Recently, the implementation of machine learning algorithms in the analysis of voice disorders has become crucial for providing non-invasive identification of voice disorders using only audio signals. However, some voice pathology detection systems still face challenges such as working with limited acoustic databases, achieving low accuracy, and relying on constant parameters like a single number of hidden nodes. Therefore, this paper presents a method for the identification of voice pathology based on a machine learning algorithm. The voice signals are collected from a voice pathology database called the Malaysian Voice Pathology Database (MVPD). This database is created recently. The features of voices are extracted by using the Mel-Frequency Cepstral Coefficient (MFCC). Furthermore, the proposed method uses the Fast-Learning Network (FLN) algorithm for the classification part. In the proposed method, the FLN algorithm uses a different number of hidden nodes, where it starts with 20 nodes and finishes with 200 nodes. The performance of the proposed method is assessed in terms of many performance metrics. The results show that the proposed FLN algorithm achieves the highest results at the hidden nodes of 110. The FLN algorithm obtains promising results in detecting voice pathology.
Waveform-based localization approaches are considered the most reliable for vehicles in complex heterogeneous environments. However, this paper addresses millimeter-wave (mm-wave) propagation for vehicle to infrastructure (V2I) scenario at 28 GHz and 73 GHz. Particularly, it analyses the key localization elements for waveform-based approach such as received power and angel of arrival. Additionally, this paper investigates the impact of vehicles mobility on the stability these key elements at a certain position. Ray tracing techniques have been used to model the propagation characteristics of the aforementioned bands of mm-wave at a semirealistic V2I communication scenario. The obtained results indicated a proportional correlation between the received power and the elevation angle of arrival at a given location, potentially enhancing localization performance. In the other side, the mobility speed of the vehicle showed a notable impact on the stability of the key localization elements at a certain location.
The increasing need for high data rates in vehicular networks necessitates the development of innovative technologies to improve spectral efficiency (SE) and guarantee the reliability of cellular connections. An innovative approach that uses an integration of Fully Generalized Spatial Modulation (FGSM) with hybrid beamforming (H-FGSM) to improve vehicular communication, exhibiting notable enhancements in SE. This paper explores the utilization of FGSM in conjunction with hybrid beamforming within vehicular communication systems, specifically emphasizing two predominant frameworks: millimeter-wave (mmWave) and IEEE 802.11ad. It employs a complete performance analysis to investigate the effects of H-FGSM on the aforementioned communication frameworks. It is worth mentioning that the mmWave framework has a superior SE of 7bps/Hz in comparison to IEEE 802.11ad, in addition to extended coherence time by 3 times and decreased frequency of adjustments to channel state information (CSI). These characteristics render mmWave very suitable for dynamic vehicular environments that require reliable communication routes. Consequently, the integration of FGSM with hybrid beamforming presents a potentially viable approach for addressing the requirements of advanced intelligent transportation systems in future vehicle networks, which leads to the foundation for future investigations aim at enhancing the efficiency of mmWave communication systems in dynamic vehicular situations.
The rise in high-bandwidth applications like IoT devices, smart appliances, and social networking has driven the development of 5G technology. This research focuses on enhancing the energy efficiency of Time and Wavelength Division Multiplexing Passive Optical Networks (TWDM-PON), which are widely used in PON systems. TWDM-PON is crucial for 5G fronthaul networks, which need to support a thousandfold increase in capacity, ten to a hundred times higher data rates, and reduced latency. To address these needs, the research proposes an algorithm for a flexible TWDM-PON- based mobile fronthaul (MFH) architecture. This system allows Optical Network Units (ONUs) to dynamically allocate bandwidth for efficient 5G fronthaul usage. The proposed network topology facilitates dynamic resource management based on demand, with ONUs positioned near Remote Radio Heads (RRHs) to manage wavelength allocation. Each ONU uses a laser suited to any available wavelength in the transmission link. The study shows that this architecture helps maintain average latency while significantly saving energy. The proposed algorithm optimizes wavelength allocation, achieving a 60% reduction in energy consumption per active wavelength while meeting various latency requirements without compromising 5G fronthaul performance.
The extinction of freshwater fish is mainly due to environmental destruction, which has led to an increase in fish breeding activities in the aquaculture sector. This study presents the implementation of a machine learning approach using the YOLOv5 model to facilitate the fish breeding process by classifying the sex of Mahseer (Kelah) and the gravid broodstock of Tilapia. Traditional methods of sex and broodstock identification are often time-consuming and prone to errors due to a lack of expertise and equipment. A collection of X-ray images of Tilapia fish was gathered and labeled to develop a dataset for training and evaluating the YOLOv5 model. Due to the high cost of obtaining original data for Mahseer, publicly available images from Google were used to build the dataset. The results demonstrated that the YOLOv5 model achieved a precision of 94% and 99.4% for classifying sex and broodstock, respectively, using a maximum dataset of 430 images and 40 images. This finding indicates that YOLOv5 can significantly improve the efficiency and accuracy of broodstock management in aquaculture, contributing to the growth of the aquaculture sector.
Reconfigurable Intelligent Surfaces (RIS) and Un-manned Aerial Vehicles (UAVs) have emerged as promising technologies for the 6th-Generation (6G) network. The integration of RIS with the UAV (RIS-UAV) can enhance ground communication by providing a 360° panoramic reflection. Existing RIS-UAV mainly considers passive elements which suffer from double path loss problems. This motivates the use of the hybrid RIS-UAV equipped with both active and passive RIS elements. This paper investigates the rate maximisation problem for the hybrid RIS-UAV by optimising the UAV's altitude, transmit power allocation at the base station and hybrid RIS, subject to the hardware power dissipation, transmission power, and flight power. The non-convex joint optimisation problem is addressed using a Genetic Algorithm (GA). The numerical results show that the joint-optimised hybrid RIS-UAV can achieve a rate twice the unoptimised hybrid RIS-UAV and 22 times higher than the conventional passive RIS-UAV.
The rise of cloud-based disaster recovery systems has reshaped disaster recovery practices. Integrated into the Hadoop framework with core components like Hadoop Distributed File System (HDFS), MapReduce, and Yet Another Resource Negotiator (YARN), these systems ensure data integrity and business continuity during crises while enhancing data reliability and operational efficiency, surpassing traditional recovery systems. This study presents a comprehensive comparative analysis of traditional and Hadoop-integrated cloud recovery systems, focusing on hardware architecture and performance. Architecturally, it highlights the stark differences between conventional models and Hadoop-based innovations. The results show that the Hadoop-backed recovery system exhibits remarkable improvement, reducing Recovery Time Objective (RTO) from over 18 seconds to under 7 seconds while preserving data integrity as per Recovery Point Objective (RPO) standards. This finding underscores the academic importance of evaluating cloud-based disaster recovery within the Hadoop context, offering insights into its transformative potential for disaster recovery strategies.
Amidst the rapidly evolving landscape of digital healthcare, integrating blockchain and Internet of Medical Things (IoMT) technologies creates new opportunities and challenges. This study compares the RBMCA (Robust Blockchain-based Model for Cloud Architecture) and traditional cloud-based healthcare systems. It demonstrates that the RBMCA model outperforms conventional systems in security metrics. The decentralized, blockchain-integrated architecture of the RBMCA model significantly improves response times, data integrity, and system uptime and effectively mitigates cybersecurity threats. These improvements are crucial in healthcare environments where data accuracy, system reliability, and security are paramount. Using advanced simulation tools and real-world cloud environments, this research highlights the potential of the RBMCA model to transform healthcare data management, providing a robust and scalable solution for challenges posed by the Internet of Medical Things devices and cloud-based infrastructures. The findings indicate that adopting the RBMCA model can significantly enhance healthcare systems’ security, reliability, and efficiency, improving patient outcomes and safer data management practices. Additionally, this research sets the stage for future developments by advocating for the refinement of the RBMCA model to address additional cybersecurity challenges, explore applications beyond healthcare, and integrate emerging technologies such as artificial intelligence and machine learning to enhance cybersecurity further and optimize healthcare delivery.
The paper focuses on deploying the YOLOv5 model on Jetson Nano using C++ and evaluating the mean average precision (mAP) index. It includes the deployment process and environment configuration, performance testbed, and assessing mAP to evaluate the model. Besides, we perform to track objects on Jetson Nano by integrating YOLOv5 in C++. The results show that the model achieves a fairly high level of accuracy while evaluated on the visual object classes (VOC) dataset. We achieve real-time image processing from 10 to 12 frames per second (FPS) compared with the traditional TensorRT (4 to 5 frames) on Jetson Nano with limited resources. Results will provide important information for using Jetson Nano in embedded applications that require image processing and object detection.
Fluid Antenna System (FAS), Non-Orthogonal Multiple Access (NOMA), and Unmanned Aerial Vehicle (UAV) have been explored as promising technologies for sixth-generation (6G) wireless networks. However, a single technology alone will not be sufficient to meet the stringent performance requirements of 6G. Thus, in this paper, we propose a NOMA-based UAV Relay-Aided FAS to address this challenge. Specifically, we jointly optimize the active port, power allocation, and UAV altitude to maximize the achievable sum-rate of Ground Users (GUs) while satisfying minimum rate requirements. To tackle this non-convex optimization problem, we decouple it into two sub-problems: a large-scale fading optimization problem and a small-scale fading optimization problem. We demonstrate that the large-scale fading problem can be solved using a one-dimensional search method, while the small-scale fading problem can be addressed with closed-form expressions. Our simulation results show that FAS consistently outperforms Traditional Antenna System (TAS). Moreover, the combination of FAS and NOMA greatly benefits from the joint optimization of both small-scale and large-scale fadings in UAV relay-aided communications, resulting in substantial performance improvements over the existing TAS and Orthogonal Multiple Access (OMA) combination.
Inter-cell interference (ICI) and limited sky coverage are major challenges to the development of cellular-connected unmanned aerial vehicle (UAV). To address these challenges, we consider the applications of reconfigurable intelligent surface (RIS). RIS is a promising solution capable of reconfiguring the phase of a reflected radio signal so as to enhance the desired signal strength or mitigate the ICI. In this paper, we propose a coordinated RIS-aided cellular-connected UAV scheme to provide ubiquitous coverage in the sky and mitigate the ICI of aerial users (AUs) while improving the performance of existing terrestrial users (TUs). We further develop an optimization algorithm to obtain the optimal coordinated beamforming and phase shift that maximize the weighted sum-rate (WSR) of the TUs subject to the AUs' rate requirement. Our simulation results show that the proposed coordinated RIS-aided cellular-connected UAV scheme is a promising solution where the WSR of the TUs increases with respect to the number of reflecting elements while satisfying the AUs' rate requirement. Compared to the baseline scheme without RIS, the proposed scheme also provides a WSR gain up to 48%.
Ultra-Dense Networks (UDNs), which offer previously unheard-of connectivity and capacity, have emerged in Beyond 5G (B5G) systems as a result of the rapid evolution of wireless communication technology. On the other hand, substantial interference is introduced by the dense base station deployment in UDNs, creating severe performance problems. The Successive Interference Cancellation (SIC) method has drawn interest as a potential solution to improve system performance to overcome this problem. The possibility for SIC implementation in 5G UDNs is examined in this study. The main contributions of this work are as follows: First, to validate the theoretical analysis of Successive Interference Cancellation (SIC) in a multi-tier 5G Ultra-Dense Network (UDN), an enhanced SIC technique is developed. This method employs a stochastic geometry model that accurately simulates real-world conditions by accounting for the random distribution of users and base stations within the UDN. As a result, it improves the precision of calculations for Success Probability and Ergodic Capacity. Second, numerical simulations are run to show how the suggested SIC strategy outperforms conventional approaches in terms of effectiveness. The theoretical conclusions are validated by comparing the simulation results with Zero Forcing (ZF) and No-SIC to assess the concept's validity and application. The knowledge and use of SIC in UDNs are being furthered by this research, opening the door to enhanced network capacity and performance in next wireless communication systems.
The rapid growth of the Internet of Things (IoT), expected to exceed 29 billion devices by 2027, presents a significant security challenge. Device fingerprinting, which identifies devices through unique network traffic patterns, is a valuable security tool but can be exploited by attackers for undetectable reconnaissance. This paper explores the feasibility of passive traffic fingerprinting attacks from an attacker's perspective, focusing on a small-scale testbed with four devices from an undisclosed single vendor. Using minimal resources and open-source tools, traffic patterns were analyzed, confirming unique device behaviors. A Random Forest (RF) classifier was developed, demonstrating high precision with instances where 100 % classification accuracy can be achieved in a specific experimental scenario using just two features: Source Port and Destination Port. Key aspects of the model included feature selection, hyperparameter tuning, and a tree depth optimization of 20. Notably, forward feature selection proved more effective than Principal Component Analysis (PCA). These preliminary results underscore the vulnerabilities of single-vendor IoT ecosystems and highlight the simplicity and replicability of this low-cost attack methodology. The ease with which attackers can implement this approach underscores the urgent need for robust defenses as IoT devices proliferate across various sectors.
This closed loop rotate modulation goal is to improve the multicarrier systems performance in slow fading channels, by utilizing complex channel state information feedback. The rotate modulation process uses the precoding principles applied to multicarrier systems with the Inverse Discrete Fourier Transform (IDFT) and Discrete Fourier Transform (DFT). If the deep fade channel is an input for the rotate modulation process that is inversely proportional to the channel response of each subcarrier, it will produce a signal with a large power. To limit the peak power of the system output, we propose the level limiter algorithm for feedback channels to anticipate deep fade conditions. The simulation results show the addition of a limiter can reduce the peak of the output power of the multicarrier modulator, which means the peak-to-average power can be reduced. However, it has a consequence that the bigger the limiter will result in an increased probability error, so we propose the optimal limiter level is 0.01 volt. The probability of error in low-level modulation like Binary Phase Keying shows good results even though the limiter's level process is added. At high-level modulation like 16-quadrature Amplitude Modulation, the greater level limiter causes the probability of error to worsen.
This paper explores a rule based mediation approach to optimize the balance between coverage and capacity Key Performance Indicators (KPIs) in Self Organizing Networks (SONs). Traditional methods often provide mitigation solutions to resolve conflicts between these KPIs. In contrast, this study introduces a mediation optimization technique that dynamically adjusts electrical antenna tilt in response to changing user densities. Through the simulations, the proposed approach demonstrates significant improvements in network performance, effectively reducing the coverage and capacity losses typically observed when one KPI is prioritized over the other. This dynamic adjustment method offers a more balanced solution for optimizing SON performance.
Digital predistortion (DPD) is the most widely used technique for linearizing power amplifiers. However, the increasing demand for high-bandwidth applications poses a challenge for traditional DPD as it usually entails oversampling a signal by up to five times its bandwidth. To address this, hybrid DPD techniques, which split the predistortion algorithm between the digital and analog domains, offer a promising solution. This study evaluates the robustness of a hybrid digital/analog predistorter in terms of its error vector magnitude (EVM) and adjacent channel leakage ratio (ACLR) performances. Specifically, it examines the impacts of a limited gain resolution in the analog predistortion implementation and imperfect delay matching between the input and control signals of the analog predistortion function. These effects were first considered independently of each other. Later, the performance of the linearized amplifier was assessed in presence of both imperfections. The findings show that the ACLR is sensitive to an imperfect implementation of the analog predistortion function. However, the EVM performance is much more resilient to such imperfections.
In the age of technology, the use of technological devices has become an indispensable part of daily life. It combines technological devices with artificial intelligence models to exploit their potential. Combining traffic surveillance cameras with artificial intelligence is receiving attention and research from countries worldwide. To continue this research, we use surveillance cameras combined with artificial intelligence to build a system to identify and authenticate violating vehicles. However, applying the system to different environmental conditions is always challenging. This system uses deep learning models and image processing techniques that require fast speed and high reliability. The proposed method for this system is to use the Retina-LPD and GFP-GAN models to improve the license plate recognition model, add license plate alignment methods, and build an additional image quality recovery model. height from the license plate is aligned to help increase model reliability. The system is deployed on the ICOMM server, and evaluated on video from traffic surveillance cameras from Hanoi city surveillance cameras. The system has achieved better results than the old system up to 99.74%, helping to increase system reliability. Although the system has made a lot of progress in vehicle authentication, many challenges and issues must be resolved to ensure effective performance and response to all environmental conditions.
This work provides an investigation for the design and analysis of frequency-switching planar reflecting surface unit cells designed in X-band and Ku-band frequency ranges. Slot and gap-embedded configurations have been proposed to obtain optimum reflection phase performance from the unit cells. The proposed design also helps to minimize the mutual coupling in the array environment. A PIN diode has been attached to the patch surface of each unit cell for frequency switching. Various design parameters have been investigated in order to study their effect on the performance of proposed unit cell designs. Based on the frequency variation using PIN diodes frequency tunability of 0.43 GHz and 0.35 GHz have been demonstrated for X-band and Ku-band frequency ranges.
One of the most significant advancements in urban infrastructure management is the integration of smart sensors into manhole covers. These sensors play a crucial role in providing real-time information on the condition of manhole covers and offering insights into the state of utility and sewer systems. However, developing a reliable and efficient networking architecture poses a significant challenge to the successful deployment of these sensors. This project aims to address ‘Connectivity for Manhole Cover Sensors,’ a critical aspect of urban infrastructure management that offers a revolutionary approach with far-reaching implications. The proposed system enables intelligent management of urban manhole covers, reducing the workload for managers and maintenance staff while contributing to the development of smart cities. Additionally, it helps mitigate safety risks to some extent.
Dual connectivity (DC) can increase network capacity by allowing a user to simultaneously connect to more than one Radio Access Technology (RAT) at the same time. Although it offers higher network capacity, power management is a constraint in DC enabled heterogeneous networks (HetNet). This work aims to find the best trade-off between network capacity and power efficiency in DC deployed Fifth Generation New Radio (5G-NR) and Long Term Evolution (LTE) HetNet. Past works have not conducted a simultaneous optimization of power saving and capacity enhancement for DC in a Non Standalone(NSA) setup. A novel base station pairing technique is used to determine Next Generation Node B (gNB) sleep patterns. Two algorithms, the Super Greedy algorithm and the Integer Linear Programming (ILP) - Greedy algorithm are proposed to find the optimal way to preserve network power and enhance the user throughput. The ILP-Greedy uses ILP optimization for power minimizing through cell sleeping and a greedy algorithm for user association. Based on comparison with past baseline works, the ILP-Greedy algorithm is found to be superior for DC implementation in a 4G/5G NSA HetNet, saving 7.02 % of network power and still providing comparable network capacity improvement.
