Intelligent Application for User Equipment Positioning in Open-RAN

In this paper, we propose a novel artificial intelligence (AI) application in open radio access network (Open-RAN) framework to predict positions of user equipment (UE) in mobile networks. Specifically, we first analyze the characteristics of measured value (i.e., signal-to-noise ratio (SNR) value) at the mobile network, which will serve as the foundational input features for the AI application. Based on the analyzed results, we implement the moving average-based preprocessing technique in AI applications to eliminate randomness that could negatively impact prediction accuracy. Then, we design the AI model that utilizes the preprocessed data as input to predict UE positions. The evaluation results demonstrate that the proposed AI application predicts UE positions more accurately than an AI model without the preprocessing method.