar e mitigar ataques de...

With the constant technological evolution, every day human beings look for automated ways to guarantee the safety of certain places. Inevitably, the use of face recognition becomes an important mechanism for access control, since each person has practically unique face features. However, it is also necessary to develop tools together for a validati...