This study investigates Downlink/Uplink Decoupling (DUDe) in 5G networks, a framework that allows user equipment to select its uplink serving cell independently of the downlink anchor. This approach is designed to alleviate the "macro bias" and pathloss issues that typically degrade performance for Internet of Things (IoT) traffic. We propose a framework managed by Mobile Edge Computing (MEC) that operates on a per-Transmission Time Interval (TTI) basis, incorporating stability mechanisms such as hysteresis and Time to Trigger to prevent frequent, unnecessary handovers. The performance is evaluated using a system-level simulator across two scenarios: a high-density urban IoT deployment and an Industry 4.0 smart factory environment. Our results demonstrate that the proposed framework significantly improves uplink throughput and reduces tail latency compared to traditional coupled association methods. Furthermore, an ablation study confirms that these performance gains are derived from the structural decoupling of links, providing a scalable path for improving connectivity in 5G and beyond.

Massive Multiple Input Multiple Output (MIMO) is an essential technology that can significantly improve the performance of 5G wireless networks by using multiple antennas in base stations, improving coverage, reducing interference, and increasing data throughput. In this comprehensive study, we propose and analyze advanced optimization techniques for resource allocation in 5G MIMO networks, focusing on three distinct approaches: simple sorting, Hungarian Algorithm, and Minimum Cost Flow Algorithm. Simulations are performed using the publicly available DeepMIMO dataset, where we evaluate each method under both static and dynamic scenarios, aiming to optimize bandwidth distribution and minimize power consumption. A key contribution of this work is the formulation and comparative evaluation of the resource allocation problem as an assignment-based model, allowing the examined methods to be compared under common DeepMIMO-based static and dynamic scenarios. The technical contribution of this work lies in the common assignment-based formulation and comparative evaluation of simple sorting, Hungarian, and Minimum Cost Flow allocation methods under the same DeepMIMO-based static and dynamic 5G MIMO scenarios. Our comparative analysis shows that, under the evaluated DeepMIMO-based scenarios, the examined assignment-based methods exhibit different trade-offs in throughput, energy-consumption-related performance, bandwidth utilization, and adaptability to varying user demands, offering useful insights for 5G MIMO resource allocation studies.

Ultra-dense 5G networks require advanced traffic steering to maintain performance and balance load amid growing user and base station (gNB) densities. Traditional heuristics such as nearest-base-station and Signal-to-Interference-plus-Noise Ratio (SINR)-based selection provide simple solutions but struggle to adapt to dynamic user mobility, diverse traffic, and fluctuating radio conditions at the mobility-control level, often leading to inefficient handovers and degraded network quality. We propose a deep reinforcement learning (DRL) framework to dynamically tune a global handover hysteresis margin that governs handover triggering decisions, optimizing handover success, reducing failures, and enhancing throughput and fairness. Implemented in Python using Stable Baselines3 and NumPy, our custom simulation environment models key mobility-related 5G dynamics at a high level, including user mobility, pathloss-based signal degradation, and interference. We evaluate DRL agents-Deep Q-Network (DQN) and Proximal Policy Optimization (PPO)-against heuristic and hysteresis-based baselines. Results show that DRL-based hysteresis optimization provides strong and robust performance under the considered ultra-dense mobility conditions in handover success rate, average SINR, throughput, and fairness, with PPO demonstrating the most consistent behavior across configurations. This work offers a reproducible simulation framework for further research into adaptive mobility management.

This paper presents a reinforcement learningbased approach for optimizing the performance of $\mathbf{5 G}$ mobile networks. By leveraging Deep Q-Networks (DQN), our system autonomously tunes network parameters across macro, micro, and pico cells, adapting to the dynamic distribution in a heterogeneous network environment. The agent is tasked with optimizing several Key Performance Indicators (KPIs) such as throughput, latency, interference, and Quality of Service (QoS). Each cell in the network can perform actions such as adjusting power levels, changing handover thresholds, allocating bandwidth, and performing interference mitigation. Our approach demonstrates significant improvements in user experience, resource utilization, and network efficiency over traditional static optimization methods. The results show that the proposed reinforcement learning -based algorithm not only reduces latency and interference but also ensures better load balancing and throughput optimization across heterogeneous cells.

This study investigates the prediction of Doppler shift variations in high-speed rail (HSR) environments using advanced deep learning and classical timeseries models. By simulating Doppler shifts at a 5G carrier frequency under noisy conditions, we evaluate and compare the performance of Bidirectional Long Short-Term Memory (LSTM) networks, Gated Recurrent Unit (GRU) networks, and an optimized Auto-Regressive Integrated Moving Average (ARIMA) model. The results highlight the strengths and limitations of each model, providing a detailed comparison between data-driven and statistical forecasting methods in dynamic 5G communication scenarios.

This paper explores the application of Game Theory techniques to Downlink-Uplink Decoupling (DUDe) in 5G networks, with the goal of optimizing resource allocation. DUDe technology enables the independent management of uplink and downlink connections, introducing new challenges in the distribution of network resources such as bandwidth and energy. To address these challenges, the study investigates two specific game-theoretic algorithms: the Gale-Shapley algorithm, which models a matching game between users and base stations, and the Nash Bargaining algorithm, which dynamically adjusts bandwidth allocation based on individual user demands. By applying these algorithms, the paper demonstrates how Game Theory can offer effective and adaptive solutions to complex resource management problems in next-generation networks. The results highlight improvements in network performance, energy efficiency, and user experience. This work underscores the relevance of Game Theory in the design of intelligent 5G infrastructures and provides a foundation for future research in autonomous resource allocation in distributed network environments.

To meet the escalating data demands of 5G and beyond networks, densified Heterogeneous Networks (HetNets) provide a promising solution, deploying small base stations for improved spectral and energy efficiency. However, HetNets pose challenges, particularly in user association. This journal introduces the Downlink/Uplink Decoupling (DUDe) approach, which enhances uplink performance in HetNets by allowing different access points for uplink and downlink associations. We assess DUDe's energy efficiency through extensive simulations across various scenarios, demonstrating substantial energy savings compared to centralized 5G systems. Our findings underscore the importance of energy-efficient design for reducing network operational costs and carbon footprint in 5G networks. In addition to energy efficiency gains, DUDe also offers improved resource allocation and network flexibility, making it a valuable solution for evolving wireless communication ecosystems.

In the landscape of 5G networks, efficient resource allocation (RA) stands as a critical factor in meeting the diverse demands of applications and users. This paper delves into optimizing RA within 5G Multiple Input Multiple Output (MIMO) networks by leveraging Downlink/Uplink Decoupling (DUDe) techniques. MIMO technology, enabling the simultaneous transmission of multiple data streams, holds promise for boosting spectral efficiency. However, accommodating the dynamic and diverse user requirements poses a significant challenge in resource allocation. By employing advanced DUDe techniques, this study dynamically allocates resources in 5G MIMO Heterogeneous Networks (HetNets), seeking to enhance throughput, minimize latency, and optimize user satisfaction. The paper includes scenarios involving varying User Equipment (UE) densities and mobility to evaluate system performance under different load conditions. Through simulation-based analysis, this paper highlights the efficacy of the proposed approach in significantly improving network performance, energy efficiency, and resource utilization.

Next-generation 5G networks with massive Multiple Input Multiple Output (MIMO) must efficiently allocate radio resources to mobile users whose channel conditions change rapidly due to movement. This paper proposes a novel game-theory Reinforcement Learning (RL) framework for mobility-aware resource allocation in 5G MIMO systems. We model the resource allocation problem as a dynamic game between network entities and integrate a predictive deep RL agent that anticipates User Equipment (UE) mobility patterns. By forecasting UE movement, the RL agent proactively assists a game-theory optimization of MIMO resource allocation before channel quality degrades. The combination of game theory with predictive RL enables the network to reach a near-equilibrium resource distribution that is both adaptive and fair, improving convergence stability compared to standalone learning or game approaches. Simulation results in a high-mobility 5G scenario demonstrate that the proposed approach significantly boosts user Quality of Service (QoS) for example, increasing average throughput and reducing latency and handover failures relative to conventional reactive allocation strategies. Specifically, the proposed framework delivers a 17–22

The rapid expansion of Internet of Things (IoT) devices in various sectors, from healthcare to industrial automation, has heightened the need for secure and efficient communication networks. Multiple Input Multiple Output (MIMO) systems, integral to 5G networks, offer high throughput and low latency essential for real-time IoT applications. However, the integration of robust security mechanisms into MIMO systems often leads to increased latency, potentially undermining the performance of timesensitive IoT applications. This paper proposes a framework for enhancing the security of MIMO networks specifically designed for IoT environments while maintaining ultra-low latency. The study explores lightweight encryption techniques, physical layer security methods, and optimized beamforming strategies that collectively safeguard data integrity and confidentiality without compromising network performance. Through theoretical analysis and extensive simulations, this research demonstrates that it is possible to achieve a secure MIMO-based IoT network that balances both latency and security requirements.

Efficient resource allocation is essential in 5G MIMO networks due to increasing demands for high-quality communications. This paper compares four game theory algorithms: Stackelberg, Nash Bargaining, Mean Field Game, and Potential Game, evaluating their effectiveness in allocating resources dynamically. A simulation environment is developed to represent realistic user mobility by continuously updating user equipment (UE) positions. Each algorithm is assessed based on UE distribution, fairness, bandwidth consumption, and energy efficiency. The simulation results show clear differences among the algorithms, highlighting specific advantages and limitations that help inform resource allocation strategies in practical 5G network scenarios.

The rapid increase in IoT (Internet of Things) devices in 5 G networks poses an increasing challenge for managing uplink (UL) traffic, especially in dense and heterogeneous scenarios. Traditional uplink management strategies may not achieve fairness or efficiency in resource allocation under realistic, dynamic conditions related to IoT traffic. This paper provides a procedural framework for Downlink and Uplink Decoupling (DUDe) which produces performance benefits for the uplink in 5G networks leveraging IoT devices. The framework is examined via simulation in MATLAB in a variety of scenarios and the performance of DUDe is compared with existing uplink management strategies in the literature. The results presented demonstrate that DUDe achieves substantial gains in fairness with regard to resource allocation, with minimal deterioration in spectral efficiency, and moderate impact on throughput. Finally, operational challenges related to deploying DUDe in practice (e.g. hardware compatibility and operational complexity) are considered, along with potential solutions. The paper can be seen as an informative reference for future potential practical implementation of DUDe in real world 5G IoT networks.

In recent years, Generative Adversarial Networks (GANs) have emerged as powerful tools for improving signal processing in advanced communication systems, particularly in the context of 5G networks. In this paper, we present a novel approach for distinguishing signal from noise in 5G Multiple Input Multiple Output (MIMO) systems using GANs. Our method leverages the generative capabilities of GANs to produce realistic noise signals and the discriminative power of GANs to accurately identify real signals amidst noise. By training the GAN on a combination of real-world noisy signals and pure noise, our model achieves robust signal detection and classification. We evaluate our approach using synthetic data, demonstrating significant improvements over other techniques such as the autoencoders. Our results highlight the potential of GANs in enhancing the reliability and performance of 5 G MIMO communications.

The increasing density of Industrial Internet of Things (IIoT) devices in 5G networks introduces severe uplink congestion and fairness challenges. Traditional coupled access—where a user equipment (UE) uses the same base station for both downlink and uplink—can lead to suboptimal performance, particularly in smart factories characterized by heterogeneous traffic patterns and dynamic interference. This paper investigates uplink decoupling, a lightweight association mechanism that allows independent selection of the uplink serving cell to optimize link quality and resource utilization. We develop a MATLAB-based simulation that models a 5G smart factory and compare the proposed decoupled uplink approach against conventional proportional-fair (PF) scheduling. Results show consistent improvements in throughput, 95th-percentile latency, and fairness across multiple traffic classes. These findings demonstrate that uplink decoupling can effectively mitigate congestion, improve reliability, and enhance service differentiation in Industry 4.0 networks.

Fifth-Generation (5G) networks deal with dynamic fluctuations in user traffic and the demands of each connected user and application. This creates a need for optimized resource allocation to reduce network congestion in densely populated urban centers and further ensure Quality of Service (QoS) in (5G) environments. To address this issue, we present a framework for both predicting user traffic and allocating users to base stations in 5G networks using neural network architectures. This framework consists of a hybrid approach utilizing a Long Short-Term Memory (LSTM) network or a Transformer architecture for user traffic prediction in base stations, as well as a Convolutional Neural Network (CNN) to allocate users to base stations in a realistic scenario. The models show high accuracy in the tasks performed, especially in the user traffic prediction task, where the models show an accuracy of over 99%. Overall, our framework is capable of capturing long-term temporal features and spatial features from 5G user data, taking a significant step towards a holistic approach in data-driven resource allocation and traffic prediction in 5G networks.

The advent of 5 G technology has ushered in a new era of wireless communication, characterized by its promise of high data rates, low latency, and enhanced connectivity. In this context, Multiple-Input Multiple-Output (MIMO) systems have emerged as a key enabler, leveraging advanced antenna arrays to simultaneously serve multiple users with increased spectral efficiency. This paper investigates the dynamic resource allocation problem in a MIMO 5 G environment, where each user possesses distinct bandwidth requirements. The focus is on optimizing user allocation while considering the limited bandwidth and user capacity of base stations. By harnessing the power of deep learning techniques, the proposed solution aims to efficiently manage the allocation of users to base station antennas, thereby maximizing overall network performance while accommodating heterogeneous user demands.

In this paper we evaluate the energy efficiency of Downlink/Uplink Decoupling (DUDe) systems by considering the energy consumption of each Base Station (BS) and the communication overhead between the BSs and User Equipment (UE). Through extensive simulations, the energy efficiency of DUDe systems is evaluated under various scenarios, such as different UE densities, Base Stations (BSs) configurations, and traffic patterns. The simulation results indicate that the proposed DUDe system can achieve significant energy savings compared to traditional centralized 5G systems. The impact of various parameters on energy efficiency is also analyzed, providing guidelines for designing energy-efficient DUDe systems. The paper emphasizes the significance of energy-efficient design in 5G networks, as it can significantly reduce the operational costs and carbon footprint of the network.

The advent of 5G technology has brought unprecedented advancements in wireless communication, promising higher data rates and improved user experiences. However, the implementation of Multiple-Input Multiple-Output (MIMO) techniques in 5G networks has significantly increased power consumption, posing a challenge to sustainability and operational costs. This study conducts a comparative analysis of energy efficiency in 5G MIMO networks, specifically between Downlink Uplink Coupling (DUCo) and Downlink Uplink Decoupling (DUDe) techniques. By examining the remaining power in the Base Stations (BSs) of the implemented Heterogeneous Network (HetNet), this analysis evaluates the potential of DUDe over DUCo in managing power consumption. The findings of this study aim to assist network operators, researchers, and policymakers in developing greener and more sustainable 5G infrastructures for a cleaner, energy-efficient future.

Chatbots, the pioneering conversational artificial intelligence (AI) agents, have experienced remarkable growth and integration in various domains. In modern societies, chatbots have emerged as transformative digital entities, revolutionizing the way humans interact with technology. These conversational AI agents have transcended their initial applications to become integral parts of various industries and daily life. One of the most prominent roles of chatbots is in customer service, where they offer round-the-clock assistance, swift issue resolution, and personalized interactions. By handling routine queries and tasks, chatbots free up human agents to focus on complex and specialized issues, thus optimizing overall efficiency and customer satisfaction. To this end, this paper aims to present and describe the architecture of a novel chatbot generator with improved functionality in terms of quality of communication with end users and level of provided services, with a specialized infrastructure understanding the Greek language. The chatbot generator was developed in the framework of a research project and will be pilot tested by two end-users, the National Bank of Greece (NBG) and the General Secretariat for Information Systems & Digital Governance (GSIS-DG).