(SoE) School of 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

<p>This paper proposes a dual-polarized and high gain, four-element based compact multiple-input-multiple-output (MIMO) antenna operating at 5.2 GHz. First, a hammer-shaped antenna has been designed with a gain of 5.3 dBi, impedance bandwidth of 400 MHz, and broadside radiation. A mathematical analysis for radiated electric field and an equivalent circuit model of the hammer-shaped antenna are developed. Using the hammer-shaped antenna as an element, four element MIMO design with shorting walls is proposed. The shorting walls near non-radiating edges improve isolation between the elements by changing the direction of the major lobe. The proposed MIMO design has an envelope correlation coefficient (ECC) &lt; 0.15, measured gain of 5.5 dBi, and mean effective gain (MEG) ∼−3dB. This design has a low profile and single layer planar structure of area 65 mm × 65 mm, which makes it a good contender for portable devices or low-profile hand-held applications in WLAN band. © 2021, Electromagnetics Academy. All rights reserved.</p>

World University root node

Electrical Engineering

Progress in Electromagnetics Research Letters

Hammer-shaped element-based compact mimo antenna for wlan application

Epilepsy is a chronic brain disorder that is characterized by intermittent epileptic seizures that can be identified in an Electroencephalogram (EEG) signal. This paper proposes a low computational complex method to classify epileptic EEG signals by using a suitable Ramanujan Periodic Subspace (RPS). Initially, this method divides the given single channel EEG signal into multiple non-overlapping EEG blocks, which are projected onto a particular RPS. The energy of the projection is used as a feature to classify each block into epileptic or non-epileptic, using an SVM binary classifier. Here, in order to choose that particular RPS, a few sample blocks from the epileptic (ictal), interictal, and healthy EEG signal are projected onto the divisor RPSs of that block. The difference between the average subspace energy of healthy vs ictal or interictal vs ictal blocks is used as a measure to choose the suitable RPS. Finally, the class of each block of the EEG signal is combined using the majority voting scheme to classify the epileptic EEG signal. A publicly available benchmark EEG database from Bonn University, Germany, is used to evaluate the performance of the proposed method. Further, the EEG signals are added with White Gaussian Noise (WGN) and ocular artifact for testing the robustness of the method against noise. Evaluation results in terms of accuracy, sensitivity, specificity, and F-score demonstrate that the proposed method is comparable with the state-of-the-art techniques and also robust against artifacts and noise. IEEE

IEEE Sensors Letters

Ramanujan Periodic Subspace Based Epileptic EEG Signals Classification

A working person on an average spends 1.5–2 h every day traveling either to their places of work or for other daily activities, using metros, trams, buses, and cars, as common modes of travel. Most of such commuters regularly suffer from health conditions like headache, breathless condition, drowsiness, etc. Numerous accidents have been reported due to drowsiness while driving, which may occur due to the build-up of carbon-dioxide (CO2) build in the vehicle chamber. This paper attempts to monitor, analyze, and predict air quality inside the vehicle. This work proposes a sensing system using an off-the-shelf sensor Sensordrone which is connected to an Android Smartphone using Bluetooth Low Energy. The data obtained from the proposed sensing system are then utilized to perform predictive analysis of CO2 build-up inside the vehicular chamber using Auto Regressive Integrated Moving Average (ARIMA) and Support Vector Regression (SVR). Root-Mean-Square Error for SVR and ARIMA models is 47.91 ppm and 55.32 ppm CO2, respectively, indicating that SVR outperformed ARIMA in predicting the CO2 build-up inside the vehicle. © 2021, The Author(s), under exclusive licence to Springer Nature Switzerland AG part of Springer Nature.

Computer Science and Engineering

Journal of Reliable Intelligent Environments

Modeling vehicle indoor air quality using sensor data analytics

Advanced CMOS technology is highly susceptible to ageing effects such as negative bias temperature instability (NBTI) and process variability. This article focuses on investigating the ‘combined impact’ of NBTI degradation along with the process, voltage and temperature (PVT) variations on SRAM metrics. These issues impact the performance of SRAM (Static Random Access Memory) cell in many ways but the two main dynamic metrics are critical read-stability and write-ability. The dynamic metrics can, in turn, be characterised by their static metrics, that determine the static noise margins (SNM) of SRAM. The dynamic metrics were analysed for symmetric and asymmetric NBTI degradation of two inverters of SRAM. NBTI degradations were incorporated on the pull-up transistors of SRAM cell. Furthermore, results of the correlated performance metrics are presented for varied data storage patterns in SRAM cell over its lifetime. This analysis aims to provide an insight into the effects of integrated PVT and NBTI degradation on SRAM cell performance, enabling robust designs and help designers at the early stage of design to mitigate system failure. © 2021 Informa UK Limited, trading as Taylor & Francis Group.

International Journal of Electronics

Integral impact of PVT variation with NBTI degradation on dynamic and static SRAM performance metrics

Transfer learning enables solving a specific task having limited data by using the pre-trained deep networks trained on large-scale datasets. Typically, while transferring the learned knowledge from source task to the target task, the last few layers are fine-tuned (re-trained) over the target dataset. However, these layers are originally designed for the source task that might not be suitable for the target task. In this paper, we introduce a mechanism for automatically tuning the Convolutional Neural Networks (CNN) for improved transfer learning. The pre-trained CNN layers are tuned with the knowledge from target data using Bayesian Optimization. First, we train the final layer of the base CNN model by replacing the number of neurons in the softmax layer with the number of classes involved in the target task. Next, the CNN is tuned automatically by observing the classification performance on the validation data (greedy criteria). To evaluate the performance of the proposed method, experiments are conducted on three benchmark datasets, e.g., CalTech-101, CalTech-256, and Stanford Dogs. The classification results obtained through the proposed AutoTune method outperforms the standard baseline transfer learning methods over the three datasets by achieving 95.92%, 86.54%, and 84.67% accuracy over CalTech-101, CalTech-256, and Stanford Dogs, respectively. The experimental results obtained in this study depict that tuning of the pre-trained CNN layers with the knowledge from the target dataset confesses better transfer learning ability. The source codes are available at https://github.com/JekyllAndHyde8999/AutoTune_CNN_TransferLearning. © 2020 Elsevier Ltd

Fulltext

Neural Networks

AutoTune: Automatically Tuning Convolutional Neural Networks for Improved Transfer Learning

Deep convolutional neural networks (CNN) have evolved as popular machine learning models for image classification during the past few years, due to their ability to learn the problem-specific features directly from the input images. The success of deep learning models solicits architecture engineering rather than hand-engineering the features. However, designing state-of-the-art CNN for a given task remains a non-trivial and challenging task, especially when training data size is less. To address this phenomena, transfer learning has been used as a popularly adopted technique. While transferring the learned knowledge from one task to another, fine-tuning with the target-dependent fully connected (FC) layers generally produces better results over the target task. In this paper, the proposed AutoFCL model attempts to learn the structure of FC layers of a CNN automatically using Bayesian Optimization. To evaluate the performance of the proposed AutoFCL, we utilize five pre-trained CNN models such as VGG-16, ResNet, DenseNet, MobileNet, and NASNetMobile. The experiments are conducted on three benchmark datasets, namely CalTech-101, Oxford-102 Flowers, and UC Merced Land Use datasets. Fine-tuning the newly learned (target-dependent) FC layers leads to state-of-the-art performance, according to the experiments carried out in this research. The proposed AutoFCL method outperforms the existing methods over CalTech-101 and Oxford-102 Flowers datasets by achieving the accuracy of 94.38 % and 98.89 % , respectively. However, our method achieves comparable performance on the UC Merced Land Use dataset with 96.83 % accuracy. The source code of this research is available at https://github.com/shabbeersh/AutoFCL. © 2021, Springer-Verlag London Ltd., part of Springer Nature.

Neural Computing and Applications

AutoFCL: automatically tuning fully connected layers for handling small dataset

Slurry erosion–corrosion is a severe issue for many engineering components used in marine, petrochemical, and agricultural sectors. The deleterious effects of the slurry erosion–corrosion significantly lower the service life and enhance the maintenance cost. For limiting the slurry erosion–corrosion effects, there is a need for high-performance advanced materials. In the present work, equimolar MoNbTaTiZr high-entropy alloy (HEA) was developed, and its slurry erosion–corrosion behavior was investigated. For comparison, conventionally used stainless steel SS316L was also investigated. The detailed microstructural characterization showed the presence of a two-phase bcc crystal structure in the HEA. The major bcc phase was predominantly composed of Ta, Mo, and Nb with the interdendritic region being rich in Ti and Zr. The MoNbTaTiZr HEA showed 2 times higher hardness than the SS316L steel. Under slurry erosion–corrosion condition, the HEA showed 3.5 times higher resistance while under erosive conditions, two times better performance than the reference material. The analysis of the eroded surface morphology showed the presence of a mixed ductile–brittle erosion response for the HEA. The improved performance of the HEA is predominantly related to the high hardness and extremely high corrosion resistance. Electrochemical corrosion testing revealed that the HEA has 80 times lower current density than SS316L and high pitting resistance, resulting in low corrosion rates. The electrochemical impedance spectroscopy (EIS) findings showed a denser and highly stable passive layer. The results indicated that MoNbTaTiZr HEA could be effectively used for countering the slurry erosion–corrosion and corrosive conditions. © 2021, The Author(s), under exclusive licence to Springer Nature Switzerland AG.

Mechanical Engineering

Journal of Bio- and Tribo-Corrosion

Slurry Erosion–Corrosion Resistance of MoNbTaTiZr High Entropy Alloy

Abstract: The results from self-consistent k· p computation of optical gain characteristics of AlSb/InGaAsP/GaAsSb type-II ultra-thin quantum-well heterostructures show a marked improvement in optical gain as compared to the InGaAsP/GaAsSb type-II ultra-thin quantum well heterostructures. The AlSb/InGaAsP/GaAsSb type-II ultra-thin quantum well heterostructures were designed to obtain enhanced optical gain as compared to InGaAsP/GaAsSb type-II quantum well heterostructures. An improvement in optical gain of 948 cm- 1 and a shift in peak energy of 0.03 eV is attributed to interband resonant tunnelling effect and the band alignment due to the presence of GaAsSb layer. Also, a narrower optical gain spectrum is observed in AlSb / In 0.5Ga 0.5As 0.8P 0.2/ Ga As 0.5Sb 0.5 QW heterostructure as compared to the In 0.5Ga 0.5As 0.8P 0.2/ Ga As 0.5Sb 0.5 QW heterostructures. Furthermore, the effect of variation in well width has been studied at AlSb layer thickness 1 nm for optical gain enhancement and wavefunction confinement where the resonant tunnelling effect is observed. Graphic Abstract: Fig. Optical gain in AlSb/ In0.5Ga0.5As0.8P0.2/ GaAs0.5Sb0.5 heterostructures [Figure not available: see fulltext.] © 2021, The Author(s), under exclusive licence to EDP Sciences, SIF and Springer-Verlag GmbH Germany, part of Springer Nature.

European Physical Journal B

Optical gain enhancement and wavefunction confinement tuning in AlSb/InGaAsP/GaAsSb heterostructures

As the proportion of time spent by humans in indoor environment increases, it becomes challenging to maintain good air quality for healthy and productive life. The need to develop a context aware, reliable system capable of providing real time information and alerts on indoor air quality is addressed in this paper. The proposed Internet of Things (IoT) system serves to collect data, predict ventilation states, and provide alerts and recommendations to the end user. A novel method for determination of ventilation states using three indoor pollutants PM2.5, PM10 and CO is proposed. Multi-level logistic regression is first used to define indoor ventilation states using ventilation rate which is calculated with the help of indoor CO2 concentration. K-NN classification technique then predicts indoor ventilation state with the help of 3 input attributes, PM2.5, PM10 and CO. Context-aware information about indoor environment and current ventilation state is conveyed to the end-user in form of an alert, through a smartphone application. The system is found to determine the poor ventilation state with accuracy, precision, recall and F1 score values of 94.34\%, 0.91, 0.88, and 0.89, respectively. IEEE

IEEE Internet of Things Journal

Context-Aware Monitoring and Control of Ventilation Rate in Indoor Environments Using Internet of Things

There has been a significant disruption in the energy market because of the pandemic. Now the focus is shifting towards investing more in renewable energy to address the present health and energy crisis. Wind power has gained special prominence in our clean energy transition as wind energy is the easiest to harvest, most efficient resources, and lowest carbon emitter among the available technologies. The prime focus of this study is towards implementing an array configuration of Savonius turbines beside the train track to maximize the power production. A model is set up on OpenFOAM platform and studied for the different arrangements of turbine clusters. The effect of energy production by the series of turbines was assessed by varying distance among turbines in a staggered or inline manner and the location from the train as well. A systematic characterization of the optimum Savonius turbine cluster in a railway tunnel, which is relatively an unexplored area in the energy harvesting community, has been taken up in this study. It is exciting to note that our computer-oriented result matches well with the benchmark solution and proved to be a viable green energy resource that can possibly erase the energy crisis at remote places. © 2021 Elsevier Ltd