Identifying DoS attacks on IoT devices using deep learning

Presented by: Deborah Mesquita
The ESP32 is a series of microcontrollers with integrated Wi-Fi. They offer an opportunity to empower a variety of IoT projects because they’re both low-cost and low-power. Security should be an essential part of the development of IoT systems, and these devices are easy targets for cyberattacks because of the poor cycles of update and maintenance. In this talk, we’ll see how we used deep learning and network traffic analysis knowledge to train and embed a model on the ESP32 capable of preventing DoS attacks. IoT and machine learning combined can solve a lot of today’s problems (including problems of developing countries) and our main goal is to show that with low-cost devices we can bring IoT project ideas to real life.