Electrical Engineering

<p>Shiv Nadar University is a comprehensive, multidisciplinary, research-focused, and student-centric University offering a full range of academic programs at the undergraduate, postgraduate and doctoral level. With state-of-the-art infrastructure, the University comprises of academic wings, sophisticated labs, international standard sports facilities, amphitheaters, auditorium, conference rooms and smart classrooms. The University&rsquo;s goal is to become internationally recognized for the quality of its research and creative endeavors and their applicability to improving quality of life, generating new insights and expanding the boundaries of human knowledge creativity. Committed to excellence in teaching, research and service, the University aims to serve the higher education needs of India and the world beyond.</p>

Shiv Nadar University

Selective feature extraction from single-channel electroencephalogram signal provides appropriate classification of mental tasks, which is crucial for designing mobile brain-computer interface and neuro-bio-feedback systems. However, existing features deteriorate in the presence of artifacts. Therefore, we propose a mental task classification method using variational mode decomposition (VMD)-based novel feature extraction from single-channel EEG, with three stages: Signal decomposition using VMD; computation of proposed variational mode energy ratio; classification using adaptive boosting algorithm. The proposed method is evaluated using artifact-free and contaminated EEG signals from EEG during mental arithmetic task (EEGMAT) database and self-acquired (SA) database recorded using single-channel device. Average subject-specific accuracies of 93\% and 96\% for classification of baseline and serial-subtraction task have been achieved in EEGMAT and SA databases respectively. Extensive comparative analysis exhibits the superiority of proposed feature as compared to existing features in terms of accurate classification of baseline-mental task, and robustness under artifactual EEG signals. © 2020 IEEE.

World University root node

(SoE) School of Engineering

2020 IEEE 17th India Council International Conference, INDICON 2020

Variational Mode Decomposition Based Mental Task Classification from Electroencephalogram

Resistive Random Access Memories (RRAMs) or Memristor has been a revolution in current nonvolatile memory technology. This work demonstrates an analytical model that shows the effect of applied bias on the metal-semiconductor interface to affect resistive switching of an RRAM device. The applied bias modulates the corresponding interface in terms of various interfacial electrical parameters. Besides, the change in bias affects the distribution of bulk defects primarily oxygen vacancies as well as non-lattice oxygen ions. This, in turn, also affects the corresponding resistance of bulk as well as other electrical parameters at the interface. Further conduction mechanism of the device with interfacial oxide formation as well as dissolution to impact resistive switching behavior has been elaborated. © 2020 IEEE.

2020 International Conference on Emerging Frontiers in Electrical and Electronic Technologies, ICEFEET 2020

Effect of applied bias on Schottky interface to affect resistive switching in a memristive device

According to a report by the World Health Organization(WHO) of diseases responsible for deaths, heart disease tops the chart. It is revealed that diagnosing heart disease in an earlier stage is an important issue and researchers across the world have investigated to develop intelligent support systems that could help a physician in better medical discernment beforehand. Machine Learning can be used to develop a medical diagnosis system for predicting heart disease which not only promises a more accurate diagnosis but also reduces the cost of diagnosis. In this paper, we are automating the task of predicting heart disease by using statistical methods/machine learning algorithms. We are exploiting some of the most important algorithms used these days like Logistic Regression, Support Vector Machine (SVM), Decision trees and also Ensemble Learning algorithms (Bagging, Boosting and Random Forests). We conduct these experiments on Cleveland Heart Dataset available on UCI KDD Archive. The dataset distribution judged by our research fits well with decision trees and gives better results with ensemble learning methods, we justify this claim by comparing ensemble results with other Machine Learning approaches. Random forest classifier works best for this data set distribution, the random-nature of this ensemble approach fits the dataset well providing good test accuracies of upto 96.26\% © 2020 IEEE.

Proceedings of the 2020 International Conference on Computing, Communication and Security, ICCCS 2020

Ensemble learning classification for medical diagnosis

This work models high current snapback behavior in n-FET transistors with bottom body contact under high current stress at the drain for ZRAM (Zero capacitor RAM). We analyze 2D current flow in n-FET near the pinch-off region and relate the results to the S-shaped snapback characteristics under high injection avalanche generated carriers conditions. The role of surface bipolar effects on the first snapback phenomenon in the GG-NMOS (Gate-Grounded NMOS) is investigated. A novel physical insight of the bipolar activity is modeled through current flow and barrier lowering at the source-substrate junctions. Moreover, the coupling of electron and hole injection is described. Finally, a model for surface potential inside the substrate before and after snapback is derived and compared with TCAD simulation results. A very good agreement is observed between the model and the TCAD results. © 2020 The Author(s).

Fulltext

Journal of Physics Communications

Bipolar effects in snapback mechanism in advanced n-fet transistors under high current stress conditions

Nowadays robots have been inching towards living beings in terms of performance, agility, accuracy and performance. But there is one domain which is yet to come to limelight i.e. autonomous rescue robots. This involves a parallel use of robot design and control systems to get it executed. Basically, in simpler terms, there is a risk of life is involved in such scenarios and it's always hoped that a robot deployed for such operations be working with high precision and recall. So, in this paper we have described the design and control of an autonomous amphibian all terrain six-legged robot with adaptive gait for rescue operations. To fulfill this, we have used LiDAR and camera in parallel to assist bot in locomotion and detect the victim. Achieving all of this needs a significant amount of on-board processing power, but at the same time has to be compact enough. Keeping this in mind, NVIDIA Jetson Nano seems to be promising choice to get all the processing done onboard. © 2020 IEEE.

2020 6th International Conference on Control, Automation and Robotics, ICCAR 2020

Designing of an Amphibian Hexapod with Computer Vision for Rescue Operations

The non-orthogonal multiple access (NOMA) scheme has been considered as a study item in 3rd generation partnership project (3GPP) for its possible use in fifth-generation new radio (5G NR) and beyond. In this paper, we explore the available scalable bandwidth feature of 5G NR and propose a hybrid NOMA-OMA scheme based downlink data transmission in 5G NR. The hybrid NOMA-OMA based seamless data transmission over the same 5G NR frame is carried out with help of different numerology factor which indirectly controls the subcarrier spacing, without affecting the conventional OMA based transmission. Performance of the proposed method in conventional 5G NR system is analyzed in terms of the average bit error rate (BER) and ergodic capacity performance for various modulation schemes. The results are illustrated for different μ used in 5G NR specifications. Moreover, the obtained results are corroborated with the analytically obtained expressions. © 2020 IEEE.

Performance Analysis of Hybrid NOMA-OMA Scheme for 5G NR System

Power domain NOMA SISO system achieve more capacity compared to OMA system for a given power. To enhance the overall system capacity, NOMA-MIMO based downlink system with superposition coding (SC) coding is considered in this work. With the incorporation of MIMO, the detector subject to interuser as well as interantenna interference. Conventionally, after receiving the signal from the base station (BS), each user eliminates the interantenna interference using zero-forcing (ZF) based linear detector. This paper proposes low-complexity based QR-LRL detector to overcome jointly the interuser interference and interantenna interference. The performance of the proposed detector under BER and capacity metrics compared with ZFSIC and ML detectors. Simulation results show that the proposed receiver guarantees near ML performance with lower complexity for NOMA-MIMO downlink systems. It presents the sumrate improvement for various MIMO configurations with different modulation orders. It also provides optimal power allocation factor $(alpha)$ to experience the same BER at both the near user (NU) and the far user (FU) using ZF and QR-LRL detector. © 2020 IEEE.

International Symposium on Advanced Networks and Telecommunication Systems, ANTS

Low Complexity Based QR-LRL OSIC Detector for Downlink NOMA-MIMO Systems

In signal processing applications the information about the signal such as frequency (or) period is known a prior for most of the practical signals like ECG, EEG, speech, etc. Inspired by this, in this paper, we propose a new signal representation to estimate the period and frequency information of a given signal with low computational complexity. We achieve this by representing a finite-length discrete-time signal as a linear combination of signals belongs to Ramanujan subspaces. Further, we evaluate the performance of the proposed representation with a simulated example and also by addressing the problem of reducing Power Line Interference (PLI) in an ECG signal. Finally, for a given integer-valued signal, we show that the computational complexity of the proposed transform is quite low in comparison with the existing transforms, and it is quite comparable for a given real (or) complex-valued signal. © 2020 IEEE.

SPCOM 2020 - International Conference on Signal Processing and Communications

Signal Representation Using Ramanujan Subspaces Utilizing A Prior Signal Information

In a high power, grid connected photovoltaic (PV) array system a large number of PV modules need to be connected in series and parallel to get the required power level and dc bus voltage of around 700 - 760 V. Commercially available PV modules are provided with bypass diodes across each string of about 20-cells in, say, a 60-cells module. Among the series connected modules, some may get partially shaded mainly because of dust particles, clouds, bird droppings etc. The partially shaded module(s) or string(s) may get bypassed through the bypass diode(s), especially at higher output current. This may also cause multiple power peaks and may pose difficulty for the maximum power point tracking (MPPT) algorithm. As a result the overall power yield of the PV array may reduce. In this paper a circuit topology working on differential power processing (DPP) concept has been analysed. This topology employs an equalizer circuit which injects extra current in parallel with the shaded modules. This allows the series connected modules to operate at their individual maximum power points and the PV string sees a single peak in the power versus voltage curve. The present equalizer circuit topology has some additional advantages over various other equalizer topology reported in literature. Equalizer modeling equations in different modes of operation are reproduced and working of converter circuit is described based on derived dc equivalent circuit. The transformer turn ratio effect of the implemented converter topology is analyzed and discussed. Matlab simulation and experimental results show the efficacy of the employed converter. © 2019 IEEE.

IEEE Region 10 Annual International Conference, Proceedings/TENCON

SPV Power Optimization Using Half Bridge Inverter and Voltage Multiplier

In this article, Frequency Modulated Thermal Wave Imaging (FMTWI) [1–6] is introduced for the first time for determining power distribution of electromagnetic waves on plane surfaces. The advantage with this technique is that we can extract multiple amplitude and phase images from a single run of experiment. The applied excitation signal in this technique is a frequency modulated chirp signal instead of a single frequency signal used in conventional lock-in infrared (IR) thermography [7–11]. The thermal images obtained using FMTWI can be used qualitatively, e.g., to detect field leakage near electromagnetic junctions and microstrip feed lines. As a practical demonstration of this technique, an example of 2 × 2 patch antenna array at 8 GHz is considered. First, amplitude images at various modulation frequencies are obtained. Next, signal to noise ratio (SNR) values at each frequency are calculated. It is seen that SNR is lower at higher frequencies. It is observed that at higher modulation frequencies, micro-strip lines feeding the individual patch antennas of the array, are not visible in amplitude images, while at lower frequencies they are clearly visible Mathematical modeling of the microwave absorption screen has also been carried out to show variations of incident, reflected, and transmitted powers as a function of screen surface impedance. It is also observed that the screen minimally perturbs the electromagnetic fields. © 2017, © 2017 American Society for Nondestructive Testing.