KUET Institutional Repository (Khulna University of Engineering & Technology)
Not a member yet
1045 research outputs found
Sort by
Efficient Mode Selection Scheme for Video Coding
No full textThis dissertation is submitted to the Department of Electrical and Electronic Engineering, Khulna University of Engineering & Technology in partial fulfillment of the requirements for the degree of Doctor of Philosophy, November 2016.Cataloged from PDF Version of dissertation.Includes bibliographical references (pages 149-159).Video coding plays vital role in communications and multimedia applications. The appearance
and development of various new multimedia services have need for higher coding
efficiency. H.264/AVC is the latest international video coding standard that can achieve
considerably highest coding efficiency than previous standards and uses rate-distortion optimized
(RDO) mode selection technique. This means that the encoder has to code the
video by exhaustively trying all the mode combinations including the different intra and
inter prediction modes. However, this mode selection process also makes the encoding
process extremely complex, especially in the computation of the rate-distortion cost function,
which includes the computations of the sum of squared difference (SSD) between
the original and reconstructed image blocks and context-based entropy coding of the block.
So, the highly complex mode selection process limits the practical implementation of codec’s.
This dissertation proposes six different fast and efficient schemes to reduce the computational
complexity of mode (Intra and Inter) selection process. It also can effectively reduce
the spatial and temporal redundancy. In H.264/AVC intra coding, DC mode is used to predict
regions with no unified direction and the predicted pixel values are same and thus smooth the
varying regions which are not well de-correlated. A modified DC prediction (MDCP) mode
is suggested to replace the existing DC mode based on the distance between the predicted
and reference pixels. By analyzing the impact of scanning, a new Z-scan scheme is proposed
which can remarkably reduce bit rates compared with AVS codec using conventional zigzag
scan, while the PSNR of video sequences are maintained. The entire macroblock is divided
into simple motion or complex motion contents and their cost functions are calculated by
using the developed rate-based fast and efficient mode selection scheme, which plays a vital
role as no additional operation module is required. To avoid the entropy coding method
during intra and inter mode decision, a rate estimation method proposed and readjusted to fit
the characteristic of low complexity distortion functions of SATD and SAITD that use more
accurate estimation while achieving higher prediction performance with low computational
requirement of feature. In addition, a relative sum of absolute difference (RSAD) based inter
mode decision algorithm is proposed to determine the best inter mode without calculating
xiv
rate-distortion costs. To further speedup SKIP mode early termination algorithm is adopted.
Based on the probability characteristics, a fast inter mode decision algorithm is suggested
which can determine the best inter mode only by calculating the Hadamard Transform
(H-SAD) cost instead of rate distortion cost and avoid most of the complex computation
processes like quantization, variable length coding, pixel reconstruction etc. To make the
motion estimation process fast, a search point reduction scheme is introduced. Then the
ranking and performance of the commonly used scheme are analyzed for different kinds
of video. Experimental results indicate that the proposed efficient mode selection schemes
reduce the computational complexity of H.264/AVC codec without sacrificing the quality of
video which can be applied in implementing real time encoder for low-power mobile devices
and general purpose computers.Md. Salah Uddin YusufDoctor of Philosophy (Ph.D.
Design of Symmetric 10 Gbps Bi-Directional Wavelength Reused Optical Access Networks
No full textThis thesis is submitted to the Department of Electronics and Communication Engineering, Khulna University of Engineering & Technology in partial fulfillment of the requirements for the degree of Master of Science in Electronics and Communication Engineering, August 2016.Cataloged from PDF Version of Thesis.Includes bibliographical references.This research was conducted to deal with the problem of finding cost‐effective solutions for
Fiber‐to‐the‐Home (FTTH) network deployment. In the FTTH network, the transceiver at the
user premises and the deployment of fiber at the last mile are the major barriers.
A novel approach is demonstrated for reducing the noise of residual modulation and Rayleigh
backscattering (RB) in bidirectional single fiber wavelength division multiplexing passive
optical network (WDM-PON) with 10 Gb/s symmetric differential phase shift keying
(DPSK) signal in downstream (DS) and OFDM re-modulated signal in upstream (US).
Centralized wavelength reused WDM-PON produces re-modulation noise and bidirectional
single fiber generates RB noise. For simplicity, first approach only studied the effect of remodulation
noise and the second approach is the final design of this thesis work.
The first approach proposed a 10 Gb/s symmetric bidirectional dual fiber wavelength reuse
WDM-PON with DPSK signal in DS and reflective semiconductor optical amplifier (RSOA)
re-modulation with orthogonal frequency division multiplexing (OFDM) in US. RSOA is
used for its colorless and cost effective property and dual fiber is used to avoid RB noise.
Similarly, DPSK is used for its constant envelop (CE) property to reduce re-modulation
noise. The results show that the proposed first approach can achieve good performance over
25 km fiber transmission with error free operation in DS and bit error rate (BER) lower than
forward error correction (FEC) limit in US.
But the dual fiber approach is not cost effective. It increases the use of optical resources and
network size outside of the plan. Therefore, another approach of single fiber is also presented
in this thesis.
The second proposed approach is a 10 Gb/s symmetric bidirectional single fiber wavelength
reuse WDM-PON with DPSK signal in DS and Mach-zhender modulator (MZM) remodulation
with OFDM signal in US. MZM is used to overcome the bandwidth limitation to
reach 10 Gb/s. But this scheme severely affected by RB noise. Because RB noise spectra
overlaps with OFDM signal near the DC frequency. This in-band coherent noise severely
degrades the system performance. To overcome this limitation, wavelength shifted approach
is used to reduce the spectral overlap. It is found that by shifting only 375 MHz, system can
achieve a significant improvement.Tanvir Zaman KhanMaster of Science in Electronics and Communication Engineerin
Studies on Redox Interaction among the Dihydroxybenzene Isomers
No full textThis thesis is submitted to the Department of Chemistry, Khulna University of Engineering & Technology in partial fulfillment of the requirements for the degree of Master of Science in Chemistry, May 2016.Cataloged from PDF Version of Thesis.Includes bibliographical references (pages 164-168)Redox interaction of catechol, hydroquinone and resorcinol has been investigated in
different pH (3-9) media using acetate buffer, phosphate buffer and in only supporting
electrolyte containing KCl media at platinum (Pt), glassy carbon (GC) and gold (Au)
electrode. The study has been carried out using cyclic voltammetry (CV) and differential
pulse voltammetry (DPV) techniques at various scan rates and various concentrations. The results suggest that the all three electrodes exhibited an excellent electrocatalytic effect on the redox behaviors of the dihydroxybenzene isomers. Catechol and hydroquinone show one pair of redox peaks in buffer solution of pH 7 at GC and Pt electrode. But catechol and hydroquinone show two pairs of redox peaks at Au electrode at different pH. In KCl media, catechol and hydroquinone also shows two pairs of peaks at GC and Pt electrode.
Resorcinol shows irreversible anodic peak at GC electrode but it shows quasi-reversible
voltammogram in Pt and Au electrodes in all electrolytic media. The electrochemical
process in all the isomers was controlled by diffusion process.
The electrochemical behavior of catechol, resorcinol and hydroquinone in presence of
sulfanilic acid has been studied by cyclic voltammetry and differential pulse voltammetry
using glassy carbon (GC), gold (Au) and platinum (Pt) electrodes. Voltammetric
appearance reflected that sulfanilic acid has been formed adduct with catechol and
hydroquinone. But no reaction has been occurred between sulfanilic acid and resorcinol.
The peak position of the redox couple is also found to be dependent upon pH. Electron
transfer is most favorable in neutral media. The slopes of the peak potential, Ep vs pH plot was determined graphically as the anodic peak of dihydroxybenzene isomers or adducts at 0.1V/s at different electrode. The value of slope is close to the theoretical value (60mV/pH) for one-electron, one-proton transfer process. This indicates that the oxidation reaction of dihydroxybenzene isomers proceeded via the 1e−/1H+ process. This also suggests that during the reaction not only electron but also proton are released from the oxidation of dihydroxybenzene isomers.
Among the dihydroxybenzene isomers, the electron transfer of hydroquinone is easier than the catechol and catechol is easier than the resorcinol. As -OH sites in resorcinol are connected via meta linkage, so the electron transfer at meta position is unfavorable
(electron deficiency) to that at a para (hydroquinone) or ortho (catechol) –position. The
interaction energy of catechol and hydroquinone in KCl media are 37.63J and 38.60J
respectively. So the extent of delocalization of charge through phenyl ring of hydroquinone is higher than the catechol. Resorcinol shows only one anodic peak, -OH sites in resorcinol are connected via meta linkage, so the electron transfer at meta position is unfavorable due to the deficiency of electron. The redox interactions of the dihydroxybenzene isomers are media dependent.Afsana AfrinMaster of Science in Chemistr
A Mathematical Meta Model of Academic Attainment Based on Class Attendance and Socio-Economic Changes
No full textThis thesis is submitted to the Department of Mathematics, Khulna University of Engineering & Technology in partial fulfillment of the requirements for the degree of Master of Science in Mathematics, September 2016.Cataloged from PDF Version of Thesis.Includes bibliographical references (pages 67-70).The performance of students depend on many factors such as, attendance, motivation, level of
engagement etc. which may be considered as the students attributes towards learning and some
other attributes of the teachers on the process. No research is available relating the academic
attainment and class attendance of our universities, especially at KUET where a 60%
mandatory attendance is imposed to appear in the final examination. Though there exist many
factors on student academic attainment, namely Class Test Marks and Final Grade, here we
are interested and selected one important factor namely class attendance. There are many
departments in this university (KUET) and a lot of subjects are taught. But Mathematics is
common to all. Hence Mathematics is chosen for this study. This study is done among the
students of first year and second year in several engineering departments regarding
mathematics courses during the period 2000 -2013. It is of no doubt that for the existence of
many proxy variables (Teacher’s attribute, student’s attribute, subjects, socio-economic
environment etc.), it is very difficult to assess the impact of class attendance on academic
attainment. Moreover, due to impose of mandatory percentage on class attendance, it becomes
much more difficult to find out the impact of attendance on Class Test Marks as well as on
Final Grade.
In this study we have considered only existing old data (student attendance and academic
performance), where the effect of proxy variables are ignored. Moreover, for better
comparison as well as for finding some test statistics, all the data are normalized. We have
rigorously studied about the correlation between class Attendance and Class Test Marks for
each course. We have also investigated thoroughly in each course for the existence of
correlation between class Attendance and Final Grade. In spite of 60% mandatory attendance,
from the experimental results, it revealed that attendance has a great effect on academic
attainments. But the effect is varying department to department as well as semester to
semester. Though there exist correlations between Attendance and Class Test Marks but there
exist relatively much strong correlation between Attendance and Final Grade in perceptive of
all departments. It reveals that class attendance grow some intuitive knowledge to the students
which affect on their final Grade. According to the existence of correlations, some Meta
regression models are proposed regarding both Class Test Marks and Final grade depending
on attendance. Though for the lacking of continuous data, we could not find out socioeconomic
effect on academic attainment but it is revealed from the experimental study that
introducing any new system effect on Attendance as well as academic performance.Nazmin NaherMaster of Science in Mathematic
Development of Glutamate Sensor Based on Nickel Oxide Nano Particle
No full textThis thesis is submitted to the Department of Chemistry, Khulna University of Engineering & Technology in partial fulfillment of the requirements for the degree of Master of Science in Chemistry, July 2016.Cataloged from PDF Version of Thesis.L-glutamic acid is one of the 20 standard amino acids used by all organisms. It plays an
important role in clinical applications and in food processing and well known as a flavor
enhancer commonly found in various foods. The excessive intake of this flavor enhancer
can cause toxic effect to the human health. Therefore, development of sensors or
biosensors for the determination of glutamate has been of great interest over the past two
decades owing to its importance in food and biomedical industries.
In this study nickel oxide nanocrystal and nickel oxide hollowsphere have synthesized by
using sol-gel and hydrothermal methods. Morphological characterization of nickel oxide
particle was carried out using scanning electron microscopy (SEM) and X-ray diffraction
(XRD). The crystallite size of nickel oxide prepared by sol-gel method has been found 60
nm. Nickel oxide nanocrystal shows higher selectivity towards glutamate detection
compare to nickel oxide hollowsphere. The enzyme-free sensor has been fabricated
through modification of glassy carbon using the mixture of nickel oxide nanocrystals and
citosan (NiO/GCE). Cyclic voltammetry (CV) and amperometry were used to investigate
the electrochemical behavior and catalytic properties of the assembled sensor for glutamate
electro-oxidation in alkaline media. It has been found that NiO/GCE showed remarkably
enhanced electrocatalytic activity towards glutamate and the electrochemical reaction is
diffusion controlled. The sensitivity of NiO/ GCE has been found to be 11 μA/mM/cm2.
Under optimal detection conditions, the sensor exhibited linear behavior for glutamate
detection in the concentration up to 8 mM with a limit of detection of 272 μM. The
interference study of NiO/GCE has been found a significant current response for glutamate
compared to the uric acid and ascorbic acid. The stability of NiO/GCE has been studied
over a period of one week and found to retain 65% of its original activity towards
glutamate detection. Experimental results show that NiO/GCE has significant promise for
fabricating cost effective, enzyme-less, sensitive, stable and selective sensor platform.Sumon ChakrabartyMaster of Science in Chemistr
Determination of Muscle Activity and Work-Done Through Emg Analysis During Human Movements in Salat
No full textThis thesis is submitted to the Department of Biomedical Engineering, Khulna University of Engineering & Technology in partial fulfillment of the requirements for the degree of Master of Science in Biomedical Engineering, May 2016.Cataloged from PDF Version of Thesis.Includes bibliographical references (pages 64-68).In this thesis, human movements during Salat (the ritual prayer of Muslim) are taken in account to observe this muscle movement patterns as exercise. Since surface electromyography (sEMG) is a proven method to observe the muscle activity based on the
generated action potential by the muscles, sEMG method is applied to measure the physical
activities from different muscles of the different steps of Salat. From these measurements, it
has been shown that muscle activity during Salat can perform better exercise than walking in treadmill based on muscle fatigue index. The results specify that during Salat, Biceps Brachii (BB) and Erector Spine (ES) produce improved EMG level for both male and female subjects in opposition to Treadmill exercise. Therefore, the aftermath of this work gives us a message that the person who performs Salat five times in a day is doing exercise of upper limb muscles especially the BB and ES muscles. Furthermore, a modified power spectrum
analyzing method is developed. By this proposed method it is proved that mean frequency
(MNF) based EMG power spectrum analyzing method is comparatively efficient method than
the previously proposed to determine Salat associated muscle fatigue and indices. For finding the relation between work done and corresponding sEMG voltage a number of subjects (male and female) are studied with a known work done and their corresponding sEMG voltages are measured. As a result, some approximate mathematical functions are developed by appropriate curve fitting to relate the work done and corresponding EMG voltage. These mathematical functions are proposed for upper limb (Bicep Bracii) muscle and lower limb (Biceps Femoris and Medial Gastrocnemius) muscle. The estimated relationship between work done and sEMG voltage show minimum error when it actually follows second degree function. Based on the work, it is observed that the functions vary from muscle to muscle which means for upper limb and lower limb the relationships follow different mathematical functions. It is also found that the sEMG voltage generated from muscle undergoes saturation after increasing to a certain work done. At last it can be said so far that this work focuses on human movement in Salat to prove it as a fresh exercise (no muscle fati(Iue) mathematically.In addition with that, analytical techniques are developed to determine its muscle indices and fatigue, as well as mathematical relations are established between work done and corresponding sEMG voltage to observe the work done during Salat.Farzana KhanamMaster of Science in Biomedical Engineerin