International Journal of Communication Networks and Information Security (IJCNIS)
Not a member yet
1021 research outputs found
Sort by
Prediction of Alzheimer Disease using LeNet-CNN model with Optimal Adaptive Bilateral Filtering
Get PDFAlzheimer's disease is a kind of degenerative dementia that causes progressively worsening memory loss and other cognitive and physical impairments over time. Mini-Mental State Examinations and other screening tools are helpful for early detection, but diagnostic MRI brain analysis is required. When Alzheimer's disease (AD) is detected in its earliest stages, patients may begin protective treatments before permanent brain damage has occurred. The characteristics of the lesion sites in AD affected role, as identified by MRI, exhibit great variety and are dispersed across the image space, as demonstrated in cross-sectional imaging investigations of the disease. Optimized Adaptive Bilateral filtering using a deep learning model was suggested as part of this study's approach toward this end. Denoising the pictures with the help of the suggested adaptive bilateral filter is the first stage (ABF). The ABF improves denoising in edge, detail, and homogenous areas separately. After then, the ABF is given a weight, and the Adaptive Equilibrium Optimizer is used to determine the best possible value for that weight (AEO). LeNet, a CNN model, is then used to complete the AD organization. The first step in using the LeNet-5 network model to identify AD is to study the model's structure and parameters. The ADNI experimental dataset was used to verify the suggested technique and compare it to other models. The experimental findings prove that the suggested method can achieve a classification accuracy of 97.43%, 98.09% specificity, 97.12% sensitivity, and 89.67% Kappa index. When compared against competing algorithms, the suggested model emerges victorious
Analysis and Exploration of the Development and Application Status of 5G Communication Technology
Get PDFThe advent of 5G has invited opportunities for many socioeconomic benefits. The core principle of 5G is an array of contemporary technologies, which makes 5G render efficient networks, fostering new services, creating new ecosystems, and increasing revenues. The technologies in 5G continue to be developed from the new 5G ecosystem, which can transform the vertical industries. This work provides a comprehensive treatment of the 5G technologies, with its characteristics, new technologies that are a part of 5G, comparison with existing technology with a special mention of the technological advantages and disadvantages. The work analyses the salient features of 5G in the context of latency reduction, more connectivity, imparting spectral efficiency, and providing network services without compromising the QoS to a wider range of populations. The limitations of 5G technology are also discussed in the work, which throws light on futuristic research directions
Vehicle Detection and Speed Estimation Using Semantic Segmentation with Low Latency
Get PDFComputer vision researchers are actively studying the use of video in traffic monitoring. TrafficMonitor uses a stationary calibrated camera to automatically track and classify vehicles on roadways. 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. In this work, we discuss some state-of-the-art ways to estimating the speed of vehicles, locating vehicles, and tracking objects. 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 IoU of "54.27 per cent" and an average fps 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 second to "30.19" on GPU with such a mean IoU of "47.65%". Because the GPU was utilised "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
The Necessity of Digital Technology in the Supply Chain Finance Network Based on Digital Integration
Get PDFModern enterprise information consultancy and management firms are evolving with a fresh paradigm. This model emphasizes primary businesses and oversees the capital, data, and logistical operations of small to medium-sized entities. It aims to transform unpredictable risks into manageable supply chain enterprise risks, ensuring the most minimal financial service risks. Additionally, supply chain finance offers a broad spectrum of financial solutions for businesses throughout the supply chain. As technology advances, this has given rise to a novel supply chain financial ecosystem. This network can effectively execute supply chain finance operations. Notably, supply chain finance is inherently a credit-based financing system. Conventional techniques fall short in addressing the trust issues within the financial network of supply chain finance. This study introduces a digital methodology for financial network scrutiny. Initially, computer systems are employed to probe the trustworthiness challenges of the financial network, segmenting indices based on the network's demands to mitigate interfering elements. Subsequently, these systems evaluate the financial trust impacts on the supply chain, establish a financial network blueprint, and undertake a holistic examination of the financial network outcomes. Simulations in MATLAB indicate that, when assessed under specific criteria, the digital technology's financial network trust in supply chain finance surpasses traditional approaches in network reliability
Clustering Based Energy Efficient Routing for Wireless Sensor Network Using Particle Swarm Optimization
Get PDFThe clustering strategy is the most effective and efficient way to preserve energy in the Wireless Sensor Network (WSN). However, the cluster heads in the hierarchical clustering approach use the majority of the energy that is required to carry out the operations. These operations include receiving the data from the sensor nodes, aggregating it, and then eventually transmitting it to the base station. When choosing the appropriate cluster head, you can play a significant part in reducing the amount of energy that is consumed by the WSN and, as a result, extending its lifespan. A technique for the selection of energy-efficient cluster heads that is based on the particle swarm optimization method is proposed in this study (PSO-EECH). For the method that has been proposed to measure the amount of energy used, we need to take into account the cluster distance, the distance between each sensor node and the nodes that are nearby, and the amount of residual energy that is left in sensor nodes. The aforementioned structure is also capable of doing cluster building, in which the non-cluster head node can follow its CH based on the determined weight function. The proposed PSO-EECH approach has been put through extensive testing, and the results have shown that it possesses a high degree of accuracy in every scenario. The outputs of the proposed algorithm are compared with those of other clustering-based algorithms already in existence, and the conclusions of this comparison have reported that our method outperforms the other existing methods
Promotion of Intangible Cultural Heritage in China during the Epidemic Using Modern Visual Technologies with Information Security
Get PDFThe transition from physical work to cyberspace has been happening in a very rampant phase in recent years, which has changed and transformed the notion of traditional physical museums into digital but more interactive museums. The worldwide shutdown due to the pandemic has led to the closure of museums and other cultural heritage artifacts. However, a major advantage that could be reaped from technological advancement is the protection and inheritance of age-old Intangible cultural heritage without disrupting its originality by deploying immersive technologies. This study proposes a holistic three-layered framework that considers the possible technologies, storage options, and user views without compromising information security. An overview of the widely deployed technologies is also presented in the work along with its potential usage. The work enumerates the popular technologies, software, and tools that can find their application in the establishment of digital museums. Finally, the work discusses the important challenges and limitations that should be confronted by the digitization of the elements of museums, which are the future research directions. These limitations are a blessing in disguise that can be convolved into the de facto design of the future digital museums to provide a more realistic and appealing experience to the visitors so that the ICH is preserved for generations
SECURE ENCRYPTION BASED UNICAST ROUTING WITH GENETIC APPROACH
No full textThe wireless or mobile ad hoc network is one of the types of decentralized network. MANets are self-contained, multi- hop networks. MANet protocol’s main objective is to construct an accurate and effective route between two nodes so that messages can be delivered promptly. The process of natural selection and evolution served as the inspiration for genetic algorithms (GAs), which are optimization algorithms. these algorithms are used to identify the best solutions to challenging situations. The route-repair function fixes the route locally during this crucial time, or if not, it sends the information to source nodes to help discover a new route well in advance. This method forecasts the availability time in the link, which indicates a brief time before the link fails. Unauthorized nodes can intervene and interrupt the entire network during the selection process by sending erroneous data, or completely restricting the content as well as other services. In order to ensure data confidentiality in MANets, GAs can be employed to optimize the secure distribution of keys using key chaining method. The act is being performed by GA’s and Newton’s interpolation method and utilizing the network simulator, results are produced
FSGS NET: EARLY DETECTION OF FOCAL SEGMENTAL GLOMERULOSCLEROSIS VIA MULTI MODAL DEEP NEURAL NETWORKS
No full textIntroduction Focal SegmentalGlomerulosclerosis (FSGS) is a devastating form ofkidney disease that commonly progresses to end stagerenal disease in an aggressive manner if not diagnosedappropriately. Laborious, invasive and not conducteduntil after significant symptoms have alreadydeveloped (e.g. kidney biopsies) The aim of this studyis to put forward a novel machine learning basedapproach presenting a FSGS NET model is a MMDNNfor early and efficient detection of FSGS. Usinggenetic profiles, clinical data, imaging scans andbiomarkers from a va riety of sources, the algorithmcan identify FSGS in its earliest stages before thefirst symptoms of disease have appeared. The networkarchitecture builds three data specific branches, fullyconnected layers to process genetic data, dense layersfor cl inical data, and convolutional neural networksfor imaging data. The result of the above modelappears as a shared fused layer to combineinformation from all branches. Its output also providesthe probability of FSGS, thus allowing for a moreaccurate dia gnosis with less need to biopsy. At the endof implementation, the proposed framework is testedon five parameters: accuracy, sensitivity (recall),specificity, predictive value and Area Under the ROCCurve (AUC). Our multi modal strategy should helpincre ase the rate of early detection, offering a noninvasive means for clinicians to identify FSGS andbetter devise personalized treatment plans. This modelis expected to increase the performance of patientoutcomes as well as reduce invasive diagnosticproc edures, such as colonoscopies
A DDoS Attack Detection using PCA Dimensionality Reduction and Support Vector Machine
Get PDFDistributed denial-of-service attack (DDoS) is one of the most frequently occurring network attacks. Because of rapid growth in the communication and computer technology, the DDoS attacks became severe. So, it is essential to research the detection of a DDoS attack. There are different modes of DDoS attacks because of which a single method cannot provide good security. To overcome this, a DDoS attack detection technique is presented in this paper using machine learning algorithm. The proposed method has two phases, dimensionality reduction and model training for attack detection. The first phase identifies important components from the large proportion of the internet data. These extracted components are used as machine learning’s input features in the phase of model detection. Support Vector Machine (SVM) algorithm is used to train the features and learn the model. The experimental results shows that the proposed method detects DDoS attacks with good accuracy
Impact of Social Network Sites on Developing Brand Communication in Xian City
Get PDFThe digital era mobilises the world cities with various traits and symbolic resources to brand the images of the cities through propagation in Social Media (SM) platforms. This work examines Xian, the fastest developing city in China, which is in the process of building its brand through SM platforms. The model proposed in this work uses elaborative variables that are categorised into four indicators which characterises the behaviours of users in the SM. Each variable is weighted based on the entropy method as each one has a varied level of intensity in the model. The detailed results indicate that the size of the network is a significant factor in construction of brand communication through the SM platforms. Further, the study also includes the descriptive statistics and correlation analysis of individual explanatory variables with the model’s output value. The result of this analysis indicates that connections with friends and a number of followers as crucial factors in board communication of Xian city. This work can guide the local municipal authorities to advertise and influence SM users to create a positive brand of the city