چكيده به لاتين
The Internet of Things (IoT) technology, as a global network, has enabled communication between humans,
between humans and objects, and between objects themselves by providing a unique identity for each object.
However, IoT devices generate a vast amount of data, and processing this data is challenging due to limitations in storage, processing power, and battery life. Therefore, offloading data and processing to edge servers is essential to ensure faster processing and timely decision-making. Edge computing, by providing computational and storage resources closer to IoT devices, reduces latency compared to cloud computing. In traditional methods, data generated by IoT devices is sent to a central server, analyzed, and then feedback is returned to the device. These methods not only lack real-time feedback but also risk delivering results that lose their value due to feedback delays, which can lead to significant errors. In this context, digital twins, as virtual representations of physical objects, play a critical role. They enable complex modeling, the transfer of large amounts of data, prediction, monitoring, control, and decision-making, allowing for remote testing in a virtual environment with lower costs and greater confidence. However, integrating digital twins with mobile IoT devices presents challenges, as mobility can disrupt real-time interactions and service delivery. Therefore, the placement of digital twins must take into account the mobility patterns of IoT devices to reduce latency and improve the quality of service for mobile users. A review and comparison of existing works revealed challenges that were not addressed, including IoT device mobility, energy consumption, access and communication latency, digital twin migration, and the resource limitations of edge servers. Hence, this research considers aspects such as the mobility and prediction of IoT devices, transmission latency, and energy consumption of IoT devices. The aim of this research is to propose a mobility-based method for the placement of digital twins in the IoT environment. Given the presence of a network of IoT devices and edge servers, as well as the time complexity of existing methods, this method uses a graph-based algorithm to minimize latency and energy consumption as the primary goals. The algorithm is used to allocate IoT devices to edge servers. This method has been implemented in Python, and evaluation results show up to 30% improvement in latency and 31% in energy consumption compared to
the works under comparison.
