Theses, Dissertations & Reports

Permanent URI for this community

http://dspace.iiuc.ac.bd/handle/88203/66

Browse

Now showing 1 - 20 of 327

A Transfer Learning Approach with Modified VGG16 for Driving Behavior Detection in Intelligent Transportation Systems
(International Islamic University Chittagong, 2024-06) Nur, Abdullah Hafez
The complex nature of human behaviour and the numerous factors contributing to distractions on the road, particularly in terms of the limited ability to provide precise and timely early warnings, are addressed in this paper through the introduction of a novel method. The proposed method aims to enhance the detection of distracted driver behaviour using modified CNN transfer learning. By employing transfer learning within the Convolutional Neural Networks framework, the method seeks to tackle the intricate challenges associated with identifying distracted driving behaviours. The experimental results demonstrate an overall accuracy of approximately 99%, with the highest achieved accuracy reaching 99.46% on a publicly available dataset. This highlights the effectiveness of the proposed approach in significantly improving the precision of distracted driver behaviour detection
An Improved Positioning System for 6G Cellular Network
(International Islamic University Chittagong, 2024-02) Sifat, Md. Zahidul Islam
Information about the whereabouts of oneself or our cell phones has always been crucial to humans. Technological advancements have supported us across various domains, spanning from interior environments to outdoor global positioning systems. However, achieving real-time indoor positioning has remained a formidable task. The requirement for enhanced positioning has gained prominence in the context of the emergence of 6G cellular networks. The introduction of novel radio technologies, characterized by reduced end-to-end latency, specialized control protocols, and increased processing capacity at the network edge, has opened doors to harnessing the 6G cellular network’s full potential for precise localization in indoor and outdoor settings. Within the 6G cellular network context, this study successfully implemented the classic signal fingerprinting approach using the Received Signal Strength Indicator in a combination with machine learning algorithms. Consequently, it proposed an improved positioning strategy designed for indoor localization scenarios in the next 6G cellular network environment. Using this method, with two datasets, the mean error distance of 0.08025523, meaning an accuracy of 99.92% and 0.10692412 was achieved, which is 99.89% accuracy respectively.
Analysis of a S-shape Metamaterial Absorber Operating at 3.5 GHz
(International Islamic University Chittagong, 2024-06) sajjad, Sajjadul Alam
Analysis of a metamaterial absorber with an S-shaped structure is suggested to reduce the specific absorption rate (SAR) at 3.5 GHz. This absorber exhibits single negative metamaterialcharacteristics and achieves an absorption rate of over 90% for all electromagnetic polarization angles. In the age of 5G, numerous applications are functioning at various frequency ranges, emitting substantial amounts of electromagnetic energy (EM) that possess an elevated SAR value. This radiation has an impact on the human body, specif ically the head and hands, which are exposed to it. In Malaysia, a frequency band of 3.5 GHz has been designated for 5G communication. However, this has resulted in the emission of undesirable signals due to high Specific Absorption Rate (SAR) values. To address this issue, a proposed absorber would beimplemented to reduce SAR levels.
analysis of the impacts of load shifting dependent dsm strategy in a neighboring locality in bangladesh
(Department of Electrical and Electronic Engineering, 2022-06) REZA, MD. SHAMIUN; ISLAM, MD. RYHANUL
In recent years, end users management of power usage has been given more importance to ensure efficient and effective use of electricity. One such tool, that can improve the end user power use scenario is Demand Side Management (DSM). The main theme of DSM is to reduce power wastage, improve efficiency and to reduce the price of electricity to the consumer. In this work,We have investigated the effects of residential customers’ load shifting compared to their historical patterns of energy consumption. We have followed the data collection methodology of a similar study to take realistic load profile of the consumers. We have collected 10 individual house data from different locations to keep variety in the load profiles of the consumers and to find out an overall scenario. Initially, we have used an excel spreadsheet to take consumer data then we made simulink model for each user. The simulink model has been used to simulate the load shifting and its impacts. From the study we found that 6.7% of the total load can be shifted from peak electric hour to off peak hour. This study can make an important role in applying load shifting among mass consumer and to examine its impacts.
Analysis the Performance of mm Wave Microstrip Patch Antenna (MPA) with Array for 5G Applications
(Department of Electronic and Telecommunication Engineering, 2022-11) Arfath, Md. Yeasin, T181059
The evaluation of Microstrip antenna performance has changed from a simple free space evaluation, where things like efficiency, gain, bandwidth, and matching were looked at, to a much more complicated process that takes into account the end-user and real-world environments. This process of evaluating is still going on, and both academics and industry people are coming up with better ways to evaluate antennas. This thesis gives new ideas about how to improve antennas parameters. Microstrip antennas have become a part and parcel of today’s mobile communication world due to their low cost, low profile and ease of fabrication in the circuit boards. However, poor performance, such as low power handling efficiency, limited bandwidth, low gain, etc. The next 5th generation (5G) application’s communication will experience significant route loss since high frequency bands will be employed. So, a high gain antenna is necessary for this issue. Since microstrip antennas are small, they are a great choice for 5G technology applications.Therefore, the primary goal of this research project is to create a high-gain & high efficient antenna using a ambient microstrip patch antenna. The antenna operates on the 28 GHz band and was built and tested using CST Microwave Studio. Laminate material is used as a surface close to 2.2 with relative permittivity. Analysis the performance by implementing the MIMO and array. The gain, bandwidth, return loss, VSWR and efficiency of the 1x4 array antenna designed is 9.32 dBi, 9.5 GHz, 34.55 dB, 1.03 and nearly 94% respectively.
Analyzation of Q-factor for Free Space Optical Communication System Using Neural Network
(Department of Electronic and Telecommunication Engineering, 2021-10) Mohammad Nur Uddin Sikder, T171023; Lokman Hossen, T171039; Fahim Ikbal Sakib T171045
Analyzation of Q-factor for Free Space Optical Communication System Using Neural Network There are many options for optical fiber communication system configurations according to user satisfaction, but to establish this system it is essential to simulate the system to avoid unexpected output. To predict the output, among other mathematical tools, many researchers found the neural network is remarkable and one of the most promising methodological approaches to perform network-data analysis and enable automated network self-configuration and fault management. In this research paper, we proposed a neural network to perform optical communications and networking analysis, summarize and compare some classical research using the MATLAB tools. For training and test data, as the maximum quality factor can be achieved using Gaussian optical filter compared to different optical filters simulated by optical communication system design software product by the optical-wave team, we’ve used the Gaussian optical filter throughout the procedure. We also have used different data rates and diameter aperture. And the system performance is evaluated using the BER curve, Q factor, eye diagram, and spectrum analyzer output
An Approach to Examine Blood Glucose Ramification on Human Health by Assembling Machine Learning Classifiers
(Department of Electronic and Telecommunication Engineering, 2022-11) Roy, Arnab
Approach to Improving Machine Learning Models for Intrusion Detection System
(International Islamic University Chittagong, 2022-07) Labib, Ahmad Ibtisam; Chy, Shamsuddin Ahmmed; Hossain, Md. Shahriar
In today's digital world, there are several security risks that digital assets must contend with. Systems for detecting intrusions (IDS) are essential security tools that protect digital assets. But their usefulness depends on meeting strict accuracy requirements, and their effectiveness depends on timely alarms. This study offers a novel IDS model that combines deep learning and machine learning methods as a solution to these problems. The study applies several classification techniques, such as Gaussian Naive Bayes (GNB), Random Forest (RF), Decision Tree, K-Nearest Neighbors (KNN), Soft Voting, and Hard Voting, using the well known KDD Cup-1999 dataset. After a large-scale dataset was processed, the Decision Tree method performed better than the others, with a 99.9% accuracy rate. This study aims to investigate the effects of soft voting and hard voting, a novel application in IDS. Decision Tree proved to be the better performance in spite of these efforts. By offering information about algorithmic efficacy, the research advances the field of intrusion detection and helps decision-makers in the design and deployment of intrusion detection systems. These findings have implications for improving digital asset protection against changing cyber threats.
An Approach To Perform Comprehensive Wind Turbine Performance Analytics By Means Of Machine Learning (Sci-Kit Learn)
(Department of Electrical and Electronic Engineering, 2022-06) ZUBAIR, MUHAMMAD; MASUD, FAISAL IBNE; YEAKUB, MAHAMMAD
With the present day’s emphasis on sustainable and secure energy supply, wind power sector is growing rapidly all over the world. Along with the rapid expansion of the wind turbine sector, the wind turbine industry is also growing. Understanding the power response of these systems to the variations in wind velocity is essential for the optimal selection and efficient management of these turbines. This is defined by the power analysis and prediction of wind turbines output. Previous approaches such as mechanical aspects of a turbine are used to work with wind turbine power output. However, these techniques are not capable of analyzing big dataset with hefty number of parameters. In this regard, Machine Learning is a handy tool to analyze a dataset and design a model that can perform prediction related to turbines parameters. Python is one of the Open Source convenient tools to implement ML tasks. In this thesis work, we proposed Scikit Learn based Machine Learning models, which is based on Python for the power analysis. Three different machine learning methods such as XGBoost, LightGBM, Catboost were used for the modeling. The comprehensive dataset based on Western Wind Firms are being used to design the analysis model. The accuracies of these models are validated by estimating the error between the model output and the field observations from these turbines from the comprehensive dataset.
Bachelor Of Science In Electronic And Telecommunication Engineering
(Department of Electronic and Telecommunication Engineering, 2022-11) Alamgir, Md Khobaib
The perception of Sensor Technology for plant advancement and natural organization is a promising new field of research. This projet an arrangement of reliably interfaced sensors is used to support data went for giving more profitable plant improvement and an unrivaled circumstance. In this part, we present a system where eight sorts of sensors are used to measure the air and soil quality. Our structure utilizes dispersed capacity for keeping the assembled sensor data which by then gets organized on the web in order to make exact figures on nature and plants using an auto-in reverse facilitated moving typical count. Likewise, the structure has been arranged with a web interface and data portrayal, engaging people to obtain the persistent normal information to take better decisions for plant advancement and biological organization
Bachelor Of Science In Electronic And Telecommunication Engineering
(Department of Electronic and Telecommunication Engineering, 2022-01) Nayem, Tanvir Mohammad; T173002; Rakibul Islam, T173035; Mohammad Yesin, T173038
With the evaluation of technology, medical science is no longer static. We have seen many important electronic devices that help a lot in the patient health observation and operation monitoring system. Patients are in danger of dying unexpectedly owing to a specific cause of heart difficulties and attacks, which is caused by a lack of proper medical care provided to patients at the appropriate time. An ICU patient needs proper observation. Mostly they are being observed on basis of ECG, Blood pressure, body temperature, oxygen saturation, and diabetes sugar level. Only ICU patients get access to such amenities. But we are going to develop a system in which a patient outside of an ICU or from home can also be under these observations remotely with the help of IoT. The data gathered from the patients will be automatically stored on a server, which can subsequently be accessed via the app. Even if the doctor is a significant distance away, an IoT health monitoring system can help with disease prevention and precise diagnosis of one's present state of health
Bandwidth Enhancement of a Microstrip Patch Antenna for 5G Applications,
(Department of Electronic and Telecommunication Engineering, 2022-01) Salauddin T173016; Amin Hasan Mahmud T173018; Abir Roy T173025
The main disadvantage of a Microstrip Patch antenna is its narrow bandwidth. This research aims to improve the bandwidth of a microstrip patch antenna (MPA) for 5G applications. A variety of factors such as substrate element, substrate height, feeding method, and so on, which have an impact on the antenna's performance. If those parameters are chosen correctly, the 5G application requirements will be met. This work defined MPA's expanded bandwidth, which will be favorable for 5G applications. The insertion gap concept is used to extend the bandwidth of a microstrip patch antenna in this research. It would also be possible to attain a bandwidth that matches the demands and expectations of 5G applications by adjusting the size of the insertion gap. Finally, a single microstrip patch antenna operating at 28 GHz with a center frequency of 28.075 GHz has been designed. The suggested antenna's gain, bandwidth, return loss, VSWR, directivity, and efficiency are 5.83 dB, 928 MHz, - 41.282 dB, 1.017, 7.67 dBi, and 66.7 percent respectively
“ BANGLA TEXT CLASSIFICATION THROUGH MACHINE LEARNING ALGORITHM”
(International Islamic University Chittagong, Department of Computer Science and Engineering, 2023-07) Tarek, Ashraf Uddin; Chowdhury, Satirtha; Hossain, Mosharaf
The increasing number of social media users and e-commerce platforms have a great impact on people's daily life. People share their emotions and opinion through the social media platform. These emotions and comments now take the important place of analysis because based on these emotions, opinion, business farm make them plan what will produce, consumer decide what will he/she buy etc. Lots of work have been carried out to analysis the sentiment or emotion in English language. Due to the complexity of the Bangla language, few work has been done on it but in recent years several researchers have been carrying out various research based on the Bangla language. In this paper we conduct sentiment analysis (SA) on Bangla language by using the machine leaning model (ML). Most of the work basically divide the sentiment into three categories in this paper we divide the sentiment into four categories namely strong positive, positive, negative, and strongly negative. In this paper we use logistic regression (LR), decision tree (DT), Random Forest (RF), linear support vector machine (LSVM), confusion matrix, and kernel support vector machine (KSVM) algorithm of machine learning (ML). From support vector machine (SVM) we mainly used gaussian kernel radial basis function (RBF). The sentiments are converted to NumPy array to use the sentiment in machine learning. Since the NumPy array is numeric we train our model by these data to get the proper prediction about any given sentiment whether that sentiment positive or negative. The data used are all raw data or primary data collected from the different microblog websites and social media platforms. The total number of raw data 10851 most of them collected from Facebook and YouTube due to their popularity in our country some of the data collected from twitter as well. All the ML model applied for the single word of the sentences first, known as unigram feature analysis best on the single word the logistic regression model and RBF SVM provide the highest accuracy 71.26% and 71.91% respectively. By using two words each model works almost like unigram models, in bigram models LR again shows the highest accuracy, but the accuracy level little bit dropped for RBF SVM. The accuracy for LR and RBF SVM 72.04% and 68.79% respectively. Later we used models for three words defined as trigram feature analysis in that time get highest 70.87% accuracy for LR. Most of the papers basically use SentiWordNet to assess the polarity of the sentiments but in this paper, we use word by word analysis which hardly 6 seen to any paper. This paper will help the business farm as well as the consumers to make their decision and will work as guideline for the new researcher in this topic.
A Blockchain-Based Smartgrid Monitoring, Protection, And Electricity Trading Scheme
(Department of Electrical and Electronic Engineering, 2020-12) DAS, SOURAV; CHOWDHURY, PLABON
The electricity demand is at its highest peak with the rapid growth of modernization. To mitigate the ever-increasing demand for electricity, microgrid becomes quite popular in recent days. In nature, the traditional power market is unidirectional, i.e. power plants provide the consumers with electricity that contributes to centralization and monopoly. With the emergence of prosumers, microgrid is popular with its bidirectional configuration of energy supply. But the challenges also arise in terms of microgrid monitoring and trading mechanism. Furthermore, the traditional model of microgrid becomes vulnerable prey to cyber-attack. Incorporating blockchain can address all those challenges with minimal effort. Blockchain is an immutable, secure data management technology that is decentralized in nature. This thesis presents a triple-layered blockchain model constructed for real-time microgrid monitoring. A blockchain-enabled differential protection system is also designed for virtual protection of the grid in the event of physical relay failure. The protection model also prevents false data injection and cyber threats to the microgrid. To facilitate trade between prosumers, algorithms are designed to form coalitions and negotiate electricity trading. We have created a three-layered blockchain using Python. Also, we have designed a simulink model for blockchain-enabled monitoring and protection scheme using MATLAB. Our proposed thesis work (Smartgrid) itself is an application. It can not only implement on Smartgrid but also implement on Microgrid, low voltage DC Grid and all types of Power Grid.
Book Valley [An Online Web-Based Book Publication & Selling System]
(International Islamic University Chittagong, 2022-07) Chowdhury, Md. Hayat Hossain; Hoque, Md. Sohidul; Nadim
Nowadays most of the work, entertainment facilities, and education is done by digital tools. Now we are living in a digital world where time is so valuable, that people want to lead hassle-free life. For a hassle-free centralized Online Bookselling flat form (Book Valley) for bookworm people and businessmen. This system gathers readers, writers, and publishers in one flat form. Readers can find a huge amount of books from anywhere on the earth and carry all the books with them all the time wherever they are. The E-book system is a blessing for book reader because it saves their time, carrying hassle, and storage costs. A writer can contact and send their creation to any publication over the world. Mostly new writers can easily reach the publisher for their desired publication. Book Valley provides e-books for readers for that reason the cost of publication will decrease and publishers can easily publish their books and sell them in online stores. By utilizing Book Valley’s features as a centralized system we can get a hassle-free book ecosystem where readers, writers, and publishers will benefit equally.
Cancer prognosis & Survival rate prediction using machine learning algorithm.
(Department of Electronic and Telecommunication Engineering, 2022-10) Alif, Borhan Uddin
Roughly 10 million deaths, or nearly one in six deaths, were caused by cancer in 2020, making it the top cause of death globally. Breast, lung, colon, rectum, and prostate cancers are the most prevalent types of cancer. Any disease that can affect any region of the body is referred to as cancer. Neoplasms and malignant tumors are other words that are used. One characteristic of cancer is the quick development of aberrant cells that expand outside of their normal borders, infiltrate other body components, and eventually move to other organs. This process is known as metastasis. The main reason why cancer patients die is because of widespread metastases [1]. The research was held on two different datasets of breast cancer. To train our model initially we used a primary data set from UCI which includes 569 instances with several attributes [2]. This was for cancer diagnosis prediction/detection. Then we used another dataset that includes 4024 instances with 14 different attributes which we mainly used for death rate prediction with total analysis [3]. Here 6 algorithms were incorporated including KNN, Random Forest, J48, etc., and their results were compared.
A Compact and Efficient mmWave Microstrip Patch Antenna for 5G Communication
(Department of Electronic and Telecommunication Engineering, 2022-12) Ahmed, Shahriar, T-181052
Several well-known advantages of microstrip patch antennas are their low profile, affordability, light weight, simplicity of manufacture, and conformance. Microstrip antennas are becoming a common fixture in the field of wireless communication because of their low cost, low profile, and ease of manufacture on circuit boards. The use of high-speed millimeter wave (mm Wave) wireless technology in high-speed and secure data transfer has become commonplace. A millimeter wave microstrip patch antenna and related array are proposed for 5G applications. The Rogers RT-Rogers 5880 substrate, which has a relative dielectric constant (∈ r) of 2.2, is used to make the 5G Microstrip patch. The antenna's resonance frequency is 27.38 GHz, it has a 2.4 GHz bandwidth, and a return loss of -58.84 dB. The antenna can be utilized for 5G mobile communication due to its tiny size. The proposed antenna has better gain and directivity, offers excellent VSWR and efficiency, as well as having a reasonable enough bandwidth at the resonance frequency under consideration, according to comparative research that was also carried out. The proposed antenna's efficiency is 90%
A Compact Multiband Antenna for Sub-6 GHz 5G mMIMO Applications
(Department of Electronic and Telecommunication Engineering, 2022-11) Bhuiyan, Md. Emdadul Hoque
In this paper, we have designed a compact antenna for massive MIMO applications. The design has been carried out using various parameters such as left-hand tweak and right-hand tweak. The simulations were carried out with different hands simultaneously. The design and simulation of the proposed MIMO antenna are performed in the CST Studio Suite. The antenna is designed at an operational frequency of 2.6 GHz with a dielectric substrate and a conducting ground plane (W×L=19×38.2 mm2). The material of the substrate is FR-4 (lossy), with a relative permittivity of 4.3, a height of 1.6 mm, and a loss tangent of 0.025. The material of the ground plane is copper (annealed) with a thickness (t) of 0.035 mm, and the ground plane is cut into a smaller one to achieve a broader bandwidth. The optimized dimensions of the antenna are Wp=17.17 mm and Lp=22.13 mm. The width of the feedline (Flow) connected to the patch is 2.42 mm, and the length (Fl) is 14.93 mm.
A Compact size mmWave multiband microstrip patch antenna design for 5G communication”
(Department of Electronic and Telecommunication Engineering, 2022-11) Bhuiyan, Mohammad Abu Sayeed
A “+” shape slotted tri-band microstrip patch antenna for 5G mm-Wave & V-band application is designed and addressed in this paper. The designed antenna can be performed 5G mm-Wave and V-band applications. The proposed microstrip patch antenna has a compact size rectangular shape structure of 5.02 mm  4.03 mm with a simple geometrical structure. Rogers RT 5880 (lossy) with a dielectric constant of 2.2 is used as substrate in this antenna with a thickness of .499 mm. The designed antenna covers multi-band operation. It resonates at 36.72 GHz, 41.73 GHz & 49.89 GHz mm-Wave bands with gains 6.615 dB, 6.017 dB & 5.752 dB respectively. The preferred antenna is designed with the benefits of low cost, compact size, good gain & efficiency by using microstrip feeding techniques
A Compact Wideband Circularly Polarized Patch Antenna for 5G Applications
(Department of Electronic and Telecommunication Engineering, 2023-07) Mahmud, Shakil
The Internet of Things, or IoT, is a network of mechanical and digital computing devices that are networked and have the ability to send data across a network without the need for human or computer involvement. IoT devices are being used more often every day. The addition of 5G technology will increase the responsiveness of IoT devices. As a consequence, a lot of study has been conducted, with researchers offering their ideas for a good antenna design. However, it is challenging for wireless communication experts to create small, wideband antennas that can concurrently cover all of the new 5G radio frequencies. This thesis paper reveals a wideband antenna that can simultaneously cover all of the 5G radio frequencies. The proposed antenna's dimensions for mm wave operation at 28GHz are 8.066×6.96×0.8 mm3 . Rogers RT5880LZ, which has a thickness of 0.8 millimeters, is the substrate that is being utilized. The 3.2 GHz bandwidth of 28 GHz corresponds to an obtained gain of 4.484 dBi. The resonance reflection coefficient of the suggested antenna is -22.25 dB. CST Microwave Studio 2018 was used to create and test the suggested antenna. The simulation results are suitable for common 5G applications.

Browse

Browsing Theses, Dissertations & Reports by Title

Results Per Page

Sort Options