Electrical Enginerring

As embedded systems have become more complex and designers are required to develop new products faster while using fewer chips. FPGAs are a good choice because they offer flexibility to design on-chip devices as well as the embedded systems that are reprogrammable and reconfigurable. Today's tougher cost, higher performance, and lower power consumption requirements demand even more efficient design strategies. Dynamic partial reconfiguration (PR) is a good approach to meet these requirements because it extends the inherent flexibility of the FPGA by allowing partial regions of the FPGA to be dynamically reconfigured with new functionality while other applications are still running in the remainder of the device. Partial reconfiguration offers significant advantages compared with the traditional full reconfiguration. Designers can keep the hardware resource utilization low because they can dynamically load more logic circuits into single device in a sequential fashion. Moreover, designers can improve the productivity and scalability of their systems because they need only modify, or revise those functions that are required by a particular application [1]. In the last few decades, computational intelligence has been applied to transform human behavior and experience into mathematical representations that can be interpreted by computer programs. The development of intelligent control systems and intelligent decision support systems are part of this trend. In this Thesis, VHDL and Zynq-7000 FPGA board have been used to design all hardware modules, and a hardware accelerator of fuzzy automata-based decision support system in the context of eye-hand coordination testing for handicapped children was defined and developed.

Solar energy, one of many types of renewable energy, is considered to be an excellent alternative to non-renewable energy sources. Its popularity is increasing rapidly, especially because fuel energy consumes and depletes finite natural resources, polluting the environment, whereas solar energy is low- cost and clean. To produce a reliable supply of energy, however, solar energy must also be consistent. The energy we derive from a photovoltaic (PV) array is dependent on changeable factors such as sunlight, positioning of the array, covered area, and status of the solar cell. Every change adds potential for the creation of error in the array. Therefore, thorough research and a protocol for fast, efficient location and correction of all kinds of errors must be an urgent priority for researchers.For this project we used machine learning (ML) with voltage and current sensors to detect, localize and classify common faults including open circuit, short circuit, and hot-spot. Using the pro...

Learning networks have become extremely powerful tools for data classification. One drawback in many of these applications, however, is that they require a large dataset for efficient performance. Medical imaging, for example, cannot make large enough databases available due to strong data privacy concerns and high data collection costs. Fortunately, data augmentation can provide a solution to improve the classification performance of these systems. This paper proposes a novel method for data augmentation based on the compressive sensing technique. Our approach uses sensing matrices to manipulate data features into different spaces in a compressed form, which are then combined and used as extensions of the original limited dataset. This new feature representation is allowing for a more robust learning network that thereby enables improved data classification. When applying our approach to binary-class and multi-class datasets, results showed that our approach can significantly improve the classifier performance and enhance classification results for both datasets.

Steganography is considered the first line of defense in information security as it hides a secret message (payload) inside an innocent looking file (container) to transfer the payload under the adversary's nose without noticing it. Steganographic systems only use the container to hide the payload. In this paper, we present a steganographic system that uses the container not only to hide the payload, but also to give misleading information to the adversary. To achieve this goal, we use quick response (QR) code as a container. QR codes generated by our proposed system can carry its ordinary message in addition to the payload. Anyone can read the message, but the payload can only be obtained using a secret key. The message and the payload are unrelated; i.e. any message can be generated regardless of the payload and vise versa. We can take advantage of that by generating a message that gives misleading information to the adversary. We test the proposed system and show that the generated QR code is (valid) i.e indistinguishable from an ordinary QR code which makes it look innocent and less susceptible to an adversary's attack. Moreover, it is space-efficient, has an acceptable level of noise immunity and is prone to steganalysis attacks.

Powerful cryptographic systems based on mathematically hard problems are utilized to ensure tighter security for data communication purposes. However, these traditional cryptographic systems are bound to fail in the ensuing era of quantum computing. Thus, Artificial Intelligence (AI) inspired security methods are needed to secure communications in the era of quantum computing. This article presents a challenge-response password-based authentication system based on the Completely Automated Public Turing test to tell Computers and Humans Apart (CAPTCHA) AI hard problem. In this system, a server sends a challenge text to a client, then the client generates a random image and blends the challenge text inside this random image using his password. Then the client sends the generated image to the server. The server extracts the challenge text from the sent image using his copy of the client’s password. If the extracted challenge text is the same as the sent challenge text, then both the cl...

Machine Learning (ML) techniques have been combined with modern technologies across medical fields to detect and diagnose many diseases. Meanwhile, given the limited and unclear statistics on the Coronavirus Disease 2019 (COVID-19), the greatest challenge for all clinicians is to find effective and accurate methods for early diagnosis of the virus at a low cost. Medical imaging has found a role in this critical task utilizing a smart technology through different image modalities for COVID-19 cases, including X-ray imaging, Computed Tomography (CT) and magnetic resonance image (MRI) that can be used for diagnosis by radiologists. This paper combines ML with imaging analysis in an artificial deep learning approach for COVID-19 detection. The proposed methodology is based on convolutional long short term memory (ConvLSTM) to diagnose COVID-19 automatically from X-ray images. The main features are extracted from regions of interest in the medical images, and an intelligent classifier is used for the classification task. The proposed model has been tested on a dataset of X-ray images for COVID-19 and normal cases to evaluate the detection performance. The ConvLSTM model has achieved the desired results with high accuracy of 91.8%, 95.7%, 97.4%, 97.7% and 97.3% at 10, 20, 30, 40 and 50 epochs that will detect COVID-19 patients and reduce the medical diagnosis workload.

In this paper, a joint spatio–radio frequency resource allocation and hybrid beamforming scheme for the massive multiple-input multiple-output (MIMO) systems is proposed. We consider limited feedback two-stage hybrid beamformimg for decomposing the precoding matrix at the base-station. To reduce the channel state information (CSI) feedback of massive MIMO, we utilize the channel covariance-based RF precoding and beam selection. This beam selection process minimizes the inter-group interference. The regularized block diagonalization can mitigate the inter-group interference, but requires substantial overhead feedback. We use channel covariance-based eigenmodes and discrete Fourier transforms (DFT) to reduce the feedback overhead and design a simplified analog precoder. The columns of the analog beamforming matrix are selected based on the users’ grouping performed by the K-mean unsupervised machine learning algorithm. The digital precoder is designed with joint optimization of intra-...

This article primarily focuses on the performance evaluation of a new methodology, imputation by feature importance (IBFI), to serve its imputed dataset in further regression scenarios when dealing with soil radon gas concentration (SRGC) time-series data. The time-series data have been collected spanning over fourteen(14) months period, which included four seismic events, and have been used for experimentation. The imputation by feature importance (IBFI) has been experimented and obtained results are found more efficient in the imputation of missing patterns in investigated time series when compared to traditionally used imputation methods viz. mean, median, mode, predictive mean matching (PMM), and hot-deck imputation.The IBFI methodology has been used in a variety of settings, such as data missing not at random (MNAR), missing completely at random (MCAR), and missing at random (MAR), with missingness percentages ranging from 10% to 30%. In this study, the imputed datasets, 9 for each imputation method, have been used further to predict the attribute of interest (radon concentration (RN)) keeping others as independent attributes such as thoron, temperature, relative humidity, and pressure time series. Support vector machine (SVM) with linear kernel has been used as a learning algorithm and its performance was evaluated based on the fact that how efficient and unbiased values were imputed. Statistical performance evaluation measures viz. root mean squared log error (RMSLE), root mean square error (RMSE), mean squared error (MSE),and mean absolute percentage error (MAPE) have been calculated for the assessment of performance. The findings of our study show that the IBFI imputed dataset has provided a better-fitted model. The model generation and predictions upon IBFI imputed time series result in more accurate predictions when compared to mean, median, mode, PMM, and hot-deck imputed time series. Furthermore, PMM and median imputed time series also perform closer to the IBFI imputed time series.

Due to the wide range of applications, Wireless Sensor Networks (WSN) are increased in day to day life and becomes popular. WSN has marked its importance in both practical and research domains. Energy is the most significant resource, the important challenge in WSN is to extend its lifetime. The energy reduction is a key to extend the network's lifetime. Clustering of sensor nodes is one of the well-known and proved methods for achieving scalable and energy conserving WSN. In this paper, an energy efficient protocol is proposed using metaheuristic Echo location-based BAT algorithm (ECHO-BAT). ECHO-BAT works in two stages. First Stage clusters the sensor nodes and identifies tentative Cluster Head (CH) along with the entropy value using BAT algorithm. The second stage aims to find the nodes if any, with high residual energy within each cluster. CHs will be replaced by the member node with high residual energy with an objective to choose the CH with high energy to prolong the network's lifetime. The performance of the proposed work is compared with Low-Energy Adaptive Clustering Hierarchy (LEACH), Power-Efficient Zoning Clustering Algorithm (PEZCA) and Chaotic Firefly Algorithm CH (CFACH) in terms of lifetime of network, death of first nodes, death of 125th node, death of the last node, network throughput and execution time. Simulation results show that ECHO-BAT outperforms the other methods in all the considered measures. The overall delivery ratio has also significantly optimized and improved by approximately 8%, proving the proposed approach to be an energy efficient WSN.

Recent advancements of the intelligent transportation system (ITS) provide an effective way of improving the overall efficiency of the energy management strategy (EMSs) for autonomous vehicles (AVs). The use of AVs possesses many advantages such as congestion control, accident prevention, and etc. However, energy management and traffic flow prediction (TFP) still remains a challenging problem in AVs. The complexity and uncertainties of driving situations adequately affect the outcome of the designed EMSs. In this view, this paper presents novel sustainable energy management with traffic flow prediction strategy (SEM-TPS) for AVs. The SEM-TPS technique applies type II fuzzy logic system (T2FLS) energy management scheme to accomplish the desired engine torque based on distinct parameters. In addition, the membership functions of the T2FLS scheme are chosen optimally using the barnacles mating optimizer (BMO). For accurate TFP, the bidirectional gated recurrent neural network (Bi-GRNN) model is used in AVs. A comprehensive experimental validation process is performed and the results are inspected with respect to several evaluation metrics. The experimental outcomes highlighted the supreme performance of the SEM-TPS technique over the recent state of art approaches.

The developments of multi-core systems (MCS) have considerably improved the existing technologies in the field of computer architecture. The MCS comprises several processors that are heterogeneous for resource capacities, working environments, topologies, and so on. The existing multi-core technology unlocks additional research opportunities for energy minimization by the use of effective task scheduling. At the same time, the task scheduling process is yet to be explored in the multi-core systems. This paper presents a new hybrid genetic algorithm (GA) with a krill herd (KH) based energy-efficient scheduling technique for multi-core systems (GAKH-SMCS). The goal of the GAKH-SMCS technique is to derive scheduling tasks in such a way to achieve faster completion time and minimum energy dissipation. The GAKH-SMCS model involves a multiobjective fitness function using four parameters such as makespan, processor utilization, speedup, and energy consumption to schedule tasks proficiently. The performance of the GAKH-SMCS model has been validated against two datasets namely random dataset and benchmark dataset. The experimental outcome ensured the effectiveness of the GAKH-SMCS model interms of makespan, processor utilization, speedup, and energy consumption. The overall simulation results depicted that the presented GAKH-SMCS model achieves energy efficiency by optimal task scheduling process in MCS.

Recently, cyber physical system (CPS) has gained significant attention which mainly depends upon an effective collaboration with computation and physical components. The greatly interrelated and united characteristics of CPS resulting in the development of cyber physical energy systems (CPES). At the same time, the rising ubiquity of wireless sensor networks (WSN) in several application areas makes it a vital part of the design of CPES. Since security and energy efficiency are the major challenging issues in CPES, this study offers an energy aware secure cyber physical systems with clustered wireless sensor networks using metaheuristic algorithms (EASCPS-MA). The presented EASCPS-MA technique intends to attain lower energy utilization via clustering and security using intrusion detection. The EASCPS-MA technique encompasses two main stages namely improved fruit fly optimization algorithm (IFFOA) based clustering and optimal deep stacked autoencoder (OSAE) based intrusion detection. Besides, the optimal selection of stacked autoencoder (SAE) parameters takes place using root mean square propagation (RMSProp) model. The extensive performance validation of the EASCPS-MA technique takes place and the results are inspected under varying aspects. The simulation results reported the improved effectiveness of the EASCPS-MA technique over other recent approaches interms of several measures.

Human fall detection (FD) acts as an important part in creating sensor based alarm system, enabling physical therapists to minimize the effect of fall events and save human lives. Generally, elderly people suffer from several diseases, and fall action is a common situation which can occur at any time. In this view, this paper presents an Improved Archimedes Optimization Algorithm with Deep Learning Empowered Fall Detection (IAOA-DLFD) model to identify the fall/non-fall events. The proposed IAOA-DLFD technique comprises different levels of pre-processing to improve the input image quality. Besides, the IAOA with Capsule Network based feature extractor is derived to produce an optimal set of feature vectors. In addition, the IAOA uses to significantly boost the overall FD performance by optimal choice of CapsNet hyperparameters. Lastly, radial basis function (RBF) network is applied for determining the proper class labels of the test images. To showcase the enhanced performance of the IAOA-DLFD technique, a wide range of experiments are executed and the outcomes stated the enhanced detection outcome of the IAOA-DLFD approach over the recent methods with the accuracy of 0.997.

Hyperspectral imaging (HSI) plays a major role in agricultural remote sensing applications. Its data unit is the hyperspectral cube that contains spatial data in 2D but spectral band data of all the pixels in 3D. The classification accuracy of HSI was significantly enhanced by deploying either spatial or spectral features. HSIs are developed as a significant approach to achieve growth data monitoring and distinguish crop classes for precision agriculture, based on the reasonable spectral response to the crop features. The latest developments in deep learning (DL) and computer vision (CV) approaches permit the effectual detection and classification of distinct crop varieties on HSIs. At the same time, the hyperparameter tuning process plays a vital role in accomplishing effectual classification performance. The study introduces a dandelion optimizer with deep transfer learning-based crop type detection and classification (DODTL-CTDC) technique on HSI. The DODTL-CTDC technique makes use of the Xception model for the extraction of features from the HSI. In addition, the hyperparameter selection of the Xception model takes place using the DO algorithm. Moreover, the convolutional autoencoder (CAE) model is applied for the classification of crops into distinct classes. Furthermore, an arithmetic optimization algorithm (AOA) can be employed for optimal hyperparameter selection of the CAE model. The performance analysis of the DODTL-CTDC technique is assessed on the benchmark data set. The experimental outcomes demonstrate the betterment of the DODTL-CTDC method in the crop classification process. INDEX TERMS Hyperspectral imaging, convolutional neural network, crop classification, agriculture, parameter tuning.

Learning networks have become extremely powerful tools for data classification. One drawback in many of these applications, however, is that they require a large dataset for efficient performance. Medical imaging, for example, cannot make large enough databases available due to strong data privacy concerns and high data collection costs. Fortunately, data augmentation can provide a solution to improve the classification performance of these systems. This paper proposes a novel method for data augmentation based on the compressive sensing technique. Our approach uses sensing matrices to manipulate data features into different spaces in a compressed form, which are then combined and used as extensions of the original limited dataset. This new feature representation is allowing for a more robust learning network that thereby enables improved data classification. When applying our approach to binary-class and multi-class datasets, results showed that our approach can significantly improve the classifier performance and enhance classification results for both datasets.

Human fall detection (FD) acts as an important part in creating sensor based alarm system, enabling physical therapists to minimize the effect of fall events and save human lives. Generally, elderly people suffer from several diseases, and fall action is a common situation which can occur at any time. In this view, this paper presents an Improved Archimedes Optimization Algorithm with Deep Learning Empowered Fall Detection (IAOA-DLFD) model to identify the fall/non-fall events. The proposed IAOA-DLFD technique comprises different levels of pre-processing to improve the input image quality. Besides, the IAOA with Capsule Network based feature extractor is derived to produce an optimal set of feature vectors. In addition, the IAOA uses to significantly boost the overall FD performance by optimal choice of CapsNet hyperparameters. Lastly, radial basis function (RBF) network is applied for determining the proper class labels of the test images. To showcase the enhanced performance of the IAOA-DLFD technique, a wide range of experiments are executed and the outcomes stated the enhanced detection outcome of the IAOA-DLFD approach over the recent methods with the accuracy of 0.997.