Scanning the Issue

Welcome to the June issue of 2023.

This month’s Issue of the IETE Journal of Research (Volume 69, Number 6, June 2023) includes seventy articles. The articles that follow describe ongoing research and new advancements in the fields of Biomedical Engineering (10 articles), Communications (12 articles), Computers (6 articles), Control (7 articles), Electromagnetics (10 articles), Electronics, Circuits, Devices, and Components (5 articles), Instrumentation (1 article), and Power Electronics (18 articles).

In accordance with the aforementioned order of areas of study, the opening paper of the issue, "Channel Contributions of EEG in Emotion Modelling based on Multivariate Adaptive Orthogonal Signal Decomposition," presents a model for emotion identification by decomposing EEG signals using MEMD. The recovered Intrinsic Mode Functions (IMFs) using the MEMD technique is quasi-orthogonal. The extracted IMFs are therefore orthogonalized using the Gram-Schmidt method. The number of orthogonal components reveals the number of modes employed in the second phase of the proposed method, in which the Empirical Wavelet Transform is used to investigate various characteristics of the IMFs.

The second research paper, titled “Automatic Skin Tumor Detection Using Online Tiger Claw Region Based Segmentation - A Novel Comparative Technique,” aims to develop a method for efficient enhancement and tumor detection from unaltered regions. This work relies heavily on benign or malignant computed tomography (CT) images of cutaneous tumors, which have been efficiently implemented. In this research paper, a novel method called Online Tiger Claw Region Based Segmentation (OTCRBS) is proposed for detecting the boundary of unaffected Skin Cell, similar to how tigers use their claws to tear off the skin of their prey while searching for sustenance. Various epidermis cell anomaly detection properties can be formulated using metric for region. The results of the experiments show that the proposed flow works better than any other segmentation method currently used to find tumour cells.

In recent years, drowsiness while working has gained prominence as a research topic. The different sleepiness detection methods that have been documented in the literature are all thoroughly reviewed in the next article titled, “Deep Review of Machine Learning Techniques on Detection of Drowsiness using EEG Signal.” The performance of 15 different machine learning algorithms is examined by the authors using a collection of EEG features they independently collected.

In the next paper, titled "A Multi-View SVM Approach for Seizure Detection from Single Channel EEG signals", seizure detection is carried out using a single channel EEG dataset. For assessing the suggested model, performance estimation parameters such as Accuracy, Sensitivity, Specificity, F1-score, and AUC value have been computed. Using k-fold cross validation, the model correctly identified seizure and non-seizure over the sets A vs E and B vs E with a precision of more than 99%. When employing the same features, multi-view SVM produces classification accuracy that is 1–4% higher than that of single view SVM. The suggested model is contrasted with current single view SVM models as well. In comparison to a single view SVM model over the same features, it was shown that the multi view SVM model performed noticeably better.

Mammograms can assess breast density and breast cancer risk. Next research paper, titled “Predicting Breast Density of Digital Breast Tomosynthesis from 2D Mammograms”, builds a CAD method to estimate breast density using DBT images. LDA was trained on 2D mammograms. Load DBT projection picture to forecast breast density. LDA outperforms one rule, naive Bayes, decision tree, and support vector machine in experiments. DBT predicts breast density 80% accurately.

The next study combined and compared all phases of BCI signal processing experiments using the 2b EEG data set from the BCI Competition IV. This paper titled, “Determination of Effective Signal Processing Stages for Brain Computer Interface on BCI Competition IV Data Set 2b: A Review Study”, structured into common components and showed how changing the four steps affects classification performance to be an effective review. Kappa values compared performance classification from the gathered research. Results show that combining approaches improves performance. This work highlights BCI Competition IV data set 2b signal processing research to improve BCI performance

In the paper, titled "Deep Learning-based Brain Tumour Segmentation", the authors investigate various FCNN-based Semantic Segmentation techniques in order to develop a deep learning model capable of segmenting tumours in brain MRI images with a high degree of precision and accuracy. To improve the accuracy of their model and acquire the optimal model for their classification, they experiment with a variety of architectures and loss functions, such as dice loss function, hierarchical dice loss function cross entropy, etc.

To achieve a low-cost and efficient platform, the authors of "An Integrated Co-Design of Flow-based Biochips Considering Flow-Control Design Issues and Objectives" propose a unified flow-control co-design methodology that takes into account multiple flow-control cost drivers and design cycle reduction. The suggested co-design method reduces the number of valves by 30%, control-channel length by 48.5%, and flow-channel crossings by 72.2% compared to existing works, on average.

The next article “Oblique and Polarization Independent Metasurface based Absorber for Bio-Sensing Applications” describes a submillimeter/THz metasurface polarization/oblique independent absorber for bio-sensing applications. Periodic modified split ring resonator unit cells form the absorber. Symmetric absorbers neglect incidence and polarisation angles. TM or TE polarisation absorbs nearly perfectly at 300 GHz. The absorber works for normal and oblique incident angles (over 90% at 45 degrees). 20 GHz is always the maximum half-width bandwidth. The absorber has a low-profile thickness of = 0.0002 λ₀ and a unit cell size of 0.0052 λ₀ × 0.0052 λ₀ at the operating frequency. Absorber-based biosensors are shown. Changing absorption frequency reveals the thickness of a covered analyte. 24.9%RIU^-1 is most sensitive at 50 μm profile width. Authors use analytical frequency shift methods. Results are examined using three-dimensional full-wave EM simulations.

In the paper "Data Analytics for Risk of Hospitalization of Cardiac Patients," the authors make a prediction model to figure out how likely it is that a cardiac patient will end up in the hospital. They explain and measure the importance of each factor that increases a heart patient's risk of being hospitalized. The purpose of the proposed model is to identify and validate the factors associated with the elevated risk of hospitalization in cardiac patients.

The authors of "Honey Comb Structured Angularly Stable Band Stop Frequency Selective Surface Based on Hexagonal Loop Unit Cells" propose a single-layer FSS with a small periodic element. The suggested unit cell has hexagonal concentric rings connected as a loop by a flexible segmented modification. Increased electrical length and space efficiency make the unit cell 7.29mm x 12.63mm. The suggested FSS also responds well to TE and TM polarisation modes and oblique incidences. Thus, the design is angle- and polarisation-independent. Full-wave EM simulates and PCB fabricates the hexagonal unit cell FSS. Measurements verify simulations.

In the paper, “Local Diagonal Maxima-Minima Pattern Based Edge Detection Technique for Ultrasound and Digital Radiography Images”, local diagonal and non-diagonal maximaminima patterns are used to create a low-dimensional medical images edge descriptor. The approaches use a middle pixel's local diagonal or non-diagonal maximum and minimum. The method significantly reduces feature vector length without affecting edge map quality. Digital X-rays for fractured bones, dental imaging, and ultrasound images test the approaches. The improved depth local binary pattern, anisotropic diffusion, and Canny edge detection are compared using Canny and Sobel methods. The local diagonal and non-diagonal maxima-minima patterns resist noise and retain valuable edge information. The trials reveal that the suggested methods outperform the present methods in accuracy, Jaccard similarity index, specificity, sensitivity, and dice similarity coefficient.

Initial uplink synchronisation finds new users in wireless communication networks. Due of the high user count, resource sharing is problematic. Early multi-user synchronisation efforts didn't estimate signal parameters. These concerns affect OFDMA spectrum usage and data transfer speed. The article “Joint Initial Uplink Synchronisation and Interference Aware Resource Allocation for D2D Communication in MU-OFDMA System” addresses these concerns using JIUS-IRA. JIUS-IRA requires parameter estimation, antenna selection, resource allocation, and resource sharing. JIUS-IRA uses FBLMS-PES to estimate user parameters initially. JIUS-IRA chooses the best power-efficient 5G MIMO base station-user antenna. IROAS selects. SignRank Oriented Resource Allocation (SO-RA) delivers cellular users sub-channel resources. D2D OFDMA users use H 2ORSS. D2D resource sharers identify the best cellular user. Numerical simulation results are confidential. Other metrics get improved

The following paper "WMFLICM: A Robust Algorithm for SAR Image Segmentation using Hybrid Spatial Information" proposes Weighted Membership Fuzzy Local Information C-Means, which uses spatial context information to incorporate implicit and explicit neighbourhood information in terms of feature similarity. SAR images are notoriously difficult to analyse due to the presence of multiplicative speckle noise. Although the Fuzzy C- Means (FCM) segmentation methodology and its derivatives perform admirably for images affected by additive noise, for speckle-contaminated SAR images, these algorithms and other intensity-based traditional methods do not yield promising results. Incorporating spatial context information can boost the segmentation performance of SAR images. When applied to the task of SAR image segmentation, textural aspects of images prove to be more reliable and appropriate.

In the next paper, titled “Electronically Tunable All pass Filter Using Voltage Variable Inductor and Linear VCO Design”, a novel electronically tunable allpass filter (ETAF) is introduced using a voltage variable inductor (VVI). The architecture is based on a composite CFA (Current Feedback Amplifier)-voltage tunable current conveyor transconductance amplifier (VTCCTA) analogue building block (ABB). The phase (θ)-response of this first-order filter is tuned via the VVI, while the signal-gain (K) is independently adjustable via a single resistor. The phase-error (θ e) induced by device parasitic capacitors is estimated analytically. Experiments utilising PSPICE simulation and hardware configuration reveal a low θ_e (∼3°) with a gain of 10dB at pole frequency 12 MHz. Next, a linear voltage-controlled quadrature oscillator (LVCQO) is constructed by cascading the ETAF to a voltage-tuned integrator in a feedback-loop arrangement.

In the past, many realizations were produced by intuition, which meant choosing the circuit layout beforehand and then guessing at component values that would result in the desired voltage or current transfer function orders. The author of the work, titled "Synthesis of Some Specific Types of Voltage/Current Transfer Functions with Minimum Number of Passive Elements and One Active Device" presents a genuine synthesis for three distinct kinds of functions. It is shown that only bi-linear functions may have their minimal realisation achieved by grounding all passive components, while a broad variety of functions can be synthesised when all capacitors are grounded.

By this paper “Local Triangular Coded Pattern: A Texture Descriptor for Image Classification”, the authors introduce a new descriptor called Local Triangular Coded Pattern (LTCP), which is computed by looking at the connections between a group of pixels in the triangular neighbourhood of a region. The suggested descriptor takes into account several pixels as centres inside the provided region to derive the binary pattern, unlike many of the current local binary descriptors. Image classification on KTH-TIPS, Outex, Brodatz, and Kylergb, as well as CK+ and JAFFE for face emotion, are used to evaluate the efficacy of the LTCP descriptor.

Nonlinear measurement equations govern Doppler-bearing passive target tracking (DBT). Measurement equations make target monitoring difficult. To address these challenges, many researchers have used nonlinear estimation methods such sigma-point Kalman filters. These filters excel in nonlinear system parameter estimation. The present paper “Adaptive Transformed Unscented Simplex Cubature Kalman Filter for Target Tracking”, improves sigma-point Kalman filters. This work next proposes a transformed unscented simplex cubature Kalman filter that uses adaption approaches to lessen its dependence on process and measurement disturbance statistical information. A range-parameterized Unscented Kalman filter initialises the filter. The Cramér –Rao lower bound (CRLB), which establishes the greatest estimator accuracy, is utilised to validate the suggested filter in this article. Monte Carlo simulations demonstrate the filter's efficacy.

A CMOS-implemented two-stage, totally differential, low-voltage, low-power transconductance amplifier is suggested in the paper, titled “DTMOS Based Low Power Adaptively Biased Fully Differential Transconductance Amplifier with Enhanced Slew-Rate and its Filter Application”. This study uses the dynamic threshold voltage MOSFET (DTMOS) and a novel adaptive biassing approach to improve slew rate and power supply overhead. This MOS-based completely differential OTA uses a ±0.5 V dual supply voltage. It consumes 0.104 mW and slews at 168 V/μS. The circuit has 73.86 dB dc gain and 72° phase margin. The proposed OTA's figure of merit for a 10 pF discharge capacitor is 16.15 [(V/μs).pF/μW], a significant improvement over earlier studies. A universal voltage mode filter is constructed and simulated to test the design. The suggested circuit was tested using Mentor Graphics Eldospice and 0.18 μm TSMC level 53 CMOS technology.

The paper "Design of Microstrip Dual-Mode Wideband Bandpass Filter with Controlled Centre Frequency and Bandwidth Using Bandstop Filter Topology" describes a microstrip dual-mode, sharp-skirt, wideband bandpass filter that uses a bandstop filter and has a controlled centre frequency and bandwidth. A reduced bandstop filter (BSF) is made from a meander closed loop resonator with orthogonal feed lines that are directly connected to each other. L-shaped open stubs placed along the diagonal line D-D¹ make a bandstop filter with a wide passband inside the rejection band (0.93GHz to 4.13GHz) and sharp stopbands on each side. Passband ends are changed by L-shaped, open stubs. Use a quarter-wavelength stepped impedance shunt stub (SISS) linked at θ=170° to make the filter response more sharp. The suggested filter's middle frequency and width are set by the impedance ratio (R) of the SISS element. The suggested filter is made, and a Vector Network Analyzer (VNA) is used to test it. Both simulation and recording go well together.

The next paper, titled, “A Framework for Filtering Step of Number Field Sieve and Function Field Sieve” provides a complete filtering framework for NFS class algorithms. This work divides filtering steps into two parts. The steps are "construction of the matrix" and "reduction of the matrix." The first sub-phase arranges the relationships found in the relation collecting stage into a matrix. This phase's main purpose is to build a relation from the relation collecting phase's smooth sections and other NFS class algorithm inputs. Matrix reduction is the second sub-phase. Removing duplicates, singletons, cliques, and merging reduces matrix size while maintaining sparsity. The experiments and results leverage CADO-NFS relationships. This work's matrix representation filter map layout improves filtering module efficiency. The improved matrix duplication handling improves the filtering duplicate removal module. This study optimizes clique removal to improve filtering efficiency.

In the next paper, "Deep Insights of Erroneous Bengali-English Code-mixed Bilingual Language," the authors first describe how to write Bengali-English in English script and then provide rules for writing Bengali-English code-mixed language in English script. Language experts believe typographical and cognitive errors cause the massive amount of incorrect data. The methodology was demonstrated with code-mixed Hindi-English Indic language errors. LSTM-based two-level attention-based deep networks discover faults, rectify them, and translate code-mixed words into single-language utterances. Accuracy, ROUGE, and BLEU scores are given for Bengali-English and Hindi-English code-mix languages at the word and sentence level.

In the subsequent paper, “An Efficient Obstacle-Avoiding Rectilinear Steiner Tree Construction Method using PB-SAT”, the authors present a satisfiability (SAT)-based strategy for effectively creating an obstacle-avoiding Rectilinear Steiner tree for a given set of pins in the presence of rectilinear obstacles. A spanning tree is constructed without obstructions, and the network is divided into 2-pin sub-networks. Each subnet's RST is built while barriers are examined. RST is created using Pseudo-Boolean (PB) SAT. The proposed technique was tested and verified by multiple benchmark circuits. In many cases, the proposed method can reduce wire lengths.

The authors of the paper titled "Neuro-Fuzzy Model for Quantified Rainfall Prediction using Data Mining and Soft Computing Approaches" discuss a method that predicts the quantitative value of rainfall. The proposed algorithm makes predictions using a combination of data mining and neuro fuzzy inference system. According to data from the Tropical Meteorological Department, the prediction accuracy is deemed to be high.

In the paper,” MQTT Vulnerabilities, Attack Vectors and Solutions in The Internet of Things”, the authors focus on MQTT protocol, which uses "publisher/subscriber" concept and let devices effortlessly exchange data. The system architecture makes reliable security challenging to build and maintain. Denial of service, identity spoofing, data leaking, elevation of authority, and manipulation are issues. Internal and external causes can cause these issues. Scientists developed many mechanisms and safeguards to address these issues. Security precautions have increased startup and shutdown times, which will affect battery-powered IoT devices. This requires professionals to streamline techniques without compromising safety.

Wireless communication using MIMO channels increases transmission rate by splitting the input stream into multiple simultaneous data streams. Precoding at the transmitter uses channel decomposition to split the channel into many uncorrelated subchannels to send data streams in parallel. In the paper, “Analysis of Precoder Decomposition Algorithms for MIMO System Design”, the MIMO precoder performance and computational complexity are examined using various channel decomposition techniques. Singular Value Decomposition (SVD), Geometric Mean Decomposition (GMD), LDL^H, LU, Schur, QR, and Jordan decomposition are studied for channel decomposition. The simulation and analytical results show that precoding for MIMO channels decomposed by the QR scheme outperforms all other channel decomposition methods in terms of Bit Error Rate (BER) performance and FLOP usage.

In the paper "A Scalable Key Pre-Distribution Scheme based on the Unital Design for the Internet of Things Security," a key pre-distribution scheme (KPS) based on the combinatorial design for IoT security is proposed. The suggested KPS was made to make networks easier to grow. Because of this, a complex design called the "unital design" has been used to make keys. In the current plan, there is a kind of mapping from the unital design to the key setup. This makes for a network that is easy to scale. The results show that the current plan makes the network much more scalable and has a high level of robustness.

The purpose of this paper titled, “A Neural Network based Time Series Model for Predicting Global Solar Radiations”, is to use time series based neural networks to estimate the global solar radiations from actual experimental data collected over time. MATLAB is used to create the model. The model makes use of solar radiation data collected at regular intervals to create a time series. Three models are constructed in this study; one for daily forecast of global solar radiations and two for hourly prediction of global solar radiations (with and without night hours). Statistical study of the resulting data proved the efficacy of applying nonlinear autoregressive models for foretelling global solar radiations.

In the paper, “Advanced STATCOM Control with Optimized FOPTID-MPC Controller”, the authors use load reactive power to operate a fuel cell power plant static adjustment system. 24 IGBTs form the 5-level static compensator inverter. Advanced and hybrid controllers are tested. In the study, fractional order proportional-integral-derivative controller (MPC), tilt-integral-derivative controller (MPC), and hybrid fractional order proportional-tilt-integral-derivative controller (FOPTID-MPC) controllers are used to compare their transient response characteristics and error-based performance functions. Pathfinder Optimization Algorithm also modifies processor parameters. The FOPTID-MPC controller system is compared.

The main objective of the paper, titled “A Study of Anti-swing Fuzzy LQR Control of a Double Serial Link Rotary Pendulum” is to analyze the anti-swing FLQR controller and compare it to the classical LQR controller. SimMechanics/MATLAB simulated a dynamic mechanical model of the DLRP. Mathematically, this model is the single input and multiple output non-linear dynamic equations. Both anti-swing controllers are tested using time-domain control criteria. To demonstrate controller performance, simulation and experiment results are compared. The FLQR anti-swing controller outperforms classical LQR in time-domain control criteria. In the presence of external force disturbance, robustness analysis contrasted the controllers' dynamic responses. Here, the FLQR anti-swing controller outperforms others in vibration suppression.

The next research uses multiscale bubble entropy and power metric for feature extraction and MLA and stability analysis to produce a reliable multichannel seizure detection technique. After representing the multichannel EEG data as a 2D matrix, the AM FM model exploits the decomposed EEG. Using multiscale bubble entropy analysis, the authors calculate the complexity coefficient from the decomposed EEG wave. A second feature set is created using a simple and efficient power approach to calculate absolute and relative power index. The proposed epileptic seizure classification approach is tested using two machine learning methods. Multiscale bubble entropy is tested on a typical EEG dataset. Their method detects epileptic seizures and outperforms the KNN classifier with the ANN classifier. This approach can detect more seizure types and improve diagnosis because it is more discriminating and stable.

The next paper, titled, “DFIG Based Wind Energy System Robust Optimal Control by Using of Novel LMI-Based Adaptive MPC”, proposes integrating model predictive control with adaptive control to achieve the required performance under varied scenarios. By decoupling active and reactive power control, an adaptive technique estimates uncertainty, disturbance, nonlinear factors, and internal feedback design. LMI-based robust model predictive control provides external feedback. The Lyapunov approach shows system stability under the suggested controller. Simulations demonstrate the method's efficacy.

In the paper, “Hybrid Backstepping Sliding Mode Controller for Stick-Slip Vibrations Mitigation in Rotary Drilling Systems”, the authors aim to reduce stick-slip vibrations in the drilling system so the drill bit can follow the Top Drive's nominal angular velocity quickly. Thus, reduce NPT and safeguard drilling equipment. A sliding controller and backstepping technique are offered to attain this goal. In addition, a hybridization of sliding mode control with a backstepping strategy has been proposed for the so-called hybrid backstepping sliding mode controller. Rotary drilling system literature has never suggested the latter. Two system torsion models for different degrees of freedom have been discussed. Compared to other controllers, the hybrid controller's fast response helps prevent drilling equipment from slip phase vibrations. Therefore, it is advised to use the proposed controller in smart rotary drilling systems for the petroleum business.

The authors proposed a new mixed order reduction strategy for LSLDS, in the paper, “Renovation in the Modified Pole Clustering Technique for the Linear Dynamic Systems”. The revised pole clustering technique reduces the denominator of the original LSLDS, while the Pade approximation reduces the numerator. The method improves Vishwakarma's modified pole clustering method. The original LSLDS and low order reduction system (LORS) match better when pole cluster centres are low. Thus, the improved pole clustering strategy yields fewer pole cluster centres than any other pole cluster-based methods. The suggested solution maintains the stability and steady-state value of the Original LSLDS in LORSs with deficient integral square error (ISE) values. The temporal and frequency responses of the recommended LORS are computed using MATLAB/Simulink and compared to other well-known reduction techniques. The step time qualitative attributes and performance index (ISE) of the proposed LORS are compared to several current well-known approaches to demonstrate its superiority and simplicity. The method works for MIMO LSLDSs also.

This paper, titled “Performance Comparison of ANFIS, FOPID-PSO and FOPID-Fuzzy Tuning Methodology for Optimizing Response of High-Performance Drilling Machine” tunes the FOPID controller with PSO, fuzzy logic, and ANFIS to find a stable and regulated structure. FOPID tuning compensates for PID controller shortcomings. Different optimization strategies compensate for big overshoots and extended settling times. PSO, fuzzy logic, and ANFIS improved high-performance FOPID-controlled drilling machines is shown in this study. Based on the analysis and comparison of simulation results, it can be seen that the FOPID-PSO method outperforms the Ziegler-Nichols (ZN)-FOPID and the other intelligent techniques in terms of less target settling time (ts=0.823s) and optimized peak overshoot (Mp=2.44%).

In the paper, “UHF RFID indoor Localization Based on Phase Difference”, the authors present an indoor positioning system with UHF RFID based on phase difference (PD). Multiple groups of phases are acquired by adjusting the reader's transmission frequency in order to reduce phase ambiguity. With imprecise PDs, two positioning algorithms are proposed: pattern matching with PD correction and machine learning without PD correction. Also presented and discussed are two experiments in which machine learning achieves a superior positioning precision of 0.43 m.

This paper “Multi -objective Naked Mole -Rat Algorithm for UWB Antenna Design”, introduces a novel multi-objective naked mole-rat (MONMR) algorithm for the optimization of two printed monopole patch antenna structures for ultrawideband (UWB) applications. The design incorporates fundamental ultrawideband and triple band-notch antennas. The efficacy of MONMR is evaluated based on two criteria: the minimization of pass-band signal reflection and the maximisation of antenna gain. By mapping solutions to the Pareto-front (PF) boundary in the objective space, MONMR provided multiple optimal antenna designs as opposed to a singular optimal antenna design. MONMR has proven to be an effective method in the field of electromagnetics, as the simulated results corresponded with the experimental results. Furthermore, in terms of gain and reflection coefficient, the obtained PF designs are superior to the reported UWB antennas.

In the next article “An Asymmetric Meandered-Line based Dual-Band ENG-TL Antenna Loaded with Complementary Closed Ring Resonators for Gain Enhancement”, a dual-band ENG-TL antenna that is fed by an asymmetric coplanar waveguide (CPW) and has its gain increased by complementary closed ring resonators (CCRRs) has been developed and tested. An asymmetric rectangular patch, a meandered-line inductor, coplanar ground planes, square-shaped CCRRs, and a radial stub make up the suggested antenna. The suggested antenna works well at 1.28 GHz (1.26-1.29 GHz), 2.61 GHz (2.57-2.65 GHz), and 4.3 GHz (4.15-4.49 GHz), according to simulations. However third band was found as weak resonance and consequently not shown in the paper.

The proposed antenna has a simulated gain of 1.5 and 2.01 dB at 1.28 and 2.61 GHz, respectively, in the direction of maximumradiation. Above reflection and far-field characteristics makes this antenna suitable for aeronautical radionavigation (960–1215 MHz), space to earth radio-navigational satellite (1215–1240 MHz) communications.

In the beginning of this paper “Modified Designs of U-slot Cut Microstrip Antennas for Wider Bandwidth”, a comprehensive study is presented to explain the effects of using a wider U-slot in the rectangular patch to obtain a wider bandwidth. To accomplish impedance matching using a wider U-slot, a further modified design of U-slot-cut rectangular microstrip antenna is presented. The effects of varying the lengths of the vertical U-slots have been studied. Further, a pair of rectangular slot-cut designs of modified U-slot cut antennas is presented, yielding an additional 5% bandwidth increase. Similar designs with circular patch are also proposed, which increase bandwidth by more than 5%. In addition to increasing bandwidth, proposed designs reduce patch copper area by nearly 20%. The proposed antennas produce a perpendicular radiation pattern with a 9 dBi peak gain.

Numerous studies analysing the EMI performance of natural convection heatsinks are investigated in the paper, titled “A Review of the EMI Effect on Natural Convection Heatsinks.” If their electrical dimensions are comparable to λ-λ/20 wavelengths, these structures behave like semi-antennas, especially in devices operating at high frequencies. Consequently, they function as both monopole and patch antennas. To eliminate the EMI effect of a heatsink, the techniques of grounding, shielding, and filtering are studied. Using absorbers and shielding techniques, it is possible to attain 20 dB enhancements in EMI for frequencies between 1 and 40 GHz. Lastly, the effects of heatsink geometry, design parameters, fin varieties, and excitation points on RE are examined.

The authors offer a "Simple Model to Investigate a Three Layer Electromagnetically Coupled Equilateral Triangular Patch Antenna for Enhancement in Bandwidth and Gain" in their study. An analytical model calculates the resonance frequency, quality parameters, bandwidth, input impedance, and gain of electromagnetically coupled stacked equilateral triangular patch antennas. The model is simple, fast, and CAD-compatible. Their experimental findings for varied microstrip structure relative permittivity and thickness are used to calculate the model's accuracy. Commercial software validates the model.

ISM band applications require a metamaterial-loaded circularly polarised fractal boundary patch antenna. In the paper “Meta-Material Loaded Circularly Polarized Fractal Antenna for 2.4 GHz Frequency Applications”, Minkowski fractal over metamaterial-inspired reactive impedance substrate is offered. For circular polarisation (CP), asymmetrical Minkowski fractal curves are placed along the margins of a square patch. Later, metamaterial reactive impedance substrate antenna loading increases CP bandwidth. Adjusting Minkowski fractals yields pure CP radiation. CP emission is examined with the metamaterial-loaded koch fractal antenna. Minkowski antenna with iteration 1 metamaterial fractal reactive impedance surface is also tested.

For use in implantable biotelemetry systems, a novel planar inverted-F antenna (PIFA) based on a Gosper curve fractal shape has been proposed in the paper, “Implantable Antenna Design Based on Gosper Curve Fractal Geometry”. The antenna's MICS band (402-405 MHz) frequencies are optimized for use inside human muscle tissue. The antenna is constructed using a Rogers RO 3010 substrate that is 10 mils thick and 5 mm in radius and two radiating patch layers. The suggested antenna will require up about 60 mm³ of space. The antenna is first conceived inside a simplified model of a single layer of human muscle, and then optimised within a more realistic model with many layers. Bessel filter function-based lumped equivalent circuit models have also been proposed, with an inaccuracy of 2.1% when compared to the proposed antenna design. The suggested numerical antenna model shows good agreement with the performance of the built antenna prototype within the human muscle-mimicking liquid phantom.

The authors of the paper "Performance Analysis of Highly Efficient Two Port MIMO Antenna for 5G Wearable Applications" propose a low-profile dual band 2x2 wide-band wearable slot antenna (smart watch) for the frequency spectrum of 3.4 - 3.6 GHz (LTE Band 42). The proposed antenna is manufactured and its efficacy is measured experimentally. Calculating the Envelope Correlation Coefficient (ECC) between the antenna elements validates the proposed MIMO performance. The measured antenna parameters corresponds well to the results of the simulation.

In the paper, “Design and Testing of Graphene Based Screen Printed Antenna on Flexible Substrates for Wireless Energy Harvesting Applications”, the energy-harvesting patch antenna's radiating part was made using a screen-printing method. Waxed paper and polyester were looked into as flexible substrates for mounting the graphene radiating element. As a comparison, FR4, which is a common material used in electrical circuitry, was also thought of as a substrate. The effectiveness of the antenna was tested with a standard Vector network analyzer. Using the material-based software COMSOL Multiphysics, the antenna's performance was simulated, and the predicted results are compared to the prototyped results.

In the paper "Experimental Study of Parametric Dependency of ZnO Nanorods based Vibration Sensor," Zinc oxide (ZnO) nanorods were grown on a rigid substrate (Fluorine-doped tin oxide, FTO) with different molar concentrations using a low-temperature hydrothermal process. The XRD study showed that ZnO nanorods with a hexagonal wurtzite structure and a (002) plane c-axis are formed. Morphological study using FESEM showed that ZnO nanorods with hexagonal top surfaces had been made. Using photoconductivity and impedance tests, the electrical properties of the devices made are identified.

In the next paper titled “Design and Optimization of Reversible Logic based Magnitude Comparator using Gate Diffusion Input Technique”, some suggested reversible magnitude comparators were made by using existing reversible gates and the gate diffusion input (GDI) technique to put them into action. For creating an N-bit comparator, there are also suggestions for how to do it. The main goal of this paper is to design and build a reversible magnitude comparator using some suggested methods and compare it with existing circuits in terms of constant input, garbage output, number of reversible gates, and quantum cost. EDA Tanner tools use a mix of the CMOS and GDI techniques to build the proposed comparators' transistors.

In the paper "A Hardware-Based Memory-Efficient Solution for Pair-Wise Compact Sequence Alignment," the authors describe a design for sequence alignment that is based on hardware and uses as little memory as possible. The FPGA board is used to simulate and produce the whole architecture. Existing alignment methods take a lot more time than the suggested alignment engine. It takes ≈ 64% −95% less time and ≈ 85% −99% less memory space.

The next paper “Low Loss Voltage Equalization Scheme for Series Connected BiMOSFETs for Pulsed Power Applications,” talks about how to make a circuit for static and dynamic voltage equalization in high voltage BiMOSFETs connected in series and working in pulsed mode. Authors find a general formula for the voltage equalization circuit parameters, and active clamping is used to protect the standby state. The dynamic equalization circuit wastes little power, and the system can work even if one or more of the gadgets fails to turn on or off. A prototype 4 kV series switch has been generated in PSpice, built, and tested both with and without the gate pulse to the lower device in the stack. This work is part of making high voltage pulse modulators at the Pulsed High Power Microwave Division, Raja Ramanna Centre for Advanced Technology in Indore.

In the paper “Plasmonic Grating based Refractive Index Sensor with High Sensitivity”, the authors propose Plasmonic grating-based refractive index sensors. In the plasmonic grating, a defect region is formed. The stop band of the plasmonic grating triggers a unique transmission mode near λ = 1550 nm. A nano-metallic slit in the flaw causes fano resonance. Engineering the device's structural elements shows how Fano resonance is tuned. FDTD numerical investigation is used to understand sensing. Sensitivity determines this sensor's success. The proposed sensor has a 10 nm line width and S=1250 nm/RIU sensitivity. The study enables bio- and chemical-sensing optical sensors.

In the paper, “Digital Holographic Interferometry for Temperature Measurement of Flame without Phase Unwrapping”, Digital holographic interferometry was utilized to measure the temperature profile of a candle flame and a group of candle flames. This saved time and effort by avoiding phase unwrapping. The temperature profile of a single candle flame and a connected candle flame is also affected by wick thickness. With a thicker wick, a single candle flame and a group of candle flames drop down. When flames of different thicknesses were fused together, it was the combined flames of the thinnest wick that reached the highest peak temperature (103 K). The thermocouple reading and the experimentally determined temperature are very close to one another.

This paper “Investigation on Torque Sharing Function for Torque Ripple Minimization of Switched Reluctance Motor: A Flower Pollination Algorithm Based Approach”, talks about the new version of the Torque Sharing Function (TSF) and the intelligent controller based on the flower pollination algorithm to reduce the ripples in the torque of a Switched Reluctance Motor (SRM). TSF numbers depend on the angles of commutation during turn on, turn off, and overlap. The suggested TSF and flower pollination algorithm uses a model of the machine to evaluate the performance parameters and control the ripples in torque for a wide range of speeds. Also, the best results are achieved by using three TSF functions, and helpful tips are given for choosing the right TSF. The suggested TSF with a flower pollination-based intelligent controller is good for figuring out the best angles for turning on, turning off, and overlapping in real time.

Next paper “Vector Controlled Dual Stator Multiphase Induction Motor Drive for Energy Efficient Operation of Electric Vehicles”, suggests installing an Indirect Field Oriented Controlled Dual Stator Multiphase Induction Motor (IFOC-DSMIM) as the vehicle's traction unit to improve mileage and carrying capacity. A MATLAB/Simulink lab loads the drive during free-running, acceleration, and deceleration to assess its efficiency. Researchers calculate efficiency curves for its three excitation modes using a steady-state equivalent circuit approach. These evaluations reveal the drive's load sharing, speed tracking, and torque generating capabilities. Thus, the DSMIM design could reduce energy expenses and increase EV range by saving energy.

Direct-Sequence Spread-Spectrum (DSSS) and Frequency-Hopping Spread-Spectrum (FHSS) modulation techniques are proposed in the paper "Suppression of EMI Using Cost Effective FPGA Based Digital Communication Modulation Techniques in Power Converters" for the operation of three-phase Voltage-Source Rectifiers (VSR) to mitigate EMI in AC Grid. This study contrasts the conventional Sinusoidal Pulse Width Modulation with the suggested modulation approaches. This research demonstrates that the power quality and EMI level can be enhanced by employing the FHSS and QPSK approaches proposed.

The HAVC energy harvester is designed in the paper, titled “Multi-Perforated Energy Efficient Piezoelectric Energy Harvester Using Improved Stress Distribution.” Cantilever constructions of various forms and sizes have been constructed and investigated at 1g (9.8 ms^-2). Due to better stress distribution, the multiperforated cantilever construction produces higher output voltage and power.

In next paper, “Online Identification of Underlying Causes for Multiple and Multi-Stage Power Quality Disturbances using S-Transform”, the suggested method identifies multi-stage power quality (PQ) issues and their causes. This work presents a computationally efficient S-Transform with decision tree (DT)-based online PQ monitoring technique. This online program analyzes sample-based 40-dB noisy multiple and multi-stage PQ disturbance signals simulated in MATLAB with various IEEE-1159-compliant underlying sources. From the modified contours, residual voltage, instantaneous phase-angle jump (IPAJ), and number of zero crossings (NZC) were retrieved. Thus, most typical reasons including three-phase fault, induction motor starting, and capacitor bank energizing have been accurately identified. An online PQ monitoring method is assessing the type and cause of a multi-stage disturbance for the first time. Real-time PQ disturbance data from the lab validated the suggested approach.

The subsequent paper, “Genetic Algorithm & Fuzzy Logic based Condition Monitoring of Induction Motor through Estimated Motor Losses”, offers computing induction motor losses to monitor its condition and performance. The proposed approach collects data without separating the motors from the load, which is novel. Many motor losses can be tracked as a single figure. This research's approach separates losses, which can be used to diagnose the motor instantly. This method evaluates induction motors' stator, rotor, core, bearing, and other internal issues. A Genetic Algorithm (GA) assesses motor and equivalent circuit losses. Defects are recorded if anticipated motor loss data differs from NEMA & IE Standards. Fuzzy method for fault identification and severity evaluation uses data. LT motors from 5 to 200 horsepower were employed for real-time testing and assessment of the suggested approach in various circumstances.

The next paper “Differential DC Component-based Relaying Scheme for Transmission Lines”, demonstrated that differential DC helps determine if a defect is inside or outside. Based on this differential DC feature, current transmission line defects are found. Simulation research using EMTDC/PSCAD software tested the suggested criterion for several fault types with different fault position, resistance, inception angle, and compensation level. The suggested algorithm is further tested for rare scenarios such problems that alter over time, faults that occur across nations, measurement noise, and measurement mistakes. For series-compensated and uncompensated lines, the approach is selective, accurate, and reliable. Low-resistance faults are accurate for shunt-compensated lines.

In the paper “A Novel Flux State Based Control Method for Improved Dynamic Performance of Induction Motor”, the authors say that many techniques have been made, such as direct and indirect field-oriented control, direct torque control, and direct power control. However, most of these techniques use approximation criteria to make the linear system work best. It makes it hard to tune the controls for the current, torque, and flux, which hampers the dynamic performance. In this paper, a new nonlinear control method based on flux states is introduced. This method improves the dynamic performance of an induction motor drive. The proposed control method is tested on a 0.75kW IM by simulating it in MATLAB/Simulink and then testing it in the real world. The results show that the suggested control method based on flux states works to improve the dynamic performance of an induction motor drive.

The simultaneous regulation of SCUC and OUPFC is a major focus of this paper, “Multi-objective Mathematical Programming to Simultaneously Control Security-Constrained Unit Commitment (SCUC) and Optimal Unified Power Flow Controller (OUPFC) to Improve Power-system Controllability”. This is due to the fact that minimizing energy loss and freeing up transmission line capacity can both be accomplished by combining SCUC and OUPFC and then adjusting the effective variables. To optimize the total fuel cost, power losses, and cost of OUPFC installation, the SCUC problem with OUPFC is solved as a Nonlinear Programming with Discontinuous Derivatives (DNLP) framework, and the proposed algorithm is implemented on two IEEE 14-bus and IEEE 30-bus systems using General Algebraic Modeling System (GAMS) software. Simulation findings demonstrate that the proposed approach can optimize the placement and parameters of the OUPFC to enhance power system control in the SCUC problem.

The authors in their paper, titled “Energy Transmission Modes and Output Ripple Voltage of Quadratic Buck-Boost Converters with Switched Inductor Network”, propose a novel Quadratic Buck-Boost Converter with Switched Inductor Network to study output voltage ripple. The input and storage energy inductors determine how a Quadratic Buck-Boost Converter with Switched Inductor Network operates. Quadratic Buck-Boost Converters with Switched Inductor Network study storage energy inductor energy transfer in continuous conduction mode (CCM) and discontinuous conduction mode (DCM). Based on the difference between the minimum current flowing through the energy storage inductor and the output current, it is suggested that the energy transmission of the energy storage inductor working in CCM is split into complete power supply mode (CISM) and incomplete power supply mode (IISM) and the border inductance between them is obtained. Thus, Quadratic Buck-Boost Converters with Switched Inductor Network are CISM-CCM, IISM-CCM, and DCM. The border inductance of "complete power supply mode" and "incomplete power supply mode" is the minimal inductance to lower converter output ripple voltage.

In the paper "Implementation of Incremental Conductance MPPT Algorithm with Integral Regulator by using Boost Converter in Grid-Connected PV Array," the authors describe a fast, effective, and linear incremental conductance (IC) algorithm for chasing the maximum power point (MPP) of a grid-connected photovoltaic (PV) array. A method for determining the optimal values of the PV array's design parameters has been proposed. By figuring out the best numbers for the inductor, capacitor, and duty cycle of the boost converter, this method makes it possible to match the PV system to the configuration of the DC/AC converter. There is an analysis of the different ways to find the maximum power point (MPPT), and the full design of the PV converter system is also included.

The authors of "Feature Extraction and Classification Techniques for Power Quality Disturbances in Distributed Generation: A Review" investigate the common occurrences of power quality problems in this setting. Unpredictable and temporary, Power Quality (PQ) problems are common. The authors employ various intelligent system technologies, such as wavelet transformations, expert systems, and artificial neural networks, to pinpoint the precise sites of the faults.

In the paper, “MVSI & AVSI Supported DSTATCOM for PQ Analysis”, the authors present a Distributed Static Compensator (DSTATCOM) on a dual voltage source inverter (DVSI) that improves power quality (PQ) in the power utility system (PUS). The proposed DSTATCOM uses two voltage source inverters (VSI) to manage reactive power. MVSI and AVSI are their names. The control approach generates reference source currents and switching pulses for i CosØ. MATLAB/Simulink analyzes the control algorithm mathematically, and case studies compare the outcomes. The DVSI provides better voltage regulation, voltage balancing, source current harmonic reduction, and power factor correction under diverse loading conditions than any theoretical results. Reliability, stress balancing of MVSI and AVSI, cost reduction due to filter size, and enhanced PUS function are also accomplished. To evaluate its performance, the DVSI-based DSTATCOM is compared to the IEEE-519-2014 and IEC-61000-1 grid code.

In the next paper, “New Method to Detect Loss of Excitation in Synchronous Generators”, the authors present a new LOE fault detection approach. The derivative of the terminal voltage and derivative power angle of a synchronous generator are used to determine the problem. Full computer studies have been done to test the proposed approach by taking into account the synchronous generator's fault modes and conditions. The proposed method finds LOE defects faster and more precisely, according to these studies. It also helps distinguish LOE and SPS faults.

In the paper, “Standalone Solar Photovoltaic Electricity Supply to Rural Household in Tanzania”, the authors carry out hybrid optimization of multiple energy resources (HOMER) pro program to perform technical and non-technical analyses to discover the best renewable energy source configuration. Particle swarm optimization (PSO) in MATLAB reduces net present cost (NPC). In HOMER, a diesel generator (DG) costs $28.116, whereas a solar photovoltaic (SPV) system with the same capacity costs $3,013. PSO reduces those values to $19,488 and $2,089. This research will boost rural energy.

In “Improved Switched-Capacitor Switched-Inductor Z-source Inverter for Increasing Boost Factor and Decreasing Voltage Capacitors Stress”, the authors uses a switched-inductor Z-source inverter to improve a switched-capacitor inductor ZSI. The recommended inverter replaces side diodes of the one switched inductor Z-source inverter with capacitors. The suggested inverter outperforms the switched-inductor Z-source and quasi-Z-source inverters in voltage gain and boost factor. Switched-inductor-capacitor Z-source inverters reduce capacitor voltage stress. This paper's inverter can be scaled up to n-cascade level for higher boost factor and lower duty cycle. They determine the voltage equations by testing the proposed structure in different modes. The switches, capacitors, diodes, inductors, voltage gain, and boost factor of the suggested inverters are calculated and compared to similar inverters. Lab converters are made to test the proposed converter. The enhanced switched-capacitor-inductor ZSI converter outputs 200 voltage with a boost factor of 1.67, proving the converters perform as planned.

In next paper, “Hopfield Neural Network-based Average Current Mode Control of Synchronous SEPIC Converter”, the authors propose Generalized Hopfield Neural Network (GHNN) tuned PI controllers for ACM control of synchronous single ended primary inductance converter (SEPICs). The dynamic converter model is developed using state-space averaging and all converter flaws. To make controllers, this model is needed. The modified Hankel matrix method reduces the derived converter model to first-order to simplify the design.The transient and steady-state performance of the converter with the proposed controller is studied and compared with controllers tuned using Zeigler Nicholas (ZN) method, Genetic Algorithm (GA), Particle Swarm Optimization (PSO), and Reduced-Order Linear Quadratic Regulator (ROLQR) control scheme for changes in input voltage, reference voltage, and load using MATLAB/Simulink R2015b software tool.

In the paper titled "Application of Novel Six-phase Doubly fed induction generator for Open Phasesthrough Modeling and Simulation," the authors discuss the mathematical and simulation-based modeling of a novel 6-ϕ-DFIG fed induction generator (DFIG) operating with open-phase faults. In addition, the benefits of this DFIG are illustrated. Simulation results validate the mathematical model, something the authors assert has never been done previously.

The issue ends with the article from Power Electronics, titled, “Stability Testing and Restoration of a DEIG based Wind Power Plant with Indirect Grid Control Strategies” describes the design and implementation of indirect grid control for a wind power plant using grid and rotor side converter characteristics. The proposed system includes DEIG, WTS, and MPCS-EPCS. Transmission line faults (symmetrical and asymmetrical) produce power imbalances. The MPCS and EPCS can monitor and regulate grids under various defects. Converter topologies that reduce DEIG-WTS faults and improve power grid resiliency are discussed. DEIG-WTS fault voltage ride-through enhances system stability, resource utilization, and efficiency.

Additional information

Notes on contributors

Ranjan K Mallik

Ranjan K Mallik (FIETE, FIEEE, FIET, FTWAS, FNAE, FNA, FNASc, FASc) is an Institute Chair Professor in the Department of Electrical Engineering, Indian Institute of Technology (IIT) Delhi. He received the BTech degree from IIT Kanpur and the MS and PhD degrees from the University of Southern California, Los Angeles, all in electrical engineering. He has worked as a scientist in the Defence Electronics Research Laboratory, Hyderabad, India, and as a faculty member in IIT Kharagpur and IIT Guwahati. His research interests are in diversity combining and channel modelling for wireless communications, space-time systems, cooperative communications, multiple-access systems, power line communications, molecular communications, difference equations, and linear algebra. He is a recipient of the Shanti Swarup Bhatnagar Prize, the Hari Om Ashram Preriot Dr Vikram Sarabhai Research Award, the Khosla National Award, the IETE Ram Lal Wadhwa Award, the IEI-IEEE Award for Engineering Excellence and the J C Bose Fellowship. He is a member of Eta Kappa Nu and a Fellow of IEEE, the Indian National Academies INAE, INSA, NASI, and IASc, TWAS, the West Bengal Academy of Science and Technology, IET (UK), IETE (India), The Institution of Engineers (India) and the Asia-Pacific Artificial Intelligence Association. He served as an Area Editor and an Editor for the IEEE Transactions on Wireless Communications, and as an Editor for the IEEE Transactions on Communications. He was a Technical Program Committee (TPC) Co-Chair for the Wireless Communications Symposium of IEEE GLOBECOM 2008 and IEEE ICC 2010, a TPC Co-Chair for the PHY Track of IEEE WCNC 2013, and a TPC co-chair for the Communication Theory Symposium of IEEE ICC 2021. He is currently a Deputy Editor-in-Chief of IETE Journal of Research. E-mail: [email protected]

Shiban K Koul

Shiban K Koul is currently an emeritus professor at the Indian Institute of Technology, Delhi. He served as deputy director (Strategy and Planning) at IIT Delhi from 2012-2016 and mentor deputy director (Strategy & Planning, International Affairs) at IIT Jammu from 2018-2021. He also served as the chairman of Astra Microwave Products Limited, Hyderabad from 2009-2019 and Dr R P Shenoy Astra Microwave chair professor at IIT Delhi from 2014-2019. His research interests include RF MEMS, high frequency wireless communication, microwave engineering, microwave passive and active circuits, device modelling, millimetre and sub-millimetre wave IC design, body area networks, flexible and wearable electronics, medical applications of sub-terahertz waves and reconfigurable microwave circuits including miniaturized antennas. He successfully completed 38 major sponsored projects, 52 consultancy projects and 61 technology development projects. He has authored/co-authored 601 research papers, 22 state-of-the art books, 4 book chapters and 2 e-books. He holds 26 patents, 6 copyrights and one trademark. He has guided 30 PhD thesis and more than 100 Master’s theses. He is a Life Fellow of IEEE and Fellow of INAE and IETE. He is the chief editor of IETE Journal of Research, associate editor of the International Journal of Microwave and Wireless Technologies, Cambridge University Press. He served as a Distinguished Microwave Lecturer of IEEE MTT-S for the period 2012-2014.

Recipient of numerous awards including IEEE MTT Society Distinguished Educator Award (2014); Teaching Excellence Award (2012) from IIT Delhi; Indian National Science Academy (INSA) Young Scientist Award (1986); Top Invention Award (1991) of the National Research Development Council for his contributions to the indigenous development of ferrite phase shifter technology; VASVIK Award (1994) for the development of Ka- band components and phase shifters; Ram Lal Wadhwa Gold Medal (1995) from the Institution of Electronics and Communication Engineers (IETE); Academic Excellence Award (1998) from Indian Government for his pioneering contributions to phase control modules for Rajendra Radar, Shri Om Prakash Bhasin Award (2009) in the field of Electronics and Information Technology, VASVIK Award (2012) for the contributions made to the area of Information, Communication Technology (ICT) and M N Saha Memorial Award (2013) from IETE. His name has recently figured in the Scopus Elsevier top 2% Scientists under the Category “Year 2021”.E-mail: [email protected]

Arun Kumar

Arun Kumar is with the Centre for Applied Research in Electronics, Indian Institute of Technology, Delhi, since 1997. He became professor in 2008 and has served as head of Centre for more than 7 years. He obtained the BTech, MTech and PhD degrees from Indian Institute of Technology Kanpur in 1988, 1990 and 1995, respectively. He was visiting research rate the University of California, Santa Barbara, USA from 1994 to 1996 before joining IIT Delhi. His research interests are in digital signal processing, under water and air acoustics, human and machine speech communication, and multi-sensor data fusion. Professor Arun Kumar is an inventor on 10 granted US patents. He has guided 16 PhD theses and 180 Master’s theses. He has authored/co-authored 160 papers in peer reviewed journals and conferences. He has been project investigator/co-investigator for 72 funded R&D projects from industry and government. These projects have led to several technology and know-how transfers. Many of the technologies co-developed by him are deployed in the field and are in practical use. Professor Arun Kumar has served on several technical and organization committees of conferences, and on national level committees in electronics and defence fields. He is co-founder and director of a company that develop signal processing and AI based technologies and products for speech-based and multi-modal human and machine. He is currently a Deputy Editor-in-Chief of IETE Journal of Research.E-mail: [email protected]