Exploración de los problemas de seguridad que presenta el IoT en el contexto del Edge Computing.

The rapid growth of Internet of Things devices and the large amount of data generated at the network perimeter have caused the traditional Cloud Computing model to become a bottleneck, due to bandwidth limitations and high latency between the interaction of smart devices and centralised servers. This problem has led in recent years to the advancement of Edge Computing or Fog Computing, which provides analysis and processing capabilities closer to where data is generated, thus offering great services to IoT environments that require immediate response or those that have very limited connectivity to the cloud. Taking into consideration the amount of information that will be exposed, the need arises to study in depth the security in the scenarios posed by these new paradigms, based on the hypothesis that the Internet itself is not secure; therefore, the Cloud and Edge Computing that make use of it, will not be the exception; for this reason, the need to conduct a research where the new paradigm called Edge Computing will provide sufficient security in the processing, storage and response of data arises. Consequently, the purpose of this degree project was to carry out a Systematic Literature Review (SLR) that allowed us to establish the security problems that currently affect the IoT in the context of Edge Computing, based on the three phases of the SLR methodology proposed by Torres-Carrión, which is based on the Kitchenham and Bacca protocol. From the studies analysed, 24 security problems were identified, which are identified as attacks that mostly affect the IoT and are also the most frequent in Edge and Fog Computing. Furthermore, as an additional contribution, some solutions proposed by the authors to prevent certain security attacks were identified; likewise, it was determined that: location awareness, geographic distribution, decentralised architecture, mobility support, scalability and reduced energy consumption are characteristics that serve as entry points for security attacks to occur. Finally, it was concluded that the most frequent security problems in IoT such as: DoS attack, DDoS, man-in-the-middle, spoofing, replay, collision, physical tampering, fake node injection and device cloning, also affect Edge Computing by its very nature, since this paradigm is born thanks to the existence of IoT and is implemented on the basis of it, which itself has many security flaws, and most of these problems are because manufacturers and developers focus more on the utility that the devices can give to the user than on security; Also, it is important to emphasise that the security mechanisms widely studied and used in traditional computing cannot be directly implemented in Edge Computing, because most of them require a lot of computational capacity, and also, due to the heterogeneity of the devices that make up the Edge nodes in the different application environments, it is not possible to design generic security mechanisms for all Edge Computing environments.

• Autor Principal: Solano Chamba., Nora Cecivel (author)
• Formato: bachelorThesis
• Idioma: spa
• Publicado: 2022
• Subjects: <INTERNET OF THINGS > <FOG COMPUTING> <EDGE COMPUTING> <FOG>< EDGE >
• Acceso en liña: https://dspace.unl.edu.ec/jspui/handle/123456789/24789