The next generation of wireless networks demands transformative solutions to achieve ultra-reliable, high-capacity, and energy-efficient communications. Conventional systems based solely on relay-assisted transmission or intelligent reflecting surfaces (IRS) face inherent trade-offs in spectral efficiency, coverage, and energy consumption. In this work, we propose a novel cooperative hybrid communication system that integrates IRS with relay-assisted transmission to form a robust, scalable, and energy-aware design for beyond 5G (B5G) networks. Unlike traditional architectures, the proposed system enables mobile users to receive signals from the base station through three cooperative transmission paths: direct relay transmission, IRS reflection, and relay-assisted IRS reflection. This cooperative signal propagation enhances path diversity, improves link robustness, and increases spatial efficiency. Additionally, we introduce a joint optimization algorithm to determine the optimal IRS placement and reflecting element configuration, maximizing system throughput under practical rate constraints. Simulation results under Rayleigh fading conditions demonstrate the effectiveness of the proposed system across various deployment scenarios, including SISO, MISO-OMA, and MISO-NOMA. In the MISO-NOMA setting with 40 dBm transmit power, the system achieves a sum rate of 49.80 bits/s/Hz and an energy efficiency of 16.60 bits/Joule, outperforming benchmark hybrid relay-IRS-aided and relay-dominant cooperative hybrid systems. These findings establish the proposed system as a high-performing and energy-efficient solution for future wireless networks.

Hyperspectral imaging (HSI) has evolved as an important tool for many applications, including remote sensing, crime investigation, target detection, disease diagnosis, and anomaly detection. Among these, hyperspectral underwater target detection (HUTD) presents unique challenges due to spectral distortions caused by water absorption and scattering. Conventional methods often struggle with spectral variability, complicating detection accuracy. This paper introduces spectral variability-aware hybrid autoencoder (SVHAE) for HUTD, a novel autoencoder-based unmixing network incorporating parallel linear and nonlinear decoders to improve underwater target detection. Our method effectively reduces the effect of spectral distortions and addresses variability using a combined loss function integrating Kullback-Leibler divergence, mean squared error, and spectral angle distance. Experimental validation on real-world and simulated datasets demonstrates that our proposed SVHAE outperformed state-of-the-art methods by achieving superior AUC values. These advancements contribute to the progressing field of HUTD, making the way for robust solutions in marine exploration and detecting targets under the water.

Microalgae-based wastewater treatment has resulted in a paradigm shift toward nutrient removal and simultaneous resource recovery. However, traditionally used microalgal biomass quantification methods are time-consuming and costly, limiting their large-scale use. The aim of this study is to develop a simple and cost-effective image-based method for microalgae quantification, replacing cumbersome traditional techniques. In this study, preprocessed microalgae images and associated optical density data were utilized as inputs. Three feature extraction methods were compared alongside eight machine learning (ML) models, including linear regression (LR), random forest (RF), AdaBoost, gradient boosting (GB), and various neural networks. Among these algorithms, LR with principal component analysis achieved an R2 value of 0.97 with the lowest error of 0.039. Combining image analysis and ML removes the need for expensive equipment in microalgae quantification. Sensitivity analysis was performed by varying the train-test splitting ratio. Training time was included in the evaluation, and accounting for energy consumption in the study leads to the achievement of high model performance and energy-efficient ML model utilization.

The Internet of Things (IoT) and artificial intelligence (AI) enabled IoT is a significant paradigm that has been proliferating to new heights in recent years. IoT is a smart technology in which the physical objects or the things that are ubiquitously around us are networked and linked to the internet to deliver new services and enhance efficiency. The primary objective of the IoT is to connect all the physical objects or the things of the world under a common infrastructure, allowing humans to control them and get timely, frequent updates on their status. These things or devices connected to IoT generate, gather and process a massive volume of binary data. This massive volume of data generated from these devices is analyzed and learned by AI algorithms and techniques that aid in providing users with better services. Thus, AI-enabled IoT or artificial IoT (AIoT) is a hybrid technology that merges AI with IoT and is capable of simplifying complicated and strenuous tasks with ease and efficiency. The various machine learning (ML) and deep learning (DL) algorithms in IoT are necessary to ensure the IoT network’s improved security and confidentiality. Furthermore, this paper also surveys the various architectures that form the backbone of IoT and AIoT. Moreover, the myriad state-of-the-art ML and DL-based approaches for securing IoT, including detecting anomalies/intrusions, authentication and access control, attack detection and mitigation, preventing distributed denial of service (DDoS) attacks, and analyzing malware in IoT, are also enlightened. In addition, this work also reviews the role of AIoT in optimizing network efficiency, securing IoT infrastructures, and addressing key challenges. Furthermore, it explores cutting-edge technologies like blockchain, 6G-enabled AIoT, federated learning (FL), and hyperdimensional (HD) computing, indicating their potential in advancing IoT and AIoT-driven applications within sectors like healthcare, autonomous systems, and industrial automation. Therefore, based on the plethora of prevailing significant works, the objective of this manuscript is to provide a comprehensive survey that expounds on AIoT in terms of security, architecture, applications, emerging technologies, and challenges.

Biomedical hyperspectral imaging (HSI) is a powerful tool for disease diagnosis and medical research, offering rich spectral information for precise tissue characterization. However, the secure transmission and storage of these sensitive images, along with associated patient data, pose significant challenges. This research introduces a novel deep learning-based steganographic method to embed confidential patient data within hyperspectral medical images, ensuring both security and data integrity. Unlike conventional least significant bit (LSB) methods, which are prone to distortion and detection, we employ a GAN-based approach to generate imperceptible and high-capacity steganographic images. To further enhance security, patient data is first encrypted using XChaCha with Argon2 key derivation before embedding, ensuring that access is restricted to authorized users. This research advances privacy-preserving techniques in medical imaging, enabling secure and efficient patient data transmission while maintaining the diagnostic utility of hyperspectral medical images.

In this letter, we investigate an optimization design for reconfigurable intelligent surface (RIS)-assisted amplify-and-forward relay network. Specifically, we formulate an energy-efficiency (EE) maximization problem by simultaneously optimizing the beamforming, phase shifts, and transmit power. To solve this problem, we design an alternative optimization algorithm based on successive convex approximation (SCA) and Dinkelbach transformation. Firstly, through the SCA and Dinkelbach transformation, we can obtain the optimal beamforming. Then, with the optimal beamforming, we use the semi-definite relaxation and Gaussian randomization to obtain the optimal phase shifts. Subsequently, with the optimal beamforming and phase shifts, we further apply the Karush-Kuhn-Tucker conditions to obtain the optimal transmit power. Simulation results illustrate that our optimization design can effectively enhance the network EE.

Hyperspectral image processing, a critical domain in remote sensing and Earth observation, has witnessed exponential growth in feature extraction methods. In light of this, our study delves into the ongoing debate between spectral and spatial feature extraction strategies, a question of paramount importance. To address this, we embarked on a rigorous investigation using two eminent deep learning architectures, namely 3DCNN(Three dimensional convolutional neural network) and ResNet(Residual Neural Network) and employed two benchmark hyperspectral datasets: Indian Pines and Salinas. Our research was meticulously structured, encompassing spectral feature extraction methods such as mean, mode, standard deviation, and skewness, as well as spatial feature extraction techniques employing advanced convolutional operations. The spectral standard deviation feature extraction method using 3DCNN yields an overall accuracy of 98.71% for the Indian Pines dataset. In the Salinas dataset, the spectral and spatial mean feature extraction methods achieve accuracies of 99.78% and 99.89% respectively. When employing the ResNet architecture, the spectral skewness feature extraction method attains an accuracy of 99.83% for the Indian Pines dataset. Meanwhile, in the Salinas dataset, spectral standard deviation feature extraction excels with an accuracy of 99.99%. Analyzing computational times, the 3DCNN on the Indian Pines data set shows that spatial mode feature extraction has the shortest time of 0.33 s, while in the Salinas dataset, spatial skewness feature extraction requires 1.60 s. For ResNet, Indian Pines’ spatial mode is the quickest at 2.10 s, and Salinas’ spatial mode is 10.66 s. In the comparison of spectral and spatial feature extraction methods, the likelihood of achieving lower computational times favors spectral methods. Additionally, the probability of attaining higher accuracy is notably higher with spectral feature extraction methods. Therefore, spectral approaches emerge as a superior choice.

Hyperspectral Image (HSI) processing stands at the forefront of remote sensing technologies, offering a wealth of detailed information crucial for various environmental and agricultural applications. The complexity of HSI datasets demands sophisticated techniques for accurate and efficient analysis. In this research, we investigate the efficiency of feature selection and extraction methods for HSI classification using Fast-3D Convolutional Neural Networks (CNN) and Residual Neural Networks (ResNet). Our motivation stems from the necessity to enhance accuracy and computational efficiency in remote sensing applications. We conducted experiments on two prominent HSI datasets, Indian Pines and Salinas, leveraging feature selection techniques like Correlation-based Feature Selection (CFS), Mutual Information Feature Selection (MIFS), Linear Discriminant Analysis (LDA), and Recursive Feature Elimination (RFE), alongside feature extraction methods such as Principal Component Analysis (PCA), Incremental PCA (IPCA), Quadratic Discriminant Analysis (QDA), and Non-negative Matrix Factorization (NMF), implemented within both Fast-3D CNN and ResNet architectures. Our results indicate that in hyperspectral data classification for both Indian Pines and Salinas datasets, IPCA and QDA excelled in accuracy (99.68%, 99.14%) within the Fast-3D CNN architecture, with IPCA demonstrating speed (0.40 s, 1.40 s). CFS displayed notable efficiency (99.27%, 90.88%) with minimal processing time (0.56 s, 1.35 s) in the Fast-3D CNN architecture. RFE achieved high accuracy (99.88%, 96.39%) with reasonable processing times (2.28 s, 11.53 s) in the ResNet architecture. While IPCA led in extraction (99.90%, 99.12%), it required more time (21.17 s, 2.23 s) in the ResNet architecture. These results underscore the practical significance of feature extraction methods like IPCA for accuracy-focused tasks and the efficiency of feature selection methods like CFS in scenarios prioritizing computational time.

Cholangiocarcinoma, a form of liver bile duct cancer, is challenging to detect due to its critically low 5-year survival rate. Conventional imaging modalities, such as Computed Tomography (CT) and Magnetic Resonance Imaging (MRI), are widely used, but recent advancements in Hyperspectral Imaging (HSI) offer a promising, non-invasive alternative for cancer diagnosis. However, supervised learning methods often require large annotated datasets that can be difficult to obtain. To alleviate this limitation, we propose an unsupervised learning strategy using Generative Adversarial Networks (GANs) for cholangiocarcinoma detection. This approach, named Unsupervised Spectral and Spatial Attention-based GAN (USSGAN), employs an unsupervised Spectral-Spatial attention-based GAN to classify and segment cancerous regions without relying on labeled training data. The integration of an adaptive step size into Tasmanian Devil Optimization (TDO) enhances the convergence speed and effectively captures diverse cancerous features. Enhanced Tasmanian Devil Optimization (ETDO) further improves segmentation performance, making the framework robust and computationally efficient. The proposed method was tested on a publicly available multidimensional choledochal cholangiocarcinoma dataset, achieving superior performance compared with existing techniques in the literature. USSGAN demonstrated high accuracy across key metrics such as overall accuracy (OA), average accuracy (AA), and Cohen’s Kappa. Ablation studies confirmed the critical contributions of the proposed enhancements to the overall performance. With an overall accuracy of 98.03%, the USSGAN closely aligns with the assessments of experienced pathologists while maintaining minimal computational requirements. Its lightweight nature ensures real-time deployment, providing results within a minute, making it a practical and effective solution for clinical applications.

In seismic data processing, denoising is one of the essential steps to identifying the earth's subsurface layer information. The noise present in the seismic data is categorized into two types: random and erratic noise. The random noise is distributed uniformly over the seismic data. The erratic noise attenuation is always challenging due to the unknown distribution of high-amplitude peaks over seismic data. The existing double sparsity dictionary learning (DSDL) method performs with analytical and adaptive transforms; both the transforms include iterative algorithms with K-singular-value decomposition (SVD); it is computationally costly, and the dictionary is initialized with trained data. To address these limitations, we propose a novel method of dictionary learning with double regularization (DLDR) to denoise both random and erratic noise from seismic data. In double regularization, we used with l1 -norm and nuclear norm. The denoised data is applied to the alternating direction method of multipliers (ADMMs) to improve denoising while preserving the signal features from seismic data while reducing the computational cost. We evaluated the performance of the proposed method using signal-to-noise ratio (SNR), mean squared error (MSE), and local similarity map. The numerical results demonstrated that the proposed method resulted in higher SNR, lower MSE, and less signal leakage from seismic data. The method gives precise interpretation from the denoised seismic data.

The rapid expansion of Internet of Things (IoT) applications has driven advancements in networking technologies like Low-Power Wide-Area Networks (LPWANs) to extend coverage and enhance the lifespan of IoT devices (IoDs). However, real-world IoT networks are typically heterogeneous, comprising static and dynamic IoDs leading to variations in network topology. These fluctuations cause challenges like increased data latency and energy imbalances, which hinder efficient information flow. To overcome these issues, this paper presents a novel approach that integrates Small-World Characteristics (SWC), inspired by social network theory, into heterogeneous LPWANs using reinforcement learning. Specifically, the Q-learning technique is employed to introduce new long-range links into the network, enhancing connectivity and optimizing performance. Different conventional networks with varying numbers of mobile nodes are studied in this work followed by their subsequent transformation to small-world versions. The performance of the networks is optimized in terms of energy efficiency and latency in data routing. It is observed that, irrespective of the network (conventional or small-world), the performance is better if the number of static nodes is greater. Furthermore, independent of the degree of dynamicity, the SW-LPWAN is more energy efficient and has lower transmission delay than the corresponding conventional network. Numerically, SWLPWANs achieve up to 14.6% faster data transmission speeds, supporting 19.7% more active IoDs, and maintaining 15.5% higher residual energy compared to conventional networks.

This paper proposes a secure medical data hiding system that integrates hybrid AES-256 and ECC-256 encryption with Hamming-LSB-OPAP steganography. Patient information is first encrypted and structured into a QR code, which is then invisibly embedded into grayscale medical images. The system is evaluated based on imperceptibility metrics such as PSNR, SSIM, and MSE under various attack conditions, including Gaussian noise, salt-and-pepper noise, and JPEG compression. Experimental results demonstrate that the proposed method achieves an average PSNR of 51.31 dB, SSIM of 0.9965, and MSE of 0.48, indicating minimal distortion and high visual quality. Although QR decoding from the stego images was unsuccessful, the limitation is acknowledged, and future work will explore transform-domain embedding to improve robustness against distortion.

In the era of cloud-based healthcare and the Internet of Medical Things (IoMT), securing medical images is essential to protect patient privacy and ensure data integrity. This paper proposes a robust encryption scheme tailored for medical images, incorporating a 3D chaos-based system with layered security mechanisms. The approach integrates chaotic sequence generation, histogram equalization, spatial permutation, DNA- based encoding, and secure XOR operations. By employing chaotic maps for both key generation and permutation, alongside dynamically selected DNA rules, the method achieves high levels of confusion and diffusion. Experimental evaluations demonstrate strong performance in entropy, resistance to statistical and differential attacks, and robustness against brute-force and chosen-plaintext attacks. The proposed system offers a secure and efficient solution for protecting sensitive medical data in IoMT environments.

This paper examines the integration of Intelligent Reflecting Surfaces (IRS) in beyond 5G (B5G) communication networks, where the IRS reflects signals with adjustable phase shifts. By optimizing these phase shifts, called passive beamforming, substantial improvements in communication performance can be achieved. We maximize energy efficiency in the uplink communication, utilizing the IRS. However, including an IRS introduces complexities, particularly in channel estimation. To address this, we examine two innovative approaches to minimize the channel estimation overhead: the first leverages a grouping strategy for the reflecting elements. In contrast, the second approach utilizes positioned-based phase optimization. Simulation results confirm that the IRS significantly enhances energy efficiency compared to the traditional system.

Hyperspectral unmixing (HU) is crucial for extracting material information from hyperspectral images (HSI) obtained through remote sensing. Although linear unmixing methods are widely used due to their simplicity, they only address linear mixing effects. Nonlinear mixing models, while more complex, often focus solely on the nonlinear aspects affecting individual pixels. However, in practice, light reflected from materials within a pixel experiences linear and nonlinear interactions, necessitating a hybrid mixing model (HMM) that leverages spatial and spectral information. This work proposes a novel deep learning-based autoencoder (AE) with dual-stream decoders to enhance spectral unmixing. Our approach employs multitask learning (MTL) to process spatial and spectral information concurrently. Specifically, one decoder stream performs linear unmixing from HSI patches, while the other stream utilizes fully connected layers to capture and model the nonlinear interactions within the data. By integrating linear and nonlinear information, our method improves the accuracy of unmixing the mixed spectrum. We validate the effectiveness of our architecture on three real-world HSI datasets and compare its performance against various baseline methods. Experimental results consistently demonstrate that our approach outperforms existing methods, as evidenced by superior spectral angle distance (SAD) and mean squared error (MSE) metrics.

In seismic data processing, denoising is one of the important steps to identify the Earth’s subsurface layer information. The erratic noise attenuation is always challenging due to the unknown distribution of high-amplitude peaks over seismic data. In literature, the double sparsity dictionary learning (DSDL) methods were used for erratic and random noise attenuation. Here, analytical and adaptive transformations are performed sequentially to attenuate erratic and random noises. However, the DSDL technique leads to a high computational cost due to K-SVD. Therefore, we propose a double dictionary learning (DDL) method to denoise both random and erratic noise by preserving the signal features from seismic data. The method uses two parallel adaptive dictionaries for simultaneous denoising, and both dictionaries are concatenated further to form a comprehensive dictionary. The regularized K-SVD was used to update the dictionary and sparse coefficients for signal preservation. The DDL method effectively reduced the computational costs. The DDL method was applied to different synthetic and field datasets for denoising. The numerical results show that the proposed method provides a higher signal-to-noise ratio (SNR), lower mean-squared error (mse), and less signal leakage than existing state-of-the-art denoising methods.

Recently, reconfigurable intelligent surface (RIS) technology has provided new opportunities to rescale the performance metric and boost network coverage for sixth-generation (6G) wireless communications. In RIS-assisted wireless communication systems, traditional analog beamforming requires a huge number of phase shifters, which increases exponential and computational complexity. In order to accomplish computationally efficient hybrid beamforming for 6G communications, we therefore propose a branchwise phase shifters (BPS) architecture of analog beamforming in this paper. To establish the best baseband configuration for wireless 6G communications, we first consider an RIS element and a transmitter antenna branchwise millimeter wideband (B-MW) channel model. Following that, we design a BPS-based analog beamforming architecture using a power splitting technique, which splits the transmit power associated with each RF chain and allows each RF chain to utilize two PS branches. The B-MW channel can easily provide a low-dimensional hybrid beamforming solution with the BPS architecture, which allows easy rescaling of the beamforming gain, thus enhancing the achievable rate. We then present an iterative algorithm with zero-forcing to tackle a joint optimization problem of RIS-assisted hybrid beamforming. Lastly, the beam coverage and achievable rate of the proposed BPS architecture are verified by the simulation results, which outperform the traditional single, double, and combined phase-shifter architecture.

Intelligent Reflecting Surfaces (IRS) offer a promising solution for enhancing sum rates in wireless networks by dynamically adjusting signal reflections to optimize propagation paths. When combined with Non-Orthogonal Multiple Access (NOMA), which enables multiple users to share the same frequency band, significant improvements in spectral efficiency can be achieved. However, as the number of users increases in IRS-NOMA systems, ensuring consistently high data rates for all users becomes challenging due to coverage limitations and inefficient power allocation in static network configurations, leading to performance degradation in multi-user scenarios. To address these limitations, we propose a novel IRS-NOMA cooperative system designed to optimize sum rates through an intelligent power allocation algorithm, nearby users, and IRS to assist the base station in delivering signals and expanding network coverage. The proposed system operates in two phases: during the first phase, the base station transmits signals directly to users and indirectly through the IRS. In the second phase, nearby users assist in relaying signals to enhance coverage and reliability. The proposed system adopts a cascaded channel model to accurately capture the interactions between the base station, IRS, and users. By leveraging our optimization algorithm, the proposed system ensures efficient resource allocation, achieving superior spectral efficiency and fairness among users compared to traditional models. Numerical results validate the effectiveness of the proposed system, demonstrating its potential for next-generation IoT networks.

Integrating reconfigurable intelligent surfaces (RISs) in wireless communication systems holds tremendous promise for revolutionizing connectivity by offering scalability, cost-efficiency, and energy neutrality. However, navigating the complexities of dynamic environments poses significant challenges for power control in RIS-empowered wireless networks. The proposed methodology involves implementing a cooperative deep reinforcement learning (DRL) system with two interconnected networks, DRL-M and DRL-S. We called it as DRL master and slave DRL(M-S), which aims to optimize system performance and energy efficiency (EE). RL-M optimizes system performance by adjusting transmit beamforming and phase shift. The results show that increasing the transmit power (from 0 to 10 to 20 dB) leads to a proportional increase in the average reward, reaching approximately values of (2.5, 4.8, 7.8). This average reward serves as feedback for the DRL-S network, assisting it in intelligently managing power transmission to adapt to changing environmental conditions by leveraging the reward feedback from DRL-M, facilitating dynamic adjustment of power transmission based on variations in these rewards, either increasing or decreasing power transmission accordingly. This chapter contributes to advancing RIS-integrated wireless systems with enhanced power control capabilities, offering a robust solution to address the challenges of power control in RIS-enabled wireless systems operating in dynamic environments.

In the 6G wireless communication, intelligent reflecting surface (IRS) has emerged as a transformative technology in a new era of intelligent and efficient wireless networks. IRS can manipulate radio waves, which means they can help to improve communication in terms of coverage, capacity, and energy efficiency. IRS can overcome obstacles such as signal blockage, path loss, and interference, improving communication reliability and performance. IRS can adaptively reconfigure the wireless propagation environment according to changing conditions. IRS can adjust its reflective properties dynamically in real time, optimizing signal propagation based on user location and channel conditions. Propagation measurements are essential for understanding signal propagation processes and describing wireless channel behavior. These measurements involve collecting data on signal strength, fading, delay spread, and other channel parameters in various environments. Channel modeling techniques aim to represent wireless channel behavior in mathematical models accurately. These models incorporate factors such as path loss, multipath fading, shadowing, and interference to simulate the propagation of electromagnetic waves in different scenarios. Wireless channels are inherently nonstationary, evolving unpredictably in response to environmental changes. This unpredictability poses a significant challenge for propagation measurements, which aim to characterize the behavior of wireless channels over time and space. Overcoming these challenges requires integrating IRS into 6G wireless communication systems, which promises to make a big difference in performance. Thus, this chapter aims to comprehensively review the propagation measurements and channel modeling techniques in 6G wireless communication via an IRS.

The advent of sixth-generation (6G) wireless networks holds the promise of revolutionizing the landscape of the Internet of Things (IoT), expanding the horizons of wireless communication and ushering in a new era of IoT applications with unprecedented performance and reliability. However, a crucial requirement in this field is the need for precise positioning and localization of IoT devices, which is a fundamental necessity for a plethora of applications. Nevertheless, the existing positioning and localization methods used in 6G-IoT pose challenges due to blockages of the line-of-sight signals and interference and difficulties arising from multipath propagation, which results in new requirements for positioning and localization. These fundamental necessities for precise positioning and localization can be fulfilled with a reconfigurable intelligent surface (RIS), a potential candidate technology for the future 6G wireless communication. Thus, integrating RIS in the IoT can enhance the accuracy of positioning while offering the added benefits of being economical and energy-efficient. In this chapter, the role of RIS-assisted 6G-IoT networks in wireless positioning and localization is explained initially. Then, the fundamental localization principles and the RIS-aided localization algorithms are explored. After that, the state-of-the-art research on positioning and localization, comprising RIS-assisted millimeter-wave positioning systems, RIS for indoor, near-field, outdoor, and far-field localization, and RIS for terahertz communication, is elaborated in detail. Finally, this chapter concludes by discussing the potential challenges and future research directions of RIS-aided 6G-IoT for wireless positioning and localization.

The advent of reconfigurable intelligent surfaces (RISs) technology has ushered in a new era of wireless communication, promising unprecedented capabilities and opportunities. However, implementing RIS-based wireless communication systems raise significant security and privacy concerns. This work delves into the multifaceted landscape of privacy and security issues associated with RIS deployments. Privacy concerns stem from the manipulation of wireless signals, raising issues of data leakage, location privacy, user profiling, and surveillance. In parallel, security challenges encompass unauthorized access, data tampering, signal jamming, physical infrastructure vulnerabilities, and regulatory compliance issues. Addressing these issues requires robust encryption, authentication mechanisms, intrusion detection, rigorous privacy and security regulations adherence. This research outlines a comprehensive strategy for various attacks and threats, ensuring data confidentiality, integrity, and availability in RIS-enabled networks. Additionally, the topic of physical layer security for RIS-assisted networks is being addressed. Incorporating physical layer security measures into RIS deployments enhances the confidentiality and integrity of wireless communication, making it more resilient against eavesdropping and unauthorized access. Multiple challenges are identified for future research to fully utilize the benefits of the IRS in physical layer security and covert communications. This chapter offers insights into the evolving domain of RIS, shedding light on the imperative need to balance its transformative potential with protecting individual privacy and system security.

This paper proposes a cooperative reinforcement learning-based framework to jointly optimize active and passive beamforming in a passive Intelligent Reflecting Surface (PIRS)-assisted wireless communication system for green and secured communications. The framework employs two Deep Deterministic Policy Gradient (DDPG) agents: one at the Base Station (BS) for active beamforming control and the other at the PIRS for phase shift adjustments in passive beamforming. The BS agent optimizes beamforming for both Line-of-Sight (LOS) and Non-Line-of-Sight (NLOS) paths, while the PIRS agent adjusts phase shifts to improve the constructive contribution of the reflected signals. The user assesses the combined direct and reflected signals, using a secure rate (Rsec) based reward to guide the learning process of both agents. Through channel state information (CSI) from BS-PIRS, PIRS-user, and BS-user links, the agents learn coordinated actions to maximize the secure rate, boosting signal strength for the intended user and reducing eavesdropping risks. Simulations reveal that the proposed framework achieves substantial secured data rate efficiency gains with BS antenna configurations of 4, 8, and 16. However, further increases in antenna count require BS power adjustments for optimal performance. This joint optimization approach significantly improves secure rate and signal quality, positioning it as a valuable solution for next-generation wireless networks, such as 6G, that demand high data rates, enhanced security, and reliable connectivity.

In biometric authentication, traditional methods like RFID cards, PINs, or single-modal biometrics are susceptible to spoofing and unauthorized access. This paper proposes a novel biometric system that integrates face and voice recognition to enhance identity verification, combining facial and vocal characteristics for a more secure and reliable authentication process. Utilizing deep learning models trained on extensive datasets, the system extracts and matches unique facial and vocal features, significantly reducing fraudulent access while maintaining high accuracy in diverse real-world conditions. It is designed to handle variations in facial expressions, lighting conditions, background noise, and facial masks, ensuring robust verification even when masks are worn. By incorporating a facial mask dataset, the system effectively distinguishes between masked and unmasked faces, enhancing security and reliability. Additionally, the system is optimized for real-time processing, making it highly scalable and efficient for applications requiring fast and precise identity verification. The integration of face and voice recognition, along with the use of a facial mask dataset, improves both security and authentication speed, offering a comprehensive and sophisticated response to modern identity verification challenges. Experimental results demonstrate that this approach outperforms traditional single-modal biometric systems, reducing the risk of spoofing and ensuring high performance even in dynamic environments.

Oral cancer is a significant global health concern, often leading to high mortality rates due to late-stage diagnosis and the lack of effective early detection methods. Despite advances in medical science, the absence of reliable early diagnostic tools remains a critical challenge. Hyperspectral imaging (HSI) has emerged as a powerful noninvasive technology, capturing detailed spectral information across a wide range of wavelengths. This allows for accurate differentiation between cancerous and healthy tissues, improving early detection and enhancing treatment outcomes. In this study, we propose the use of HSI for early oral cancer diagnosis. To address the scarcity of labeled data, we developed a synthetic hyperspectral dataset that includes spectral signatures of both normal and cancerous tissues. The dataset was generated using a bilinear mixing model, with key spectral features extracted through Vertex Component Analysis (VCA) and abundances computed using Non-Negative Least Squares (NNLS). The model's performance was evaluated using Spectral angle distance (SAD) and Root mean square error (RMSE) metrics. Our findings demonstrate that HSI significantly improves the accuracy of early oral cancer detection, outperforming traditional methods. This work highlights the potential of advanced imaging technologies in revolutionizing cancer diagnosis, offering a robust framework for non-invasive detection and showcasing the effectiveness of synthetic datasets in medical imaging research.

In crime investigations, rapid bloodstain identification is crucial. Hyperspectral imaging (HSI) offers a non-destructive solution. Our investigation into preprocessing techniques to improve classification accuracy and reduce computation time reveals that the best options are max normalization and mean filter.

This survey paper provides a comprehensive overview of integrating Multiple-Input Multiple-Output (MIMO) with Intelligent Reflecting Surfaces (IRS) in wireless communication systems. IRS is known as reconfigurable metasurfaces, have emerged as a transformative technology to enhance wireless communication performance by manipulating the propagation environment. This work delves into the fundamental concepts of MIMO and IRS technologies, exploring their benefits and applications. It subsequently investigates the synergies of resource allocation and energy efficiency that emerge when these technologies are combined, elucidating the IRS improved in MIMO systems through signal manipulation and beamforming. Through an in-depth analysis of various techniques and cutting-edge algorithms in resource allocation and energy efficiency can explore the key research areas such as optimization techniques, beamforming strategies and practical implementation consideration. Furthermore, it provides open research directions, individually addressing topics such as limitations of resource allocation and energy efficiency in the MIMO IRS system. This paper offers insights into MIMO-enabled IRS systems challenges and future trends. Through presenting a consolidated view of the current state-of-the-art, this survey underscores their potential to revolutionize wireless communication paradigms, ushering in an era of enhanced connectivity, spectral efficiency and improved coverage.

Liver bile-duct cancer is also called as cholangio- carcinoma that stands a significant global health hazard, due of its low 5-year survival rate that is about (2-24%). So Precise and prompt diagnoses is vital in order to improve patient diagnosis and increase survival rates. Hyperspectral imaging (HSI) offers a promising avenue for improving liver cancer diagnosis due to its ability to capture detailed continuous spectral plus spatial information that is beyond the visible range of the human eye. Classifying cholangiocarcinoma through HSI is complex because of its high dimensionality. To solve this,a network called as MedisawHSI is introduced in this article. Inspired from Jigsaw HSI that demonstrates superior performance compared to other Neural Networks. In this article we present Medisaw-based clas- sification involves dividing the hyperspectral image into smaller non - overlapping patches, which are then classified individually based on their spectral characteristics. Results demonstrate that we have achieved better results in comparison with the literature. This will help the surgeons in image - guided surgery, ultimately reducing the burden of liver cancer on global healthcare systems.

This article introduces a novel semi-supervised learning method for Cholangiocarcinoma detection using inherent statistical parameters of the image on the multidimensional Choledochal dataset. Results closely match the pathologist’s annotations, validated by image similarity indices.

Healthcare is a vital sector influencing societal well-being and economic stability. The COVID-19 pandemic has highlighted the critical need for innovative solutions, such as remote monitoring and real-time health tracking, to address emerging challenges. This paper examines the transformative potential of wireless technology in revolutionizing healthcare systems, emphasizing advancements in communication, remote surgeries, patient engagement, and cost efficiency. It explores the role of 6G technology in enabling high-speed data transfer, ultra-reliable connectivity, and low latency, providing the foundation for intelligent, connected healthcare ecosystems. Key challenges, including seamless connectivity, data privacy, and network scalability, are analyzed alongside strategies to overcome them, such as adopting 6G-enabled Internet of Everything (IoE), Intelligent Reflecting Surfaces (IRS) to counter signal blockages, and advanced latency reduction techniques. By reviewing state-of-the-art developments and real-world case studies, the paper demonstrates the indispensable role of wireless technology in enhancing patient outcomes, reducing healthcare costs, and ensuring universal access to high-quality care. It concludes with actionable recommendations for healthcare organizations to embrace these innovations for a resilient and efficient future.

Pepper is a valuable medicinal substance and an expensive aromatic. For profit purposes, some vendors adulterate dried papaya seeds with black pepper due to their physical similarities. This impurity can lead to various health issues. Several existing methods are available to detect this adulteration, but they have some limitations. To overcome these challenges, the study employed a technique called Hyperspectral Imaging (HSI) by using machine learning classification algorithms. This research experimented with various machine learning classification algorithms, including Decision Tree, Random Forest, and Linear Discriminant Analysis (LDA). Among these algorithms, the Decision Tree algorithm stood out as the most effective in achieving an impressive classification accuracy of 99.93%, with a computational time of 6.76 seconds. This hyperspectral imaging analysis and the machine learning classification hold significant promise in enhancing food quality assurance, ensuring consumer health, and reinforcing trust within the industry.

Ghee, a popular clarified butter widely consumed around the world, particularly in India, is valued for its taste and health benefits. However, some vendors adulterate it with cheaper substances such as vanaspati to increase profits, which can be harmful to consumers. This requires robust methods for quality assurance. In response to this challenge, this article presents a noninvasive method for detecting ghee adulteration with vanaspati using hyperspectral imaging (HSI). We created a data set consisting of hyperspectral images with different proportions of ghee and vanaspati. This data set was tested on various machine-learning algorithms. The results were impressive, showing a highly accurate detection of adulteration (99. 35%) with the K-Nearest Neighbor (KNN) and Random Forest algorithms. These methods were quick to converge, facilitating faster results.

In this age of rapid digitalization, secure storage and transmission of sensitive data have become crucial. This study introduces a novel encryption technology that embeds critical data within a hyperspectral image (HSI) to ensure secure storage and transmission. The technology takes advantage of hyperspectral images' complex, high-dimensional nature to conceal the underlying data, successfully shielding it from unauthorized users. By combining encryption and steganography, sensitive data is masked so that even if the image is intercepted, it seems to be a typical hyperspectral image with no visible anomalies. This deceptive strategy confuses attackers, making it extremely difficult to determine the presence of encrypted data, let alone where it's located within the image. Furthermore, the data is connected to a unique key, providing an additional layer of protection. Without this key, any attempts to decode the data will fail, adding an extra layer of security against unauthorized access. This research investigates the use of hyperspectral images as a medium for secure data transmission and storage, presenting a strong solution for protecting sensitive information in various applications.

In this chapter, we investigate the energy efficiency (EE) of the intelligent reflecting surface (IRS)–assisted intelligent transportation system (ITS) under both the Rayleigh and Nakagami-m fading conditions. Since its inception, ITS is growing rapidly, as it helps to provide seamless data transfer between vehicles, avail safe transportation, and avoid accidents. Nevertheless, the amount of data processed in ITS demands more transmission power. To address this, IRS blocks with several passive reflective elements have been recognized as a promising technology to reduce power consumption and enhance EE. In recent years, with the rapid increase in data usage, mobile users demand more transmission power. It is difficult to satisfy all users with limited transmission power; however, IRS has the ability to solve the power requirement problem by using the reflecting elements. In our chapter, two different fading environments (i.e., Rayleigh, Nakagami-m) are adopted to meet the needs for the practical implementations of IRS-assisted ITS. In addition to this, the phase-shift optimization of each IRS element becomes a challenging task which also makes it difficult to estimate the channel for IRS-assisted ITS. To overcome the above challenges and optimize the EE, we develop a novel IRS element clustering method and a passive beamforming technique based on the desired location of the ITS. Furthermore, we analyze the EE and also spectral efficiency of the IRS-assisted ITS with multiple IRS blocks. Numerical results show that the multiple IRS blocks can significantly improve the ITS performance in terms of EE

Globally, the rapid development of fifth-generation (5G) networks is being widely deployed. Meanwhile, efforts of extensive research from industries and academia have begun to explore sixth-generation (6G) communication networks beyond 5G. The design and optimization of 6G networks with high levels of intelligence have been made easier with the aid of artificial intelligence (AI) techniques. These wireless 6G emerging technologies, such as optical-free technology, quantum technology, native network slicing, integrated access backhaul networks, and holographic beam forming, enable the wireless propagation environment via signal transmissions and receptions are discussed. Based on the edge AI requirements, the product of combining AI with 6G communication technologies were introduced optimally. This technique focused on federated, decentralized, split, distributed reinforcement, and trustworthy learning models that can improve the potential of 6G communication networks in various applications. Edge AI deployed the centralized, decentralized, hybrid, self-learning, and end-to-end architecture to the edge nodes that provide the user's service to fully exploit the power of edge computing and it also achieves the maximum potential in data analytics. Due to the complexity of the virtual environment, edge AI faced higher demands on interacting with the metaverse framework. The main intention of this work is to bring a virtual and real-world conversation with billions of people who can interact with one another. In contrast to conventional intelligent applications, this framework explores the system architecture, virtual communication platforms, software, and hardware facilities to achieve low latency, ultra-high bandwidth, and reliability in user-defined networks. To be more precise, this chapter demonstrates the edge cloud metaverse (ECM), mobile ECM, and decentralized metaverse architectures that utilize the edge AI empowered by the 6G to address resource and compute restrictions in the framework. Additionally, a summary of the technical features and applications presented in the metaverse is established. Finally, the technical convergence of edge computing AI with 6G technologies is emphasized along with future research possibilities.

Hyperspectral imaging (HSI) has gained prominence in various fields of science. In particular, it has spurred much interest in biomedical imaging especially cancer (such as skin, breast, oral, colon, pancreatic, and prostate) detecting applications. Of them, breast cancer (BC) is known to be the second-largest cause of mortality throughout the world. According to the Cancer Registry Program, over 1.3 million people in India are suffering from BC, and more recently, the numbers seem to be growing exponentially. Currently, no permanent cure for metastatic BC is reported; nevertheless, detecting it at an earlier stage and treating accordingly is shown to reduce its severity, i.e., increasing the survival rate. To effectively detect BC, several optical techniques including mammography, ultrasound imaging, computed tomography, positron emission tomography, and magnetic resonance imaging are widely used. Note that these methods have their own merits and demerits such as the false-negative results, usage of higher-energy radiation, and poor soft tissue contrast, to name a few. Therefore, to validate the imaging results, a biopsy (using surgical excisions) is often performed, which is painful, troublesome, and may cause discomfort for a longer period. For this reason, cancer detection via non-invasive imaging methods is highly sought. Techniques such as thermal imaging, photo-acoustic imaging, and, more recently, HSI are shown to be providing satisfactory results at the laboratory scale. This chapter comprehensively reviews the utilization of HSI technique for the detection of various stages of breast cancer. We also review the state-of-the-art deep learning frameworks that are employed for automated breast cancer detection.

Object detection is a crucial task with numerous applications. The ability to detect changes in an object requires monitoring its behavior over time to recognize any alterations. This task is crucial in various domains, ranging from basic image analysis to remote sensing applications, where understanding geographic changes is of utmost importance. For example, in the production of printed circuit boards and integrated circuits, detecting component errors is essential. Similarly, in astronomy, tracking the movement of astronomical objects and changes in land cover due to tectonic plate deviations are of great interest. Change detection and tracking models are therefore in high demand. However, current models that use Earth Mover' Distance (EMD) for binary classification of object changes have limited applications. Therefore, an alternate position change identification model that can function as a substitute for deep learning methods is required. In this study, we propose a model that utilizes Mean Square Error (MSE)in place of EMD and considers the variation in image intensity from pixel to pixel to improve accuracy. Moreover, to overcome the limitations of binary classification our model categorizes images into multiple groups based on their chronological position. This enables us to identify the differences between various time periods more accurately. To train and evaluate our model, we use synthetic images, allowing us to create a model that can function with less data compared to current methods. Overall, our proposed model can significantly improve object change detection in various domains, making it a valuable addition to the field.

Cholangiocarcinoma is one of the rarest yet most aggressive cancers that has a low 5-year survival rate (2%-24%) and thus often requires an accurate and timely diagnosis. Hyperspectral Imaging (HSI) is a recently developed, promising spectroscopic-based non-invasive bioimaging technique that records a spatial image (x, y) together with wide spectral (λ) information. In this work, for the first time we propose to use a three-dimensional (3D)U-Net architecture for Hyperspectral microscopic imaging-based cholangiocarcinoma detection and classification. In addition to this architecture, we opted for a few preprocessing steps to achieve higher classification accuracy (CA) with minimal computational cost. Our results are compared with several standard unsupervised and supervised learning approaches to prove the efficacy of the proposed network and the preprocessing steps. For instance, we compared our results with state-of-the-art architectures, such as the Important-Aware Network (IANet), the Context Pyramid Fusion Network (CPFNet), and the semantic pixel-wise segmentation network (SegNet). We showed that our proposed architecture achieves an increased CA of 1.29% with the standard preprocessing step i.e., flat-field correction, and of 4.29% with our opted preprocessing steps.

In this paper, we have investigated the performance of energy efficiency (EE) for Intelligent Transportation Systems (ITS), which recently emerged and advanced to preserve speed as well as safe transportation expansion via a cooperative IRS-relay network. To improve the EE, the relay model has been integrated with an IRS block consisting of a number of passive reflective elements. We analyze the ITS in terms of EE, and achievable rate, with different signal-to-noise ratio (SNR) values under Nakagami-m fading channel conditions that help the system to implement in a practical scenario. From the numerical results it is noticed that the EE for the only relay, IRS, and proposed cooperative relay-IRS-aided network at SNR value of 100 dBm is 30, 17, and 48 bits/joule respectively. In addition, we compare the impact of multi-IRS with the proposed cooperative IRS-relay and conventional relay-supported ITS. Simulation results show that both the proposed cooperative IRS-relay-aided ITS network and multi-IRS-aided network outperform the relay-assisted ITS with the increase in SNR.

Cancer is one of the leading causes of mortality in the world with 9.6 million deaths recorded globally for the year 2018 alone. It involves uncontrolled cell division due to the activation of carcinogen genes and causes disorders in the growth of the tissue, which can occur in any part of the human body. Oral cancer (OC) is one of the prominent cancer types, especially in India, where 11.54% of new cases and 10.16% of deaths are caused by OC. To date, there is no promising treatment to cure cancer. Early detection of cancer can increase the chances of survival and quality of life after the treatment. Nowadays, various imaging and non-imaging diagnosis techniques are available. Imaging techniques became popular due to their non-invasiveness, nonpainful nature, and repetitiveness. X-ray, ultrasound, magnetic resonance imaging (MRI), computed tomography (CT), positron emission tomography (PET), and fluorescence imaging are some of those techniques. Fluorescence imaging uses fluorescence contrast agents, whereas all other techniques use ionizing radiation, which is harmful when repetitive imaging is required. However, all these techniques have their pros and cons. Recently, the research community has been working on thermal imaging, photoacoustic imaging, and hyperspectral imaging (HSI) to overcome such limitations. HSI is a promising technique for in vivo diagnosis, due to its multi-band capturing capability. It can capture the same location tissue with a higher spatial and spectral resolution, for a wide range of wavelengths from visible to near-infrared (NIR). It provides an ionization-free diagnosis, is less dependent on skilled pathologists, and produces quick results, and it is even safe for one to undergo this procedure many times. HSI can also be used for the effective identification of resection margin while operating to remove the OC tumor. It normally generates a huge three-dimensional data cube, where the effective processing of these data can produce good results. Currently, the research community is working on the OC HIS data using deep learning techniques like CNN, 3DCNN, R-CNN, Mask R-CNN, Customized CNN, etc. In this chapter, we present state-of-the-art works employing HSI with deep learning techniques for the early detection of OC and propose future research directions to the OC research community.

This study emphasizes the potential for hyper-spectral imaging in identifying and classifying blood stains in forensic science without physical sampling of crucial evidence. The chemical processes currently used for blood identification and classification can affect DNA analysis, making it necessary to explore novel approaches. Developing algorithms for blood detection is difficult due to the high dimensionality of hyper-spectral imaging and the scarcity of training sample data. This issue is addressed with a new hyperspectral blood detection data set. The proposed work emphasizes 8 dimensionality reduction methods as a preprocessing technique on hyperspectral data. Evaluation of these methods is done using state-of-the-art fast and compact 3D CNN and Hybrid CNN models. The experimental results and analyses demonstrate the challenges of blood detection in hyperspectral data and provide recommendations for future research in this area. Furthermore, this paper highlights the significance of Factor Analysis as a statistical tool for identifying underlying factors that explain patterns and relationships among observed variables.

Maintaining optimal tip clearance or tip gap is challenging in the Gas Turbine Engine (GTE). Meanwhile, the rotor blades should not rub the casing. When the capacitive sensor is used to measure the tip clearance in the form of a single peak signal for every blade pass, often the signal will be affected by stationary and non-stationary noises during engine running. This leads to distorted multiple peaks for every blade pass. In this work, the wavelet denoising technique removes the noise, and then the peak frequency in each blade pass is detected through a short-time Fourier transform (STFT). Finally, the cubic spline interpolation technique is employed to obtain the continuous time domain blade pass signal. This work uses the compressor stage of GTE data collected from the Gas Turbine Research Establishment (GTRE), DRDO, Bangalore. From the experimental analysis, this paper observes that the proposed methodology produces substantial results compared to the expected results.

In this paper we are presenting about a specific type of solar cell which has both organic and inorganic light harvesting layers made up of a halide-based material. Due to the limited sources of energies available, solar is the only abundant cheap promising source of renewable energy. Research is going on to find the highly efficient solar cell technologies. We have seen that mostly silicon has been the common semiconductor material in the solar cells which are expensive and sensitive towards the climatic changes. Perovskite solar cells solves these issues since they are cheap and easy to assemble, strong and flexible. We are going to implement the software which is used to stimulate light harvesting devices like OLED, OFET, Organic solar cells etc. So, we are also going to stimulate organic solar cell to compare their efficiencies with respect to the current-voltage characteristics.

In case of automobiles, safety is critical issue in order to reduce number of incidents in speed-restricted zones. According to recent polls, within the Accidents around school zones have grown in recent years. Due to their haste to reach to the desired location as soon as possible. As a result, limiting vehicle control speed has been a major concern. To thought about, our project seeks to provide a practical and compact solution. Also the development of an automatic vehicle speed system is simple. This must be implemented in jones like schools and hospitals to bring down the accident number. This speed control method is automated, and it is built on the Arduino based microcontroller board. The prescribed ordinance was incorporated into the transmitter unit that transmits the signals, and it was taken by the receiver which is located in the vehicle using a wireless communication technology Zigbee, and thus vehicle speed was controlled automatically by the received input massage of the receiver, with the assistance of devices like speed encoder. Accidents decreased at a faster pace when this method was installed, and some drivers complained less. The primary goal of this approach is to reduce accidents. We discovered the significant accidents i.e., 80 percentage by analysing some of the papers

Chest X-Rays (CXRs) are the most performed radiological procedure, accounting for roughly one-third of all radiological procedures. These images are used to study various structures such as the heart and lungs to diagnose diseases like lung cancer, tuberculosis, and pneumonia. Anatomical structure segmentation in chest X-rays is a critical component of computer-aided diagnostic systems. The measurements of irregular shape and size and total lung area can provide insight into early signs of life-threatening conditions such as cardiomegaly and emphysema. Lung segmentation is a challenge due to variance caused by age, gender, or health status; it becomes even more difficult when external objects like cardiac pacemakers, surgical clips, or sternal wire are present. As a result, accurate lung field segmentation is regarded as an important task in medical image analysis. A comparison of the efficacy of two deep-learning algorithms to detect lung-related pathologies via an investigation into the size of the lungs is enumerated herein. Utilizing X-ray images and the accompanying masks, Deep Learning Models were employed to predict the lung masks respective to the X-Ray Images with an exceptional level of accuracy achieved by one of the Deep Learning models at a 99.64%, determining the lung condition if it is normal or abnormal by calculating the sizes of the lung mask.

Intelligent Transportation Systems (ITS) play an increasingly significant role in our life, where safe and effective vehicular networks supported by sixth-generation (6G) communication technologies are the essence of ITS. Vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) communications need to be studied to implement ITS in a secure, robust, and efficient manner, allowing massive connectivity in vehicular communications networks. Besides, with the rapid growth of different types of autonomous vehicles, it becomes challenging to facilitate the heterogeneous requirements of ITS. To meet the above needs, intelligent reflecting surfaces (IRS) are introduced to vehicular communications and ITS, containing the reflecting elements that can intelligently configure incident signals from and to vehicles. As a novel vehicular communication paradigm at its infancy, it is key to understand the latest research efforts on applying IRS to 6G ITS as well as the fundamental differences with other existing alternatives and the new challenges brought by implementing IRS in 6G ITS. In this paper, we provide a big picture of deep learning enabled IRS for 6G ITS and appraise most of the important literature in this field. By appraising and summarizing the existing literature, we also point out the challenges and worthwhile research directions related to IRS aided 6G ITS.

As global competitiveness in the semiconductor sector intensifies, companies must continue to improve manufacturing techniques and productivity in order to sustain competitive advantages. In this research paper, we have used machine learning (ML) techniques on computational data collected from the sensors in the manufacturing unit to predict the wafer failure in the manufacturing of the semiconductors and then lower the equipment failure by enabling predictive maintenance and thereby increasing productivity. Training time has been greatly reduced through the proposed feature selection process with maintaining high accuracy. Logistic Regression, Random Forest Classifier, Support Vector Machine, Decision Tree Classifier, Extreme Gradient Boost, and Neural Networks are some of the model-building techniques that are performed in this work. Numerous case studies were undertaken to examine accuracy and precision. Random Forest Classifier surpassed all the other models with an accuracy of over 93.62%. Numerical results also show that the ML techniques can be implemented to predict wafer failure, perform predictive maintenance and increase the productivity of manufacturing the semiconductors.

Seismic interpretation is essential to obtain infor-mation about the geological layers from seismic data. Manual interpretation, however, necessitates additional pre-processing stages and requires more time and effort. In recent years, Deep Learning (DL) has been applied in the geophysical domain to solve various problems such as denoising, inversion, fault estimation, horizon estimation, etc. In this paper, we propose an Attention-based Deep Convolutional Neural Network (ACNN) for seismic lithology prediction. We used Continuous Wavelet Transform (CWT) to obtain the time-frequency spectrum of seismic data which is further used to train the network. The attention module is used to scale the features from the convolutional layers thus prioritizing the prominent features in the data. We validated the results on blind wells and observed that the proposed method had shown improved accuracy when compared to the existing basic CNN.

In seismic data processing, denoising and reconstruction are the two steps for identification of resources in the earth subsurface layers. The seismic data quality is affected by random noise and interference during acquisition. Further, the noisy data may be incomplete with missing traces. In this work, we propose a method for incomplete seismic data denoising and reconstruction based on double sparsity dictionary learning (DSDL) with structure oriented filtering (SOF). The main function of the DSDL step is denoising and SOF is used for residual noise attenuation and filling the missing data points. The proposed method is tested on 2-D synthetic and field datasets. The test results show that the DSDL-SOF method has better noise attenuation and reconstruction in terms of signal-to-noise ratio and mean squared error as compared to existing methods.

A method to capture three-dimensional (3D) objects image data under extremely low light level conditions, also known as Photon Counting Imaging (PCI), was reported. It is demonstrated that by combining a PCI system with computational integral imaging algorithms, a 3D scene reconstruction and recognition is possible. The resulting reconstructed 3D images often look degraded (due to the limited number of photons detected in a scene) and they, therefore, require the application of superior image restoration techniques to improve object recognition. Recently, Deep Learning (DL) frameworks have been shown to perform well when used for denoising processes. In this paper, for the first time, a fully unsupervised network (i.e., U-Net) is proposed to denoise the photon counted 3D sectional images. In conjunction with classical U-Net architecture, a skip block is used to extract meaningful patterns from the photons counted 3D images. The encoder and decoder blocks in the U-Net are connected with skip blocks in a symmetric manner. It is demonstrated that the proposed DL network performs better, in terms of peak signal-to-noise ratio, in comparison with the classical TV denoising algorithm.

A simplified (i.e., single shot) method is demonstrated to generate a Fourier hologram from multiple two-dimensional (2D) perspective images (PIs) under low light level imaging conditions. It was shown that the orthographic projection images (OPIs) can be synthesized using PIs and then, following incorporation of corresponding phase values, a digital hologram can be generated. In this work, a fast dictionary learning (DL) technique, known as Sequential Generalised K-means (SGK) algorithm, is used to perform Integral Fourier hologram reconstruction from fewer samples. The SGK method transforms the generated Fourier hologram into its sparse form, which represented it with a linear combination of some basis functions, also known as atoms. These atoms are arranged in the form of a matrix called a dictionary. In this work, the dictionary is updated using an arithmetic average method while the Orthogonal Matching Pursuit algorithm is opted to update the sparse coefficients. It is shown that the proposed DL method provides good hologram quality, (in terms of peak signal-to-noise ratio) even for cases of ~ 90% sparsity.

Intelligent Reflecting Surfaces (IRS) have been recognized as presenting a highly energy-efficient and optimal solution for future fast-growing 6G communication systems by reflecting the incident signal towards the receiver. The large number of Internet of Things (IoT) devices are distributed randomly in order to serve users while providing a high data rate, seamless data transfer, and Quality of Service (QoS). The major challenge in satisfying the above requirements is the energy consumed by IoT network. Hence, in this paper, we examine the energy-efficiency (EE) of a hybrid relay-IRS-aided wireless IoT network for 6G communications. In our analysis, we study the EE performance of IRS-aided and DF relay-aided IoT networks separately, as well as a hybrid relay-IRS-aided IoT network. Our numerical results showed that the EE of the hybrid relay-IRS-aided system has better performance than both the conventional relay and the IRS-aided IoT network. Furthermore, we realized that the multiple IRS blocks can beat the relay in a high SNR regime, which results in lower hardware costs and reduced power consumption.

To accelerate future intelligent wireless systems, we designed an energy-efficient Massive multiple-input-multiple-output (MIMO)- non-orthogonal multiple access (NOMA) aided internet of things (IoT) network in this paper to support the massive number of distributed users and IoT devices with seamless data transfer and maintain connectivity between them. Massive MIMO has been identified as a suitable technology to implement the energy efficient IoT network in beyond 5G (B5G) communications due to its distinct characteristics with large number of antennas. However, to provide fast data transfer and maintain hyper connectivity between the IoT devices in B5G communications will bring the challenge of energy deficiency. Hence, we considered a massive MIMO–NOMA aided IoT network considering imperfect channel state information and practical power consumption at the transmitter. The far users of the base stations are selected to investigate the power consumption and quality of service. Then, calculate the power consumption which is non-convex function and non-deterministic polynomial problem. To solve the above problem, fractional programming properties are applied which converted polynomial problem into the difference of convex function. And then we employed the successive convex approximation technique to represent the non-convex to convex function. Effective iterative based branch and the reduced bound process are utilized to solve the problem. Numerical results observe that our implemented approach surpasses previous standard algorithms on the basis of convergence, energy-efficiency and user fairness.

Now-a-days the concept and the use of Internet Of Things is gaining huge popularity with increase of smart cities. To increase the productivity and reliability of urban infrastructure consistent development is being made in the field of IoT. The population in the smart cities is huge and most of the people living in these smart cities own their vehicle. Due to the limited parking facilities problems such as traffic congestion is being continued in these smart cities. Due to this people waste their time in finding the parking slots. Also while parking the vehicle in multi complex areas people will be charged to park their vehicle. During their exit they should pay the amount charged for parking their vehicle and here with the use of physical money the payment process gets delayed and hence it leads to the traffic congestion. In this paper, an IoT based smart parking system with E-ticketing was proposed. Here, In this parking system we are using Arduino UNO as the processing unit and RFID cards to identify each vehicle individually and deduct the charge for the parking before they enter into parking area. Only if there is sufficient amount in the account of that particular vehicle owner, it will be deducted and a message will be sent to their mobile phone and the gate will open to park their vehicle. Also the slots that are available for parking will be shown on the display so that the user can directly head towards that slot without wasting much time. By this we can minimize the time that is being wasted by the user in finding a vacant parking slot to park the vehicle.

With the emergence of new viral infections and the rapid spread of chronic diseases in recent years, the demand for integrated short-range wireless technologies is becoming a major bottleneck. Implementation of advanced medical telemonitoring and telecare systems for on-body sensors needs frequent recharging or battery replacement. This paper discusses a priority-based resource allocation scheme and smart channel assignment in a wireless body area network capable of energy harvesting. We investigate our transmission scheme in regular communication, where the access point transmits energy and command while the sensor simultaneously sends the information to the access point. A priority scheduling nonpreemptive algorithm to keep the process running for all the users to achieve the maximum reliability of access by the decision-maker or hub during critical situations of users has been proposed. During an emergency or critical situation, the process does not stop until the decision-maker or the hub takes a final decision. The objective of the proposed scheme is to get all the user processes executed with minimum average waiting time and no starvation. By allocating a higher priority to emergency and on data traffic signals such as critical and high-level signals, the proposed transmission scheme avoids inconsistent collisions. The results demonstrate that the proposed scheme significantly improves the quality of the network service in terms of data transmission for higher priority users.

The internet has facilitated a wide range of equipment and gadgets, making it a significant component of our lives. We employ Internet of Things (IoT) technologies to remotely monitor, control, and operate these devices in our daily lives even from far distances. Smart health applications became a rapidly growing sector, especially in the past few years. And hence such types of technology which are both easy to use and understand are in high demand. For example, in individuals with heart disease, body temperature (BT), heart rate (HR) and respiration rate (RR) are all vital indicators that must be monitored on a regular basis. In our study, a Wi-Fi module-based application that may operate as a continuous monitor is built. HR, BT, and RR parameters for heart and lung patients that need to be monitored on a regular basis are achieved with this monitor. There are many problems as such which can be addressed and IoT makes it possible. So in this paper, we addressed some of the problems such as monitoring pulse rate, temperature, and respiration and notify the contacts and alert surroundings with one single click.

From the past few years due to the activities done by humans and industrialization the air pollution has become so dangerous in many countries especially in India as of the developing country. The main concern of people's health is the particulate matter which is also known as PM 2.5 which is significant between the pollutant index. The particulate matter(PM) diameter is equal to or less than 2.5m is one of the major health issues when seen with all other air pollutants. The PM2.5 is one of those tiny particles which reduces one's lucency and also the air becomes smoky when the elevation happens. In the urban areas, the PM2.5 hang on many factors,corresponding to the concentration on other pollutants and also on meteorology. To show up these factors there are some techniques which were introduced in some other air quality researches as well. These used approaches such as the neural network and Long Short-Term Memory (LSTM), to check every air pollutant level situated on traffic variables obtained and weather conditions. In our experiments, the results of our proposed method hybrid CNN-LSTM gives the most accurate prediction when compared to all other methods present and also performs a cut above than the guessing performance.

We developed a deep stacked undercomplete autoencoder (i.e., supervised) network to denoise the noisy 3D sectional images. Results demonstrate the feasibility of our proposed model in terms of peak-signal-to-noise ratio.

This paper proposes an inter-vehicular communication model using Bluetooth for information transfer.V2V technology proposes a variety of solutions for passenger safety. According to research,1.35 million people die each year due to road crashes [1]. Present vehicle system uses radars, cameras to detect collisions and gives potential warnings to the drivers, leaving the decision to the driver. Our main motivation is to avoid crash collisions, reduce fatal accidents, traffic congestion. The proposed idea enhances the current systems by upgrading from alerting the drivers to communication between vehicles, helps the vehicle to take control over the situation and control its state. In this paper, the idea is demonstrated using two prototype models designed with an Ultrasonic sensor to detect nearby vehicles and objects, Bluetooth module which uses Bluetooth for real-time data transfer of mobility parameters such as speed, distance, etc. providing 360-degree awareness to the vehicle. Bluetooth can be replaced with any highly advanced wireless technologies according to requirements. Designed prototype models are tested under 3 common real-life scenarios such as slowdown, abrupt stop, overtaking. The average reaction brake time for a driver is 2.3 sec. Replacing the driver with the vehicle taking control over the situation when required helps us in reducing this reaction time which is a major cause of accidents, reduces traffic congestion.

Agriculture has consistently been our most noteworthy part of human endurance, however in later years an increment in the population has likewise expanded the mechanical progress improvement and bringing about a deficiency of a high number of laborers in the agricultural sector. The aim/objective of this report is to propose a Voice Automation Agricultural System which assists farmers to monitor and gives the live feed (Soil Moisture/Temperature) to his/her mobile and users can use voice commands to execute any preferred actions (Watering using Sprinklers) accordingly. The IoT based Voice Automation Agricultural System being proposed via this report is a combination of two NodeMCUs with DHT11(Temperature & Humidity), FC28(Soil Moisture) Sensors, and an inbuilt ESP8266(Wifi module) which helps in producing live data feed that can be obtained online from Blynk Application and performing actions using IFTTT (If This Then That). IFTTT is an automation platform that uses applets to automate our tasks. After getting feed to the user's mobile, the user can decide to choose an action like watering the plants (via sprinklers).

Massive MIMO with large number of antennas at the BS has the ability to serve many number of users with large data rate requirements. Energy Efficiency (EE) and spectral efficiency (SE) has been considered as the major performance measures for the advanced wireless communication systems. In this paper, we analysed the performance of EE while considering the practical power consumption at the base station (BS). The results suggest that the EE can be enhanced by finding the optimal power consumption at BS and antennas in massive MIMO system.

Low-density multiple-input multiple-output code (LDMC) can reduce the complexity of tree-search detection in MIMO systems. In this paper, we present a new modified progressive edge-growth (PEG) algorithm to construct large girth LDMCs, which are referred to as PEG-LDMCs. We analyze the complexity of the LDMC constrained sphere decoding (SD) and show that the LDMC constrained SD detection can be used for high reliability of the data transmission in MIMO systems. Furthermore, we propose two new efficient iterative decoding algorithms for LDMCs, which are high speed serial decoding and fast convergence shuffled decoding. Finally, we compare the bit error rate (BER) performance of PEG-LDMCs to that of the existing LDMCs. The simulation results show that the PEG-LDMCs can achieve better BER performance than that of the existing LDMCs.

In this paper, the energy efficiency maximization problem is tackled in a multiple-input multiple-output (MIMO)-non-orthogonal multiple access (NOMA) heterogeneous cellular network (HCN) with Internet of Things (IoT) for wireless communication systems. A fractional non-convex optimization problem is formulated to maximizes the energy efficiency (EE) subject to the transmit power constraints and the minimum rate requirement for the cell edge (CE) users present in both macro base station (MBS) and pico base station (PBS). The above problem is hard to solve due to its nonlinear fractional objective function. Fractional programming properties is firstly employed to convert the non-convex problem into its parametric form. In addition, an efficient iterative algorithm is proposed established on the branch and reduced bound (BRB) approach that achieves convergence to the above problem, mitigates the inter tier interference and also improves the fairness between the users. Comprehensive numerical results emphasize that the proposed scheme achieves higher energy efficiency as compared with the existing NOMA scheme and the conventional orthogonal multiple access (OMA) scheme.

In this paper, we present a sub-optimal antenna selection technique to enhance the channel capacity in the case of High SNR MIMO correlated downlink channel. Most related work has thus far considered uncorrelated MIMO channel with antenna selection technique and used a larger radio frequency (RF) module. Thus, we propose the MIMO correlated downlink MIMO channel with sub-optimal antenna selection method to employ a smaller number of RF module. In this paper, we first design and develop the Toeplitz channel correlation matrices which reflect the correlations between the transmitter antennas, then apply a sub-optimal transmit antenna selection technique to improve the channel capacity. Mote Carlo simulation results show that the channel capacity significantly enhance considering equal power transmission.

In this paper, we present a robust beamforming design to examine the weighted sum-rate maximization (WSRM) problem in a multicell massive MIMO downlink system for 5G communications. This work assume imperfect channel state information (CSI) by adding uncertainties to channel matrices with worst-case models i.e. singular value uncertainty model (SVUM) and ellipsoidal uncertainty model (EUM). In SVUM, WSRM problem is formulated subject to the transmit power constraints. While, the problem is devised in EUM by alternatively considering its dual power minimization problem subject to the worst-case signal-to-interference-plus-noise ratio (SINR) constraints for all mobile stations. The designed problem for both SVUM and EUM are known as non-deterministic polynomial (NP) problem which is difficult to solve. We propose an iterative algorithm established on majorization minimization (MM) technique that solves and achieves convergence to stationary point of these two problems. In EUM, the convergence point is obtained after converting the infinite number of SINR constraints into linear matrix inequalities (LMI) by employing S-Procedure. Extensive numerical results are provided to show that the proposed iterative algorithm significantly increases performance in terms of sum-rate and also attains faster convergence as compared with the conventional polynomial time algorithm.

In this paper, we characterise the joint interference alignment (IA) and power allocation strategies for a K-user multicell multiple-input multiple-output (MIMO) Gaussian interference channel. We consider a MIMO interference channel with blind-IA through staggered antenna switching on the receiver. We explore the power allocation and feasibility condition for cooperative cell-edge (CE) mobile users (MUs) by assuming that the channel state information is unknown. The new insight behind the transmission strategy of the proposed scheme is premeditated (randomly generated transmission strategy) and partial cooperative CE MUs, where the transmitter is equipped with a conventional antenna, the receiver is equipped with a reconfigurable multimode antenna (staggered antenna switching pattern), and the receiver switches between preset T modes. Our proposed scheme assists and aligns the desired signals and interference signals to cancel the common interference signals because the received signal must have a corresponding independent signal subspace. The capacity for a K-user multicell MIMO Gaussian interference channel with reconfigurable multimode antennas is completely characterised. Furthermore, we show that the proposed K-user multicell MIMO scheduling and K-user L-cell CEUs partial cooperation algorithms elaborate the generalisation of K-user IA and power allocation strategies. The numerical results demonstrate that the proposed intercell interference scheme with partial-cooperative CE MUs achieves better capacity and signal-to-interference plus noise ratio (SINR) performance compared to noncooperative CE MUs and without intercell interference schemes.

In this paper, we examine a joint user pairing and dynamic power allocation (JUPDPA) design to maximize the energy efficiency (EE) in the multicell multiple-input multiple-output (MIMO) - non-orthogonal multiple access (NOMA) downlink system. A novel JUPDPA scheme is proposed based on median and the euclidean norm of the channel vectors. In addition, base station dynamically allocates the transmission power to the paired users. Comprehensive numerical results illustrate that the proposed scheme attains higher energy efficiency as compared with the existing NOMA schemes and the conventional orthogonal multiple access (OMA) scheme.

In this paper, we examine the robust beamforming design to tackle the energy efficiency (EE) maximization problem in a 5G massive multiple-input multiple-output (MIMO)-non-orthogonal multiple access (NOMA) downlink system with imperfect channel state information (CSI) at the base station. A novel joint user pairing and dynamic power allocation (JUPDPA) algorithm is proposed to minimize the inter user interference and also to enhance the fairness between the users. This work assumes imperfect CSI by adding uncertainties to channel matrices with worst-case model, i.e., ellipsoidal uncertainty model (EUM). A fractional non-convex optimization problem is formulated to maximize the EE subject to the transmit power constraints and the minimum rate requirement for the cell edge user. The designed problem is difficult to solve due to its nonlinear fractional objective function. We firstly employ the properties of fractional programming to transform the non-convex problem into its equivalent parametric form. Then, an efficient iterative algorithm is proposed established on the constrained concave-convex procedure (CCCP) that solves and achieves convergence to a stationary point of the above problem. Finally, Dinkelbach’s algorithm is employed to determine the maximum energy efficiency. Comprehensive numerical results illustrate that the proposed scheme attains higher worst-case energy efficiency as compared with the existing NOMA schemes and the conventional orthogonal multiple access (OMA) scheme.

In this paper, we study the emergence of topological interference alignment and the characterizing features of a multi-user broadcast interference relay channel. We propose an alternative transmission strategy named the relay space-time interference alignment (R-STIA) technique, in which a K-user multiple-input-multiple-output (MIMO) interference channel has massive antennas at the transmitter and relay. Severe interference from unknown transmitters affects the downlink relay network channel and degrades the system performance. An additional (unintended) receiver is introduced in the proposed R-STIA technique to overcome the above problem, since it has the ability to decode the desired signals for the intended receiver by considering cooperation between the receivers. The additional receiver also helps in recovering and reconstructing the interference signals with limited channel state information at the relay (CSIR). The Alamouti space-time transmission technique and minimum mean square error (MMSE) linear precoder are also used in the proposed scheme to detect the presence of interference signals. Numerical results show that the proposed R-STIA technique achieves a better performance in terms of the bit error rate (BER) and sum-rate compared to the existing broadcast channel schemes.

In this paper, we present a robust beamforming design to tackle the weighted sum-rate maximization (WSRM) problem in a multicell multiple-input multiple-output (MIMO) – non-orthogonal multiple access (NOMA) downlink system for 5G wireless communications. This work consider the imperfect channel state information (CSI) at the base station (BS) by adding uncertainties to channel estimation matrices as the worst-case model i.e., singular value uncertainty model (SVUM). With this observation, the WSRM problem is formulated subject to the transmit power constraints at the BS. The objective problem is known as non-deterministic polynomial (NP) problem which is difficult to solve. We propose an robust beamforming design which establishes on majorization minimization (MM) technique to find the optimal transmit beamforming matrix, as well as efficiently solve the objective problem. In addition, we also propose a joint user clustering and power allocation (JUCPA) algorithm in which the best user pair is selected as a cluster to attain a higher sum-rate. Extensive numerical results are provided to show that the proposed robust beamforming design together with the proposed JUCPA algorithm significantly increases the performance in term of sum-rate as compared with the existing NOMA schemes and the conventional orthogonal multiple access (OMA) scheme.

The heterogeneous cellular network (HCN) is most significant as a key technology for future fifth-generation (5G) wireless networks. The energy efficient design of HCNs consist of different types of base stations (BSs), which has drawn significant attention to technologies for future 5G wireless networks. The cellular networks have faced a great deal of challenges to meet sharply rising demand for higher network capacity and higher data rates as well as far more power consumption which results in operating costs caused by the number of users accessing the cellular networks concurrently. BS is the main part of power consumption, so reducing energy consumption of the BS can obviously reduce the total energy consumption. Recently, the power consumption of the BSs has been attracted in cellular networks. In this chapter, we propose switching off/on systems for the efficient power consumption at the BSs in the cellular networks which introduce active/sleep modes in the macro BSs (MBSs) and femto BSs (FBSs). The active/sleep modes reduce the interference and power consumption as well as improve the energy efficiency of the cellular networks. Moreover, we derive the two-tier HCNs under different sleeping policies as well as formulate power consumption minimization for the MBSs and FBSs. An optimization problem is formulated to maximize the energy efficiency subject to throughput outage constraints as well as solved by the Karush-Kuhn- Tucker (KKT) conditions in terms of the femto tier BS density. Furthermore, the energy efficiency of cellular network is analyzed and modeled based on Markovian wireless channels.

In this paper reveals the linear interference networks, for both wired and wireless networks, with no channel state information at the transmitters (CSIT) which has been studied under the topological interference management (TIM) framework except for the knowledge of the connectivity graph. Different solutions are classified in this paper, for the 5 user topology network, interference avoidance approach and interference alignment approach are proposed. We investigate alignment schemes for a multicast traffic over time varying circularly symmetric fading channels. For 6 and 8 user topology interference network, we propose a multicast alignment approach. Where we aligned the user interferences by grouping the common interferences with distributed broadcast channel settings.

In this paper, we mainly focused on achieve the desired degree of freedom. Channel state information at the transmitter (CSIT) plays major role for K-user Multiple Input and Multiple Output (MIMO) M × N interference channels. Span Interference alignment method is used in order to align the interference. Each transmitter has two antennas which help to communicate with transmitter and relay node in an efficient way. In addition, by adding the relay nodes with the desired system model is helps to recover the Degree of Freedom (DoF) which is ideal for these models with global channel state information (CSI). Recently studies have shown that relay cannot improve the DoF of wireless interference networks but it helps to give some partial solutions to approach the interference alignment. The relay nodes which helps to steer directions of transmitter signals to expedite the interference alignment in order to achieve optimal DoF. A new performance method has been proposed in order to estimate the efficiency of an interference alignment.