Cristiano Antonio Souza

Federal University of Santa Catarina | UFSC · Departamento de Informática e Estatística

In the “first part” will be considered the authentication of devices with restricted memory that still presents significant problems, since the memory consumption is high in the mutual authentication using cryptographic protocols. According to previous studies, the development of an efficient method to perform mutual authentication, with multifactor, remains a challenge for IoT-Fog-Cloud environments. The present work aims to improve a multifactor mutual authentication method, using a variable and adjustable response time, challenge and nonce response function. With these factors, the method can be improved for Fog and Cloud Computing contexts. Future results will be compared with evaluations carried out in related works and in our previous works, seeking to obtain a satisfactory result in terms of energy consumption and processing and communication costs. We will also use the Proverif tool and do an informal analysis to provide the security assessment. In the “second part”, in Internet of Things (IoT) systems, information of various types is continuously captured, processed and transmitted by systems usually interconnected by the Internet and distributed solutions. Attacks to capture information and overload services are common. This fact makes security techniques indispensable in IoT environments. Intrusion detection is one of the vital security points, designed to identify attack attempts. The characteristics of IoT devices make it impossible to apply these solutions in this environment. Furthermore, existing anomaly-based methods for multiclass detection do not have acceptable accuracy. An intrusion detection architecture will be proposed that will operate in the fog computing layer (Fog), aiming to classify events into specific types of attacks or non-attacks, for the execution of countermeasures. We will improve a hybrid method of binary classification called DNN-kNN. The approach will be based on Deep Neural Networks (DNN) and the k-Nearest Neighbor (kNN) algorithm. In the experiments, the public databases NSL-KDD and CICIDS2017 will be used in order to obtain greater precision about classical machine learning approaches and recent advances in intrusion detection for IoT systems. In the “third part” we will improve an Autonomic System to manage energy consumption in Internet of Things (IoT) and Fog Computing devices. The proposal will introduce advanced orchestration mechanisms to manage dynamic duty cycles for extra energy savings. The solution will work by adjusting the cycles as “Home (being at home)” and “Away (being away from home)” change states based on contextual information such as environmental conditions, user behavior, behavior variation, usage regulations energy and network resources, among others. A performance evaluation will be carried out through a proof of concept, considering average energy savings when increasing a scheduling system, and variables of long sleep cycles. We will also aim to promote autonomous management as a solution to develop more efficient buildings for energy use and smart cities, contributing to generate more sustainability.