Search

In this paper, we address the problem of efficiently answering predicate queries on encrypted databases, those secured by Trusted Execution Environments (TEEs), which enable untrusted providers to process encrypted user data without revealing its contents. A common strategy in modern databases to accelerate predicate queries is the use of indexes, which map attribute values (keys) to their corresponding positions in a sorted data array. This allows for fast lookup and retrieval of data subsets that satisfy specific predicates. Unfortunately, indexes cannot be directly applied to encrypted databases due to strong data dependent leakages. Recent approaches apply differential privacy (DP) to construct noisy indexes that enable faster access to encrypted data while maintaining provable privacy guarantees. However, these methods often suffer from large storage costs, with index sizes typically scaling linearly with the key space. To address this challenge, we propose leveraging learned indexes, a trending technique that repurposes machine learning models as indexing structures, to build more compact DP indexes.

We present a comprehensive spectro-temporal analyses of recurrent outbursting black hole sources GX 339$-$4 and H 1743$-$322 using available AstroSat and NuSTAR archival observations during 2016-2024. The nature of the outburst profiles of both sources are examined using long-term MAXI/GSC and Swift/BAT lightcurves, and failed as well as successful outbursts are classified. Wide-band (0.5-60 keV) spectral modelling with disc (diskbb) and Comptonized (Nthcomp) components indicates that GX 339-4 transits from hard ($kT_{bb}=0.12-0.77$ keV, $\Gamma_{nth}=1.54-1.74$ and $L_{bol}=0.91-11.56$\% $L_{Edd}$) to soft state ($kT_{in}~[\approx{kT}_{bb}]=0.82-0.88$ keV, $\Gamma_{nth}=1.46-3.26$, $L_{bol}=19.59-30.06\%L_{Edd}$) via intermediate state ($kT_{in}~[\approx{kT}_{bb}]=0.56-0.88$ keV, $\Gamma_{nth}=1.76-2.66$, $L_{bol}=2.90-16.09\%L_{Edd}$), whereas H 1743-322 transits from quiescent to hard state ($\Gamma_{nth}=1.57-1.71$, $L_{bol}=2.08-3.48\%L_{Edd}$). We observe type-B and type-C Quasi-periodic Oscillations (QPOs) in GX 339-4 with increasing frequencies ($0.10-5.37$ Hz) along with harmonics. For H 1743-322, prominent type-C QPOs are observed in frequency range 0.22-1.01 Hz along with distinct harmonics. Energy-dependent power spectral studies reveal that fundamental QPO and harmonics disappear beyond 20 keV in GX 339-4, whereas fundamental QPO in H 1743-322 persists up to 40 keV. We also observe that type-C $QPO_{rms}\%$ decreases with energy for both sources although such variations appear marginal for type-B QPOs. Additionally, we report non-monotonic behavior of photon index with plasma temperature and detection of annihilation line. Finally, we discuss the relevance of the observational findings in the context of accretion dynamics around black hole binaries., Comment: 20 pages, 9 figures, 5 tables. Accepted for publication in MNRAS

With the rapid surge in the prevalence of Large Language Models (LLMs), individuals are increasingly turning to conversational AI for initial insights across various domains, including health-related inquiries such as disease diagnosis. Many users seek potential causes on platforms like ChatGPT or Bard before consulting a medical professional for their ailment. These platforms offer valuable benefits by streamlining the diagnosis process, alleviating the significant workload of healthcare practitioners, and saving users both time and money by avoiding unnecessary doctor visits. However, Despite the convenience of such platforms, sharing personal medical data online poses risks, including the presence of malicious platforms or potential eavesdropping by attackers. To address privacy concerns, we propose a novel framework combining FHE and Deep Learning for a secure and private diagnosis system. Operating on a question-and-answer-based model akin to an interaction with a medical practitioner, this end-to-end secure system employs Fully Homomorphic Encryption (FHE) to handle encrypted input data. Given FHE's computational constraints, we adapt deep neural networks and activation functions to the encryted domain. Further, we also propose a faster algorithm to compute summation of ciphertext elements. Through rigorous experiments, we demonstrate the efficacy of our approach. The proposed framework achieves strict security and privacy with minimal loss in performance.

We report the first detection of evolving Low-Frequency Quasi-periodic Oscillation (LFQPO) frequencies in hard X-rays upto $100$ keV with $AstroSat/LAXPC$ during `unusual' outburst phase of Swift J1727.8-1613 in hard-intermediate state (HIMS). The observed LFQPO in $20 - 100$ keV has a centroid $\nu_{_{\rm QPO}}=1.43$ Hz, a coherence factor $Q= 7.14$ and an amplitude ${\rm rms_{_{\rm QPO}}} = 10.95\%$ with significance $\sigma = 5.46$. Type-C QPOs ($1.09-2.6$ Hz) are found to evolve monotonically during HIMS of the outburst with clear detection in hard X-rays ($80 - 100$ keV), where ${\rm rms_{_{\rm QPO}}}$ decreases ($\sim 12-3\%$) with energy. Further, $\nu_{_{\rm QPO}}$ is seen to correlate (anti-correlate) with low (high) energy flux in $2-20$ keV ($15-50$ keV). Wide-band ($0.7 - 40$ keV) energy spectrum of $NICER/XTI$ and $AstroSat/LAXPC$ is satisfactorily described by the `dominant' thermal Comptonization contribution ($\sim 88\%$) in presence of a `weak' signature of disk emissions ($kT_{\rm in} \sim 0.36$ keV) indicating the harder spectral distribution. Considering source mass $M_{\rm BH}=10M_\odot$ and distance $1.5 < {\rm d~(kpc)} < 5$, the unabsorbed bolometric luminosity is estimated as $\sim 0.03-0.92\%L_{\rm Edd}$. Finally, we discuss the implications of our findings in the context of accretion dynamics around black hole X-ray binaries., Comment: 10 pages, 5 figures, 3 tables, accepted for publication in MNRAS

The legacy beam management (BM) procedure in 5G introduces higher measurement and reporting overheads for larger beam codebooks resulting in higher power consumption of user equipment (UEs). Hence, the 3rd generation partnership project (3GPP) studied the use of artificial intelligence (AI) and machine learning (ML) in the air interface to reduce the overhead associated with the legacy BM procedure. The usage of AI/ML in BM is mainly discussed with regard to spatial-domain beam prediction (SBP) and time-domain beam prediction (TBP). In this study, we discuss different sub-use cases of SBP and TBP and evaluate the beam prediction accuracy of AI/ML models designed for each sub-use case along with AI/ML model generalization aspects. Moreover, a comprehensive system-level performance evaluation is presented in terms of user throughput with integrated AI/ML models to a 3GPP-compliant system-level simulator. Based on user throughput evaluations, we present AI/ML BM design guidelines for the deployment of lightweight, low-complexity AI/ML models discussed in this study., Comment: This work has been submitted to the IEEE for possible publication

The far-infrared (IR) region is rich with information needed to characterize interstellar dust and to investigate the cold outer planets of the solar system and their icy moons. The proposed sub-orbital observatory the Balloon Experiment for Galactic INfrared Science (BEGINS) will utilize cryogenic instruments to map spectral energy distributions (SEDs) of interstellar dust in the Cygnus molecular cloud complex. A future high priority flagship mission Uranus Orbiter and Probe carrying a net flux radiometer (NFR) will study the in situ heat flux of the icy giants atmosphere to 10 bar pressure. These instruments require far-IR filters to define the instrument spectral bandwidths. Our ultimate goal is to define the instrument bands of BEGINS and the NFR with linear-variable filters (LVFs) and discrete-variable filters (DVFs). The LVFs and DVFs will be made of metal mesh band-pass filters (MMBF) comprised of a 100 nm thick gold film with cross-shaped slots of varying sizes along a silicon (Si) substrate with cyclic olefin copolymer (COC) anti-reflection (AR) coatings. We present our progress towards LVFs and DVFs with simulated and measured transmission of a room temperature, non-AR coated, single-band 44 $\mu$m MMBF filter. We have successfully fabricated, measured, and modeled a non-AR coated, room temperature 44 $\mu$m MMBF. The transmission at room temperature and non-AR coated was measured to be 27\% with a resolving power of 11. When COC-AR coated on both sides the transmission is expected to increase to 69\% with a resolving power of 10.

A microscopic control over the origin and dynamics of localised spin centres in lower dimensional solids turns out to be a key factor for next generation spintronics and quantum technologies. With the help of low temperature electron paramagnetic resonance (EPR) measurements, supported by the first-principles calculations within density functional theory (DFT) formulation, we found the origin of different high-spin paramagnetic intrinsic charge-centres, Mo3+(4d3) and Mo2+(4d4) present in the nano-crystalline sulfur deficit hexagonal molybdenum disulfide (2H-MoS_(2-x)), against the established notion of spin-1/2 , Mo5+ centres. A critical strain generated in the nano-structured 2H-MoS_(2-x) was found to be very crucial for spin-localization in this layered material. Indeed, computationally effective proposition of the PBE+U exchange-correlations within DFT including D3-dispersion corrections found to be more viable than expensive higher rung of exchange-correlation functionals, explored earlier. It is also found that the oxygen vacancy of the reduced oxide phase, embedded in 2H-MoS_(2-x) host lattice, has the longest relaxation times. Moreover, the temperature dependence of spin-lattice relaxation measurements reveals a direct process for interstitial spin centres and a Raman process for both sulfur and oxygen vacancy sites. We expect such observation would be a valuable pillar for better understanding of the next generation quantum technologies and device applications., Comment: This manuscript has 38 pages in single column and 7 figures. Supplementary information is not attached

The proximity-effect, whereby materials in contact appropriate each others electronic-properties, is widely used to induce correlated states, such as superconductivity or magnetism, at heterostructure interfaces. Thus far however, demonstrating the existence of proximity-induced charge-density-waves (PI-CDW) proved challenging. This is due to competing effects, such as screening or co-tunneling into the parent material, that obscured its presence. Here we report the observation of a PI-CDW in a graphene layer contacted by a 1T-TaS2 substrate. Using scanning tunneling microscopy (STM) and spectroscopy (STS) together with theoretical-modeling, we show that the coexistence of a CDW with a Mott gap in 1T-TaS2 coupled with the Dirac-dispersion of electrons in graphene, makes it possible to unambiguously demonstrate the PI-CDW by ruling out alternative interpretations. Furthermore, we find that the PI-CDW is accompanied by a reduction of the Mott gap in 1T-TaS2 and show that the mechanism underlying the PI-CDW is well-described by short-range exchange-interactions that are distinctly different from previously observed proximity effects., Comment: 35 pages, 16 figures. arXiv admin note: text overlap with arXiv:2201.09195

In the recent past, binary similarity measures have been applied in solving biometric identification problems, including fingerprint, handwritten character detection, and in iris image recognition. The application of the relevant measurements has also resulted in more accurate data analysis. This paper surveys the applicability of binary similarity and distance measures in various fields.

We present the results on broadband X-ray properties of persistent black hole binaries GRS 1758$-$258 and 1E 1740.7$-$2942 using AstroSat, NuSTAR and Swift-XRT observations carried out during 2016$-$2022. We perform spectral modeling of both sources after eliminating the contamination in their \textit{LAXPC} spectra from nearby X-ray sources. Preliminary spectral modelling using Comptonization and line emission ($\sim$ 6.4 keV) models suggest that GRS 1758$-$258 occupies both dim-soft state ($kT_{bb}=0.37\pm0.01$ keV, $\Gamma\sim5.9$, $L_{bol}=1 %$ of Eddington luminosity L$_{Edd}$) and hard state ($\Gamma=1.64-2.22$, $kT_{e}$=4$-$45 keV, $L_{bol}$=1$-$5 % L$_{Edd}$) that requires a multi-colour disc blackbody model ($kT_{in}=0.54\pm0.01$ keV) occasionally. 1E 1740.7$-$2942 instead is found only in hard state ($\Gamma$=1.67$-$2.32, $kT_{e}$=5$-$16 keV, $L_{bol}$=1$-$2 % L$_{Edd}$). Reflection properties of both sources are studied by applying relativistic reflection model RELXILL to the broadband spectra. Our results from \textit{AstroSat} and \textit{NuSTAR} consistently unveiled the presence of a Comptonizing region along with an ionized reflection region (ionization parameter $log\xi$=2.7$-$3.8 and 2.7$-$4.7 erg cm s$^{-1}$ in GRS 1758$-$258 and 1E 1740.7$-$2942 respectively) in both sources. Reflection modeling revealed GRS 1758$-$258 to have a high metal abundance ($A_{fe}=3.9^{+0.4}_{-0.3}$ times solar metal abundance) and inclination angle ($i$) of $61\pm2^{\circ}$. In case of 1E 1740.7$-$2942, $i$ is constrained to be $55\pm1^{\circ}$. Finally, we discuss the implication of our findings in the context of accretion dynamics by comparing our results with the previous studies., Comment: Accepted for publication in MNRAS

We have carried out the first spectro-polarimetric study of the bright NS-LMXB GX 9+9 using IXPE and AstroSat observations. We report a significant detection of polarization of $1.7\pm 0.4\%$ over the $2-8$ keV energy band, with a polarization angle of $63^{\circ}\pm 7^{\circ}$. The polarization is found to be energy-dependent, with a $3\sigma$ polarization degree consistent with null polarization in $2-4$ keV, and $3.2\%$ in $4-8$ keV. Typical of the spectra seen in NS-LMXBs, we find that a combination of soft thermal emission from the accretion disc and Comptonized component from the optically thick corona produces a good fit to the spectra. We also attempt to infer the individual polarization of these components, and obtain a $3\sigma$ upper limit of $\sim 11\%$ on the polarization degree of the thermal component, and constrain that of the Comptonized component to $\sim 3\%$. We comment on the possible corona geometry of the system based on our results., Comment: 5 pages, 6 figures, Accepted for publication in MNRAS Letters

Artificially twisted heterostructures of semiconducting transition metal dichalcogenides (TMDs) offer unprecedented control over their electronic and optical properties via the spatial modulation of interlayer interactions and structural reconstruction. Here we study twisted MoS2 bilayers in a wide range of twist angles near 0{\deg} using Scanning Tunneling Microscopy/Spectroscopy. We investigate the twist angle-dependence of the moir\'e pattern which is dominated by lattice reconstruction for small angles (<2{\deg}) leading to large triangular domains with rhombohedral stacking. Local spectroscopy measurements reveal a large moir\'e-potential strength of 100-200 meV for angles <3{\deg}. In reconstructed regions we see a bias-dependent asymmetry between neighboring triangular domains which we relate to the vertical polarization which is intrinsic to rhombohedral stacked TMDs. This viewpoint is urther supported by spectroscopy maps and ambient Piezoresponse measurements. Our results provide a microscopic perspective on this new class of interfacial ferroelectrics and can offer clues for designing novel heterostructures which harness this effect., Comment: 34 pages and 10 figures

In this paper, we have carried out a detailed study of the `Clocked' burster GS $1826-238$ using $\sim$ 90 ks broad-band (0.7 - 60.0 keV) data obtained with {\it AstroSat} observatory. The source was observed during a soft spectral state and traced a `banana' type track in the colour-colour diagram (CCD). We find that a combination of thermal component (multi-colour disc/bbodyrad) and Comptonized component is statistically good description for all the sections of the track in the CCD. The corona becomes optically thick ($\tau$ increases from $\sim$ 5 to 21) and cooler ($kT_e$ decreases from $\sim$ 4.8 to 2.2 keV) as the source moves up in the `banana' branch. Probably cooling is caused by increase in the supply of soft-seed photons from the disc/boundary-layer. Reflection signature is observed at upper `banana' branch of the source. Two type-I X-ray bursts are detected during the {\it AstroSat} observations. During the bursts, hard X-rays increased unlike previous observations where a reduction in hard X-rays is observed during the bursts. Decrease in the electron temperature and increase in the optical depth are observed during the bursts. The PSD (Power Spectral Density) of all the sections of the CCD can be represented by a pure power-law component. The strength of this component increases from $\sim$ 1\% to 4.5\% as the source moves up in the `banana' track. Search for burst oscillations gave a null result. We discuss the implications of our results in the context of previous findings., Comment: accepted for publication in MNRAS, 14 pages, 13 figures, 6 tables