Almost instant time inference for hybrid partially dynamic Bayesian networks

A Bayesian network (BN) is a compact representation for probabilistic models and inference. They have been used successfully for many military and civilian applications. It is well known that, in general, the inference algorithms to compute the exact a posterior probability of a target node given observed evidence are either computationally infeasible for dense networks or impossible for general hybrid networks. In those cases, one either computes the approximate results using stochastic simulation methods or approximates the model using discretization or a Gaussian mixture model before applying an exact inference algorithm. This paper combines the concept of simulation and model approximation to propose an efficient algorithm for those cases. The main contribution here is a unified treatment of arbitrary (nonlinear non-Gaussian) hybrid (discrete and continuous) BN inference having both computation and accuracy scalability. The key idea is to precompile the high-dimensional hybrid distribution using a hypercube representation and apply it for both static and dynamic BN inference. Since the inference process essentially becomes a combination of table look-up and some simple operations, the method is shown to be extremely efficient. It can also he scaled to achieve any desirable accuracy given sufficient preprocessing time and memory for the cases where exact inference is not possible     

The Time-of-Flight Energy,Angle, Mass Spectrograph (Teams) Experiment for Fast

The Time-of-flight Energy Angle Mass Spectrograph (TEAMS) is being flown on the FAST Small Explorer mission to measure the 3-dimensional distribution function of the major ion species present in the lower magnetosphere. The instrument is similar to time-of-flight plasma analyzer systems that have been designed and planned for flight as CODIF (COmposition and DIstribution Function analyzer) on the four European Space Agency Cluster-II spacecraft and, as ESIC (Equator-S Ion Composition instrument) on Equator-S. This instrument allows the 3-dimensional distribution functions of individual ion species to be determined within spin period (2.5 s). Two-dimensional distributions are measured in 80 ms. These capabilities are crucial for the study of selective energization processes in the auroral regions of the magnetosphere. The design, operational characteristics, and test and calibration results for this instrument are presented. The sensor consists of a toroidal top-hat electrostatic analyzer with instantaneous acceptance of ions over 360° in polar angle. After post-acceleration of the incoming ions by up to 25 kV, a time-of-flight mass spectrograph discriminates the individual species. It has been demonstrated through calibration that the instrument can easily separate H⁺, He²⁺, He⁺, O⁺ and, for energies after post-acceleration of > 20 keV, even O₂ ⁺ molecules. On-board mass discrimination and the internal accumulation of several distinct data quantities combined with the spacecraft's flexible telemetry formatting allow for instrument data rates from 7.8 kb s^–1 to 315 kb s^–1 to be telemetered to ground through the FAST centralized Instrument Data Processor.     

Stimulated electromagnetic emissions by high-frequency electromagnetic pumping of the ionospheric plasma

A high frequency electromagnetic pump wave transmitted into the ionospheric plasma from the ground can stimulate electromagnetic radiation with frequencies around that of the ionospherically reflected pump wave. The numerous spectral features of these stimulated electromagnetic emissions (SEE) and their temporal evolution on a wide range of time scales are reviewed and related theoretical, numerical, and simulation results are discussed. On long (thermal) time scales the SEE constitutes a self-organization of the ionospheric plasma which depends on the interaction of nonlinear processes in a hierarchy of time scales in response to the electromagnetic pumping. Particularly, the appearance of the rich SEE spectrum is associated with the slow self-structuring of the plasma density into a spectrum of magnetic field-aligned density striations. The dependence of the SEE on electron gyroharmonic effects and the presence of density striations suggests that the existence of a magnetic field in the plasma is important for plasma turbulence to dissipate into non-thermal electromagnetic radiation during the long time quasi-stationary state of the turbulence evolution.     

Ring Currents and Internal Plasma Sources

The discovery of terrestrial O⁺ and other heavy ions in magnetospheric hot plasmas, combined with the association of energetic ionospheric outflows with geomagnetic activity, led to the conclusion that increasing geomagnetic activity is responsible for filling the magnetosphere with ionospheric plasma. Recently it has been discovered that a major source of ionospheric heavy ion plasma outflow is responsive to the earliest impact of coronal mass ejecta upon the dayside ionosphere. Thus a large increase in ionospheric outflows begins promptly during the initial phase of geomagnetic storms, and is already present during the main phase development of such storms. We hypothesize that enhancement of the internal source of plasma actually supports the transition from substorm enhancements of aurora to storm-time ring current development in the inner magnetosphere. Other planets known to have ring current-like plasmas also have substantial internal sources of plasma, notably Jupiter and Saturn. One planet having a small magnetosphere, but very little internal source of plasma, is Mercury. Observations suggest that Mercury has substorms, but are ambiguous with regard to the possibility of magnetic storms of the planet. The Messenger mission to Mercury should provide an interesting test of our hypothesis. Mercury should support at most a modest ring current if its internal plasma source is as small as is currently believed. If substantiated, this hypothesis would support a general conclusion that the magnetospheric inflationary response is a characteristic of magnetospheres with substantial internal plasma sources. We quantitatively define this hypothesis and pose it as a problem in comparative magnetospheres.     

Comparison of two measurement fusion methods forKalman-filter-based multisensor data fusion

Currently there exist two commonly used measurement fusion methods for Kalman-filter-based multisensor data fusion. The first (Method I) simply merges the multisensor data through the observation vector of the Kalman filter, whereas the second (Method II) combines the multisensor data based on a minimum-mean-square-error criterion. This paper, based on an analysis of the fused state estimate covariances of the two measurement fusion methods, shows that the two measurement fusion methods are functionally equivalent if the sensors used for data fusion, with different and independent noise characteristics, have identical measurement matrices. Also presented are simulation results on state estimation using the two measurement fusion methods, followed by the analysis of the computational advantages of each method     

Huygens Probe Aerosol Collector Pyrolyser Experiment

ACP's main objective is the chemical analysis of the aerosols in Titan's atmosphere. For this purpose, it will sample the aerosols during descent and prepare the collected matter (by evaporation, pyrolysis and gas products transfer) for analysis by the Huygens Gas Chromatograph Mass Spectrometer (GCMS). A sampling system is required for sampling the aerosols in the 135'32 km and 22'17 km altitude regions of Titan's atmosphere. A pump unit is used to force the gas flow through a filter. In its sampling position, the filter front face extends a few mm beyond the inlet tube. The oven is a pyrolysis furnace where a heating element can heat the filter and hence the sampled aerosols to 250 °C or 600 °C. The oven contains the filter, which has a thimble-like shape (height 28 mm). For transferring effluent gas and pyrolysis products to GCMS, the carrier gas is a labeled nitrogen ¹⁵N₂, to avoid unwanted secondary reactions with Titan's atmospheric nitrogen. Aeraulic tests under cold temperature conditions were conducted by using a cold gas test system developed by ONERA. The objective of the test was to demonstrate the functional ability of the instrument during the descent of the probe and to understand its thermal behavior, that is to test the performance of all its components, pump unit and mechanisms. In order to validate ACP's scientific performance, pyrolysis tests were conducted at LISA on solid phase material synthesized from experimental simulation. The chromatogram obtained by GCMS analysis shows many organic compounds. Some GC peaks appear clearly from the total mass spectra, with specific ions well identified thanks to the very high sensitivity of the mass spectrometer. The program selected for calibrating the flight model is directly linked to the GCMS calibration plan. In order not to pollute the two flight models with products of solid samples such as tholins, we excluded any direct pyrolysis tests through the ACP oven during the first phase of the calibration. Post probe descent simulation of flight results are planned, using the much representative GCMS and ACP spare models. This revised version was published online in August 2006 with corrections to the Cover Date.     

Antideuteron based dark matter search with GAPS: Current progress and future prospects

The General Antiparticle Spectrometer (GAPS) is a new approach to the indirect detection of dark matter. It relies on searching for primary antideuterons produced in the annihilation of dark matter in the galactic halo. Low energy antideuterons produced through Standard Model processes, such as collisions of cosmic-rays with interstellar baryons, are greatly suppressed compared to primary antideuterons. Thus a low energy antideuteron search provides a clean signature of dark matter. In GAPS antiparticles are slowed down and captured in target atoms. The resultant exotic atom deexcites with the emission of X-rays and annihilation pions, protons and other particles. A tracking geometry allows for the detection of the X-rays and particles, providing a unique signature to identify the mass of the antiparticle. A prototype detector was successfully tested at the KEK accelerator in 2005, and a prototype GAPS balloon flight is scheduled for 2011. This will be followed by a full scale experiment on a long duration balloon from Antarctica in 2014. We discuss the status and future plans for GAPS.     

Longitudinal variation of the equatorial ionosphere: Modeling and experimental results

We describe a new version of the Parameterized Regional Ionospheric Model (PARIM) which has been modified to include the longitudinal dependences. This model has been reconstructed using multidimensional Fourier series. To validate PARIM results, the South America maps of critical frequencies for the E (foE) and F (foF2) regions were compared with the values calculated by Sheffield Plasmasphere-Ionosphere Model (SUPIM) and IRI representations. PARIM presents very good results, the general characteristics of both regions, mainly the presence of the equatorial ionization anomaly, were well reproduced for equinoctial conditions of solar minimum and maximum. The values of foF2 and hmF2 recorded over Jicamarca (12°S; 77°W; dip lat. 1°N; mag. declination 0.3°) and sites of the conjugate point equatorial experiment (COPEX) campaign Boa Vista (2.8°N; 60.7°W; dip lat. 11.4°; mag. declination −13.1°), Cachimbo (9.5°S; 54.8°W; dip lat. −1.8°; mag. declination −15.5°), and Campo Grande (20.4°S; 54.6°W; dip lat. −11.1°; mag. declination −14.0°) have been used in this work. foF2 calculated by PARIM show good agreement with the observations, except during morning over Boa Vista and midnight-morning over Campo Grande. Some discrepancies were also found for the F-region peak height (hmF2) near the geomagnetic equator during times of F₃ layer occurrences. IRI has underestimated both foF2 and hmF2 over equatorial and low latitude sectors during evening-nighttimes, except for Jicamarca where foF2 values were overestimated.     

Universal scaling laws for fully-developed magnetic field turbulence near and far upstream of the Earth’s bow shock

We analyze the multifractal scaling of the modulus of the interplanetary magnetic field near and far upstream of the Earth’s bow shock, measured by Cluster and ACE, respectively, from 1 to 3 February 2002. The maximum order of the structure function is carefully estimated for each time series using two different techniques, to ensure the validity of our high-order statistics. The first technique consists of plotting the integrand of the pth order structure function, and the second technique is a quantitative method which relies on the power-law scaling of the extreme events. We compare the scaling exponents computed from the structure functions of magnetic field differences with the predictions obtained by the She–Lévêque model of intermittency in anisotropic magnetohydrodynamic turbulence. Our results show a good agreement between the model and the observations near and far upstream of the Earth’s bow shock, rendering support for the modelling of universal scaling laws based on the Kolmogorov phenomenology in the presence of sheet-like dissipative structures.