Abstract

In this paper, an improved wavelet based medical image watermarking algorithm is proposed. Initially, the proposed technique decomposes the cover medical image into ROI and NROI regions and embedding three different watermarks into the Non-Region Of Interest (NROI) part of the transformed DWT cover image for compact and secure medical data transmission in E-health environment. In addition, the method addressing the problem of channel noise distortion may lead to faulty watermark by applying Error Correcting Codes (ECCs) before embedding them into the cover image. Further, the Bit Error Rates (BER) performance of the proposed method is determined for different kind of attacks including ‘Checkmark’ attacks. Experimental results indicate that the Turbo code performs better than BCH (Bose-Chaudhuri-Hochquenghem) error correction code. Furthermore, the experimental results validate the effectiveness of the proposed framework in terms of BER and embedding.
Full Paper

Abstract

Artificial Intelligence has spurred the advent of self-driving cars. Commercially viable autonomous cars inculcate lot of complex sensors in order to perceive the world. These cars are built for more standardized road conditions with distinct features. So, this paper cites an approach undertaken to suit more mercurial environment. A children’s toy car is used to maneuver over the university campus road which is built of Reinforced Cement Concrete (RCC) and has very few features that too with subtle difference. Environmental noise like shadows and varied daylighting condition posits a grim challenge. A real-time system is built with algorithmic software driven approach. Also, brute logic is coded to mimic the behavior of actual self-driving car. For example, gradual acceleration, gradual braking, sudden braking in case of emergency. Concepts of Pulse Width Modulation (PWM) have been utilized to control the acceleration-velocity dynamics of car. Transfer Learning is used to classify and detect obstacles in the path and algorithm is written to aid subsequent decision making. Overall, a rather brute force algorithmic approach is used to build this real time system.
Full Paper

Abstract

Today Big data is the source of much valuable information. It is extensively used by different agencies like business organizations, governments, academicians, policy makers, etc., for knowledge discovery. Though it is beneficial to many, but all are facing crucial challenges also from it. These challenges are originating from the characteristics of big data and they can be studied from different angles; such as data challenges, process challenges and management challenges. The main objective of this paper is to review the big data challenges from the characteristics of data itself by exploring its sources and identify potential research issues developing from it.
Full Paper

Abstract

A dashboard is a kind of a graphical user interface which is used to analyze and display the key performance indicators, metrics and important data points associated with a business process or an application. Geographic Information Systems (GIS) are computer based systems to collect, manage, analyze and present large amounts of geo-spatial data.Operations Dashboard for ArcGIS is a module which can be integrated in to the Portal for ArcGIS to build map based dashboards. It contains various widgets like home, zoom in/out, legend, pie chart, graph, map, indicator, gauge, list, details etc. The widgets can be connected and interlinked to each other using “actions”. Operations ashboard for ArcGIS is a simple application with the help of which a map enabled dashboard can be created in a short period of time and that to without any programming skills.This paper reviews a simple and easy to use methodology to create a dashboard using ArcGIS Enterprise tools like Operations Dashboard for ArcGIS, Portal for ArcGIS and Server for ArcGIS.
Full Paper

Abstract

The explosive growth of wireless communication nodes has encountered a spectrum scarcity problem which led to emergence of intelligent communication technology known as Cognitive Radio (CR). CR provides opportunistic utilization of spectrum band by unlicensed users when licensed users are idle. Security of Cognitive Radio Networks (CRN) is very challenging which needs to be modelled effectively considering the dynamic nature of the environment. In this study, an effective authentication mechanism is proposed addressing emulation attack, denial of Service attack and withstanding replay attack.
Full Paper

Abstract

Recent advances in wireless networking and big data technologies, such as 5G networks, med-ical big data analytics, and the Internet of Things, along with recent developments in wearable com-puting and artificial intelligence, are enabling the development and implementation of innovative diabetes monitoring systems and applications. Due to the life-long and systematic harm suffered by diabetes patients, it is critical to design effective methods for the diagnosis and treatment of diabe-tes. Based on our comprehensive investigation, this article classifies those methods into Diabetes 1.0 and Diabetes 2.0, which exhibit deficiencies in terms of networking and intelligence. Thus, our goal is to design a sustainable, cost-effective, and intelligent diabetes diagnosis solution with person-alized treatment. In this article, we first propose the 5G-Smart Diabetes system, which combines the state-of-the-art technologies such as wearable 2.0, machine learning, and big data to generate comprehensive sensing and analysis for patients suffering from diabetes. Then we present the data sharing mechanism and personalized data analysis model for 5G-Smart Diabetes. Finally, we build a 5G-Smart Diabetes testbed that includes smart clothing, smartphone, and big data clouds. The experimental results show that our system can effectively provide personalized diagnosis and treatment suggestions to patients.
Full Paper

Abstract

Plagiarism, in recent times, is one of the serious problems due to which the uniqueness in the writings of different authors is being impaired. Plagiarism detection systems have to be developed to help preserve an author’s identity, as there is a lot of effort put in by him to write a document. The aim is to study and develop an automated tool to detect cases of plagiarism in a document through our project.
Extrinsic plagiarism is where a suspicious document is compared with a given set of source documents and phrases, sentences, etc. which appear in both the documents. It may not simply mean copying text from a document, but copying it and using it in a modified form with a misconception of not being detected as plagiarism. This system will be able to detect extrinsic plagiarism. The text in the document is first to split into k-grams. The hash values i.e. fingerprints of all the k-grams are then generated using the Karp-Rabin string matching algorithm. A subset of all the fingerprints is calculated using the Winnowing algorithm. A similar process is carried out for the suspicious and source documents and then the fingerprints of both documents are compared using Jaccard similarity to get the percentage of plagiarism.
Full Paper

Abstract

The main aim of this project is FPGA implementation of the convolutional encoder and Viterbi decoder which will allow the receiver to detect and correct any errors without the need for retransmission. The Convolutional Encoder and the maximum likelihood Viterbi decoder solve the problem of the transmission of data in an erroneous channel. This project is aimed at designing and implementing the encoder and decoder on an FPGA so that the receiver can detect and correct any errors without the need for retransmission.
Full Paper

Abstract

Internet Network contains different yet reliable Protocol for Internet Routing. Recently there has been a substantial increase in the number of Internet Routing Protocols. However, the performance of these different types of protocols is not up to the mark, from this survey paper different types of protocols have been investigated including MIPv6, FMIPv6, HMIPv6, FHMIPv6, PMIPv6, and DHMIPv6 are investigated in the DMM environment. The effective solution for mobility management mechanism is needed for the end users while changing their location. For supporting mobility, Mobility management protocols have been introduced was evolved from Host-based approach to Network based approach. In network-based protocols, host was shielded by transferring the mobility – related signaling to the network entities whereas host-based protocols the mobility-related signaling is involved with MN. The purpose of the mobility protocol is to enable network applications to operate continuously with the good quality of services for both wireless networks.
Full Paper

Abstract

Distributed denial of service (DDoS) is still one of the main threats of the online services. Attackers are able to run DDoS with simple steps and high efficiency in order to prevent or slow down users’ access to services. In this paper, we propose a novel hybrid framework based on data stream approach for detecting DDoS attack with incremental learning. We use a technique which divides the computational load between client and proxy sides based on their resource to organize the task with high speed. Client side contains three steps, first is the data collecting of the client system, second is the feature extraction based on forward feature selection for each algorithm, and the divergence test. Consequently, if divergence got bigger than a threshold, the attack is detected otherwise data processed to the proxy side. We use the naïve Bayes, random forest, decision tree, multilayer perceptron (MLP), and k-nearest neighbors (K-NN) on the proxy side to make better results. Different attacks have their specific behavior, and because of different selected features for each algorithm, the appropriate performance for detecting attacks and more ability to distinguish new attack types is achieved. The results show that the random forest produces better results among other mentioned algorithms.
Full Paper

Abstract

The major cause for death in human beings is because of cancer. Lung cancer is one of the most common and serious types of cancer that severely harms the human body. In order to cure the cancer early cancer detection is required. If lung cancer is diagnosed at early stages many lives will be saved. The other name for lung cancer is lung carcinoma, an uncontrolled malignant tumor distinguished by undisciplined cell growth in lung cells. There are many people suffering from this kind of cancer and confining to death. If this is left untreated, this may grow later than lung by metastasis into other parts of body. Many of the cancers starts from lungs, called as primary lung carcinoma. There are two types of small cell lung carcinoma (SCLC), non small cell lung carcinoma(NSCLC). The main reason for lung cancer is smoking of cigarette. There are many researches targeting on exact approaches for treating cancer. To predict the survival rate for NSCLC patients data mining techniques can be used with selection of algorithms. The algorithms used to detect the lung cancer are Support vector machine (SVM), Decision tree, k-Nearest neighbour, Random forest, Logistic regression. In this paper By implementing 2 different datasets and various packages and libraries in python, it is compared and on implementation found suitable algorithms have more accuracy on certain data sets for optimum prediction rate of lung cancer.
Full Paper