Low-Power Security Solutions for IOT Applications

Abstract

The Internet of Things (IoT) is a technical advancement that has transformed civilization. The Internet of Things will irrevocably transform our use of basic objects into intelligent, fully functional devices. IoT devices may execute and automate routine domestic and occupational operations using basic sensors. Despite the advantages of these gadgets, they remain susceptible to infringements, including privacy concerns and security breaches. This study seeks to elucidate the vulnerabilities of IoT devices and the contemporary dangers they face. Additionally, the technologies used in the IoT are analyzed, along with the various communication layers of the IoT and their operations. The results indicate that IoT devices are susceptible to several software and hardware vulnerabilities, as well as the associated problems of IoT. Proposed solutions to these difficulties include the implementation of anomaly-based intrusion detection systems, which are essential elements of network security. Employing machine learning (ML) for the identification of possible assaults is advisable. Numerous suggested anomaly-based detection systems use various machine learning algorithms and methodologies. Nonetheless, there exists no standardized baseline for comparing them regarding power usage. A benchmark is provided to assess both accuracy and power usage for evaluating each algorithm's implementation.