International Journal of Communication Networks and Information Security (IJCNIS)
Not a member yet
1021 research outputs found
Sort by
Mathematically Engineered Adams Optimizer for Energy Efficient and Optimal Routing Approach for the Wireless Sensor Network
Get PDFWireless Sensor Network (WSN) comprises numerous Sensor Nodes (SN) scattered across a network to observe the atmospheric condition where the SN is the minimum cost. Vast usage of energy during data transmission is the primary design issue in WSNs that can be addressed by routing and clustering techniques. WSN has diverse transmission paths with unstable nodes, where the energy consumption is high, and the Quality of Service (QoS) is affected. The high data transmission delay and inefficient throughput indicate the ineffectiveness of WSN. To overcome this issue, this research concentrates on energy-efficient optimal routing formulated with the assistance of a mathematical approach and Adams optimizer. The mathematics’ based Pareto optimization is utilized to optimize the Adam Moment Estimation (Adam) that trains Deep Learning (DL) network that is deployed in both heterogeneous and homogeneous networks. The learning process is enhanced with the support of Pareto optimization, and the multi-objective problem is efficiently handled. In this context, Pareto optimization balances the path construction, and the equitable distribution issue is rectified by the Adams-based DL network. The proposed Pareto-integrated Adams Optimizer for Energy Efficient Routing (PAOEER) sustains the WSN performance by enhancing network parameters. The PAOEER achieves a higher Packet Delivery Ratio (PDR) of 97.18% and minimal Energy Consumption (EC) of 112.34 J. The simulation analysis shows that the proposed PAOEER is effective and outperforms the existing state-of-the-art techniques, indicating PAOEER is a promising alternative
Research on the Development and Innovative Strategy of Knowledge Payment Platform in the Internet Era
Get PDFThe widespread adoption of knowledge-based payment systems and numerous online knowledge-based paid items is made possible by the quick development of knowledge information and networking technology. Paid columns were quickly introduced by platforms like Himalaya FM, Zhihu, and Qingting FM, while pay-for- knowledge businesses like Fenda, iGet, and Qianliao quickly went live online. Users rapidly increased, and it was assumed that the knowledge payment progressive trend had been met. This paper aims to examine the emergence of edge computing into the mobile information system into the presence and inevitability of the knowledge payment platform, evaluate the advantages, challenges, and pathways for knowledge payment platform optimization, and try to provide a conceptual suggestion for aiding in its advancement. During data analysis, the independence of the polynomial characteristics was evaluated using Pearson's chi- squared tests. A brand-new meta-heuristic optimizer dubbed Harris Hawks optimization is inspired by how Harris hawks seek food in the wild. According to the experimental results, 98 of the survey's respondents—or 19.1% of all respondents—were under the age of 18; 201 were between the ages of 18 and 29; 142 were between the ages of 30 and 39; and 73 were over 40, or 14.2% of all respondents. The findings indicate a younger age distribution for the sample, with the concentration being highest among those between the ages of 18 and 29, then 30 to 39. The mobile information system edge-based knowledge payment platform has successfully undergone continuous use behavior analysis
Video Semantic Segmentation Network with Low Latency Based on Deep Learning
Get PDFRecently, new advances in deep learning algorithms have yielded some fascinating results in the field of computer vision technology. As a result, it can now perform activities that formerly required the use of human vision and the brain. Classification, object identification, and semantic segmentation have all seen substantial advancements in deep learning architecture in the last few years. For still images and movies, there has been a major advancement in the field of semantic segmentation. In practical uses like autonomous vehicles, segmenting semantic video continues to be difficult due to high-performance standards, the high cost of convolutional neural networks (CNNs), and the significant need for low latency. An effective machine-learning environment will be developed to meet the performance and latency challenges outlined above. The use of deep learning architectures like SegNet and FlowNet2.0 on the CamVid dataset enables this environment to conduct pixel-wise semantic segmentation of video properties while maintaining low latency. As a result, it is ideally suited for real-world applications since it takes advantage of both SegNet and FlowNet topologies. The decision network determines whether an image frame should be processed by a segmentation network or an optical flow network based on the expected confidence score. In conjunction with adaptive scheduling of the key frame approach, this technique for decision-making can help to speed up the procedure. Using the ResNet50 SegNet model, a mean Intersection on Union (IoU) of "54.27 percent" and an average frame per second of "19.57" were observed. Aside from decision network and adaptive key frame sequencing, it was discovered that FlowNet2.0 increased the frames processed per second9(fps) to "30.19" on GPU with a mean IoU of "47.65%". Because the GPU was utilized "47.65%" of the time, this resulted. There has been an increase in the speed of the Video semantic segmentation network without sacrificing quality, as demonstrated by this improvement in performance
Named Multipath Depth-First Search: An SDN-based Routing Strategy for Efficient Failure Handling and Content Delivery in NDN
Get PDFInformation-centric networking (ICN) architectures, such as named data networking (NDN), have emerged as potential solutions for efficiently retrieving and delivering content. However, challenges remain regarding routing scalability, resilience, and caching efficiency. Software-defined networking (SDN) offers opportunities to optimize NDN implementations through centralized control and programmability. In this paper, we propose Named Multipath DFS, an SDN-based routing and caching scheme for NDN networks. NMDFS leverages a centralized controller to pre-compute multipath routes and implement coordinated caching. We evaluate NMDFS on an emulated topology testbed against default NDN and Named-data link state routing. The results demonstrate significant improvements with NMDFS, reducing overhead signalling costs by 94% and 78%, respectively, compared with other schemes. Round-trip latencies for content retrieval were reduced by up to 98%. The SDN controller’s global network view and control are leveraged to optimize content caching through packet loss-driven adaptation and eliminate redundant messaging, leading to substantial performance gains
The Judgement of Genetic Algorithm on the Process of Maternal Image Change in Script Creation
Get PDFThe position of mother image change in script creation is very important, changing the logic of script creation, and even changing the task layout of the script. However, in the process of changing the mother's image, there are problems such as a large amount of analysis data and complex image construction. The main reasons are that the characteristics of the change of the mother image in different scripts are not summarized in place, the feedback of change data is not timely, and the data mining is not deep enough. Therefore, this paper proposes a mother image change method based on genetic algorithm to summarize the characteristics of mother image at different stages. The mother image data is collected by genetic algorithm, and the change data is summarized with the help of remote coding and multimedia network to complete the iterative calculation of the mother image data and identify the commonality and personality characteristics in the process of change. The calculation results show that under the conditions of remote coding and multimedia network, the genetic algorithm can improve the calculation level of the change of the mother's image, effectively find out the characteristics in the process of change, and meet the requirements of script creation
Research on the Influence Factors of ESG Performance on Corporate Value Based on Digital Information Technology
Get PDFThis research delves into the enhancement of ESG (Environmental, Social, Governance) performance evaluation in corporate contexts through advanced digital technologies. Initially, it involves the development of a comprehensive system designed to assess the impact of ESG factors on corporate value. This system particularly emphasizes the efficiency and reliability of transmitting digital information and performance-related data. The study then progresses to an in-depth analysis of various elements influencing individual ESG ratings. By employing sophisticated information technology tools, the research successfully pinpoints critical ESG performance indicators and implements a mechanism for the dynamic modification of their thresholds in response to evolving corporate and environmental dynamics. The results of the study are quite telling; they highlight the pivotal role of information technology in the accurate and effective evaluation of ESG factors. Specifically, the study demonstrates a remarkable 65% accuracy in the determination of influencing factors, surpasses 80% in overall judgment precision, and achieves an impressive effectiveness exceeding 90% in the assessment of influence on corporate value. These outcomes underscore the significant advancements in the field of corporate ESG evaluation, showcasing how digital technology can substantially refine and enhance the understanding and measurement of ESG performance's impact on corporate value. This breakthrough offers a promising avenue for corporations to more accurately gauge and improve their ESG contributions, ultimately leading to a more sustainable and socially responsible business environment
Using Gradient Descent to An Optimization Algorithm that uses the Optimal Value of Parameters (Coefficients) for a Differentiable Function
Get PDFDeep neural networks (DNN) are commonly employed. Deep networks' many parameters require extensive training. Complex optimizers with multiple hyper parameters speed up network training and increase generalisation. Complex optimizer hyper parameter tuning is generally trial-and-error. In this study, we visually assess the distinct contributions of training samples to a parameter update. Adaptive stochastic gradient descent is a variation of batch stochastic gradient descent for neural networks using ReLU in hidden layers (aSGD). It involves the mean effective gradient as the genuine slope for boundary changes, in contrast to earlier procedures. Experiments on MNIST show that aSGD speeds up DNN optimization and improves accuracy without added hyper parameters. Experiments on synthetic datasets demonstrate it can locate redundant nodes, which helps model compression
Intelligent agents Model with JADE for scheduling analysis and correction of Real-Time Systems
Get PDFThis research proposes a new model for analyzing and correcting non-schedulable partitions in real-time multiprocessor systems, specifically in the context of fault tolerance in distributed networks. The need for such a model arises from current techniques for correcting non-schedulable partitions that must be revised and repartitioning all tasks across processors. The proposed model is based on intelligent agents and implemented using the JADE platform. The model consists of (1) a supervisor agent in the first layer that distributes tasks and manages system correction when a non-schedulable partition is detected; and (2) a second layer composed of partition agents that analyze schedulability, request corrections, and negotiate with the supervisor for additional tasks to correct the entire system. The effectiveness of the proposed model is demonstrated through a case study. Quantitative analysis shows that the proposed model improves fault tolerance in distributed systems and has the potential for further enhancement by adding communicative tasks, heterogeneous processors, and other improvements
Analyzing the Users' Awareness of Data Privacy and Security on Online Social Network in Saudi Arabia: A Systematic Literature Review
Get PDFThe rise of Online Social Networks has brought about numerous benefits, in particular in the Kingdom of Saudi Arabia, has comes an inherent danger that cannot be ignored., the danger from both sides, the users themselves and the attackers. In the literature, we observed that there are various factors influencing the user's behavior regarding their information, privacy and security on OSN. Moreover, in Saudi Arabia, most studies focused only on one aspect of users' awareness which is privacy awareness to assessing the level of users' behavior within OSNs. In addressing this gap, we aim to systematically evaluate the literature on current research to provide a thorough understanding of the factors influencing the user's awareness behavior regarding their privacy and security on OSNs, whether world-wide or in Saudi Arabia studies. The results show interesting inferences e.g. we concentrate on empirical studies only to extract the privacy and security-related factors and determine the relationships between them and their impact on the users' awareness behavior on Online Social Networks. Further, our findings declared that most of the selected research studies adopt the quantitative approach for their investigation
Alzheimer’s And Parkinson’s Disease Classification Using Deep Learning Based On MRI: A Review
Get PDFNeurodegenerative disorders present a current challenge for accurate diagnosis and for providing precise prognostic information. Alzheimer’s disease (AD) and Parkinson's disease (PD), may take several years to obtain a definitive diagnosis. Due to the increased aging population in developed countries, neurodegenerative diseases such as AD and PD have become more prevalent and thus new technologies and more accurate tests are needed to improve and accelerate the diagnostic procedure in the early stages of these diseases. Deep learning has shown significant promise in computer-assisted AD and PD diagnosis based on MRI with the widespread use of artificial intelligence in the medical domain. This article analyses and evaluates the effectiveness of existing Deep learning (DL)-based approaches to identify neurological illnesses using MRI data obtained using various modalities, including functional and structural MRI. Several current research issues are identified toward the conclusion, along with several potential future study directions