Under the rapid development of the Internet of Things (IoT), vehicles can be recognized as mobile smart agents that communicating, cooperating, and competing for resources and information. The task between vehicles is to learn and make decisions depending on the policy to improve the effectiveness of the multi-agent system (MAS) that deals with the continually changing environment. The multi-agent reinforcement learning (MARL) is considered as one of the learning frameworks for finding reliable solutions in a highly dynamic vehicular MAS. In this paper, we provide a survey on research issues related to vehicular networks such as resource allocation, data offloading, cache placement, ultra-reliable low latency communication (URLLC), and high mobility. Furthermore, we show the potential applications of MARL that enables decentralized and scalable decision making in vehicle-to-everything (V2X) scenarios.

Massive multiple-input-multiple-output (MIMO) systems support advanced applications with high data rates, low latency, and high reliability in next-generation mobile networks. However, using machine learning in massive MIMO resource allocation is challenging due to quality of service (QoS) and network complexity across layers. This work presents a novel framework for adapting the scheduling and antenna allocation in massive MIMO systems using deep reinforcement learning (DRL). Rather than directly assigning execution parameters, the proposed framework utilizes DRL to select combinations of algorithms based on the current traffic conditions. The DRL model is trained using a specialized Markov decision process (MDP) model with a componentized action structure and is evaluated in realistic traffic scenarios. The results show that the proposed framework increases satisfied users by 7.2% and 12.5% compared to static algorithm combinations and other cross-layer adaptation methods. This demonstrates the effectiveness of the framework in managing and optimizing resource allocation in a flexible and adaptable manner.

Cache placement is a critical issue in fog networks. It is essential to simultaneously consider the quality of network connection, the demand of contents, and the users' activities. This paper proposes an efficient cache placement by placing the files based on popularity categories within a fog node cluster. Requested files are categorized into three popularity levels and strategically cached in fog nodes of various activity levels This work aims to reduce energy consumption by reducing the number of cells to serve the users based on content popularity. Another contribution of this paper is the clustering method for the fog nodes to select the node to host for the cache contents. The effectiveness of the algorithm is tested using the simulation regarding energy efficiency.

Managing cellular networks in the Beyond 5G (B5G) era is a complex and challenging task requiring advanced deep learning approaches. Traditional models focusing on internet traffic (INT) analysis often fail to capture the rich temporal and spatial contexts essential for accurate INT predictions. Furthermore, these models do not account for the influence of external factors such as weather, news, and social trends. This study proposes a multi-source CNN-RNN (MSCR) model that leverages a rich dataset, including periodic, weather, news, and social data to address these limitations. This model enables the capture and fusion of diverse data sources for improved INT prediction accuracy. An advanced deep learning model, the transformer-enhanced CNN-RNN (TE-CNN-RNN), has been introduced. This model is specifically designed to predict INT data only. This model demonstrates the effectiveness of transformers in extracting detailed temporal-spatial features, outperforming conventional CNN-RNN models. The experimental results demonstrate that the proposed MSCR and TE-CNN-RNN models outperform existing state-of-the-art models for traffic forecasting. These findings underscore the transformative power of transformers for capturing intricate temporal-spatial features and the importance of multi-source data and deep learning techniques for optimizing cell site management in the B5G era.

Cloud computing is a remarkable model for permitting on-demand network access to an elastic collection of configurable adaptive resources and features including storage, software, infrastructure, and platform. However, there are major concerns about security-related issues. A very critical security function is user authentication using passwords. Although many flaws have been discovered in password-based authentication, it remains the most convenient approach that people continue to utilize. Several schemes have been proposed to strengthen its effectiveness such as salted hashes, one-time password (OTP), single-sign-on (SSO) and multi-factor authentication (MFA). This study proposes a new authentication mechanism by combining user's password and modified characters of CAPTCHA to generate a passkey. The modification of the CAPTCHA depends on a secret agreed upon between the cloud provider and the user to employ different characters for some characters in the CAPTCHA. This scheme prevents various attacks including short-password attack, dictionary attack, keylogger, phishing, and social engineering. Moreover, it can resolve the issue of password guessing and the use of a single password for different cloud providers.

Cloud computing is a significant model for permitting on-demand network access to shared data, software’s, infrastructure, and platform resources. However, cloud storage needs a certain level of availability, confidentiality, and integrity. Information sensitivity and value encourage users to select a highly secure protocol. This work proposes a new mechanism to increase the user trust in cloud computing using the secret sharing technique. The proposed algorithm is using Base64 encoding to convert any file type to ASCII strings. Base64 strings do not need any extra process to be compressed and this can speed up the share building process. Each string is used to produce N shares (using Shamir Secret Sharing Scheme) where each share is stored in a separate location in the cloud.