A Secure Cloud-Based Infrastructure for Virtual Sensors in IoT Environments Tahira Jehan Makda College of Business Management University of Doha for Science and Technology Doha, Qatar tahira.makda@udst.edu.qa Ana Luiza Barros Computer Science Program State University of Cear´ a Fortaleza, Brazil analuiza.barros@uece.br Selma Dilek Department of Computer Engineering Hacettepe University Ankara, Turkey selmadilek@hacettepe.edu.tr Abstract—Recent work has proven the effort of researchers to integrate small sensors and a cloud environment, delivering the Internet of Things (IoT). Sensors as a service are one of the leading research concerns in this context. Nevertheless, security is becoming one of the most significant attributes of the IoT as sensors become more human-independent and are being extensively used to monitor human lives. That way, IoT brings many key security challenges that need attention, some of which we address in this position paper. We present a cloud- based infrastructure that can deliver sensors and actuators as a service, providing secure communication between them and the control nodes on which IoT applications rely while implementing Big Data algorithms. For mapping our proposal, two scenarios related to Health Assistance are discussed, considering secure communications in a sensor network. In conclusion, we propose a scope for future research in this field considering digital twin concepts. Since domains exploiting IoT technologies can benefit from adopting Digital Twin, our goal is to evolve a virtual sensor and actuator system into this technology. Index Terms—Security, Virtual Sensors, Actuators, Internet of Things (IoT), Big Data Algorithms, Digital Twin I. I NTRODUCTION The Internet of Things (IoT) is a new paradigm of commu- nication that is quickly gaining space in telecommunications. The main idea behind IoT is the ubiquitous presence of a variety of objects or “things” such as Radio Frequency Identification (RFID), sensors, actuators, and smartphones, which can reach a common goal through cooperation [1]. IoT is a progression of the Internet to an interconnected network of objects that not only collect environmental information and interact with the physical world, but also use existing Internet protocols to provide services related to information transfer, data analysis, applications, and communication [2]. This evolution will lead to an interconnection between people on an unprecedented level of scale and rhythm, forming an interconnected environment between objects to create an intelligent environment. Unquestionably the IoT paradigm will have a high socioeconomic and political impact on various aspects of humanity, such as in agriculture, health, transport, industry, education, vehicles, and domiciles. The International Data Corporation (IDC) company has forecasted that there will be a total of 41.6 billion connected IoT devices of all kinds, referred to as “things”, by 2025, primarily driven by industrial and automotive equipment and embracement of smart home and wearable devices [3]. Similarly, IoT brings many key security challenges that need attention, including hacking internet devices, surveillance issues, and privacy concerns, which have already captured public interest [4]. Additionally, IoT devices and servers that are poorly secured are prone to cyber-attacks [4]. Such an enormous number of connected objects generates a massive amount of data. Related tasks such as collecting, classifying, and analyzing these data have opened possibilities for new IoT applications that support real-time environment analysis for intelligent decision-making, health monitoring, and other utilities. Data centers provide a powerful and robust infrastructure to process a large amount of data in a scalable way at a low cost. As proposed by the authors of [5], the virtualization of sensors is a starting point to deliver them as a service to IoT applications. A virtual sensor is a virtual replica of a sensor that can monitor and control its physical representation. Virtual sensors can stay in the firmware of a physical sensor, in a gateway, or in a cloud. In [6], the authors implemented what is proposed by [5]; i.e., that IoT applications stay in the cloud and that virtual sensors are embedded in gateways. Security, however, still needs to be addressed. Several issues related to security are discussed in [7]. They proposed that three dimensions of problems can be addressed: preprocessing, processing and runtime. Preprocessing creates commands that a sensor can receive. Processing comprises identity management, access management, and data privacy. Furthermore, runtime ensures that gathered data are not forged and keys are not compromised. In our work, to the best of our knowledge, we propose the first solution, which can deliver virtual sensor networks and all aspects of the processing dimension. In [8], the authors addressed the concern of protecting IoT units from network-based attacks, thus ensuring a high level of availability of the IoT apparatus. The implementation stage is designed with an exclusive function schema, “mirrored” in a cloud computing executed machine, which is resistant to attacks, thus offering high availability. The IoT apparatus adopts a dedicated protocol synchronized to communicate with the mirror machine where all IoT state information and modifications are transmitted, while any end-user applications 156 2023 Sixth International Conference of Women in Data Science at Prince Sultan University (WiDS PSU) 978-1-6654-7723-9/23/$31.00 ©2023 IEEE DOI 10.1109/WiDS-PSU57071.2023.00041 2023 Sixth International Conference of Women in Data Science at Prince Sultan University (WiDS PSU) | 978-1-6654-7723-9/23/$31.00 ©2023 IEEE | DOI: 10.1109/WiDS-PSU57071.2023.00041 Authorized licensed use limited to: Qatar National Library. Downloaded on August 22,2023 at 12:48:33 UTC from IEEE Xplore. Restrictions apply.