ECE-Theses-MSc

Permanent URI for this collectionhttps://iorep.miuegypt.edu.eg/handle/20.500.13071/11

Browse

Now showing 1 - 12 of 12

Restricted
An Artificial intelligence-based system for automatic diagnosis of siseases via EEG signals /
Hanafy, Mennato-Allah Talaat Mostafa,; Supervisor : Medhat Hussein Ahmed Awadalla, Lamiaa Sayed Abdel-Hamid. Includes Arabic Summary.
Alzheimer’s disease (AD) is known for being the main type of dementia, distinguished by developing descent in cognition and amnesia. Early diagnosis can assist in disease management and enhance patients’ overall quality of life. Electroencephalogram (EEG) has emerged as a non-invasive tool for detecting AD that has the benefit of having a high temporal resolution. AD causes several significant changes to the patients’ EEG recordings including reduced complexity, slower EEG rhythms, and changes in synchrony. This thesis explores the use of EEG signals combined with machine learning techniques to develop a computer-aided diagnosis (CAD) tool for the improvement of AD diagnosis. Although Recurrence Quantification Analysis (RQA) features have demonstrated promising results in various EEG analysis methods such as emotion detection, they have been scarcely implemented for AD detection. In this thesis, RQA features are computed to investigate their usefulness for AD detection. Specifically, four feature groups are considered for the detection of AD from EEG recordings which are (1) RQA, (2) Hjorth, (3) Statistical, and (4) Power Spectral features. Multiple classifiers are compared including Support vector machines (SVM), K-Nearest Neighbors (KNN) and Random Forest (RF). Cross-validation (CV) methods, such as 10-fold and leave one-subject-out (LOSO) CV, are used to evaluate model performance. For investigating the relevance of features extracted from original EEG vs. from decomposed EEG, results reveal that features extracted from decomposed brain frequency sub-bands significantly enhance classification accuracy when compared to those extracted from the original EEG signal. An improvement ranging from 7% to 25% is observed for 10-fold CV and from 4% to 16% for LOSO CV. Next, two feature selection methods are applied and compared. In general, both feature selection methods yielded consistent results, leading to performance improvement by 1% to 3% in all experiments. Throughout all performed investigations, RQA features results in II the best accuracies in which its accuracies outperform Hjorth, Statistical and Power Spectral features with up to 15% and 25% for 10-fold CV and LOSO CV, respectively. These results highlight the usefulness of RQA features for AD detection from EEG signals. Best results are achieved by combining best-performing features from RQA and statistical group of features extracted from the decomposed EEG signals in which achieved accuracies were 99.2% and 96.7% for 10-fold CV and LOSO, respectively, using the SVM classifier. This research contributes to the development of more reliable AD diagnostic tools and highlights the potential of EEG-based methods in clinical practice.
Restricted
Facial expression recognition using Deep Learning /
El-Khashab, Omar Mohammed,; Supervisor : Elsayed Mostafa Saad; Alaa Mahmoud Hamdy, Ayman Mahmmoud Ahmed. Includes Arabic Summary.
Restricted
Performance evaluation of intelligent reflecting surface in heterogeneous wireless networks : (5G and beyond) /
El-Kholy, Enji Mamdouh Rashad Mahmoud,; Supervisor : Fawzy Ibrahim Abd El-Ghany, Mahmoud Mohamed El-Mesalawy, Mehaseb Ahmed Mehaseb. Includes Arabic Summary.
Simultaneous Wireless Information and Power Transfer (SWIPT) is recognized as a promising technique for addressing the energy constraints in wireless communication systems. Meanwhile, Intelligent Reflecting Surfaces (IRS) can customize wireless channels to enhance the spectral and energy efficiency of wireless communication systems. Integrating IRS into SWIPT systems can enable long-distance wireless communication with high data rates, device density, reliability, and energy endurance, particularly in the forthcoming Sixth Generation (6G) era. This thesis investigates a Multiple-Input Single-Output (MISO) IRS-assisted SWIPT system. This system involves a Base Station (BS) mounted on an Unmanned Aerial Vehicle (UAV) that concurrently sends information to Information Decoding Receivers (IDRs) and transfers energy to Energy Harvesting Receivers (EHRs). Unlike conventional SWIPT systems with fixed transmitters, UAV-mounted BSs offer flexible wireless service delivery to ground users. The main objective of the proposed system is to maximize the spectral efficiency of the IDRs by simultaneously optimizing the phase shifts of the reflecting elements at the IRS and the precoding vectors at the UAV. To achieve this, the thesis proposes the utilization of Dinkelbach’s algorithm, which allows the optimization of all the variables simultaneously with an acceptable complexity. MATLAB-based simulation results show that integrating the IRS into the SWIPT system enhances the spectral efficiency of IDRs by 60% and the energy efficiency of EHRs by almost 20 μW compared to the system without IRS. Furthermore, numerical analysis reveals that the proposed method significantly exceeds the performance of the AO-based SCA algorithm by approximately 78%
Restricted
Design of PLL-based frequency synthesiser for communication applications /
Daif, Esraa Nashaat Ali Ahmed,; Supervisor : Elsayed Mostafa Saad, Hassan Ahmed El Ghitani, Ghazal Abdelaty Fahmy. Includes Arabic Summary.
Restricted
Development of a contact layer for LEO constellation /
Azzam, Amr Ahmed,; Supervisor : Mohammed Medhat Mokhtar, Al-Emam Said, Mehaseb Ahmed. Includes Ararbic Summary.
In addition to providing truly global coverage, LEO satellite constellations can have significantly decreased the end-to-end path delays compared to geostationary satellites through Inter Satellite Links (ISLs) that form a constellation network. The main challenge is the process burden on the satellites is selecting which satellite will act as network nodes or Host terminals, where each satellite performs both traffic forwarding, and its data downloading. This network is subjected to long delays, disruption, intermittent links, and network partitioning due to its dynamic nature. Delay Tolerant Network (DTN) protocol is recently proposed for different satellite-based networks in such server environment. It relies on strategies of traffic forwarding to expected satellite neighbors according to a contact plan. This gives rise to a concept of On-Board. Contact Plane (CP) to be in charge of forwarding tasks. The Thesis presents a model of a Layer 2 Contact Plane (CP) in charge of traffic management. The design of the CP is based on the multiprocessor system on chip (MPSoc) FPGA envisioned technologies to perform both Host and network node satellite tasks. The reason behind this choice is the need for On-Board Orchestrator to manage numerous tasks in two main categories: Network’s Management and Control Tasks (NMCTs), and Application’s Management and Control Tasks (AMCTs). Therefore, the design is fragmented in three domains using three Cortex-A53 processors of Zynq UltraScale MPSoC ZU3EG Xilinx ultra96-v2 kit. The first domain is the Orchestrator while the others are NMC for NMCTs and AMC for AMCTs. The Orchestrator uses Arm GIC-400 Generic Interrupt Controller (GIC) to perform these tasks.
Restricted
Design and Implementation of Multi Input Multi Output (MIMO) Antenna in 5G Mobile Communication Systems /
Khalifa, Manar Mohamed,; Supervisor : Fawzy Ibrahim, Hesham ElBadawy, Lamia Khashan.
The mobile access technology is going through a revolutionary change every ten years. Each generation of mobile technology has provided significant performance enhancements. These rapid changes are in response to the capacity demands resulting from the massive data growth over the last ten years. 5G aims to a revolutionary leap forward in terms of data rates, efficiency, capacity and massive connectivity. The requirements of 5G are expected to be met by utilizing the bandwidths in both the sub 6-GHz and mm-wave using very large antenna array and MIMO technology. The MIMO antenna technology is widely known for its ability to enhance the system capacity, as well as mitigate the negative effect of the multipath interference. Even though the 5G MIMO is capable of boosting the spectrum efficiency and the data throughput, it is not an easy task to integrate multiple antenna elements into a smartphone or even a base station. To meet the requirement of a closely packed antenna arrangement and to optimize the isolation, the decoupling techniques have become an underlying need for future antenna designs. Furthermore, considering the future demand of multi-band and multi-mode massive MIMO applications, besides the 4G main and the diversity antennas, additional operating bands and resonant modes is supported by the 5G antenna elements, which is now a challenging topic for antenna engineer. The main objective of this thesis is to design a novel antenna that operates for both 4G/5G mobile communication systems. The total dimension of the SCB is 150 x 80 mm2 that would fit in the commercial cellular phones. The size of proposed antenna is 11.7 x 15.9 mm2. It is constructed of four monopoles of various shapes and dimensions with two rectangular bricks on the top layer, and a stub and open slot that are etched in the ground plane. Each part of the antenna contributed to different frequency bands. Different bandwidth enhancement V techniques were investigated and their effect on the total bandwidth is recorded. This single element antenna structure covers frequency ranges from 1.4 to 2.64 GHz and from 3.32 to 4.64 GHz, which is equivalent to 23 bands of the total (32) 5G NR bands and 30 bands of the total (76) LTE bands. The antenna has an efficiency of 80% and a gain of 5 dBi. The proposed antenna element satisfies the desired design specifications. Consequently, it is arrayed to form an 8-element MIMO antenna for the smartphones on a SCB of 150 x 80 mm2 and an 8-element MIMO antenna for the base station on a SCB of 140 x 140 mm2. Finally, after designing 8 x 8 MIMO antenna system, the average channel capacity is calculated at the lower boundary and is found to be 17.857 bps/Hz at SNR 10 dB. However, the peak channel capacity is calculated to be 22.9187 bps/Hz at SNR 10 dB. A great enhancement has been done when it is compared with the results of the single element, where the enhancement is 573.7121%
Restricted
Design of Low Phase Noise RF Oscillators for Modern Wireless Communication Systems /
Abouyoussef, Mahmoud Samy,; Supervisor : Hassan El-Ghitani, Ayman M. El-Tager.
In this work, a low phase noise RF oscillator is designed, simulated and measured. The size of the designed oscillator is compact enough to be used in different applications. Moreover, it can be designed with low fabrication facilities. First, a novel quad spiral resonator is designed. Two resonator structures were introduced. One with a wide separation between the spirals, and another with a narrow gap. Parametric analyses were performed on many physical parameters of the two structures. Different orientations were simulated and compared with each other until the one that gives the best performance was chosen. Another parametric sweep is done on the minimum dimension length, width, and separation used. Each dimension gives a resonator with its own oscillation frequency that can be used according to the desired application. The proposed resonator is compared with other published work and shows a significant improvement in the Q factor of about 400. Second, a negative resistance method is utilized in designing two oscillators, including nonlinear Harmonic Balance simulation, EM-Circuit-Co-simulation. The active device used in the proposed oscillators is ATF-13786. A biasing network as well as an output circuit are designed and optimized for both oscillators. The first oscillator works at 14.4 GHz with a low phase noise of -125 dBc/Hz @ 1 MHz. It has an output power of 10.6 dBm. The second is a harmonic oscillator at 20 GHz. The active circuit is designed to oscillate at 10 GHz, but the output circuit is matched to the second harmonic. The circuit oscillates at 19.8 GHz with a low phase noise of -101 dBc/Hz @ 10 KHz The circuit has an output power -3 dBm. The two circuits are compared with other published work in the same frequency range. The advantage of low phase noise is achieved at the expense of the area.
Restricted
Design and Implementation of Wirless Sensor Network in a Distributed Control System for Emergency Response Notification /
Bekdash, Omar Mokhtar,; Supervisor : Hassan Ahmed El-Ghitani, Alaa Mahmoud Hamdy.
Today’s world is faced with many different types of emergencies in various types of environment. Response to such emergencies is critical to protect resources including human life. The research carries a study on how the wireless sensor network can solve problems that can be difficult and high risk for wired sensor network and how it beats the wired system on both time and cost matter. The research also carries a comparative study between different types of emergen cy communication systems and shows the most suitable system for each case depending on danger, cost and field used in. Furthermore, in this work, an emergency response system is proposed which is easy to deploy and can report the emergency to the users i n various forms, emergency response notification will be the preferred emergency system in the future. A design for an emergency response system is presented using temperature, level, and pressure sensors as a proof of concept. The automated system achieve s better performance than manual human response, it overcomes delayed informed danger, alarm priority and incorrectly estimating the type of danger or emergency.
Restricted
Design and implementation of an ECG readout circuit based on advanced CMOS operational floating current conveyor /
Eldeeb, Mohammed Abdel Raouf.; Supervisor : Hassan Ahmed El-Ghitani, Yehya Hassan Ghallab.
Portable medical equipment like heart rate, blood pressure and glucose monitors has become a common accessory. This led to a rise in need for low power single ended supply analog frontend biosensor circuits. They are required to operate from a single battery and occupy as small area as possible. In this thesis a few of those techniques are analyzed to find the most suitable for an Electrocardiogram (ECG) readout circuit. An ECG readout circuit consists of a low noise Instrumentation Amplifier (INA) to amplify the weak ECG signal followed by an Analog to Digital Converter (ADC) for processing or viewing on oscilloscopes and a couple of filters. The entire chip is based on the Operational Floating Current Conveyor (OFCC), which is a versatile analog building block capable of achieving most functions the Operational Amplifier (opamp) can realize with fewer external components. In this thesis, implementation of an ECG frontend based on the OFCC is presented and discussed. This implementation lead to a very small chip area compared to other designs in the literature. A low noise INA is constructed using 2 OFCC to amplify the signal with chopper technique to reduce flicker noise to 3 µVrms for bandwidth 0.05 – 100Hz for monitoring applications and 3.7 µVrms for bandwidth 0.05-150 Hz for diagnostic purposes. The high pass filter is constructed using a pseudo resistor. The ADC is a first order continuous time Sigma Delta Modulator (∑∆ ADC) with Oversampling Ratio (OSR) 128 equivalent to sampling frequency of 38.4 kHz. The ECG signal bandwidth is at most 150 Hz thus the 38.4 kHz sampling frequency is sufficient to achieve a resolution of 8 bits. The fully integrated overall chip area is merely 0.3 mm2. The area could be reduced to 0.078 mm2 by using an external capacitor for the ADC. Powered by a 0.4 V supply the circuit consumes 28.5 µW.
Restricted
Design and Simulation of Antennas for Millimeter-Wave Communication Systems /
Elsebai, Mayar Raafat Wageeh Elsebai,; Supervisor : Fawzy Ibrahim, Hesham ElBadawy, Tamer Mostafa Abuelfadl.
Design an antenna in the Milli-Meter Wave (MMW) band could be a very hard task, the designing rules that are used in the designing process for the low band frequencies do not apply in case of high frequencies (from 30 GHz to 300 GHz) which is called Extremely High Frequency (EHF) band. The main objective of this thesis is to propose a technique that could make the designing process in MMW band much easier than before. The main idea of the proposed technique is designing new antennas based on a previously designed one called the reference antennas in order not to start from scratch every time a desired antenna is designed. This technique is applied and tested to design several antennas both single and dual band antenna designs. The utilization of proposed technique proved that it is fast, generic, easy and reliable one. For the single band designed antennas, four high performance antennas are designed and simulated based on this approach using IE3D simulation tool. The reference antenna operates at 77 GHz; its frequency is transformed to make the antennas operate at 35 GHz, 60 GHz, 94 GHz and 120 GHz for each design. For the dual band antennas, the reference antenna operates at dual band of 77 GHz and 98 GHz. Then by applying the same approach, three antennas are designed and simulated based on the previously designed reference one using CST Microwave Studio. The resulted desired first antenna works at dual band of 60 GHz and 77 GHz, the second one works at dual band of 77 GHz and 94 GHz and the last one works at dual band of 94 GHz and 120 GHz. Some modifications are applied to the mentioned technique and to the design in order to increase the limit of operation for the designed MMW dual band antenna and to make maximum benefits from the design without the need to design antennas from scratch; this was demonstrated by two examples, first antenna operate at 60 and 72 GHz while the second one operates at 90 and 117 GHz. I After proving the validity of this technique through this work; the designing process will be easier, especially the optimization step, which could be done also by applying this approach instead of tedious optimization based on the HF software packages.
Restricted
Retinal Image Analysis using Image Processing Techniques /
EL-Mahmoudy, Safinaz Mohamed Ahmed EL-Sayed,; Supervisor : Salwa Hussein ElRamly, Ahmed Mohamed Ibrahim El-Rafei, Lamiaa Sayed Abdel-Hamid.
Retinal images provide a simple non-invasive method for the detection of several silent ocular diseases such as diabetic retinopathy and age-related macular degradation. Silent ocular diseases have early symptoms that slowly advance, and are usually unobserved by the patients. In the late stages, these diseases become recognizable after severe vision impairments have already occurred within the patient’s retina. Early detection and treatment of the silent ocular diseases can reduce their progression and improve the quality of life for their patients. However, many factors can result in the degradation of the images’ quality, such as the retinas’ structure, pupil dilation, ocular media opacity, and the experience of the operating person. In case of poor quality retinal images, such early disease symptoms, could be hardly detectable by the computer aided diagnosis (CAD) systems or the ophthalmologists leading to false negatives or misdiagnosis of the silent ocular diseases. Accordingly, retinal image enhancement is an essential preprocessing step necessary to increase the reliability of the performed diagnosis by improving the overall appearance of the different retinal structures within the retinal images. Wavelet transform has the benefit of being consistent with the human visual system in identifying the retinal structures. In addition, wavelet transform is a multi-resolution technique that reveals the finer image details related to the different retinal structures within its subsequent levels. It separates the retinal images’ luminance and edge information in its approximation and detail subbands, respectively. Thus, wavelet decomposition allows for the improvement of the contrast and illumination through manipulation of its low frequency approximation subbands. In addition, it permits for the enhancement of the retinal structures edges for sharper image as well as noise removal using its high frequency detail subbands. In this work, a wavelet-based retinal image enhancement algorithm is proposed that addresses the four most common quality issues within retinal images (1) contrast enhancement, (2) illumination enhancement, (3) noise removal, and (4) sharpness enhancement. Contrast and illumination enhancement involve applying contrast limited adaptive histogram equalization (CLAHE) and the proposed luminance boosting method to the approximation subband, respectively. Noise removal and sharpness enhancement are performed by processing the wavelet detail subbands, such that the upper detail coefficients are eliminated, whereas bilinear mapping is used to enhance the lower detail coefficients based on their relevance. Several analyses were performed for each of the four considered quality issues in order to tune their parameters. The proposed wavelet-based retinal image enhancement algorithm was tested on the public High-Resolution Fundus (HRF) dataset, which is characterized by its blurry, low contrast and poorly illuminated retinal images. Six different retinal image quality measures were considered to assess the proposed algorithm and to compare its performance against four other methods from literature. The comparison showed that the introduced method resulted in the highest overall image improvement followed by spatial CLAHE for all the considered quality measures. Thus, proving the superiority of the proposed wavelet-based enhancement method.
Restricted
Performance Assessment for 5g Access Techniques /
Al-Masry, Ahmed Ali Fawzy,; Supervisor : Fawzy Ibrahim, Hesham ElBadawy
Nowadays, Non-Orthogonal Multiple Access (NOMA) has gained great attention due to its higher spectral efficiency (SE) over Orthogonal Multiple Access (OMA). NOMA can play a vital role in improving the capacity of future networks. Moreover, power domain NOMA multiplexes the users in power domain utilizing the Superposition Coding (SC) and allows them to access the whole spectrum simultaneously. Furthermore, at the receiver side Successive Interference Cancelation (SIC) is applied for signal detection. In this thesis, the Nakagami-m fading channel is more suitable to study the channel model. Therefore, the generation of the Nakagami-m fading channel represents the main contribution of this work. This is why the special Nakagami-m function will be generated. This thesis addresses two main channel models. First of all, the effect of conventional channel model on dual NOMA user (conventional NOMA system two users) is studied. This will be done to get complete modelling of the NOMA system. Secondly, the channel model is changed to be based on a statistical Nakagami-m generation function. The channel statistical model is used by taking into consideration the correlation coefficient matrix. Finally, the statistical channel model is deployed with multi user and perform the assessment technique. The obtained results show that the maximum efficiency that may be reached for conventional NOMA (two users NOMA) will occur at transmitted power of 25dBm. Whereas, by increasing the number of users up to sixteen users the maximum Energy Efficiency (EE) will be achieved at power level of 32dBm. Therefore, increasing more power the system will suffer from more interference, so that the obtained EE will be degraded.

Browse

Recent Submissions