Researchers from the Institute for Infocomm Research (I2R) in Singapore, led by Lei Li and including Yanqing Xu, Ye Xue, Feng Yin, Chao Shen, Rui Zhang, and Tsung-Hui Chang, have developed a novel framework aimed at enhancing the efficiency and autonomy of mobile networks. Their work, titled “PEMNet: Towards Autonomous and Enhanced Environment-Aware Mobile Networks,” was published in the IEEE Journal on Selected Areas in Communications.
As 5G networks continue to be deployed and the industry looks towards 6G, mobile networks face the challenge of making rapid decisions in highly dynamic environments. These decisions must consider factors such as latency, energy consumption, and spectrum usage, all while adapting to changes in wireless channels and traffic demands. To address this, the researchers propose a framework called the perception embedding map (PEM). This framework creates a localized, joint representation of radio propagation and user demand within a base station’s coverage area. By integrating this knowledge, PEM supports optimization across various layers of the network, including the physical layer (PHY), media access control layer (MAC), and network layer.
One of the key advantages of PEM is its practicality. It is built using standard-compliant measurements, such as measurement reports and scheduling/quality-of-service logs, which are readily available in current networks. This means that PEM can be deployed and maintained at scale with minimal additional cost. Moreover, PEM emphasizes the joint embedding of channel and traffic information, which is crucial for network optimization. This joint approach sets PEM apart from existing site-specific channel maps and digital-twin replicas.
The researchers highlight that PEM significantly reduces training overhead and signaling, making it a promising solution for enhancing the autonomy and efficiency of mobile networks. By providing a fine-grained, environment-aware optimization framework, PEM could help network operators better manage the complexities of 5G and future 6G networks, ultimately leading to improved performance and user experience.
The research was published in the IEEE Journal on Selected Areas in Communications, a reputable publication in the field of telecommunications and networking. This work represents a significant step forward in the development of autonomous, efficient, and adaptive mobile networks, with practical applications for the energy sector in managing and optimizing network resources.
This article is based on research available at arXiv.