Alakai Defense Systems has developed several standoff ultra-violet (UV) Raman systems over the years to enable detection of hazardous chemicals from a safe distance. These systems have traditionally used classical non-machine-learning- based algorithms, but Alakai together with its partner Systems & Technology Research (STR) are currently developing the Agnostic Machine learning Platform for Spectroscopy (AMPS). AMPS, implemented using PyTorch, automatically creates and optimizes tailored one-dimensional (1D) convolutional neural networks (CNN) when trained on simulated or measured data. Several emerging and novel techniques, including advanced domain adaptation approaches, have been implemented to increase model robustness and minimize training data requirements. While the created models are optimized for a specific modality, AMPS itself is agnostic—it can be used for any spectroscopic modality that produces 1D spectra. AMPS has shown promising results for long-wave infrared (LWIR) reflectance spectroscopy as well as UV and near-infrared (NIR) Raman. This talk will focus on AMPS models created using both simulated UV Raman data as well as measured UV Raman data taken with Alakai’s Portable Raman Improvised Explosives Detection (PRIED) system. Performance between AMPS and Alakai’s legacy algorithms will be compared.

We present new results on inferring the hidden states in trackable weak models. A weak model is a directed graph where each node has a set of colors which may be emitted when that node is visited. A hypothesis is a node sequence consistent with a given color sequence. A weak model is trackable if the worst case number of hypotheses grows polynomially in the sequence length. We show that the number of hypotheses in strongly-connected trackable models is bounded by a constant. We also consider the problem of reconstructing which branch was taken at a node with same-colored out-neighbors, and show that it is always eventually possible to identify which branch was taken if the model is strongly connected and trackable. We illustrate these properties by employing standard tools for analyzing Markov chains. In addition, we present new results for the entropy rates of weak models. These theorems indicate that the combination of trackability and strong connectivity simplifies the task of reconstructing which nodes were visited. This work has implications for problems which can be described in terms of an agent traversing a colored graph, such as the reconstruction of hidden states in a hidden Markov model (HMM).

Analysis and modeling of rotation reversal hysteresis experiments show that a single turbulent bifurcation is responsible for the Linear to Saturated Ohmic Confinement (LOC/SOC) transition and concomitant intrinsic rotation reversal on Alcator C-Mod. Plasmas on either side of the reversal exhibit different toroidal rotation profiles and therefore different turbulence characteristics despite the profiles of density and temperature, which are indistinguishable within measurement uncertainty. Elements of this bifurcation are also shown to persist for auxiliary heated L-modes. The deactivation of subdominant (in the linear growth rate and contribution to heat transport) ion temperature gradient and trapped electron mode instabilities is identified as the only possible change in turbulence within a reduced quasilinear transport model across the reversal, which is consistent with the measured profiles and inferred heat and particle fluxes. Experimental constraints on a possible change from strong to weak turbulence, outside the description of the quasilinear model, are also discussed. These results indicate an explanation for the LOC/SOC transition that provides a mechanism for the hysteresis through the dynamics of subdominant modes and changes in their relative populations and does not involve a change in the most linearly unstable ion-scale drift-wave instability.

We present results showing that software programs which are not part of the training set can be characterized into broad classes using involuntary RF side channels. This extends previous work on program identification through analog side channels focused on identifying the specific program out of the training set or flagging previously-unseen programs as "anomalous." This new approach enables an intrusion detection system to be robust to benign changes such as software updates and eliminates the need for an exhaustive training set which covers all possible device functions and states. We have applied our approach to a variety of devices under test, ranging from microcontrollers to laptop computers, and identify program classes such as processor-bound, signal processing, database access, etc. This approach is particularly applicable for defending devices which lack the computational resources to run traditional cybersecurity solutions, including industrial control systems (ICS) and internet of things (IoT) devices.

A colored graph is a directed graph in which nodes or edges have been assigned colors that are not necessarily unique. Observability problems in such graphs consider whether an agent observing the colors of edges or nodes traversed on a path in the graph can determine which node they are at currently or which nodes were visited earlier in the traversal. Previous research efforts have identified several different notions of observability as well as the associated properties of graphs for which those observability properties hold. This paper unifies the prior work into a common framework with several new results about relationships between those notions and associated graph properties. The new framework provides an intuitive way to reason about the attainable accuracy as a function of lag and time spent observing, and identifies simple modifications to improve the observability of a given graph. We show that one form of the graph modification problem is in NP-Complete. The intuition of the new framework is borne out with numerical experiments. This work has implications for problems that can be described in terms of an agent traversing a colored graph, including the reconstruction of hidden states in a hidden Markov model (HMM).

The use of involuntary analog side-channel emissions to remotely identify the internal state of digital platforms has recently emerged as a valuable tool in the arsenal of defensive measures against intrusion and malicious attacks, as well as hardware modifications. In particular RF emissions have been shown to be effective in this task. One of the key challenges is identifying and selecting useful features from the noisy signals which simultaneously enable the detection of the internal digital state reliably while minimizing the complexity of this operation. Our team has developed such sensors and we show the ability to optimally select features as well as optimally select bands of operation from which features can be drawn. Optimality here is in the sense of maximizing the mutual information between the features and the true state of the devices under test. In addition to being optimal in the sense of performance and low complexity for the real-time operation, the process of finding the optimal features is parsimonious and amenable to deployment in adaptive real-time sensors. In these proceedings we describe specific examples related to the detection of intended vs unintended programs on IoT devices and FPGAs as well as identification of other internal device settings. We show near-perfect identification of such internal states, achieved in real-time at distances of several feet in challenging environments.

We present an analysis which suggests that model selection is a critical ingredient for successful reconstruction of impurity transport coefficient profiles, D and V, from experimental data. Determining these quantities is a challenging nonlinear inverse problem. We use synthetic data to show that this problem is ill-posed, and hence D and V are not recommended for use in validation metrics unless the data analysis procedure goes to great lengths to account for the possibility that there are multiple possible solutions. In particular, inferred profiles which are very different from the true ones yield seemingly reasonable goodness-of-fit for synthetic x-ray spectrometer data. We present a Bayesian approach for inferring D and V which provides a rigorous means of selecting the level of complexity of the inferred profiles, thereby enabling successful reconstruction of the profiles.

Analysis and modeling of a new set of rotation reversal hysteresis experiments unambiguously show that changes in turbulence are responsible for the intrinsic rotation reversal and the linear to saturated ohmic confinement (LOC/SOC) transition on Alcator C-Mod. Plasmas on either side of the reversal exhibit different toroidal rotation profiles and therefore different turbulence characteristics despite profiles of density and temperature that are indistinguishable within measurement uncertainty. The deactivation of subdominant (in linear growth rate and heat transport) ion-temperature gradient and trapped electron mode-like instabilities in a mixed-mode state is identified as the only possible change in turbulence within a quasilinear transport approximation across the reversal which is consistent with the measured profiles and the inferred heat and particle fluxes. This indicates an explanation for the LOC/SOC transition that provides a mechanism for hysteresis through the dynamics of subdominant modes and changes in their relative populations, and does not involve a change in most (linearly) unstable ion-scale drift-wave instability.

X-ray spectra of n = 3 to 1 transitions in He-like ions (and sat ellites) from calcium, argon and chlorine have been measured in the core of Alcator C -Mod tokamak plasmas using high wavelength resolution x-ray spectrometer syste m . The intensity ratio of the intercombination line y 3 (1s3pP1 1s S0) to the resonance line w 3 (1s3pP1 1s S0) is found to be much larger than what is expected if collision al excitation out of the ground state is considered as the only population mechan ism for the upper levels. This suggests that recombination and cascades from higher l ev ls with n≥ 4 are important. Modeling with the MARIA code is in good agreement wi th the observations, demonstrating the importance of recombination population of the upper level for y 3. The intensity ratio y 3/w3 has been studied over a large range of core electron temperature and density, and radial position in the plasma. The obse rved ratio decreases with increasing Te, increases with increasing Z and is independent of n e, in agreement with modeling.

Changes in the core intrinsic toroidal rotation velocity fo llowing Lto Hand Lto I-mode transitions have been investigated in Alcator C-M od tokamak plasmas. The magnitude of the co-current rotation increments is found to increase with the pedestal temperature gradient and q 95, and to decrease with toroidal magnetic field. These results are captured quantitatively by a model of fluctuation e ntropy balance which gives the Mach number Mi ∼= ρ∗/2 Ls/LT ∼ ∇T q95/BT in an ITG turbulence dominant regime. The agreement between experiment and theory gives c onfidence for extrapolation to future devices in similar operational regimes. Co re thermal Mach numbers of ∼0.07 and∼0.2 are expected for ITER and ARC, respectively.

Active mid-infrared spectroscopy with tunable lasers is a leading technology for standoff detection and identification of trace chemicals. Information-theoretic optimal selection of the laser wavelength offers the promise of increased detection confidence at lower abundances and with fewer wavelengths. Reducing the number of wavelengths required enables faster detections and lowers sensor power consumption while keeping the optical power under eye safety limits. This paper presents an approximation to the mutual information which operates ~40000x faster than traditional techniques, thereby making near-optimal real-time sensor control computationally feasible. Application of this technique to synthetic data suggests it can reduce the number of wavelengths needed by a factor of two relative to an evenly-spaced grid, with even higher gains for chemicals with weak signatures.

Many dedicated embedded processors do not have memory or computational resources to coexist with traditional (host-based) security solutions. As a result, there is interest in using out-of-band analog side-channel measurements and their analyses to accurately monitor and analyze expected program execution. In this paper, we describe an approach to this problem using externally observable multi-band radio frequency (RF) measurements to make inferences about a program's execution. Because it is very difficult to identify individual instructions solely from their RF emissions, we compare RF measurements with the constrained execution logic of the program so that multiple RF measurements over time can effectively track program execution dynamically. In our approach, a program's execution is modeled by control flow graphs (CFG) and transitions between nodes of such graphs. We demonstrate that tracking performance can be improved through applications program modifications such as changing basic block transition properties and/or adding new basic blocks that are highly observable. In addition to demonstrating these principled approaches on some simple programs, we present initial results on the complexity and structure of real-world applications programs, namely gzip and md5sum, in this modeling framework.

Recent attempts to measure impurity transport in Alcator C-Mod using an x-ray imaging crystal spectrometer and laser blow-off impurity injector have failed to yield unique reconstructions of the transport coefficient profiles. This paper presents a fast, linearized model which was constructed to estimate diagnostic requirements for impurity transport experiments. The analysis shows that the spectroscopic diagnostics on Alcator C-Mod should be capable of inferring simple profiles of impurity diffusion DZ and convection VZ accurate to better than ±10% uncertainty, suggesting that the failure to infer unique DZ and VZ from experimental data is attributable to an inadequate analysis procedure rather than the result of insufficient diagnostics. Furthermore, the analysis reveals that even a modest spatial resolution can overcome a low time resolution. This approach can be adapted to design and verify diagnostics for transport experiments on any magnetic confinement device.

Helium majority experiments on Alcator C-Mod were performed to compare with deuterium discharges, and inform ITER early operations. ELMy H-modes were produced with a special plasma shape at B-T = 5.3 T, I-P = 0.9 MA, at q(95) similar to 3.8. The He fraction ranged over, n(He,L)/n(L) = 0.2-0.4, with n(D,L)/n(L) = 0.15-0.26, compared to D plasmas with n(D,L)/n(L) = 0.85-0.97. The power to enter the H-mode in He was found to be greater than similar to 2 times that for D discharges, in the low density region < 1.4 x 10(20)/m(3). However, it appears to follow the D threshold for higher densities. The stored energies in the He discharges were about 80% of those in D, and about 40% higher net power was required to sustain them compared to D. Global particle confinement times for tungsten of tau(W)*/tau(E) similar to 4 were obtained with ELMy H-modes in He, however accumulation occurred when the ELMs were irregular and infrequent. The electron temperatures and densities in the pedestal were similar between D and He discharges, and the Delta T-e/T-e and Delta n(e)/n(e) values were similar or larger in He than D. The higher net power required to access the H-mode, and sustain it in flattop, for He discharges in C-Mod, imply some limitations for He operation in ITER.

Real-time, standoff detection of trace chemicals on surfaces in the presence of unknown interferent contaminants using active IR spectroscopy poses significant challenges. The measurement time and computational burden can be prohibitive due to the number of spatial pixels, hundreds of potential wavenumbers, and size of the chemical library containing thousands of signatures. Therefore it is advantageous to optimally sample a small subset of the possible wavenumbers, where optimality is meant in the sense of detection and classification performance. Our approach accomplishes this by selecting wavenumbers which maximize the information about chemical identity. This is done by using submodular optimization, a technique which guarantees near-optimality at vanishingly low computational burden. Therefore, the methods shown here lend themselves to the time and resource constrained problems of data acquisition. This is in contrast to more traditional dimensionality reduction approaches such as lowering spectral resolution, random sparse sampling, and principal component analysis which degrade detection performance. In this work we describe methods for optimal illumination wavenumber selection to address the time constraints while addressing the challenges imposed by hardware and environmental artifacts (e.g., atmospheric effects).

As plasma physics research for fusion energy transitions to an increasing emphasis on cross-machine collaboration and numerical simulation, it becomes increasingly important that portable tools be developed to enable data from diverse sources to be analyzed in a consistent manner. This paper presents eqtools, a modular, extensible, open-source toolkit implemented in the Python programming language for handling magnetic equilibria and associated data from tokamaks. eqtools provides a single interface for working with magnetic equilibrium data, both for handling derived quantities and mapping between coordinate systems, extensible to function with data from different experiments, data formats, and magnetic reconstruction codes, replacing the diverse, non-portable solutions currently in use. Moreover, while the open-source Python programming language offers a number of advantages as a scripting language for research purposes, the lack of basic tokamak-specific functionality has impeded the adoption of the language for regular use. Implementing equilibrium-mapping tools in Python removes a substantial barrier to new development in and porting legacy code into Python. In this paper, we introduce the design of the eqtools package and detail the workflow for usage and expansion to additional devices. The implementation of a novel three-dimensional spline solution (in two spatial dimensions and in time) is also detailed. Finally, verification and benchmarking for accuracy and speed against existing tools are detailed. Wider deployment of these tools will enable efficient sharing of data and software between institutions and machines as well as self-consistent analysis of the shared data. Program summary Program title: eqtools Catalogue identifier: AFBICv1_0 Program summary URL: http://cpc.cs.qub.ac.uk/summaries/AFBICv1_0.html Program obtainable from: CPC Program Library, Queen's University, Belfast, N. Ireland Licensing provisions: GNU GPL v3 No. of lines in distributed program, including test data, etc.: 27204 No. of bytes in distributed program, including test data, etc.: 1217844 Distribution format: tar.gz Programming language: Python, C. Computer: PCs. Operating system: Linux, Macintosh OS X, Microsoft Windows. RAM: Several megabytes," depends on resolution of data Classification: 19.4. External routines: F2PY [1], matplotlib [2], MDSplus [3], NumPy [4], SciPy [5] Nature of problem: Access to results from magnetic equilibrium reconstruction code, conversion between various coordinate systems tied to the magnetic equilibrium. Solution method: Data are stored in an object-oriented data structure with human-readable getter methods. Coordinates are converted using bivariate or trivariate splines. Running time: Coordinate transformations on a 66x66 point spatial grid take between 1 and 5 ms per time slice, depending on the transformation used and how many intermediate results have been stored. References: [1] P. Peterson, F2PY: a tool for connecting Fortran and Python programs, International Journal of Computational Science and Engineering 4 (4) (2009) 296-305. [2] J. D. Hunter, Matplotlib: A 2D graphics environment, Computing in Science and Engineering, 9 (3) (2007) 90-95. [3] J. A. Stillerman, T. W. Fredian, K. A. Klare, G. Manduchi, MDSplus data acquisition system, Review of Scientific Instruments 68 (1) (1997) 939-942. [4] S. van der Walt, S. C. Colbert and G. Varoquaux, The NumPy array: a structure for efficient numerical computation, Computing in Science and Engineering 13 (2) (2011) 22-30. [5] E. Jones, T. Oliphant, P Peterson, et al., SciPy: Open source scientific tools for Python (2001-). (C) 2016 Elsevier B.V. All rights reserved.

New experiments on Alcator C-Mod reveal that the favorable impurity screening characteristics of the high-field side (HFS) scrape-offlayer (SOL), previously reported for single null geometries, is retained in double null configurations, despite the formation of an extremely thin SOL. In balanced double-null, nitrogen injected locally into the HFS SOL is better screened by a factor of 2.5 compared to the same injection into the low field side (LFS) SOL. This result is insensitive to plasma current and Greenwald fraction. Nitrogen injected into the HFS SOL is not as well screened (only a factor of 1.5 improvement over LFS) in unbalanced double-null discharges, when the primary divertor is in the direction of B x. B. In this configuration, impurity 'plume' emission patterns indicate that an opposing E xB drift competes with the parallel impurity flow to the divertor. In balanced double-null plasmas, the dispersal pattern exhibits a dominant E xB motion. Unbalanced discharges with the primary divertor opposite the direction of B x. B exhibit excellent HFS screening characteristics - a factor of 5 enhancement compared to LFS. These data support the idea that future tokamaks should locate all RF actuators and close-fitting wall structures on the HFS and employ near-double-null magnetic topologies, both to precisely control plasma conditions at the antenna/ plasma interface and to maximally mitigate the impact of local impurity sources arising from plasma-material interactions. (C) 2016 The Authors. Published by Elsevier Ltd.

Changes in the core intrinsic toroidal rotation velocity following L-to H-and L-to I-mode transitions have been investigated in Alcator C-Mod tokamak plasmas. The magnitude of the co-current rotation increments is found to increase with the pedestal temperature gradient and q(95), and to decrease with toroidal magnetic field. These results are captured quantitatively by a model of fluctuation entropy balance which gives the Mach number M-i congruent to rho(*)/2L(s)/L-T similar to del Tq(95)/B-T in an ITG turbulence dominant regime. The agreement between experiment and theory gives confidence for extrapolation to future devices in similar operational regimes. Core thermal Mach numbers of similar to 0.07 and similar to 0.2 are expected for ITER and ARC, respectively.