LIGO Summer Undergraduate Research Fellowship Project Proposals and Abstracts
Summer 2014
Draft: February 12, 2014

AT CALTECH:

__________________________________________________________________________
Investigating Mechanical Upconversion noise in Steel

Mentors: Gabriele Vajente, Eric Quintero and Rana Adhikari
Abstract: Elastic materials such as metals that can be described with very good accuracy with linear models. However, material defects can introduce non-linear behavior that can cause excess noise. One important case is the displacement noise generated by the motion of crystal dislocations (crackling) which can be triggered by the external stress applied to the material. A low frequency variation of the stress can modulate the noise level at much higher frequencies. Such effect may be relevant in the suspension system for the advanced gravitational wave detectors being built. Based on a seismically isolated in-vacuum laser interferometer, a dedicated experiment has been built to try to measure this crackling noise. The student will contribute to the tuning of the suspension system and of the laser control, learning the basics of feed-back control and interferometry, which are the basis of the present gravitational wave detector operations. Ze will also contribute to the analysis of the results, with the goal of measuring the level of crackling noise as a function of the stress for different materials.
[1] http://www.nature.com/nature/journal/v410/n6825/full/410242a0.html

__________________________________________________________________________
Global Optimization of the interferometers

Mentors: Gabriele Vajente, Jenne Driggers
Abstract: In a Michelson interferometer the sensitivity to the differential displacement can be limited by many sources of noise. The experiment we are currently operating is a in-vacuum, seismically isolated Michelson interferometer with balanced output ports. The main sensitivity limitations are seismic noise and laser frequency noise. The coupling of these noises to the output signals depends on parameters of the system that can be controlled: the alignment of the optical elements, the difference between the two arm lengths, the seismic isolation control system, etc. Normally these parameters are tuned by hand in a trial-and-error way. The goal is to develop an automatic global optimization algorithm which is capable of finding the best parameters for the actual experiment. As a first step a simulation of the system and of the algorithm will be created. When the algorithm is validated, the student will implement it in the real experiment to improve its sensitivity.

__________________________________________________________________________
A Fiber Optic System for Automatic System Identification of the Frequency Response of the Photodetectors in an Interferometric Gravitational Wave Detector

Mentors: Eric Gustafson, Jameson Rollins and Rana Adhikari
Abstract: Before you can build an Interferometric Gravitational Wave Detector that works at the limits set by Quantum Mechanics it is necessary to first build a Detector that you can control and read out optically. There are several photodiodes in such a detector used to sense various degrees of freedom of the system to provide feedback signals for the control of the detector. In addition there is the main interferometric Gravitational Wave Signal that is read out with a photodiode. In this precision physics experiment it is necessary to treat the photodiode and its readout electronics as systems whose performance including the frequency response can change over time and with changing operating conditions. In this project we will build an automatic frequency response measurement system for the gravitational wave detector photodiodes. This system will use a modulated diode laser coupled through a fiber optic distribution system to illuminate the photodiodes and then the computer control system will measure the frequency response of the photo-receivers sequentially using a network analyzer. This experiment will involve hands-on work with lasers, precision mechanical systems, optical equipment, fiber optic components, digital controls and a network analyzer. Skills in python, Matlab, electronics, and laser optics would be helpful.

__________________________________________________________________________
Thermal noise in Ultra-Stable Fabry-Perot cavities

Mentors: Evan Hall, Tara Chalermsongsak, Rana Adhikari
Abstract: Brownian thermal noise from silica/tantala dielectric coatings on mirrors is a limiting noise source in LIGO (as well as other precision experiments) in the most sensitive band (50-300 Hz). However, the theoretical calculation of coating noise at such frequency has yet to be verified. We aim to use ultra stable dual Fabry-Perot cavities in order to measure the coating noise. In this experiment, the student will learn how to suppress various technical noise e.g. ambient thermal fluctuation, seismic noise, random photon absorption by employing feedback control system as well as general laboratory skill in optical Physics.
[1] Phys. Rev. Lett. 93, 250602 (2004) (http://link.aps.org/doi/10.1103/PhysRevLett.93.250602)

__________________________________________________________________________
Cryogenic Silicon Cavities: Laser Stabilization and Macroscopic Quantum Mechanics

Mentors: Nic Smith, David Yeaton-Massey, Rana Adhikari
Abstract: Fabry-Perot cavities with exceptional length stability are essential tools in the frequency stabilization of lasers for precision spectroscopy, optical atomic clocks, quantum optics experiments, and interferometric gravitational wave detectors. The stability of the cavity is ultimately limited by a multitude of noise sources, including thermal noise of the mirror coatings, vibration induced elastic deformation, laser power fluctuations, and temperature fluctuations of the cavity itself. In this project, we will be pushing towards the theoretical limitations on the performance of a pair of 4 inch silicon cavities at cryogenic temperatures (120K), and investigating whatever noise sources limit us.
[1] Phys. Rev. Lett. 105, 070403 (2010)

__________________________________________________________________________
Kalman Filter based state estimation for Thermal Adaptive Optics

Mentors: Aidan Brooks
Abstract: Absorption induced thermal distortions in the optics of next generation interferometers will diminish the overall performance of these devices. In Advanced LIGO, a thermal compensation system (TCS) is employed to correct thermal lensing. For both 2nd and 3rd generation devices, we must model and understand the 2D thermal lenses in conjunction with our correction system and we must develop a system to determine the optimum compensation to apply. FEA Modeling, PDE solving, and strong computing skills are recommended.
[1] http://dx.doi.org/10.1109/PACC.2011.5978971

__________________________________________________________________________
Mechanical Quality Factor of Cryogenic Silicon Ribbons

Mentors: Nicolas Smith-L
Abstract: Advanced LIGO will be limited by thermal noise in a large frequency band where gravitational wave signals are expected to exist. A large contribution to thermal noise is caused by internal friction of the mirror and suspension elements. To further push down the contribution of thermal noise, future detectors will require new materials with extremely high mechanical quality. Silicon at cryogenic temperatures shows the promise to provide the required mechanical quality, though research is required to determine the effects how surface etching on mechanical quality. The student will use finite element modeling to predict the mechanical properties of our samples, and will learn silicon etching techniques and perform mechanical quality factor (Q) measurements on silicon samples at cryogenic temperatures using the resonant ringdown technique.
[1] http://iopscience.iop.org/0264-9381/31/2/025017/

__________________________________________________________________________
Design of a coating-less reference cavity with total internal reflection

Mentors: Matt Abernathy, Koji Arai
Abstract: Thermal noise is considered one of the fundamental noise sources in optomechanics and laser frequency stabilization using an external reference cavity[1]. In particular, thermal noise associated with the reflective coatings can be a noise source that limits the stability of the reference cavity. It is also considered to be a sensitivity limit in second-generation interferometer gravitational wave detectors.
One of the ideas to resolve the coating thermal-noise issue is to eliminate the coatings and instead utilizing total internal reflection (TIR) for the mirror surfaces[2]. A monolithic, polished substrate is used as a reference cavity while the input and output beams interact with the cavity via an evanescent coupling, in a process called Frustrated Total Internal Reflection (FTIR). Improvement in thermal noise may be achieved by this technique as the other thermal noises (Brownian, thermo-elastic, and thermo-refractive noises) of the cavity substrate can be reduced or canceled out by choosing the material, geometry, and operating temperature of the cavity body.
In this project, we search for an optimal design of the TIR reference cavity for thermal noise reduction. We plan to use COMSOL and MATLAB as the analysis environment. The student should have completed courses on mechanics and thermodynamics.
[1] Physical Review Letters 93, 250602 (2004)
[2] Optics Letters 17, 378 (1992)

__________________________________________________________________________
Automatic Alignment of High Finesse Optical Ring Cavities

Mentors: Nicolas Smith-L, Gabriele Vajente
Abstract: The 40m Prototype interferometer has a suspended Fabry-Perot optical filter cavity known as the mode cleaner which acts to condition the laser beam. It is important that the laser is well aligned with the mode cleaner cavity at all times. The student will implement the reflection phase heterodyne wavefront sensing alignment technique for designing and commissioning a system for automatically maintaining optimal alignment of the laser beam with the suspended mode cleaner optical cavity at the 40m Prototype. The student will make a comprehensive noise model of the mode cleaner including radiation pressure effects with the goal of maintaining alignment fluctuations below 10 picoradians. This project will include techniques related to radio frequency (RF) electronics, laser optics, feedback control systems, and computer modeling in MATLAB. Experience in any of these areas is desirable but not required.
[1] Applied Optics, Vol. 33, Issue 22, pp. 5041-5049 (1994) (http://dx.doi.org/10.1364/AO.33.005041)

__________________________________________________________________________
Phase Locked Loop for Dual Wavelength Laser Stabilization

Mentors: Manasa T., Eric Quintero and Koji Arai
Abstract: The Arm Length Stabilization system of LIGO uses auxiliary lasers which are frequency doubled NPROs locked to both the arm cavities. The transmitted auxiliary light from the arm cavities are then combined with the frequency-doubled main pre-stabilized laser(PSL) to sense the arm cavity lengths independent of the rest of the interferometer. The project will involve two students working at LIGO 40m prototype designing a phase-locked-loop (PLL) to lock the frequency of the auxiliary NPRO with the PSL using the beat note frequency detected at an RF photodetector. The PLL will drive the temperature of the auxiliary NPRO via a digital PID loop. A remote readout for the beat frequency and amplitude will also be implemented. Prior hands-on-experience in laser optics, feedback control systems and electronics will be useful.
[1] http://dx.doi.org/10.1364/AO.18.003165

__________________________________________________________________________
Automatic System Identification of the Photodetectors in a Gravitational Wave Detector

Mentors: Eric Gustafson, Jameson Rollins
Abstract: Before you can build an Interferometric Gravitational Wave Detector that works at the limits set by Quantum Mechanics it is necessary to first build a Detector that you can control and read out optically. There are several photodiodes in such a detector used to sense various degrees of freedom of the system to provide feedback signals for the control of the detector. In addition there is the main interferometric Gravitational Wave Signal that is read out with a photodiode. In this precision physics experiment it is necessary to treat the photodiode and its readout electronics as systems whose performance including the frequency response can change over time and with changing operating conditions. In this project we will build an automatic frequency response measurement system for the gravitational wave detector photodiodes. This system will use a modulated diode laser coupled through a fiber optic distribution system to illuminate the photodiodes and then the computer control system will measure the frequency response of the photo-receivers sequentially using a network analyzer. This experiment will involve hands-on work with lasers, precision mechanical systems, optical equipment, fiber optic components, digital controls and a network analyzer. Skills in python, Matlab, electronics, and laser optics would be helpful.

__________________________________________________________________________
Testing the strong field dynamics of general relativity using compact binary systems

Mentors: Tjonnie Li, Alan Weinstein
Abstract: Collisions of stellar mass compact objects, such as neutron stars or black holes, are among the most promising sources of gravitational waves for Advanced LIGO. These systems have strong gravitational fields and are highly relativistic, making them excellent laboratories for testing the strong field dynamics of general relativity.
In this project, we explore the ability of Advanced LIGO to constrain alternative theories of general relativity. In particular, we simulate gravitational wave signals from various alternative theories, and quantify the sensitivity of Advanced LIGO to these deviations from general relativity.
For this project, experience with computer programming and Unix-like environments is highly recommended. Moreover, understanding of general relativity and statistics is preferred. In turn, students will learn about the generation of gravitational waves by compact binary systems, alternative theories of general relativity, and advanced statistical tools to extract information from tiny signals buried deep in the noise.

__________________________________________________________________________
Characterization of hardware injections in LIGO data

Mentors: Jonah Kanner, Alan Weinstein
Abstract: LIGO data includes simulated astrophysical signals that are added to the detectors by physically moving the test masses. These "hardware injections" appear in the data at labeled times exactly as real astrophysical signals, and so represent the best end-to-end consistency checks of LIGO calibration, data processing, and signal identification and characterization.  We are preparing the latest LIGO data for public release, and wish to include a complete record of the included hardware injections.  The student will write software to implement a matched filter search for hardware injections, and attempt to detect these signals in the data, as well as characterize their basic parameters.  The student will gain experience with gravitational wave data analysis, signal processing, and programming. Some experience programming with Python is a plus.


__________________________________________________________________________
Extracting physics from the stochastic gravitational wave background

Mentors: Eric Thrane, Tjonnie Li
Abstract: A gravitational-wave background is expected to arise from the superposition of many gravitational-wave signals, which are too weak to detect individually, but which combine to create a "stochastic" gravitational-wave glow. By measuring the stochastic background, we can probe a wide range of interesting science, from neutron stars to the inflationary epoch shortly after the Big Bang.
In this project, we will study the stochastic gravitational wave background that we hope to detect with the Advanced LIGO-Virgo detector network. In particular, we will develop software to analyse data from the Advanced LIGO-Virgo network, and infer the characteristics of the stochastic gravitational wave background.
For this project, experience with computer programming, Unix-like environments, and statistics is desirable. In turn, students will learn about the broad spectrum of sources that contribute to the stochastic gravitational wave background, and advanced statistical tools to extract information from weak signals buried deep in the noise.

__________________________________________________________________________
Cutting Edge Computing for the Extraction of Astrophysical Parameters from Gravitational-Wave Observations.

Mentors: Kent Blackburn, Vivien Raymond, Rory Smith.
Abstract: The detection of gravitational waves will usher in a new era in astrophysics and astronomy. After the eventual detection of gravitational waves we will want to infer the rich astrophysical information that they contain about the sources which emitted them; a field known as parameter estimation. Coalescing compact binary sources, consisting of neutron stars and/or black holes, are prime scientific targets for astrophysics with gravitational waves because of the detailed families of model available, and the relatively high expected detection rates. With parameter estimation we will be able to estimate the masses and spins of the compact objects, as well as their location in the universe. Based on those models, a complete inference software has been developed with in the LIGO-Virgo Collaboration, seevhttp://arxiv.org/pdf/1201.1195.pdf and reference therein. This tool's speed has been recently greatly improved using more efficient algorithms, see http://arxiv.org/pdf/1304.0462v2.pdf. To be optimal, these algorithms need to take advantage of specialized highly parallel hardware such as Many Integrated Core processors available on XSEDE (Extreme Science and Engineering Discovery Environment), the National Science Foundation computational system. 
The student will utilize these computational resources to contribute to ongoing efforts to enable the rapid extraction of astrophysical information from gravitational waves. The project will also involve managing virtual imaging to provide portability across time sharing computing resources and robustness using virtual imaging as a cutting edge check pointing infrastructure. All those cutting-edge computing application to parameter estimation will allow the exploration of previously intractable physical questions on the measurement capability of the advanced gravitational-wave detectors.
This project will develop experience and skills in the C programming language, parallel computing and the Linux/Unix environment. The student will gain experience in Bayesian inference, sampling techniques and high-performance computing (all ubiquitous in science).
__________________________________________________________________________
Searching for gravitational waves from the coalescence of high mass black hole binaries. 

Mentors: Alan Weinstein, Tjonnie Li, Surabhi Sachdev
Abstract: We aim to detect gravitational wave signals from the coalescence (inspiral, merger and final black hole ringown) of compact binary systems (neutron stars and/or black holes) with data from the advanced detectors (LIGO, Virgo, KAGRA). The merger signal from the coalescence of Low-mass systems (binary neutron stars) tends to lie above the LIGO frequency band; only the inspiral phase is detectable. For higher mass systems (involving black holes, each of mass greater than 5 solar masses), the merger and final ringdown are also detectable. We search for these signals using analysis pipelines which filter all the data, identify "triggers" of interest, form coincident triggers between multiple detectors in the network, and attempt to optimally separate signal from detector background noise fluctuations. The size of these noise fluctuations in the advanced detectors is currently unknown. We use simulated signal injections to evaluate the sensitivity of the search pipeline. The analysis pipeline has numerous parameters that can be tuned to improve the sensitivity. In this project, we will run high-statistics simulations to evaluate the search sensitivity as the analysis parameters are tuned, to arrive at optimal settings under different anticipated noise fluctuation conditions. 
This project will develop experience and skills in statistical analysis, high throughput computing and the Linux/Unix environment. The student will learn about the physics and astrophysics of compact binary coalescence, and gain experience with modern analysis techniques with large data samples.

__________________________________________________________________________
Improving the Detection Rate of Gravitational Waves from Coalescing Binary Black Holes with a Template Bank Consistency Test

Mentor: Stephen Privitera
Binaries consisting of combinations of neutron stars and black holes are expected to radiate so strongly in gravitational waves that they eventually coalesce into a single black hole. These systems involve incredibly strong and rapidly changing gravitational fields and are not only a natural laboratory for testing General Relativity, but serve as prime targets for gravitational wave astronomy. Even for these strongly gravitating systems, their signals hide beneath the detector noise and the search for their signals in LIGO data requires sensitive techniques to separate the background from the actual signal. In this project, we will explore the use of a new background discrimination statistic (a template bank chisq) for the search of gravitational wave signals from coalescing binary black holes. If successful, the use of this technique could significantly increase the detection rate of these systems in the upcoming Advanced LIGO observations.

__________________________________________________________________________
Identifying Glitches in the Advanced LIGO detector subsystems.

Mentors: Alan Weinstein, Vivien Raymond
Abstract: The search for transient gravitational wave signals from astrophysical sources such as the merger of compact binary systems, the core collapse of massive stars, or cosmic string cusps is greatly complicated by the presence of large non-Gaussian fluctuations (glitches) in the detector noise. These can arise from multiple instrumental sources in the complex advanced LIGO detectors, including the pre-stabilized laser, the mirror seismic isolation or suspension systems, the detector control system, scattered light, or the physical environment. In this project, the student will develop and run analysis procedures to identify strongly non-Gaussian glitches in data from the advanced LIGO detector subsystems, and correlate them with glitches in the gravitational wave data stream. 
This project will develop experience and skills in statistical analysis, high throughput computing and the Linux/Unix environment. The student will learn about the advanced LIGO detector subsystems, and gain experience with modern analysis techniques with large data samples.
  
__________________________________________________________________________
Measuring the stochastic gravitational-wave background in the presence of correlated noise

Mentor: Eric Thrane
A gravitational-wave background is expected to arise from the superposition of many gravitational-wave signals, which are too weak to detect individually, but which combine to create a "stochastic" gravitational-wave glow.  By measuring the stochastic background, we can probe a wide range of interesting science, from neutron stars to the inflationary epoch shortly after the Big Bang.  Studies are planned or underway in order to realize the full potential of Advanced LIGO stochastic analyses, and interested students can contribute to this effort in a number of ways.  One project of particular interest involves developing strategies to deal with correlated noise at widely separated detectors (created by the resonances in the Earth's magnetic field), which can masquerade as apparent gravitational-wave signals.  Students will gain expertise in programming, statistics, and signal processing while making an important contribution to Advanced LIGO science.  Experience with matlab is a plus.

__________________________________________________________________________
Estimating the parameters of a population of coalescing compact binaries

Mentors: Larry Price, Vivien Raymond and Jonah Kanner
The aLIGO network stands to make hundreds of detections over the lifetime of the project.  While there is much to be learned from the parameters of single events, the parameter distribution of the population of events is also of astrophysical interest.  The goal of this project is to develop the tools for estimating such population distributions and investigating the requirements on the number and quality of events to make useful inferences.
The successful applicant will gain a working knowledge of Bayesian inference, the astrophysics of compact binaries, and python programming in a UNIX environment.

__________________________________________________________________________
Modeling and minimizing errors in the Advanced LIGO calibration system.

Mentors: Alan Weinstein, Jamie Rollins
Abstract: The response of an advanced LIGO detector to a gravitational wave is encoded in digitized electro-optic signals from the detector; these must then be converted to gravitational wave strain through a calibration procedure. In Initial LIGO, the calibration procedure achieved an accuracy on the order of 10%. We aim for much better accuracy for advanced LIGO. Also, the detector response is significantly more complex for advanced LIGO because of the more complex optical configuration of the detectors. We will develop and work with computer models and simulations to quantify the accuracy of the calibration procedure.
 
__________________________________________________________________________
Numerical simulations of black-hole binaries.

Mentors: Mark Scheel, Bela Szilagyi
Abstract: Numerical models of gravitational-wave sources are important both for improving detection algorithms for LIGO, and for understanding and interpreting future LIGO detections. In this project, we will use the parallel computer code "SpEC" to solve Einstein's equations for colliding black holes and to compute the resulting waveforms.  In particular we will concentrate on cases (such as large-kick configurations) where small changes in initial parameters (masses and spins of the black holes) might lead to large differences in the waveforms.  The waveforms will be used as part of an ongoing effort to produce a fast, accurate model for black-hole binary waveforms for use with LIGO.  This project will develop experience and skills in high-performance parallel computing, the C++ programming language, and the Linux/Unix programming environment.  The student will learn about numerical methods, black hole physics, and phenomenology of black-hole binaries.

__________________________________________________________________________
Dynamics and Gravitational Wave Signatures of Magnetized Neutron Stars

Mentors: A. Nishizawa and Y. Chen
Abstract: Advanced LIGO is expected to make the first gravitational wave detection in the near future.  The student will look at how the magnetic field of a neutron star may influence the dynamics of a neutron-star-black-hole binary ? in which the black hole is large enough so that the neutron star does not get tidally disrupted.  The student should have taken courses on undergraduate-level general relativity. It is preferable that applicants have some experience on post-Newtonian theory and/or black hole perturbation theory.

_________________________________________________________________________
Gravitational Wave Signatures of Alternative Theories of Gravity

Mentors: A. Nishizawa and Y. Chen
Abstract: Advanced LIGO is expected to make the first gravitational wave detection in the near future.  The student will look at how gravitational wave emission and propagation can be modified by alternative theories of gravity --- and how LIGO data can be used to constrain such modifications. The student should have taken courses on undergraduate-level general relativity. It is preferable that applicants have some experience on post-Newtonian theory and/or black hole perturbation theory.

__________________________________________________________________________
__________________________________________________________________________
__________________________________________________________________________
At LIGO Livingston (Louisiana)

__________________________________________________________________________
GW Signal Calibration of the End Mirror Actuators

Mentors: Adam Mullavey, Joseph Betzwieser (LIGO Livingston Observatory)
Abstract: The Advanced LIGO interferometer has a green laser, normally used for auxiliary locking of the individual 4km long arm cavities. We are proposing an alternative use, namely making a short Michelson interferometer with one arm including the end test mass and the other reference arm located on the end station table. This will allow sensitive calibration of the end test mass actuators, increasing our chances of making gravitational wave detections and correctly estimating parameters of astrophysical sources to hunt for new physics.

__________________________________________________________________________
Modernizing the Interferometer's Control Loops

Mentors: Valera Frolov, Denis Martynov (LIGO Livingston Observatory)
Abstract: Installation of the Advanced LIGO interferometers is almost complete. Instrument commissioning is well underway at LLO. We expect the first interferometer lock in the spring 2014. The full instrument consists of two 4 km arm cavities and vertex section that includes a power recycling cavity, short Michelson interferometer and signal recycling cavity. These five major degrees of freedom are controlled to keep the interferometer in the linear regime. Beyond that Advanced LIGO has hundreds of auxiliary control loops that combine feedback and feedforward techniques. The main goal of the project is to improve current control loops to minimize noise couplings to the main gravitational wave channel. Interferometer lock can be acquired quicker and maintained longer once the control system is optimized. We plan to use modern control techniques such as H-infinity, LQR, Neural Networks and learning algorithms to achieve the goal. Experimental and analytical skills will be used. Knowledge of control theory is required.

__________________________________________________________________________
__________________________________________________________________________
__________________________________________________________________________
At LIGO Hanford (Washington)

__________________________________________________________________________
Investigations using the physical environment monitor system

Mentor: Robert Schofield (LIGO Hanford Observatory)
Abstract: Advanced LIGO will detect gravitational wave radiation by measuring test mass motions of as little as 1e-20 m/sqrt(Hz). It is important to keep motions produced by environmental signals, such as acoustic vibrations and ambient magnetic fields from overwhelming the motions produced by astrophysical signals, and of course it is important to be able to distinguish between the two. The student will join the group that is setting up the environmental monitoring system for this purpose. Tasks will include installing, calibrating and documenting seismometers, microphones, magnetometers etc. The student will also help investigate and reduce the coupling of environmental signals to the developing aLIGO interferometer.

__________________________________________________________________________
Dynamical tuning of signal recycling

Mentor: Kiwamu Izumi (LIGO Hanford Observatory)
Abstract: The signal recycling is an indispensable technique in the advanced gravitational-wave interferometers. It allows us to tune the observational bandwidth by introducing a microscopic offset in the optical length of the signal recycling cavity. As opposed to the conventional tuning in which a static offset is introduced, we study a possible application of the dynamical tuning of the offset [1] in a realistic interferometer. We will build a numerical time-domain model in order to study the feasibility and possible technical issues.
[1] D.A.Simakov, arXiv (2013) http://arxiv.org/abs/arXiv:1311.2766

_________________________________________________________________________
__________________________________________________________________________
__________________________________________________________________________
__________________________________________________________________________
__________________________________________________________________________