During the past decade, Industry 4.0 has greatly promoted the improvement of industrial productivity by introducing advanced communication and network technologies in the manufacturing process. With the continuous emergence of new communication technologies and networking facilities, especially the rapid evolution of cellular networks for 5 G and beyond, the requirements for smarter, more reliable, and more efficient cellular network services have been raised from the Industry 5.0 blueprint. To meet these increasingly challenging requirements, proactive and effective allocation of cellular network resources becomes essential. As an integral part of the cellular network resource management system, cellular traffic prediction faces severe challenges with stringent requirements for accuracy and reliability. One of the most critical problems is how to improve the prediction performance by jointly exploring the spatial and temporal information within the cellular traffic data. A promising solution to this problem is provided by Graph Neural Networks (GNNs), which can jointly leverage the cellular traffic in the temporal domain and the physical or logical topology of cellular networks in the spatial domain to make accurate predictions. In this paper, we present the spatial-temporal analysis of a real-world cellular network traffic dataset and review the state-of-the-art researches in this field. Based on this, we further propose a time-series similarity-based graph attention network, TSGAN, for the spatial-temporal cellular traffic prediction. The simulation results show that our proposed TSGAN outperforms three classic prediction models based on GNNs or GRU on a real-world cellular network dataset in short-term, mid-term, and long-term prediction scenarios.

Benefiting from the large-scale commercial use of 5G, smart campuses have attracted increasing research attention in recent years and are expected to revolutionize traditional campus activities. However, there are some obstacles that hinder the practical deployment of MEC (multi-access edge computing). First, traditional information infrastructures on campus cannot support latency-sensitive and computing-intensive smart applications, such as AR/VR, live interactive lectures and digital twin experiments. In addition, the mixture of old and new applications, isolated data islands and heterogeneous equipment management introduce more challenges. Moreover, the existing MEC framework proposed by ETSI and 3GPP cannot meet the specific deployment requirements of smart campuses, e.g., educational data security, real-time interactive applications, heterogeneous connections, and others. In this paper, we propose a 5G-based architecture for smart education information infrastructure; a new dedicated cloud architecture eMEC (educational multi-access edge computing) is defined. It consists of a UGW (universal access gateway) and an eMEP (educational multi-access edge computing platform), making it possible to satisfy education-specific requirements and long-term evolution. Furthermore, we implement the framework and conduct real-world field tests for eMEC in a university campus. Based on the framework and practical field tests, we also conduct a measurement study to unveil the spatial-temporal characteristics of mobile users in the smart campus and discuss exploiting them for better network performance. The experimental results show that the system achieves satisfactory performance in terms of both throughput and latency.

One fundamental problem of content caching in edge computing is how to replace contents in edge servers with limited capacities to meet the dynamic requirements of users without knowing their preferences in advance. Recently, online deep reinforcement learning (DRL)-based caching methods have been developed to address this problem by learning an edge cache replacement policy using samples collected from continuous interactions (trial and error) with the environment. However, in practice, the online data collection phase is often expensive and time-consuming, thus hindering the practical deployment of online DRL-based methods. To bridge this gap, we propose a novel Agile edge Cache replacement method based on Offline-online deep Reinforcement learNing (ACORN), which can efficiently learn an edge cache replacement policy offline from a training dataset collected by a behavior policy (e.g., Least Recently Used) and then improve it with fast online fine-tuning. We also design a specific convolutional neural network structure with multiple branches to effectively extract content popularity knowledge from the dataset. Experimental results show that the offline policy generated by ACORN outperforms the behavior policy by up to 38%. Through online fine-tuning, ACORN also achieves the number of cache hits as good as that of several advanced DRL-based methods while significantly reducing the number of training epochs by up to 40%.