http://ilp-www.mit.edu/ List of 40 most recently entered publications, as compiled by the Industrial Liaison Office. en-us Copyright 2009 MIT ILP Sat, 4 Jul 2009 06:25:00 GMT http://ilp-www.mit.edu/favicon.ico http://ilp-www.mit.edu/display_page.a4d?key=P5d http://ilp-www.mit.edu/display_paper.a4d?paperId=83960 In this paper, we consider the problem of multifunctional compression with side information. The problem is how we can compress a source X so that the receiver is able to compute some deterministic functions f1(X; Y1),...,fm(X; Ym), where Yi is available at the receiver as side information. Wyner and Ziv considered this problem for the special case of m = 1 and f1(X; Y1) = X and derived a rate-distortion function. Yamamoto extended this result to the case of having one general function f1(X; Y1) . Both of these results were in terms of an auxiliary random variable. For the case of zero distortion, Orlitsky and Roche gave an interpretation of this variable in terms of properties of characteristic graph which led to a particular coding scheme. This result was extended by providing an achievable scheme based on colorings of the characteristic graph. These previous works only considered the case that the receiver only wants to compute one function (m=1). Here, we want to consider the case that the receiver wants to compute several functions with different side information random variables and zero distortion. Results do not depend on the fact that all functions are desired in one receiver and one can apply them to the case of having several receivers with different desired functions (i.e., functions are separable). We define a new concept named the multi-functional graph entropy which is an extension of the graph entropy defined by Korner. We show that the minimum achievable rate for this problem is equal to the conditional multi-functional graph entropy of random variable X given the side informations. We also propose a coding scheme based on graph colorings to achieve this rate. 07/01/09 Muriel Medard http://ilp-www.mit.edu/display_paper.a4d?paperId=83973 A Motional Stark Effect (MSE) diagnostic system has been installed on the Alcator C-Mod tokamak to measure the plasma internal magnetic pitch angle profile. The diagnostic utilizes polarization patterns from Doppler-shifted Balmer-alpha decay emission from an energetic neutral beam injected into a magnetically confined plasma. This dissertation consists of three parts: (1) the current status of the C-Mod MSE diagnostic which includes major upgrades in the hardware and calibration techniques; (2) the elimination of the spurious drift in the polarization measurements due to thermal-stress induced birefringence; and (3) the measurement of current density profiles in Lower Hybrid Current Drive (LHCD) experiments. The major hardware upgrades include replacement of photomultiplier tubes (PMT’s) with avalanche photodiodes (APD’s) which enhanced the quantum efficiency; installation of a wire-grid polarizer to verify small Faraday rotation in the diagnostic; installation of steep edge filters to minimize pollution by the thermal Balmer-alpha signals; rotation of the Diagnostic Neutral Beam (DNB) which significantly reduced the anomalous effect from the secondary beam neutrals during the beam-into-gas calibrations. The new calibration techniques include two plasma calibrations: plasma current sweeping and the plasma size sweeping whose feasibility was experimentally proven; and an absolute intensity calibration which measured the real optical throughput of the system. A large database study indicates the signal-to-background ratio larger than 100 is required to have the measurement uncertainty under 0.1 degrees. The spurious drift in the measurement has been identified as the thermal-stress induced birefringence imposed on the in-vessel lenses. By modeling this effect as a single wave plate, an in-situ calibration method has been proposed and its feasibility was experimentally verified. Based on the experiments that characterized the thermal response of the system, a single-layer heat shield with gold plating and a lens holder which reduces the thermal conduction path to the lens have been designed and fabricated. A more rigorous model that includes an intrinsic phase shift by mirrors reveals the thermal phase shift can be greatly magnified by the intrinsic phase shift. The current density profiles from LHCD experiments have been obtained from the MSE data corrected by a baseline magnetic equilibrium whose internal profile is constrained by the sawtooth inversion radius. The resultant profiles successfully demonstrate several standard predictions of LHCD theory such as the dependence of efficiency on the parallel refractive index and the off-axis current drive. 06/30/09 Steven Scott http://ilp-www.mit.edu/display_paper.a4d?paperId=83972 Many programming domains involve the manipulation of values distributed through a manifold— examples include sensor networks, smart materials, and biofilms. This paper describes a programming semantics for manifolds based on the amorphous medium abstraction, which places a computational device at every point in the manifold. This abstraction enables the creation of programs that automatically scale to networks of different size and device density. This semantics is currently implemented in our language Proto and compiles for execution on Mica2 Motes and several other platforms. 06/30/09 Jacob Beal http://ilp-www.mit.edu/display_paper.a4d?paperId=83971 Presentation and graphics software enables users to experiment with variations of illustrations. They can revisit recent editing operations using the ubiquitous undo command, but they are limited to sequential exploration. We propose a new interaction metaphor and visualization for operation history. While editing, a user can access a history mode in which actions are denoted by graphical depictions appearing on top of the document. Our work is inspired by the visual language of film storyboards and assembly instructions. Our storyboard provides an interactive visual history, summarizing the editing of a document or a selected object. Each view is composed of action depictions representing the user’s editing actions and enables the user to consider the operation history in context rather than in a disconnected list view. This metaphor provides instant access to any past action and we demonstrate that this is an intuitive interface to a selective undo mechanism. 06/30/09 Sylvain Paris http://ilp-www.mit.edu/display_paper.a4d?paperId=83969 Disruptions are one of the largest problems facing tokamaks. In a large-scale exper- iment such as ITER, disruptions may cause crippling damage and severe setbacks in experimentation. One method for disruption mitigation involves the use of a gas jet which has been tested on both stable plasmas and vertical displacement events (VDEs) on Alcator C-Mod. In both cases, the jet was successful in mitigating dis- ruption effects. The gas jet has not yet been tested on other types of disruptions. Locked-mode disruptions are easily created in C-Mod and could be used to test the ef- fectiveness of the gas jet as a mitigation method if the jet could be fired early enough. It has been empirically observed that the electron cyclotron emissions (ECE) signal displays a flattening of the normally-present sawteeth before the current quench oc- curs in certain locked-mode disruptions. A procedure has been written which detects the ECE sawtooth suppression by calculating changes in the standard deviation of the signal over a moving time-window. This procedure has been programmed into the digital plasma control system (DPCS) for real-time testing. The procedure suc- cessfully located the locked modes present during a run. 06/30/09 Robert Granetz http://ilp-www.mit.edu/display_paper.a4d?paperId=83968 We have developed a design for a new rotated 4-strap ICRF antenna. The design Is based on a modification of the existing C-Mod antennas with the antenna rotated 10° such that the entire structure Is perpendicular to total magnetic field. This rotation is implemented in an attempt to reduce ICRI? impurity production. The rotation results in an antenna with less surface area than the previous antennas and therefore a higher power density will be obtained for a given input power of 3 MW. The power density would reach 1SMW/m2, near the world record power density obtained In Tore Supra. We will describe the RF of this new design. The RF analysis was accomplished using the CST computer code. The fields were studied to minimize E field breakdown in the feed system and to accomplish the symmetry required for reduced impurity production in the four straps. 06/30/09 Peter Koert http://ilp-www.mit.edu/display_paper.a4d?paperId=83967 The Ion Cyclotron RF Transmitter System (ICRF) at Alcator C-Mod comprises four separate transmitters each capable of driving 2 MW of power into plasma loads. Four separate transmission lines guide RF power into three antennas, each mounted in a separate horizontal port, in the C-Mod Tokamak. Protection for the antennas, matching elements and transmission line is accomplished by two unique but interdependent subsystems encompassed by the ICRF Fault System. The Antenna Protection System evaluates antenna phasing and voltage, sets fault thresholds, generates fault signals, and passes fault information to the Master Fault Processor. During operation, the Master Fault Processor is responsible for detecting hazards along the transmission line, generating faults, processing faults from the Antenna Protection System, terminating RF drive and extinguishing faults within 10£gs. In addition, the system controls various delays and sets the boundaries for RF retries. The ICRF Control System provides amplitude regulation for all antennas and phase control for a four-strap antenna. We are modifying some of the fault processing components and control elements of these systems in an effort to improve reliability and serviceability, and increase flexibility. This upgrade will reduce wired interconnections, add remote features to improve access to key operating parameters, improve RF isolation with new switching components, simplify phase control, and expand the RF regulation system to an active control regime whereby plasma parameters may become direct feedback elements for RF regulation. Details of the proposed upgrade to the system will be presented, and implementation of any new technological tools will be discussed. 06/30/09 Miklos Porkolab http://ilp-www.mit.edu/display_paper.a4d?paperId=83966 An upgraded phase contrast imaging (PCI) diagnostic is used to study turbulence and transport in Alcator C-Mod. The upgraded PCI system is capable of measuring density fluctuations with high temporal (2 kHz-5 MHz) and wavenumber (0.5-55 cm-1) resolution. An upgrade of the system has enabled PCI to localize the short wavelength turbulence in the electron temperature gradient (ETG) range and resolve the direction of propagation (i.e., electron vs. ion diamagnetic direction) of the longer wavelength turbulence in the ion temperature gradient (ITG) and trapped electron mode (TEM) range. Nonlinear GYRO simulations have also been performed and the predicted fluctuation is compared against experimental measurements through a synthetic PCI diagnostic method. Both L-Mode and H-Mode plasmas are examined through these numerical and experimental methods. The L-Mode experiments were carried out over the range of densities covering the "neo-Alcator" (linear confinement time scaling with density, electron transport dominates) to the "saturated ohmic" regime. The key role played by the ITG turbulence has been verified. In the saturated ohmic regime, the simulated ion and electron thermal diffusivities also agree with experiments after varying the ion temperature gradient within experimental uncertainty. However, in the linear ohmic regime, GYRO does not agree well with experiments, showing significantly larger ion thermal transport and smaller electron thermal transport. Our study shows that although the short wavelength turbulence in the ETG range is unstable in the linear ohmic regime, the nonlinear simulation with k£c£ls up to 4 does not raise the electron thermal diffusivity to the experimental level, where k£c is the poloidal wavenumber and £ls is the ion-sound Larmor radius. The H-Mode studies focus on plasmas before and during internal transport barrier formation in an enhanced D(alpha) H-Mode plasma. The simulated fluctuations from GYRO agree with experimental measurements in the ITG regime. GYRO also shows good agreement in transport predictions with experimental measurements after reducing the ion temperature gradient (~15%) and adding E X B shear suppression, all within the experimental uncertainty. 06/30/09 Miklos Porkolab http://ilp-www.mit.edu/display_paper.a4d?paperId=83965 Magnetic reconnection is an important process in magnetized plasmas ranging from the laboratory to astrophysical scales. It enables the release of magnetic energy believed to power solar ares and magnetospheric substorms. Reconnection also controls the evolution of the topology of the magnetic eld, enabling deleterious instabilities, such as the sawtooth instability in fusion experiments, to transport plasma across the experiment's minor radius. Notably, simple estimates of the finite reconnection rate due to classical resistivity fail to explain the fast and explosive nature of reconnection observed in these systems. A major goal of reconnection research is to determine which mechanisms enable \fast" reconnection to occur. This thesis studied the fluctuations arising in the plasma during magnetic reconnection experiments on the Versatile Toroidal Facility (VTF), with a primary goal of testing whether \anomalous resistivity" due to micro-instabilities can speed the reconnection process. Fluctuations were studied using impedance-matched, high-bandwidth Langmuir probes. Strong, broadband fluctuations, with frequencies extending from near the lower-hybrid frequency [fLH = (fcefci)1=2] to the electron cyclotron frequency fce were found to arise during the reconnection events. Based on frequency and wavelength measurements, lower-hybrid waves and Trivelpiece-Gould waves were identified. The lower-hybrid waves appear to be driven by strong perpendicular drifts or gradients which arise due to the reconnection events; an appealing possibility is strong temperature gradients. The Trivelpiece-Gould modes were found to result from kinetic, bump-on-tail instability of a runaway electron population energized by the reconnection events. Nonlinear, spiky turbulence was also observed, and attributed to the creation of \electron phase-space holes," a class of nonlinear solitary wave known to evolve from a strong beam-on-tail instability. Overall, these instabilities were found to be a consequence of reconnection, specifically the strong energization of electrons, leading to steep gradients in both coordinate- and velocity-space. However, it was not established that these modes had a strong feedback on the reconnection process: fluctuation power varied strongly between discharges and was observed to systematically trail the reconnection events. Finally, crude estimates (using quasi-linear theory) of the anomalous resistivity due to these modes did not appear large enough to substantially impact the reconnection process. 06/30/09 William Fox http://ilp-www.mit.edu/display_paper.a4d?paperId=83964 It is commonly accepted that ExB velocity shear is responsible for the suppression of edge turbulence, which reduces the losses of both energy and particles across magnetic field lines and results in the formation of edge transport barriers and high-confinement mode (H-mode) in tokamak plasmas. However, the self consistent evolution of the radial electric field profile (Er), pedestal shape and improvement in plasma confinement is not well understood. A better understanding of pedestal physics and the interplay between Er, turbulence suppression and pedestal formation should enable better control of edge transport and improve core confinement. A new, high-resolution, charge exchange recombination spectroscopy (CXRS) diagnostic has been installed on Alcator C-Mod to provide measurements of the B5+ population in the pedestal region. This diagnostic is capable of measuring the boron temperature, density, and poloidal and toroidal velocity with 3mm radial resolution and 5ms temporal resolution. These profiles, coupled with knowledge of the toroidal and poloidal magnetic fields, enable the determination of the edge radial electric field through the radial force balance equation. The new CXRS diagnostic has provided the first spatially resolved calculations of the radial electric field in the C-Mod edge and has made possible significant contributions to the study of pedestal physics. Detailed measurements of the boron population have been made in a variety of plasma regimes. The measured rotation profiles connect the SOL and core measurements and are consistent with both. The CXRS boron temperature profiles are observed to agree well with the Thomson Scattering electron temperature profiles in both shape and magnitude over a wide range of collisionalities. In H-mode plasmas both the boron temperature and density profiles form clear pedestals, similar to what is observed in the electron channel. The edge toroidal rotation increases in the co-current direction at the onset of H-mode confinement and the poloidal rotation in the pedestal region increases in the electron diamagnetic direction forming a narrow peak (3-4mm) just inside of the LCFS. In Ohmic L-mode plasmas Er is positive near the last closed ux surface (LCFS) and becomes more negative with distance into the plasma. In H-mode plasmas Er is positive in the core, but forms a deep negative well, relative to its L-mode values, just inside of the LCFS. These results are qualitatively consistent with the observations made on other machines. However, the C-Mod H-mode Er wells are unprecedented in depth (up to 300kV/m) and the narrow Er well widths (5mm), as compared to results from other tokamaks, suggest a scaling with machine size. The measured Er well widths have been compared to theoretical scalings for the edge pedestal and no significant correlation was observed with any of the predictions. In fact, very little variation of the Er well width is observed in general. However, the depth of the Er well, or alternatively the magnitude of the Er shear (constant width), shows a strong correlation with improved plasma energy confinement. It also correlates well with the edge electron temperature and pressure pedestal heights (and gradients). It is not, however, very sensitive to variation in the edge electron density pedestal height. These results are an indication that the energy and particle transport have different relationships to Er, with energy transport more directly linked. The radial electric field results from ELM-free H-mode and I-mode plasmas support this interpretation. 06/30/09 Bruce Lipschultz http://ilp-www.mit.edu/display_paper.a4d?paperId=83963 The collapse of strong convergent shocks in spherical geometry is observed using measurements of induced nuclear production and x-ray emission. Precise and absolute measurements of the timing and yield of nuclear production induced by the collapse of laser-driven shocks give the same results when shocks are launched by uniform (<2% rms) or non-uniform (up to 32% rms) laser illumination. The observation was repeated for both low-mode (dominated by spherical harmonic modes ÜP = 1-2) and high mode (ÜP = 31-500) drive asymmetries. For low-mode nonuniform drive, the center of collapse as observed through x-ray emission shifts away from target center toward the direction of low intensity. The x-ray emission brightness is seen to drop precipitously with larger low-mode drive asymmetry, in stark contrast to the drive-uniformity insensitivity of nuclear yields at the time of shock collapse. 06/30/09 Johan Frenje http://ilp-www.mit.edu/display_paper.a4d?paperId=83962 A distinctive way of quantitatively imaging inertial fusion implosions has resulted in the characterization of two different types of electromagnetic configurations and in the measurement of the temporal evolution of capsule size and areal density. Radiography with a pulsed, monoenergetic, isotropic proton source reveals field structures through deflection of proton trajectories, and areal densities are quantified through the energy lost by protons while traversing the plasma. The two field structures consist of (i) many radial filaments with complex striations and bifurcations, permeating the entire field of view, of magnetic field magnitude 60 tesla, and (ii) a coherent, centrally directed electric field of order 109 volts per meter, seen in proximity to the capsule surface. Although the mechanism for generating these fields is unclear, their effect on implosion dynamics is potentially consequential. 06/30/09 Fredrick Seguin http://ilp-www.mit.edu/display_paper.a4d?paperId=83961 While many developing-country policymakers see heavy fertilizer subsidies as critical to raising agricultural productivity, most economists see them as distortionary, regressive, environmentally unsound, and argue that they result in politicized, inefficient distribution of fertilizer supply. We model farmers as facing small fixed costs of purchasing fertilizer, and assume some are stochastically present-biased and not fully sophisticated about this bias. Even when relatively patient, such farmers may procrastinate, postponing fertilizer purchases until later periods, when they may be too impatient to purchase fertilizer. Consistent with the model, many farmers in Western Kenya fail to take advantage of apparently profitable fertilizer investments, but they do invest in response to small, time-limited discounts on the cost of acquiring fertilizer (free delivery) just after harvest. Later discounts have a smaller impact, and when given a choice of price schedules, many farmers choose schedules that induce advance purchase. Calibration suggests such small, time-limited discounts yield higher welfare than either laissez faire or heavy subsidies by helping present-biased farmers commit to fertilizer use without inducing those with standard preferences to substantially overuse fertilizer. 06/30/09 Esther Duflo http://ilp-www.mit.edu/display_paper.a4d?paperId=83959 For a tokamak, we consider gyrokinetic quasineutrality limitations when evaluating the axisymmetric radial electric field; provide an insight by considering the gyrokinetic entropy production restriction on an ion temperature pedestal like that of ITER; and an present an improved hybrid gyrokinetic-fluid treatment valid on slowly evolving transport time scales. 06/30/09 Peter Catto http://ilp-www.mit.edu/display_paper.a4d?paperId=83958 Unique detection of electromagnetic fields, and identification of field type and strength as a function of position, were used to determine the nature of self-generated fields in a novel experiment with laser- generated plasma bubbles on two sides of a plastic foil. Field-induced deflections of monoenergetic 1 5-MeV probe protons passing through the two bubbles, measured quantitatively with proton radiography, were combined with Lorentz mapping to provide separate measurements of magnetic and electric fields. The result was absolute identification and measurement of a toroidal magnetic field around each bubble and determination that any electric field component parallel to the foil was below measurement uncertainties. 06/30/09 Richard Petrasso http://ilp-www.mit.edu/display_paper.a4d?paperId=83957 Groups of frequency chirping modes observed between sawtooth crashes in the Alcator C-Mod tokamak are interpreted as reversed shear Alfvén eigenmodes near the q=1 surface. These modes indicate that a reversed shear q profile is generated during the relaxation phase of the sawtooth cycle. Two important parameters, and its radial position, are deduced from comparisons of measured density fluctuations with calculations from the ideal MHD code NOVA. These studies provide valuable constraints for further modeling of the sawtooth cycle. 06/30/09 E M Edlund http://ilp-www.mit.edu/display_paper.a4d?paperId=83956 Reversed shear Alfvén eigenmodes (RSAEs) have been observed with the phase contrast imaging diagnostic and Mirnov coils during the sawtooth cycle in Alcator C-mod' plasmas with minority ion cyclotron resonance heating. Both down-chirping RSAEs and up-chirping RSAEs have been observed during the sawtooth cycle. Experimental measurements of the spatial structure of the RSAEs are compared to theoretical models based on the code NOVA 2, and used to derive con- straints on the q profile. It is shown that the observed RSAEs can be understood by assuming a reversed shear q profile (up-chirping) or a q profile with a local maximum (down-chirping) with q 1. 06/30/09 Miklos Porkolab http://ilp-www.mit.edu/display_paper.a4d?paperId=83955 Recent advances in gyrokinetic simulation of core turbulence and associated transport requires an intensified experimental effort to validate these codes using state of the art synthetic diagnostics to compare simulations with experimental data. A phase contrast imaging (PCI) diagnostic is used to study H-Mode plasmas in Alcator C-Mod. The PCI system is capable of measuring density fluctuations with high temporal (2 kHz-5 MHz) and wavenumber (0.5-55 cnf') resolution. Recent upgrades have enabled PCI to localize the short wavelength turbulence in the electron temperature gradient (ETG) range and resolve the direction of propagation (i.e., electron vs. ion diamagnetic direction) of the longer wavelength turbulence in the ion temperature gradient (ITG) and trapped electron mode (TEM) range. The studies focus on plasmas before and during internal transport barrier formation in an enhanced Da H-Mode plasma assisted with ion cyclotron resonance frequency (ICRF) heating. Nonlinear GYRO simulations have also been performed [J. Candy and R. E. Waltz, Phys. Rev. Lett. 91, 045001 (2003)] and the predicted fluctuation is compared against experimental measurements through a synthetic PCI diagnostic method. The simulated fluctuations from GYRO agree with experimental measurements in the ITG regime. GYRO also shows good agreement in transport predictions with experimental measurements after reducing the ion temperature gradient (-45%) and adding EXB shear suppression, all within the experimental uncertainty. 06/29/09 Miklos Porkolab http://ilp-www.mit.edu/display_paper.a4d?paperId=83954 Recent advances in gyrokinetic simulation have allowed for quantitative predictions of core turbulence and associated transport. However, numerical codes must be tested against experimental results in both turbulence and transport. In this paper, we present recent results from ohmic plasmas in the Alcator C-Mod tokamak using phase contrast imaging (PCI) diagnostic, which is capable of measuring density fluctuations with wave-numbers up to 55 cm'. The experiments were carried out over the range of densities covering the "neo-Alcator" (linear confinement time scaling with density, electron transport dominates) to the "saturated ohmic" regime. We have also simulated these plasmas with the gyrokinetic code GYRO and compared numerical predictions with experimentally measured turbulence through a synthetic PCI diagnostic method. The key role played by the ion temperature gradient (ITG) turbulence has been verified, including measurements of turbulent wave propagation in the ion diamagnetic direction. It is found that the intensity of density fluctuations increase with density, in agreement between simulation and experiments. The absolute fluctuation intensity agrees with simulation within experimental error (+1-60%). In the saturated ohmic regime, the simulated ion and electron thermal diffusivities also agree with experiments after varying the ion temperature gradient within experimental uncertainty. However, in the linear ohmic regime, GYRO does not agree well with experiments, showing significantly larger ion thermal transport and smaller electron thermal transport. Our study shows that although the short wavelength turbulence in the electron temperature gradient (ETG) range is unstable in the linear ohmic regime, the nonlinear time, it is not known if even shorter wavelength turbulence would account for the measured electron transport. 06/29/09 Miklos Porkolab http://ilp-www.mit.edu/display_paper.a4d?paperId=83953 Nonlinear gyrokinetic simulations of trapped electron mode (TEM) turbulence, within an internal particle transport barrier, are performed and compared with experimental data. The results provide a mechanism for transport barrier control with on-axis radio frequency heating, as demonstrated in Alcator C-Mod experiments [S. J. Wukitch at at, Phys. Plasmas 9(5) 2149 (2002)[. Off-axis heating produces an internal particle and energy transport barrier after the transition to enhanced Dc high confinement mode. The barrier foot reaches the half-radius, with a peak density 2.5 times the edge density. While the density profile peaks, the temperature profile remains relatively unaffected. The peaking and concomitant impurity accumulation are controlled by applying modest central heating power late in the discharge. Cyrokinetic turbulence simulations of the barrier formation phase, using the 032 code (W. Dorland at al., Phys. Rev. Lett. 85, 5579 (2000)1 show that toroidal ion temperature gradient driven modes are suppressed inside the barrier foot, but continue to dominate in the outer half-radius. As the density gradient steepens further, trapped electron modes are driven unstable. The onset of TEM turbulence produces an outflow that strongly increases with the density gradient, upon exceeding a new nonlinear critical density gradient, which significantly exceeds the linear critical density gradient. The TEM turbulent outflow ultimately balances the inward Ware pinch, leading to steady state. Moreover, the simulated turbulent particle diffusivity matches that inferred from particle balance using measured density profile data and the calculated Ware pinch. This turbulent diffusivity exhibits a strong unfavorable temperature dependence that allows control with central heating. 06/29/09 Darin Ernst http://ilp-www.mit.edu/display_paper.a4d?paperId=83952 Trapped electron mode (TEM) turbulence exhibits a rich variety of collisional and zonal flow physics. This work explores the parametric variation of zonal flows and underlying mechanisms through a series of linear and nonlinear gyrokinetic simulations, using both particle-in-cell (the GEM code) and continuum (the GS2 code) methods. The two codes are shown to closely agree except at the largest values of il, = dlnTe/dlnnc > 5, where qualitative agreement is found. A new stabil- ity diagram for electron modes is presented, identifying = 1 as a critical boundary separating long and short wavelength TEMs, A scan shows fine scale structure appears when il, E 1, con- sistent with linear expectations. Zonal flows are weak when 1 exceeds unity. For Tl > I, trans- port levels fall inversely with a power law in i. For i>> 1, bispectral analysis supports a sim- ple analytic model in which the dominant primary mode couples to itself to drive zonal fluctuations. 06/29/09 Darin Ernst http://ilp-www.mit.edu/display_paper.a4d?paperId=83951 Two research subjects: (1) excitation of "forced ion acoustic waves," and (2) "simul-taneous excitation of plasma density fluctuations and geomagnetic field fluctuations" are reported in my M.S. thesis. The data was acquired in our experiments con-ducted at Gakona, Alaska from summer 2007 to winter 2008, using DoD/NSF-funded HAARP facilities and our own optical (ASIS) and radio instruments (VLF receiv- ing system of IRIS) aided by GPS satellites as well as AMISR radar at Poker Flat, Alaska. We suggest that "Forced ion acoustic waves" detected by MUIR radar on October 29 during 6:20-6:30 UT arise from key electron precipitation associated with the occurrence of green aurora. Work shows, for the first time, that MUIR radar is suitable for probing naturally occurring space plasma processes and not limited to HF heater-induced effects. This would extend the usage of MUIR for the investigation of space weather together with AMISR radar at Poker Flat, to advance our knowledge in space plasma turbulence. The research on "simultaneous excitation of plasma density fluctuations and ge- omagnetic field fluctuations" is an extension of my B.S. thesis research on thermal filamentation instability, which started in our summer Gakona experiments in 2005. Large plasma sheets (also known as sheet-like filaments) can be excited by HF 0- mode and X-mode heater waves via thermal filamentation instability. The dominant nonlinearity is provided by the differential Joule heating acting on electrons, which subsequently gives rise to a cross-field thermal pressure force, to concomitantly gener- ate spatially varying plasma density fluctuations and geomagnetic field fluctuations. It is interesting to find that the fractional density fluctuations are approximately equal to the fractional magnetic field fluctuations. This gives us the theoretical basis to use ground-based magnetometer measurements to infer the density fluctuations in space plasma turbulence. Such a remote sensing technique for probing the space plasma is much more effective and economic than using a beacon satellite. 06/29/09 Joel Cohen http://ilp-www.mit.edu/display_paper.a4d?paperId=83949 Internal particle and thermal energy transport barriers are produced in Alcator C-Mod with off-axis ICRF heating, with core densities exceeding 1021, m3, without core fueling, and with little change in the temperature profile. Applying on-axis ICRP heating controls the core density gradient and rate of rise. The present study employs linear and nonlinear gyrokinetic simulations of trapped electron mode (TEM) turbulence to explore mechanisms for 1TB formation and control in Alcator C-Mod ITB experiments. Anomalous pinches are found to be negligible in our simulations; further, the collisional Ware pinch is sufficient to account for the slow density rise, lasting many energy confinement times. The simulations have revealed new nonlinear physics of TEM turbulence. The critical density gradient for onset of TEM turbulent transport is nonlinearly up-shifted by zonal flows. As the density profile peaks, during ITB formation, this nonlinear critical gradient is eventually exceeded, and the turbulent particle difflisivity from GS2 gyrokinetic simulations matches the particle diffusivity from transport analysis, within experimental errors. A stable equilibrium is then established when the TEM turbulent diffusion balances the Ware pinch in the ITB. This equilibrium is sensitive to temperature through gyroBohm scaling of the TEM turbulent transport, and the collisionality dependence of the neoclassical pinch, providing for control of the density rate of rise with on-axis RP heating. With no core particle fueling, and '-4 mm between density spatial channels, the C-Mod experiments provide a nearly ideal test bed for particle transport studies. The pure TEM is the only unstable drift mode in the ITB, producing particle transport driven by the density gradient. 06/29/09 Darin Ernst http://ilp-www.mit.edu/display_paper.a4d?paperId=83948 Designing robust algorithms for mobile agents with reliable communication is difficult due to the distributed nature of computation, in mobile ad hoc networks (MANETs) the matter is exacerbated by the need to ensure connectivity. Existing distributed algorithms provide coordination but typically assume connectivity is ensured by other means. We present a connectivity service that encapsulates an arbitrary motion planner and can refine any plan to preserve connectivity (the graph of agents remains connected) and ensure progress (the agents advance towards their goal). The service is realized by a distributed algorithm that is modular in that it makes no assumptions of the motion-planning mechanism except the ability for an agent to query its position and intended goal position, local in that it uses 1-hop broadcast to communicate with nearby agents but doesn’t need any network routing infrastructure, and oblivious in that it does not depend on previous computations. We prove the progress of the algorithm in one round is at least (min(d, r)), where d is the minimum distance between an agent and its target and r is the communication radius. We characterize the worst case configuration and show that when d r this bound is tight and the algorithm is optimal, since no algorithm can guarantee greater progress. Finally, we show all agents get "-close to their targets within O(D0/r + n2/") rounds where n is the number of agents and D0 is the sum of the initial distances to the targets. 06/24/09 Nancy Lynch http://ilp-www.mit.edu/display_paper.a4d?paperId=83947 Meeting future world energy needs while addressing climate change requires large-scale deployment of low or zero greenhouse gas (GHG) emission technologies such as wind energy. The widespread availability of wind power has fueled legitimate interest in this renewable energy source as one of the needed technologies. For very large-scale utilization of this resource, there are however potential environmental impacts, and also problems arising from its inherent intermittency, in addition to the present need to lower unit costs. To explore some of these issues, we use a threedimensional climate model to simulate the potential climate effects associated with installation of wind-powered generators over vast areas of land or coastal ocean. Using windmills to meet 10% or more of global energy demand in 2100, could cause surface warming exceeding 1oC over land installations. In contrast, surface cooling exceeding 1oC is computed over ocean installations, but the validity of simulating the impacts of windmills by simply increasing the ocean surface drag needs further study. Significant warming or cooling remote from both the land and ocean installations, and alterations of the global distributions of rainfall and clouds also occur. These results are influenced by the competing effects of increases in roughness and decreases in wind speed on near-surface turbulent heat fluxes, the differing nature of land and ocean surface friction, and the dimensions of the installations parallel and perpendicular to the prevailing winds. These results are also dependent on the accuracy of the model used, and the realism of the methods applied to simulate windmills. Additional theory and new field observations will be required for their ultimate validation. Intermittency of wind power on daily, monthly and longer time scales as computed in these simulations and inferred from meteorological observations, poses a demand for one or more options to ensure reliability, including backup generation capacity, very long distance power transmission lines, and onsite energy storage, each with specific economic and/or technological challenges. 06/24/09 Chien Wang http://ilp-www.mit.edu/display_paper.a4d?paperId=83946 In order to examine the underlying longer-term trends in greenhouse gases, that are driven for example by anthropogenic emissions or climate change, it is useful to remove the recurring effects of natural cycles and oscillations on the sources and/or sinks of those gases that have strong biological (e.g., CO2, CH4, N2O) and/or photochemical (e.g. CH4) influences on their global atmospheric cycles. We use global observations to calculate monthly estimates of greenhouse gas levels expressed as CO2 equivalents, and then fit these estimates to a semi-empirical model that includes the natural seasonal, QBO, and ENSO variations, as well as a second order polynomial expressing longer-term variations. We find that this model provides a reasonably accurate fit to the observation-based monthly data. We also show that this semiempirical model has some predictive capability; that is it can be used to provide a reasonably reliable estimate of CO2 equivalents at the current time using validated observations that lag real time by a few to several months. 06/24/09 Jin Huang http://ilp-www.mit.edu/display_paper.a4d?paperId=83945 Conical stromatolites are thought to be robust indicators of the presence of photosynthetic and phototactic microbes in aquatic environments as early as 3.5 billion years ago. However, phototaxis alone cannot explain the ubiquity of disrupted, curled, and contorted laminae in the crests of many Mesoproterozoic, Paleoproterozoic, and some Archean conical stromatolites. Here, we demonstrate that cyanobacterial production of oxygen in the tips of modern conical aggregates creates contorted laminae and submillimeter- to-millimeter-scale enmeshed bubbles. Similarly sized fossil bubbles and contorted laminae may be present only in the crestal zones of some conical stromatolites 2.7 billion years old or younger. This implies not only that cyanobacteria built Proterozoic conical stromatolites but also that fossil bubbles may constrain the timing of the evolution of oxygenic photosynthesis. 06/24/09 Biqing Liang http://ilp-www.mit.edu/display_paper.a4d?paperId=83944 The human eye lens is composed of fiber cells packed with crystallins up to 450 mg/ml. Human gammaD-crystallin is a monomeric, twodomain protein of the lens central nucleus. Both domains of this long-lived protein have double Greek key beta-sheet folds with wellpacked hydrophobic cores. Three mutations resulting in amino acid substitutions in the gamma-crystallin buried cores -- two in the N-terminal domain, and one in the C-terminal domain -- cause early-onset cataracts in mice. Unfortunately, it has not been possible to identify the aggregating precursor within lens tissues. To compare in vivo cataract-forming phenotypes with in vitro unfolding and aggregation of gamma-crystallins, the mouse mutant substitutions were introduced into HgammaD-Crys. The HgammaD-Crys mutant proteins L5S, V75D and I90F were expressed and purified from E. coli. WT HgammaD-Crys unfolds in vitro through a threestate pathway, exhibiting an intermediate with the N-terminal domain unfolded, and the Cterminal domain native-like. L5S and V75D in the N-terminal domain also displayed threestate unfolding transitions, with the first transition, unfolding of the N-terminal domain, shifted to significantly lower denaturant concentrations. I90F destabilized the Cterminal domain, shifting the overall unfolding transition to lower denaturant concentrations. During thermal denaturation, the mutant proteins exhibited lowered thermal stability compared with WT. Kinetic unfolding experiments showed that the N-td’s of L5S and V75D unfolded faster than WT. I90F was globally destabilized and unfolded more rapidly. These results support models of cataract formation in which generation of partially unfolded species are precursors to the aggregated cataractous states responsible for light scattering. 06/24/09 Jonathan King http://ilp-www.mit.edu/display_paper.a4d?paperId=83943 Today’s situational awareness requirements in the undersea environment present severe challenges for acoustic communication systems. Acoustic propagation through the ocean environment severely limits the capacity of existing underwater communication systems. Specifically, the presence of internal waves coupled with the ocean sound channel creates a stochastic field that introduces deep fades and significant intersymbol interference (ISI) thereby limiting reliable communication to low data rates. In this paper we present a communication architecture that optimally predistorts the acoustic wave via spatial modulation and detects the acoustic wave with optimal spatial recombination to maximize reliable information throughput. This effectively allows the system to allocate its power to the most efficient propagation modes while mitigating ISI. Channel state information is available to the transmitter through low rate feedback. New results include the asymptotic distribution of singular values for a large number of apertures. Further, we present spatial modulation at the transmitter and spatial recombination at the receiver that asymptotically minimize bit error rate (BER). We show that, in many applications, the number of apertures can be made large enough so that asymptotic results approximate finite results well. Additionally, we show that the interference noise power is reduced proportional to the inverse of the number of receive apertures. Finally, we calculate the asymptotic BER for the sparse aperture acoustic system. 06/19/09 Pierre Lermusiaux http://ilp-www.mit.edu/display_paper.a4d?paperId=83942 This work presents four partitioning strategies, or patterns, useful for decomposing a serial application into multiple concurrently executing parts. These partitioning strategies augment the commonly used task and data parallel design patterns by recognizing that applications are spatiotemporal in nature. Therefore, data and instruction decomposition are further distinguished by whether the partitioning is done in the spatial or in temporal dimension. Thus, this work describes four decomposition strategies: spatial data partitioning (SDP), temporal data partitioning (TDP), spatial instruction partitioning (SIP), and temporal instruction partitioning (TIP), while cataloging the benefits and drawbacks of each. In addition, the practical use of these strategies is demonstrated through a case study in which they are applied to implement several different parallelizations of a multicore H.264 encoder for HD video. This case study illustrates both the application of the patterns and their effects on the performance of the encoder. 06/17/09 Anant Agarwal http://ilp-www.mit.edu/display_paper.a4d?paperId=83941 We describe a modeling framework and collection of foundational composition results for the study of probabilistic distributed algorithms in synchronous radio networks. Existing results in this setting rely on informal descriptions of the channel behavior and therefore lack easy comparability and are prone to error caused by definition subtleties. Our framework rectifies these issues by providing: (1) a method to precisely describe a radio channel as a probabilistic automaton; (2) a mathematical notion of implementing one channel using another channel, allowing for direct comparisons of channel strengths and a natural decomposition of problems into implementing a more powerful channel and solving the problem on the powerful channel; (3) a mathematical definition of a problem and solving a problem; (4) a pair of composition results that simplify the tasks of proving properties about channel implementation algorithms and combining problems with channel implementations. Our goal is to produce a model streamlined for the needs of the radio network algorithms community. 06/17/09 Nancy Lynch http://ilp-www.mit.edu/display_paper.a4d?paperId=83940 A diversity of possible communication assumptions complicates the study of algorithms and lower bounds for radio networks. We address this problem by defining an Abstract MAC Layer. This service provides reliable local broadcast communication, with timing guarantees stated in terms of a collection of abstract \emph{delay functions} applied to the relevant contention. Algorithm designers can analyze their algorithms in terms of these functions, independently of specific channel behavior. Concrete implementations of the Abstract MAC Layer over basic radio network models generate concrete definitions for these delay functions, automatically adapting bounds proven for the abstract service to bounds for the specific radio network under consideration. To illustrate this approach, we use the Abstract MAC Layer to study the new problem of Multi-Message Broadcast, a generalization of standard single-message broadcast, in which any number of messages arrive at any processes at any times.We present and analyze two algorithms for Multi-Message Broadcast in static networks: a simple greedy algorithm and one that uses regional leaders. We then indicate how these results can be extended to mobile networks. 06/17/09 Nancy Lynch http://ilp-www.mit.edu/display_paper.a4d?paperId=83939 Depth of field (DOF), the range of scene depths that appear sharp in a photograph, poses a fundamental tradeoff in photography -- wide apertures are important to reduce imaging noise, but they also increase defocus blur. Recent advances in computational imaging modify the acquisition process to extend the DOF through deconvolution. Because deconvolution quality is a tight function of the frequency power spectrum of the defocus kernel, designs with high spectra are desirable. In this paper we study how to design effective extended-DOF systems, and show an upper bound on the maximal power spectrum that can be achieved. We analyze defocus kernels in the 4D light field space and show that in the frequency domain, only a low-dimensional 3D manifold contributes to focus. Thus, to maximize the defocus spectrum, imaging systems should concentrate their limited energy on this manifold. We review several computational imaging systems and show either that they spend energy outside the focal manifold or do not achieve a high spectrum over the DOF. Guided by this analysis we introduce the lattice-focal lens, which concentrates energy at the low-dimensional focal manifold and achieves a higher power spectrum than previous designs. We have built a prototype lattice-focal lens and present extended depth of field results. 06/17/09 Anat Levin http://ilp-www.mit.edu/display_paper.a4d?paperId=83938 This paper presents remote store programming (RSP). This paradigm combines usability and efficiency through the exploitation of a simple hardware mechanism, the remote store, which can easily be added to existing multicores.Remote store programs are marked by fine-grained and one-sided communication which results in a stream of data flowing from the registers of a sending process to the cache of a destination process. The RSP model and its hardware implementation trade a relatively high store latency for a low load latency because loads are more common than stores, and it is easier to tolerate store latency than load latency. This paper demonstrates the performance advantages of remote store programming by comparing it to both cache-coherent shared memory and direct memory access (DMA) based approaches using the TILEPro64 processor. The paper studies two applications: a two-dimensional Fast Fourier Transform (2D FFT) and an H.264 encoder for high-definition video. For a 2D FFT using 56 cores, RSP is 1.64x faster than DMA and 4.4x faster than shared memory. For an H.264 encoder using 40 cores, RSP achieves the same performance as DMA and 4.8x the performance of shared memory. Along with these performance advantages, RSP requires the least hardware support of the three. RSP's features, performance, and hardware simplicity make it well suited to the embedded processing domain. 06/17/09 Anant Agarwal http://ilp-www.mit.edu/display_paper.a4d?paperId=83936 This paper proposes a new market-based distributed planning algorithm for multi-agent systems under uncertainty, called MIRA (Market-based Iterative Risk Allocation). In large coordination problems, from power grid management to multi-vehicle missions, multiple agents act collectively in order to optimize the performance of the system, while satisfying mission constraints. These optimal plans are particularly susceptible to risk when uncertainty is introduced. We present a distributed planning algorithm that minimizes the system cost while ensuring that the probability of violating mission constraints is below a user-specified level. We build upon the paradigm of risk allocation (Ono and Williams, AAAI-08), in which the planner optimizes not only the sequence of actions, but also its allocation of risk among each constraint at each time step. We extend the concept of risk allocation to multi-agent systems by highlighting risk as a good that is traded in a computational market. The equilibrium price of risk that balances the supply and demand is found by an iterative price adjustment process called tatonnement (also known as Walrasian auction). The simulation results demonstrate the efficiency and optimality of the proposed distributed planner. 06/17/09 Brian Williams http://ilp-www.mit.edu/display_paper.a4d?paperId=83935 The problem of energy technology innovation has moved to the forefront of public and political debate for the first time since the energy crises of the 1970s. Politicians of both major parties, scientists, business leaders, and pundits are calling for a major new national commitment to accelerating energy innovation. They argue that this is an essential ingredient of any successful strategy for dealing with the three major and interconnected energy-related challenges confronting the nation: escalating energy prices, oil supply vulnerabilities, and global climate change. To address these challenges will require the country to make a transition away from its current heavy dependence on petroleum for transportation and on high-carbon fuels (mostly coal) for electricity generation, towards a much more energy-efficient economy based on alternative fuels and low or zero-carbon electrical generating technologies. 06/17/09 Richard Lester http://ilp-www.mit.edu/display_paper.a4d?paperId=83934 The statistical analysis of firm-level data on U.S. multinational companies was conducted at the Bureau of Economic Analysis, U.S. Department of Commerce under arrangements that maintain legal confidentiality requirements. The views expressed are those of the authors and do not reflect official positions of the U.S. Department of Commerce. We thank Anil Kashyap for inspiration for the title, and Heitor Almeida, Alan Auerbach, Jennifer Blouin, Tom Brennan, Alex Brill, Robin Greenwood, Michelle Hanlon, Jim Hines, David Weisbach, Rohan Williamson, Jeff Wurgler, Bill Zeile, and seminar and conference participants at Harvard, MIT, the NBER, the National Tax Association, Rutgers, and the University of North Carolina Tax Symposium for helpful comments and suggestions. Foley thanks the Division of Research of the Harvard Business School for financial support. First draft: September, 2008. The views expressed herein are those of the author(s) and do not necessarily reflect the views of the National Bureau of Economic Research. 06/17/09 Kristin Forbes http://ilp-www.mit.edu/display_paper.a4d?paperId=83932 Homogeneous nucleation of the crystal phase in n-octane melts was studied by molecular simulation with a realistic, united-atom model for n-octane. The structure of the crystal phase and the melting point of n-octane were determined through molecular dynamics simulation and found to agree with experimental results. Molecular dynamics simulations were performed to observe the nucleation events at constant pressure and constant temperature corresponding to about 20% supercooling. Umbrella sampling Monte Carlo simulations were used to calculate the nucleation free energy for three temperatures, ranging from 8% to 20% supercooling, and to reveal details of the critical nucleus for the first time. The cylindrical nucleus model was found to provide a better quantitative description of the critical nucleus than the spherical nucleus model. The interfacial free energies of the cylinder model were calculated from the simulation data. As the temperature increased, the interfacial free energy of the side surface remained relatively unchanged, at about 8 mJ/m2, whereas the interfacial free energy of the end surface decreased significantly from 11.4 mJ/m2 to about 2 mJ/m2. These results, and the methods employed, provide valuable and quantitative information regarding the rate limiting step during the solidification of chain molecules, with ramifications for both short alkanes and polymers. 06/16/09 Gregory Rutledge http://ilp-www.mit.edu/display_paper.a4d?paperId=83931 The origin and scaling of the current measured during steady electrospinning of polymer solutions in organic solvents is considered. For a specified electric field strength, E, flow rate, Q, and conductivity, K, the total measured current is shown empirically to scale as ITOTAL ~EQ0.5K0.4, for a wide variety of polymer solutions with different electrical conductivities. It is also shown that ITOTAL is composed of two distinct components, one that varies linearly with E, and another that is independent of E, but varies with the conductivity, K, of the fluid and the flow rate Q. The experimental evidence suggests that the latter component arises due to a secondary electrospray emanating from the surface of the jet. The consequence of this secondary electrospray mechanism on the final fiber size achieved during the electrospinning process is also discussed. 06/16/09 Gareth McKinley http://ilp-www.mit.edu/display_paper.a4d?paperId=83930 Boron Neutron Capture Therapy (BNCT) effectiveness depends on the preferential sequestration of boron in cancer cells relative to normal tissue cells. We present a novel strategy for sequestering boron using an adenovirus expressing the sodium iodide symporter (NIS). Human glioma grown subcutaneously in athymic mice and orthotopic rat brain tumors were transfected with NIS using a direct tumor injection of adenovirus. Boron bound as sodium tetrafluoroborate (NaBF4) was administered systemically several days after transfection. Tumors were excised hours later and assessed for boron concentration using inductively coupled plasma atomic emission spectroscopy. In the human glioma transfected with NIS, boron concentration was more than 10 fold higher with 100 mg/kg of NaBF4, compared to tumor not transfected. In the orthotopic tumor model, the presence of NIS conferred almost 4 times the boron concentration in rat tumors transfected with human virus compared with contralateral normal brain not transfected. We conclude that adenovirus expressing NIS has the potential to achieve tumor boron levels in the range suitable for therapy. The new delivery approach should be explored for future therapeutic applications. 06/16/09 Peter Binns