AI Based RSSI Algorithm for Localization in the IoT Era

The utilization of the Internet of Things (IoT) is growing, spanning various domains such as smart homes, sophisticated automotive technologies, and smart healthcare. It is crucial to ascertain the source location of transmitted data, as information lacking a known source lacks significance. The applications under development require positioning without reliance on the global positioning system (GPS) due to the absence of signals in indoor areas and challenging environmental conditions. The most popular localization algorithm in wireless sensor networks (WSNs) is received signal strength indicator (RSSI) algorithm. The RSSI Path Loss Model approach relies on triangulation to determine the positions of wireless nodes; however, the accuracy attained has been unsatisfactory. Absolutely, artificial intelligence (AI) can play a crucial role in improving positioning accuracy. By leveraging AI, algorithms can be developed to learn from existing data, such as anchor nodes, and make predictions about the positions of unknown sensor nodes. This approach allows for more intelligent and adaptive positioning solutions, enhancing accuracy in various applications. This paper presents the AI-RSSI algorithm, which enhances localization accuracy without requiring additional hardware. The simulations clearly show that the proposed algorithm exhibits significantly improved the localization accuracy compared to the original RSSI algorithm.