A probabilistic constrained nonlinear optimization framework to optimize RED parameters

The random early detection (RED) technique has seen a lot of research over the years. However, the functional relationship between RED performance and its parameters viz,, queue weight (omega(q)), marking probability (max(p)), minimum threshold (min(th)) and maximum threshold (max(th)) is not analytically availa ble. In this paper, we formulate a probabilistic constrained optimization problem by assuming a nonlinear relationship between the RED average queue length and its parameters. This problem involves all the RED parameters as the variables of the optimization problem. We use the barrier and the penalty function approaches for its Solution. However (as above), the exact functional relationship between the barrier and penalty objective functions and the optimization variable is not known, but noisy samples of these are available for different parameter values. Thus, for obtaining the gradient and Hessian of the objective, we use certain recently developed simultaneous perturbation stochastic approximation (SPSA) based estimates of these. We propose two four-timescale stochastic approximation algorithms based oil certain modified second-order SPSA updates for finding the optimum RED parameters. We present the results of detailed simulation experiments conducted over different network topologies and network/traffic conditions/settings, comparing the performance of Our algorithms with variants of RED and a few other well known adaptive queue management (AQM) techniques discussed in the literature.

Vertical structure and horizontal gradients of aerosol extinction coefficients over coastal India inferred from airborne lidar measurements during the Integrated Campaign for Aerosol, Gases and Radiation Budget (ICARB) field campaign

Quantitative estimates of the vertical structure and the spatial gradients of aerosol extinction coefficients have been made from airborne lidar measurements across the coastline into offshore oceanic regions along the east and west coasts of India. The vertical structure revealed the presence of strong, elevated aerosol layers in the altitude region of similar to 2-4 km, well above the atmospheric boundary layer (ABL). Horizontal gradients also showed a vertical structure, being sharp with the e(-1) scaling distance (D-0H) as small as similar to 150 km in the well-mixed regions mostly under the influence of local source effects. Above the ABL, where local effects are subdued, the gradients were much shallower (similar to 600-800 km); nevertheless, they were steep compared to the value of similar to 1500-2500 km reported for columnar AOD during winter. The gradients of these elevated layers were steeper over the east coast of India than over the west coast. Near-simultaneous radio sonde (Vaisala, Inc., Finland) ascents made over the northern Bay of Bengal showed the presence of convectively unstable regions, first from surface to similar to 750-1000 m and the other extending from 1750 to 3000 m separated by a stable region in between. These can act as a conduit for the advection of aerosols and favor the transport of continental aerosols in the higher levels (> 2 km) into the oceans without entering the marine boundary layer below. Large spatial gradient in aerosol optical and hence radiative impacts between the coastal landmass and the adjacent oceans within a short distance of < 300 km (even at an altitude of 3 km) during summer and the premonsoon is of significance to the regional climate.

Axially homogeneous, zero mean flow buoyancy-driven turbulence in a vertical pipe

We report an experimental study of a new type of turbulent flow that is driven purely by buoyancy. The flow is due to an unstable density difference, created using brine and water, across the ends of a long (length/diameter = 9) vertical pipe. The Schmidt number Sc is 670, and the Rayleigh number (Ra) based on the density gradient and diameter is about 10(8). Under these conditions the convection is turbulent, and the time-averaged velocity at any point is `zero'. The Reynolds number based on the Taylor microscale, Re-lambda, is about 65. The pipe is long enough for there to be an axially homogeneous region, with a linear density gradient, about 6-7 diameters long in the midlength of the pipe. In the absence of a mean flow and, therefore, mean shear, turbulence is sustained just by buoyancy. The flow can be thus considered to be an axially homogeneous turbulent natural convection driven by a constant (unstable) density gradient. We characterize the flow using flow visualization and particle image velocimetry (PIV). Measurements show that the mean velocities and the Reynolds shear stresses are zero across the cross-section; the root mean squared (r.m.s.) of the vertical velocity is larger than those of the lateral velocities (by about one and half times at the pipe axis). We identify some features of the turbulent flow using velocity correlation maps and the probability density functions of velocities and velocity differences. The flow away from the wall, affected mainly by buoyancy, consists of vertically moving fluid masses continually colliding and interacting, while the flow near the wall appears similar to that in wall-bound shear-free turbulence. The turbulence is anisotropic, with the anisotropy increasing to large values as the wall is approached. A mixing length model with the diameter of the pipe as the length scale predicts well the scalings for velocity fluctuations and the flux. This model implies that the Nusselt number would scale as (RaSc1/2)-Sc-1/2, and the Reynolds number would scale as (RaSc-1/2)-Sc-1/2. The velocity and the flux measurements appear to be consistent with the Ra-1/2 scaling, although it must be pointed out that the Rayleigh number range was less than 10. The Schmidt number was not varied to check the Sc scaling. The fluxes and the Reynolds numbers obtained in the present configuration are Much higher compared to what would be obtained in Rayleigh-Benard (R-B) convection for similar density differences.

First principles study of magnetism in divalent Eu perovskites

A comparative first principles study has been carried out for EuLiH3 (ELH) and EuTiO3 (ETO) using the generalized gradient approximation +U approach. While ELH exhibits ferromagnetic ground state for all volumes, the magnetic ground state of ETO has the tendency to switch from G-type antiferromagnetic to a ferromagnetic state with change in volume. The marked difference in magnetic behavior and magnitude of the nearest neighbors exchange interaction of both the compounds are shown to be related to the difference in their respective electronic structure near the Fermi level. The Ti 3d states are shown to play predominant role in weakening the strength of the exchange interaction in ETO.

Unsteady incompressible boundary layer flow of a micropolar fluid at a stagnation point

The flow, heat and mass transfer on the unsteady laminar incompressible boundary layer in micropolar fluid at the stagnation point of a 2-dimensional and an axisymmetric body have been studied when the free stream velocity and the wall temperature vary arbitrarily with time. The partial defferential equations governing the flow have been solved numerically using a quasilinear finite-difference scheme. The skin friction, microrotation gradient and heat transfer parameters are found to be strongly dependent on the coupling parameter, mass transfer and time, whereas the effect of the microrotation parameter on the skin friction and heat transfer is rather weak, but microrotation gradient is strongly affected by it. The Prandtl number and the variation of the wall temperature with time affect the heat-transfer very significantly but the skin friction and micrortation gradient are unaffected by them.

Auxiliary coils for generating magnetic field gradients for a Faraday magnetometer

A set of coils has been designed and constructed for generating magnetic field gradients for a Faraday magnetometer. We have obtained a gradient of magnitude -1 1 kOe m-' (8.75 x lo5 A m-') in an air gap of 42 mm for a current of 12 A passing through the coils.

Zeeman effect of chlorine nuclear quadrupole resonance and hydrogen bonding in 2,6-, 2,5- and 3,5-dichlorophenols

The Zeeman effect of chlorine nuclear quadrupole resonance in polycrystalline samples of 2,6-, 2,5 and 3,5-dichlorophenol has been investigated at room temperature in order to study the effect of hydrogen bonding on the electric field gradient asymmetry parameter n. While the two n.q.r. lines in 3,5-dichlorophenol gave an asymmetry parameter of 10%, those in 2,6- and 2,5-dichlorophenol gave different values of n for the two chlorines. The chlorine atom which is ortho to the OH group and involved in hydrogen bonding (i.e., corresponding to the low frequency line) gave an asymmetry parameter of 0.21 in 2,6-dichlorophenol and 0.17 in 2,5-dichlorophenol while the other chlorine (i.e., corresponding to the high frequency line) gave a lower value of 0.12 in 2,6-dichlorophenol and 0.11 in 2,5-dichlorophenol. These values of n are discussed in terms of hydrogen bonding and bond parameters.

A Fuzzy Approximation Scheme for Sequential Learning in Pattern Recognition

An adaptive learning scheme, based on a fuzzy approximation to the gradient descent method for training a pattern classifier using unlabeled samples, is described. The objective function defined for the fuzzy ISODATA clustering procedure is used as the loss function for computing the gradient. Learning is based on simultaneous fuzzy decisionmaking and estimation. It uses conditional fuzzy measures on unlabeled samples. An exponential membership function is assumed for each class, and the parameters constituting these membership functions are estimated, using the gradient, in a recursive fashion. The induced possibility of occurrence of each class is useful for estimation and is computed using 1) the membership of the new sample in that class and 2) the previously computed average possibility of occurrence of the same class. An inductive entropy measure is defined in terms of induced possibility distribution to measure the extent of learning. The method is illustrated with relevant examples.

Co-induction of sucrose transport and invertase activities in a thermophilic fungus,Thermomyces lanuginosus

The incorporation of sucrose into the thermophilic fungus,Thermomyces lanuginosus, occurred only in mycelia previously exposed to sucrose or raffinose. Sucrose uptake and invertase were inducible. Both activities appeared in sucrose-induced mycelia at about the same time. Both activities declined almost simultaneously following the exhaustion of sucrose in the medium. The sucrose-induced uptake system was specific for \beta -fructofuranosides as revealed by competition with various sugars. The induction of sucrose uptake system was blocked by cycloheximide, showing that it was dependent on new protein synthesis. Transport of sucrose did not seem to be dependent on ATP. Rather, uptake of this sugar seemed to be driven by a proton gradient across the plasma membrane. The uptake system showed Michaelis-Menten kinetics.

Bull sperm plasma and acrosomal membranes: Fluorescence studies of lipid phase fluidity

Bull sperm plasma and outer acrosomal membranes have been isolated by discontinuous sucrose density gradient centrifugation and Ca2+-ATPase activity has been determined for both the membranes. Pyrene excimer fluorescence and diphenylhexatriene fluorescence polarization studies show that the lipid phase of the sperm plasma membranes is more fluid than the lipids of the outer acrosomal membranes. Approximately, a three fold increase in the cholesterol content has been found in the outer acrosomal membranes as compared to that in the plasma membranes.

Regional differentiationnext term in previous termbull sperm plasma membranesnext term

previous termBull spermnext term heads and tails have been separated by proteolytic digestion (trypsin) and previous termplasma membranesnext term have been isolated, using discontinuous sucrose density gradient centrifugation. previous termPlasma membranenext term bound Ca2+-ATPase is shown to be associated mostly with the tail previous termmembranes.next term Pyrene excimer fluorescence and diphenylhexatriene fluorescence polarization experiments indicate a more fluid lipid phase in the tail region. Differences in surface charge distribution have been found, using 1-anilinonaphthalene-8-sulfonate and Tb3+ as fluorescent probes.

Unsteady Incompressible Two-Dimensional and Axisymmetric Turbulent Boundary Layer Flows

The unsteady turbulent incompressible boundary-layer flow over two-dimensional and axisymmetric bodies with pressure gradient has been studied. An eddy-viscosity model has been used to model the Reynolds shear stress. The unsteadiness is due to variations in the free stream velocity with time. The nonlinear partial differential equation with three independent variables governing the flow has been solved using Keller's Box method. The results indicate that the free stram velocity distribution exerts strong influence on the boundary-layer characteristics. The point of zero skin friction is found to move upstream as time increases.

Experimental evidence of the kinematic Cosserat effect in dense granular flows

To capture shear localization in the flow of dense granular materials in a continuum description, it has earlier been proposed that granular materials be treated as Cosserat, or micropolar, continua. Here, we provide experimental verification of the kinematic Cosserat effect, or the deviation of the particle spin from the material spin induced by the velocity gradient. Contrary to earlier belief, we find this effect to be sizable even outside the shear layers. Remarkably, the particles and material elements spin in opposite directions in flow through a hopper.

Invited paper Application of the least mean fourth (LMF) adaptive algorithm to signals associated with gaussian noise

This paper considers the applicability of the least mean fourth (LM F) power gradient adaptation criteria with 'advantage' for signals associated with gaussian noise, the associated noise power estimate not being known. The proposed method, as an adaptive spectral estimator, is found to provide superior performance than the least mean square (LMS) adaptation for the same (or even lower) speed of convergence for signals having sufficiently high signal-to-gaussian noise ratio. The results include comparison of the performance of the LMS-tapped delay line, LMF-tapped delay line, LMS-lattice and LMF-lattice algorithms, with the Burg's block data method as reference. The signals, like sinusoids with noise and stochastic signals like EEG, are considered in this study.

Numerical studies on columnar-to-equiaxed transition in directional solidification of binary alloys

A numerical study on columnar-to-equiaxed transition (CET) during directional solidification of binary alloys is presented using a macroscopic solidification model. The position of CET is predicted numerically using a critical cooling rate criterion reported in literature. The macroscopic solidification model takes into account movement of solid phase due to buoyancy, and drag effect on the moving solid phase because of fluid motion. The model is applied to simulate the solidification process for binary alloys (Sn-Pb) and to estimate solidification parameters such as position of the liquidus, velocity of the liquidus isotherm, temperature gradient ahead of the liquidus, and cooling rate at the liquidus. Solidification phenomena under two cooling configurations are studied: one without melt convection and the other involvin thermosolutal convection. The numerically predicted positions of CET compare well with those of experiments reported in literature. Melt convection results in higher cooling rate, higher liquidus isotherm velocities, and stimulation of occurrence of CET in comparison to the nonconvecting case. The movement of solid phase aids further the process of CET. With a fixed solid phase, the occurrence of CET based on the same critical cooling rate is delayed and it occurs at a greater distance from the chill.