POLARIZATION CONTROL USING POLARIZATION MODULATION
BACKGROUND OF THE INVENTION
The present invention relates to controlling polarization state of an optical signal.
There are number of technologies used in constructing polarization control devices such as rotating fixed-birefringence plates; mechanically squeezed optical fibers (manually or piezo-electrically activated); lithium niobate based technology where optical waveguides with appropriate electrode configuration can convert transverse electric (TE) to transverse magnetic (TM) polarized light; and birefringence plates made of electro-optic materials such as lead lanthanum zirconate titanate PLZT, or liquid crystals.
Each one of these approaches has its own advantages. The physically rotating, fixed- birefringence plates can be controlled very precisely and are therefore very well suited for instrumentation such as, for example Polarimeters, polarization dependent loss (PDL) and polarization mode dispersion (PMD) meters. The response time of mechanically rotating plates is in the few seconds range.
The mechanically squeezed optical fibers technology can be made to respond very fast, on the order of a few microseconds. In this technology fibers are squeezed using piezo-electric actuators. Application of voltage changes the pressure on the fiber, which in turn changes its birefringence. The absolute value of retardation is not easily controllable and can vary from one birefringent segment to another. This makes the control strategy very complex and a smooth rewind process difficult. Reliability of this technology is also difficult to achieve, since pressure is applied to the polymeric fiber- protecting coating. Repeated pressure degrades the coating creating difficulty in achieving a reliable operation.
Lithium Niobate is the most well-developed electro-optic technology available currently. Waveguides formed in Lithium Niobate can act as Variable birefringence devices (VBDs) under voltage. This technology is very fast and very attractive for network application. Its shortcomings are the cost and complexity of control, relatively large insertion loss (3-4 dB), relatively large polarization dependent loss and drift in tuning
voltage.
The fourth option is to implement VBDs using electro-optic plates. These are non- waveguiding devices in which light propagates some distance through a material whose degree of birefringence can be controlled by application of voltage. Examples include but are not limited to, liquid crystal plates, slabs of PLZT or slabs of Lithium Niobate.
A state of polarization (SOP) can be commonly represented as a point on the Poincare sphere where each point corresponds to a specific state of polarization. The Poincare sphere is used to describe the polarization and changes in polarization of the propagating electromagnetic wave. Polarization controllers (PC) are devices, which convert any input SOP into a fixed, predetermined SOP or into an arbitrary output SOP. The conversion is done by a sequential use of polarization modifying stages such as, for example, variable birefringence devices (VBD), which can be comprised of, without limitation, phase retardation plates, segmented waveguides, or polarization rotators oriented in a specific relation with respect to each other.
In voltage controlled PCs, in contrast to mechanically controlled devices, each device is actuated by a control signal, such as, for example, voltage, which adjusts either the degree of birefringence, the actual or effective device orientation or the relative phase between polarization components. Application of a control signal to a PC plate changes the output SOP. The corresponding point moves on the sphere accordingly. A sequence of steps is needed to move a point on the sphere from one location to any other location by means of rotation about a fixed axis. This principle governs all PC devices.
In order to transform an arbitrary input SOP into the desired output SOP, or equivalently in order to move a point on the Poincare sphere from any location to a desired coordinate, at least two axes of rotation are needed. Each axis of rotation can be implemented using a single variable retardation plate, a specific type of VBD. Since the order of rotations is important and may differ for different input SOP, at least three plates are needed to perform such transformation in a preferred embodiment of the present invention. In practice, because of limited range of rotation or variety of other practical limitations, PCs are made with a larger number of plates, or segments in the
case of waveguides. The actual number of VBDs depends on the technology used.
Most current polarization control devices are based on multiple polarization- transforming elements (plates or waveguide segments) where each element moves the SOP (state of polarization) point on the Poincare sphere over a finite extent trajectory. A sequential activation of elements brings the SOP to the desired output position. Some control technologies require an additional control step, called rewind, which resets the control voltages applied to the elements when these reach a maximum value, without affecting the output polarization state.
SUMMARY OF THE INVENTION
It is an object of the invention to provide an improved polarization control. The object is solved by the independent claims. Preferred embodiments are shown by the dependent claims.
According to one embodiment of the present invention, a polarization controller (provided to convert an input state of polarization of a received input optical signal into an output state of polarization) comprises a plurality of polarization modifying stages for modifying the input state of polarization into the output state of polarization. Each stage allows converting an optical signal from a first state of polarization to a second state of polarization, respectively. A modulation unit is provided to modulate operation of at least one of the plurality of polarization modifying stages. A control unit receives a signal in response to the modulation, and derives therefrom an error signal indicative of a deviation of the output state of polarization from a target state of polarization. The control unit further controls operation of at least one of the plurality of polarization modifying stages in order reduce the deviation of the output state of polarization from the target state of polarization.
Preferably, the output state of polarization is modulated, and more preferably around the output state of polarization in the Poincare sphere.
A detector might be provided for deriving a signal indicative of the power in the target state of polarization of the optical signal, or at least a portion thereof, having the output state of polarization.
A polarizer might be provided oriented in the target state of polarization for receiving the optical signal, or at least a portion thereof, having the output state of polarization. A detector might further be provided for detecting the output of the polarizer and deriving a signal indicative of the power in the target state of polarization of the optical signal, or at least a portion thereof, having the output state of polarization.
The error signal might be derived from the signal indicative of the power in the target state of polarization. In such case, the error signal might represent a deviation in at least one of amplitude and phase in the signal indicative of the power in the target state of polarization. In other words, to generate the error signal, the polarization modulation is converted into intensity modulation e.g. by a polarizer. The amplitude and phase of this AC signal contain all the information needed for the control circuit to decide the distance and the direction (on the Poincare sphere) of the actual SOP from the target point. The control logic than adjusts all plates at once to move the SOP to the target. This allows fast response time needed e.g. for real time monitoring of networks.
In a preferred embodiment, each polarization modifying stage modulated by the modulation unit is located at the end of the polarization modifying stages in the propagation direction from receipt of the input optical signal towards an output of an optical signal with the output state of polarization.
In a preferred embodiment, each polarization modifying stage modulated by the modulation unit is located in a control path separated from an output or the main path where an optical signal with the output state of polarization is output or exits the device.
A beam splitter might be provided for splitting an optical signal with the output state of polarization into a first portion coupled into a control path, wherein the at least one of the plurality of polarization modifying stages modulated by the modulation unit is located, and into a second portion coupled into an output path providing an output of the polarization controller.
The polarization controller of a preferred embodiment comprises a plurality of first polarization modifying stages adapted for modifying the input state of polarization into the output state of polarization, and at least one second polarization modifying stage adapted to be modulated by the modulation unit. Preferably, the at least one second
polarization modifying stage allows modifying state of polarization substantially faster than each one of the plurality of first polarization modifying stages, e.g. 10 times faster or more.
In one embodiment, at the output of the PC, the SOP is monitored directly (e.g. via a polarization beam splitter) or indirectly (e.g. via any polarization dependent effect) and an error signal is sent to the PC controller. This in turn minimizes the error voltage by adjusting the control voltage on each one of the VBDs. The provision of voltage adjustment might be embedded in the control algorithm and the performance of the PC depends on this algorithm or sequence of instructions run by the control processor. A particular embodiment uses the peak search algorithm, where each VBD is activated and adjusted so as to locally minimize the error signal. Subsequently, the next VBD is activated and its control voltage readjusted, again to minimize the error signal, and so on. This method can have more complex variants but in principle it requires multiple trial-and-error steps to bring the SOP point on the sphere to the desired location. Imperfections in plate orientations create local minima in the output space, causing the control system to optimize the operation at the wrong point of the parameter space. The number of steps needed to bring the SOP to its target multiplied by the time per step defines the response time of the PC.
A preferred embodiment of the present invention includes a polarization control device, which can be constructed in an array configuration. The plates used to transform the input SOP can be built using either a segmented electrode configuration in the case of solid electro-optic material or in the case of liquid crystal it can be constructed by depositing on the glass substrate an arrayed pattern of transparent electrodes, for example, but not limited to, Indium Tin Oxide (ITO). The PC array is particularly useful for multi-channel applications where each of the channels needs to be controlled separately. The polarization detection can be made either by using a polarization beam splitter for each channel or a long prism shaped polarization beam splitter, one to cover all channels.
In another preferred embodiment, detecting polarization in a multi-channel application includes modulating each channel with its own distinct tone (frequency tagging) and separating the channels electronically using a single photodetector and a single tap
mounted on the multiplexed output fiber. The control electronics in the case of arrayed PC includes separate sections to control each channel individually.
An arrayed PC can be used to align the polarization states of different wavelengths such that their polarization is at a selected relation to each other. The PC can be designed to provide the selected polarization states for the different wavelengths either at its output or at some other point along the fiber. In this embodiment the SOP data supplied to the PC array may be separated out by wavelength. This can be done using wavelength tagging by low frequency dither, a de-multiplexing technique to separate the channels and to monitor each one individually, or by means of a tunable filter where each channel is monitored sequentially.
In one embodiment, a device for transforming an arbitrary input polarization state of a beam into a fixed output polarization state comprises several transforming birefringent plates that, in conjunction with electronic controls, maintain a fixed output polarization state. In order to achieve fast tracking of changes in the incoming polarization state, two (or more) additional polarization modulating plates are added. These generate a fast error signal in a time scale much shorter than the duration of a single control step of the transforming plates. The modulating plates are driven such that multiple waveforms, one for each plate, are applied at 90 degrees out of phase thus creating a closed trajectory on the Poincare sphere. A polarization beam splitter converts the modulated polarization state to time-varying signal whose amplitude, phase, and offset provides an error signal. The control electronics utilize this signal to precisely control the transforming birefringent plates to maintain a linear output state of polarization in a minimum timescale. .
The invention can be partly or entirely embodied or supported by one or more suitable software programs, which can be stored on or otherwise provided by any kind of data carrier, and which might be executed in or by any suitable data processing unit.
BRIEF DESCRIPTION OF THE DRAWINGS
Other objects and many of the attendant advantages of the present invention will be readily appreciated and become better understood by reference to the following detailed description when considering in connection with the accompanied drawings.
Features that are substantially or functionally equal or similar will be referred to with the same reference sign(s).
Figure 1 illustrates a flow chart of the PC control logic in accordance with a preferred embodiement of the present invention.
Figure 2 illustrates a schematic diagram of the polarization controller optical unit wherein S1 , S2, S3 indicate for each plate the axis around which it rotates the SOP on the Poincare sphere in accordance with a preferred embodiment of the present invention.
Figure 3 is a block diagram of the polarization control system in accordance with a preferred embodiment of the present intention.
Figure 4 illustrates a Poincare Sphere Representation of SOP drift in accordance with a preferred embodiment of the present invention.
Figure 5 illustrates an array configuration of polarization control devices for use in multi-channel polarization transformation in accordance with a preferred embodiment of the present intention.
Figure 6 illustrates a launch polarization-optimizing device in accordance with a preferred embodiment of the present intention.
DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS OF THE
INVENTION
Figure 2 depicts the design of a fast polarization controller based on polarization modulation. An input optical fiber 10 carries an optical signal. Lens 16 collimates the light beam. A set of birefringent plates oriented at 45 degrees with respect to each other (S1 , S2) is placed inside a box 60, which provides a mechanical support and an enclosure to the device. The collimated beam passes then through a set of polarization transforming plates, which can be made of: liquid crystal twisted nematic elements or voltage controlled birefringent plates (VCB) or ferro-electric liquid crystal plates. Alternatively they could be constructed using birefringent materials such as Lithium Niobate or PLZT. Each of these plates 20 has a range of rotation on the Poincare
sphere of at least 180 degrees. The number of plates used can vary for example between 6-8 depending on the range of each plate, the temperature of operation and other tradeoffs in usage. If liquid crystal plates are used with 180 degrees of retardation per plate then eight plates would be needed. The transforming plates are followed by two polarization modulating plates 30. The function of these plates is to provide a continuous, small-amplitude polarization rotation, which can be detected by the monitoring photodetector 50, and used as an error signal. Because of the small amplitude of the modulation signal its frequency of response can be much higher than that of the transforming plates 20. Furthermore, the polarization modulating plates can be made using materials such as ferro-electric liquid crystals that have a very fast response time but limited rotation amplitude. The fast modulation enables the control logic to compute the voltage corrections needed to activate the plates and apply the correction in a single step. The operation of the polarization modulating plates is depicted in Figure 4 where a Poincare sphere representation of the modulated polarization is shown. Use of polarization modulating plates speeds up the response time of the polarization controller by a factor of 10 or more. To monitor the output state of polarization a polarization beam splitter 40 is used. Light polarized in the plane of the drawing is transmitted and light polarized perpendicular to the plane of the drawing is reflected toward a monitoring photodetector 50.
Figure 3 shows the block diagram of the polarization control system. The monitoring photo-detector 50 is connected to a control electronic board 90 where the photocurrent is amplified and converted to error signal in the following way. The amplitude of the error signal measured at the modulation frequency is proportional to the distance of the target SOP from the measured SOP. The phase of this signal depends on the direction in which the output SOP deviates from the target value. Using these two correction parameters the control electronics 90 computes the values of voltage corrections for each one of the plates. The correction voltage is then applied to all plates simultaneously in one step.
Figure 4 shows the trajectories traced by a polarization-state representing point on the Poincare sphere. A target output SOP (T) is continuously modulated and thus producing a circle CT around the target point. This circle shifts from its optimal position
CT, as marked, when input SOP changes. The photodetector current is proportional to
the projection of this trajectory on the S2 axis. This projection remains constant as a function of time if the trajectory is centered. If the circular trajectory CO is off-set from the center, the projection moves along the axis generating AC photocurrent at the detector. This AC component of the photocurrent is the error signal fed back to the control circuitry.
Figure 4 is a drawing of the Poincare sphere, showing the axes S1 , S2, S3 and the closed trajectory generated by the modulated birefringence plates. The optimal target trajectory is centered on the tip of the S2 axis while an error-signal generating trajectory is shown offset from the optimal position.
Figure 1 illustrates an example of a logic diagram. An error signal ES2 in S2 is monitored in a step 400 and derived from the AC projection on S2. The polarization modulation is converted into intensity modulation to generate the error signal ES2. As long as the error signal ES2 is zero, not further control action has to be provided. In case the error signal ES2 is not zero, the amplitude and phase of ES2 is detected in step 410. The amplitude and phase of this AC signal contain all the information needed for the control circuit to decide the distance and the direction (on the Poincare sphere) of the actual SOP from the target point. In step 420, plate voltages for achieving ES2=0 are calculated. The control logic then adjusts all plates at once to move the SOP to the target by applying all voltages simultaneously in step 430.
Figure 5 shows the array implementation of the polarization controller. Functionally each member of the array performs in the same way as the single polarization controller. The control electronics in the case of array Polarization Controller need to be made such that each member of the array is controlled independently of all other members. The input fibers 11 are arranged in an array. The fiber array is attached to a lens array 17. The light exiting the lens array 17 propagates to the polarization transforming plates 21 where each plate is made of an array of active areas. One implementation of such polarization transforming arrays are plates made using liquid crystal technology where each array element is defined by electrodes formed using e.g. ITO on the encapsulating glass substrate. The light propagates than to a set of modulated birefringent plates 41 which are made of arrays of active areas. Light is then split into its two polarization components by the polarization beam splitter 41 and one
of these components is detected by the detector array 51. The transmitted light is focused by the lens array 71 into the output fiber array 81. The whole assembly is encapsulated by box 91.
Figure 6 shows another mode of use of use of a polarization controller where multiple channel fibers 11 are aligned and collimated 17 into a fiber array. Each fiber transmits one communication channel. The output of the array propagates to transforming plates 21 where each plate is made of an array of active areas. One implementation of such polarization transforming arrays are plates made using liquid crystal technology where each array element is defined by electrodes formed using e.g. ITO on the encapsulating glass substrate. The light propagates then to a set of modulated birefringent plates 32 which are made of arrays of active areas. An array of lenses images the output light beams onto an input surface of multiplexing device 52 where all channels are combine into a single light beam. This light beam is coupled into an optical fiber 62. The output fiber is than tapped by e.g. fiber coupler and a polarizer or a polarization beam splitter 72 at some point outside the device enclosure 92. The tapped part of the beam passes through a tunable optical filter, which sequentially selects the channels from the combined light stream. The control logic can than be used to: minimize the power in each channel and thus aligning the launch polarization state of all channels to each other. Alternatively, the control logic can be programmed to minimize and maximize the tapped power in alternating channels and thus making them orthogonal pair wise. This mode of operation creates launch conditions, which may be advantageous in networks deploying large number of amplifiers. Orthogonal launch in that case reduces the polarization dependent gain.