The development of innovative mobile technologies with intelligent processes has made everything interconnected through the Internet of Things (IoT). Thus, cloud technology is considered the basis for IoT. Yet, in reality, numerous security challenges face these network advancements in the form of network anomalies and threats. In fact, IoT-Cloud networks are vulnerable to attacks due to their wireless communication features and the participation of IoT devices. Thus, security has become an important issue for the basic functionality of such networks. Ensuring the reliability of collected data remains a challenging issue. IoT devices especially are too vulnerable to attackers due to their limited security resources. In this context, traditional intrusion detection systems (IDS) are becoming incompatible with the new network environment, while systems related to machine learning and deep learning are emerging. Artificial Intelligence (AI) tools are powerful techniques that could be used to achieve this purpose. In this work, we propose applying a machine-learning technique with a real IoT dataset to develop an intelligent intrusion detection system (IIDS). Data transferred between interconnected systems in the IoT-cloud networks will be encrypted. Moreover, we propose an optimal intelligent solution for the IoT energy consumption problem. This solution is based also on a machine learning tool. Our proposed solution, which appears to be more scalable than other attack detection schemes, shows promising results in boosting both the performance of security systems and the quality of services in IoT-cloud environments.
Chahira Lhioui (Sat,) studied this question.