The Einstein-Hilbert-Palatini formalism in Pseudo-Finsler Geometry

A systematic development of the so-called Palatini formalism is carried out for pseudo-Finsler metrics $L$ of any signature. Substituting in the classical Einstein-Hilbert-Palatini functional the scalar curvature by the Finslerian Ricci scalar constructed with an independent nonlinear connection $mathrm{N}$, the metric and affine equations for $(mathrm{N},L)$ are obtained. In Lorentzian signature with vanishing mean Landsberg tensor $mathrm{Lan}_i$, both the Finslerian Hilbert metric equation and the classical Palatini conclusions are recovered by means of a combination of techniques involving the (Riemannian) maximum principle and an original argument about divisibility and fiberwise analyticity. Some of these findings are also extended to (positive definite) Riemannian metrics by using the eigenvalues of the Laplacian. When $mathrm{Lan}_i eq 0$, the Palatini conclusions fail necessarily, however, a good number of properties of the solutions remain. The framework and proofs are built up in detail.

Can Evil IoT Twins Be Identified? Now Yes, a Hardware Behavioral Fingerprinting Methodology

The connectivity and resource-constrained nature of IoT, and in particular single-board devices, opens up to cybersecurity concerns affecting the Industrial Internet of Things (IIoT). One of the most important is the presence of evil IoT twins. Evil IoT twins are malicious devices, with identical hardware and software configurations to authorized ones, that can provoke sensitive information leakages, data poisoning, or privilege escalation in industrial scenarios. Combining behavioral fingerprinting and Machine/Deep Learning (ML/DL) techniques is a promising solution to identify evil IoT twins by detecting minor performance differences generated by imperfections in manufacturing. However, existing solutions are not suitable for single-board devices because they do not consider their hardware and software limitations, underestimate critical aspects during the identification performance evaluation, and do not explore the potential of ML/DL techniques. Moreover, there is a dramatic lack of work explaining essential aspects to considering during the identification of identical devices. This work proposes an ML/DL-oriented methodology that uses behavioral fingerprinting to identify identical single-board devices. The methodology leverages the different built-in components of the system, comparing their internal behavior with each other to detect variations that occurred in manufacturing processes. The validation has been performed in a real environment composed of identical Raspberry Pi 4 Model B devices, achieving the identification for all devices by setting a 50% threshold in the evaluation process. Finally, a discussion compares the proposed solution with related work and provides important lessons learned and limitations.

Federated Learning for Malware Detection in IoT Devices

This work investigates the possibilities enabled by federated learning concerning IoT malware detection and studies security issues inherent to this new learning paradigm. In this context, a framework that uses federated learning to detect malware affecting IoT devices is presented. N-BaIoT, a dataset modeling network traffic of several real IoT devices while affected by malware, has been used to evaluate the proposed framework. Both supervised and unsupervised federated models (multi-layer perceptron and autoencoder) able to detect malware affecting seen and unseen IoT devices of N-BaIoT have been trained and evaluated. Furthermore, their performance has been compared to two traditional approaches. The first one lets each participant locally train a model using only its own data, while the second consists of making the participants share their data with a central entity in charge of training a global model. This comparison has shown that the use of more diverse and large data, as done in the federated and centralized methods, has a considerable positive impact on the model performance. Besides, the federated models, while preserving the participants privacy, show similar results as the centralized ones. As an additional contribution and to measure the robustness of the federated approach, an adversarial setup with several malicious participants poisoning the federated model has been considered. The baseline model aggregation averaging step used in most federated learning algorithms appears highly vulnerable to different attacks, even with a single adversary. The performance of other model aggregation functions acting as countermeasures is thus evaluated under the same attack scenarios. These functions provide a significant improvement against malicious participants, but more efforts are still needed to make federated approaches robust.

A Survey on Device Behavior Fingerprinting: Data Sources, Techniques, Application Scenarios, and Datasets

In the current network-based computing world, where the number of interconnected devices grows exponentially, their diversity, malfunctions, and cybersecurity threats are increasing at the same rate. To guarantee the correct functioning and performance of novel environments such as Smart Cities, Industry 4.0, or crowdsensing, it is crucial to identify the capabilities of their devices (e.g., sensors, actuators) and detect potential misbehavior that may arise due to cyberattacks, system faults, or misconfigurations. With this goal in mind, a promising research field emerged focusing on creating and managing fingerprints that model the behavior of both the device actions and its components. The article at hand studies the recent growth of the device behavior fingerprinting field in terms of application scenarios, behavioral sources, and processing and evaluation techniques. First, it performs a comprehensive review of the device types, behavioral data, and processing and evaluation techniques used by the most recent and representative research works dealing with two major scenarios: device identification and device misbehavior detection. After that, each work is deeply analyzed and compared, emphasizing its characteristics, advantages, and limitations. This article also provides researchers with a review of the most relevant characteristics of existing datasets as most of the novel processing techniques are based on machine learning and deep learning. Finally, it studies the evolution of these two scenarios in recent years, providing lessons learned, current trends, and future research challenges to guide new solutions in the area.

Structure of globally hyperbolic spacetimes with timelike boundary

Globally hyperbolic spacetimes with timelike boundary $(overline{M} = M cup partial M, g)$ are the natural class of spacetimes where regular boundary conditions (eventually asymptotic, if $overline{M}$ is obtained by means of a conformal embedding) can be posed. $partial M$ represents the naked singularities and can be identified with a part of the intrinsic causal boundary. Apart from general properties of $partial M$, the splitting of any globally hyperbolic $(overline{M},g)$ as an orthogonal product ${mathbb R}times bar{Sigma}$ with Cauchy slices with boundary ${t}times bar{Sigma}$ is proved. This is obtained by constructing a Cauchy temporal function $tau$ with gradient $ abla tau$ tangent to $partial M$ on the boundary. To construct such a $tau$, results on stability of both, global hyperbolicity and Cauchy temporal functions are obtained. Apart from having their own interest, these results allow us to circumvent technical difficulties introduced by $partial M$. As a consequence, the interior $M$ both, splits orthogonally and can be embedded isometrically in ${mathbb L}^N$, extending so properties of globally spacetimes without boundary to a class of causally continuous ones.

On the definition and examples of cones and Finsler spacetimes

A systematic study of (smooth, strong) cone structures $C$ and Lorentz-Finsler metrics $L$ is carried out. As a link between both notions, cone triples $(Omega,T, F)$, where $Omega$ (resp. $T$) is a 1-form (resp. vector field) with $Omega(T)equiv 1$ and $F$, a Finsler metric on $ker (Omega)$, are introduced. Explicit descriptions of all the Finsler spacetimes are given, paying special attention to stationary and static ones, as well as to issues related to differentiability. In particular, cone structures $C$ are bijectively associated with classes of anisotropically conformal metrics $L$, and the notion of {em cone geodesic} is introduced consistently with both structures. As a non-relativistic application, the {em time-dependent} Zermelo navigation problem is posed rigorously, and its general solution is provided.

Ehlers-Kundt Conjecture about Gravitational Waves and Dynamical Systems

Ehlers-Kundt conjecture is a physical assertion about the fundamental role of plane waves for the description of gravitational waves. Mathematically, it becomes equivalent to a problem on the Euclidean plane ${mathbb R}^2$ with a very simple formulation in Classical Mechanics: given a non-necessarily autonomous potential $V(z,u)$, $(z,u)in {mathbb R}^2times {mathbb R}$, harmonic in $z$ (i.e. source-free), the trajectories of its associated dynamical system $ddot{z}(s)=- abla_z V(z(s),s)$ are complete (they live eternally) if and only if $V(z,u)$ is a polynomial in $z$ of degree at most $2$ (so that $V$ is a standard mathematical idealization of vacuum). Here, the conjecture is solved in the significative case that $V$ is bounded polynomially in $z$ for finite values of $uin {mathbb R}$. The mathematical and physical implications of this {em polynomial EK conjecture}, as well as the non-polynomial one, are discussed beyond their original scope.

Wind Riemannian spaceforms and Randers metrics of constant flag curvature

Recently, wind Riemannian structures (WRS) have been introduced as a generalization of Randers and Kropina metrics. They are constructed from the natural data for Zermelo navigation problem, namely, a Riemannian metric $g_R$ and a vector field $W$ (the wind), where, now, the restriction of mild wind $g_R(W,W)<1$ is dropped. Here, the models of WRS spaceforms of constant flag curvature are determined. Indeed, the celebrated classification of Randers metrics of constant flag curvature by Bao, Robles and Shen, extended to the Kropina case in the works by Yoshikawa, Okubo and Sabau, can be used to obtain the local classification. For the global one, a suitable result on completeness for WRS yields the complete simply connected models. In particular, any of the local models in the Randers classification does admit an extension to a unique model of wind Riemannian structure, even if it cannot be extended as a complete Finslerian manifold. Thus, WRSs emerge as the natural framework for the analysis of Randers spaceforms and, prospectively, wind Finslerian structures would become important for other global problems too. For the sake of completeness, a brief overview about WRS (including a useful link with the conformal geometry of a class of relativistic spacetimes) is also provided.

Sensitivity Projections for Dark Matter Searches with the Fermi Large Area Telescope

The nature of dark matter is a longstanding enigma of physics; it may consist of particles beyond the Standard Model that are still elusive to experiments. Among indirect search techniques, which look for stable products from the annihilation or decay of dark matter particles, or from axions coupling to high-energy photons, observations of the $gamma$-ray sky have come to prominence over the last few years, because of the excellent sensitivity of the Large Area Telescope (LAT) on the Fermi Gamma-ray Space Telescope mission. The LAT energy range from 20 MeV to above 300 GeV is particularly well suited for searching for products of the interactions of dark matter particles. In this report we describe methods used to search for evidence of dark matter with the LAT, and review the status of searches performed with up to six years of LAT data. We also discuss the factors that determine the sensitivities of these searches, including the magnitudes of the signals and the relevant backgrounds, considering both statistical and systematic uncertainties. We project the expected sensitivities of each search method for 10 and 15 years of LAT data taking. In particular, we find that the sensitivity of searches targeting dwarf galaxies, which provide the best limits currently, will improve faster than the square root of observing time. Current LAT limits for dwarf galaxies using six years of data reach the thermal relic level for masses up to 120 GeV for the $bbar{b}$ annihilation channel for reasonable dark matter density profiles. With projected discoveries of additional dwarfs, these limits could extend to about 250 GeV. With as much as 15 years of LAT data these searches would be sensitive to dark matter annihilations at the thermal relic cross section for masses to greater than 400 GeV (200 GeV) in the $bbar{b}$ ($tau^+ tau^-$) annihilation channels.

GLACE survey: OSIRIS/GTC Tuneable Filter H$alpha$ imaging of the rich galaxy cluster ZwCl 0024.0+1652 at z = 0.395. Part I -- Survey presentation, TF data reduction techniques and catalogue

The cores of clusters at 0 $lesssim$ z $lesssim$ 1 are dominated by quiescent early-type galaxies, whereas the field is dominated by star-forming late-type ones. Galaxy properties, notably the star formation (SF) ability, are altered as they fall into overdense regions. The critical issues to understand this evolution are how the truncation of SF is connected to the morphological transformation and the responsible physical mechanism. The GaLAxy Cluster Evolution Survey (GLACE) is conducting a study on the variation of galaxy properties (SF, AGN, morphology) as a function of environment in a representative sample of clusters. A deep survey of emission line galaxies (ELG) is being performed, mapping a set of optical lines ([OII], [OIII], H$beta$ and H$alpha$/[NII]) in several clusters at z $sim$ 0.40, 0.63 and 0.86. Using the Tunable Filters (TF) of OSIRIS/GTC, GLACE applies the technique of TF tomography: for each line, a set of images at different wavelengths are taken through the TF, to cover a rest frame velocity range of several thousands km/s. The first GLACE results target the H$alpha$/[NII] lines in the cluster ZwCl 0024.0+1652 at z = 0.395 covering $sim$ 2 $times$ r$_{vir}$. We discuss the techniques devised to process the TF tomography observations to generate the catalogue of H$alpha$ emitters of 174 unique cluster sources down to a SFR below 1 M$_{odot}$/yr. The AGN population is discriminated using different diagnostics and found to be $sim$ 37% of the ELG population. The median SFR is 1.4 M$_{odot}$/yr. We have studied the spatial distribution of ELG, confirming the existence of two components in the redshift space. Finally, we have exploited the outstanding spectral resolution of the TF to estimate the cluster mass from ELG dynamics, finding M$_{200}$ = 4.1 $times$ 10$^{14}$ M$_{odot} h^{-1}$, in agreement with previous weak-lensing estimates.