Search

Wireless Sensor Node (WSN) helps to track inpatient and remote patient (home/working) health information. Mishandling of the electronic system, patient behaviour and environmental changes which are all lead to incorrect data generation while using WSN for medical purposes. It leads to a false alarm being raised, network resource wastage, a false node priority level and low reliability. We have introduced the Mutual Trust Model (MTM) for Wireless Body Area Network (WBAN) with the help of Fog-Node (FN) to address these issues and to ensure the trustworthiness of the information acquired. In this, First-Hand Trust Method calculates the confidence value of the individual sensor node. Then, with neighbor node support, the Stigmercy Trust Method (STM) is implemented to reinforce the trust source node. Ultimately, the individual patient’s confidence value for the MTM model is determined. With the assistance of the wireless-mininet network emulator and the RYU controller, the network environment model implement, and the results have been obtained. MTM predicts the confidence level of the collected data significantly and produces an accuracy of 92.3 percentage to prevent the emergency band from being used dispensable.

As the big data and internet growing huge day by day, there is in need to overwhelm people by the information in large. These kind of issues makes researchers to create a technique that can generate automatic text. This research is very great interesting topic and the summarization of text is the process which undergoes finding important information within a document. Automatic text summarization is the process where a subset is created which represents the relevant content using the original text trained. So, text summarization is an intimidating challenge in Natural Language Processing because getting a good summary is a very big challenge as it is based on various aspects like metrics, non-redundancy, readability etc., At the outset, the text is trained into the system, next the trained text is fed into encoder decoder process where additional techniques are employed between the result in best coherent summary. The aim of automatic text summarization is to the produce finest and top quality text summary and our model also produces supreme quality summary.

From the recent decade, high degree of attention is drawn over Sentiment Analysis (SA) in which sentiments are collected, investigated and aggregated from the text. The growth of SA has led to the introduction of different subdomains, each handling a diversified degree of investigation or research issue. This review concentrates on the Aspect Level Sentimental Analysis (ALSA), where the objective is to identify and aggregate sentiments or entities specified in the documents or the dimensions of them. An in-depth overview of the existing state-of-the-art methods of SA that portray tremendous growth in estimating the opinion or target which may relate to an entity and its associated sentiment is presented. This review also presents the different ALSA methods that include Frequency-based, Syntax-based, Supervised and Unsupervised machine learning based and Hybrid techniques. It also presents the future scope of research that could be possibly conducted using ALSA in diversified fields during its implementation.

Generally, a data or an information of a device will play a key role to find the owner and the things were done in that device from starting day up to the day it was tested were shown in the information which we get from the device. And getting that data was done by data extractors and the problem with the extractor is, for every type of data it needs a different type of extractors to overcome this problem, and this bulk extractor gets all types of data of a device without using any other extractor. This bulk extractor saves the time of testing and gathering information and gets all types of data easily.

Internet of things (IoT) remains one of the most electrifying and fast burgeoning fields. It permits most of the things like any objects are controlled with very less manpower and we can control things anywhere and anytime by using network. In this paper that going to be constructing that the “Detection of Blockage in Manhole Using IoT”. Detecting a struck in the manhole pipe, due to blockage of sludge in a pipe. Here by using detecting the length of the pipe using ultrasonic sensor till where the sludge struck with the help of capturing image of sludge to aware of, due to gloominess inside a manhole by provide a luminescence, which helps to capture picture of particular sludge. And with the help of developing an application to control the device, by using that application data access can be done (captured picture) using a cloud. In traditional method, sewers have to do cleaning in all the cities (areas). In gutters, workers stand in the waste, which reaches chest level and with the help of long wooden sticks and broomsticks they clear blockage. In some areas workers crawl into a sewage, without wearing any protective gear. And those workers are treated very badly by the society. Thus, sewers need to work a lot to clear blockage. But now by using this technique it will be very much easier to reduce a man power and also incremental steps can be added further.

The notion of Cloud computing has reshaped the field of a distributed system as well as that changed the organizations to use processing and computing today. The educational institutions have been utilizing the computing resources through the e-learning technologies, which overcomes the traditional pedagogy in learning but the majority of the understudies in the village are not able to get their instructive properties and the highlighting issue that emerges inside the training framework is missing of value in teaching and that are resolved through the e-learning assets utilization. The accessibility of the asset is not as simple for the bucolic based environment and they were not aware of cloud resource utilization. It can be made accessible through the custom built educational portal. Thus it enables understudies of common domains to get to the electronic utilization with the help of tablets and mobile phones.

In a Vehicular Ad-hoc Network (VANET), vehicles communicate with each other and possibly with road-side infrastructure nodes. Node connectivity and the amount of data that can be exchanged are limited by the duration and quality of the communication links established among nodes, which are determined by the space and time dynamics of moving vehicles. We analyze the connectivity dynamics of vehicular ad-hoc networks in a generic signalized urban route. Given the velocity profile of an urban route as a function of space and time, we utilize a fluid model to characterize the general vehicular traffic flow, and a stochastic model to capture the randomness of individual vehicle which is determined by Mesoscopic,and Microscopic mobility model.Mesoscopic is a technique which is used to find time headway between two vehicles and microscopic technique is used to find a distance headway between the two vehicles.

Image processing for medical field has contributed a max beyond limits and many fuzzy solutions have also been proposed. In current scenario of technological addiction cataract based problems are encountered due to prolong exposure towards sharp light emitting devices. 55% of vision impairment is reported from cataract under urban region. In this paper, a cataract density ration is analyzed under color image processing for acquiring affected and unaffected region. The paper has reached the accuracy rate of 87.3% under medical samples.

Wireless spread spectrum is a standout amongst the most valuable assets. With the fast development of the quantity of remote gadgets, interchanges over the unlicensed range groups get to be extremely swarmed. Data dissemination in the wireless environment is usually characterized as the spreading of data to numerous goals through communicating. The principle target is to achieve the most extreme number of neighbors with each sent parcel. The main objective of the paper is to find the best routing and resource allocation strategies in order to minimize the average end-to-end delay of multiple data connections in the cognitive radio based wireless mesh network. Wireless mesh nodes utilize the cognitive radio to share the spectrum with primary users. Depending on the primary user's activities, the spectrum resources available to the cognitive mesh nodes are varying in both space and time. In this joint design scheme can accommodate the traffic load, or achieve half the delay compared to the disjoint methods.

Energy saving is one of the major concerns in the present era. The project highlights the necessity of Power Audit in the residential feeders and the need of local MicroGrid design. The present study deals with Power Audit Analysis of 132 residential houses and 10 commercial buildings, connected to the 250 Kva distribution transformer feeders. The Detailed Power Audit is carried out for 10 houses connected to the feeder and preliminary audit is carried out for 122 houses. Recommendations are proposed further with Micro Grid for reducing the dependency of at least 20% loads of houses on Main Grid and for encouraging people to generate their own power. The suggested implementation tested with ETAP simulation and concluded with the improvement of energy efficiency of Residential houses and thereby reducing the energy wastage.

A Wireless sensor network is a collection of Wireless nodes with limited energy capabilities that may be mobile or stationary and are located randomly on a dynamically changing environment. How to effectively control the power of the sensor nodes and extend the life time of the whole network become a very important issue. This project presents the clustering approach which includes the Wireless sensor network. In the clustering approaches share the loads with cluster head nodes. To extend the cluster head's life and extend the life time of network based on the distance between the cluster head and base station. Energy usage is minimized by exploiting density estimation, sleeping modes and duty cycle control and distance between the cluster head and nodes. In this work, we proposed clustering mechanism. Nodes associate with the nearby CH. Nodes are classified in to critical nodes and non-critical nodes. Critical nodes are continuously transmission between the CH and nodes. so energy usage of critical node is high at the same time non-critical node based on the open loop system. It is not continiously transmit, so minimized energy usage of non-critical nodes. The proposed mechanism will effectively reduce the frequency of the transmission of the sensor nodes more close to sink and achieve the effectiveness of saving power and extending the lifetime of network.

In this paper, we present a Modified Revised Wilson Current Mirror Level Shifter (MRWCMLS) to convert subthreshold voltage to supply voltage with low leakage power. Increasing speed and complexity of today's designs resulted in increase in power consumption. System on Chip (SoC) may have few modules which are supplied with subthreshold voltage to achieve low power consumption and the remaining work with above threshold voltage supply. The level shifter which can convert subthreshold voltage level to supply voltage level is designed to minimize circuit leakage power with optimum number of transistors in the circuit. MRWCMLS has been implemented in 180 nm CMOS technology and is simulated using Cadence Virtuoso Simulator and the results are compared with the existing level shifters. In comparison with Revised Wilson Current Mirror Level Shifter (RWCMLS), the leakage power with normal transistors and with high Vth transistors type of implementation of the proposed MRWCMLS, is reduced by 34% and 43% respectively.

The process of demonstrating, organizing and evaluating the pictures regarding the information despite of evaluating pictures is the field of Content Based Image Retrieval (CBIR). Here we work on the salvage of images based not on keywords or explanations but on features haul out directly from the image data. The well-organized algorithms of salvage algorithms are already proposed. Content Based Image Retrieval has replaced Text Based Image Retrieval. CBIR is processed by more methods and research scientists are working to improve the accuracy of the technique. The project presents that the ROI from an image is retrieved and it retains the image based on Teacher Learning Based Optimization genetic algorithm. The retrieval of the image improves the efficiency based on two metrics such as precision and recall which is the main advantage of the project. The issue of Content Based Image Retrieval systems to provide the semantic gap and to determine the variation between the structure of visual objects and definition of semantics. From the human visual system the visual courtesy is more projected for the purpose of Content Based Image Retrieval. The new similarity based matching method is described based on the saliency map which retains the courtesy values and the regions of interest are hauled out.

Authentication is an obligatory factor in the technological world regarding security. Two factor authentication secures access network and validates and protects user identities. CAPTCHA-"Completely Automated Public Turing" is a challenge response test used in evaluating to check the user is human or not. Modern Cryptography is a practice of representing the assumptions in which the cryptographic protocols are secured in a clear manner. Graphical user Authentication is in the form of cued-click points. However, they do not deal much with shoulder surfing attacks. It is a security option that is designed based on hotspot concepts and the hidden patterns.

This paper aimed to compare the analysis of two different face annotation techniques. There is a sudden powerful research in mining weakly labeled web facial images on the internet to tackle the existing longtime research challenges in computer vision and image processing. In this paper, unified learning framework for auto face annotation by mining web facial images and another scheme Mining weakly labeled web facial images for search-based face annotation are analyzed and the former designed with two learning schemes such as inductive and transductive learning's which are used for this framework. For inductive, weak label laplacian support vector machine is proposed and for transductive weak label regularized local coordinate coding is used. By combining inductive and transductive learning schemes produces a maximum efficacy. For search -based face annotation two algorithms such as ULR algorithm is used to refine the weak labels of web facial images using machine learning techniques and clustering algorithm is used to improve the scalability of the web facial images.

Earlier, web services are used to retrieve data for small amount of user only, but now a day's more data retrieved through web services by more number of users in all the actions of day to day life. There are numerous web services available in the internet to collect the data. Today internet plays a vital role in every body's life. Internet provides all the information one can access through the web service. To make the retrieval more efficient, Hidden Markov Models (HMM) &Mediator Agent Model are used in the project. The Hidden Markov Models can be used to measure and predict the behavior of Web Services in terms of response time, and thus can be used to rank services quantitatively. The mediator agent Model is used to select web services with the help of ranking method. The Mediator Agent Model will provide web service to the user from the web service provider. It will be processed in such a way that optimal result is obtained. If many users refer the optimal site then server loading will occur to reduce that problem server redirection is used in this project.