Optical: systems and elements – Optical modulator
Reexamination Certificate
2000-08-31
2002-04-16
Ben, Loha (Department: 2873)
Optical: systems and elements
Optical modulator
C359S199200, C359S250000, C359S490020, C359S256000, C359S249000, C359S618000, C356S367000, C356S489000, C250S201100, C349S009000, C348S751000
Reexamination Certificate
active
06373614
ABSTRACT:
FIELD OF THE INVENTION
The invention relates to sensing and control of polarization in optical beams using feedback, and to optical telecommunication systems where control of polarization and of polarization mode dispersion are important.
BACKGROUND OF THE INVENTION
It is routine to measure polarization by means of serial measurements of intensity while a polarization sensitive element such as a waveplate or polarizer is mechanically rotated, or engaged and disengaged from the beam. Electro-optic approaches are used, where variable retarders such as pockels cells, Kerr cells, liquid crystal cells, optical rotators, and the like are driven to different settings while the beam is analyzed at a fixed polarizer or other element. The mechanical systems are unwieldy, and the electro-optic systems are costly; both are inherently slow and vulnerable to time-fluctuations in the beam being measured since they employ serial measurements of intensity.
Other approaches are possible, such as the use of two detectors at the outputs of a Wollaston prism. This is rugged and provides rapid information about the state of polarization (SOP) of a beam, but the information is incomplete, being typically limited to at most two of the four Stokes parameters. The cost of a Wollaston prism or similar displacing element is considerable, another problem with this approach.
Azzam teaches in U.S. Pat. No. 4,681,450 an apparatus for sensing polarization of a monochromatic beam, based on several photodiodes in a three-dimensional configuration that reflects light from a first detector onto a second detector and finally onto a third. The reflections are at significantly non-normal incidence, and the detectors have polarization-dependent reflection coefficients, which enable calculation of the incident SOP from the three detector readings. The apparatus is dependent on the surface properties of the detectors, and often requires calibration of each instrument. Further, it is only suitable for use over a limited range of wavelengths, since the reflection coefficients are wavelength-dependent. Finally, the configuration of several detectors is awkward to construct and costly to package, compared to simple mounting of detectors on a printed-circuit board or the like. These factors all vitiate against low cost in a mass-production fashion. The alternative arrangement of Azzam in U.S. Pat. No. 4,725,145 requires fewer photodetectors (only one or two, rather than four) but they must be mechanically rotated to derive the state of polarization.
In short, all prior-art method for sensing SOP suffer one or more of the following limitations: high cost, slow speed, mechanical moving parts, incomplete information about the SOP, need for calibration, critical dependence on detector reflection properties, and inability to sense SOP for a polychromatic beam. There is at present no rugged and economical means for sensing complete SOP information.
Turning to polarization control, Clark teaches a polarization controller in U.S. Pat. No. 5,005,952, based on a stack of multiple liquid crystal variable-retarder cells controlled by means of a feedback signal. He further teaches means for transforming a continuously varying SOP to a fixed linear state by use of four such cells, or by a pair of three-cell stacks with switching means to direct the incident beam to one or another of the stacks.
Rumbaugh et. al. teach a polarization controller in U.S. Pat. No. 4,979,235, based on a stack of three liquid crystal variable-retarder cells. He further discloses a method for generating a continuously-varying SOP from a fixed linear SOP using only a single three-cell stack with no switching means. The method is based on optimizing the output of a homodyne detector by noting the change in output while the drive signal is adjusted in a certain sense for each of two liquid crystal cells, one after the other. The detector output is monitored after each adjustment, and the sense of adjustment is reversed in subsequent loop iterations for a given cell if the previous adjustment decreased the detector output. Thus, if the first adjustment to a given cell has the wrong sense to produce the desired SOP, a total of four adjustments are required—two adjustments to each of the two cells-before the feedback begins to be beneficial .
This complicated and unwieldy control method is inherent in such a system where there is only one feedback signal from which adjustment of two cells must be derived. And, since the signal attains a local maximum at the desired operating point, the feedback sense changes from negative (stable) to positive (unstable) when the controller passes through the optimum response. An equivalent situation occurs if the signal attains a local minimum at the desired operating point. Overall, the servo action is not deterministic, but works by trial-and-error ‘hunting’: it hunts to determine which retarder needs adjustment, and it hunts to determine whether an increase or decrease in retardance is needed.
Miller teaches feedback control of a liquid crystal variable retarder in U.S. Pat. No. 4,848,877, based on an optical signal passing through the retarder.
Various prior-art references describe the use of mechanical means to adjust the SOP of light in an optical fiber by squeezing the fiber. Firms supplying such equipment commercially include Oz Optics (Carp, Ontario, Canada), FiberPro (Taejon, Korea), and Optics For Research (Caldwell, N.J.).
Systems for SOP compensation and control of have been proposed based on the opto-ceramic materials sold by NZ Applied Technologies (Woburn, Mass.).
Dithering systems of various kinds are known for maximizing signals which have a periodic sinusoidal (or similar) dependence on a control variable. Some dither the control at a frequency F
d
which is faster than the servo response of the system F
s
, and look for a minimum in the envelope of response at F
d
, which indicates that the flat top of the periodic response has been attained; or, they monitor the envelope of response at frequency 2F
d
and seek a maximum, or combinations of these themes and variations upon them. However, these methods cannot be exploited when the control element has inherently slow time response and cannot be dithered at or above the servo response frequency.
Thus while various methods have been shown for polarization control, or for control of liquid crystal variable retarders, none provides for deterministic servo control that is free from hunting, nor that produces a stable output with rapid time response to changes in the input SOP, and that exhibits low output error. It is the aim of the present invention to provide these capabilities. It is further an aim to provide this in a controller with only three variable retarder stages, which nonetheless has the capacity to transform a continuously-variable SOP to a desired state. It is yet a further aim of this invention to provide a control action that is robust in the face of intensity changes in the incident beam, and to achieve a high optical efficiency or throughput.
SUMMARY OF THE INVENTION
The invention consists of an apparatus that samples the SOP of a beam, that may optionally be placed in series with a polarization compensator to sample the SOP of the compensated beam. The compensator may be built using any type of control element, including liquid crystal cells, fiber optic squeezers, opto-ceramic modulators, lithium niobate modulators, or any other device which acts as a polarization compensator. In a preferred embodiment, the compensator comprises three retarders that transform incident light with a continuously-varying SOP to be linearly polarized along a specified polarization axis. The first and third variable retarder have their fast or slow axes oriented at 45° to the exit polarization axis, to which either the fast or slow axis of the middle variable retarder is parallel. The middle variable retarder may be constructed as two liquid crystal cells with parallel (or perpendicular) slow axes, so they act in concert as a single retarder whose retardance is the sum (or
Ben Loha
Cambridge Research & Instrumentation Inc.
Cohen & Pontani, Lieberman & Pavane
LandOfFree
High performance polarization controller and polarization... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with High performance polarization controller and polarization..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and High performance polarization controller and polarization... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2889392