Thesis 2017
Browse
56 results
Search Results
Item Feature Extraction and Analysis of Large Biological Data(East West University, 2017-04-23) Islam, Md. Imranul; Binti, Umme KulSum; Kuddus, Md. AbdulDNA sequencing is the process of determining the sequence of nucleotide bases (adenine, guanine, cytosine, and thymine) in a piece of DNA that is represented as A, G, C and T respectively. Today, with the right equipment and materials, sequencing a short piece of DNA is relatively straightforward. The advent of rapid DNA sequencing methods has greatly accelerated biological and medical research and discovery. Knowledge of DNA has become indispensable for basic biological research, and in numerous applied fields such as medical diagnosis, biotechnology, forensic biology, virology and biological systematics. The rapid speed of sequencing and searching attained with modern DNA pattern searching technology has been instrumental. Nowadays analysis of large biological dataset using searching a pattern from DNA needs faster and cost effective machines to attain more accurate result within a short time. It is so much difficult to handle large set of biological data. Searching a specified pattern is now automated and works faster within a short piece of sequenced DNA. In this paper, modified version of Clustering and KMP algorithm is used to search a specific pattern in a large DNA dataset. Overall process also includes the total number of matching pattern found in DNA dataset.Item Real Time Heart Disease Monitoring System(East West University, 2018-05-05) Hosen, Md. Solaiman; Mahmud, HasanDuring the recent decade, rapid advancements in healthcare services and low cost wireless communication have greatly assisted in coping with the problem of fewer medical facilities. The main purpose of this research work is to develop a wireless sensor network system that can continuously monitor and detect cardiovascular disease experienced in patients at remote areas. One of the most prevalent healthcare problems today is the poor survival rate of out-of-hospital sudden cardiac arrests. The Objective of this study is to present a Wearable Body Area Network System to continuously capture and sent the ECG signal to patient’s Mobile Phone. By analyzing the signal critical situation will be identified and alert will be sent to doctor, relatives and Ambulance services using data processing algorithm implemented on patient’s mobile phone. A wireless transmission system is also proposed for continuous data transmitting to a server system where a doctor can monitor the patient Electrocardiography (ECG) from a long distance. In this project we developed a wearable ECG device and a real time Brachycardia, tachycardia, and sinus arrhythmia detection based android mobile application. ECG signals from patient’s body is collected by the mini ECG device and sent through a Bluetooth module to Android Mobile Application. On Android application processed data analysis based Pan Tompkins algorithms to detect complex QRS ECG signal and heart beats. From the number of heart rate can be detected abnormalities. Upon completing the system, we tested the system using signals generated by Fluke PS400 and real data. There are three categories of abnormalities under study: Brachycardia, tachycardia, and sinus arrhythmia. Normal heart signal is also included in the test. We have tested this application in real time by collecting the ECG from the patient in stationary as well as simulated data. In both situations the application fulfills requirements of the proposed system.Item Cost and Energy Efficient Real Time Location Tracking Android Application Framework(2017-08-08) Sarkar, Saikat KumarObject tracking systems plays a vital role in monitoring and surveillance systems. The main target of location tracking systems is to provide object locations at real time with exact accuracy using personalized setup and location sensitivities. Shipping industries at first developed tracking systems to determine ship location at a given time. But those location data used to be stored in local storage which is accessible only when the ship is arrived. But with the flow of time and advancement of technology, now we want to know object location at real time. As the local storage cannot provide real time data, we moved to server based systems for real time data parsing. Today we have smart devices like android phones that can detect location data by itself. But to share and monitor those data we need an online system. Here comes the server based location tracking system. But many factors like battery consumption, data load cost, accuracy of data make this system a bit complex. We have tried to solve these problems and improve existing systems by proposing an android based framework that uses cloud services for real time data storage and processing and firebase cloud messaging for device to device communication at real time. We also propose a cost and energy efficient framework for android devices for location tracking and data transmiting.Item eHut Online Shopping(East West University, 12/17/2017) Gaffar, AbduleHut is an e-commerce web application that allows user to select products to buy from a range of product available for purchasing in the store. It also includes shopping cart and online bill payment facilities for the customers. The site administrator can manage the products and user roles like, subscribers and administrators. Beside its core features, this application provides a better way of creating list of customer email addresses to send promotional emails to the with the information of new products based on their purchasing behavior. This would ensure more effective customer responses by allowing the emails to be sent to appropriate customers.Item IoT Based Real Time Monitoring and Automated Agricultural Storage System(East West University, 12/31/2017) Haque, S. M. Mominul; Hossain, Muhammad Ramzaniv Abstract In this project, an IoT based smart storage system is developed for agricultural purpose in order to monitor and maintain storage environment both manually and automatically. Data is stored in MySQL database and the communication with the sensor data and web server is established with the help of Ethernet shield for Arduino. The project is supported by both website for computers and android devices via android app which increases its mobility. Monitoring and controlling can be done from website or android app. The web application also developed considering privacy and security issues so it makes it more secured. The system is tested on current environment and result was as expected.Item AN IOT FRAMEWORK FOR SOIL BORNE DISEASES DETECTION SYSTEM(East West University, 12/13/2017) Hasan, Amit; Ahmmed, FiruzNow a days Internet of things (IOT) is pairing up with our daily life. The Internet of Things helps to peek into the virtual world of digitization where every smart device, every sensor is connected with the internet and therefore can be accessed all over the world. This paper propose a secure IoT framework with multiple sensor nodes that sends data to the web cloud server and allows users to access those data via a mobile application through a unique user id and password or API’s. In this paper, we have built an IOT framework for detecting soil borne diseases SBDDS by using pH, temperature and moisture sensor. Soil borne diseases can create serious damage without showing any visible symptoms. Our device can sense the soil condition through the sensors and send data to the web cloud. After that our own build android application can process those various data and based on classification algorithm can generate result whether the soil contain disease or not. Among these factors pH value is the most sensitive and important factor for detecting exact disease. Different type of pH value represent the health of soil and present of bacteria, fungus or nematode. When soil is affected by these then the pH, moisture and temperature also get changed and create damage to the seeds, roots and soil nutrition. So this framework is highly efficient to detect the soil borne disease and prevent the damage.Item IoT Based Information Collection System of Victims From Crash Sites(East West University, 12/16/2017) Hossain, Md. Iqbal; Ghose, RajibIn this project, an IoT based information collection system of victims from crash sites is developed in order to minimize the death toll of the victims who are still trapped alive in the crash site through capturing images and sending the images continuously to cloud storage and also taking necessary information through various sensors and send the information to web server. The communication between the web server and the hardware is established via motion, sound, humidity and temperature sensors and another connection is established between hardware and cloud storage via camera. The project is developed for real time data fetching from hardware by sensors and camera and storing the information in web server and cloud storage which are used by the rescuers to have the exact information of victims in order to save their lives in time.Item Remote Health Monitoring For Critically Ill Patient(East West University, 12/17/2017) Sultana, Farhana; Nila, Maria Yasmin; Islam, NahidaIt’s our fundamental right to get quality Health Care. But now a day’s due to inadequate number of well qualified physician and vast number of population growth it has become very difficult to provide quality Health care services to everyone. So, in this paper we proposed a portable health care device that can be used as wearable or one can easily carry it with themselves that can be used to continuously monitor that patient’s bio status. Our proposed system monitors a patient’s oxygen saturation, blood pressure, blood glucose, body temperature and skin moisture and sends it to cloud. So that Patients with critical health condition can be remotely monitored by their preferred doctor from anywhere. In case of any emergency that doctor and hospital authority will be notified to prevent any loss of life.Item Mobile Target Tracking Using Modified Leach-r Protocol In Wireless Sensor Networks(East West University, 12/7/2017) Islam, A. K. M. Shamiul; Bristy, Farhana Nasrin; Alam, SamiraWireless Routing in Wireless Sensor Network (WSN) is usually an important area of research. Rapidly increasing applications of sensor nodes in monitoring of physical environment has led to an increased significance of WSN. Modified LEACH-R introduces the energy efficient clustering scheme for Wireless Sensor Network which is dependent on Low Energy Adaptive Clustering Hierarchy structure and removed many of the limitations of the traditional LEACH protocol. Modified LEACH-R uses the remaining energy of the current cluster head which makes the process of routing better and efficient. In heterogeneous network, a cluster head node aggregate and collects data from their member nodes and send that aggregated data towards the base station. The proposed protocol uses advanced nodes. The role of advanced nodes is to act as cluster head and to perform the task of data aggregation while consuming low power during operation. Modified LEACH-R protocol is tested and simulated in MATLAB software package. This protocol increases the frequency of packets that can be transmitted between base stations to cluster heads and between cluster head and nodes. The findings of the simulation revealed that Modified LEACH-R provides better results as compared to traditional LEACH protocol. Modified LEACH-R protocol enhances the time of first dead node and as well as increases the duration of operation of the last node and hence enhances the overall network lifetime.Item Analysis of Feature Tree using Miranda(East West University, 12/31/2017) Badhan, Ferdous Hussain; Islam, Md SharifulFeature models are used to specify the variability of software product lines. We study a special subset of trees, called generalised feature trees, and show how they can be used to compute properties of the corresponding software product lines. We introduce our paper the concept of generalised feature trees, which are feature trees where features can have multiple occurrences. It is shown how an important class of feature models can be transformed into generalised feature trees. We present algorithms which, after transforming a feature model to a generalised feature tree, compute properties of the corresponding software product line. We discuss the computational complexity of these algorithms and provide executable specifications in the functional programming language Miranda.