Search

Nowadays, a massive amount of data leads to cause network traffic and inflexible mobility in future mobile networks. A new Group Mobility Model (GMM) named MoMo is introduced that addresses the issue of the aforementioned problems. Even though, software defined network (SDN) is functional with network-rooted mobility protocols that enhance the network efficiency. Some existing network-rooted mobility administration methods still undergo handover delay, packet loss, and high signaling cost through handover processing. In this research work, SDN-based fast handover for GMM is proposed. Here, the neighbor number of evolving node transition probabilities of the mobile node (MN) and their obtainable resource probabilities are estimated. This makes a mathematical framework to decide the preeminent number of the evolving nodes and then allot these to mobile nodes virtually with all associations finished by the exploit of Open-Flow tables. The performance examination demonstrates that the proposed SDN rooted GMM technique has the enhanced performance than the conventional handover process and further technique by handover latency, signaling cost, network throughput, and packet loss. [ABSTRACT FROM AUTHOR]

Difficulty in communicating and interacting with other people are mainly due to the neurological disorder called autism spectrum disorder (ASD) diseases. These diseases can affect the nerves at any stage of the human being in childhood, adolescence, and adulthood. ASD is known as a behavioral disease due to the appearances of symptoms over the first two years that continue until adulthood. Most of the studies prove that the early detection of ASD helps improve the behavioral characteristics of patients with ASD. The detection of ASD is a very challenging task among various researchers. Machine learning (ML) algorithms still act very intelligent by learning the complex data and predicting quality results. In this paper, ensemble ML techniques for the early detection of ASD are proposed. In this detection, the dataset is first processed using three ML algorithms such as sequential minimal optimization with support vector machine, Kohonen self-organizing neural network, and random forest algorithm. The prediction results of these ML algorithms (ensemble) further use the bagging concept called max voting to predict the final result. The accuracy, sensitivity, and specificity of the proposed system are calculated using confusion matrix. The proposed ensemble technique performs better than state-of-the art ML algorithms. [ABSTRACT FROM AUTHOR]

With the arrival of on-demand computing, data centre requirements are extensive, with fluid boundaries. Loaded Internet applications, service-oriented architectures, virtualization and security provisioning are the major operations of a data centre. Security is an absolute necessity of any network architecture, and the virtual IT data centre is no exception. At the boundary, security is focused on securing the terminals of the data centre from external threats and providing a secure gateway to the Internet. The paradigm shift towards a new computing environment makes communications more complicated for Infrastructure Providers (InP). This complexity includes the security of the data centre's components to protect data from malicious attacks or from being compromised. Threats/attacks are inevitable if the data are generated from a public network, such as Wi-Fi in an Airport, Railway station and other public places. Since these places create enormous amounts of data from anonymous and naive users, it is essential to store the information in a data centre. In this article, we propose an efficient, secure, and privacy-preservation information access algorithm to access data centres in public wifi networks. This algorithm is based on the primitive root approach for sending and receiving credentials through the anonymous authentication of the users and ensuring protected data access from the data centre. Security and Performance Analysis and its evaluation prove that our approach is successful with respect to security, privacy preservation and computational complexity. [ABSTRACT FROM AUTHOR]

Existing research works for Alzheimer's disease (AD) can predict the prevalence of disease only after the advancement of the disease. With these existing prediction models, it is possible only to reduce and delay the symptoms of the disease. The exact usefulness is when the presence of the disease is identified at an early stage and this early detection makes a great impact in subjects' recovery. Thus, early detection of controls at high risk of development of Alzheimer's disease is of a key objective of the proposed work. Existing machine learning and deep learning algorithms derive only limited predictive accuracy. Also, they derive results based on expensive machine learning algorithms that had hard-to-collect features and classifying becomes complex with numerous overfitting in choosing decision boundaries. The proposed study intends to develop a learning algorithm for the prediction of Alzheimer's disease at an early stage. It also classifies the features if the subjects with Mild Cognitive impairment (MCI) and Pre-Mild Cognitive Impairment (Pre-MCI)has the likelihood to develop Alzheimer's disease. A dataset of AD controls was used to train different machine learning algorithms. Onset information like social behavior, demographic characteristics, neurological test scores, clinical cardiovascular index, and brain atrophy ratio can also be used as the extract predictor. A validation procedure was applied to identify a relevant subset of predictors. The conversion to AD in MCI and Pre MCI subjects are based only on non-invasively and effectively collectible predictors. [ABSTRACT FROM AUTHOR]

With the ageing people community, Alzheimer's disease (AD) is the most common disease in prevalence. It affects the memory of the affected individual by causing neuron degeneration. Such cognitive disability leads people to roam around places since they are ignorant about location. They may go afar from the safe area to traffic prone areas which cause serious, life threatening issues to them. Proposed study introduces a technology that involves a wearable device for Alzheimer disease affected elderly patients and listening device given to care takers or doctors. This system detects and notifies caregivers about the departure of people to risk locations from their residence zone. This system uses Bluetooth Low energy (BLE) communication technology that utilizes transmitter and receiver fixed to elderly individual as wearable device and in residence respectively. Based on the signal strength between the transmitter and receiver, the distance estimation between the elderly and the receiver is calculated with the help of a Raspberry Pi(Rpi) and an alert notification is sent to doctors/caretakers when the distance exceeds safety limit. [ABSTRACT FROM AUTHOR]

Grid Interoperation is a short term solution to join different grid middlewares, which works in the concept of developing bridges/gateways to overcome the dissimilarities of the grid middleware functionalities. Grid interoperability is a long term effort which aims at building standard components pluggable with the native middleware and thereby making it seamlessly connected with other standard middlewares. In this paper, we discuss a grid interoperation mechanism, Extensible Service Based Grid Information Aggregator, which aggregates the resource information from different middlewares viz. GlobusTookit 2.4(GT2), Globus Toolkit 4.0(GT4), UNICORE and gLite. The proposed information aggregator is service based, extensible and pluggable with resource brokers and Meta schedulers. The salient feature in this work is that there is no need for any modifications at the middleware level. The performance of Extensible Service based Grid Information Aggregator (EGSIA) is tested by providing the aggregated resource information to a simulation model which runs deadline driven grid application. [ABSTRACT FROM AUTHOR]

• Feature Selection process perform better selection from UCI heart disease dataset. • ResNet-50 classifier approach is used for secured transmission of heart disease features. • Enhanced ElGamal method for encryption and decryption of data with key generation. • ResNet-50 with near 50 layers classifier for better classification process. • Comparative analysis in terms of performance metrics are performed. Heart disease is increasing, and their detection is a significant concern. With the several technologies developed, the approaches for detection mechanisms can be improved further with better and improved algorithms. Data related with patient's health are of huge amount and stored in the larger space of cloud storage. Accessing cloud storage is an easy task, where the data stored is available to many users of cloud and there comes the need of security. Generally security can be improved using encryption and decryption algorithms. In this paper, a framework using the ResNet-50 classifier approach for secured transmission of heart disease features is performed. This study focused on an enhanced ElGamal encryption-decryption method for the encryption of data with a generated private key and a public key for decryption to better access the data. The data encrypted are then decrypted when the user request data. With Convolutional Neural Network classifier of ResNet-50 with its near 50 layers, the refinement or classification process is performed. The heart disease dataset from the UCI heart disease repository is considered for the evaluation of the proposed work. Further, with feature selection method, better selection of input can be filtered from the selected dataset. The results are obtained with respect to various performance measures, then compared and analyzed with some of the existing methodologies. The results proved to be better than other existing frameworks. [ABSTRACT FROM AUTHOR]