ILPA: An Intelligent Location Preference Assignment Framework for MCS in Metaverse and Digital Twins Environments

Abstract

With the rapid development of metaverse and digital twins, consumer electronics, like smartphones and wearable devices, are pivotal in merging the physical and digital realms. This integration has led to the prominence of Mobile CrowdSensing (MCS) as a critical data collection method. In MCS, the efficiency with which workers sense tasks directly determines the timeliness of data collection, which affects the capabilities of digital twin services and service consumers’ interests. However, improving efficiency often requires workers to accelerate data collection, which may sacrifice precise processing, affecting overall data quality. Conversely, to ensure high-quality data, workers may need to spend more time on meticulous task execution, which can reduce efficiency. Therefore, simultaneously achieving high efficiency and high quality in data collection presents a typical multi-objective optimization problem, where efficiency and quality goals often conflict. Against this background, an Intelligent Location Preference Assignment (ILPA) framework is proposed to strengthen the robustness of crowdsensing while maximizing the efficiency and quality of data collection. Firstly, a Location Preference Optimization Algorithm (LPOA) is developed to optimize quality and efficiency, obtaining workers’ location preferences. Then, under multi-location tasking, a Location Preference Assignment Scheme (LPAS) is constructed to achieve optimal assignment effectively. Comprehensive experiments conducted with two real-world datasets validate the effectiveness and applicability of the ILPA framework within the digital twins model.