International Conference

Permanent URI for this communityhttp://dspace.iiuc.ac.bd/handle/88203/27

Browse

Now showing 1 - 20 of 39

11th International Conference on World Pease and Security-Role of Islam (ICWPSRI)
(Faculty of Shari'ah and Islamic Studies, 2017-12-29) Shaoque, Shaker Alam
Adopting Factors of Electronic Human Resource Management: Evidence from Bangladesh
(IEEE, 2016-10-28) Masum, Abdul Kadar Muhammad; Alam, Md. Golam Rabiul; Alam, Mohammed Shamsul; Azad, Md. Abul Kalam
The incorporation of information technology (IT) instincts the legacy of human resource management (HRM) towards agile HRM. To achieve agility, this paper explores the factors or determinants inducing the organisation decisions to adopt electronic human resource management (eHRM) in organisations of Bangladesh through structural equation modeling (SEM) of data science. To realize the influencing determinants, a research model was developed based on technology-organisation-environment (TOE) model. A total number of 320 respondents were participated from 48 organisations in Bangladesh using simple random sampling. The SEM results indicate that perceived compatibility, perceived cost, top management support, organisational culture, centralisation, IT vendor support, and government support have significant influence on management decision of e-HRM adoption. The applied implication of the findings and the scope of future studies are deliberated at the end of this paper.
Analysis of Charge-Shared Matchline Sensing Schemes and Current Race Scheme in High-Speed Ternary Content Addressable Memory (TCAM)
(IEEE, 2016-10-28) Kadir, Raqqib Bin; Anwar, Ahmed Selim; Haque, Muntasim-Ul
Due to emerging of hi-speed communication larger TCAMs with higher speed is needed. A comparative analysis of different Matchline sensing schemes in high speed Ternary content addressable memory (TCAM) is presented in this paper. With the conventional current race scheme, two different methods of charge sharing matchline sensing schemes are being analyzed. The power is distributed along the matchline by dividing it and sharing charge. This two charge shared schemes also improve search time and voltage margin. Simulations are performed using 180nm 1.8V CMOS logic in HSPICE.
Automated Weather Event Analysis with Machine Learning
(IEEE, 2016-10-28) Hasan, Nasimul; Uddin, Md. Taufeeq; Chowdhury, Nihad Karim
Weather forecasting has numerous impacts in our daily life from cultivation to event planning. Previous weather forecasting models used the complicated blend of mathematical instruments which was insufficient in order to get higher classification rate. In contrast, simple analytical models are wellsuited for weather forecasting tasks. In this work, we focus on the weather forecasting by means of classifying different weather events such as normal, rain, and fog by applying comprehensible C4.5 learning algorithm on weather and climate features. The C4.5 classifier classifies weather events by building the decision tree using information entropy from the set of training samples. We conducted experiments on LA weather history dataset; from evaluation results, it is revealed that C4.5 classifier classifies weather events with f-score of around 96.1%. This model also indicates that climate features such as rainfall, visibility, temperature, humidity, and wind speed are highly discriminative toward events classification.
Bio Inspired Cyber Security Architecture for Smart Grid
(IEEE, 2016-10-28) Faisal, Muhammad Mostafa Amir; Chowdhury, Muhammad Ariful Islam
Smart grid is an advanced and intelligent form of conventional power grid with high fidelity power-flow control, self-healing, and energy reliability using advance computer and communication technologies. The idea of integrating power systems with complex computer communication has introduced serious cyber security concerns as it requires significant dependence on secured communication infrastructures. Because of the wide nature of smart grid, it is very risky to compromise any component of the grid which may lead to serious damage to the electrical infrastructure, energy theft and unnecessary expenditure. Studying various smart grid security architectures, efficient bio-inspired complete security architecture for smart grid is proposed in this work which can be easily implemented in the smart grid without changing any element of its infrastructure. Bio-inspired cyber security architecture for smart grid is developed with discussion on the core elements of the architecture. The threat detection is tested with simulation using popular KDDDump dataset and real time gas pipe line SCADA data.
Bomb disposal Robot Discarding explosive through wireless controlled method
(IEEE, 2016-10-28) Motaleb, Abdul Kadir Bin; Hoque, Mohammad Busayeed; Hoque, Md. Ahsanul
Bomb disposal robot has been developed by different experts around the world to make an affordable and safe device which will be useful for emergency rescue support. A 6 DOF articulated robotic arm mounted on a moveable base is been developed to help bomb disposal squad to dispose bomb safely from distance. The whole operation of the robot can be performed wirelessly from a computer. DC servo motor is used as actuator of the arm, and the servo is controlled by the PWM signal generated by microcontroller. This report is concerned with the mathematical modeling of a 6 DOF robotic arm along with the methodology of the entire prototype development. Architectural and circuitry development are separately deliberated. Lastly, the performance of the developed robot has been discussed.
Combining a Rule-based Classifier with Weakly Supervised Learning for Twitter Sentiment Analysis
(IEEE, 2016-10-28) Siddiqua, Umme Aymun; Ahsan, Tanveer; Chy, Abu Nowshed
Microblog, especially Twitter, have become an integral part of our daily life, where millions of user sharing their thoughts daily because of its short length characteristics and simple manner of expression. Monitoring and analyzing sentiments from such massive amount of twitter posts provide enormous opportunities for companies and other organizations to learn about what user think and feel about their products and services. But the ever-growing unstructured and informal user-generated posts in twitter demands sentiment analysis tools that can perform well with minimum supervision. In this paper, we propose an approach for sentiment analysis on twitter, where we combine a rule-based classifier with weakly supervised NaiveBayes classifier. To classify the tweets sentiment, we introduce a set of rules for the rule-based classifier based on the occurrences of emoticons and sentiment-bearing words, whereas several sentiment lexicons are applied to train the Naive-Bayes classifier. We conducted our experiments based on the Stanford sentiment140 dataset. Experimental results demonstrate the effectiveness of our method over the baseline in terms of recall, precision, F1 score, and accuracy.
A Compact Patch Antenna for Ultrawideband Application
(IEEE, 2016-10-28) Islam, M. Tarikul; Samsuzzaman, M.; Islam, M. M.; Mahmmud, M.Z.; Islam, M.T.
A compact microstrip line fed tapered-shaped slotted ultra-wideband(UWB) antenna is presented. The new design is composed of a rectangular slotted patch with a compact size of 21.44 × 23.53 mm2 and fed by microstrip transmission line of 50 Ω. The proposed antenna is simulated onto a less expensive FR4 substrate which has a height of 1.6 mm. There achieved a wider bandwidth of 8.51 GHz (3.49-12 GHz) with good impedance matching, constant gain, 5.76 dBi of maximum gain and stable radiation pattern results in the proposed antenna more appropriate for using UWB communication applications compared to existing antennas. The proposed antenna is designed and simulated in HFSS and CST Microwave Studio software to see the antenna performance.
A Comparative Analysis among PWM Control Z-source Inverter with Conventional PWM Inverter for Induction Motor Drive
(IEEE, 2016-10-28) Islam, Md. Khurshedul; Zaved, Md. Mustafa; Siddiky, Air Mohammad; Mamun, Khandakar Abdulla Al
The uses of induction motor is ever-growing day by day because of its superiority and high efficiency over dc machine. So for wide range of use in the industry, this machine requires an efficient driver circuitry arrangement. Currently conventional Voltage Source Inverter (VSI) or Current Source Inverter (CSI) is dealing as key part in the field of induction motor driver circuit. These converters fail to perform at our desire level due to some crucial drawbacks like they can perform as either buck or boost operation and they contain considerable amount of harmonics as well as EMI noise. So we are trying to replace these traditional inverters by PWM control Z-Source Inverter (ZSI) which offers buck-boost operation capability by utilizing shoot through state and provides less EMI noise. This paper presents a comparative study among these three inverters. Firstly three MATLAB simulation model of Induction motor driver by using VSI, CSI and ZSI (having almost same environment) have arranged. Then some important motor characteristics such as rotor speed, electromagnetic torque, rotor current and stator current have observed for three system models at various loading conditions. The model using Z-source inverter exhibits efficient performance in all cases compared to the other models.
Design and implementation of a digital calling bell with door lock security system using fingerprint
(IEEE, 2016-10-28) Kader, M.A.; Haider, Md. Yousuf; Karim, Md.Rezaul; Islam, Md. Saiful; Uddin, Mohammad Mamun
The term ‘Home Security’ numerously growing as a major concerning issue in today’s life. The challenge of developing a home security device is to make it user friendly which can reduce human effort as well as ensure safety and security of people and their home. In this paper, a security device is proposed which is developed from the concept of conventional calling bell. The device analyzes the fingerprint of a person and matches it with stored fingerprint database to categorize the person as home member or known guest or unknown guest. When it detects a home member the entrance door automatically opens. In case, the person categorize as known guest, the device sounds a tune but don’t open the door automatically rather people inside home can open the entrance door from anywhere of the home. As guest is known people don’t need to walk towards door to open it. And when device detects the person as unknown guest, it generates a tunes which is different from previous one making aware the people inside that an unknown person wish to enter your home.
Design Evolution of Planar Slot Antennas for Ultrawideband Wireless Communication
(IEEE, 2016-10-28) Azim, Rezaul; Islam, Mohammad Tariqul
The design evolution of planar ultra-wideband slot antennas is presented in this paper. The basic antenna with a Tshaped radiator and ground plane with wide slot is able to exhibit UWB characteristics. Insertion of a curve slit in the front side of the basic antenna helps to generate a stop band at around 3.5GHz while etching of a pair of symmetric slits in the rear side helps to produce a stop band for WLAN. Dual stop band centered at 3.5GHz and 5.5GHz can be obtained by putting both types of slits together. The insertion of slits does not alter the size and shape of basic antenna which gives us an upper hand in the designing of UWB antenna with notch band/s.
Design of a low-cost lighting system for the rural areas of Bangladesh
(IEEE, 2016-10-28) Yasin, Mohammad Salman; Islam, Mohammad Sharier; Biswas, Tanmoy; Rahman, M. Saidur
This study was conducted to find a solution to solve the problem of deprivation of light present in the rural society of Bangladesh. More than 50percent of our rural population are deprived of electricity. This in turns results in absence of light in houses. A device was designed using bottles, Light Emitting Diode, solar panel and battery for an alternative lighting system. This device is not so costly and can be reproduced very easily. Different variations of this device was tested using a light meter (Model-LX 1102). A brief overview of this newly proposed lighting system is discussed in the study.
Design of an universal numeric segmented display and implementation of its decoder circuit in FPGA
(IEEE, 2016-10-28) Kader, M.A.; Hossain, Kazi Sahadat; Munir, Mohammad Serajum; Dhar, Rajvi Sutra; Rahaman, Md. Arifur
Segment display is very important in displaying numeric data. Although matrix display can be used to show any number with better font but it has some major disadvantages compared to segment display. Matrix display uses scanning techniques to show something, this technique is complex to implement and needs more memory. Here, a 20-segment display is proposed which can show numbers and mathematical symbols of 14 different languages. Also a decoder circuit is proposed which takes BCD input and generates output binary combination for 20-segment display of selected language. Finally a prototype of 20-segment display is implemented by LED’s and the decoder circuit is implemented in FPGA. The system is checked for 14 different languages and some mathematical symbols.
Design of N-Segmented Display system for Bengali vowels
(IEEE, 2016-10-28) Hannan, Saif; Shohag, Ahasan Ullah
Different segmented display systems have been proposed for representing Bangla and English alphabets and numerals in past few years. In order to distinguish Bengali Vowels in precise form, they must be represented with their unavoidable curved shapes. Accurate recognition of Vowels is difficult due to their complex shapes. In this paper, we tried to represent a total of 11 Bangla vowels in simpler and accurate look. To improve their accuracy, we have proposed a 58-segmented display. And we derived a very simple, scientific logic to find segments out of matrix display. We have introduced a 4-bit input line to represent all the Bengali Vowels, analyzed all the 58-sements against each and every Vowel, and designed the suitable circuit for each of the 58 segments.
Design, Fabrication and Performance Evaluation of a Three Electrode ECG Recorder
(IEEE, 2016-10-28) Chowdhury, Mehdi Hasan; Hossain, Quazi Delwar; Saha, Pradipta; Rahaman, Md. Mushfiqur
Frequent monitoring is required for chronically ill cardiac patients to ensure superior treatment. This paper presents a low-cost, novel wearable electrocardiogram (ECG) measuring unit for consistent use. In this study, the ECG signal is accumulated by a three-electrode data acquisition system and measured by an integrated signal conditioning block AD8232. The analog signal from AD8232 is converted to digital data and recorded on a computer using unique software developed for this study. This automated software can store electronic records of the patients. These patient particulars can be forwarded to a proper medical support system if needed. To confirm the quality of ECG recorded by this prototype, clinical tests are conducted. The results of these assessments demonstrate the viability and efficacy of this system.
Development of a telemedicine model with low cost portable tool kit for remote diagnosis of rural people in Bangladesh
(IEEE, 2016-10-28) Prodhan, Uzzal Kumar; Rahman, Mohammad Zahidur; Abid, Ahsin; Bellah, Mohtasim
In this paper we have developed a Telemedicine model with portable tool kit for remote patients to collect vital signs of patients which are used for Telemedicine services. This developed system is low cost, portable, and easily maintainable and can be integrated with any complex health system. We have used the GNU health where local doctors can communicate with a low cost terminal. Expert doctors can also take part through this terminal and deliver treatment to the patients. The patient’s medical history is stored in GNU health database and accessed from the remote terminal. We have successfully designed the system and collected the patient's data. Through our developed android apps, the data will be stored in the staging server. From the staging server, any health system can collect the data and give the services to the rural people. Finally we can conclude that, Telemedicine service can be given effectively by using our portable tool kit in a cost effective manner which improves the quality and accessibility especially in rural areas.
Effect of Acceptor Concentration on Performance of CdTe Solar Cell from Numerical Analysis
(IEEE, 2016-10-28) Noman, M.A.A.; M.S., Islam; Abden, M.J.; Arifuzzaman, M.; Islam, M.A.
In this study, the absorber layer of Cadmium Telluride (CdTe) ultra-thin solar cell was analyzed by using the one dimensional simulator Analysis of Microelectronic and Photonic Structures (AMPS 1D) software. A novel structure of ultra-thin CdS:O/CdTe solar cell is suggested in this research that emphases on conversion efficiency. The maximum conversion efficiency of ~ 26.24% (JSC = 27.05 mA/cm2, VOC = 1.2 V, FF = 0.886) is achieved with 600 nm CdTe absorber layer at the acceptor concentration of 5 × 1019 cm-3. The temperature effect is also examined with an aim to realize the environmental impact on performance of the cell. Our outcomes offer innovative research guidelines for resolving persistent challenges of CdTe photovoltaics.
Effect of Higher Carrier Injection Rate on Charge Transport and Recombination In Mixed-Host Organic Light Emitting Diode
(IEEE, 2016-10-28) Chowdhury, Riku; Haq, Md. Rashedul; Chowdhury, Md. Sarwar Uddin; Afrose, Sharmin; Paul, Sukanta
The effect of anode surface modification on performance of uniformly mixed-host (MH) emissive layer (EML) based organic light emitting diode (OLED) has been investigated by the numerical simulation. Indium Tin Oxide (ITO) used as anode in the device. Due to proposed surface modification technique of ITO, the energy barrier at anode/organic layer interface is reduced which consummately enhanced the hole injection rate that leads to balance of carriers transportation and recombination in the EML. Through the numerical simulation, the electrical characteristics and internal device physics of uniform MH-OLED have been analyzed quantitatively. Calculated current balance factor which is related to the external quantum efficiency also confirmed the efficiency enhancement of MH-OLED by the proposed ITO work function modification technique.
Effects of Buffer Size and Mobility Models on the Optimization of Number of Message Copies for Multi-Copy Routing Protocols in Scalable DelayTolerant Networks
(IEEE, 2016-10-28) Hossen, Md. Sharif; Ahmed, Md. Tofael; Rahim, Muhammad Sajjadur
In Delay-Tolerant Network (DTN), there exists a lack of direct path from source to destination, and it is also featured by extremely high bit error rate, unwanted delay, limited resource, etc. For such network, we have investigated the effects of buffer size and mobility models on the optimization of the number of message copies for multi-copy routing protocols such as Spray-and-Focus (SNF) and Spray-and-Wait (SNW). This investigation is essential for providing us with the insight on the number of message copies that would provide better delivery, lower latency and lower overhead considering the effects of buffer size and mobility models in scalable delay-tolerant networks. In this investigation, we have considered three performance metrics, namely delivery probability, average latency, and overhead ratio. Simulation is done using Opportunistic Network Environment (ONE) simulator, which is designed basically for evaluating the performance of DTN routing strategies. Here we have considered three mobility models, namely Shortest Path Map Based (SPMB) movement, Random Walk, and Random Direction. Simulation results show that for increasing buffer size and number of nodes on these considered mobility models for a particular value of number of message copies, indicated by L, SNF routing exhibits satisfactory performance, especially in the case of SPMB movement using only L = 2 copies. Since our purpose is to provide a satisfactory performance, i.e., higher delivery, lower latency and lower overhead, SNW routing shows overall good performance using L = 10 copies. Therefore, it would be a good optimization using Shortest Path Map Based movement model using L = 2 copies for SNF routing, and L = 10 copies for SNW in scalable DTN scenario, where there may be a high possibility of varying buffer size with number of mobile nodes.
Frequency Domain Linear Prediction-Based Robust Text-Dependent Speaker Identification
(IEEE, 2016-10-28) Islam, M. A.
Speaker identification is a biometric technique of determining an unknown speaker's identity among a number of speakers using distinguish latent information of uttered speech. Crime investigation, security control, telephone banking and trading, and information reservation are some applications of this technique. Frequency Domain Linear Prediction (FDLP) is a time-frequency-based feature has been derived using 2-D autoregressive model. This feature was constructed from sub-bands short frame energies estimation. FDLP has been used in this study to propose a robust text-dependent speaker identification technique. The clean features were used to obtain speaker behavioural model. Support vector machine has been used to train the proposed method. This presented study was tested in both clean and noisy conditions to validate the method extensively. The proposed method got significant improved performance over all traditional methods performances in noisy conditions. The obtained performance was indicated; the proposed method was very robust to noises and showed consistent performance irrespective to noises.