-
Certified Adversarial Robustness of Machine Learning-based Malware Detectors via (De)Randomized Smoothing
Authors:
Daniel Gibert,
Luca Demetrio,
Giulio Zizzo,
Quan Le,
Jordi Planes,
Battista Biggio
Abstract:
Deep learning-based malware detection systems are vulnerable to adversarial EXEmples - carefully-crafted malicious programs that evade detection with minimal perturbation. As such, the community is dedicating effort to develop mechanisms to defend against adversarial EXEmples. However, current randomized smoothing-based defenses are still vulnerable to attacks that inject blocks of adversarial con…
▽ More
Deep learning-based malware detection systems are vulnerable to adversarial EXEmples - carefully-crafted malicious programs that evade detection with minimal perturbation. As such, the community is dedicating effort to develop mechanisms to defend against adversarial EXEmples. However, current randomized smoothing-based defenses are still vulnerable to attacks that inject blocks of adversarial content. In this paper, we introduce a certifiable defense against patch attacks that guarantees, for a given executable and an adversarial patch size, no adversarial EXEmple exist. Our method is inspired by (de)randomized smoothing which provides deterministic robustness certificates. During training, a base classifier is trained using subsets of continguous bytes. At inference time, our defense splits the executable into non-overlap** chunks, classifies each chunk independently, and computes the final prediction through majority voting to minimize the influence of injected content. Furthermore, we introduce a preprocessing step that fixes the size of the sections and headers to a multiple of the chunk size. As a consequence, the injected content is confined to an integer number of chunks without tampering the other chunks containing the real bytes of the input examples, allowing us to extend our certified robustness guarantees to content insertion attacks. We perform an extensive ablation study, by comparing our defense with randomized smoothing-based defenses against a plethora of content manipulation attacks and neural network architectures. Results show that our method exhibits unmatched robustness against strong content-insertion attacks, outperforming randomized smoothing-based defenses in the literature.
△ Less
Submitted 1 May, 2024;
originally announced May 2024.
-
Machine Learning for Windows Malware Detection and Classification: Methods, Challenges and Ongoing Research
Authors:
Daniel Gibert
Abstract:
In this chapter, readers will explore how machine learning has been applied to build malware detection systems designed for the Windows operating system. This chapter starts by introducing the main components of a Machine Learning pipeline, highlighting the challenges of collecting and maintaining up-to-date datasets. Following this introduction, various state-of-the-art malware detectors are pres…
▽ More
In this chapter, readers will explore how machine learning has been applied to build malware detection systems designed for the Windows operating system. This chapter starts by introducing the main components of a Machine Learning pipeline, highlighting the challenges of collecting and maintaining up-to-date datasets. Following this introduction, various state-of-the-art malware detectors are presented, encompassing both feature-based and deep learning-based detectors. Subsequent sections introduce the primary challenges encountered by machine learning-based malware detectors, including concept drift and adversarial attacks. Lastly, this chapter concludes by providing a brief overview of the ongoing research on adversarial defenses.
△ Less
Submitted 29 April, 2024;
originally announced April 2024.
-
A Robust Defense against Adversarial Attacks on Deep Learning-based Malware Detectors via (De)Randomized Smoothing
Authors:
Daniel Gibert,
Giulio Zizzo,
Quan Le,
Jordi Planes
Abstract:
Deep learning-based malware detectors have been shown to be susceptible to adversarial malware examples, i.e. malware examples that have been deliberately manipulated in order to avoid detection. In light of the vulnerability of deep learning detectors to subtle input file modifications, we propose a practical defense against adversarial malware examples inspired by (de)randomized smoothing. In th…
▽ More
Deep learning-based malware detectors have been shown to be susceptible to adversarial malware examples, i.e. malware examples that have been deliberately manipulated in order to avoid detection. In light of the vulnerability of deep learning detectors to subtle input file modifications, we propose a practical defense against adversarial malware examples inspired by (de)randomized smoothing. In this work, we reduce the chances of sampling adversarial content injected by malware authors by selecting correlated subsets of bytes, rather than using Gaussian noise to randomize inputs like in the Computer Vision (CV) domain. During training, our ablation-based smoothing scheme trains a base classifier to make classifications on a subset of contiguous bytes or chunk of bytes. At test time, a large number of chunks are then classified by a base classifier and the consensus among these classifications is then reported as the final prediction. We propose two strategies to determine the location of the chunks used for classification: (1) randomly selecting the locations of the chunks and (2) selecting contiguous adjacent chunks. To showcase the effectiveness of our approach, we have trained two classifiers with our chunk-based ablation schemes on the BODMAS dataset. Our findings reveal that the chunk-based smoothing classifiers exhibit greater resilience against adversarial malware examples generated with state-of-the-are evasion attacks, outperforming a non-smoothed classifier and a randomized smoothing-based classifier by a great margin.
△ Less
Submitted 26 February, 2024; v1 submitted 23 February, 2024;
originally announced February 2024.
-
On the effect of the luni-solar gravitational attraction on trees
Authors:
Jean-Louis Le Mouël,
Dominique Gibert,
Jean-Baptiste Boulé,
Pierpaolo Zuddas,
Vincent Courtillot,
Fernando Lopes,
Marc Gèze,
Alexis Maineult
Abstract:
In this paper, we revisit old electrical measurements conducted on a poplar tree in 2003 (Gibert et al. 2006), which showed the presence of a diurnal electrical signal, attributed to an electrokinetic phenomenon correlated with sap flow, even in winter. We reanalyze these data using the singular spectrum analysis method and demonstrate that the electrical signal measured at various locations (root…
▽ More
In this paper, we revisit old electrical measurements conducted on a poplar tree in 2003 (Gibert et al. 2006), which showed the presence of a diurnal electrical signal, attributed to an electrokinetic phenomenon correlated with sap flow, even in winter. We reanalyze these data using the singular spectrum analysis method and demonstrate that the electrical signal measured at various locations (roots, trunk, and branches) on the poplar tree can be decomposed to over 80\% into a sum of 7 pseudo-periods, all linked to luni-solar tides. The precision of the extracted periods exceeds that of models. To validate these old measurements, we have reproduced the 2003 protocol since 2018 at the Jardin des Plantes in Paris, on 3 oak trees and 3 hornbeams. The electrical signals today exhibit the same characteristics as those of 2003. Tidal forces could be a driving force of sap flow in trees.
△ Less
Submitted 12 February, 2024;
originally announced February 2024.
-
On the tilt of the Earth's polar axis (climat): Some 'impressionist' remarks
Authors:
V. Courtillot,
F. Lopes,
V. Kossobokov,
P. Zuddas,
D. Gibert,
J. -B. Boulé,
J. -L. Le Mouël
Abstract:
In this lengthy letter, we wanted to discuss the concept of climate based on definitions established for over a century and direct observations that we have been collecting for more than a century as well. To do this, we present and discuss the remarkably stable maps over time of the various physical parameters that make up the climate corpus: solar temperature, atmospheric pressure, winds, precip…
▽ More
In this lengthy letter, we wanted to discuss the concept of climate based on definitions established for over a century and direct observations that we have been collecting for more than a century as well. To do this, we present and discuss the remarkably stable maps over time of the various physical parameters that make up the climate corpus: solar temperature, atmospheric pressure, winds, precipitation, temperature anomalies. This impressionistic tableau that we are gradually sketching as our reflection unfolds leads us to the following proposition: What if, as Laplace first proposed in 1799 and later Milankovi\{'}c in 1920, ground temperature were merely a consequence of climate and not a separate parameter of climate in its own right?
△ Less
Submitted 4 December, 2023; v1 submitted 4 October, 2023;
originally announced October 2023.
-
Towards a Practical Defense against Adversarial Attacks on Deep Learning-based Malware Detectors via Randomized Smoothing
Authors:
Daniel Gibert,
Giulio Zizzo,
Quan Le
Abstract:
Malware detectors based on deep learning (DL) have been shown to be susceptible to malware examples that have been deliberately manipulated in order to evade detection, a.k.a. adversarial malware examples. More specifically, it has been show that deep learning detectors are vulnerable to small changes on the input file. Given this vulnerability of deep learning detectors, we propose a practical de…
▽ More
Malware detectors based on deep learning (DL) have been shown to be susceptible to malware examples that have been deliberately manipulated in order to evade detection, a.k.a. adversarial malware examples. More specifically, it has been show that deep learning detectors are vulnerable to small changes on the input file. Given this vulnerability of deep learning detectors, we propose a practical defense against adversarial malware examples inspired by randomized smoothing. In our work, instead of employing Gaussian or Laplace noise when randomizing inputs, we propose a randomized ablation-based smoothing scheme that ablates a percentage of the bytes within an executable. During training, our randomized ablation-based smoothing scheme trains a base classifier based on ablated versions of the executable files. At test time, the final classification for a given input executable is taken as the class most commonly predicted by the classifier on a set of ablated versions of the original executable. To demonstrate the suitability of our approach we have empirically evaluated the proposed ablation-based model against various state-of-the-art evasion attacks on the BODMAS dataset. Results show greater robustness and generalization capabilities to adversarial malware examples in comparison to a non-smoothed classifier.
△ Less
Submitted 17 August, 2023;
originally announced August 2023.
-
A living forest of Tibetan Juniper trees as a new kind of astronomical and geophysical observatory
Authors:
Vincent Courtillot,
Jean-Baptiste Boulé,
Jean-Louis Le Mouël,
Dominique Gibert,
Pierpaolo Zuddas,
Alexis Maineult,
Marc Gèze,
Fernando Lopes
Abstract:
The trajectory of Earth about the Sun is perturbed by torques exerted by the Moon and Sun, and also the four giant planets. These provoke variations of insolation at Earth surface, known as kyr-long Milanković cycles. The concept has been extended to the shorter time scales of years to centuries, that are relevant to tree growth. This paper focuses on iSSA of results of the dendro-chronological st…
▽ More
The trajectory of Earth about the Sun is perturbed by torques exerted by the Moon and Sun, and also the four giant planets. These provoke variations of insolation at Earth surface, known as kyr-long Milanković cycles. The concept has been extended to the shorter time scales of years to centuries, that are relevant to tree growth. This paper focuses on iSSA of results of the dendro-chronological study of a forest of long-lived Tibetan junipers. From this, we determine a median curve of tree growth rates, that is analyzed by iSSA. We obtain a rich set of (pseudo-) periods, from 3.3 yr up to more than 1000 years, that compare with the specific spectral signature found in the sunspot and length-of-day time series. We discuss in detail the record from a single tree that spans almost completely the 357-2000 AD interval. The 90 yr Gleissberg, 22 yr and 30 yr components are quite prominent. The Oort, Wolf, Spörer, Maunder and Dalton climate extrema all correspond quite precisely to extrema of the Gleissberg cycle. The well-known Medieval Climate Optimum, Little Ice Age and Modern Climate Optimum all seem to be mainly forced by variations in the envelope of the Gleissberg cycle. The Gleissberg cycle is strongly modulated with a period of 500-600 years. The node near a small gap in the data is very close to the Medieval Climate Optimum. Observations in different parts of Earth are in favor of a global extension of the MCO. In the same way that the Milankovic mathematical theory of climate allows one to relate climate change and length of day, through changes in inclination of Earth's rotation axis and solar insolation, it is reasonable to propose that the set of pseudo-periods that are evidenced in the Tibetan tree ring growth rates simply corresponds to short period Milanković cycles. The Dulan forest could be considered as a good candidate for a continuous, global geophysical observatory.
△ Less
Submitted 7 July, 2023; v1 submitted 20 June, 2023;
originally announced June 2023.
-
Query-Free Evasion Attacks Against Machine Learning-Based Malware Detectors with Generative Adversarial Networks
Authors:
Daniel Gibert,
Jordi Planes,
Quan Le,
Giulio Zizzo
Abstract:
Malware detectors based on machine learning (ML) have been shown to be susceptible to adversarial malware examples. However, current methods to generate adversarial malware examples still have their limits. They either rely on detailed model information (gradient-based attacks), or on detailed outputs of the model - such as class probabilities (score-based attacks), neither of which are available…
▽ More
Malware detectors based on machine learning (ML) have been shown to be susceptible to adversarial malware examples. However, current methods to generate adversarial malware examples still have their limits. They either rely on detailed model information (gradient-based attacks), or on detailed outputs of the model - such as class probabilities (score-based attacks), neither of which are available in real-world scenarios. Alternatively, adversarial examples might be crafted using only the label assigned by the detector (label-based attack) to train a substitute network or an agent using reinforcement learning. Nonetheless, label-based attacks might require querying a black-box system from a small number to thousands of times, depending on the approach, which might not be feasible against malware detectors. This work presents a novel query-free approach to craft adversarial malware examples to evade ML-based malware detectors. To this end, we have devised a GAN-based framework to generate adversarial malware examples that look similar to benign executables in the feature space. To demonstrate the suitability of our approach we have applied the GAN-based attack to three common types of features usually employed by static ML-based malware detectors: (1) Byte histogram features, (2) API-based features, and (3) String-based features. Results show that our model-agnostic approach performs on par with MalGAN, while generating more realistic adversarial malware examples without requiring any query to the malware detectors. Furthermore, we have tested the generated adversarial examples against state-of-the-art multimodal and deep learning malware detectors, showing a decrease in detection performance, as well as a decrease in the average number of detections by the anti-malware engines in VirusTotal.
△ Less
Submitted 16 June, 2023;
originally announced June 2023.
-
On variations of global mean surface temperature: When Laplace meets Milankovic
Authors:
V. Courtillot,
F. Lopes,
D. Gibert,
J. B Boulé,
J. L Le Mouël
Abstract:
In his mathematical theory, Milankovic finds a link between the heat received by the Earth surface per unit time as a function of the solar ephemerids and derives a model of climate changes at periods longer than a few thousand years and more. In this paper, we investigate the potential connections of global temperature and Earth rotation at much shorter periods, in the complementary range of one…
▽ More
In his mathematical theory, Milankovic finds a link between the heat received by the Earth surface per unit time as a function of the solar ephemerids and derives a model of climate changes at periods longer than a few thousand years and more. In this paper, we investigate the potential connections of global temperature and Earth rotation at much shorter periods, in the complementary range of one to a few hundred years. For temperature, we select the HadCrut05. For Earth rotation, defined by pole coordinates and length of day, we use the IERS data sets. Using iterative Singular Spectrum Analysis (iSSA), we extract the trend and quasi-periodic components of these time series. The first quasi-periodic components (period ~80-90 years) are expressions of the Gleissberg cycle and are identical (at the level of uncertainty of the data). Taken together, the trend and Gleissberg components allow one to reconstruct 87% of the variance of the data for lod and 48% for temperature. The next four iSSA components, with periods ~40, 22, 15 and 9 years. The Lagrange and Laplace theories imply that the derivative of pole motion should be identical to lod variations: this strong check is passed by the trend + Gleissberg reconstructions. The annual oscillations of pole motion and lod are linked to annual variations in Sun-Earth distance, in agreement with an astronomical, but not a climatic origin. The results obtained in this paper for the observed temperature/rotation couple add to the growing list of evidence of solar and planetary forcings of gravitational nature on a number of geophysical processes (including sea-level, sea-level pressure, sea-ice extent, oceanic climate indices).
△ Less
Submitted 4 December, 2023; v1 submitted 6 June, 2023;
originally announced June 2023.
-
On the external forcing of global eruptive activity in the past 300 years
Authors:
J-L. Le Mouël,
D. Gibert,
V. Courtillot,
S. Dumont,
J. de Bremond Ars,
S. Petrosino,
P. Zuddas F. Lopes,
J-B. Boulé,
M. C. Neves,
S. Custódio,
G. Silveira,
V. Kossobokov,
L. Coen,
M. Geze
Abstract:
The decryption of the temporal sequence of volcanic eruptions is a key step in better anticipating future events. Volcanic activity is the result of a complex interaction between internal and external processes, with time scales spanning multiple orders of magnitude. We review periodicities that have been detected or correlated with volcanic eruptions/phenomena and interpreted as resulting from ex…
▽ More
The decryption of the temporal sequence of volcanic eruptions is a key step in better anticipating future events. Volcanic activity is the result of a complex interaction between internal and external processes, with time scales spanning multiple orders of magnitude. We review periodicities that have been detected or correlated with volcanic eruptions/phenomena and interpreted as resulting from external forces. Taking a global perspective and longer time scales than a few years, we approach this interaction by analyzing three time series using singular spectral analysis: the global number of volcanic eruptions (NVE) between 1700 and 2022, the number of sunspots (ISSN), a proxy for solar activity, the polar motion (PM) and length of day (lod), two proxies for gravitational force. Several pseudo-periodicities are common to NVE and ISSN, in addition to the 11-year Schwabe cycle that has been reported in previous work, but NVE shares even more periodicities with PM. These quasi-periodic components range from ~5 to ~130 years. We interpret our analytical results in light of the Laplace's paradigm and propose that, similarly to the movement of Earth's rotation axis, global eruptive activity is modulated by commensurable orbital moments of the Jovian planets, whose influence is also detected in solar activity.
△ Less
Submitted 19 April, 2023;
originally announced April 2023.
-
Is the Earth's magnetic field a constant ? a legacy of Poisson
Authors:
J-L. Le Mouël,
F. Lopes,
V. Courtillot,
D. Gibert,
J-B. Boulé
Abstract:
In the report he submitted to the Académie des Sciences, Poisson imagined a set of concentric spheres at the origin of the Earth's magnetic field. It may come as a surprise to many that Poisson as well as Gauss both considered the magnetic field to be constant. We propose in this study to test this surprising assertion for the first time evoked by Poisson (1826). First, we will present a developme…
▽ More
In the report he submitted to the Académie des Sciences, Poisson imagined a set of concentric spheres at the origin of the Earth's magnetic field. It may come as a surprise to many that Poisson as well as Gauss both considered the magnetic field to be constant. We propose in this study to test this surprising assertion for the first time evoked by Poisson (1826). First, we will present a development of Maxwell's equations in the framework of a static electric field and a static magnetic field in order to draw the necessary consequences for the Poisson hypothesis. In a second step, we will see if the observations can be in agreement with Poisson (1826). To do so, we have chosen to compare 1) the polar motion drift and the secular variation of the Earth's magnetic field, 2) the seasonal pseudo-cycles of day length together with those of the sea level recorded by different tide gauges around the globe and those of the Earth's magnetic field recorded in different magnetic observatories. We then propose a mechanism, in the spirit of Poisson, to explain the presence of the 11-year in the magnetic field. We test this mechanism with observations and finally we study closely the evolution of the g10 coefficient of the IGFR over time.
△ Less
Submitted 23 March, 2023; v1 submitted 16 March, 2023;
originally announced March 2023.
-
On distributions law of planetary rotations and revolutions as a function of aphelia, following Lagrange's formulation
Authors:
F. Lopes,
V. Courtillot,
D. Gibert,
J. -L. Le Mouël,
J. -B. Boulé
Abstract:
Earth rotation is determined by polar motion (PM) and length of day (lod). The excitation sources of PM are torques linked to fluid circulations ("geophysical excitations"), and those of lod to luni-solar tides ("astronomical excitations"). We explore the links between the rotations and revolutions of planets, following Lagrange's (1853) presentation of mechanics. The energy of a planet in motion…
▽ More
Earth rotation is determined by polar motion (PM) and length of day (lod). The excitation sources of PM are torques linked to fluid circulations ("geophysical excitations"), and those of lod to luni-solar tides ("astronomical excitations"). We explore the links between the rotations and revolutions of planets, following Lagrange's (1853) presentation of mechanics. The energy of a planet in motion in a central field is the sum of kinetic, centrifugal (planet dependent) and centripetal (identical for all planets) energies. For each planet, one can calculate a "constant of gravitation" Gp . For the giant planets, Gp decreases as a function of aphelia. There is no such organized behavior for the terrestrial planets. The perturbing potential of other planets generates a small angular contribution to the displacement : this happens to be identical to Einstein's famous formula for precession. Delays in the planet's perihelia follow a (-5/2) power law of a. This is readily understood in the Lagrange formalism (the centrifugal term takes over for small distances). The telluric planets have lost energy, probably transferred to the planets rotations. The ratio of areal velocities to rotation obeys a -5/2 power law of a. The ratio of areal velocity to integrated period R also fits a -5/2 power dependence, implying linearity of the energy exchange between revolution and rotation. For Einstein deformation of space-time by the Sun is the origin of the field perturbation. For Lagrange the perturbation could only be due to the interactions of torques. The perihelion delays, the areal velocities and the planetary rotations display power laws of aphelia, whose behavior contrasts with that of the kinetic moment. The areal velocity being linearly linked to the kinetic moment of planets, this must be the level at which the transfer is achieved.
△ Less
Submitted 23 March, 2023; v1 submitted 15 September, 2022;
originally announced September 2022.
-
On the Slow Drift of Solstices: Milankovic Cycles and Mean Global Temperature
Authors:
F. Lopes,
V. Courtillot,
D. Gibert,
J-L. Le Mouël
Abstract:
The Earth's revolution is modified by changes in inclination of its rotation axis. Despite the fact that the gravity field is central, the Earth's trajectory is not closed and the equinoxes drift. Milankovic (1920) argued that the shortest precession period of solstices is 20,7kyr: the Summer solstice in one hemisphere takes place alternately every 11kyr at perihelion and at aphelion. We have subm…
▽ More
The Earth's revolution is modified by changes in inclination of its rotation axis. Despite the fact that the gravity field is central, the Earth's trajectory is not closed and the equinoxes drift. Milankovic (1920) argued that the shortest precession period of solstices is 20,7kyr: the Summer solstice in one hemisphere takes place alternately every 11kyr at perihelion and at aphelion. We have submitted the time series for the Earth's pole of rotation, global mean surface temperature and ephemeris to iterative Singular Spectrum Analysis. iSSA extracts from each a trend, a 1yr and a 60yr component. Both the apparent drift of solstices of Earth around the Sun and the global mean temperature exhibit a strong 60yr oscillation. The "fixed dates" of solstices actually drift. Comparing the time evolution of the Winter and Summer solstices positions of the rotation pole and the first iSSA component (trend) of the temperature allows one to recognize some common features. A basic equation from Milankovic links the derivative of heat received at a given location on Earth to solar insolation, known functions of the location coordinates, solar declination and hour angle, with an inverse square dependence on the Sun-Earth distance. We have translated the drift of solstices as a function of distance to the Sun into the geometrical insolation theory of Milankovic. Shifting the inverse square of the 60yr iSSA drift of solstices by 15 years with respect to the first derivative of the 60yr iSSA trend of temperature, that is exactly a quadrature in time, puts the two curves in quasi-exact superimposition. The probability of a chance coincidence appears very low. Correlation does not imply causality when there is no accompanying model. Here Milankovic's equation can be considered as a model that is widely accepted. This paper identifies a case of agreement between observations and a mathematical formulation.
△ Less
Submitted 19 July, 2022;
originally announced July 2022.
-
On the Nature and Origin of Atmospheric Annual and Semi-annual Oscillations
Authors:
V. Courtillot,
J-L. Le Mouël,
F. Lopes,
D. Gibert
Abstract:
This paper proposes a joint analysis of variations of global sea-level pressure and of Earth's rotation (RP), expressed as the coordinates of the rotation pole and length of day. Sea-Level-Pressure (SLP) extracted components are a weak trend, eleven quasi-periodic or periodic components ($\sim$130, 90, 50, 22, 15, 4, 1.8 yr), an annual cycle and its first three harmonics. These periods are charact…
▽ More
This paper proposes a joint analysis of variations of global sea-level pressure and of Earth's rotation (RP), expressed as the coordinates of the rotation pole and length of day. Sea-Level-Pressure (SLP) extracted components are a weak trend, eleven quasi-periodic or periodic components ($\sim$130, 90, 50, 22, 15, 4, 1.8 yr), an annual cycle and its first three harmonics. These periods are characteristic of the space-time evolution of the Earth's rotation axis and are present in many characteristic features of solar and terrestrial physics. We focus mainly on the annual and to a lesser extent semi-annual components. Maps of the first three components of SLP (that together comprise more than 85% of the data variance) reveal interesting symmetries. The trend is very stable and forms a triskel structure that can be modeled as Taylor-Couette flow of mode 3. The annual component is characterized by a large negative anomaly extending over Eurasia in the NH Summer (and the opposite in the NH Winter) and three large positive anomalies over Australia and the southern tips of South America and South Africa in the SH Spring (and the opposite in the SH Autumn) forming a triskel. The semi-annual component is characterized by three positive anomalies (an irregular triskel) in the NH Spring and Autumn (and the opposite in the NH Summer and Winter), and in the SH Spring and Autumn by a strong stable pattern consisting of three large negative anomalies forming a clear triskel within the 40$^{\circ}$-60$^{\circ}$ annulus formed by the southern oceans. A large positive anomaly centered over Antarctica, with its maximum displaced toward Australia, and a smaller one centered over Southern Africa complement the pattern.
△ Less
Submitted 24 June, 2022;
originally announced June 2022.
-
Fusing Feature Engineering and Deep Learning: A Case Study for Malware Classification
Authors:
Daniel Gibert,
Carles Mateu,
Jordi Planes,
Quan Le
Abstract:
Machine learning has become an appealing signature-less approach to detect and classify malware because of its ability to generalize to never-before-seen samples and to handle large volumes of data. While traditional feature-based approaches rely on the manual design of hand-crafted features based on experts knowledge of the domain, deep learning approaches replace the manual feature engineering p…
▽ More
Machine learning has become an appealing signature-less approach to detect and classify malware because of its ability to generalize to never-before-seen samples and to handle large volumes of data. While traditional feature-based approaches rely on the manual design of hand-crafted features based on experts knowledge of the domain, deep learning approaches replace the manual feature engineering process by an underlying system, typically consisting of a neural network with multiple layers, that perform both feature learning and classification altogether. However, the combination of both approaches could substantially enhance detection systems. In this paper we present an hybrid approach to address the task of malware classification by fusing multiple types of features defined by experts and features learned through deep learning from raw data. In particular, our approach relies on deep learning to extract N-gram like features from the assembly language instructions and the bytes of malware, and texture patterns and shapelet-based features from malwareś grayscale image representation and structural entropy, respectively. These deep features are later passed as input to a gradient boosting model that combines the deep features and the hand-crafted features using an early-fusion mechanism. The suitability of our approach has been evaluated on the Microsoft Malware Classification Challenge benchmark and results show that the proposed solution achieves state-of-the-art performance and outperforms gradient boosting and deep learning methods in the literature.
△ Less
Submitted 12 June, 2022;
originally announced June 2022.
-
On the annual and semi-annual components of variations in extent of Arctic and Antarctic sea-ice
Authors:
F. Lopes,
V. Courtillot,
D. Gibert,
JL Le Mouël
Abstract:
The time series of northern hemisphere (NHSI) and southern hemisphere (SHSI) sea-ice extent are submitted to singular spectral analysis (SSA). The components are analyzed with Laplace's formulation of the Liouville-Euler system. As already shown in a previous work, the trends observed in the time series are quasi linear, decreasing for NHSI and increasing for SHSI. The amplitude of annual variatio…
▽ More
The time series of northern hemisphere (NHSI) and southern hemisphere (SHSI) sea-ice extent are submitted to singular spectral analysis (SSA). The components are analyzed with Laplace's formulation of the Liouville-Euler system. As already shown in a previous work, the trends observed in the time series are quasi linear, decreasing for NHSI and increasing for SHSI. The amplitude of annual variations in the Antarctic is double that in the Arctic, they are in phase opposition, modulated. The semi-annual components are in quadrature. The first 3 components of both NHSI and SHSI at 1, 1/2 and 1/3 yr account for more than 95% of the signal variance. We complement previous analyses of variations in pole position (PM = m1, m2) and length of day (lod). Whereas SSA of lod is dominated by the same first 3 components as sea-ice, SSA of PM contains only the 1 yr and the Chandler components. The 1 yr component of NHSI is in phase with that of lod and in phase opposition with m1. The reverse holds for the 1 yr component of SHSI. We note that the semi-annual component appears in lod not in m1. The annual and semi-annual components of NHSI and SHSI are much larger than the trends observed since 1978, that leads us to test whether a first order geophysical or astronomical forcing should not be preferred to the mechanisms generally suggested as a forcing factor of the trends. The lack of modulation of the largest forced component suggests an alternate mechanism. In Laplace's paradigm, the torques exerted by the Moon, Sun and planets play the leading role as the source of forcing of many geophysical phenomena. These forces lead to changes in the inclination of the Earth's rotation axis, setting Earth masses in motion and resulting in thermal dissipation. It is variations in inclination of the rotation axis that lead to the large annual components of melting and re-freezing of sea-ice.
△ Less
Submitted 19 May, 2022;
originally announced May 2022.
-
On two formulations of polar motion and identification of its sources
Authors:
F. Lopes,
V. Courtillot,
JL. Le Mouël,
D Gibert
Abstract:
Differences in interpretation may arise from differences in formulation of the equations of celestial mechanics. This paper focuses on the Liouville system of differential equations. In a "modern" presentation of the equations, variations in polar motion (PM) and variations in length of day (lod) are decoupled. Their source terms result from redistribution of masses and torques. Elasticity of the…
▽ More
Differences in interpretation may arise from differences in formulation of the equations of celestial mechanics. This paper focuses on the Liouville system of differential equations. In a "modern" presentation of the equations, variations in polar motion (PM) and variations in length of day (lod) are decoupled. Their source terms result from redistribution of masses and torques. Elasticity of the Earth, large earthquakes, or external forcing by the fluid envelopes have been successively invoked in the "modern" approach. In the "classical" presentation, PM is governed by the inclination of Earth's rotation pole and the derivative of its declination (close to the derivative of lod). These are coupled by the Liouville system and all "source" terms are astronomical. The "classical" approach also implies that there should be a link between the rotations and the torques exerted by the planets of the solar system. A sufficiently strong source of energy can be found. There is remarkable agreement between the sum of forces exerted by the four Jovian planets and components of Earth's polar motion. Since 1970, lod tends to decrease, but a recent acceleration of rotation velocity, that contradicts previous models, finds a simple explanation with the "classical" formalism. Analysis of lod finds nine components, all with physical sense: first comes a "trend", then pseudo-oscillations with periods 80 yr , 18.6 yr, 11 yr , 1 year and 0.5 yr, 27.54 days, 13.66 days, 13.63 days and 9.13 days. The longer periods, 1 yr, 11 yr, 18.6 yr and 80 yr are common to lod and PM. The Liouville system of equations in the "classical" analysis implies a strong link between the two: the straightening of the inclination of the axis of rotation should accompany a decrease in lod.Also, if this link is valid, all the components with extraterrestrial periods should be present in the series of sunspots and indeed they are.
△ Less
Submitted 25 April, 2022;
originally announced April 2022.
-
Enhancing the Insertion of NOP Instructions to Obfuscate Malware via Deep Reinforcement Learning
Authors:
Daniel Gibert,
Matt Fredrikson,
Carles Mateu,
Jordi Planes,
Quan Le
Abstract:
Current state-of-the-art research for tackling the problem of malware detection and classification is centered on the design, implementation and deployment of systems powered by machine learning because of its ability to generalize to never-before-seen malware families and polymorphic mutations. However, it has been shown that machine learning models, in particular deep neural networks, lack robus…
▽ More
Current state-of-the-art research for tackling the problem of malware detection and classification is centered on the design, implementation and deployment of systems powered by machine learning because of its ability to generalize to never-before-seen malware families and polymorphic mutations. However, it has been shown that machine learning models, in particular deep neural networks, lack robustness against crafted inputs (adversarial examples). In this work, we have investigated the vulnerability of a state-of-the-art shallow convolutional neural network malware classifier against the dead code insertion technique. We propose a general framework powered by a Double Q-network to induce misclassification over malware families. The framework trains an agent through a convolutional neural network to select the optimal positions in a code sequence to insert dead code instructions so that the machine learning classifier mislabels the resulting executable. The experiments show that the proposed method significantly drops the classification accuracy of the classifier to 56.53% while having an evasion rate of 100% for the samples belonging to the Kelihos_ver3, Simda, and Kelihos_ver1 families. In addition, the average number of instructions needed to mislabel malware in comparison to a random agent decreased by 33%.
△ Less
Submitted 18 November, 2021;
originally announced November 2021.
-
A multi-decadal view of the heat and mass budget of a volcano in unrest: La Soufrière de Guadeloupe (French West Indies)
Authors:
David E. Jessop,
Séverine Moune,
Roberto Moretti,
Dominique Gibert,
Jean-Christophe Komorowski,
Vincent Robert,
Michael J. Heap,
Alexis Bosson,
Magali Bonifacie,
Sébastien Deroussi,
Céline Dessert,
Marina Rosas-Carbajal,
Arnaud Lemarchand,
Arnaud Burtin
Abstract:
Particularly in the presence of a hydrothermal system, many volcanoes output large quantities of heat through the transport of water from deep within the edifice to the surface. Thus, heat flux is a prime tool for evaluating volcanic activity and unrest. We review the volcanic unrest at La Soufrière de Guadeloupe (French West Indies) using an airborne thermal camera survey, and in-situ measurement…
▽ More
Particularly in the presence of a hydrothermal system, many volcanoes output large quantities of heat through the transport of water from deep within the edifice to the surface. Thus, heat flux is a prime tool for evaluating volcanic activity and unrest. We review the volcanic unrest at La Soufrière de Guadeloupe (French West Indies) using an airborne thermal camera survey, and in-situ measurements of temperature and flow rate through temperature probes, Pitot-tube and MultiGAS measurements. We deduce mass and heat fluxes for the fumarolic, ground and thermal spring outputs and follow these over a period spanning 2000--2020. Our results are compared with published data and we performed a retrospective analysis of the temporal variations in heat flux over this period using the literature data.
We find that the heat emitted by the volcano is 36.5 +/- 7.9 MW, of which the fumarolic heat flux is dominant at 28.3 +/- 6.8 MW. Given a total heated area of 26780 m2, this equates to a heat flux density of 627 +/- 94 W/m2, which is amongst the highest established for worldwide volcanoes with hydrothermal systems, particularly for dome volcanoes. A major change at La Soufrière de Guadeloupe, however, is the development of a widespread region of ground heating at the summit where heat output has increased from 0.2 +/- 0.1 MW in 2010 to 5.7 +/- 0.9 MW in 2020. This change is concurrent with accelerating unrest at the volcano, and the emergence of two new high-flux fumaroles in recent years. Our findings highlight the importance of continued and enhanced surveillance and research strategies at La Soufrière de Guadeloupe, the results of which can be used to better understand hydrothermal volcanism the world over.
△ Less
Submitted 21 December, 2020; v1 submitted 22 October, 2020;
originally announced October 2020.
-
Middle-atmosphere dynamics observed with a portable muon detector
Authors:
Matias Tramontini,
Marina Rosas-Carbajal,
Christophe Nussbaum,
Dominique Gibert,
Jacques Marteau
Abstract:
In the past years, large particle-physics experiments have shown that muon rate variations detected in underground laboratories are sensitive to regional, middle-atmosphere temperature variations. Potential applications include tracking short-term atmosphere dynamics, such as Sudden Stratospheric Warmings. We report here that such sensitivity is not only limited to large surface detectors under hi…
▽ More
In the past years, large particle-physics experiments have shown that muon rate variations detected in underground laboratories are sensitive to regional, middle-atmosphere temperature variations. Potential applications include tracking short-term atmosphere dynamics, such as Sudden Stratospheric Warmings. We report here that such sensitivity is not only limited to large surface detectors under high-opacity conditions. We use a portable muon detector conceived for muon tomography for geophysical applications and we study muon rate variations observed over one year of measurements at the Mont Terri Underground Rock Laboratory, Switzerland (opacity of ~700 meter water equivalent). We observe a direct correlation between middle-atmosphere seasonal temperature variations and muon rate. Muon rate variations are also sensitive to the abnormal atmosphere heating in January-February 2017, associated to a Sudden Stratospheric Warming. Estimates of the effective temperature coefficient for our particular case agree with theoretical models and with those calculated from large neutrino experiments under comparable conditions. Thus, portable muon detectors may be useful to 1) study seasonal and short-term middle atmosphere dynamics, especially in locations where data is lacking such as mid-latitudes; and 2) improve the calibration of the effective temperature coefficient for different opacity conditions. Furthermore, we highlight the importance of assessing the impact of temperature on muon rate variations when considering geophysical applications. Depending on latitude and opacity conditions, this effect may be large enough to hide subsurface density variations due to changes in groundwater content, and should therefore be removed from the time-series.
△ Less
Submitted 4 September, 2019;
originally announced September 2019.
-
Abrupt changes of hydrothermal activity in a lava dome detected by combined seismic and muon monitoring
Authors:
Y. Le Gonidec,
M. Rosas-Carbajal,
J. de Bremond d'Ars,
B. Carlus,
J. -C. Ianigro,
B. Kergosien,
J. Marteau,
D. Gibert
Abstract:
The recent 2014 eruption of the Ontake volcano in Japan recalled that hydrothermal fields of moderately active volcanoes have an unpredictable and hazardous behavior that may endanger human beings. Steam blasts can expel devastating ejecta and create craters of several tens of meters. The management of such hydrothermal hazards in populated areas is problematic because of their very short time con…
▽ More
The recent 2014 eruption of the Ontake volcano in Japan recalled that hydrothermal fields of moderately active volcanoes have an unpredictable and hazardous behavior that may endanger human beings. Steam blasts can expel devastating ejecta and create craters of several tens of meters. The management of such hydrothermal hazards in populated areas is problematic because of their very short time constants. At present no precursory signal is clearly identified as a potential warning of imminent danger. Here we show how the combination of seismic noise monitoring and muon density tomography allows to detect, with an unprecedented space and time resolution, the increase of activity of a hydrothermal focus located 50 to 100 m below the summit of an active volcano, the La Soufrière of Guadeloupe, in the Lesser Antilles. The present study deals with hydrothermal activity events at timescales of few hours to few days. We show how the combination of those two methods improves the risk evaluation of short-term hazards and the localization of the involved volumes in the volcano. We anticipate that the deployment of networks of various sensors including temperature probes, seismic antennas and cosmic muon telescopes around such volcanoes could valuably contribute to early warning decisions.
△ Less
Submitted 17 November, 2018;
originally announced November 2018.
-
Forward scattering effects on muon imaging
Authors:
H. Gómez,
D. Gibert,
C. Goy,
K. Jourde,
Y. Karyotakis,
S. Katsanevas,
J. Marteau,
M. Rosas-Carbajal,
A. Tonazzo
Abstract:
Muon imaging is one of the most promising non-invasive techniques for density structure scanning, specially for large objects reaching the kilometre scale. It has already interesting applications in different fields like geophysics or nuclear safety and has been proposed for some others like engineering or archaeology. One of the approaches of this technique is based on the well-known radiography…
▽ More
Muon imaging is one of the most promising non-invasive techniques for density structure scanning, specially for large objects reaching the kilometre scale. It has already interesting applications in different fields like geophysics or nuclear safety and has been proposed for some others like engineering or archaeology. One of the approaches of this technique is based on the well-known radiography principle, by reconstructing the incident direction of the detected muons after crossing the studied objects. In this case, muons detected after a previous forward scattering on the object surface represent an irreducible background noise, leading to a bias on the measurement and consequently on the reconstruction of the object mean density. Therefore, a prior characterization of this effect represents valuable information to conveniently correct the obtained results. Although the muon scattering process has been already theoretically described, a general study of this process has been carried out based on Monte Carlo simulations, resulting in a versatile tool to evaluate this effect for different object geometries and compositions. As an example, these simulations have been used to evaluate the impact of forward scattered muons on two different applications of muon imaging: archaeology and volcanology, revealing a significant impact on the latter case. The general way in which all the tools used have been developed can allow to make equivalent studies in the future for other muon imaging applications following the same procedure.
△ Less
Submitted 13 December, 2017; v1 submitted 15 September, 2017;
originally announced September 2017.
-
Three-dimensional density structure of La Soufrieère de Guadeloupe lava dome from simultaneous muon radiographies and gravity data
Authors:
Marina Rosas-Carbajal,
Kevin Jourde,
Jacques Marteau,
Sébastien Deroussi,
Jean-Christophe Komorowski,
Dominique Gibert
Abstract:
Muon imaging has recently emerged as a powerful method to complement standard geophysical tools in the understanding of the Earth's subsurface. Muon measurements can yield a radiography of the average density along the muon path, allowing to image large volumes of a geological body from a single observation point. Here we jointly invert muon data from three simultaneous telescope acquisitions toge…
▽ More
Muon imaging has recently emerged as a powerful method to complement standard geophysical tools in the understanding of the Earth's subsurface. Muon measurements can yield a radiography of the average density along the muon path, allowing to image large volumes of a geological body from a single observation point. Here we jointly invert muon data from three simultaneous telescope acquisitions together with gravity data to estimate the three-dimensional density structure of the La Soufrière de Guadeloupe lava Dome. Our unique dataset allows us to achieve an unprecedented spatial resolution with this novel technique. The retrieved density model reveals an extensive, low-density anomaly where the most active part of the volcanic hydrothermal system is located, supporting previous studies that indicate this region as the most likely to be involved in a partial edifice collapse.
△ Less
Submitted 9 July, 2017;
originally announced July 2017.
-
DIAPHANE: Muon tomography applied to volcanoes, civil engineering, archaelogy
Authors:
Jacques Marteau,
Jean de Bremond d'Ars,
Dominique Gibert,
Kevin Jourde,
Jean-Christophe Ianigro,
Bruno Carlus
Abstract:
Muography techniques applied to geological structures greatly improved in the past ten years. Recent applications demonstrate the interest of the method not only to perform structural imaging but also to monitor the dynamics of inner movements like magma ascent inside volcanoes or density variations in hydrothermal systems. Muography time-resolution has been studied thanks to dedicated experiments…
▽ More
Muography techniques applied to geological structures greatly improved in the past ten years. Recent applications demonstrate the interest of the method not only to perform structural imaging but also to monitor the dynamics of inner movements like magma ascent inside volcanoes or density variations in hydrothermal systems. Muography time-resolution has been studied thanks to dedicated experiments, e.g. in a water tower tank. This paper presents the activities of the DIAPHANE collaboration between particle- and geo-physicists and the most recent results obtained in the field of volcanology, with a focus on the main target, the Soufrière of Guadeloupe active volcano. Special emphasis is given on the monitoring of the dome's inner volumes opacity variations, that could be inferred to the hydrothermal system dynamics (vaporization of inner liquid water in coincidence with the appearance of new fumaroles at the summit). I also breifly present results obtained in the fields of civil engineering (study of urban underground tunnels) and archaelogy (greek tumulus scanning).
△ Less
Submitted 12 December, 2016;
originally announced December 2016.
-
Muon dynamic radiography of density changes induced by hydrothermal activity at the La Soufrière of Guadeloupe volcano
Authors:
Kevin Jourde,
Dominique Gibert,
Jacques Marteau,
Jean de Bremond d'Ars,
Jean-Christophe Komorowski
Abstract:
Imaging geological structures through cosmic muon radiography is a newly developed technique particularly interesting in volcanology. Here we show that muon radiography may be efficient to detect and characterize mass movements in shallow hydrothermal systems of low-energy active volcanoes like the La Soufrière lava dome. We present an experiment conducted on this volcano during the Summer $2014$…
▽ More
Imaging geological structures through cosmic muon radiography is a newly developed technique particularly interesting in volcanology. Here we show that muon radiography may be efficient to detect and characterize mass movements in shallow hydrothermal systems of low-energy active volcanoes like the La Soufrière lava dome. We present an experiment conducted on this volcano during the Summer $2014$ and bring evidence that huge density changes occurred in three domains of the lava dome. Depending on their position and on the medium porosity the volumes of these domains vary from $1 \times 10^6 \; \mathrm{m}^3$ to $7 \times 10^6 \; \mathrm{m}^3$. However, the mass changes remain quite constant, two of them being negative ($Δm \approx -0.6 \times 10^9 \; \mathrm{kg}$) and a third one being positive ($Δm \approx +2 \times 10^9 \; \mathrm{kg}$). We attribute the negative mass changes to the formation of steam in shallow hydrothermal reservoir previously partly filled with liquid water. This coincides with the apparition of new fumaroles on top of the volcano. The positive mass change is synchronized with the negative mass changes indicating that liquid water probably flowed from the two reservoirs invaded by steam toward the third reservoir.
△ Less
Submitted 10 June, 2016; v1 submitted 9 June, 2016;
originally announced June 2016.
-
Muon tomography applied to active volcanoes
Authors:
Jacques Marteau,
Bruno Carlus,
Dominique Gibert,
Jean-Christophe Ianigro,
Kevin Jourde,
Bruno Kergosien,
Pascal Rolland
Abstract:
Muon tomography is a generic imaging method using the differential absorption of cosmic muons by matter. The measured contrast in the muons flux reflects the matter density contrast as it does in conventional medical imaging. The applications to volcanology present may advantadges induced by the features of the target itself: limited access to dangerous zones, impossible use of standard boreholes…
▽ More
Muon tomography is a generic imaging method using the differential absorption of cosmic muons by matter. The measured contrast in the muons flux reflects the matter density contrast as it does in conventional medical imaging. The applications to volcanology present may advantadges induced by the features of the target itself: limited access to dangerous zones, impossible use of standard boreholes information, harsh environmental conditions etc. The Diaphane project is one of the largest and leading collaboration in the field and the present article summarizes recent results collected on the Lesser Antilles, with a special emphasis on the Soufrière of Guadeloupe.
△ Less
Submitted 18 October, 2015;
originally announced October 2015.
-
Monitoring temporal opacity fluctuations of large structures with muon tomography : a calibration experiment using a water tower tank
Authors:
Kevin Jourde,
Dominique Gibert,
Jacques Marteau,
Jean de Bremond d'Ars,
Serge Gardien,
Claude Girerd,
Jean-Christophe Ianigro
Abstract:
Usage of secondary cosmic muons to image the geological structures density distribution significantly developed during the past ten years. Recent applications demonstrate the method interest to monitor magma ascent and volcanic gas movements inside volcanoes. Muon radiography could be used to monitor density variations in aquifers and the critical zone in the near surface. However, the time resolu…
▽ More
Usage of secondary cosmic muons to image the geological structures density distribution significantly developed during the past ten years. Recent applications demonstrate the method interest to monitor magma ascent and volcanic gas movements inside volcanoes. Muon radiography could be used to monitor density variations in aquifers and the critical zone in the near surface. However, the time resolution achievable by muon radiography monitoring remains poorly studied. It is biased by fluctuation sources exterior to the target, and statistically affected by the limited number of particles detected during the experiment. The present study documents these two issues within a simple and well constrained experimental context: a water tower. We use the data to discuss the influence of atmospheric variability that perturbs the signal, and propose correction formulas to extract the muon flux variations related to the water level changes. Statistical developments establish the feasibility domain of muon radiography monitoring as a function of target thickness (i.e. opacity). Objects with a thickness comprised between $\simeq$ 50 $\pm$ 30m water equivalent correspond to the best time resolution. Thinner objects have a degraded time resolution that strongly depends on the zenith angle, whereas thicker objects (like volcanoes) time resolution does not.
△ Less
Submitted 19 March, 2016; v1 submitted 9 April, 2015;
originally announced April 2015.
-
Joint inversion of muon tomography and gravimetry - a resolving kernel approach
Authors:
Kevin Jourde,
Dominique Gibert,
Jacques Marteau
Abstract:
Both muon tomography and gravimetry are geophysical methods that provide information on the density structure of the Earth's subsurface. Muon tomography measures the natural flux of cosmic muons and its attenuation produced by the screening effect of the rock mass to image. Gravimetry generally consists in measurements of the vertical component of the local gravity field. Both methods are linearly…
▽ More
Both muon tomography and gravimetry are geophysical methods that provide information on the density structure of the Earth's subsurface. Muon tomography measures the natural flux of cosmic muons and its attenuation produced by the screening effect of the rock mass to image. Gravimetry generally consists in measurements of the vertical component of the local gravity field. Both methods are linearly linked to density, but their spatial sensitivity is very different. Muon tomography essentially works like medical X-ray scan and integrates density information along elongated narrow conical volumes while gravimetry measurements are linked to density by a 3-dimensional integral encompassing the whole studied domain. We develop the mathematical expressions of these integration formulas -- called acquisition kernels -- to express resolving kernels that act as spatial filters relating the true unknown density structure to the density distribution actually recoverable from the available data. The resolving kernels provide a tool to quantitatively describe the resolution of the density models and to evaluate the resolution improvement expected by adding new data in the inversion. The resolving kernels derived in the joined muon/gravimetry case indicate that gravity data are almost useless to constrain the density structure in regions sampled by more than two muon tomography acquisitions. Interestingly the resolution in deeper regions not sampled by muon tomography is significantly improved by joining the two techniques. Examples taken from field experiments performed on La Soufrière of Guadeloupe volcano are discussed.
△ Less
Submitted 19 November, 2014;
originally announced November 2014.
-
Implementation of sub-nanoseconds TDC in FPGA: applications to time-of-flight analysis in muon radiography
Authors:
J. Marteau,
J. De Bremond D'ars,
D. Gibert,
K. Jourde,
S. Gardien,
C. Girerd,
J. -C. Ianigro
Abstract:
Time-of-flight (tof) techniques are standard techniques in high energy physics to determine particles propagation directions. Since particles velocities are generally close to c, the speed of light, and detectors typical dimensions at the meter level, the state-of-the-art tof techniques should reach sub-nanosecond timing resolution. Among the various techniques already available, the recently deve…
▽ More
Time-of-flight (tof) techniques are standard techniques in high energy physics to determine particles propagation directions. Since particles velocities are generally close to c, the speed of light, and detectors typical dimensions at the meter level, the state-of-the-art tof techniques should reach sub-nanosecond timing resolution. Among the various techniques already available, the recently developed ring oscillator TDC ones, implemented in low cost FPGA, feature a very interesting figure of merit since a very good timing performance may be achieved with limited processing ressources. This issue is relevant for applications where unmanned sensors should have the lowest possible power consumption. Actually this article describes in details the application of this kind of tof technique to muon tomography of geological bodies. Muon tomography aims at measuring density variations and absolute densities through the detection of atmospheric muons flux's attenuation, due to the presence of matter. When the measured fluxes become very low, an identified source of noise comes from backwards propagating particles hitting the detector in a direction pointing to the geological body. The separation between through-going and backward-going particles, on the basis of the tof information is therefore a key parameter for the tomography analysis and subsequent previsions.
△ Less
Submitted 16 October, 2013;
originally announced October 2013.
-
Effects of upward-going cosmic muons on density radiography of volcanoes
Authors:
K. Jourde,
D. Gibert,
J. Marteau,
J. De Bremond D'ars,
S. Gardien,
C. Girerd,
J. -C. Ianigro,
D. Carbone
Abstract:
Muon tomography aims at deriving the density structure of geological bodies from their screening attenuation produced on the natural cosmic muons flux. Because of their open-sky exposure, muons telescopes are subject to noise fluxes with large intensities relative to the tiny flux of interest. A recognized source of noise flux comes from fake tracks caused by particles that fortuitously trigger th…
▽ More
Muon tomography aims at deriving the density structure of geological bodies from their screening attenuation produced on the natural cosmic muons flux. Because of their open-sky exposure, muons telescopes are subject to noise fluxes with large intensities relative to the tiny flux of interest. A recognized source of noise flux comes from fake tracks caused by particles that fortuitously trigger the telescope detectors at the same time. Such a flux may be reduced by using multiple-detector telescopes so that fortuitous events become very unlikely. In the present study, we report on a different type of noise flux caused by upward-going muons crossing the detectors from the rear side. We describe field experiments on La Soufriére of Guadeloupe and Mount Etna, and give details on the high-resolution clocking system and the statistical procedure necessary to detect upward-going muons. We analyse several data sets acquired either in calibration or in volcano tomography situation. All data sets are shown clearly biased by upward-going noise flux whose intensity may amount to 50% of the measured total flux in given directions. Biases produced on density radiographies by this kind of flux are quantified and correction procedures are detailed. Examples for La Soufriére and Mount Etna are given.
△ Less
Submitted 25 July, 2013;
originally announced July 2013.
-
Muons tomography applied to geosciences and volcanology
Authors:
J. Marteau,
D. Gibert,
N. Lesparre,
F. Nicollin,
P. Noli,
F. Giacoppo
Abstract:
Imaging the inner part of large geological targets is an important issue in geosciences with various applications. Dif- ferent approaches already exist (e.g. gravimetry, electrical tomography) that give access to a wide range of informations but with identified limitations or drawbacks (e.g. intrinsic ambiguity of the inverse problem, time consuming deployment of sensors over large distances). Her…
▽ More
Imaging the inner part of large geological targets is an important issue in geosciences with various applications. Dif- ferent approaches already exist (e.g. gravimetry, electrical tomography) that give access to a wide range of informations but with identified limitations or drawbacks (e.g. intrinsic ambiguity of the inverse problem, time consuming deployment of sensors over large distances). Here we present an alternative and complementary tomography method based on the measurement of the cosmic muons flux attenuation through the geological structures. We detail the basics of this muon tomography with a special emphasis on the photo-active detectors.
△ Less
Submitted 31 January, 2012;
originally announced January 2012.
