Showing 1–50 of 131 results for author: Pal, M

Emergence of a radio jet in the changing-look AGN 1ES 1927+654

Authors: Eileen T. Meyer, Sibasish Laha, Onic I. Shuvo, Agniva Roychowdhury, David A. Green, Lauren Rhodes, Amelia M. Hankla, Alexander Philippov, Rostom Mbarek, Ari laor, Mitchell C. Begelman, Dev R. Sadaula, Ritesh Ghosh, Gabriele Bruni, Francesca Panessa, Matteo Guainazzi, Ehud Behar, Megan Masterson, Haocheng Zhang, Xiaolong Yang, Mark A. Gurwell, Garrett K. Keating, David Williams-Baldwin, Justin D. Bray, Emmanuel K. Bempong-Manful , et al. (10 additional authors not shown)

Abstract: We present multi-frequency (5-345 GHz) and multi-resolution radio observations of 1ES 1927+654, widely considered one of the most unusual and extreme changing-look active galactic nuclei (CL-AGN). The source was first designated a CL-AGN after an optical outburst in late 2017 and has since displayed considerable changes in X-ray emission, including the destruction and rebuilding of the X-ray coron… ▽ More We present multi-frequency (5-345 GHz) and multi-resolution radio observations of 1ES 1927+654, widely considered one of the most unusual and extreme changing-look active galactic nuclei (CL-AGN). The source was first designated a CL-AGN after an optical outburst in late 2017 and has since displayed considerable changes in X-ray emission, including the destruction and rebuilding of the X-ray corona in 2019-2020. Radio observations prior to 2023 show a faint and compact radio source not unusual for radio-quiet AGN. Starting in February 2023, 1ES 1927+654 began exhibiting a radio flare with a steep exponential rise, reaching a peak 60 times previous flux levels, and has maintained this higher level of radio emission for nearly a year. The 5-23 GHz spectrum is broadly similar to gigahertz-peaked radio sources, which are understood to be young radio jets less than ~1000 years old. Recent high-resolution VLBA observations at 23.5 GHz now show resolved extensions on either side of the core, with a separation of 0.14 pc, consistent with a new and mildly relativistic bipolar outflow. A steady increase in the soft X-ray band (0.3-2 keV) concurrent with the radio may be consistent with jet-driven shocked gas, though further observations are needed to test alternate scenarios. This source joins a growing number of CL-AGN and tidal disruption events which show late-time radio activity, years after the initial outburst. △ Less

Submitted 26 June, 2024; originally announced June 2024.

Comments: Submitted to ApJ Letters 24 June 2023. 15 pages, 4 figures, 1 table

An asymptotic expansion for a Lambert series associated to Siegel cusp forms of degree $n$

Authors: Babita, Abhash Kumar Jha, Bibekananda Maji, Manidipa Pal

Abstract: Utilizing inverse Mellin transform of the symmetric square $L$-function attached to Ramanujan tau function, Hafner and Stopple proved a conjecture of Zagier, which states that the constant term of the automorphic function $y^{12}|Δ(z)|^2$ i.e., the Lambert series $y^{12}\sum_{n=1}^\infty τ(n)^2 e^{-4 πn y}$ can be expressed in terms of the non-trivial zeros of the Riemann zeta function. This study… ▽ More Utilizing inverse Mellin transform of the symmetric square $L$-function attached to Ramanujan tau function, Hafner and Stopple proved a conjecture of Zagier, which states that the constant term of the automorphic function $y^{12}|Δ(z)|^2$ i.e., the Lambert series $y^{12}\sum_{n=1}^\infty τ(n)^2 e^{-4 πn y}$ can be expressed in terms of the non-trivial zeros of the Riemann zeta function. This study examines certain Lambert series associated to Siegel cusp forms of degree $n$ twisted by a character $χ$ and observes a similar phenomenon. △ Less

Submitted 27 May, 2024; originally announced May 2024.

Comments: 15 pages, comments are welcome! arXiv admin note: text overlap with arXiv:2305.07412

MSC Class: Primary 11M06; 11M26; 11F46; Secondary 11N37

Speech enhancement deep-learning architecture for efficient edge processing

Authors: Monisankha Pal, Arvind Ramanathan, Ted Wada, Ashutosh Pandey

Abstract: Deep learning has become a de facto method of choice for speech enhancement tasks with significant improvements in speech quality. However, real-time processing with reduced size and computations for low-power edge devices drastically degrades speech quality. Recently, transformer-based architectures have greatly reduced the memory requirements and provided ways to improve the model performance th… ▽ More Deep learning has become a de facto method of choice for speech enhancement tasks with significant improvements in speech quality. However, real-time processing with reduced size and computations for low-power edge devices drastically degrades speech quality. Recently, transformer-based architectures have greatly reduced the memory requirements and provided ways to improve the model performance through local and global contexts. However, the transformer operations remain computationally heavy. In this work, we introduce WaveUNet squeeze-excitation Res2 (WSR)-based metric generative adversarial network (WSR-MGAN) architecture that can be efficiently implemented on low-power edge devices for noise suppression tasks while maintaining speech quality. We utilize multi-scale features using Res2Net blocks that can be related to spectral content used in speech-processing tasks. In the generator, we integrate squeeze-excitation blocks (SEB) with multi-scale features for maintaining local and global contexts along with gated recurrent units (GRUs). The proposed approach is optimized through a combined loss function calculated over raw waveform, multi-resolution magnitude spectrogram, and objective metrics using a metric discriminator. Experimental results in terms of various objective metrics on VoiceBank+DEMAND and DNS-2020 challenge datasets demonstrate that the proposed speech enhancement (SE) approach outperforms the baselines and achieves state-of-the-art (SOTA) performance in the time domain. △ Less

Submitted 27 May, 2024; originally announced May 2024.

Differentially Private Ad Conversion Measurement

Authors: John Delaney, Badih Ghazi, Charlie Harrison, Christina Ilvento, Ravi Kumar, Pasin Manurangsi, Martin Pal, Karthik Prabhakar, Mariana Raykova

Abstract: In this work, we study ad conversion measurement, a central functionality in digital advertising, where an advertiser seeks to estimate advertiser website (or mobile app) conversions attributed to ad impressions that users have interacted with on various publisher websites (or mobile apps). Using differential privacy (DP), a notion that has gained in popularity due to its strong mathematical guara… ▽ More In this work, we study ad conversion measurement, a central functionality in digital advertising, where an advertiser seeks to estimate advertiser website (or mobile app) conversions attributed to ad impressions that users have interacted with on various publisher websites (or mobile apps). Using differential privacy (DP), a notion that has gained in popularity due to its strong mathematical guarantees, we develop a formal framework for private ad conversion measurement. In particular, we define the notion of an operationally valid configuration of the attribution rule, DP adjacency relation, contribution bounding scope and enforcement point. We then provide, for the set of configurations that most commonly arises in practice, a complete characterization, which uncovers a delicate interplay between attribution and privacy. △ Less

Submitted 22 March, 2024; originally announced March 2024.

Comments: To appear in PoPETS 2024

Efficient Fault Detection and Categorization in Electrical Distribution Systems Using Hessian Locally Linear Embedding on Measurement Data

Authors: Victor Sam Moses Babu K., Sidharthenee Nayak, Divyanshi Dwivedi, Pratyush Chakraborty, Chandrashekhar Narayan Bhende, Pradeep Kumar Yemula, Mayukha Pal

Abstract: Faults on electrical power lines could severely compromise both the reliability and safety of power systems, leading to unstable power delivery and increased outage risks. They pose significant safety hazards, necessitating swift detection and mitigation to maintain electrical infrastructure integrity and ensure continuous power supply. Hence, accurate detection and categorization of electrical fa… ▽ More Faults on electrical power lines could severely compromise both the reliability and safety of power systems, leading to unstable power delivery and increased outage risks. They pose significant safety hazards, necessitating swift detection and mitigation to maintain electrical infrastructure integrity and ensure continuous power supply. Hence, accurate detection and categorization of electrical faults are pivotal for optimized power system maintenance and operation. In this work, we propose a novel approach for detecting and categorizing electrical faults using the Hessian locally linear embedding (HLLE) technique and subsequent clustering with t-SNE (t-distributed stochastic neighbor embedding) and Gaussian mixture model (GMM). First, we employ HLLE to transform high-dimensional (HD) electrical data into low-dimensional (LD) embedding coordinates. This technique effectively captures the inherent variations and patterns in the data, enabling robust feature extraction. Next, we perform the Mann-Whitney U test based on the feature space of the embedding coordinates for fault detection. This statistical approach allows us to detect electrical faults providing an efficient means of system monitoring and control. Furthermore, to enhance fault categorization, we employ t-SNE with GMM to cluster the detected faults into various categories. To evaluate the performance of the proposed method, we conduct extensive simulations on an electrical system integrated with solar farm. Our results demonstrate that the proposed approach exhibits effective fault detection and clustering across a range of fault types with different variations of the same fault. Overall, this research presents an effective methodology for robust fault detection and categorization in electrical systems, contributing to the advancement of fault management practices and the prevention of system failures. △ Less

Submitted 9 March, 2024; originally announced March 2024.

Fault Classification in Electrical Distribution Systems using Grassmann Manifold

Authors: Victor Sam Moses Babu K., Sidharthenee Nayak, Divyanshi Dwivedi, Pratyush Chakraborty, Chandrashekhar Narayan Bhende, Pradeep Kumar Yemula, Mayukha Pal

Abstract: Electrical fault classification is vital for ensuring the reliability and safety of power systems. Accurate and efficient fault classification methods are essential for timely and effective maintenance. In this paper, we propose a novel approach for effective fault classification through Grassmann manifolds, which is a non-Euclidean space that captures the intrinsic structure of high-dimensional d… ▽ More Electrical fault classification is vital for ensuring the reliability and safety of power systems. Accurate and efficient fault classification methods are essential for timely and effective maintenance. In this paper, we propose a novel approach for effective fault classification through Grassmann manifolds, which is a non-Euclidean space that captures the intrinsic structure of high-dimensional data and offers a robust framework for feature extraction. We use simulated data for electrical distribution systems with various types of electrical faults. The proposed method involves transforming the measurement fault data into Grassmann manifold space using techniques from differential geometry. This transformation aids in uncovering the underlying fault patterns and reducing the computational complexity of subsequent classification steps. To achieve fault classification, we employ a machine learning technique optimized for the Grassmann manifold. The support vector machine classifier is adapted to operate within the Grassmann manifold space, enabling effective discrimination between different fault classes. The results illustrate the efficacy of the proposed Grassmann manifold-based approach for electrical fault classification which showcases its ability to accurately differentiate between various fault types. △ Less

Submitted 9 March, 2024; originally announced March 2024.

Dynamical phase transitions in $XY$ model: a Monte Carlo and mean-field theory study

Authors: Mainak Pal, William D. Baez, Pushan Majumdar, Arnab Sen, Trinanjan Datta

Abstract: We investigate the dynamical phases and phase transitions arising in a classical two-dimensional anisotropic $XY$ model under the influence of a periodically driven temporal external magnetic field in the form of a symmetric square wave. We use a combination of finite temperature classical Monte Carlo simulation, implemented within a CPU + GPU paradigm, utilizing local dynamics provided by the Gla… ▽ More We investigate the dynamical phases and phase transitions arising in a classical two-dimensional anisotropic $XY$ model under the influence of a periodically driven temporal external magnetic field in the form of a symmetric square wave. We use a combination of finite temperature classical Monte Carlo simulation, implemented within a CPU + GPU paradigm, utilizing local dynamics provided by the Glauber algorithm and a phenomenological equation-of-motion approach based on relaxational dynamics governed by the time-dependent free energy within a mean-field approximation to study the model. We investigate several parameter regimes of the variables (magnetic field, anisotropy, and the external drive frequency) that influence the anisotropic $XY$ system. We identify four possible dynamical phases -- Ising-SBO, Ising-SRO, $XY$-SBO and $XY$-SRO. Both techniques indicate that only three of them (Ising-SRO, Ising-SBO, and $XY$-SRO) are stable dynamical phases in the thermodynamic sense. Within the Monte Carlo framework, a finite size scaling analysis shows that $XY$-SBO does not survive in the thermodynamic limit giving way to either an Ising-SBO or a $XY$-SRO regime. The finite size scaling analysis further shows that the transitions between the three remaining dynamical phases either belong to the two-dimensional Ising universality class or are first-order in nature. The mean-field calculations yield three stable dynamical phases, i.e., Ising-SRO, Ising-SBO and $XY$-SRO, where the final steady state is independent of the initial condition chosen to evolve the equations of motion, as well as a region of bistability where the system either flows to Ising-SBO or $XY$-SRO (Ising-SRO) depending on the initial condition. Unlike the stable dynamical phases, the $XY$-SBO represents a transient feature that is eventually lost to either Ising-SBO or $XY$-SRO. △ Less

Submitted 28 February, 2024; v1 submitted 12 February, 2024; originally announced February 2024.

Comments: v2; added a few references, slightly modified abstract (see pdf version for the full abstract since it exceeds the arXiv limit of 1,920 characters) and main text; 16 pages including appendix, 6 figures

Energy Sharing among Resources within Electrical Distribution Systems: A Systematic Review

Authors: G Hari Krishna, K. Victor Sam Moses Babu, Divyanshi Dwivedi, Pratyush Chakraborty, Pradeep Kumar Yemula, Mayukha Pal

Abstract: The rapid increase in Electric Vehicle (EV) adoption provides a promising solution for reducing carbon emissions and fossil fuel dependency in transportation systems. However, the increasing numbers of EVs pose significant challenges to the electrical grids. In addition, the number of Distributed Energy Resources (DER) and Microgrids (MGs) is increasing on a global scale to meet the energy demand,… ▽ More The rapid increase in Electric Vehicle (EV) adoption provides a promising solution for reducing carbon emissions and fossil fuel dependency in transportation systems. However, the increasing numbers of EVs pose significant challenges to the electrical grids. In addition, the number of Distributed Energy Resources (DER) and Microgrids (MGs) is increasing on a global scale to meet the energy demand, consequently changing the energy infrastructure. Recently, energy-sharing methods have been proposed to share excess energy from DERs and EVs in Electric Vehicle Charging Infrastructure (EVCI) and MGs. Accommodating this sharing mechanism with the existing electrical distribution systems is a critical issue concerning the economic, reliability, and resilience aspects. This study examines the ever-changing field of EVCI and the critical role of peer-to-peer (P2P) energy trading in mitigating the problems with grid management that result from unorganized EV charging and intermittency in DER. Also, the possibility of energy sharing in electrical distribution systems for microgrids and EVCI on various energy-sharing methods and algorithms are discussed in detail. Furthermore, the application of market clearing algorithms like game theory, double auction theory, blockchain technology, optimization techniques, machine learning algorithms, and other models from the existing literature are presented. This paper discusses the policies, economic benefits, environmental impacts, societal advantages, and challenges in distribution systems related to sharing in EVCI and MGs. A roadmap for future research and sharing strategies is provided to guide policymakers, researchers, and industry stakeholders toward a sustainable, resilient, and efficient energy market by integrating P2P technology into EVCIs and MGs. △ Less

Submitted 3 January, 2024; originally announced January 2024.

Regression Based Anomaly Detection in Electric Vehicle State of Charge Fluctuations Through Analysis of EVCI Data

Authors: Mitikiri Sagar Babu, Yash Tiwari, vedantham Lakshmi Srinivas, Mayukha Pal

Abstract: With the increase in the number of electric vehicles (EV), there is a need for the development of the EV charging infrastructure (EVCI) to facilitate fast charging, thereby mitigating the EV congestion at charging stations. The role of the public charging station depot is to charge the vehicle, prioritizing the achievement of the desired state of charge (SoC) value for the EV battery or charging t… ▽ More With the increase in the number of electric vehicles (EV), there is a need for the development of the EV charging infrastructure (EVCI) to facilitate fast charging, thereby mitigating the EV congestion at charging stations. The role of the public charging station depot is to charge the vehicle, prioritizing the achievement of the desired state of charge (SoC) value for the EV battery or charging till the departure of the EV, whichever occurs first. The integration of cyber and physical components within EVCI defines it as a cyber physical power system (CPPS), increasing its vulnerability to diverse cyber attacks. When an EV interfaces with the EVCI, mutual exchange of data takes place via various communication protocols like the Open Charge Point Protocol (OCPP), and IEC 61850. Unauthorized access to this data by intruders leads to cyber attacks, potentially resulting in consequences like energy theft, and revenue loss. These scenarios may cause the EVCI to incur higher charges than the actual energy consumed or the EV owners to remit payments that do not correspond adequately to the amount of energy they have consumed. This article proposes an EVCI architecture connected to the utility grid and uses the EVCI data to identify the anomalies or outliers present in the EV transmitted data, particularly focusing on SoC irregularities. △ Less

Submitted 3 January, 2024; originally announced January 2024.

A connection between the boomerang uniformity and the extended differential in odd characteristic and applications

Authors: Mohit Pal, Pantelimon Stanica

Abstract: This paper makes the first bridge between the classical differential/boomerang uniformity and the newly introduced $c$-differential uniformity. We show that the boomerang uniformity of an odd APN function is given by the maximum of the entries (except for the first row/column) of the function's $(-1)$-Difference Distribution Table. In fact, the boomerang uniformity of an odd permutation APN functi… ▽ More This paper makes the first bridge between the classical differential/boomerang uniformity and the newly introduced $c$-differential uniformity. We show that the boomerang uniformity of an odd APN function is given by the maximum of the entries (except for the first row/column) of the function's $(-1)$-Difference Distribution Table. In fact, the boomerang uniformity of an odd permutation APN function equals its $(-1)$-differential uniformity. We next apply this result to easily compute the boomerang uniformity of several odd APN functions. In the second part we give two classes of differentially low-uniform functions obtained by modifying the inverse function. The first class of permutations (CCZ-inequivalent to the inverse) over a finite field $\mathbb{F}_{p^n}$ ($p$, an odd prime) is obtained from the composition of the inverse function with an order-$3$ cycle permutation, with differential uniformity $3$ if $p=3$ and $n$ is odd; $5$ if $p=13$ and $n$ is even; and $4$ otherwise. The second class is a family of binomials and we show that their differential uniformity equals~$4$. We next complete the open case of $p=3$ in the investigation started by G\" olo\u glu and McGuire (2014), for $p\geq 5$, and continued by Kölsch (2021), for $p=2$, $n\geq 5$, on the characterization of $L_1(X^{p^n-2})+L_2(X)$ (with linearized $L_1,L_2$) being a permutation polynomial. Finally, we extend to odd characteristic a result of Charpin and Kyureghyan (2010) providing an upper bound for the differential uniformity of the function and its switched version via a trace function. △ Less

Submitted 3 December, 2023; originally announced December 2023.

MSC Class: 12E20; 11T06; 94A60

Advancements in Arc Fault Detection for Electrical Distribution Systems: A Comprehensive Review from Artificial Intelligence Perspective

Authors: Kriti Thakur, Divyanshi Dwivedi, K. Victor Sam Moses Babu, Alivelu Manga Parimi, Pradeep Kumar Yemula, Pratyush Chakraborty, Mayukha Pal

Abstract: This comprehensive review paper provides a thorough examination of current advancements and research in the field of arc fault detection for electrical distribution systems. The increasing demand for electricity, coupled with the increasing utilization of renewable energy sources, has necessitated vigilance in safeguarding electrical distribution systems against arc faults. Such faults could lead… ▽ More This comprehensive review paper provides a thorough examination of current advancements and research in the field of arc fault detection for electrical distribution systems. The increasing demand for electricity, coupled with the increasing utilization of renewable energy sources, has necessitated vigilance in safeguarding electrical distribution systems against arc faults. Such faults could lead to catastrophic accidents, including fires, equipment damage, loss of human life, and other critical issues. To mitigate these risks, this review article focuses on the identification and early detection of arc faults, with a particular emphasis on the vital role of artificial intelligence (AI) in the detection and prediction of arc faults. The paper explores a wide range of methodologies for arc fault detection and highlights the superior performance of AI-based methods in accurately identifying arc faults when compared to other approaches. A thorough evaluation of existing methodologies is conducted by categorizing them into distinct groups, which provides a structured framework for understanding the current state of arc fault detection techniques. This categorization serves as a foundation for identifying the existing constraints and future research avenues in the domain of arc fault detection for electrical distribution systems. This review paper provides the state of the art in arc fault detection, aiming to enhance safety and reliability in electrical distribution systems and guide future research efforts. △ Less

Submitted 28 November, 2023; originally announced November 2023.

Modelling of the Electric Vehicle Charging Infrastructure as Cyber Physical Power Systems: A Review on Components, Standards, Vulnerabilities and Attacks

Authors: Sagar Babu Mitikiri, K. Victor Sam Moses Babu, Divyanshi Dwivedi, Vedantham Lakshmi Srinivas, Pratyush Chakraborty, Pradeep Kumar Yemula, Mayukha Pal

Abstract: The increasing number of electric vehicles (EVs) has led to the growing need to establish EV charging infrastructures (EVCIs) with fast charging capabilities to reduce congestion at the EV charging stations (EVCS) and also provide alternative solutions for EV owners without residential charging facilities. The EV charging stations are broadly classified based on i) where the charging equipment is… ▽ More The increasing number of electric vehicles (EVs) has led to the growing need to establish EV charging infrastructures (EVCIs) with fast charging capabilities to reduce congestion at the EV charging stations (EVCS) and also provide alternative solutions for EV owners without residential charging facilities. The EV charging stations are broadly classified based on i) where the charging equipment is located - on-board and off-board charging stations, and ii) the type of current and power levels - AC and DC charging stations. The DC charging stations are further classified into fast and extreme fast charging stations. This article focuses mainly on several components that model the EVCI as a cyberphysical system (CPS). △ Less

Submitted 14 November, 2023; originally announced November 2023.

Advancements in Enhancing Resilience of Electrical Distribution Systems: A Review on Frameworks, Metrics, and Technological Innovations

Authors: Divyanshi Dwivedi, Sagar Babu Mitikiri, K. Victor Sam Moses Babu, Pradeep Kumar Yemula, Vedantham Lakshmi Srininvas, Pratyush Chakraborty, Mayukha Pal

Abstract: This comprehensive review paper explores power system resilience, emphasizing its evolution, comparison with reliability, and conducting a thorough analysis of the definition and characteristics of resilience. The paper presents the resilience frameworks and the application of quantitative power system resilience metrics to assess and quantify resilience. Additionally, it investigates the relevanc… ▽ More This comprehensive review paper explores power system resilience, emphasizing its evolution, comparison with reliability, and conducting a thorough analysis of the definition and characteristics of resilience. The paper presents the resilience frameworks and the application of quantitative power system resilience metrics to assess and quantify resilience. Additionally, it investigates the relevance of complex network theory in the context of power system resilience. An integral part of this review involves examining the incorporation of data-driven techniques in enhancing power system resilience. This includes the role of data-driven methods in enhancing power system resilience and predictive analytics. Further, the paper explores the recent techniques employed for resilience enhancement, which includes planning and operational techniques. Also, a detailed explanation of microgrid (MG) deployment, renewable energy integration, and peer-to-peer (P2P) energy trading in fortifying power systems against disruptions is provided. An analysis of existing research gaps and challenges is discussed for future directions toward improvements in power system resilience. Thus, a comprehensive understanding of power system resilience is provided, which helps in improving the ability of distribution systems to withstand and recover from extreme events and disruptions. △ Less

Submitted 12 November, 2023; originally announced November 2023.

A Novel Approach to Threshold Quantum Images by using Unsharp Measurements

Authors: Ayan Barui, Mayukha Pal, Prasanta K. Panigrahi

Abstract: We propose a hybrid quantum approach to threshold and binarize a grayscale image through unsharp measurements (UM) relying on image histogram. Generally, the histograms are characterized by multiple overlap** normal distributions corresponding to objects, or image features with small but significant overlaps, making it challenging to establish suitable thresholds. The proposed methodology uses p… ▽ More We propose a hybrid quantum approach to threshold and binarize a grayscale image through unsharp measurements (UM) relying on image histogram. Generally, the histograms are characterized by multiple overlap** normal distributions corresponding to objects, or image features with small but significant overlaps, making it challenging to establish suitable thresholds. The proposed methodology uses peaks of the overlap** Gaussians and the distance between neighboring local minima as the variance, based on which the UM parameters are chosen, that maps the normal distribution into a localized delta function. To demonstrate its efficacy, subsequent implementation is done on noisy quantum environments in Qiskit. This process is iteratively repeated for a multimodal histogram to obtain more thresholds, which are then applied to various life-like pictures to get high-contrast images, resulting in comparable peak signal-to-noise ratio and structural similarity index measure values. The obtained thresholds are used to binarize a grayscale image by using novel enhanced quantum image representation integrated with a threshold encoder and an efficient quantum comparator (QC) that depicts the whole binarized picture. This approach significantly reduces the complexity of the proposed QC and of the whole algorithm when compared to earlier models. △ Less

Submitted 30 December, 2023; v1 submitted 16 October, 2023; originally announced October 2023.

DynamoPMU: A Physics Informed Anomaly Detection and Prediction Methodology using non-linear dynamics from $μ$PMU Measurement Data

Authors: Divyanshi Dwivedi, Pradeep Kumar Yemula, Mayukha Pal

Abstract: The expansion in technology and attainability of a large number of sensors has led to a huge amount of real-time streaming data. The real-time data in the electrical distribution system is collected through distribution-level phasor measurement units referred to as $μ$PMU which report high-resolution phasor measurements comprising various event signatures which provide situational awareness and en… ▽ More The expansion in technology and attainability of a large number of sensors has led to a huge amount of real-time streaming data. The real-time data in the electrical distribution system is collected through distribution-level phasor measurement units referred to as $μ$PMU which report high-resolution phasor measurements comprising various event signatures which provide situational awareness and enable a level of visibility into the distribution system. These events are infrequent, unschedule, and uncertain; it is a challenge to scrutinize, detect and predict the occurrence of such events. For electrical distribution systems, it is challenging to explicitly identify evolution functions that describe the complex, non-linear, and non-stationary signature patterns of events. In this paper, we seek to address this problem by develo** a physics dynamics-based approach to detect anomalies in the $μ$PMU streaming data and simultaneously predict the events using governing equations. We propose a data-driven approach based on the Hankel alternative view of the Koopman (HAVOK) operator, called DynamoPMU, to analyze the underlying dynamics of the distribution system by representing them in a linear intrinsic space. The key technical idea is that the proposed method separates out the linear dynamical behaviour pattern and intermittent forcing (anomalous events) in sequential data which turns out to be very useful for anomaly detection and simultaneous data prediction. We demonstrate the efficacy of our proposed framework through analysis of real $μ$PMU data taken from the LBNL distribution grid. DynamoPMU is suitable for real-time event detection as well as prediction in an unsupervised way and adapts to varying statistics. △ Less

Submitted 31 March, 2023; originally announced April 2023.

Journal ref: IEEE Transactions on Instrumentation and Measurement-2023

Identification of Forced Oscillation Sources in Wind Farms using E-SINDy

Authors: K. Victor Sam Moses Babu, Pratyush Chakraborty, Mayukha Pal

Abstract: The rapid growth of wind power generation has led to increased interest in understanding and mitigating the adverse effects of wind turbine wakes and forced oscillations in wind farms. In this paper, we model a wind farm consisting of three wind turbines connected to a distribution system. Forced oscillations due to wind shear and tower shadow are injected into the system. If these oscillations ar… ▽ More The rapid growth of wind power generation has led to increased interest in understanding and mitigating the adverse effects of wind turbine wakes and forced oscillations in wind farms. In this paper, we model a wind farm consisting of three wind turbines connected to a distribution system. Forced oscillations due to wind shear and tower shadow are injected into the system. If these oscillations are unchecked, they could pose a severe threat to the operation of the system and damage to the equipment. Identifying the source and frequency of forced oscillations in wind farms from measurement data is challenging. Thus, we propose a data-driven approach that discovers the underlying equations governing a nonlinear dynamical system from measured data using the Ensemble-Sparse Identification of Nonlinear Dynamics (E-SINDy) method. The results suggest that E-SINDy is a valuable tool for identifying sources of forced oscillations in wind farms and could facilitate the development of suitable control strategies to mitigate their negative impacts. △ Less

Submitted 21 June, 2023; v1 submitted 31 March, 2023; originally announced April 2023.

CerviFormer: A Pap-smear based cervical cancer classification method using cross attention and latent transformer

Authors: Bhaswati Singha Deo, Mayukha Pal, Prasanta K. Panigarhi, Asima Pradhan

Abstract: Purpose: Cervical cancer is one of the primary causes of death in women. It should be diagnosed early and treated according to the best medical advice, as with other diseases, to ensure that its effects are as minimal as possible. Pap smear images are one of the most constructive ways for identifying this type of cancer. This study proposes a cross-attention-based Transfomer approach for the relia… ▽ More Purpose: Cervical cancer is one of the primary causes of death in women. It should be diagnosed early and treated according to the best medical advice, as with other diseases, to ensure that its effects are as minimal as possible. Pap smear images are one of the most constructive ways for identifying this type of cancer. This study proposes a cross-attention-based Transfomer approach for the reliable classification of cervical cancer in Pap smear images. Methods: In this study, we propose the CerviFormer -- a model that depends on the Transformers and thereby requires minimal architectural assumptions about the size of the input data. The model uses a cross-attention technique to repeatedly consolidate the input data into a compact latent Transformer module, which enables it to manage very large-scale inputs. We evaluated our model on two publicly available Pap smear datasets. Results: For 3-state classification on the Sipakmed data, the model achieved an accuracy of 93.70%. For 2-state classification on the Herlev data, the model achieved an accuracy of 94.57%. Conclusion: Experimental results on two publicly accessible datasets demonstrate that the proposed method achieves competitive results when compared to contemporary approaches. The proposed method brings forth a comprehensive classification model to detect cervical cancer in Pap smear images. This may aid medical professionals in providing better cervical cancer treatment, consequently, enhancing the overall effectiveness of the entire testing process. △ Less

Submitted 17 March, 2023; originally announced March 2023.

A Resilient Power Distribution System using P2P Energy Sharing

Authors: K. Victor Sam Moses Babu, Divyanshi Dwivedi, Pratyush Chakraborty, Pradeep Kumar Yemula, Mayukha Pal

Abstract: The adoption of distributed energy resources (DERs) such as solar panels and wind turbines is transforming the traditional energy grid into a more decentralized system, where microgrids are emerging as a key concept. Peer-to-Peer (P2P) energy sharing in microgrids enhances the efficiency and flexibility of the overall system by allowing the exchange of surplus energy and better management of energ… ▽ More The adoption of distributed energy resources (DERs) such as solar panels and wind turbines is transforming the traditional energy grid into a more decentralized system, where microgrids are emerging as a key concept. Peer-to-Peer (P2P) energy sharing in microgrids enhances the efficiency and flexibility of the overall system by allowing the exchange of surplus energy and better management of energy resources. This work analyzes the impact of P2P energy sharing for three cases - within a microgrid, with neighbouring microgrids, and all microgrids combined together in a distribution system. A standard IEEE 123 node test feeder integrated with renewable energy sources is partitioned into microgrids. For P2P energy sharing between microgrids, the results show significant benefits in cost, reduced energy dependence on the grid, and a significant improvement in the system's resilience. We also predicted the energy requirement for a microgrid to evaluate energy resilience for the control and operation of the microgrid. Overall, the analysis provides valuable insights into the performance and sustainability of microgrids with P2P energy sharing. △ Less

Submitted 21 June, 2023; v1 submitted 16 March, 2023; originally announced March 2023.

Comments: arXiv admin note: text overlap with arXiv:2212.02318

Planning of Fast Charging Infrastructure for Electric Vehicles in a Distribution System and Prediction of Dynamic Price

Authors: K. Victor Sam Moses Babu, Pratyush Chakraborty, Mayukha Pal

Abstract: The increasing number of electric vehicles (EVs) has led to the need for installing public electric vehicle charging stations (EVCS) to facilitate ease of use and to support users who do not have the option of residential charging. The public electric vehicle charging infrastructures (EVCIs) must be equipped with a good number of EVCSs, with fast charging capability, to accommodate the EV traffic… ▽ More The increasing number of electric vehicles (EVs) has led to the need for installing public electric vehicle charging stations (EVCS) to facilitate ease of use and to support users who do not have the option of residential charging. The public electric vehicle charging infrastructures (EVCIs) must be equipped with a good number of EVCSs, with fast charging capability, to accommodate the EV traffic demand, which would otherwise lead to congestion at the charging stations. The location of these fast-charging infrastructures significantly impacts the distribution system (DS). We propose the optimal placement of fast-charging EVCIs at different locations in the distribution system, using multi-objective particle swarm optimization (MOPSO), so that the power loss and voltage deviations are kept at a minimum. Time-series analysis of the DS and EV load variations are performed using MATLAB and OpenDSS. We further analyze the cost benefits of the EVCIs under real-time pricing conditions and employ an autoregressive integrated moving average (ARIMA) model to predict the dynamic price. The simulated test system without any EVCI has a power loss of 164.36 kW and squared voltage deviations of 0.0235 p.u. Using the proposed method, the results obtained validate the optimal location of 5 EVCIs (each having 20 EVCSs with a 50kWh charger rating) resulting in a minimum power loss of 201.40 kW and squared voltage deviations of 0.0182 p.u. in the system. Significant cost benefits for the EVCIs are also achieved, and an R-squared value of dynamic price predictions of 0.9999 is obtained. This would allow the charging station operator to make promotional offers for maximizing utilization and increasing profits. △ Less

Submitted 22 September, 2023; v1 submitted 17 January, 2023; originally announced January 2023.

Machine Learning as an Accurate Predictor for Percolation Threshold of Diverse Networks

Authors: Siddharth Patwardhan, Utso Majumder, Aditya Das Sarma, Mayukha Pal, Divyanshi Dwivedi, Prasanta K. Panigrahi

Abstract: The percolation threshold is an important measure to determine the inherent rigidity of large networks. Predictors of the percolation threshold for large networks are computationally intense to run, hence it is a necessity to develop predictors of the percolation threshold of networks, that do not rely on numerical simulations. We demonstrate the efficacy of five machine learning-based regression… ▽ More The percolation threshold is an important measure to determine the inherent rigidity of large networks. Predictors of the percolation threshold for large networks are computationally intense to run, hence it is a necessity to develop predictors of the percolation threshold of networks, that do not rely on numerical simulations. We demonstrate the efficacy of five machine learning-based regression techniques for the accurate prediction of the percolation threshold. The dataset generated to train the machine learning models contains a total of 777 real and synthetic networks. It consists of 5 statistical and structural properties of networks as features and the numerically computed percolation threshold as the output attribute. We establish that the machine learning models outperform three existing empirical estimators of bond percolation threshold, and extend this experiment to predict site and explosive percolation. Further, we compared the performance of our models in predicting the percolation threshold using RMSE values. The gradient boosting regressor, multilayer perceptron and random forests regression models achieve the least RMSE values among considered models. △ Less

Submitted 25 March, 2023; v1 submitted 16 December, 2022; originally announced December 2022.

Detection of Malfunctioning Modules in Photovoltaic Power Plants using Unsupervised Feature Clustering Segmentation Algorithm

Authors: Divyanshi Dwivedi, Pradeep Kumar Yemula, Mayukha Pal

Abstract: The energy transition towards photovoltaic solar energy has evolved to be a viable and sustainable source for the generation of electricity. It has effectively emerged as an alternative to the conventional mode of electricity generation for develo** countries to meet their energy requirement. Thus, many solar power plants have been set up across the globe. However, in these large-scale or remote… ▽ More The energy transition towards photovoltaic solar energy has evolved to be a viable and sustainable source for the generation of electricity. It has effectively emerged as an alternative to the conventional mode of electricity generation for develo** countries to meet their energy requirement. Thus, many solar power plants have been set up across the globe. However, in these large-scale or remote solar power plants, monitoring and maintenance persist as challenging tasks, mainly identifying faulty or malfunctioning cells in photovoltaic (PV) panels. In this paper, we use an unsupervised deep-learning image segmentation model for the detection of internal faults such as hot spots and snail trails in PV panels. Generally, training or ground truth labels are not available for large solar power plants, thus the proposed model is highly recommended as it does not require any prior learning or training. It extracts the features from the input image and segments out the faults in the image. Here we use infrared thermal images of the PV panel as input, passed to a convolutional neural network which assigns cluster labels to the pixels. Further, optimize the pixel labels, features and model parameters using backpropagation based on iterative stochastic gradient descent. Then, we compute similarity loss and spatial continuity loss to assign the same label to the pixel with similar features and spatial continuity to reduce noises in the image segmentation process. The effectiveness of the proposed approach was examined on an online available dataset for the recognition of snail trails and hot spot failures in monocrystalline solar panels. △ Less

Submitted 30 December, 2022; originally announced December 2022.

Minimally driven Kapitza oscillator: A pedagogical perspective from Newtonian mechanics and geometry

Authors: Mainak Pal

Abstract: The unstable top-equilibrium point of a simple pendulum turns stable when its pivot point is given a fast and strong enough vertical vibration. Known as the Kapitza oscillator, it has four symmetrically spaced points of equilibrium in absence of gravity, out of which two are stable and two are unstable. This article, completely based on a geometric argument and an elementary intuition in Newtonian… ▽ More The unstable top-equilibrium point of a simple pendulum turns stable when its pivot point is given a fast and strong enough vertical vibration. Known as the Kapitza oscillator, it has four symmetrically spaced points of equilibrium in absence of gravity, out of which two are stable and two are unstable. This article, completely based on a geometric argument and an elementary intuition in Newtonian mechanics, is a visual and pedagogical exposition of (a) why the oscillator has four symmetrically spaced equilibrium points in absence of gravity, (b) which of them are stable or unstable, (c) why they are so and (d) how the stability and position and number of the equilibrium points change when gravity is turned on gradually along the line of vibration of the pivot of the oscillator. A minimal impulsive drive of the pivot is sufficient to illustrate the bare bones of the phenomenon. I propose a construction that can sustain the minimal drive passively in absence of dissipative forces, or actively if all dissipative forces can't be eliminated. In either of the cases, the discussed arguments apply. △ Less

Submitted 15 December, 2022; originally announced December 2022.

Comments: 5 pages, 4 figures

Topological superconductivity driven by correlations and linear defects in multiband superconductors

Authors: Mainak Pal, Andreas Kreisel, P. J. Hirschfeld

Abstract: There have been several proposals for platforms sustaining topological superconductivity in high temperature superconductors, in order to make use of the larger superconducting gap and the expected robustness of Majorana zero modes towards perturbations. In particular, the iron-based materials offer relatively large $T_c$ and nodeless energy gaps. In addition, atomically flat surfaces enable the e… ▽ More There have been several proposals for platforms sustaining topological superconductivity in high temperature superconductors, in order to make use of the larger superconducting gap and the expected robustness of Majorana zero modes towards perturbations. In particular, the iron-based materials offer relatively large $T_c$ and nodeless energy gaps. In addition, atomically flat surfaces enable the engineering of defect structures and the subsequent measurement of spectroscopic properties to reveal topological aspects. From a theory perspective, a materials-specific description is challenging due to the correlated nature of the materials and complications arising from the multiband nature of the electronic structure. Here we include both aspects in realistic interacting models, and find that the correlations themselves can lead to local magnetic order close to linear potential scattering defects at the surface of the superconductor. Using a self-consistent Bogoliubov-de Gennes framework in a real-space setup using a prototype electronic structure, we allow for arbitrary magnetic orders and show how a topological superconducting state emerges. The calculation of the topological invariant and the topological gap allows us to map out the phase diagram for the case of a linear chain of potential scatterers. While intrinsic spin-orbit coupling is not needed to enter the topological state in presence of spin-spiral states, it enlarges the topological phase. We discuss the interplay of a triplet component of the superconducting order parameter and the spin spiral leading effectively to extended spin orbit coupling terms, and connect our results to experimental efforts on the Fe(Se,Te) system. △ Less

Submitted 5 December, 2022; originally announced December 2022.

Comments: 13 pages, 9 figures

Evaluation of Energy Resilience and Cost Benefit in Microgrid with Peer-to-Peer Energy Trading

Authors: Divyanshi Dwivedi, K. Victor Sam Moses Babu, Pradeep Kumar Yemula, Pratyush Chakraborty, Mayukha Pal

Abstract: Integration of distributed energy resources (DERs) at a residential level has given way to the development of a self-sufficient energy-resilient system and peer-to-peer (P2P) energy trading at a community level. Several researchers have developed models to implement energy trading in the P2P network with DERs that enables prosumers to avail cost benefits. However, the impact of DERs and P2P energy… ▽ More Integration of distributed energy resources (DERs) at a residential level has given way to the development of a self-sufficient energy-resilient system and peer-to-peer (P2P) energy trading at a community level. Several researchers have developed models to implement energy trading in the P2P network with DERs that enables prosumers to avail cost benefits. However, the impact of DERs and P2P energy network on the system's resilience is not evaluated with a quantifiable measure. In this work, we use the percolation threshold as a measure to quantify energy resilience, which can be computed using complex network. We consider a standard IEEE-123 node test feeder integrated with renewable energy sources and partitioned into microgrids. The microgrid with the highest percolation threshold value is considered for P2P energy trading analysis. We use cooperative game theory to model the P2P energy trading so that all prosumers take part rationally. The simulation is carried out using time series data of houses with varying load profiles, solar generation, and energy storage capacities. The results show 5.81\% of total cost benefit in a year and 10.67\% improvement in the resilience of the system due to P2P energy trading. △ Less

Submitted 5 December, 2022; originally announced December 2022.

Identification of Surface Defects on Solar PV Panels and Wind Turbine Blades using Attention based Deep Learning Model

Authors: Divyanshi Dwivedi, K. Victor Sam Moses Babu, Pradeep Kumar Yemula, Pratyush Chakraborty, Mayukha Pal

Abstract: The global generation of renewable energy has rapidly increased, primarily due to the installation of large-scale renewable energy power plants. However, monitoring renewable energy assets in these large plants remains challenging due to environmental factors that could result in reduced power generation, malfunctioning, and degradation of asset life. Therefore, the detection of surface defects on… ▽ More The global generation of renewable energy has rapidly increased, primarily due to the installation of large-scale renewable energy power plants. However, monitoring renewable energy assets in these large plants remains challenging due to environmental factors that could result in reduced power generation, malfunctioning, and degradation of asset life. Therefore, the detection of surface defects on renewable energy assets is crucial for maintaining the performance and efficiency of these plants. This paper proposes an innovative detection framework to achieve an economical surface monitoring system for renewable energy assets. High-resolution images of the assets are captured regularly and inspected to identify surface or structural damages on solar panels and wind turbine blades. {Vision transformer (ViT), one of the latest attention-based deep learning (DL) models in computer vision, is proposed in this work to classify surface defects.} The ViT model outperforms other DL models, including MobileNet, VGG16, Xception, EfficientNetB7, and ResNet50, achieving high accuracy scores above 97\% for both wind and solar plant assets. From the results, our proposed model demonstrates its potential for monitoring and detecting damages in renewable energy assets for efficient and reliable operation of renewable power plants. △ Less

Submitted 8 January, 2024; v1 submitted 22 November, 2022; originally announced November 2022.

Simulating Superconducting Properties of Overdoped Cuprates: the Role of Inhomogeneity

Authors: Mainak Pal, Andreas Kreisel, W. A. Atkinson, P. J. Hirschfeld

Abstract: Theoretical studies of disordered $d$-wave superconductors have focused, with a few exceptions, on optimally doped models with strong scatterers. Addressing recent controversies about the nature of the overdoped cuprates, however, requires studies of the weaker scattering associated with dopant atoms. Here we study simple models of such systems in the self-consistent Bogoliubov-de Gennes (BdG) fra… ▽ More Theoretical studies of disordered $d$-wave superconductors have focused, with a few exceptions, on optimally doped models with strong scatterers. Addressing recent controversies about the nature of the overdoped cuprates, however, requires studies of the weaker scattering associated with dopant atoms. Here we study simple models of such systems in the self-consistent Bogoliubov-de Gennes (BdG) framework, and compare to disorder-averaged results using the self-consistent-T-matrix-approximation (SCTMA). Despite surprisingly linear in energy behavior of the low-energy density of states even for quite disordered systems, the superfluid density in such cases retains a quadratic low-temperature variation of the penetration depth, unlike other BdG results reported recently. We trace the discrepancy to smaller effective system size employed in that work. Overall, the SCTMA performs remarkably well, with the exception of highly disordered systems with strongly suppressed superfluid density. We explore this interesting region where gap inhomogeneity dominates measured superconducting properties, and compare with overdoped cuprates. △ Less

Submitted 23 November, 2022; originally announced November 2022.

Comments: 12 pages, 9 figures

Local Interpretable Model Agnostic Shap Explanations for machine learning models

Authors: P. Sai Ram Aditya, Mayukha Pal

Abstract: With the advancement of technology for artificial intelligence (AI) based solutions and analytics compute engines, machine learning (ML) models are getting more complex day by day. Most of these models are generally used as a black box without user interpretability. Such complex ML models make it more difficult for people to understand or trust their predictions. There are variety of frameworks us… ▽ More With the advancement of technology for artificial intelligence (AI) based solutions and analytics compute engines, machine learning (ML) models are getting more complex day by day. Most of these models are generally used as a black box without user interpretability. Such complex ML models make it more difficult for people to understand or trust their predictions. There are variety of frameworks using explainable AI (XAI) methods to demonstrate explainability and interpretability of ML models to make their predictions more trustworthy. In this manuscript, we propose a methodology that we define as Local Interpretable Model Agnostic Shap Explanations (LIMASE). This proposed ML explanation technique uses Shapley values under the LIME paradigm to achieve the following (a) explain prediction of any model by using a locally faithful and interpretable decision tree model on which the Tree Explainer is used to calculate the shapley values and give visually interpretable explanations. (b) provide visually interpretable global explanations by plotting local explanations of several data points. (c) demonstrate solution for the submodular optimization problem. (d) also bring insight into regional interpretation e) faster computation compared to use of kernel explainer. △ Less

Submitted 10 October, 2022; originally announced October 2022.

Comments: Under review

Data-Driven Approach to form Energy Resilient Smart Microgrids with Identification of Vulnerable Nodes in Active Electrical Distribution Network

Authors: D Maneesh Reddy, Divyanshi Dwivedi, Pradeep Kumar Yemula, Mayukha Pal

Abstract: With the commitment to climate, globally many countries started reducing brownfield energy production and strongly opting towards green energy resources. However, the optimal allocation of distributed energy resources (DERs) in electrical distribution systems still pertains as a challenging issue to attain the maximum benefits. It happens due to the systems complex behaviour and inappropriate inte… ▽ More With the commitment to climate, globally many countries started reducing brownfield energy production and strongly opting towards green energy resources. However, the optimal allocation of distributed energy resources (DERs) in electrical distribution systems still pertains as a challenging issue to attain the maximum benefits. It happens due to the systems complex behaviour and inappropriate integration of DERs that adversely affects the distribution grid. In this work, we propose a methodology for the optimal allocation of DERs with vulnerable node identification in active electrical distribution networks. A failure or extreme event at the vulnerable node would interrupt the power flow in the distribution network. Also, the power variation in these vulnerable nodes would significantly affect the operation of other linked nodes. Thus, these nodes are found suitable for the optimal placement of DERs. We demonstrate the proposed data-driven approach on a standard IEEE-123 bus test feeder. Initially, we partitioned the distribution system into optimal microgrids using graph theory and graph neural network (GNN) architecture. Further, using Granger causality analysis, we identified vulnerable nodes in the partitioned microgrid; suitable for DERs integration. The placement of DERs on the vulnerable nodes enhanced network reliability and resilience. Improvement in resilience is validated by computing the percolation threshold for the microgrid networks. The results show a 20.45% improvement in the resilience of the system due to the optimal allocation of DERs. △ Less

Submitted 25 March, 2023; v1 submitted 24 August, 2022; originally announced August 2022.

Evaluating the Planning and Operational Resilience of Electrical Distribution Systems with Distributed Energy Resources using Complex Network Theory

Authors: Divyanshi Dwivedi, Pradeep Kumar Yemula, Mayukha Pal

Abstract: Electrical Distribution Systems are extensively penetrated with Distributed Energy Resources (DERs) to cater the energy demands with the general perception that it enhances the system's resilience. However, integration of DERs may adversely affect the grid operation and affect the system resilience due to various factors like their intermittent availability, dynamics of weather conditions, non-lin… ▽ More Electrical Distribution Systems are extensively penetrated with Distributed Energy Resources (DERs) to cater the energy demands with the general perception that it enhances the system's resilience. However, integration of DERs may adversely affect the grid operation and affect the system resilience due to various factors like their intermittent availability, dynamics of weather conditions, non-linearity, complexity, number of malicious threats, and improved reliability requirements of consumers. This paper proposes a methodology to evaluate the planning and operational resilience of power distribution systems under extreme events and determines the withstand capability of the electrical network. The proposed framework is developed by effectively employing the complex network theory. Correlated networks for undesirable configurations are developed from the time series data of active power monitored at nodes of the electrical network. For these correlated networks, computed the network parameters such as clustering coefficient, assortative coefficient, average degree and power law exponent for the anticipation; and percolation threshold for the determination of the network withstand capability under extreme conditions. The proposed methodology is also suitable for identifying the hosting capacity of solar panels in the system while maintaining resilience under different unfavourable conditions and identifying the most critical nodes of the system that could drive the system into non-resilience. This framework is demonstrated on IEEE 123 node test feeder by generating active power time-series data for a variety of electrical conditions using simulation software, GridLAB-D. The percolation threshold resulted as an effective metric for the determination of the planning and operational resilience of the power distribution system. △ Less

Submitted 6 July, 2023; v1 submitted 24 August, 2022; originally announced August 2022.

Journal ref: Renewable Energy Focus-2023

Some new classes of (almost) perfect $c$-nonlinear permutations

Authors: Mohit Pal

Abstract: The concept of differential uniformity was recently extended to the $c$-differential uniformity. An interesting problem in this area is the construction of functions with low $c$-differential uniformity and a lot of research has been done in this direction in the recent past. Here, we present three classes of (almost) perfect $c$-nonlinear permutations over finite fields of even characteristic. The concept of differential uniformity was recently extended to the $c$-differential uniformity. An interesting problem in this area is the construction of functions with low $c$-differential uniformity and a lot of research has been done in this direction in the recent past. Here, we present three classes of (almost) perfect $c$-nonlinear permutations over finite fields of even characteristic. △ Less

Submitted 1 August, 2022; originally announced August 2022.

MSC Class: 11T06; 94A60

Effect of Non-local Grazing on Dry-land Vegetation Dynamics

Authors: Mrinal Kanti Pal, Swarup Poria

Abstract: Dry-land ecosystem has turned into a matter of grave concern, due to growing threat of land degradation and bioproductivity-loss. Self-organized vegetation patterns are a remarkable characteristic of these ecosystems; apart from being visually captivating, patterns modulate the system-response to increasing environmental stress. Empirical studies hinted that herbivory is one the key regulatory mec… ▽ More Dry-land ecosystem has turned into a matter of grave concern, due to growing threat of land degradation and bioproductivity-loss. Self-organized vegetation patterns are a remarkable characteristic of these ecosystems; apart from being visually captivating, patterns modulate the system-response to increasing environmental stress. Empirical studies hinted that herbivory is one the key regulatory mechanism behind pattern formation and overall ecosystem functioning. However most of the mathematical models have taken a mean-field strategy to grazing; foraging has been considered to be independent of spatial distribution of vegetation. To this end, an extended version of the celebrated plant-water model due to Klausmeier, has been taken as the base here. To encompass the effect of heterogeneous vegetation distribution on foraging intensity and subsequent impact on entire ecosystem, grazing is considered here to depend on spatially weighted average vegetation density, instead of density at a particular point. Moreover, varying influence of vegetation at any location over gazing elsewhere, is incorporated by choosing suitable averaging function. A comprehensive analysis demonstrates that inclusion of spatial non-locality, alters the understanding of system dynamics significantly. The grazing ecosystem is found to be more resilient to increasing aridity than it was anticipated to be in earlier studies on non-local grazing. The system-response to rising environmental pressure is also observed to vary depending on the grazer. Obtained results also suggest possibility of multi-stability due the history-dependence of system-response. Overall, this work indicates that the spatial heterogeneity in grazing intensity has a decisive role to play in the functioning of water-limited ecosystems. △ Less

Submitted 24 September, 2022; v1 submitted 14 June, 2022; originally announced June 2022.

Comments: 13 pages, 6 figures

Quantum order-by-disorder induced phase transition in Rydberg ladders with staggered detuning

Authors: Madhumita Sarkar, Mainak Pal, Arnab Sen, K. Sengupta

Abstract: $^{87}{\rm Rb}$ atoms are known to have long-lived Rydberg excited states with controllable excitation amplitude (detuning) and strong repulsive van der Waals interaction $V_{{\bf r} {\bf r'}}$ between excited atoms at sites ${\bf r}$ and ${\bf r'}$. Here we study such atoms in a two-leg ladder geometry in the presence of both staggered and uniform detuning with amplitudes $Δ$ and $λ… ▽ More $^{87}{\rm Rb}$ atoms are known to have long-lived Rydberg excited states with controllable excitation amplitude (detuning) and strong repulsive van der Waals interaction $V_{{\bf r} {\bf r'}}$ between excited atoms at sites ${\bf r}$ and ${\bf r'}$. Here we study such atoms in a two-leg ladder geometry in the presence of both staggered and uniform detuning with amplitudes $Δ$ and $λ$ respectively. We show that when $V_{\bf r r'} \gg(\ll) Δ, λ$ for $|{\bf r}-{\bf r'}|=1(>1)$, these ladders host a plateau for a wide range of $λ/Δ$ where the ground states are selected by a quantum order-by-disorder mechanism from a macroscopically degenerate manifold of Fock states with fixed Rydberg excitation density $1/4$. Our study further unravels the presence of an emergent Ising transition stabilized via the order-by-disorder mechanism inside the plateau. We identify the competing terms responsible for the transition and estimate a critical detuning $λ_c/Δ=1/3$ which agrees well with exact-diagonalization based numerical studies. We also study the fate of this transition for a realistic interaction potential $V_{{\bf r} {\bf r'}} = V_0 /|{\bf r}-{\bf r'}|^6$, demonstrate that it survives for a wide range of $V_0$, and provide analytic estimate of $λ_c$ as a function of $V_0$. This allows for the possibility of a direct verification of this transition in standard experiments which we discuss. △ Less

Submitted 10 September, 2022; v1 submitted 26 April, 2022; originally announced April 2022.

Comments: v2; 22 pages, 9 figs. Minor corrections

Journal ref: SciPost Phys. 14, 004 (2023)

Synthetic speech detection using meta-learning with prototypical loss

Authors: Monisankha Pal, Aditya Raikar, Ashish Panda, Sunil Kumar Kopparapu

Abstract: Recent works on speech spoofing countermeasures still lack generalization ability to unseen spoofing attacks. This is one of the key issues of ASVspoof challenges especially with the rapid development of diverse and high-quality spoofing algorithms. In this work, we address the generalizability of spoofing detection by proposing prototypical loss under the meta-learning paradigm to mimic the unsee… ▽ More Recent works on speech spoofing countermeasures still lack generalization ability to unseen spoofing attacks. This is one of the key issues of ASVspoof challenges especially with the rapid development of diverse and high-quality spoofing algorithms. In this work, we address the generalizability of spoofing detection by proposing prototypical loss under the meta-learning paradigm to mimic the unseen test scenario during training. Prototypical loss with metric-learning objectives can learn the embedding space directly and emerges as a strong alternative to prevailing classification loss functions. We propose an anti-spoofing system based on squeeze-excitation Residual network (SE-ResNet) architecture with prototypical loss. We demonstrate that the proposed single system without any data augmentation can achieve competitive performance to the recent best anti-spoofing systems on ASVspoof 2019 logical access (LA) task. Furthermore, the proposed system with data augmentation outperforms the ASVspoof 2021 challenge best baseline both in the progress and evaluation phase of the LA task. On ASVspoof 2019 and 2021 evaluation set LA scenario, we attain a relative 68.4% and 3.6% improvement in min-tDCF compared to the challenge best baselines, respectively. △ Less

Submitted 24 January, 2022; originally announced January 2022.

Spectral States of OJ 287 blazar from Multi-wavelength Observations with $AstroSat$

Authors: K. P. Singh, P. Kushwaha, A. Sinha, Main Pal, G. Dewangan, A. Agarwal

Abstract: We present {\it AstroSat} soft X-ray, near-UV (NUV), and far-UV (FUV) observations of a blazar, OJ~287, carried out in 2017, 2018, and 2020. The simultaneous observations with NuSTAR in 2017 provide a broad-band look encompassing NUV, FUV, soft and hard X-rays. Captured in three different broadband spectral states in three observations, the X-ray spectrum is found to be the hardest during 2018, wh… ▽ More We present {\it AstroSat} soft X-ray, near-UV (NUV), and far-UV (FUV) observations of a blazar, OJ~287, carried out in 2017, 2018, and 2020. The simultaneous observations with NuSTAR in 2017 provide a broad-band look encompassing NUV, FUV, soft and hard X-rays. Captured in three different broadband spectral states in three observations, the X-ray spectrum is found to be the hardest during 2018, while the high-energy-end of the simultaneous optical-FUV spectrum shows a steepening that is modeled with a broken power-law spectrum. The spectral energy distribution (SED) in 2017 shows a relatively flatter optical-FUV and soft X-ray spectra, implying an additional emission component. The 2020 optical-FUV spectrum is harder than in 2017 and 2018, with an extremely soft X-ray spectrum and a hardening above $\sim$1 GeV, similar to the SEDs of High-energy-peaked BL Lac objects (HBL), thereby establishing that this additional emission component has HBL-like properties. The {\it AstroSat} multi-wavelength observations trace the spectral evolution from the end-phase of the HBL component in 2017 to its disappearance in 2018 followed by its revival in 2020. A single zone leptonic model reproduces the 2018 broadband spectrum while the 2017 and 2020 SEDs require an additional HBL-like emitting zone. The spectral evolution of the high-energy-end of the optical-UV spectrum, revealed by the FUV observations in 2017 and 2018, strongly suggests that X-ray spectral changes in the normal broadband spectral state of OJ~287 are primarily due to the evolution of the optical-UV synchrotron spectrum. △ Less

Submitted 28 October, 2021; originally announced October 2021.

Comments: 13 pages, 5 Figures, 3 Tables, MNRAS (in press)

Complex optical/UV and X-ray variability in Seyfert 1 galaxy Mrk 509

Authors: Neeraj Kumari, Main Pal, Sachindra Naik, Arghajit Jana, Gaurava K. Jaisawal, Pankaj kushwaha

Abstract: We performed a detailed spectral and timing analysis of a Seyfert 1 galaxy Mrk 509 using data from the Neil Gehrels Swift observatory that spanned over ~13 years between 2006 and 2019. To study the variability properties from the optical/UV to X-ray emission, we used a total of 275 pointed observations in this work. The average spectrum over the entire duration exhibits a strong soft X-ray excess… ▽ More We performed a detailed spectral and timing analysis of a Seyfert 1 galaxy Mrk 509 using data from the Neil Gehrels Swift observatory that spanned over ~13 years between 2006 and 2019. To study the variability properties from the optical/UV to X-ray emission, we used a total of 275 pointed observations in this work. The average spectrum over the entire duration exhibits a strong soft X-ray excess above the power-law continuum. The soft X-ray excess is well described by two thermal components with temperatures of kT_BB1 ~120 eV and kT_BB2 ~460 eV. The warm thermal component is likely due to the presence of an optically thick and warm Comptonizing plasma in the inner accretion disk. The fractional variability amplitude is found to be decreasing with increasing wavelength, i.e. from the soft X-ray to UV/optical emission. However, the hard X-ray (2-8 keV) emission shows very low variability. The strength of the correlation within the UV and the optical bands (0.95-0.99) is found to be stronger than the correlation between the UV/Optical and X-ray bands (0.40-0.53). These results clearly suggest that the emitting regions of the X-ray and UV/optical emission are likely distinct or partly interacting. Having removed the slow variations in the light curves, we find that the lag spectrum is well described by the 4/3 rule for the standard Shakura-Sunyaev accretion disk when we omit X-ray lags. All these results suggest that the real disk is complex, and the UV emission is likely reprocessed in the accretion disk to give X-ray and optical emission. △ Less

Submitted 26 July, 2021; originally announced July 2021.

Comments: This article has been accepted to the Publications of the Astronomical Society of Australia (PASA). This article contains 17 pages and 13 figures

Journal ref: Publ. Astron. Soc. Aust. 38 (2021) e042

AstroSat View of Blazar OJ 287: A complete evolutionary cycle of HBL Component from end-phase to disappearance and Re-emergence

Authors: Pankaj Kushwaha, K. P. Singh, A. Sinha, Main Pal, G. Dewangan, A. Agarwal

Abstract: We report three AstroSat observations of BL Lacertae object OJ 287. The three observations caught it in very different flux states that are connected to different broadband spectral states. These observations trace the source spectral evolution from the end-phase of activity driven by a new, additional HBL like emission component in 2017 to its complete disappearance in 2018 and re-emergence in 20… ▽ More We report three AstroSat observations of BL Lacertae object OJ 287. The three observations caught it in very different flux states that are connected to different broadband spectral states. These observations trace the source spectral evolution from the end-phase of activity driven by a new, additional HBL like emission component in 2017 to its complete disappearance in 2018 and re-emergence in 2020. The 2017 observation shows a comparatively flatter optical-UV and X-ray spectrum. Supplementing it with the simultaneous NuSTAR monitoring indicates a hardening at the high-energy-end. The 2018 observation shows a harder X-ray spectrum and a sharp decline or cutoff in the optical-UV spectrum, revealed thanks to the Far-UV data from AstroSat. The brightest of all, the 2020 observation shows a hardened optical-UV spectrum and an extremely soft X-ray spectrum, constraining the low-energy peak of spectral energy distribution at UV energies -- a characteristic of HBL blazars. The contemporaneous MeV-GeV spectra from LAT show the well-known OJ 287 spectrum during 2018 but a flatter spectrum during 2017 and a hardening above ~1 GeV during 2020. Modeling broadband SEDs show that 2018 emission can be reproduced with a one-zone leptonic model while 2017 and 2020 observations need a two-zone model, with the additional zone emitting an HBL radiation. △ Less

Submitted 22 July, 2021; originally announced July 2021.

Comments: 2 figures, 7 pages

Boomerang uniformity of a class of power maps

Authors: Sartaj Ul Hasan, Mohit Pal, Pantelimon Stanica

Abstract: We consider the boomerang uniformity of an infinite class of (locally-APN) power maps and show that its boomerang uniformity over the finite field $\F_{2^n}$ is $2$ and $4$, when $n \equiv 0 \pmod 4$ and $n \equiv 2 \pmod 4$, respectively. As a consequence, we show that for this class of power maps, the differential uniformity is strictly greater than its boomerang uniformity. We consider the boomerang uniformity of an infinite class of (locally-APN) power maps and show that its boomerang uniformity over the finite field $\F_{2^n}$ is $2$ and $4$, when $n \equiv 0 \pmod 4$ and $n \equiv 2 \pmod 4$, respectively. As a consequence, we show that for this class of power maps, the differential uniformity is strictly greater than its boomerang uniformity. △ Less

Submitted 6 September, 2021; v1 submitted 10 May, 2021; originally announced May 2021.

Comments: 11 pages

MSC Class: 12E20; 11T06; 94A60

Journal ref: Designs, Codes and Cryptography, 2021

SAR and Optical data fusion based on Anisotropic Diffusion with PCA and Classification using Patch-based with LBP

Authors: Achala Shakya, Mantosh Biswas, Mahesh Pal

Abstract: SAR (VV and VH polarization) and optical data are widely used in image fusion to use the complimentary information of each other and to obtain the better-quality image (in terms of spatial and spectral features) for the improved classification results. This paper uses anisotropic diffusion with PCA for the fusion of SAR and optical data and patch-based SVM Classification with LBP (LBP-PSVM). Fusio… ▽ More SAR (VV and VH polarization) and optical data are widely used in image fusion to use the complimentary information of each other and to obtain the better-quality image (in terms of spatial and spectral features) for the improved classification results. This paper uses anisotropic diffusion with PCA for the fusion of SAR and optical data and patch-based SVM Classification with LBP (LBP-PSVM). Fusion results with VV polarization performed better than VH polarization using considered fusion method. For classification, the performance of LBP-PSVM using S1 (VV) with S2, S1 (VH) with S2 is compared with SVM classifier (without patch) and PSVM classifier (with patch), respectively. Classification results suggests that the LBP-PSVM classifier is more effective in comparison to SVM and PSVM classifiers for considered data. △ Less

Submitted 20 January, 2021; originally announced January 2021.

Spectro-Timing Analysis of a highly variable narrow-line Seyfert 1 galaxy NGC 4748 with AstroSat and XMM-Newton

Authors: Main Pal, Neeraj Kumari, Pankaj Kushwaha, K. P. Singh, Alok C. Gupta, Sachindra Naik, G. C. Dewangan, P. Tripathi, Rathin Adhikari, O. Adegoke, H. Nandan

Abstract: We present a detailed timing and spectral study of an extremely variable narrow-line Seyfert~1 galaxy NGC 4748 using observations in the year 2017 and 2014 performed with AstroSat and XMM-Newton, respectively. Both observations show extremely variable soft and hard X-ray emission that are correlated with each other. In the 2014 data set, the source retains its general behaviour of "softer when bri… ▽ More We present a detailed timing and spectral study of an extremely variable narrow-line Seyfert~1 galaxy NGC 4748 using observations in the year 2017 and 2014 performed with AstroSat and XMM-Newton, respectively. Both observations show extremely variable soft and hard X-ray emission that are correlated with each other. In the 2014 data set, the source retains its general behaviour of "softer when brighter" while the 2017 observation exhibits a "harder when brighter" nature. Such changing behaviour is rare in AGNs and is usually observed in the black hole binary systems. The "harder when brighter" is confirmed with the anti-correlation between the photon index and the 0.3-10 keV power-law flux. This suggests a possible change in the accretion mode from standard to the advection-dominated flow. Additionally, both the observations show soft X-ray excess below 2 keV over the power-law continuum. This excess was fitted with a single or multiple blackbody component(s). The origin of soft excess during the 2017 observation is likely due to the cool Comptonization as the photon index changes with time. On the other hand, the broad iron line and delayed UV emission during the 2014 observation strongly suggest that X-ray illumination onto the accretion disk and reflection and reprocessing play a significant role in this AGN. △ Less

Submitted 12 January, 2021; originally announced January 2021.

Comments: 13 pages, 9 figures, 2 tables; accepted for publication in the Journal of Astrophysics and Astronomy

Magnetic anisotropy from linear defect structures in correlated electron systems

Authors: Mainak Pal, Laetitia Bettmann, Andreas Kreisel, P. J. Hirschfeld

Abstract: Correlated electron systems, particularly iron-based superconductors, are extremely sensitive to strain, which inevitably occurs in the crystal growth process. Built-in strain of this type has been proposed as a possible explanation for experiments where nematic order has been observed at high temperatures corresponding to the nominally tetragonal phase of iron-based superconductors. Strain is ass… ▽ More Correlated electron systems, particularly iron-based superconductors, are extremely sensitive to strain, which inevitably occurs in the crystal growth process. Built-in strain of this type has been proposed as a possible explanation for experiments where nematic order has been observed at high temperatures corresponding to the nominally tetragonal phase of iron-based superconductors. Strain is assumed to produce linear defect structures, e.g. dislocations, which are quite similar to O vacancy chainlets in the underdoped cuprate superconductor YBCO. Here we investigate a simple microscopic model of dislocations in the presence of electronic correlations, which create defect states that can drive magnetic anisotropy of this kind, if spin orbit interaction is present. We estimate the contribution of these dislocations to magnetic anisotropy as detected by current torque magnetometry experiments in both cuprates and Fe-based systems. △ Less

Submitted 15 July, 2021; v1 submitted 7 December, 2020; originally announced December 2020.

Comments: 10 pages, 9 figures

Journal ref: Phys. Rev. B 103, 245132 (2021)

Adversarial defense for deep speaker recognition using hybrid adversarial training

Authors: Monisankha Pal, Arindam Jati, Raghuveer Peri, Chin-Cheng Hsu, Wael AbdAlmageed, Shrikanth Narayanan

Abstract: Deep neural network based speaker recognition systems can easily be deceived by an adversary using minuscule imperceptible perturbations to the input speech samples. These adversarial attacks pose serious security threats to the speaker recognition systems that use speech biometric. To address this concern, in this work, we propose a new defense mechanism based on a hybrid adversarial training (HA… ▽ More Deep neural network based speaker recognition systems can easily be deceived by an adversary using minuscule imperceptible perturbations to the input speech samples. These adversarial attacks pose serious security threats to the speaker recognition systems that use speech biometric. To address this concern, in this work, we propose a new defense mechanism based on a hybrid adversarial training (HAT) setup. In contrast to existing works on countermeasures against adversarial attacks in deep speaker recognition that only use class-boundary information by supervised cross-entropy (CE) loss, we propose to exploit additional information from supervised and unsupervised cues to craft diverse and stronger perturbations for adversarial training. Specifically, we employ multi-task objectives using CE, feature-scattering (FS), and margin losses to create adversarial perturbations and include them for adversarial training to enhance the robustness of the model. We conduct speaker recognition experiments on the Librispeech dataset, and compare the performance with state-of-the-art projected gradient descent (PGD)-based adversarial training which employs only CE objective. The proposed HAT improves adversarial accuracy by absolute 3.29% and 3.18% for PGD and Carlini-Wagner (CW) attacks respectively, while retaining high accuracy on benign examples. △ Less

Submitted 29 October, 2020; originally announced October 2020.

Comments: Submitted to ICASSP 2021

Blazar OJ 287 After First VHE Activity: Tracking the Re-emergence of the HBL like Component in 2020

Authors: Pankaj Kushwaha, Main Pal, Nibedita Kalita, Neeraj Kumari, Sachindra Naik, Alok C Gupta, E. M. de Gouveia Dal Pino, Minfeng Gu

Abstract: We report the re-emergence of a new broadband emission through a detailed and systematic study of the multi-wavelength spectral and temporal behavior of OJ 287 after its first-ever reported VHE activity in 2017 to date, which includes the second-highest X-ray flux of the source. The source shows high optical to X-ray flux variations, accompanied mainly by strong spectral changes. The optical to X-… ▽ More We report the re-emergence of a new broadband emission through a detailed and systematic study of the multi-wavelength spectral and temporal behavior of OJ 287 after its first-ever reported VHE activity in 2017 to date, which includes the second-highest X-ray flux of the source. The source shows high optical to X-ray flux variations, accompanied mainly by strong spectral changes. The optical to X-ray flux variations are correlated and simultaneous except for two durations when they are anti-correlated. The flux variations, however, are anti-correlated with the X-ray spectral state while correlated with optical-UV (ultraviolet). Weekly binned {\it Fermi}-LAT data around the duration of the highest X-ray activity show a few detections with a log-parabola model but none with a power-law; yet the extracted LAT spectral energy distribution (SED) of the high activity duration for both the models is similar and show a hardening above 1 GeV. Further, near-infrared (NIR) data indicate strong spectral change, resembling a thermal component. Overall, the combined optical to gamma-ray broadband spectrum establishes the observed variations to a new high-energy-peaked (HBL) broadband emission component, similar to the one seen during the highest reported X-ray flux state of the source in 2017. The observed activities indicate some peculiar features that seem to be characteristic of this emission component while its appearance, a few years around the claimed \(\sim 12\)-year optical outbursts strongly indicate a connection between the two. △ Less

Submitted 29 July, 2021; v1 submitted 27 October, 2020; originally announced October 2020.

Comments: 14 pages, 6 figures, 1 table, ApJ (in press)

Journal ref: 2021 ApJ 921 18

IRX-1D: A Simple Deep Learning Architecture for Remote Sensing Classifications

Authors: Mahesh Pal, Akshay, B. Charan Teja

Abstract: We proposes a simple deep learning architecture combining elements of Inception, ResNet and Xception networks. Four new datasets were used for classification with both small and large training samples. Results in terms of classification accuracy suggests improved performance by proposed architecture in comparison to Bayesian optimised 2D-CNN with small training samples. Comparison of results using… ▽ More We proposes a simple deep learning architecture combining elements of Inception, ResNet and Xception networks. Four new datasets were used for classification with both small and large training samples. Results in terms of classification accuracy suggests improved performance by proposed architecture in comparison to Bayesian optimised 2D-CNN with small training samples. Comparison of results using small training sample with Indiana Pines hyperspectral dataset suggests comparable or better performance by proposed architecture than nine reported works using different deep learning architectures. In spite of achieving high classification accuracy with limited training samples, comparison of classified image suggests different land cover classes are assigned to same area when compared with the classified image provided by the model trained using large training samples with all datasets. △ Less

Submitted 19 June, 2023; v1 submitted 8 October, 2020; originally announced October 2020.

Comments: Want to improve this manuscript as it is not accepted by journal in present form

The binary Gold function and its c-boomerang connectivity table

Authors: Sartaj Ul Hasan, Mohit Pal, Pantelimon Stanica

Abstract: Here, we give a complete description of the c-Boomerang Connectivity Table for the Gold function over finite fields of even characteristic, by using double Weil sums. In the process we generalize a result of Boura and Canteaut (IACR Trans. Symmetric Cryptol. 2018(3) : 290-310, 2018) for the classical boomerang uniformity. Here, we give a complete description of the c-Boomerang Connectivity Table for the Gold function over finite fields of even characteristic, by using double Weil sums. In the process we generalize a result of Boura and Canteaut (IACR Trans. Symmetric Cryptol. 2018(3) : 290-310, 2018) for the classical boomerang uniformity. △ Less

Submitted 6 September, 2021; v1 submitted 19 September, 2020; originally announced September 2020.

Comments: 20 pages

MSC Class: 12E20; 11T24; 11T06; 94A60

Journal ref: Cryptography and communications, 2022

Short-term X-ray Variability during Different Activity Phases of Blazars S5 0716+714 and PKS 2155-304

Authors: Pankaj Kushwaha, Main Pal

Abstract: We explored the statistical properties of short-term X-ray variability using long-exposure {\it XMM-Newton} data during high X-ray variability phases of blazars S5 0716+714 and PKS 2155-304. In general, hardness ratio shows correlated variations with the source flux state (count rate), but in a few cases, mainly the bright phases, the trend is complex with correlation and anti-correlation both, in… ▽ More We explored the statistical properties of short-term X-ray variability using long-exposure {\it XMM-Newton} data during high X-ray variability phases of blazars S5 0716+714 and PKS 2155-304. In general, hardness ratio shows correlated variations with the source flux state (count rate), but in a few cases, mainly the bright phases, the trend is complex with correlation and anti-correlation both, indicating spectral evolution. Stationarity tests suggest the time series as non-stationarity or have trend stationarity. Except for one, none of the histograms fit resulted in a reduced-\(χ^2 \sim 1\) for a normal and log-normal profile but a normal profile is favored in general. On the contrary, the Anderson-Darling test favors lognormal with a test-statistic value lower for log-normal over normal for all the observations, even if out of significance limits. None of the IDs show linear RMS-flux relation. The contrary inferences from widely used different statistical methods indicate that a careful analysis is needed while the complex behavior of count rate with hardness ratio suggests spectral evolution over a few 10s of kilo-seconds during bright phases of the sources. In these cases, the spectrum extracted from whole observation may not be meaningful for spectral studies and certainly not a true representation of the spectral state of the source. △ Less

Submitted 14 September, 2020; originally announced September 2020.

Comments: 14 Pages, 4 figures, 2 Tables

Journal ref: Galaxies 2020, 8(3), 66

Adversarial Attack and Defense Strategies for Deep Speaker Recognition Systems

Authors: Arindam Jati, Chin-Cheng Hsu, Monisankha Pal, Raghuveer Peri, Wael AbdAlmageed, Shrikanth Narayanan

Abstract: Robust speaker recognition, including in the presence of malicious attacks, is becoming increasingly important and essential, especially due to the proliferation of several smart speakers and personal agents that interact with an individual's voice commands to perform diverse, and even sensitive tasks. Adversarial attack is a recently revived domain which is shown to be effective in breaking deep… ▽ More Robust speaker recognition, including in the presence of malicious attacks, is becoming increasingly important and essential, especially due to the proliferation of several smart speakers and personal agents that interact with an individual's voice commands to perform diverse, and even sensitive tasks. Adversarial attack is a recently revived domain which is shown to be effective in breaking deep neural network-based classifiers, specifically, by forcing them to change their posterior distribution by only perturbing the input samples by a very small amount. Although, significant progress in this realm has been made in the computer vision domain, advances within speaker recognition is still limited. The present expository paper considers several state-of-the-art adversarial attacks to a deep speaker recognition system, employing strong defense methods as countermeasures, and reporting on several ablation studies to obtain a comprehensive understanding of the problem. The experiments show that the speaker recognition systems are vulnerable to adversarial attacks, and the strongest attacks can reduce the accuracy of the system from 94% to even 0%. The study also compares the performances of the employed defense methods in detail, and finds adversarial training based on Projected Gradient Descent (PGD) to be the best defense method in our setting. We hope that the experiments presented in this paper provide baselines that can be useful for the research community interested in further studying adversarial robustness of speaker recognition systems. △ Less

Submitted 17 August, 2020; originally announced August 2020.

Meta-learning with Latent Space Clustering in Generative Adversarial Network for Speaker Diarization

Authors: Monisankha Pal, Manoj Kumar, Raghuveer Peri, Tae ** Park, So Hyun Kim, Catherine Lord, Somer Bishop, Shrikanth Narayanan

Abstract: The performance of most speaker diarization systems with x-vector embeddings is both vulnerable to noisy environments and lacks domain robustness. Earlier work on speaker diarization using generative adversarial network (GAN) with an encoder network (ClusterGAN) to project input x-vectors into a latent space has shown promising performance on meeting data. In this paper, we extend the ClusterGAN n… ▽ More The performance of most speaker diarization systems with x-vector embeddings is both vulnerable to noisy environments and lacks domain robustness. Earlier work on speaker diarization using generative adversarial network (GAN) with an encoder network (ClusterGAN) to project input x-vectors into a latent space has shown promising performance on meeting data. In this paper, we extend the ClusterGAN network to improve diarization robustness and enable rapid generalization across various challenging domains. To this end, we fetch the pre-trained encoder from the ClusterGAN and fine-tune it by using prototypical loss (meta-ClusterGAN or MCGAN) under the meta-learning paradigm. Experiments are conducted on CALLHOME telephonic conversations, AMI meeting data, DIHARD II (dev set) which includes challenging multi-domain corpus, and two child-clinician interaction corpora (ADOS, BOSCC) related to the autism spectrum disorder domain. Extensive analyses of the experimental data are done to investigate the effectiveness of the proposed ClusterGAN and MCGAN embeddings over x-vectors. The results show that the proposed embeddings with normalized maximum eigengap spectral clustering (NME-SC) back-end consistently outperform Kaldi state-of-the-art z-vector diarization system. Finally, we employ embedding fusion with x-vectors to provide further improvement in diarization performance. We achieve a relative diarization error rate (DER) improvement of 6.67% to 53.93% on the aforementioned datasets using the proposed fused embeddings over x-vectors. Besides, the MCGAN embeddings provide better performance in the number of speakers estimation and short speech segment diarization as compared to x-vectors and ClusterGAN in telephonic data. △ Less

Submitted 19 July, 2020; originally announced July 2020.

Comments: Submitted to IEEE/ACM TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSING

Patch Based Classification of Remote Sensing Data: A Comparison of 2D-CNN, SVM and NN Classifiers

Authors: Mahesh Pal, Akshay, Himanshu Rohilla, B. Charan Teja

Abstract: Pixel based algorithms including back propagation neural networks (NN) and support vector machines (SVM) have been widely used for remotely sensed image classifications. Within last few years, deep learning based image classifier like convolution neural networks (2D-CNN) are becoming popular alternatives to these classifiers. In this paper, we compare performance of patch based SVM and NN with tha… ▽ More Pixel based algorithms including back propagation neural networks (NN) and support vector machines (SVM) have been widely used for remotely sensed image classifications. Within last few years, deep learning based image classifier like convolution neural networks (2D-CNN) are becoming popular alternatives to these classifiers. In this paper, we compare performance of patch based SVM and NN with that of a deep learning algorithms comprising of 2D-CNN and fully connected layers. Similar to CNN which utilise image patches to derive features for further classification, we propose to use patches as an input in place of individual pixel with both SVM and NN classifiers. Two datasets, one multispectral and other hyperspectral data was used to compare the performance of different classifiers. Results with both datasets suggest the effectiveness of patch based SVM and NN classifiers in comparison to state of art 2D-CNN classifier. △ Less

Submitted 21 June, 2020; originally announced June 2020.

Comments: 8 pages, 2 Figures

Decomposition in Decision and Objective Space for Multi-Modal Multi-Objective Optimization

Authors: Monalisa Pal, Sanghamitra Bandyopadhyay

Abstract: Multi-modal multi-objective optimization problems (MMMOPs) have multiple subsets within the Pareto-optimal Set, each independently map** to the same Pareto-Front. Prevalent multi-objective evolutionary algorithms are not purely designed to search for multiple solution subsets, whereas, algorithms designed for MMMOPs demonstrate degraded performance in the objective space. This motivates the desi… ▽ More Multi-modal multi-objective optimization problems (MMMOPs) have multiple subsets within the Pareto-optimal Set, each independently map** to the same Pareto-Front. Prevalent multi-objective evolutionary algorithms are not purely designed to search for multiple solution subsets, whereas, algorithms designed for MMMOPs demonstrate degraded performance in the objective space. This motivates the design of better algorithms for addressing MMMOPs. The present work identifies the crowding illusion problem originating from using crowding distance globally over the entire decision space. Subsequently, an evolutionary framework, called graph Laplacian based Optimization using Reference vector assisted Decomposition (LORD), is proposed, which uses decomposition in both objective and decision space for dealing with MMMOPs. Its filtering step is further extended to present LORD-II algorithm, which demonstrates its dynamics on multi-modal many-objective problems. The efficacies of the frameworks are established by comparing their performance on test instances from the CEC 2019 multi-modal multi-objective test suite and polygon problems with the state-of-the-art algorithms for MMMOPs and other multi- and many-objective evolutionary algorithms. The manuscript is concluded by mentioning the limitations of the proposed frameworks and future directions to design still better algorithms for MMMOPs. The source code is available at https://worksupplements.droppages.com/lord. △ Less

Submitted 3 February, 2021; v1 submitted 3 June, 2020; originally announced June 2020.

Comments: Please visit https://www.sciencedirect.com/science/article/abs/pii/S2210650221000031 or https://doi.org/10.1016/j.swevo.2021.100842

ACM Class: G.1.6; I.2.8; I.5.m; H.3.3

Journal ref: Swarm and Evolutionary Computation, 100842 (2021)

Generative Model-driven Structure Aligning Discriminative Embeddings for Transductive Zero-shot Learning

Authors: Omkar Gune, Mainak Pal, Preeti Mukherjee, Biplab Banerjee, Subhasis Chaudhuri

Abstract: Zero-shot Learning (ZSL) is a transfer learning technique which aims at transferring knowledge from seen classes to unseen classes. This knowledge transfer is possible because of underlying semantic space which is common to seen and unseen classes. Most existing approaches learn a projection function using labelled seen class data which maps visual data to semantic data. In this work, we propose a… ▽ More Zero-shot Learning (ZSL) is a transfer learning technique which aims at transferring knowledge from seen classes to unseen classes. This knowledge transfer is possible because of underlying semantic space which is common to seen and unseen classes. Most existing approaches learn a projection function using labelled seen class data which maps visual data to semantic data. In this work, we propose a shallow but effective neural network-based model for learning such a projection function which aligns the visual and semantic data in the latent space while simultaneously making the latent space embeddings discriminative. As the above projection function is learned using the seen class data, the so-called projection domain shift exists. We propose a transductive approach to reduce the effect of domain shift, where we utilize unlabeled visual data from unseen classes to generate corresponding semantic features for unseen class visual samples. While these semantic features are initially generated using a conditional variational auto-encoder, they are used along with the seen class data to improve the projection function. We experiment on both inductive and transductive setting of ZSL and generalized ZSL and show superior performance on standard benchmark datasets AWA1, AWA2, CUB, SUN, FLO, and APY. We also show the efficacy of our model in the case of extremely less labelled data regime on different datasets in the context of ZSL. △ Less

Submitted 9 May, 2020; originally announced May 2020.