Parallel Propagation Research Articles

16-channel photonic solver for optimization problems on a silicon chip

A programmable photonic solver for quadratic unconstrained binary optimization (QUBO) problems is demonstrated with a hybrid optoelectronic scheme, which consists of a photonic chip and an electronic driving board. The photonic chip is employed to perform the optical vector-matrix multiplication (OVMM) to calculate the cost function of the QUBO problem, while the electronic processor runs the heuristic algorithm to search for the optimal solution. Due to the parallel and low-latency propagation of lightwave, the calculation of the cost function can be accelerated. The photonic chip is fabricated on the silicon on insulator (SOI) substrate and integrates 16 high-speed electro-optic modulators, 88 thermo-optic phase shifters, and 16 balanced photodetectors. The computing speed of the photonic chip is 1.66 TFLOP/s. As a proof of principle, two randomly generated 16-dimensional QUBO problems are solved with high successful probabilities. These results present the potential of fast solving optimization problems with integrated photonic systems.

Closing the gap between SGP4 and high-precision propagation via differentiable programming

The simplified general perturbations 4 (SGP4) orbital propagation model is one of the most widely used methods for rapidly and reliably predicting the positions and velocities of objects orbiting Earth. Over time, SGP models have undergone refinement to enhance their efficiency and accuracy. Nevertheless, they still do not match the precision offered by high-precision numerical propagators, which can predict the positions and velocities of space objects in low-Earth orbit with significantly smaller errors.In this study, we introduce a novel differentiable version of SGP4, named ∂SGP4. By porting the source code of SGP4 into a differentiable program based on PyTorch, we unlock a whole new class of techniques enabled by differentiable orbit propagation, including spacecraft orbit determination, state conversion, covariance similarity transformation, state transition matrix computation, and covariance propagation. Besides differentiability, our ∂SGP4 supports parallel propagation of a batch of two-line elements (TLEs) in a single execution and it can harness modern hardware accelerators like GPUs or XLA devices (e.g. TPUs) thanks to running on the PyTorch backend.Furthermore, the design of ∂SGP4 makes it possible to use it as a differentiable component in larger machine learning (ML) pipelines, where the propagator can be an element of a larger neural network that is trained or fine-tuned with data. Consequently, we propose a novel orbital propagation paradigm, ML-∂SGP4. In this paradigm, the orbital propagator is enhanced with neural networks attached to its input and output. Through gradient-based optimization, the parameters of this combined model can be iteratively refined to achieve precision surpassing that of SGP4. Fundamentally, the neural networks function as identity operators when the propagator adheres to its default behavior as defined by SGP4. However, owing to the differentiability ingrained within ∂SGP4, the model can be fine-tuned with ephemeris data to learn corrections to both inputs and outputs of SGP4. This augmentation enhances precision while maintaining the same computational speed of ∂SGP4 at inference time. This paradigm empowers satellite operators and researchers, equipping them with the ability to train the model using their specific ephemeris or high-precision numerical propagation data.

The effect of dust on the propagation of a hydromagnetic solitary wave along the magnetic field in a cold collision-free plasma

The effect of dust is studied on the parallel propagation of a solitary hydromagnetic wave in a magnetized, cold, collision-free plasma. The fully nonlinear analysis is based on the Adlam–Allen method. The system of nonlinear differential equations was derived using the assumption of massive dust, namely, that the dust velocity is constant and parallel to the background magnetic field. Self-consistent solutions obtained from the model for the lighter particles dynamics and for the resulting electric and magnetic fields are presented, together with the validity range of Mach numbers as a function of dust parameters. In particular, supersonic solutions are possible in scenarios tending to two-component plasmas (positively charged ions and negatively charged dust). The stationary solitary pulse in terms of particle densities, magnitude of transverse magnetic field, transverse electron–ion velocities, and bipolar electric field is presented. The magnitude of these amplitudes is maximum at the center of wave and minimum at the boundary. Furthermore, it is shown that all the solitary wave amplitudes are decreases with dust parameter.

MLatom Software Ecosystem for Surface Hopping Dynamics in Python with Quantum Mechanical and Machine Learning Methods.

We present an open-source MLatom@XACS software ecosystem for on-the-fly surface hopping nonadiabatic dynamics based on the Landau-Zener-Belyaev-Lebedev algorithm. The dynamics can be performed via Python API with a wide range of quantum mechanical (QM) and machine learning (ML) methods, including ab initio QM (CASSCF and ADC(2)), semiempirical QM methods (e.g., AM1, PM3, OMx, and ODMx), and many types of ML potentials (e.g., KREG, ANI, and MACE). Combinations of QM and ML methods can also be used. While the user can build their own combinations, we provide AIQM1, which is based on Δ-learning and can be used out-of-the-box. We showcase how AIQM1 reproduces the isomerization quantum yield of trans-azobenzene at a low cost. We provide example scripts that, in dozens of lines, enable the user to obtain the final population plots by simply providing the initial geometry of a molecule. Thus, those scripts perform geometry optimization, normal mode calculations, initial condition sampling, parallel trajectories propagation, population analysis, and final result plotting. Given the capabilities of MLatom to be used for training different ML models, this ecosystem can be seamlessly integrated into the protocols building ML models for nonadiabatic dynamics. In the future, a deeper and more efficient integration of MLatom with Newton-X will enable a vast range of functionalities for surface hopping dynamics, such as fewest-switches surface hopping, to facilitate similar workflows via the Python API.

Duality of a coiled phononic crystal enables reflectionless interfaces

Recently, it has been demonstrated that one-dimensional bar-based phononic crystals can exhibit subwavelength Bragg bandgaps by coiling the bars and locking the nodal rotational degrees of freedom to create what is termed a coiled phononic crystal (CPnC) [C. L. Willey et al., Phys. Rev. Appl. 18, 014035 (2022)]. Here, it is shown that the CPnC exhibits duality of its dispersion curves relative to its coiling/twist angle, meaning that the dispersion curves are symmetric about a particular coiling/twist angle defined configuration. An exciting implication of this finding is that under a certain set of constraints, segments of dual unit cells with perpendicular wave propagation directions can be connected such that their wave transmission is equivalent to a finite CPnC entirely composed of identical unit cells with parallel wave propagation directions. The ability to link unit cells with different wave propagation orientations, but the same dispersion/dynamic stiffness, is used to create an elastic hierarchically coiled phononic crystal based on a fractal space-filling curve design. The novelty of this work is that it numerically demonstrates reflectionless wave propagation in large fractal architectures (created from specific combinations of dual unit cells) such that regular phononic properties (i.e., passbands and bandgaps) are preserved, allowing for propagation of broadband signals and filtering.

Radiative loss and ion-neutral collisional effects in astrophysical plasmas.

In this paper, we study the role of radiative cooling (RC) in a two-fluid model consisting of coupled neutrals and charged particles. We first analyse the linearized two-fluid equations where we include radiative losses in the energy equation for the charged particles. In a one-dimensional geometry for parallel propagation and in the limiting cases of weak and strong coupling, it can be shown analytically that the instability conditions for the thermal mode and the sound waves, the isobaric and isentropic criteria, respectively, remain unchanged with respect to one-fluid radiative plasmas. For the parameters considered in this paper, representative for the solar corona, the RC produces growth of the thermal mode and damping of the sound waves. In the weak coupling limit, the growth of the thermal instability and the damping of the sound waves is as derived in Field (Field 1965 Astrophys. J. 142, 531(doi:10.1086/148317)) using the charged fluid properties. When neutrals are included and are sufficiently coupled to the charges, the thermal mode growth rate and the wave damping both reduce by the same factor, which depends on the ionization fraction only. For a heating function that is constant in time, we find that the growth of the thermal mode and the damping of the sound waves are slightly larger. The numerical calculation of the eigenvalues of the general system of equations in a three-dimensional geometry confirm the analytic results. We then run two-dimensional fully nonlinear simulations that give consistent results: a higher ionization fraction or lower coupling will increase the growth rate. The magnetic field contribution is negligible in the linear phase. Ionization-recombination effects might play an important role because the RC produces a large range of temperatures in the system. In the numerical simulation, after the first condensation phase, when the minimum temperature is reached, the fraction of neutrals increases four orders of magnitude because of the recombination. This article is part of the theme issue 'Partially ionized plasma of the solar atmosphere: recent advances and future pathways'.

On the Wave‐Normal Distribution of Lightning‐Generated Whistlers and Their Propagation Modes

AbstractObservations from Van Allen Probes are analyzed to obtain the statistical wave normal distribution of lightning‐generated whistlers (LGWs). An automatic algorithm is developed to identify burst mode waveform data with LGW signals and analyze the wave polarization information for these LGW signals. The spatial distribution of the LGW occurrence and the probabilities of different propagation types demonstrate that most LGWs can be observed in the low L‐shell region (L < 3), where the dominant propagation type is oblique outward. Parallel propagation dominates in the high L‐shell region, but the LGW occurrence is very small. Additionally, a small group of oblique but inwardly propagating LGWs are observed at low altitudes (<0.2RE). A ray tracing simulation is performed not only to confirm predominant oblique and outward wave vectors in the vast region of the plasmasphere but also to verify the existence of these inward propagating LGW signals at low altitudes near the topside ionosphere.

MVP-Stereo: A Parallel Multi-View Patchmatch Stereo Method with Dilation Matching for Photogrammetric Application

Multi-view stereo plays an important role in 3D reconstruction but suffers from low reconstruction efficiency and has difficulties reconstructing areas with low or repeated textures. To address this, we propose MVP-Stereo, a novel multi-view parallel patchmatch stereo method. MVP-Stereo employs two key techniques. First, MVP-Stereo utilizes multi-view dilated ZNCC to handle low texture and repeated texture by dynamically adjusting the matching window size based on image variance and using a portion of pixels to calculate matching costs without increasing computational complexity. Second, MVP-Stereo leverages multi-scale parallel patchmatch to reconstruct the depth map for each image in a highly efficient manner, which is implemented by CUDA with random initialization, multi-scale parallel spatial propagation, random refinement, and the coarse-to-fine strategy. Experiments on the Strecha dataset, the ETH3D benchmark, and the UAV dataset demonstrate that MVP-Stereo can achieve competitive reconstruction quality compared to state-of-the-art methods with the highest reconstruction efficiency. For example, MVP-Stereo outperforms COLMAP in reconstruction quality by around 30% of reconstruction time, and achieves around 90% of the quality of ACMMP and SD-MVS in only around 20% of the time. In summary, MVP-Stereo can efficiently reconstruct high-quality point clouds and meet the requirements of several photogrammetric applications, such as emergency relief, infrastructure inspection, and environmental monitoring.

Linear Theory Analysis and One‐Dimensional Hybrid Simulations of High‐Frequency EMIC Waves in a Dipole Magnetic Field

AbstractNarrowband (Δf ≲ 0.1fcp), high‐frequency (0.9fcp ≲ f < fcp) electromagnetic ion cyclotron (EMIC) waves, or HFEMIC waves for short, are a relatively new type of EMIC waves, where fcp is the proton cyclotron frequency. Here, we investigate the instability threshold conditions and the nonlinear evolution of HFEMIC waves at parallel propagation. First, linear theory analysis is extended to a regime relevant to HFEMIC waves (parallel proton beta β‖p < 0.01). The instability threshold follows a similar anisotropy‐β‖p relation and requires a large value of anisotropy (T⊥p/T‖p ≳ 10) for wave growth. As a result of decreasing group velocity, the convective growth rate at a fixed threshold exhibits a similar inverse relation with β‖p. Heavy ions affect the instability only weakly, primarily through the introduction of stop bands. Second, we carry out one‐dimensional hybrid simulations in a parabolic background magnetic field, with initial parameters constrained by observation. Despite the narrow source region (within about ±3° latitude), HFEMIC waves can grow well above the thermal noise level due in large part to a small group velocity of HFEMIC waves. The saturation level is well within the range of observational amplitudes, and the quasilinear process primarily determines the wave evolution. Finally, we demonstrate that the present results compare favorably to the recent statistical results, thereby supporting anisotropic low‐energy protons as free energy source for HFEMIC waves.

Kinetic closures for unmagnetized and magnetized plasmas

Parallel and perpendicular closures with cyclotron resonance effects retained for the five-moment (density, temperature, and flow velocity) fluid equations are derived by solving the kinetic equation with the Bhatnagar–Gross–Krook operator in Fourier space. For parallel propagation, the parallel closures are reduced to those of Ji et al. [Phys. Plasmas 20, 082121 (2013)]. The closures when combined to the fluid equations reproduce the fully kinetic dispersion relation that can be directly derived from the kinetic equation. The closures for the five-moment fluid system can be utilized to derive closures for the extended fluid system, which is demonstrated by deriving closures for the ten-moment system consisting of density, flow velocity, temperature, and viscosity tensor equations.

Two step model for crosslinking polymerization with account of diffusion

The proposed model includes two parallel reactions, initiation and propagation that respectively follow the reaction order and autocatalytic kinetics. The effect of diffusion is accounted for in both steps by introducing a term that depends linearly on conversion. The term is introduced into the exponential part of the rate constants. Determining the magnitude of this term allows for estimating the diffusion effect on the overall activation energy in the final stages of crosslinking polymerization. The model is tested on five previously published datasets. It appears to fit all data very well and yield physically meaningful fit parameters. Algorithms for successful fitting of the model are suggested. It is also demonstrated that disregarding the effect of diffusion gives rise to markedly inferior results. The model is quite general and expected to be applicable to any polymerization complicated by diffusion or other processes that involve the interplay between the reaction and diffusion kinetics.

Parallel window propagation on mesh surfaces

The task of calculating the shortest distance with minimal curvature between two points on a mesh surface is a well-known problem in computational geometry. This paper aims to develop a effective algorithm for computing geodesics on large, complex models with fast processing times to facilitate interactive applications. Our contribution is developing of the parallel window propagation (PWP) algorithm, which divides the sequential MMP algorithm into four phases: front node selection, window list propagation, window list merging, and vertex update. We use two separate buffers for incoming windows on each edge to avoid data dependency and conflicts in each phase, allowing for parallel processing on a CPU. As a result, our PWP algorithm can propagate a large number of windows simultaneously and independently, leading to significant improvements in performance for real-world models.

Spark-Based Label Diffusion and Label Selection Community Detection Algorithm for Metagenome Sequence Clustering

It is a challenge to assemble an enormous amount of metagenome data in metagenomics. Usually, metagenome cluster sequence before assembly accelerates the whole process. In SpaRC, sequences are defined as nodes and clustered by a parallel label propagation algorithm (LPA). To address the randomness of label selection from the parallel LPA during clustering and improve the completeness of metagenome sequence clustering, Spark-based parallel label diffusion and label selection community detection algorithm is proposed in the paper to obtain more accurate clustering results. In this paper, the importance of sequence is defined based on the Jaccard similarity coefficient and its degree. The core sequence is defined as the one with the largest importance in its located community. Three strategies are formulated to reduce the randomness of label selection. Firstly, the core sequence label diffuses over its located cluster and becomes the initial label of other sequences. Those sequences that do not receive an initial label will select the sequence label with the highest importance in the neighbor sequences. Secondly, we perform improved label propagation in order of label frequency and sequence importance to reduce the randomness of label selection. Finally, a merge small communities step is added to increase the completeness of clustered clusters. The experimental results show that our proposed algorithm can effectively reduce the randomness of label selection, improve the purity, completeness, and F-Measure and reduce the runtime of metagenome sequence clustering.

An analysis of expert-opined strategies in multi-attribute decision-making using effort propagation

Mining the interrelationships and possible hierarchical structures among the factors in real-world multi-attribute decision-making (MADM) problems is considered one of the primary tasks. One major task is determining an optimal strategy to work on the factors to enhance the effect on the goal attribute. This paper proposes two such strategies: parallel and hierarchical effort assignment and propagation strategies. We formally define and describe the concept of effort propagation through a strategy. Both the parallel and hierarchical strategies are divided into sub-strategies based on whether the assignment of efforts to the factors is uniform or depends upon some appropriate heuristics related to the factors in the system. The adapted and discussed heuristics are the factors’ relative significance and effort propagability. The strategies are analyzed for a real-life case study regarding high school administrative factors that enhance students’ performance. Total effort propagation of around 7%–15% to the goal is seen across the proposed strategies, given a total of 1 unit of effort to the directly accessible factors of the system. A comparative analysis is adapted to determine the optimal strategy among the proposed ones to enhance student performance effectively. The highest effort propagation achieved in the work is approximately 14.4348%. The analysis establishes the necessity of research towards the direction of effort propagation analysis in case of decision-making problems.

RFDANet: an FMCW and TOF radar fusion approach for driver activity recognition using multi-level attention based CNN and LSTM network

Dangerous driving behavior is a major contributing factor to road traffic accidents. Identifying and intervening in drivers’ unsafe driving behaviors is thus crucial for preventing accidents and ensuring road safety. However, many of the existing methods for monitoring drivers’ behaviors rely on computer vision technology, which has the potential to invade privacy. This paper proposes a radar-based deep learning method to analyze driver behavior. The method utilizes FMCW radar along with TOF radar to identify five types of driving behavior: normal driving, head up, head twisting, picking up the phone, and dancing to music. The proposed model, called RFDANet, includes two parallel forward propagation channels that are relatively independent of each other. The range-Doppler information from the FMCW radar and the position information from the TOF radar are used as inputs. After feature extraction by CNN, an attention mechanism is introduced into the deep architecture of the branch layer to adjust the weight of different branches. To further recognize driving behavior, LSTM is used. The effectiveness of the proposed method is verified by actual driving data. The results indicate that the average accuracy of each of the five types of driving behavior is 94.5%, which shows the advantage of using the proposed deep learning method. Overall, the experimental results confirm that the proposed method is highly effective for detecting drivers’ behavior.

Instabilities Driven by Proton Temperature Anisotropy in the Presence of Alpha Particles: Implications for Proton-temperature-anisotropy Constraint in the Solar Wind

In situ measurements reveal that proton temperature anisotropy is ubiquitous in the solar wind. Various plasma instabilities have been proposed to regulate the distribution of the proton temperature anisotropy in the solar wind; detailed constraint processes are still unclear. In this paper, we study the effects of alpha beams on both the forward and backward proton temperature anisotropy instabilities at parallel and oblique propagation with the Vlasov theory, and compare the theoretical results with the Wind observation. As the alpha-beam drift velocity v α /v A increases, the growth rates of forward Alfvén/ion-cyclotron (FA/IC) and backward magnetosonic/whistler (BM/W) instabilities increase, those of backward Alfvén/ion-cyclotron (BA/IC) and forward magnetosonic/whistler (FM/W) instabilities decrease, and those of the mirror and forward Alfvén wave (FAW) instabilities are nearly constant. In particular, there are different constraining mechanisms on the distribution of proton temperature anisotropy for different values of the alpha-beam drift velocity. The proton temperature anisotropy instability together with the alpha beam can provide a potential explanation for the distribution of the proton temperature anisotropy in the solar wind.

Wave modes and instabilities in gravitating magnetized polytropic quantum plasmas including viscosity tensor and FLR corrections

ABSTRACT The present analytical study extends the problems of pressure anisotropy-driven instabilities and gravitational instability in space plasmas to mixed quantum polytropic gas in the interior of dense stars accounting for the effects of viscosity, finite Larmor radius (FLR) and self-gravitational effects. The generalized polytrope pressure laws are considered as adiabatic equations in which the pressure components depend upon the plasma density, magnitude of the magnetic field, and the polytrope indices. The modified properties of waves and instabilities in gravitating quantum plasmas have been analysed using the quantum magnetohydrodynamic (QMHD) fluid description in the magnetohydrodynamic (MHD) and Chew–Goldberger–Low (CGL) limits. In the parallel propagation, the Jeans instability modified by quantum diffraction parameters and firehose mode modified by FLR parameter is obtained separately. The Jeans instability condition depends upon the quantum diffraction term and polytrope index β, and it remains unaffected due to viscosity and ion Larmor frequency. The growth rate of the Jeans instability decreases due to viscosity and quantum diffraction parameters, while the growth rate of the firehose instability increases due to FLR corrections. In the transverse mode, a similar nature is observed in the growth rates; however, the instability region decreases significantly due to polytrope indices and different dispersion properties of MHD and CGL viscous quantum plasmas. The analytical results have been applied in dense degenerate stars to measure the characteristic parameters and understand the MHD wave propagation, pressure anisotropy-driven, and gravitationally driven instabilities.

Temporal characteristics of ELMs on the COMPASS divertor

The presented work shows a systematic study of the temporal characteristics of ELM events on the COMPASS divertor obtained with high temporal resolution probe measurements (∼1 μs). The resulting temporal evolution of the total ELM power on the outer target provides the values of rise (τ rise) and decay (τ decay) times for each single ELM event. It has been found that τ rise values are in the range of about 50 μs–100 μs. These values are comparable to the time of the ELM parallel propagation (τ ||) given by the sound speed and the connection length between the outer midplane and the outboard divertor. This comparison indicates that the magnetic field lines in the SOL region are not significantly ergodized during the pedestal crash on COMPASS. It also implies that the peak ELM energy fluence on the outboard divertor is dominated by the ELM parallel transport, which is confirmed by a good agreement with model prediction. In addition, the values of the ratio of τ decay and τ rise for each ELM event fit very well to the boundaries 1.5 < τ decay/τ rise < 4, as already shown on JET as well as on the HL-2A tokamak, using IR measurements. The ratio does not show any clear dependence on the relative ELM energy or line averaged electron density. It was also found that the ELM energy fluence decay length (λϵ mid) is clearly linked to this ratio.

Coupled KIPP-EDGE2D modeling of parallel transport in the SOL and divertor plasma for the ITER baseline scenario

The KInetic code for Plasma Periphery (KIPP) models the parallel (along magnetic field lines) propagation of charged particles in the scrape-off layer (SOL) and divertor of tokamaks. An iterative coupling between KIPP and a 2D edge fluid code, EDGE2D, which in turn is coupled to the Monte Carlo solver for neutrals, EIRENE, was used to achieve a converged KIPP-EDGE2D-EIRENE solution. In the iterative coupling algorithm, KIPP transfers kinetic parallel electron and ion heat conductivities to EDGE2D, whereas EDGE2D returns to KIPP 2D distributions of macroscopic plasma parameters across the computations grid. The initial EDGE2D-EIRENE solution simulated the SOL and divertor plasma of the ITER inter-edge localized mode (ELM) baseline scenario. This work employs the same methodology as an earlier study based on the analysis of JET high radiative H-mode conditions, with strong nitrogen injection leading to partial detachment at divertor targets (Chankin et al 2022 Plasma Phys. Control. Fusion 64 095007). The results are qualitatively similar to those of the JET study in demonstrating the strong heat flux limiting effect, together with the rise in electron and ion temperatures in the main SOL. At the same time, the kinetic effects of the parallel propagation of charged particles are not expected to drastically change the target power deposition at divertor targets calculated by EDGE2D-EIRENE alone.

Non-local affinity adaptive acceleration propagation network for generating dense depth maps from LiDAR.

Depth completion aims to generate dense depth maps from the sparse depth images generated by LiDAR. In this paper, we propose a non-local affinity adaptive accelerated (NL-3A) propagation network for depth completion to solve the mixing depth problem of different objects on the depth boundary. In the network, we design the NL-3A prediction layer to predict the initial dense depth maps and their reliability, non-local neighbors and affinities of each pixel, and learnable normalization factors. Compared with the traditional fixed-neighbor affinity refinement scheme, the non-local neighbors predicted by the network can overcome the propagation error problem of mixed depth objects. Subsequently, we combine the learnable normalized propagation of non-local neighbor affinity with pixel depth reliability in the NL-3A propagation layer, so that it can adaptively adjust the propagation weight of each neighbor during the propagation process, which enhances the robustness of the network. Finally, we design an accelerated propagation model. This model enables parallel propagation of all neighbor affinities and improves the efficiency of refining dense depth maps. Experiments on KITTI depth completion and NYU Depth V2 datasets show that our network is superior to most algorithms in terms of accuracy and efficiency of depth completion. In particular, we predict and reconstruct more smoothly and consistently at the pixel edges of different objects.