arXiv subjects

Atul Kumar

Publications and source records attributed to Atul Kumar.

At least 55 records · Page 3Linked to original sources

arXiv2019

A new mechanism of direct coupling of laser energy to ions

The well-known schemes (e.g. Brunel, resonance absorption, JxB heating etc.) couple laser energy to the lighter electron species of the plasma. In this work, a fundamentally new mechanism of laser energy absorption directly to the heavier ion species has been proposed. The mechanism relies on the difference between the ExB drifts of electron and ions in the oscillating electric field of the laser and an external magnetic field to create charge density perturbations. The proposed mechanism is verified with the help of Particle - In - Cell (PIC) simulations using OSIRIS4.0.

physics.plasm-ph↗

arXiv2019

Drift waves with dust acoustic wave coupling

Drift wave is a prominent mode of a magnetized plasma of inhomogeneous density. It plays an important role in the transport of particles, energy and momentum perpendicular to the ambient magnetic field. The frequency of this mode is governed by the inhomogeneity scale length and is much lower than the typical homogeneous plasma modes involving ions and electrons. In this work the possibility of coupling of this particular mode with the low frequency modes of a dusty plasma medium is considered.

physics.plasm-ph↗

arXiv2018

Analysing the efficiency of partially entangled states in Vaidman's-type games and its applications in Quantum Secret Sharing

We analyse the role of degree of entanglement for Vaidman's game in a setting where the players share a set of partially entangled three-qubit states. Our results show that the entangled states combined with quantum strategies may not be always helpful in winning a game as opposed to the classical strategies. We further find the conditions under which quantum strategies are always helpful in achieving higher winning probability in the game in comparison to classical strategies. Moreover, we show that a special class of W states can always be used to win the game using quantum strategies irrespective of the degree of entanglement between the three qubits. Our analysis also helps us in comparing the Vaidman's game with the secret sharing protocol. Furthermore, we propose a new Vaidman-type game where the rule maker itself is entangled with the other two players and acts as a facilitator to share a secret key with the two players. For practical purposes, the analysis is extended to study the proposed game under noisy conditions. In addition, the results obtained here are also generalized for multi-qubit games.

quant-ph↗

arXiv2018

Analysing the role of entanglement in the three-qubit Vaidman's game

quant-ph↗

arXiv2018

Game-theoretic perspective of Ping-Pong Protocol

We analyse Ping-Pong protocol from the point of view of a game. The analysis helps us in understanding the different strategies of a sender and an eavesdropper to gain the maximum payoff in the game. The study presented here characterizes strategies that lead to different Nash equilibriums. We further demonstrate the condition for Pareto optimality depending on the parameters used in the game. Moreover, we also analysed LM05 protocol and compared it with PP protocol from the point of view of a generic two-way QKD game with or without entanglement. Our results provide a deeper understanding of general two-way QKD protocols in terms of the security and payoffs of different stakeholders in the protocol.

quant-ph↗

arXiv2018

Role of contact work function, back surface field and conduction band offset in CZTS solar cell

We employ simulation based approach for enhancing the efficiency of Cu2ZnSnS4 (CZTS) based solar cells. Initial benchmarking of simulation with the experimentally reported solar cell in literature is performed by incorporating a suitable defect model. We then explore the effects of: (a) conduction band offset (CBO) at CZTS/CdS junction, (b) back surface field (BSF) due to an additional layer with higher carrier density, and (c) high work function back contact. Efficiency is observed to improve by about 70% upon optimizing the above three parameters. We also observe that utilizing BSF in the configuration can reduce the high work function requirement of the back contact. A work function of 5.2 eV (e.g., using Ni), a BSF layer (e.g., using SnS), and a CBO of 0.1 eV (e.g., using ZnS) constitute an optimal configuration.

physics.app-ph↗

arXiv2018

Excitation of KdV Alfven solitons by a pulsed CO2 laser in plasma in the presence of an external magnetic field

The interaction of laser with plasmas leads to many interesting phenomena which includes laser energy absorption, mode conversion, energetic particle generation etc. In this work, the excitation of Korteweg - de Vries (KdV) Alfven solitons in plasma is demonstrated with the help of 2-D Particle - In - Cell (PIC) simulations. For this purpose, the propagation of a pulsed $CO_2$ (with a wavelength of $10 μm$) laser normally incident on an overdense plasma target is considered in the presence of an external magnetic field. The magnitude of the external magnetic field is chosen so as to have the electrons magnetized and ions unmagnetized at the laser frequency. These solitons propagate stably and are observed to be responsible for energetic electron generation as the background undisturbed electrons of the medium get reflected from its front. It is also shown that subsequently, transverse modulations appear in the structure which grow with time. With recent technological advancements in the development of short pulse $CO_2$ lasers and also on the generation of magnetic fields of the order of tens of kilo - Tesla in the laboratory, these studies have practical relevance and have a possibility of getting replicated in laboratory in near future.

physics.plasm-ph↗

arXiv2018

Fast Ignition Laser Fusion Using In-Situ Ion Acceleration With Pulsed CO2 Lasers

Fast ignition is an alternative concept of laser fusion in which the task of compressing the fusion pellet to supersolid densities is accomplished by the conventional high energy nanosecond glass lasers and the task of igniting the compressed pellet is given to a high intensity, moderate energy pico second source which can set the pellet ablaze (with DT fusion reactions) by creating an adequate size hot spot in it. In this letter, we present a conceptual method by which energy from carbon dioxide ($ CO_2 $) laser could be coupled to heat up ions produced in situ in the plasma, which in turn produces the required hot spot. An efficient conversion of $ CO_2 $ energy into ion beam energy can thus give us the required source of energy for the fast ignition laser fusion. We demonstrate by using PIC (Particle - In - Cell) simulations that the use of several Kilo Tesla of an external magnetic field in the transverse direction in an inhomogeneous plasma where a cutoff is followed by a resonance can lead to coupling of laser energy to a lower hybrid ion plasma resonance. Then ions are accelerated efficiently ($\sim 1 MeV$) by the breaking of lower hybrid waves and thereby enabling the possibility of fast ignition by ions in a convenient fashion.

physics.plasm-ph↗

arXiv2017

Evidence of new finite beam plasma instability for magnetic field generation

We demonstrate by computer simulations, laser plasma experiments, and analytic theory that a hitherto unknown instability is excited in the beam plasma system with finite transverse size. This instability is responsible for the generation of magnetic fields at scales comparable to the transverse beam dimension which can be much longer than the electron skin depth scale. This counterintuitive result arises due to radiative leakage associated with finite beam boundaries which are absent in conventional infinite periodic systems considered in earlier simulations as well as theoretical analyses and may trigger a reexamination of a hitherto prevalent idea.

physics.plasm-ph↗

arXiv2017

Merger and reconnection of Weibel separated relativistic electron beam

The relativistic electron beam (REB) propagation in a plasma is fraught with beam plasma instabilities. The prominent amongst them being the collisionless Weibel destabilization which spatially separates the forward propagating REB and the return shielding currents. This results in the formation of REB current filaments which are typically of the size of electron skin depth during the linear stage of the instability. It has been observed that in the nonlinear stage the filaments size increases as they merge with each other. With the help of 2D PIC simulations in the plane perpendicular to the REB propagation, it is shown that these mergers occur in two distinct nonlinear phases. In the first phase, the total magnetic energy increases. Subsequently, however, during the second phase, one observes a reduction in magnetic energy. It is shown that the transition from one nonlinear regime to another occurs when the typical current associated with individual filaments hits the Alfven threshold. In the second nonlinear regime, therefore, the filaments can no longer permit any increase in current. Magnetic reconnection events then dissipate the excess current (and its associated magnetic energy) that would result from a merger process leading to the generation of energetic electrons jets in the perpendicular plane. At later times when there are only few filaments left the individual reconnection events can be clearly identified. It is observed that in between such events the magnetic energy remains constant and shows a sudden drop as and when two filaments merge. The electron jets released in these recon- nection events are thus responsible for the transverse heating which has been mentioned in some previous studies [Honda et al. Phys. Plasmas 7, 1302 (2000)].

physics.plasm-ph↗

arXiv2017

Observation of 1-D time dependent non-propagating laser plasma structures using Fluid and PIC codes

The manuscript reports the observation of time dependent localized and non-propagating structures in the coupled laser plasma system through 1-D fluid and PIC simulations. It is reported that such structures form spontaneously as a result of collision amongst certain exact solitonic solutions. They are seen to survive as coherent entities for a long time up to several hundreds of plasma periods. Furthermore, it is shown that such time dependence can also be artificially recreated by significantly disturbing the delicate balance between the radiation and the density fields required for the exact non-propagating solution obtained by Esirkepov et al. [1]. The ensuing time evolution is an interesting interplay between kinetic and field energies of the system. The electrostatic plasma oscillations are coupled with oscillations in the electromagnetic field. The inhomogeneity of the background and the relativistic nature, however, invariably produce large amplitude density perturbations leading to its wave breaking. In the fluid simulations, the signature of wave breaking can be discerned by a drop in the total energy which evidently gets lost to the grid. The PIC simulations are observed to closely follow the fluid simulations till the point of wave breaking. However, the total energy in the case of PIC simulations is seen to remain conserved all throughout the simulations. At the wave breaking the particles are observed to acquire thermal kinetic energy in the case of PIC. Interestingly, even after wave breaking, compact coherent structures with trapped radiation inside high-density peaks, continue to exist both in PIC and fluid simulations. Though the time evolution does not exactly match in the two simulations as it does prior to the process of wave breaking, the time-dependent features exhibited by the remnant structures are characteristically similar.

physics.plasm-ph↗

arXiv2017

A Survey on Resilient Machine Learning

Machine learning based system are increasingly being used for sensitive tasks such as security surveillance, guiding autonomous vehicle, taking investment decisions, detecting and blocking network intrusion and malware etc. However, recent research has shown that machine learning models are venerable to attacks by adversaries at all phases of machine learning (eg, training data collection, training, operation). All model classes of machine learning systems can be misled by providing carefully crafted inputs making them wrongly classify inputs. Maliciously created input samples can affect the learning process of a ML system by either slowing down the learning process, or affecting the performance of the learned mode, or causing the system make error(s) only in attacker's planned scenario. Because of these developments, understanding security of machine learning algorithms and systems is emerging as an important research area among computer security and machine learning researchers and practitioners. We present a survey of this emerging area in machine learning.

cs.AI↗

arXiv2017

Energy principle for 2D electromagnetic, relativistic, interpenetrating, counter streaming plasma flows

A relativistic electron beam propagating through plasma induces a return current in the system. Such a system of counterstreaming forward and return current is susceptible to host of instabilities out of which Weibel remains a dominant mode for destabilizing the system. Weibel instability has been widely investigated in simulations, experiments, as well as, analytically using fluid and kinetic treatments. A purely growing mode like Weibel instability can also be understood by using conservation theorems and energy principle analysis. An electrostatic analog to the Weibel instability, two stream instability in a beam plasma system has already been investigated using energy principle analysis [Lashmore-Davies 2007]. A detailed analytical description of conservation theorem for 2D Weibel instability in a beam plasma system has been carried out in this manuscript.

physics.plasm-ph↗

arXiv2017

Reducing Bias in Production Speech Models

Replacing hand-engineered pipelines with end-to-end deep learning systems has enabled strong results in applications like speech and object recognition. However, the causality and latency constraints of production systems put end-to-end speech models back into the underfitting regime and expose biases in the model that we show cannot be overcome by "scaling up", i.e., training bigger models on more data. In this work we systematically identify and address sources of bias, reducing error rates by up to 20% while remaining practical for deployment. We achieve this by utilizing improved neural architectures for streaming inference, solving optimization issues, and employing strategies that increase audio and label modelling versatility.

cs.CL↗

arXiv2017

Magnetic field generation in finite beam plasma system

For finite systems boundaries can introduce remarkable novel features. A well known example is the Casimir effect [1, 2] that is observed in quantum electrodynamic systems. In classical systems too novel effects associated with finite boundaries have been observed, for example the surface plasmon mode [3] that appears when the plasma has a finite extension. In this work a novel instability associated with the finite transverse size of a beam owing through a plasma system has been shown to exist. This instability leads to distinct characteristic features of the associated magnetic field that gets generated. For example, in contrast to the well known unstable Weibel mode of a beam plasma system which generates magnetic field at the skin depth scale, this instability generates magnetic field at the scales length of the transverse beam dimension [4]. The existence of this new instability is demonstrated by analytical arguments and by simulations conducted with the help of a variety of Particle - In - Cell (PIC) codes (e.g. OSIRIS, EPOCH, PICPSI). Two fluid simulations have also been conducted which confirm the observations. Furthermore, laboratory experiments on laser plasma system also provides evidence of such an instability mechanism at work.

physics.plasm-ph↗

arXiv2017

Laplacian matrices of weighted digraphs represented as quantum states

Representing graphs as quantum states is becoming an increasingly important approach to study entanglement of mixed states, alternate to the standard linear algebraic density matrix-based approach of study. In this paper, we propose a general weighted directed graph framework for investigating properties of a large class of quantum states which are defined by three types of Laplacian matrices associated with such graphs. We generalize the standard framework of defining density matrices from simple connected graphs to density matrices using both combinatorial and signless Laplacian matrices associated with weighted directed graphs with complex edge weights and with/without self-loops. We also introduce a new notion of Laplacian matrix, which we call signed Laplacian matrix associated with such graphs. We produce necessary and/or sufficient conditions for such graphs to correspond to pure and mixed quantum states. Using these criteria, we finally determine the graphs whose corresponding density matrices represent entangled pure states which are well known and important for quantum computation applications. We observe that all these entangled pure states share a common combinatorial structure.

quant-ph↗

arXiv2016

Usefulness of Multiqubit W-type States in Quantum Information Processing Task

We analyze the efficacy of multiqubit W-type states as resources for quantum information. For this, we identify and generalize four-qubit W-type states. Our results show that the states can be used as resources for deterministic quantum information processing. The utility of results, however, is limited by the availability of experimental setups to perform and distinguish multiqubit measurements. We, therefore, emphasize on another protocol where two users want to establish an optimal bipartite entanglement using the partially entangled W-type states. We found that for such practical purposes, four-qubit W states can be a better resource in comparison to three-qubit W-type states. For dense coding protocol, our states can be used deterministically to send two bits of classical message by locally manipulating a single qubit.

quant-ph↗

arXiv2016

A cloned qutrit and its utility in information processing tasks

We, in this paper, analyze the efficacy of an output as a resource from a universal quantum cloning machine in information processing tasks such as teleportation and dense coding. For this, we have considered the $3\otimes 3$ dimensional system (or qutrit system). The output states are found to be NPT states for certain ranges of machine parameters. Using the output state as an entangled resource, we have also studied the optimal fidelities of teleportation and capacities of dense coding protocols with respect to the machine parameters and have made a few interesting observations. Our work is mainly motivated from the fact that the cloning output can be used as a resource in quantum information processing and adds a valuable dimension to the applications of cloning machines.

quant-ph↗