This paper presents prediction path loss models in an urban environment for cellular networks with the help of machine learning methods. For this goal, Support Vector Regression (SVR), Random Forest (RF) and K-Nearest Neighbor (KNN) algorithms are exploited and assessed. The training and testing procedure is carried out with the help of a path loss dataset generated by simulated results considering a Long Term Evolution (LTE) network utilizing a digital terrain model. The simulation takes into account an urban environment for both line-of-sight (LOS) and non-LOS (NLOS) propagation condition. The results reveal that all the evaluated algorithms forecast path loss with a remarkable accuracy, providing root-mean-square errors on the order of 2. 1-2.2dB for LOS and 3. 4-4.1dB for NLOS locations, respectively. Among the examined algorithms, KNN shows the best performance, thus being an appealing option to predict path loss in urban areas. For comparison purposes, the COST231 Walfisch-Ikegami empirical model was applied, which presents the worst performance, providing the highest errors under-predicting path loss, especially in NLOS locations.

The employment of technology and digitization is crucial for cultural organizations to establish and sustain digital repositories for their cultural heritage artifacts. This exploitation is also essential in facilitating the presentation of cultural works and exhibits to a broader audience. Consequently, in this work, we propose a custom-developed digital repository that functions as software-as-a-service (SaaS), primarily promoting the safe storage, display, and sharing of cultural materials; enhancing accessibility; and fostering a deeper understanding and appreciation of cultural heritage. The proposed digital repository service is designed as a multitenant architecture, which enables organizations to expand their reach, enhance accessibility, foster collaboration, and ensure the preservation of their content. Moreover, our technology stack incorporates robust and reliable backend technologies, such as Django, to ensure data security and efficient management. Meanwhile, the frontend is powered by Angular, which guarantees a user-friendly and engaging interface for exploring and interacting with cultural materials. Specifically, this project aims to assist each cultural institution in organizing its digital cultural assets into collections and feeding other digital platforms, including educational, museum, pedagogical, and games, through appropriate interfaces. The creation of this digital repository offers a cutting-edge and effective open-access laboratory solution. It allows organizations to have a significant influence on their audiences by fostering cultural understanding and appreciation. Additionally, it facilitates the connection between different digital repositories and national/European aggregators, promoting collaboration and information sharing. By embracing this innovative solution, cultural institutions can benefit from shared resources and features, such as system updates, backup and recovery services, and data analytics tools, attributes that are currently provided by the platform.

The cooperative aerial and device-to-device (D2D) networks employing non-orthogonal multiple access (NOMA) are expected to play an essential role in next-generation wireless networks. Moreover, machine learning (ML) techniques, such as artificial neural networks (ANN), can significantly enhance network performance and efficiency in fifth-generation (5G) wireless networks and beyond. This paper studies an ANN-based unmanned aerial vehicle (UAV) placement scheme to enhance an integrated UAV-D2D NOMA cooperative network.The proposed placement scheme selection (PSS) method for integrating the UAV into the cooperative network combines supervised and unsupervised ML techniques. Specifically, a supervised classification approach is employed utilizing a two-hidden layered ANN with 63 neurons evenly distributed among the layers. The output class of the ANN is utilized to determine the appropriate unsupervised learning method—either k-means or k-medoids—to be employed. This specific ANN layout has been observed to exhibit an accuracy of 94.12%, the highest accuracy among the ANN models evaluated, making it highly recommended for accurate PSS predictions in urban locations. Furthermore, the proposed cooperative scheme allows pairs of users to be simultaneously served through NOMA from the UAV, which acts as an aerial base station. At the same time, the D2D cooperative transmission for each NOMA pair is activated to improve the overall communication quality. Comparisons with conventional orthogonal multiple access (OMA) and alternative unsupervised machine-learning based-UAV-D2D NOMA cooperative networks show that significant sum rate and spectral efficiency gains can be harvested through the proposed method under varying D2D bandwidth allocations.

Today’s terrestrial cellular communications networks face difficulties in serving coexisting users and devices due to the enormous demands of mass connectivity. Further, natural disasters and unexpected events lead to an unpredictable amount of data traffic, thus causing congestion to the network. In such cases, the addition of on-demand network entities, such as fixed or aerial base stations, has been proposed as a viable solution for managing high data traffic and offloading the existing terrestrial infrastructure. This paper presents an unmanned aerial vehicles (UAVs) aided offloading strategy of the terrestrial network, utilizing an unsupervised machine learning method for the best placement of UAVs in sites with high data traffic. The proposed scheme forms clusters of users located in the affected area using the k-medoid algorithm. Followingly, based on the number of available UAVs, a cluster selection scheme is employed to select the available UAVs that will be deployed to achieve maximum offloading in the system. Comparisons with traditional offloading strategies integrating terrestrial picocells and other UAV-aided schemes show that significant offloading, throughput, spectral efficiency, and sum rate gains can be harvested through the proposed method under a varying number of UAVs.

This letter assesses different ensemble models, exploited to forecast propagation loss in rural environments. Stacking, voting, bagging, and gradient boosted trees ensemble techniques are evaluated using path loss data recorded in various rural locations all over Greece. According to the numerical results all the introduced ensemble models increase their prediction efficacy and outperform all the single-model-based methods. Furthermore, stacking technique with 5 base learners and a customized deep neural network (DNN), as meta-learner, exhibits the best performance. Finally, it is shown that the forecast precision is gradually enhanced by increasing the number of the training features, yet under penalty of time complexity.