In 5G and emerging 6G technologies that involve massive MIMO networks, accurate channel estimation is cru-cial challenge. Channel sparsity and random projection pose additional challenges for the system. Compressive sensing (CS) has been proposed to address these challenges. However, the involvement of operations like discrete Fourier transform (DFT) result into higher complexity of the mechanism. In this paper, we used deep learning (DL) method using the discrete Rajan transform (DRT) and its inverse as alternatives to traditional DFT based CS methods for enhanced channel feedback. Through simulations, we demonstrate that the proposed mechanism outperforms established algorithms in scenarios with varying signal-to-noise ratio (SNR). Normalized mean square error (NMSE) and cosine correlation ($\rho$) versus SNR graph illustrates the superiority of our approach over other current approaches.

In this paper, we propose a double deep Q-network (DDQN) based algorithm for resource optimization and data rate enhancement in a hybrid wireless fidelity (WiFi) and light fidelity (LiFi) communication system. Multiple LiFi and a WiFi access points (APs) have been developed in indoor settings. LiFi APs can deliver exceptionally high data rates while also serving as sources of illumination. However, LiFi APs alone fail to satisfy the throughput demands due to their small cell area and non-line-of-sight (NLOS) situation inefficiency. Therefore, in practical applications, WiFi APs, which can provide widespread coverage, can be integrated with LiFi network to maintain seamless connectivity across the system. Optimal resource allocation remains a crucial issue in these systems. Simulations results verify that proposed DDQN based optimal resource allocation outperforms existing DQN learning based algorithm by 26.7

Energy acquisition from visible light communication (VLC) signals has recently received research focus in hybrid radio frequency (RF)/VLC systems. In this article, we study the effect of energy acquisition on the achievable sum-rate (ASR) of these systems. The joint uplink-downlink ASR maximization in a time-switching energy acquisitionbased hybrid RF/VLC system is investigated. We obtain fundamental results on the optimal time for performing energy acquisition. Furthermore, strict bounds on ASR pertaining to the energy acquisition conditions have also been obtained. Through simulations, it is observed that the optimal time coefficient obtained in this work outperforms the existing energy acquisition schemes by 40 Mbps in the ASR.

In massive multiple-input multiple-output (MIMO) frequency division duplex (FDD) systems, channel state information (CSI) feedback is essential. Traditional CSI methods are highly dependent on the degree of channel sparsity and are based on compressive sensing (CS). In this paper, we propose a novel deep learning (DL) based method to transform CSI into a low-dimensionality codeword. We construct DeNoNet, a DL-based denoise network, to enhance channel feedback and mitigate real-world interferences and nonlinear effects. Simulations results show that proposed DeNoNet mechanism performs better than the existing discrete Rajan transform (DRT) based DL technique, by 113.18 % in normalized mean square error (NMSE) and 72.72 % in cosine correlation (rho) particularly at varying signal-to-noise ratios (SNR).

Energy trading (ET) in vehicle-to-grid (V2G) networks enables plug-in electric vehicles (PEVs) to share energy with smart grids. However, conventional energy trading relies on financial incentives, which may restrict optimal energy distribution, particularly during emergencies. To overcome this problem, we propose an IOTA-based energy bartering framework that allows direct energy exchange between electric vehicles (EVs) and smart grids without financial transactions. Our framework enables EVs and smart grids to determine their energy imports and exports in a decentralized manner, overcoming the scalability and cost challenges of traditional blockchain-based solutions. To optimize the negotiation process during energy bartering, we employ a Stackelberg game model. In this game-theoretic approach, EVs act as leaders, setting their energy offers, while smart grids respond by selecting optimal offers based on supply-demand conditions. This structured process enhances system efficiency, ensuring rational energy allocation while maintaining network resilience. Simulation results demonstrate that our framework improves energy bartering efficiency, reducing energy costs for EVs by up to 69.23% compared to traditional trading mechanisms. It also enhances network scalability and ensures privacy-preserving, financial-transaction-free energy exchange, making it a robust solution for next-generation V2G networks.

Hybrid visible light communication (VLC) and radio frequency (RF) systems have gained significant attention for next-generation wireless networks owing to their ability to deliver high throughput and uninterrupted connectivity. Determining optimal resource allocation in such hybrid architectures, however, leads to a highly non-convex optimization problem that becomes increasingly complex in large and time-varying network scenarios. To overcome these shortcomings, this work introduces a two-stage mechanism that uses proximal policy optimization (PPO) assisted transfer learning for adaptive and efficient resource management in largescale hybrid VLC/RF systems. In the first stage, a PPO-based reinforcement learning agent learns a robust policy that ensures stable convergence even under rapidly changing link conditions. In the second stage, transfer learning uses previously trained policy parameters to quickly adapt to newly arriving mobile users, significantly minimizing retraining overhead. Extensive simulations confirm that the proposed method reduces convergence iterations by nearly 72 % compared with deep Q-network (DQN)-based approaches, while maintaining higher achievable sum-rate and stronger adaptability to network dynamics.

The depletion of fossil fuels reserves attracts attention from the scientists and researchers to develop materials which is sustainable, cost effective, green and ecofriendly. The growing demand for renewable high performance energy storage systems made to explore natural fiber precursors to develop carbon nanomaterials for next generation energy storage devices. This paper focuses on synthesis of carbon electrodes from natural fibers using various carbonization and activation methods, with the main emphasis on bio-charring process. Activation methods are mainly employed to transform surface composition of nanomaterials, thereby creating a layered and porous structure which is needed for electrodes fabrication for energy storage devices. This paper review recent development of electrode nanomaterials, from lignin, cellulose and hemicellulose nanofibers from natural fiber obtained precursors for supercapacitor applications, highlighting their preparation methods, electrochemical characteristics, performance, renewable and sustainable nature. The derived carbon nanomaterial is engineered to exhibit improved energy and power density, specifically for supercapacitor applications. The paper emphasizes nano dimensions of carbon nanomaterials from natural fiber as electrode materials. It discusses recent challenges, properties, characteristics, and future research directions in developing cost-effective and environmentally friendly supercapacitors for energy storage. Despite, the growing interest in biomass derived carbon electrodes for supercapacitors, no prior review focusing specifically on natural fiber precursors for supercapacitors. This study, therefore, provides a novel contribution to the energy storage field.

In 5G and 6G communication systems that employ multi-user MIMO architectures, accurate channel estimation remains a critical challenge. The inherent sparsity of the channel and the effects of random projection further complicate this task. Compressive sensing (CS) techniques have been explored to address these issues; however, their reliance on operations such as the channel state information network (CsiNet) often leads to increased computational complexity. In this paper, we used a deep learning (DL) based method utilizing a deep neural network denoising (DNNet-DeNo) framework for improved channel state information (CSI) feedback in FDD multi-user MIMO systems. Simulation results demonstrate that the proposed method consistently outperform existing DL-based algorithms across various signal to noise ratio (SNR) circumstances. The normalized mean square error (NMSE) and cosine similarity ($\rho$) versus SNR analyses further highlight the superior performance and robustness of our model compared to existing DL-based methods.

Wireless fidelity (WiFi) and light fidelity (LiFi) offer extensive coverage areas and high data rates, respectively. Each of them operate on distinct and non-interfering spectra. The integration of both these technologies creates a hybrid WiFi/LiFi network, capable of delivering high achievable sum rate (ASR) and enhanced network mobility. In hybrid WiFi/LiFi networks, resource allocation remains a significant challenge. This study focused on model free approach for resource allocation using deep reinforcement learning (DRL), which is simulated under realistic constraints such as signal blockage and load balancing. Simulations results show that the proposed DRL approach surpasses traditional deep Q-network (DQN)-based methods by 42.3% in terms of ASR with 35.4% lower transmission power usage, leading to better resource management and enhanced overall network performance.

The proliferation of Internet of Things (IoT) devices has raised considerable difficulties in the identification of users, optimization of power, and load balancing in hybrid RF/LiFi networks. As interconnection among devices increases, ensuring optimal performance while managing network resources efficiently becomes quite complex. This complexity arises due to accommodating the possibly diverse users' needs, fluctuating channel conditions, and varying interference levels, all necessitating sophisticated management solutions to provide seamless connectivity and dependable communication. To tackle these issues, a deep joint hybrid system (DJHS) technique is presented, which employs proximal policy optimization (PPO), a cutting-edge deep reinforcement learning (DRL) algorithm. DJHS aims to effectively handle the intricate problems surrounding user association and load balancing while optimizing power usage in dynamic contexts. DJHS continuously updates its approach based on real-time network data through adaptive learning methods, allowing it to make intelligent decisions that improve overall system performance regarding data throughput and power optimization. Simulation results demonstrate that DJHS outperforms existing approaches such as soft actor-actor (SAC), advantage actor-critic (A2C), twin-delayed deep deterministic (TD3), and trust region policy optimization (TRPO) regarding crucial metrics, including data rate and power transmission. Notably, DJHS's ability to adjust to variations in signal-to-interference-plus-noise ratio (SINR) allows for enhanced resource allocation and network stability. This flexibility ensures that users receive optimal service even in changing conditions, enhancing the overall user experience.

The reduction of channel state information feedback overhead is crucial in fifth generation and emerging sixth generation technologies involving massive multiple input multiple output (MIMO) systems. While compressive sensing (CS) has been proposed to address this challenge, the involvement of discrete Fourier transform (DFT) in CS result into higher computational complexity. This paper introduces a novel discrete Rajan transform (DRT) based deep learning (DL) to enhance channel feedback in massive MIMO systems. Through simulations, we show that the proposed mechanism outperforms the DFT based DL algorithm by 126.19% and 81.81% for cosine correlation and normalized mean square error in channel estimation, respectively, across varying signal to noise ratio (SNR) scenarios. Notably, the proposed DRT based DL algorithm excels in low SNR conditions, showing its potential for practical implementation in real-world communication systems.

Visible light communication (VLC) has emerged as a promising technology, delivering high-speed data transmission for 5G and beyond communication. Nevertheless, its susceptibility to blockages demands a co-deployment with traditional radio frequency (RF) systems to ensure uninterrupted connectivity. This co-deployment, known as a hybrid RF/VLC system, is a subset of heterogeneous networks (HetNets) and offers interoperability, energy efficiency, and optimal resource utilization. In hybrid RF/VLC, efficient resource allocation and load balancing are crucial. Existing Deep Q-Network (DQN) learning-based methods designed to address these issues, fail in large and dynamic environments. Our present study investigates alternative approaches for optimal resource allocation and load balancing in dynamic and large hybrid RF/VLC systems, to achieve maximum data rates for users. We propose two model-free on-policy deep reinforcement learning (DRL) based schemes, namely advantage actor-critic (A2C) and proximal policy optimization (PPO), for efficient resource allocation in hybrid RF/VLC. Simulation results show that the A2C and PPO based schemes outperform the DQN learning scheme by 31.3% and 32.5%, respectively, in terms of data rates. The proposed schemes also outperform the deep deterministic policy gradient (DDPG) in data rate maximization by up to 8.1% and 9.7%, respectively.

Cognitive radio (CR)-based 5G and beyond networks have emerged as a potential technology, offering secondary users (SUs) the ability to utilize radio spectrum designated for licensed primary users (PUs) in their absence. Cooperative spectrum sensing (CSS) plays a crucial role in enhancing spectrum sensing accuracy within CR networks. However, the efficacy of CSS can be undermined by potential attacks from malicious users (MUs) transmitting inaccurate sensing information to the fusion center (FC). This study introduces a novel machine learning (ML) based scheme for identifying malicious users in CR networks before their data reaches the FC, thereby enabling the complete insulation of FC from adversarial signals generated by malicious users. Through simulations, the effectiveness of the proposed deep neural network (DNN) algorithm-based ML approach in detecting MUs is evaluated. The proposed DNN model overcomes the existing schemes by classifying multiple SUs before reaching the FC.

The Internet of Vehicles (IoV) necessitates efficient resource management to meet the growing demands for high data rates, low latency, and real-time communication in Intelligent Transportation Systems (ITS). This paper presents a novel multiagent reinforcement learning framework based on the Proximal Policy Optimization (PPO) algorithm for optimizing resource allocation in clusters of cells within IoV networks. The framework dynamically allocates transmission power and bandwidth across base stations, each acting as a collaborative agent. Extensive simulations demonstrate that the multi-agent PPO approach outperforms other reinforcement learning algorithms, including Soft Actor-Critic (SAC) and Advantage Actor-Critic (A2C), regarding throughput, power efficiency, dynamic bandwidth allocation, and overall network performance.

A hybrid radio frequency (RF) and light fidelity (LiFi) network combines the strengths of RF and LiFi technologies. RF offers broad coverage, while LiFi provides high data rates. As these technologies operate on non-interfering spectra, they can co-exist without interfering with each other. This setup not only boosts data rate but also makes the network more reliable, especially when physical obstacles might block signals. However, resource management in hybrid RF/LiFi networks is challenging because of the dynamic environment and the different characteristics of the two technologies. Efficient resource allocation maximizes the data rate in these networks. In this paper, we introduce a model-free deep reinforcement learning (DRL) approach to solve the resource allocation problem in hybrid RF/LiFi networks. Our DRL model is designed to handle real-world conditions, considering factors like blockages and user mobility. Unlike traditional methods that need extensive modeling and assumptions, our approach learns directly from interacting with the environment, making it highly adaptable and robust. Through simulations, it is observed that our method improves resource utilization and overall network performance, achieving a 62.8% increase in sum rate and a 42.8% improvement in optimal transmit power compared to conventional methods.

Centralized Dynamic Spectrum Sharing (DSS) faces challenges like data security, high management costs, and limited scalability. To address these issues, a blockchain-based DSS scheme has been proposed in this paper. First, we utilize the ERC4907 standard to mint Non-Fungible Spectrum Tokens (NFSTs) that serve as unique identifiers for spectrum resources and facilitate renting. Next, we develop a smart contract for NFST auctions, ensuring secure spectrum transactions through the auction process. Lastly, we create a Web3 spectrum auction platform where users can access idle spectrum data and participate in auctions for NFST leases corresponding to the available spectrum. Experimental results demonstrate that our NFST, designed according to the ERC4907 standard, effectively meets users' secure and efficient DSS requirements, making it a feasible solution.

With the increasing adoption rate of Internet of Things (IoT) devices in smart home applications, it is vital to safeguard the privacy and security of information, communications among IoT devices, and the underlying infrastructure. In open communication scenarios, an adversary can easily tamper with data transmitted by IoT communication devices. This paper proposes a softwarized lightweight authentication key agreement scheme for smart home applications in Software-Defined IoT (SDIoT) environment. The proposed scheme comprises registration, authentication, and key selection phases. Devices are registered in the registration phase after the successful completion of validation checks, while sessions’ keys are granted to the registered devices in the authentication phase. In the final phase, controllers classify IoT devices based on pre-defined parameters and impose class-specific access control based on classification. The security of the proposed scheme is validated using the widely accepted AVISPAs verification tool against a strong adversary. Simulation results demonstrate that our proposed scheme outperforms existing schemes in terms of running time, computation complexity, and energy consumption. It is found that the proposed scheme has significant cost-effectiveness.

Light fidelity (LiFi) and wireless fidelity (WiFi) offer high data rate and extensive coverage area, respectively. Each of them operate on different and non-interfering spectra. Integration of both these technologies creates a hybrid WiFi/LiFi network, which can provide high achievable sum rate with improved network mobility. In hybrid WiFi/LiFi networks, resource allocation remains a significant challenge. In this study, we propose a model-free deep reinforcement learning (DRL) method for resource allocation, which is simulated under realistic conditions, including signal blockage, load balancing and user equipment (UE) mobility. Simulations results show that the proposed DRL approach surpasses traditional Deep Q-Network (DQN)-based methods by 52.5% in terms of achievable sum-rate with 12.5% lower transmission power usage, leading to better resource management and enhanced overall network performance.

Cooperative spectrum sensing among cognitive radio (CR) enabled smart devices improves the sensing performance in deep fading environments. However, it is prone to attacks by malicious users. These attacks become more severe when they are launched in collusion. In this paper, a collision penalty-based attack prevention method is proposed to protect honest CR enabled smart devices from individual and collusion attacks. First, an optimal decision fusion rule which maximizes the throughput of each CR enabled smart device is proposed. Then, we identify the possible malicious strategies and prevent them with appropriate bounds on the penalty. Investigations are based on the malicious utility obtained by the attackers after playing malicious strategies, and the utility reduction capability of an attack prevention mechanism. Simulations demonstrate that the proposed mechanism can achieve up to 700% of malicious utility reduction as opposed to 150% and 100% in the cases of existing Moral Hazard Principal Agent (MHPA) and friendly jammer attack prevention schemes. A game theoretic analysis of the proposed method shows that a unique Nash Equilibrium is achieved when all the CR enabled smart devices are honest.

The demand of newer technologies for fifth generation mobile communication systems is high in the current arena. A hybrid form of Free Space Optical (FSO) Communication, also called Light Fidelity (LiFi) Communication, and Wireless Fidelity (WiFi) Communication technologies, has emerged as a promising candidate to fulfill it. Such a system is termed as hybrid WiFi/LiFi Communication System. The joint optimization problem of bandwidth, user association and power for data rate maximization in these hybrid systems is non-concave. Deep Q Network (DQN) Learning based algorithms offer solution to non-concavity. However, existing DQN learning based solutions are often restricted to static networks. They face complexity issues in dynamic networks. In this paper, we address the dynamic hybrid WiFi/LiFi Communication system with DQN transfer learning algorithm. Transfer learning is used to gather information about a newly entering mobile user in the network, thereby improving the overall throughput of the network. Simulations show that the proposed algorithms perform well than the existing optimization algorithms in throughput maximization.