Optics: measuring and testing – Lamp beam direction or pattern
Reexamination Certificate
2000-12-15
2002-01-15
Pham, Hoa Q. (Department: 2877)
Optics: measuring and testing
Lamp beam direction or pattern
Reexamination Certificate
active
06339469
ABSTRACT:
TECHNICAL FIELD AND PRIOR ART
The invention relates to a method and device for determining the intensity and/or phase distribution in various cut planes of a space coherent beam, in particular of the light beam produced by a laser.
To carry out such a determination, known devices are split into three categories:
1) interferometry devices,
2) devices of Hartmann-Shack type
3) space pattern sampling devices.
Interferometry devices and methods use the combination of the wave front to be measured with a wave having a phase relation therewith.
Devices and methods of the Hartmann-Shack type use a phase mask (lens arrays, holes . . . ) to evaluate the slope of the wave surface at each point. Reconstruction then allows the wave surface to be calculated.
The problem of obtaining the phase from mere intensity profiles has already been studied in literature as the so-called “phase retrieval problem”. The study of various existing solutions is proposed in the articles “Phase retrieval algorithms: a comparison”, Applied Optics, vol. 21, p. 2758, August 1982, and “The phase retrieval problem”, IEEE Transactions on Antennas and Propagation AP29, P. 386, March 1981.
In particular, the algorithm proposed by Gerchberg and Saxton in 1971 (magazine Optik 34, 275 (1971)) is known. The authors use an image plane, and a distant field diffraction plane, at the focus of a lens.
According to this algorithm, an intensity profile is measured in a first cut plane, by assigning a phase profile that is arbitrary in every point of this plane; next, using beam propagation equations, the intensity and phase profile in a second cut plane is calculated: the intensity profile calculated in this second plane is replaced by the measured profile, and the intensity and phase profiles are again calculated in the first plane. Iteration is performed up to convergence.
According to the article by D. L. Misell (“A method for the solution of the phase problem in electron microscopy”) published in J. Phys. D. Appl. Phys., vol. 6, 1973, p. L6-L9 and the comments on this article by Gerchberg and Saxton (same references, p. L31), both planes can be any planes but close to each other.
In recent literature, numerous studies have been conducted on these methods, applied to electronic microscopy and antenna transmission. These studies suggest algorithmic improvements. E.g., we may quote the article “Radiation Pattern Evaluation from near-field Intensities on planes”, IEEE Transactions on Antennas and Propagation, vol. 44, no. 5, May 1996, which suggests not to use necessarily the object plane and the Fourier plane.
Tests performed with laser beam images and these methods have not allowed to reach results sufficiently good to make it possible to do without systems of the Hartmann-Shack type.
The problem is therefore posed of finding a method and device allowing to improve the accuracy of phase reconstruction, and compete with expensive commercially available systems, e.g. of Hartmann-Shack or Zygo type.
DESCRIPTION OF THE INVENTION
First of all, the object of the invention is an iterative method of determining the intensity and phase distribution of a light beam coherent in one plane, comprising the steps of:
measuring the intensity I
i
, i=1 . . . N, of the beam, in N planes, N≧3, including the plane wherein intensity and phase distribution is to be determined,
choosing, for plane i=1, an initial phase matrix &phgr;
1
and calculating a complex amplitude matrix, by term-wise multiplying the phase matrix e
i&phgr;1
by the corresponding amplitude matrix A
1
,
for each plane j>1:
determining a propagated complex matrix B′
j
from the measured amplitude matrix I
j−1
of plane j−1 and the phase matrix &phgr; of plane j−1,
extracting, from B′
j
, the phase matrix of plane j,
iterating the method up to convergence (and j=1 when j−1=N).
The method according to the invention therefore implements N planes, with N≧3. One of these planes is that of which intensity and phase distribution is to be determined. The method starts with a phase profile (e.g., null) on a first plane. Then, by calculation, the beam propagation from one plane to another are performed while keeping, after each iteration, the phase profile calculated for a plane, and by introducing the intensity profile measured in the same cut plane. Thus, we proceed from plane #
1
to plane #
2
, . . . to plane #N, then we return to plane #
1
. The convergence of the phase profile towards the intended profile is noted: indeed, the intensity and phase profiles calculated in one of the N planes allow the intensity profiles measured in any other plane to be retrieved exactly.
For each plane j(j>1), a phase profile is calculated from the complex matrix reflecting the state of the beam in this plane. This complex matrix is itself the result of a calculation, which reflects the equations of propagation, implementing the (measured) intensity matrix and the (calculated) phase matrix in plane j−1. In other words, for each plane j, by propagation, the intensity and phase profile in this plane is calculated, then the intensity profile calculated in this plane is replaced by the measured intensity profile. It is then possible, from this measured intensity profile and the calculated phase profile, to perform a propagation calculation for plane j+1.
It is possible to use any path among the N planes, as required.
Consequently, according to the invention, instead of moving to and fro among two cut planes, at least three cut planes are implemented and we proceed for instance from the first plane to the second one, then to the third one, . . . then to the Nth one, and we return to the first plane, and so on. The accuracy of the phase reconstruction on the intermediate planes is thus improved significantly, which allows to compete with expensive commercially available systems, e.g. of Hartmann-Shack or Zygo type.
According to a development, the phase taken into consideration in the first cut plane at each iteration is an average of the phase profiles determined from all the plane pairs in one round trip. This method allows to introduce a weighting factor into averaging, wherein this factor can depend on a merit function of the current calculation. Thus, convergence on the N profiles is optimized.
Two versions of the method according to the invention are therefore proposed according to the calculation path among the measuring planes. The first methods covers the planes sequentially, whereas the second one performs a round-trip among plane pairs, and allows the relative weight of the different measured planes to be weighted.
The invention also relates to a device for implementing the invention.
Thus, the invention relates to a device, for determining the intensity and phase of a light beam in a plane, comprising:
a means for measuring intensity I
i
(i=1, . . . N) of the beam in N planes, N≧3,
based on an initial phase matrix &phgr;
1
corresponding to a plane i=1, a calculation means programmed for:
calculating a complex amplitude matrix (B
1
) by multiplying the phase matrix e
i&phgr;1
by the corresponding amplitude matrix (A
1
),
for each plane j>1:
calculating a propagated complex matrix B′
j
from the measured amplitude matrix A
j−1
of plane j−1 and the phase matrix &phgr; of plane j−1,
extracting a phase matrix from plane j of matrix B′
j
,
iterating the above calculation by making j=1 when j−1=N.
This device uses as many cameras as there are cut planes; preferably, the measured images are digitized before being processed. It can include a system for shaping the beam, limiting the extend thereof so as to adapt it to the cameras and to the bank length; a calibration step is then integrated in the steps of the main method. The addition of a lens or any other phase object allowing the beam to be adapted to the measuring device (divergence control . . . ) makes it possible to increase the field of application of the device. Using
Belledent Jérôme
Bruel Laurent
Burns Doane , Swecker, Mathis LLP
Commissariat a l'Energie Atomique
Pham Hoa Q.
LandOfFree
Method and device for determining the distribution of... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and device for determining the distribution of..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and device for determining the distribution of... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2820019