So-called Singular Value Decomposition Research Articles

Analysis of dynamics and spatial structure on the filament during type I ELMy H-mode using VUVI system on EAST

In type I ELMy H-mode experiment, Edge localized mode (ELM) filaments are clearly captured by the high-speed vacuum ultraviolet imaging (VUVI) system which is developed on the Experimental Advanced Superconducting Tokamak. To analyze the ELM filament structures, the so-called singular value decomposition is performed on the imaging data to extract the key fluctuating components. In this work, the filament structure is characterized by the pitch angle and poloidal width. In a single ELM crash, it is found that the poloidal width increases (decreases) in the rise (decay) phase of the VUVI intensity induced by ELM crash. The pitch angle derived from the VUVI data agrees well with that calculated by the Equilibrium FITting code, indicating the filaments are aligned with the field lines. The poloidal velocity shows no obvious change during the rise and decay phases in an ELM crash. In addition, both the poloidal width and the poloidal velocity of the filament increase with the heating power. Since the filament structures are extracted from the line-integrated imaging data, all these results are obtained on the condition that the ELMs are confined to a narrow layer in the plasma.

Observation of filament-like structures in ELMy H-mode plasma with a VUV imaging system developed on the EAST tokamak

On the EAST tokamak, filament-like structures have been observed in ELMy H-mode discharges with a high-speed vacuum ultraviolet (VUV) imaging system. The topos, chronos and their weight can be obtained simultaneously by performing the so-called singular value decomposition (SVD) analysis of raw VUV imaging data. The fluctuation amplitude is observed to be suppressed and enhanced gradually in the edge localized mode (ELM) crash and pedestal recovery phase in the chronos, respectively, while filament-like structures can only be found in the pedestal recovery phase on the topos. The mode structure, i.e. m/n = 36/9 (m and n are the poloidal and toroidal mode number, respectively) with = 0.95, = 0.07 ( and denote the mode location and mode width, respectively) is derived by a comparison of the synthetic images and the experimental imaging data.

Adaptive Per-spatial Stream Power Allocation Algorithms for Single-User MIMO-OFDM Systems

This paper presents adaptive per-spatial stream power allocation algorithms for Single User Multiple-Input Multiple-Output Orthogonal Frequency Division Multiplexing (SU MIMO-OFDM) systems. Three efficient and low-complexity Greedy Power Allocation (GPA) algorithms are proposed to maximize the throughput and spectral efficiency of the SU MIMO-OFDM systems. Firstly, the low-complexity pre-coded GPA algorithms are developed for the MIMO systems. The spatial sub-channels are created by applying the so-called Singular Value Decomposition (SVD) technique on the MIMO channel matrix, and then the Pre-GPA algorithms are applied to exploit the multi-path and spatial diversities. Secondly, the spatial and frequency diversities are exploited by adaptively allocating the system sub-carriers to the spatial sub-channels followed by Per-Spatial GPA (PSGPA). Finally, spatial multiplexing-based GPA algorithms are proposed to optimize the spectral efficiency of the SU MIMO-OFDM system. An optimal two-dimensional Spatial-Frequency GPA (SFGPA) algorithm is proposed to efficiently improve the average system spectral efficiency. The high computational complexity of the optimal SFGPA solution is simplified by proposing a low-complexity Per-Spatial GPA with Excess Power Moving down (PSGPA-EPMd) algorithm, which moves the per-spatial excess power downwards to enhance the spectral efficiency of the spatial multiplexing-based SU MIMO-OFDM systems. The proposed algorithms achieve better spectral efficiency and maximize the throughput in comparison with conventional algorithms.

Identification of vortexes obstructing the dynamo mechanism in laboratory experiments

The magnetohydrodynamic dynamo effect explains the generation of self-sustained magnetic fields in electrically conducting flows, especially in geo- and astrophysical environments. Yet the details of this mechanism are still unknown, e.g., how and to which extent the geometry, the fluid topology, the forcing mechanism, and the turbulence can have a negative effect on this process. We report on numerical simulations carried out in spherical geometry, analyzing the predicted velocity flow with the so-called singular value decomposition, a powerful technique that allows us to precisely identify vortexes in the flow which would be difficult to characterize with conventional spectral methods. We then quantify the contribution of these vortexes to the growth rate of the magnetic energy in the system. We identify an axisymmetric vortex, whose rotational direction changes periodically in time, and whose dynamics are decoupled from those of the large scale background flow, that is detrimental for the dynamo effect. A comparison with experiments is carried out, showing that similar dynamics were observed in cylindrical geometry. These previously unexpected eddies, which impede the dynamo effect, offer an explanation for the experimental difficulties in attaining a dynamo in spherical geometry.

AC susceptibility of magnetic markers in suspension for liquid phase immunoassay

AC susceptibility of magnetic markers in solution was studied for biosensor applications. First, frequency dependence of the susceptibility was measured, and size distribution of the markers was estimated by analyzing the experimental result with the so-called singular value decomposition (SVD) method. The size distribution estimated with the magnetic measurement agreed with that obtained from conventional optical measurement. Next, susceptibility measurement was applied to the liquid-phase immunoassay without bound/free (B/F) separation. We performed the detection of biotin-coated polymer beads in suspension using avidin-coated magnetic markers. Changes of the susceptibility and the size distribution caused by the binding reaction were shown.

Mass attenuation coefficients of the elements Ti, V, Fe, Co, Ni, Cu and Zn for the K emission lines between 4.51 and 10.98 keV

Mass attenuation coefficients of seven elements between atomic numbers 22 and 30 were measured at 20 X-ray energies between 4.5 and 11 keV. These energies correspond to fluorescence radiation excited by monochromatic Mo Kα radiation. The measurements were made on thin metal foils, which were spread flat and perpendicular to the beam scattered from the fluorescent sample. In this arrangement, the measured attenuation includes photoelectric absorption plus average elastic and inelastic scattering. The mass per unit area of a foil was determined by weighing and the impurity concentration by fluorescence analysis. The spectral lines corresponding to the components of the fluorescent radiation and those of the secondary radiation were resolved by the so-called singular value decomposition (SVD). The probable errors of the attenuation coefficients were typically 1%. The general agreement with previous experimental and theoretical values is good, but at low energies and at energies just above the absorption edges the present values are about 5% larger than theoretical values based on relativistic HFS (Hartree–Fock–Slater) calculations.