Journal of Intelligent Systems and Internet of Things

To provide better Quality of Service (QoS), which is expected in contemporary 6G wireless networks. We project a MDMA scheme to fulfill UE-specific QoS needs with the aid of multi-dimensional radio resource cost. This method can be successfully called Multi-Dimensional Radio Resource Allocation (MDRA). Specifically, the planned scheme incorporates two novel aspects: for each UE, the choice of user-specific non-orthogonal multiple approach mode whose cost is determined by UE-specific non-orthogonal interference cancellation; and allocating multiple dimensional radio resources for co-existing UEs in dynamic network environment. To reduce the costs of using UE-specific resources, the BS mounts UEs with diverse multi-domain resources. Specific to each UE coalition by taking into consideration restrictions such as the availability of resources, the perceived quality of those resources, and the possibility for use. Every UE that is a part of the coalition has access to the radio resources that it needs, which helps to lower the costs of use while preventing resource-sharing disputes with the other nodes in the coalition. Furthermore, the allocation of multi-dimensional radio resources among co-existing user equipment makes it possible to solve the issue of maximizing the sum of cost-conscious utility. This is done to fulfil UE-specific quality of service needs as well as varied resource circumstances on the user equipment side. The gradient convexity with low complexity approximation and the Lagrange double decomposition approach are used in the development of the solution to this NP-hard issue. The efficacy of the system that we have presented is shown via the use of numerical simulations and a comparison of its performance with that of other methods.

Recently, healthcare systems integrate the power of deep learning (DL) models with the connectivity and data processing capabilities of the Internet of Things (IoT) to enhance the early recognition and diagnosis of disease. Oral cancer diagnosis comprises the detection of cancerous or pre-cancerous abrasions in the oral cavity. Timely identification is essential for successful treatment and enhanced prognosis. Here is an overview of the key aspects of oral cancer diagnosis. One potential benefit of utilizing DL for oral cancer detection is that it analyses huge counts of data fast and accurately, and it could not need clear programming of the rules for recognizing abnormalities. This can create the procedure of detecting oral cancer more effective and efficient. Thus, the study presents an Enhanced Jaya Optimization Algorithm with Deep Learning Based Oral Cancer Classification (EJOADL-OCC) method. The presented EJOADL-OCC method aims to classify and detect the existence of oral cancer accurately and effectively. To accomplish this, the presented EJOADL-OCC method initially exploits median filtering for the noise elimination. Next, the feature vector generation process is performed by the residual network (ResNetv2) model with EJOA as a hyperparameter optimizer. For accurate classification of oral cancer, a continuously restricted Boltzmann machine with a deep belief network (CRBM-DBN) model. The simulated validation of the EJOADL-OCC algorithm is tested by the series of simulations and the outcome demonstrates its supremacy over present DL approaches.

Unmanned aerial vehicle (UAV) network offers a variety of applications in public safety, disaster management, advertising and broadcasting, overload situation, etc. Due to the dynamic characteristics of MU, it is challenging to provide robust transmission services to mobile users (MU). Resource allocation (RA), including sub-channel, serving user, and transmit power, is a crucial problem; also, it is critical to enhance the coverage and energy efficiency of UAV-enabled communication protocol. Furthermore, system resources are limited (for example, spectrum, and transmission power) and UAV transmission coverage and on-board energy are limited. In order to meet the QoE of any user with limited UAV energy and limited resource system, we jointly enhance UAV trajectory, user communication scheduling, and bandwidth allocation and transmit power to satisfy user QoE requirements and increase energy efficiency. Thus, the study proposes a new mud ring optimization with deep belief network-based resource allocation scheme (MRODBN-RAS) technique for UAV-enabled wireless networks. The proposed MRODBN-RAS approach focuses on the effectual accomplishment of the computational and energy-effective decision. Besides, the MRODBN-RAS technique assumed the UAV as a learning agent by forming RA decisions as actions. In addition, the MRODBN-RAS technique designed a reward function to reduce the weighted resource utilization. The MRODBN-RAS technique uses DBN model with hyperparameter tuning using MRO algorithm to allocate the resources. The design of the MRO algorithm helps in the optimal selection of the hyperparameter related to the DBN model. The simulation results of the MRODBN-RAS method are examined under various measures. The extensive comparison study highlighted the better performance of the MRODBN-RAS approach over existing techniques.

This study investigates the significance of leveraging the incorporation of Artificial Intelligence (AI), Building Information Modeling (BIM), and the Internet of Things (IoT) to Achieve smart sustainable cities. Understanding their applications for Architecture, Engineering, and Construction (AEC) projects. The study encompasses three key dimensions: Design Optimization and Performance Simulation, Material and Life Cycle Sustainability, and Operational Efficiency and Environmental Impact. By leveraging BIM and AI, the research explores the integration of renewable energy, sustainable material selection, and smart building controls. BIM and AI experts were given a structured questionnaire, which was then analysed using SPSS. Descriptive and correlation analyses reveal significant positive correlations between energy efficiency and design visualization, construction sustainability visualization, as well as adaptability and education through visualization. The proposed framework deepens the capabilities of the combination of different technologies towards Smart Sustainable Cities. This work not only contributes theoretical insights to the field but also provides practical implications for industry professionals striving to enhance sustainable practices in AEC projects. Further studies to encourage a combination of other recent technologies to improve smart sustainable cities' performance.

Internet of Things (IoT) based Arrhythmia Classification is a cutting-edge algorithm that amalgamates the abilities of the IoT and advanced medical diagnosis to revolutionize the detection and classification of arrhythmias—irregular heartbeats that may indicate fundamental cardiovascular issues. This technique leverages IoT devices, namely connected health monitors and wearable sensors, to continuously gather electrocardiogram (ECG) information from individuals. This information, streamed in real-time, provides a great opportunity for timely and remote monitoring of cardiac health. Leveraging the abilities of deep learning and IoT, this technique provides an automated and more sophisticated means of classifying and detecting arrhythmias, improving the efficiency and accuracy of diagnoses. This article presents an Internet of Things Enabled Based Arrhythmia Classification using the Dandelion Optimization Algorithm with Ensemble Learning (AC-DOAEL) method. The presented AC-DOAEL technique utilizes IoT-based data collection with an ensemble learning-based classification process. For the arrhythmia detection and classification process, the AC-DOAEL technique follows an ensemble learning algorithm such as long short-term memory (LSTM), autoencoder (AE), and bidirectional LSTM (BiLSTM) models. To improve the recognition rate of the ensemble models, the AC-DOAEL technique uses DOA as a hyperparameter optimizer. The simulation outcomes of the AC-DOAEL method are well-studied on benchmark ECG data. The experimental result analysis inferred the greater performance of the AC-DOAEL algorithm with other techniques.

Over the last several decades, the implementation of ITS has shown to be the most efficient and successful strategy for expanding the variety of current transportation networks. Vehicle-based offloading of data going to be essential for forthcoming networking innovations like D2D and 5G due to the substantial contribution it makes to efficiently using network capability while wasting minimal power. Information transmissions that would normally need a cellular network's infrastructure may instead be made using alternative networking mechanisms including Bluetooth, WiFi, and opportunistic communications. Data offloading has the ability to significantly increase the efficiency with which network resources are used. The offloading of data from vehicles has a considerable impact on the strain on cellular networks. It helps the network achieve higher throughput by facilitating the simultaneous reception of data by a large number of users. First, we must establish that the problem of Vehicular data offloading is an NP-hard target set selection (TSS) issue before we can even begin to characterize it. Using a combination of Hybrid PSO and GWO, TSS selects a small group of nodes to do the redundant data exchange (Particle Swarm Optimization with Gray Wolf Optimization). Collaboration between individuals and ISPs to identify effective aim sets may provide useful insights. If malicious users are present in the target group, they may slow down network activity by spoofing or by reducing the network's offloading capacity. It is possible that the whole network's performance would suffer as a direct result of these malicious users. In this study, we suggest a hybrid approach to communication for specifying the intended audience. We take use of the characteristics of opinion dynamics amongst users to get around the issue of overlapping community detection. Trust-based metrics inferred from users' activities are used to ensure the safety of the target set. In order to call 911, the suggested work additionally incorporates a method of sorting and classifying the offload limitations through Radial Bias Neural Network (RBNN). The following may be determined with the use of the proposed work's performance indicators: precision, entropy, and delay.

Early detection of Lung tumors, which is lethal and equally affects men and women, is challenging. In order to decrease mortality rates and raise survival rates, early detection and classification of Lung tumors is essential. However, at the start of 2020, the entire planet would be afflicted with a coronavirus that causes a fatal sickness (COVID-19). CT imaging is a good tool to detect illness among the various COVID-19 screening techniques available. On the other hand, alternative methods of disease detection take a lot of time. Deep learning, a type of machine learning, opens up a wealth of opportunities for investigating and assessing tumor features using CT scans, allowing for improved disease prediction, diagnosis, and classification. Using CNN, DNN, and VGG-16 models, the suggested approach in this research gives unambiguous and accurate categorization.

This research employs DEMATEL analysis as a methodological approach to thoroughly examine the adverse consequences of implementing Artificial Intelligence (AI) among students enrolled at Universiti Teknologi MARA (UiTM) Negeri Sembilan, Malaysia. The analysis encompasses three distinct professional cohorts: student representatives, academic staff, and upper management. Through a systematic analysis of causal relationships between multiple factors, this study aims to identify and prioritize the fundamental elements contributing to the negative consequences associated with integrating artificial intelligence. The prominence of privacy and security concerns as a causal factor highlights the importance of implementing strong data protection measures and adhering to ethical practices related to AI. Furthermore, various factors connected with personal disconnection, restricted adaptability, dependance on technology, and insufficient emotional intelligence influence the adverse outcomes of artificial intelligence implementation among students. The results underscore the necessity of implementing focused interventions and strategies to tackle these difficulties and guarantee a harmonious and advantageous integration of artificial intelligence in students' educational journeys. Higher education institutions can effectively harness the advantages of AI while ensuring their students' welfare and educational achievements by recognizing and proactively addressing any potential limitations.

A significant proportion of one type of pattern and a relatively small quantity of another type of pattern can be found in many unbalanced real data sets. In addition, finding significant observations and excluding influential observations is effectively accomplished through diagnostic analysis. Support vector machines (SVM), a common classification technique, perform poorly on imbalanced datasets and when influential observations exist. In this research, the pigeon optimization algorithm as a metaheuristic algorithm is employed to address the influence observation issues in SVM. Experiments are done on three real sets of data. Our approach provides higher classification accuracy compared to other widely used algorithms. This approach could be used for further biological, chemical, and medical datasets.

Wireless Sensor Networks (WSNs) are crucial in several applications, highlighting the need of effective clustering and fault detection systems.  This paper introduces a novel approach that uses Reinforcement Learning (RL) and Particle Swarm Optimization (PSO) to optimize cluster head selection and enhance fault detection capabilities within WSNs. The proposed hybrid algorithm operates in two phases, combining the explorative capabilities of RL with the optimization process of PSO to select cluster heads based on residual energy and connectivity considerations. By continuously monitoring the network's residual energy state and the number of active nodes, the proposed method ensures prolonged network lifetime and improved overall performance. Our experimental results demonstrate the superior performance of the hybrid RL-PSO approach compared to traditional clustering algorithms, showcasing significant improvements in optimizer accuracy, residual energy preservation, and fault detection efficiency.