Optical beam combiner

Abstract

An optical beam combiner for combining a plurality of light beams comprises: a plurality of spherical mirrors; and a flat mirror, the plurality of spherical mirrors and the flat mirror configured to form at least one multiple pass light beam optical arrangement for receiving the plurality of light beams and for superimposing spot images of the light beams onto a single location with a single incident angle. In addition, a waveguide-based optical White cell comprises: a waveguide having front and rear edges, the inside surfaces thereof being coated with a reflective material, wherein the front edge including an input section for the passage of at least one light beam into the waveguide; at least one waveguide lens disposed in front of the inside surface of the rear edge to form a plurality of waveguide spherical mirrors at the rear edge; a plurality of angled micro mirrors disposed at the inside surface of the front edge; and the plurality of waveguide spherical mirrors and the coated front edge configured to form at least one waveguide White cell.

Description

[0001] This utility application claims the benefit of the filing date of the U.S. Provisional Application 60/588,729, entitled "Optical Beam Combiner", and filed Jul. 16, 2004.

BACKGROUND OF THE INVENTION

[0002] The present invention relates to optical devices, in general, and more particularly, to an optical beam combiner for receiving a plurality of light beams and superimposing spot images of the plurality of light beams onto a single location with a single incident angle.

[0003] Generally, an optical cross-connection device, like a White cell optical switch, for example, comprises a plurality of optical elements disposed in a predetermined spatial three dimensional pattern for directing one or more light beams from an input through a plurality of reflections to an output. Multiple light beams may bounce through various stages of the device simultaneously. A problem arises at the final or output stage of the White cell cross-connection device where the multiple light beams are ultimately directed from different spatial locations and different incidence angles. Thus, the multiple light beams will illuminate spots in various locations within the region of the output stage. Accordingly, each light beam of the multiplicity has a distinct incidence angle depending onto which region of the output stage it is being directed. This variation in the angle of incidence complicates the coupling of the light beams into an optical fiber or a light detector.

[0004] The present invention is intended to overcome or at least mitigate this drawback to the optical coupling in the output stages of optical cross-connection devices.

SUMMARY OF THE INVENTION

[0005] In accordance with one aspect of the present invention, an optical beam combiner for combining a plurality of light beams comprises: a plurality of spherical mirrors; and a flat mirror, the plurality of spherical mirrors and the flat mirror configured to form at least one multiple pass light beam optical arrangement for receiving the plurality of light beams and for superimposing spot images of the light beams onto a single location with a single incident angle.

[0006] In accordance with another aspect of the present invention, an optical beam combiner for combining an array of light beams comprises: a plurality of spherical mirrors; and a flat mirror, the plurality of spherical mirrors and the flat mirror configured to form at least one multiple pass light beam optical arrangement for receiving simultaneously the array of light beams and for superimposing spot images of each light beam of the array onto a single location with a single incident angle.

[0007] In accordance with yet another aspect of the present invention, a waveguide-based optical White cell comprises: a waveguide having front and rear edges, the inside surfaces thereof being coated with a reflective material, wherein the front edge including an input section for the passage of at least one light beam into the waveguide; at least one waveguide lens disposed in front of the inside surface of the rear edge to form a plurality of waveguide spherical mirrors at the rear edge; a plurality of angled micro mirrors disposed at the inside surface of the front edge; and the plurality of waveguide spherical mirrors and the coated front edge configured to form at least one waveguide White cell.

BRIEF DESCRIPTION OF THE DRAWINGS

[0008] FIG. 1 is an illustration of an exemplary free space White cell optical arrangement.

[0009] FIGS. 2a, 2b and 2c are top view illustrations depicting examples of operation of the exemplary White cell optical arrangement.

[0010] FIGS. 3a, 3b and 3c are front mirror illustrations depicting multiple pass light beam illuminations resulting from various operations of the exemplary White cell.

[0011] FIG. 4 is an illustration of an exemplary dual White cell optical cross-connection device.

[0012] FIG. 4a is a light beam connectivity diagram suitable for use in describing the operations of the exemplary dual White cell optical device.

[0013] FIG. 5 is an illustration depicting multiple pass light beam illuminations of the faces of the mirrors of a dual White cell device using an array of micro mirrors as a common mirror element for both White cells.

[0014] FIG. 6 is an illustration depicting an embodiment of a dual White cell device using an array of micro mirrors as a common mirror element for both White cells.

[0015] FIG. 6a is an illustration of an output region of a White cell optical device showing the illuminations from two different light beams.

[0016] FIG. 7 is an illustration depicting an alternate embodiment of a dual White cell device using an array of micro mirrors as a common mirror element for both White cells.

[0017] FIG. 8 is an illustration depicting an exemplary optical beam combiner suitable for embodying one aspect of the present invention.

[0018] FIG. 9 is an illustration of a mirror face of the optical beam combiner showing multiple pass light beam illuminations of the face thereof.

[0019] FIG. 10 is an illustration of the optical beam combiner showing the multiple pass light beam illuminations on the mirror face.

[0020] FIGS. 11 and 12 are front and isometric perspective views, respectively, of an exemplary waveguide-based White cell optical arrangement suitable for embodying another aspect of the present invention.

DETAILED DESCRIPTION OF THE INVENTION

[0021] An optical switch based on the principles of an optical White cell will exemplify an optical cross-connection device for the purposes of describing one or more embodiments of the present invention. The optical White cell is an example of a multi-pass light beam optical system for generating a series of spot illuminations in sequence for an input light beam as will be better understood from the following description. Other examples of multi-pass light beam systems include a Herriot cell or any of the alternative spot pattern generators disclosed in U.S. Pat. No. 6,266,176. For the present example, a White cell comprising a set of three spherical mirrors with identical radii of curvature will be used. The multi-pass system of spherical mirrors will refocus the beam continuously within the White cell. One of the White cell's spherical mirrors may be replaced with an array of micro mirrors which may be made using micro-electromechanical systems (MEMS) techniques and will hereinafter be referred to as the MEMS micro mirrors, MEMS array or MEMS device.

[0022] Each of the micro mirrors of the MEMS device may be independently tilted to different angles. Also, multiple light beams may be directed to reflect or bounce off of the optical elements within the White cell simultaneously, and each light beam may be focused to illuminate a spot on a different micro mirror on each bounce or pass. Thus, in the exemplary optical switch of the present embodiment, there is an opportunity to switch a light beam with the MEMS device toward a new destination on each bounce. In addition, the number of possible attainable outputs of the exemplary switch will depend on the number of bounces that the light beams make in the White Cell. So, the number of attainable outputs may be controlled by controlling the number of bounces.

[0023] This White cell technology offers a highly scaleable all-optical cross-connect switch for a large number of ports (N inputs.times.N outputs), that avoids the effects of beam divergence and high precision angle control of the MEMS micro mirrors. Because several beams may bounce inside the White cell, each one of them may be controlled individually in such a way to control the destination of each beam. That is, each beam can be directed to any of multiple output regions. As noted above, however, on the final stage each beam will have a distinct incidence angle depending on which output region a particular beam is directed, which complicates the coupling into a fiber optic core or light detector. The beam spot illumination may also land in various locations within the output region. An optical beam combiner may be included at the output stage of the exemplary optical switch to cause all the possible beam illumination spot locations to be superimposed, and to correct for the variation in the angles of incidence. Thus, with the inclusion of the beam combiner, each output light beam may be modified such that it can be coupled properly into an optical fiber or onto a light detector.

[0024] The principles of operation of an exemplary White cell on which the present optical photonic switch is based will be reviewed briefly in connection with the illustration of FIG. 1. Referring to FIG. 1, the exemplary White cell 10 comprises three spherical mirrors B, C, and M. The mirror M faces the other two mirrors B and C, and is separated from them by a distance equal to their radii of curvature R, which is the same for all three mirrors.

[0025] The center of curvature of mirror M (CC(M)) lies on the optical axis 12 thereof. Because Mirrors B and C are mounted across from mirror M and separated from it by a distance equal to the radius of curvature R, either mirror B or C images the surface of mirror M onto itself, whereas mirror M images B and C onto each other. The centers of curvature of mirror B and C (CC(B) and CC(C), respectively) are located on mirror M, at a distance .delta. left and right of the optical axis 12, respectively. Hence the centers of curvatures CC(B) and CC(C) are separated by 2.delta.. The locations of the centers of curvature are key to the operation of a binary optical cross-connection device.

[0026] An exemplary path of a single light beam 16 through the White Cell 10 is shown by light rays in the top view illustrations of FIGS. 2a, 2b and 2c. FIG. 2a shows how the light beam 16 enters the White Cell through an input turning mirror 20 located adjacent to mirror M. The input light beam 16 is focused to a spot on the input turning mirror 20. Light diverging from this input spot will propagate toward mirror C and be refocused by mirror C back onto mirror M as illustrated in FIG. 2a. The spot illumination of the input light beam 16 on the input turning mirror 20 is located at a distance d.sub.1 away from the mirror C's center of curvature CC(C), and the first image of the spot illumination of light beam 16 on mirror M will therefore be located at point 22 an equal distance d.sub.1 from mirror C's center of curvature on the other side from the input turning mirror 20.

[0027] FIG. 2b shows how the light beam 16 bounces from point 22 off mirror M towards mirror B. The light beam 16 diverges in its path towards mirror B, but is refocused by mirror B onto mirror M as a spot image at point 24. Since the first image at point 22 is located at a distance d2 from one side of the mirror B's center of curvature CC(B), and then, the second image at point 24 will appear on mirror M at an equal distance d.sub.2 from the other side B's center of curvature.

[0028] A feature of the exemplary White cell 10 is shown in FIG. 2c, where the light beam 16 from mirror C is imaged via mirror M onto mirror B. As long as these two mirrors B and C are--the same size, light can be imaged back and forth between them many times without additional diffraction losses from the edges of the mirrors. Therefore, the losses in the system are caused only by the mirrors' reflectivities.

[0029] This multiple-reflection White cell configuration 10 will result in an illumination spot pattern on the surface of mirror M. The spot pattern as shown in the front view illustrations of mirror M in FIGS. 3a, 3b, and 3c is very predictable depending only on the locations of the centers of curvature of mirrors B and C. Each front view illustration of FIGS. 3a, 3b and 3c shows a sequence of spot illuminations on mirror M for a particular input spot illumination on the input turning mirror 20. The locations of the centers of curvature of each mirror B and C are indicated in each Figure. An output turning mirror 30 has been added to the example to extract the light beam from the White cell 10 after all of the light beam reflections or bounces have been completed. The illuminating spots in the FIGS. 3a, 3b and 3c are numbered in the order in which the light "bounces" in the White Cell before finally imaging onto the output turning mirror 30. The odd-numbered spot images progress across the top to the left of the mirror M and the even-numbered spot images progress across the bottom to the right.

[0030] FIG. 3a is an illustration for a single beam White cell operation as exemplified in the previous FIGS. 1 and 2a-2c. Referring to FIG. 3a, the beam is directed to mirror C from the input turning mirror 20 and there focused onto mirror M at spot image 1. From spot image 1 the beam is directed to mirror B and there focused into mirror M at spot image 2. From spot image 2, the beam is directed back to mirror C and there focused onto mirror M at spot image 3. From spot image 3 the beam is directed back to mirror B and there focused into mirror M at spot image 4. The light beam will continue to bounce between mirrors B and C via mirror M for spot images 5 and above until the final bounce which directs the beam illumination or spot image to the output turning mirror 30.

[0031] The spacing between the illuminating spot images for a given input beam is directly related to the distance 2.delta. between the centers of curvature of mirrors B and C. The total number of spot images on mirror M is therefore dependent on the spacing .delta. and the overall size of mirror M. Note that the spot locations on mirror M depend entirely on the alignment of the two Mirrors B and C, and not on Mirror M. This will become of interest when we replace Mirror M with the MEMS micro mirrors and the beam illuminating spot images are made to land on the tilting micro mirrors thereof.

[0032] A second beam may be introduced into the White cell 10 as shown in FIG. 3b rendering a simultaneous dual beam operation. In FIG. 3b, one beam is represented by a square illuminating spot image and the other beam is represented by a triangle. Note that each input spot image from turning mirror 20 results in a different spot pattern on mirror M. In fact, as shown in FIG. 3c, it is possible to introduce a large array of spot images 40, each representing a different input signal. The spot image patterns on mirror M for each input beam are unique. In the present example, none of the bounces from any of the beams will strike any spot image from another beam.

[0033] As noted above, Mirror M may be replaced with a MEMS micro mirror array, and two additional spherical mirrors may be added to form an alternate White cell 50 as shown in the illustration of FIG. 4. In this alternate White cell 50, each spot image from each beam introduced into the White cell 10 will strike a different micro mirror of the MEMS array. Thus, in this alternate example, each beam in the array of input beams (see FIG. 3c, for example) may be independently controlled via the MEMS micro mirrors on every beam bounce as will become more evident from the following description. Thus, optical switching may be performed by allowing each input light beam to be switched between various White cell paths that alter the spot patterns on mirror M and thus, the exit location of each beam. It is possible to allow for a very large number of potential outputs for each of the input beams, but with the smallest possible number of light beam bounces within the white cell. Reducing the number of bounces reduces the loss, which will accumulate on every bounce.

[0034] Several cell configurations may be used to enhance the number of possible outputs with the least number of light beam bounces. The cell configurations may be divided in two categories: polynomial and exponential cells. In the "polynomial cells," the number of possible outputs N is proportional to the number of bounces m raised to some power. For example, in a quadratic cell N is proportional to m.sup.4, where m is the number of bounces on the MEMS device. In the "exponential cells," the number of possible outputs is proportional to a base number raised to the number of bounces (N is proportional to 2.sup.m for the binary case). The exponential approach has the advantage of providing far more connectivity for a given number of bounces (and thus loss), but the disadvantage of not having the built-in redundancy of the polynomial devices. In this application, all of these configurations will not be discussed. A binary system will be briefly discussed to ease the introduction of an optical beam combiner.

[0035] In the example of FIG. 4, an embodiment is illustrated which combines two White cells to produce an optical cross-connection device. Optical switching is performed by allowing each of a large number of input light beams to be switched between two different White cells. In this embodiment, one White cell produces two rows of spot images for each input beam, and the second White cell incorporates a spot displacement device (SDD) that will continue the spot patterns but displace them by some number of rows, thus changing the exit location of each beam. A very large number of potential outputs are provided for each of the input beams, but with the smallest possible number of bounces. Reducing the number of bounces reduces the loss, which will accumulate on every bounce. In a "binary cell," the number of possible outputs is proportional to 2.sup.m/4.

[0036] The architecture of the embodiment of FIG. 4 was originally proposed for optical true time delay devices for phased array antennas. In the exemplary White cell described in connection with FIGS. 1-3c, the location at which a spot image leaves the cell is determined by where the light beam entered the cell, and where the location of the centers of curvature of Mirrors B and C. In this alternate embodiment, the White cell is modified to control the output location of the spot illumination. To do this, Mirror M is replaced with a MEMS tilting micro-mirror array to select between two different paths on each light beam bounce. In addition, a second White cell is added in the newly available path. Both White cells produce a similar spot pattern on the MEMS array, but the illuminating spot images resulting from the second White cell are shifted such that they return in a different row of the MEMS array than if they returned from the first White cell.

[0037] Referring to FIG. 4 which illustrates an exemplary embodiment for a binary White cell device 50, mirror M is replaced with a MEMS micro mirror array 52 and a field lens 54 disposed in front thereof. The MEMS array/lens combination 52, 54 performs the imaging function of the original spherical mirror M. On either side of the MEMS micro mirror array 52 may be disposed two flat auxiliary mirrors 56 and 58, whose functions will be described supra. Each of the auxiliary mirrors 56 and 58 also has a field lens 60 and 62, respectively, disposed in front thereof to simulate a spherical mirror. These three field lenses 54, 60 and 62 may be combined into a single, larger lens as well.

[0038] The embodiment of FIG. 4 also includes four spherical mirrors 64, 66, 68 and 70 disposed in front of the mirrors 52, 56 and 58, but instead of having the centers of curvature of the spherical mirrors 64, 66, 68 and 70 on the MEMS array 52, the centers of curvature are located by design outside the MEMS array 52. In the present embodiment, the possible micro mirror tip angles of the MEMS array 52 may be .+-..theta. to the normal 72 (dashed line) of the MEMS array 52. Mirrors 64 and 66 are disposed one above the other, along an axis 74 (dashed line) at an angle of -.theta. to the normal axis 72. Mirrors 68 and 70 are also disposed one above the other along an axis 76 at an angle +3.theta. to the normal axis 72. While the mirror sets 64, 66 and 68, 70 of the present embodiment are arranged one above the other, it is understood that the mirrors of each such set may be arranged side by side on either side of the respective -.theta. or +3.theta. axis just as well. The axis of the lens 54 associated with the MEMS array 52 is disposed along the normal axis 72; the center of curvature (labeled CC.sub.AI) of the auxiliary mirror 56 and lens 60 together is disposed by design between mirrors 64 and 66, and similarly, the center of curvature CC.sub.AII of auxiliary mirror 58 and lens 62 is disposed by design between mirrors 68 and 70.

[0039] Let us assume that an input beam going from the plane of the MEMS array 52 is directed to mirror 64, for example, after light beam bounce 1. A light image reflected from this spot on mirror 64 is imaged to a new spot image on auxiliary mirror 56, in a column labeled "2" at the far left thereof as shown in FIG. 4. From there, the light beam is reflected to mirror 66, which directs the light beam back to the MEMS array 52 at a new micro mirror location, which may be in the column labeled "3", for example. If the micro mirror at that spot image of the MEMS array 52 is set to -.theta., then the light beam is directed back to mirror 64 again. So, mirrors 64 and 66 form one White cell with the MEMS array 52, lens 54, auxiliary mirror 56, and lens 60.

[0040] Accordingly, when micro mirror of the MEMS array 52 that the light beam strikes on bounce 3 is tipped to -.theta., the light returns to auxiliary mirror 56 via mirror 64 and may be focused a spot in column 4, for example. On the other hand, if the micro mirror of the MEMS array 52 that the light beam strikes at bounce 3 is instead turned to +.theta., then the light beam from mirror 66 will be reflected from the MEMS array 52 at an angle of +3.theta. along the plane of axis 76 with respect to the normal axis 72. Recall that there are two more mirrors 68 and 70 along the axis 76. So, when the reflecting micro mirror is set at +.theta., a light beam from mirror 66 will be directed to mirror 68 instead of mirror 64. In the present embodiment, a light beam is always directed to an upper mirror 64 or 68 from the MEMS array 52.

[0041] When a light beam is directed from MEMS array 52 to mirror 68, the light beam is refocused to auxiliary mirror 58 and forms a spot image in a column 4 of that mirror, for example. From there the light beam is directed to the lower mirror 70, and then back to the MEMS plane 52. Accordingly, mirrors 68 and 70 together with the MEMS array 52, lens 54, auxiliary mirror 58 and lens 62 comprise another White Cell of the embodiment. If the micro mirror in the MEMS array 52 struck by the light beam on bounce 5 is tilted to -.theta., the light beam from mirror 70 is again directed to the other White cell (specifically to mirror 64). Conversely, if the same micro mirror at bounce 5 is set tilted to +.theta., the light beam from mirror 70 is instead reflected at +4.theta., a direction that is not being used in this design, and the beam is lost.

[0042] Thus, according to the connectivity diagram shown in FIG. 4a, in the present embodiment, a light beam shown by the double arrowed line 80 may bounce continuously (and exclusively) between the MEMS array 52 and auxiliary mirror 56 via mirrors 64 and 66, a situation that doesn't occur while bouncing through mirrors 68 and 70. A light beam directed from the mirror 66 to the MEMS array 52 may be directed either back to mirror 64 (see arrowed line 80) or to mirror 68 ( see arrowed line 82) depending on the reflection angle setting of the corresponding micro mirror of the MEMS array 52. The light beam arriving at mirror 68 is returned to the mirror 70 (see arrowed line 84) via auxiliary mirror 58. Then, from mirror 70, the light beam is directed back to the MEMS array 52. Note that in the present embodiment, a light beam directed to the MEMS array 52 from mirror 70 must be directed to mirror 64 (see arrowed line 86) and auxiliary mirror 56; otherwise, it will be lost. Therefore, the light beam returning from the White cell comprising auxiliary mirror 58 needs four bounces to be directed back to auxiliary mirror 58, i.e. one bounce from the mirror 58 to the MEMS array 52 via mirror 70, a second bounce from the MEMS array 52 through mirror 64 to mirror 56, a third bounce from mirror 56 through mirror 66 to the MEMS array 52, and a fourth bounce from the MEMS array 52 to mirror 58 via mirror 68.

[0043] Note also that an input light beam may be sent to mirror 64 from the MEMS array 52 every even-numbered bounce, and to mirror 68 every fourth bounce (i.e. 4, 8, 12 . . . ). The odd-numbered bounces always appear on the MEMS array 52, and the even-number spots can appear either on auxiliary mirror 56 or auxiliary mirror 58. The light beam may be directed to auxiliary mirror 58 by the MEMS array 52 on any particular even-numbered bounce, but when the light beam is directed there, four consecutive light beam bounces are required before the light beam may be resent to auxiliary mirror 58 again.

[0044] Now, suppose that in the embodiment of FIG. 4, the auxiliary mirror 58 comprises a spot displacement device (SDD) that shifts a spot image over by some number of rows. This embodiment is exemplified in the illustrations of FIGS. 5 and 6. Referring to FIGS. 5 and 6, the SDD 58 may be divided into columns, and each column is assigned to every fourth bounce. Also, the number of elements (pixels) or rows of each column of the array of the SDD 58 by which a beam is shifted will be different for each column. That is, each column may shift a beam by a distance equal to twice that of the shift produced by the previous column. Thus, the first column will produce a shift of .DELTA., the second column a shift of 2.DELTA., the third column a shift of 4.DELTA. and so on, then producing a binary system.

[0045] Shifting the spot images can control at which row any given input light beam reaches the output turning mirror and in the present example, each row may be associated with a different output. The number of possible outputs is determined by the total number of possible shifts for a given number of bounces. In the design of FIG. 6, a shift is made every time the light beam is directed to the SDD 58, but this can only happen every four bounces. Thus the number of outputs N is given by: N.sub.binary=2.sup.m/4 (1) where m is the number of bounces.

[0046] In the mirror face diagrams of FIG. 5 is depicted a 12-bounce binary White cell system to illustrate the operation of the embodiment of FIG. 6. In this example, eight different beams, shown by various spot images, are incident on the input turning mirror 20. The patterns for the spot images for three of the eight light beams are indicated in the faces of the mirrors 52, 56 and 58 which are each divided into a grid of eight rows (for eight possible output locations) and six columns (for each bounce on the MEMS). The output column 90 constitutes a seventh column next to the MEMS array 52. In each region or pixel on the grid of the MEMS array 52 may be a group of eight micro mirrors, so that each of the eight beams may land on a different micro mirror on each bounce. Each beam may be directed either to the SDD 58 or to auxiliary mirror 56 on each bounce. The number of columns on the SDD (m/4=3), will thus determine the number of possible outputs; the other columns 92 are not used. Every four bounces allows for a shift, so 12 bounces will produce 2.sup.3=8 different outputs for each input light beam.

[0047] The example of FIG. 5 shows eight different input beams (only three, depicted by white, shaded and black symbols, being addressed in the present example) and eight possible outputs (numbered rows 0 to 7) in the output column 90. Initially, the three input beams start on row zero (0). Remember that according to the connectivity diagram of FIG. 4a, an input light beam may only go to the 68, 70 White Cell every fourth bounce (those would be the 4th, 8th and 12th bounces for a 12 bounce system). In the present example, suppose that the "white" beam is to be directed to the fifth output (row 5 of column 90), the shaded beam is to be directed to the second output (row 2 of column 90), and the black beam is to be directed to row 0 of column 90. The spot images of the three beams are shown in the respective mirror face for each bounce and the bounce numbers are shown beneath the columns of the mirror faces.

[0048] In operation, the "white" beam should be directed to the SDD 58 on the fourth and twelfth bounces, which correspond to row displacements of 4.DELTA. and .DELTA., respectively. Accordingly, the "white" beam may initially bounce in the 64, 66 White Cell (i.e. the corresponding micro mirrors on the MEMS array 52 are tilted to -.theta. position) for three bounces. Then, the "white" beam is directed to the SDD 58 on the fourth bounce (i.e. the corresponding micro mirror on the MEMS array 52 is tilted to +.theta.), and more particularly to the column in the SDD 58 that has a shift value of 4.DELTA.. After being shifted four rows in the SDD 58, the "white" beam is directed back to the MEMS array 52 on the fifth bounce and images on the row four (4) instead of row zero (0). The "white" beam is then kept bouncing in the 64,66 White cell, until the 12th bounce, when it is again directed to the SDD 58, and more specifically directed to land in the column with the shift value of .DELTA.. After being shifted an additional row in the SDD 58, the "white" beam is directed back to the MEMS array 52 on the next bounce and images on the row five (5) of the output column 90.

[0049] In a similar manner, the "shaded" beam may be shifted to the row two (2) of the output column 90 in twelve bounces (12). The "black" beam may be left unshifted throughout the 12 bounces to be output at row zero (0) of the output column 90.

[0050] In the foregoing described embodiment, it is noted that any input directed to a particular output will land in a different place within that output region. For example, in FIG. 5, the white beam was directed to output five (row 5) and appeared as a spot image in the upper right hand corner. Had the black beam been sent to output five, its spot image would appear in the lower right hand corner. Thus, once a given input has reached the correct output region, the spot images should be all made to land in the same spot, for example, for proper coupling to a light detector or a fiber core. This is non-trivial in the White cell because in addition to arriving at different locations in the output region, the light beams may arrive from different angles, a factor that will seriously affect output coupling, especially into an optical fiber. There are actually two angles of concern here. The first has to do with from which White cell of the two a beam is arriving when it reaches the output region. The other angle arises from the particular output location within that region where the spot image forms.

[0051] The first angle is the more severe than the second. FIGS. 6 and 6a show the last bounce for two different beams 100 and 102 (i.e. the "white" and "shaded" beams, respectively, of FIG. 5) for the 12-bounce system just described. The "white" beam 100 is directed to the fifth output (row 5) of column 90, meaning it was shifted on its last bounce, so it is coming from the 68, 70 White cell. On the other hand, the "shaded" beam 102, directed to the second output (row 2) of column 90, comes from the 64, 66 White cell on its last bounce. Thus, the two beams 100 and 102 are directed to their respective outputs from different White cells 64, 66 and 68, 70.

[0052] One way to solve this condition of difference in which mirror the beam comes from is to add one additional bounce to the system as shown in the illustration of FIG. 7. Then, regardless of the output row selected, all beams may be directed back to the 64, 66 White cell on their last bounce. The beams will come out at the appropriate row (i.e. output), and one column 104 over from column 90, but now all beams will arrive at their respective output regions from the same general direction, that of the final spherical mirror 66. However, while the beams are all arriving from the same White cell 64, 66, they are still directed to different outputs, i.e. rows of column 104 (FIG. 7). In addition, within each output region, e.g. row 2 or row 5, each beam 100 and 102, for example, may arrive at any of several different locations (e.g. lower corner, middle) as shown in FIG. 6a. This also creates a small difference in the angle at which a beam arrives.

[0053] Furthermore, the light input to the multi-pass, cross-connection device may be a two-dimensional spot array, having both columns and rows. Therefore, all the rows and columns of the spot array should be combined to a single spot, and this should be done taking into account the varying angles of incidence. The output should be a single spot, of substantially the same size and shape as any individual input spot, and the output should emerge at a specific angle, independent of the arrival angle of any particular beam. A method for superimposing all the potential spot images onto a single location and with a single angle using passive White cell technology will now be discussed.

[0054] An exemplary optical beam combiner 110 suitable for solving the aforementioned conditions is shown in the illustration of FIG. 8. In the present embodiment, passive (i.e. non-switching) White cell groups, which are examples of multi-pass spot generating optical systems, are disposed at each of the three output regions, which will accept as its inputs the spot arrays landing on each output region, i.e. row 0, row 2 and row 5 (see FIG. 5). Referring to FIG. 8, the light beams 100, 102 and 106 are shown arriving at their respective outputs from the optical switch 64, 66 (see FIG. 7). Included in the embodiment are three spherical mirrors 112, 114 and 116 which form multi-pass optical systems or White cells with an analog mirror 118 to the MEMS mirror 52, which may be disposed on the backside of the MEMS mirror 52. Actually, the optics of the present embodiment may be adjusted to place this analog mirror 118 in a more convenient spot, if desired. The light beams 100, 102 and 106 may pass through their respective outputs and be incident on the first 116 of three spherical mirrors 112, 114 and 116 as illustrated by the beam 120. While White cell groups are used in the present embodiment, it is understood that other suitable multi-pass spot generating optical systems may be used as noted herein above without deviating from the broad principles of the present invention.

[0055] The plane of mirror 118 comprises a passive flat mirror that has fixed tilted micro mirrors in some locations. These micro mirrors may be essentially small prisms whose hypotenuses are coated with a high reflectivity coating to direct a light beam incident at a particular pixel in a specific direction. This is in contrast to the MEMS device 52 itself, which has micro mirrors at every location that may be tilted to a variety of directions. In the beam combiner 110, the angles of the "pixels" of mirror 118 may be fixed.

[0056] Suppose that in the present embodiment the output regions of the optical switch each contains a linear array of spot image positions such as exemplified in FIG. 9, for example. FIG. 9 is an illustration of the analog mirror 118 that shows an input turning mirror 122 which is the input to the beam combiner 110 and also the output of the optical switch. A physical input turning mirror 122 may not be needed, although field lenses, not shown, may be. Each row 124 and 126 of the linear array of positions shown in FIG. 9 corresponds to a different intended output of the optical cross-connect device. In an optical cross-connection device, more than likely, only one position of the possible output spot image positions in each array 124 and 126 will actually be illuminated by an output light beam. Regardless of the position in the array 124, 126 at which the beam arrives, it should be directed to a single detector or optical fiber, corresponding to that row.

[0057] In the embodiment of FIG. 9, two different outputs of the optical switch will be considered for an exemplary description of operation of the beam combiner embodiment. One of the output beams is shown by a square symbol and the other output beam is shown by a triangle symbol in FIG. 9. In the row 126 of the linear input array 122, the fourth position from right to left is spot illuminated by the beam of the square symbol, and in the row 124, the second position from right to left is spot illuminated by the beam with the triangle symbol. Given this state as the starting state, the linear array of light passing through the output region of the optical cross-connection device or optical switch, which is the input 122 to the beam combiner 110, may be initially directed to mirror 116 as shown in FIG. 8. This spot array is imaged by mirror 116 to a new spot array in the upper right hand corner of mirror 118 as shown in the FIG. 9.

[0058] A region 130 of the mirror 118 illuminated by the new spot array includes a series of fixed micro mirrors, all tipped to some angle .theta.. The positions of the micro mirrors of region 130 correspond directly by column and row to all of the spot image locations of the imaged spot array. The tipped micro mirrors of region 130 direct the beams to mirror 114 which, in turn, directs the beams back to mirror 118 to a region 132 in the lower left hand corner thereof. Region 132 includes another series of micro mirrors, all tipped to some angle. The positions of the micro mirrors of region 132 correspond directly by column and row to all of the spot image locations of the imaged spot array from mirror 114.

[0059] At this point the entire spot array image set has been stepped sideways by some distance greater than or equal to the original spot array size. The tipped mirrors of region 132 direct the entire beam array back to mirror 116, which, in turn, directs the beam array back to mirror 118 to illuminate another set of spot images in region 134 at the top left corner thereof. At region 134, there is another corresponding set of micro mirrors which are tipped to direct the entire spot array back to mirror 116, where another set of spot images are formed.

[0060] From here on in, each of the array imaged regions of mirror 118 may include corresponding fixed micro mirror arrays that may be angled such that the light circulates only between mirror 112 and mirror 114. If it may be arranged that a flat angle, e.g. the plane of mirror 118, may be all that is needed to circulate the light beam array between mirror 112 and mirror 114, then no additional micro mirrors need to be added to mirror 118 at the array imaged regions thereof.

[0061] To achieve this result, the distance S' between the centers of curvature of mirrors 112 and 114 are set to be smaller than the centers of curvature between mirrors 114 and 116. Also, the sideways step described herein above in connection with each bounce of the beam array will be smaller to the spacing between two spot positions in the linear array. In this design configuration, some of the spot images may land on array positions or pixels that have been previously visited by another spot image of the array, but the direction of tilt of the micro-mirror is the same so there is no adverse consequence. As the beams continue to bounce, each resulting spot illumination of a bounce will move one spot position of the linear array over on each bounce. FIG. 9 shows the bounce numbers for each of the aforementioned two cases. After a predetermined number of bounces, the square symbol beam emerges from the White cell by falling through an exit port or "trap door" 136. In the present embodiment, the square symbol beam falls through the "trap door" 136 on bounce number 7. The triangle symbol beam may take somewhat longer to fall through a trap door 138, like on bounce number 11, for example.

[0062] FIG. 10 provides a three-dimensional depiction of the embodiment of the beam combiner embodiment of FIGS. 8 and 9 for a more detailed description of the operation thereof. In FIG. 10, the face of mirror 118 is laid out on a grid to show the various spot illumination patterns of the beam array between bounces and the centers of curvatures of the mirrors 112, 114 and 116 which are labeled as CC(A'), CC(B') and CC(C'), respectively. Consider the fourth spot position 140 in the linear input array 122 of FIG. 9 which is depicted at approximately index 4 in the scale of FIG. 10. The light beam from this spot position 140 is directed first to mirror 116 or C'.

[0063] Since by design this mirror's center of curvature CC(C') is located 12 units from the input spot position 140 or approximately 16 on the index scale, when the beam returns to mirror 118 from mirror 116, it is re-imaged at an approximate index location 4+2(12)=28 depicted by line 142. At position 142, there is an angled or tipped mirror 146 which directs the light beam to mirror 114 or B'. Since by design the center of curvature of mirror 114 or B' is 10 units to the left of position 142, the spot image from mirror 114 appears at an approximate location 28-2(10)=8 depicted by a line 148. The angled or tipped mirror 150 at this location 148 directs the light beam back to mirror 116 or C', creating a return spot image on mirror 118 at an approximate index location 23 depicted by line 152.

[0064] In region 152, the face of mirror 118 is flat. Thus, by design, the light beam is directed from position 152 to mirror 114 or B'. Since the center of curvature of mirror B' is set by design halfway between index locations 18 and 19, the spot image of the return beam from mirror 114 will appear approximately at an index location 14 depicted by line 154. Therein after, the light beam may circulate by design only between mirrors 114 or B' and 112 or A'. Accordingly, at the next bounce, the light beam will illuminate a spot image at approximately an index location 22 depicted by line 156, which may have already been visited on the previous bounce by the fifth positioned beam in the linear input array, but it is of no consequence. Since the centers of curvature of the mirrors 112 and 114 are spaced one-half index unit apart, the spot images of the light beam with each subsequent bounce will form one unit apart on each such bounce.

[0065] By bouncing exclusively between mirrors 112 and 114, any spot image of a particular array will scan all the array positions ahead of it, eventually landing on each one. Suppose an exit port, like a hole or "trap door", for example, is disposed at position 158 as shown in FIG. 10 (see 136 and 138 in FIG. 9), then the first spot image in the array will fall through this hole at position 158 on its third bounce, and pass to an output fiber optic cable or a light detector, for example, that may be disposed behind it. In the same bounce, the other spot images of the linear array, however, are still striking mirrors or the mirror face, and continue bouncing in the White cell formed by mirrors 112 and 114. At the fifth bounce, the second spot image of the array falls through the hole 158; at the seventh bounce, the third spot image of the array falls through the hole 158, and so on. While a hole or "trap door" is used for the exit port in the present embodiment, it is understood that other techniques, like another tipped mirror arranged to direct light out of the device, or a prism or grating cell arranged to refract or diffract light out of or from the device, for example, may be used just as well.

[0066] Note that in the foregoing described embodiment, the spot images of the linear beam array all arrive at the same exit port or hole location, with the same angle of propagation, albeit at different times. If variations in latency are a consideration, the light beams of the array may be pre-delayed in advance (in another White cell-based or other optical delay line, for example) such that when they pass through the beam combiner 110, they exit at the same time as well. The tradeoff is added complexity.

[0067] For a large cross-connection device or optical switch with many inputs and outputs, the spot images of the input beams may be in a two-dimensional array. In this case, a second White cell group may be added behind the first group 112, 114 and 116 to combine the rows of each region to a single spot. The optical losses of the beam combiner 110 are expected to be very small, since all the optical elements are passive, fixed, and may be treated with very high-reflectivity coatings.

[0068] In the operational example described in connection with the illustration of FIG. 9 herein above, an input spot pattern of two rows were used by way of example. However, it is understood that by placing an input array such that its spot images are colinear with the centers of curvatures of the mirrors 112, 114 and 116, the same operation may be performed, for one input spot array, using only one row, as described in connection with the illustration of FIG. 10. This simplification not only saves space, but also allows an implementation of the beam combiner in a planar waveguide.

[0069] FIGS. 11 and 12 illustrate an exemplary waveguide-based White cell embodiment of a beam combiner. Referring to FIGS. 11 and 12, the embodiment includes a planar waveguide 170, such that the light is guided in one dimension 172, which may be vertical direction, but acts as if it were in free space in the other direction, which may be horizontal to dimension 172. To implement a planar waveguide White cell, the implementation should include the waveguide equivalent of spherical mirrors and the waveguide equivalent of a field lens. In the present embodiment, three spherical mirrors may be implemented using three waveguide lenses 174, 176 and 178 and a tilted flat mirror or mirrors 180 behind them.

[0070] There exist different implementations of a lens for planar waveguide technology. For example, a geodesic lens, a chirped grating lens, or a Luneberg lens have all been documented in literature for several years as a suitable implementation of a waveguide lens. Any of these (or other) lens configurations may be used for the lenses 174, 176 and 178 in a planar waveguide embodiment.

[0071] Still referring to FIGS. 11 and 12, light beams may enter the waveguide 170 as spots at one end 182 thereof. Light beams from these spots are configured to travel in the direction 172 toward lens 178 or C'' and may diverge in the horizontal direction to 172. The beam is passed through a field lens 184 disposed in its path to lens 178. The combination of the waveguide lens 178 and the flat edge 180 of the waveguide back surface, which may be coated with a high-reflectivity coating, acts like a White cell objective mirror. The mirror surface 180 may be angled or "tipped," to properly locate the center of curvature of the effective mirror 178/180 or C''.

[0072] From mirror 178/180, the light beam is re-imaged at the input edge 186 of the waveguide beam combiner 170. At the first image location 188, there may be a series of angled or tipped micro mirrors 190, which could be etched into the waveguide input edge 186 or be micro prisms that are glued to the input face 186, for example. In any case, the micro mirrors 190 are also coated for high reflectivity. The tip angle of the micro mirrors 190 may be such that they send the beams to mirror 176/180 or B''.

[0073] The input edge or face 186 of the waveguide beam combiner 170 is illustrated in FIG. 11, which includes two sets of micro-prisms 190, 192 and 194, 196, for example. The rest of the input edge 186 may be left flat, but coated everywhere except for an exit port 198, which may be a hole, a gap, or "trap door", for example, which is left un-coated (or AR-coated), such that the light beam may pass through and out of the combiner 170. Other possible exit ports which have been noted herein above may be used just as well. As with the free space embodiment described supra, every spot image of the light beams exits the combiner 170 at the same point and with the same angle, but at different times.

[0074] In summary, an apparatus and method are described for combining light beams coupled from an optical cross-connection device at different spatial locations and different angles. Also, such light beams are combined to a single spot with a single arrival angle. While light beams output from a White-cell based optical cross-connection device were utilized herein above to describe various embodiments of the beam combiner by way of example, it will be appreciated that the beam combiner could be applied to other situations in which beams need to be superposed. The superposition is achieved in the exemplary embodiments by introducing all the beams into a White cell, and using the White cell to shift each beam over by one position or slot on each bounce or pass, until the light beam falls or passes through an exit port leading out to another optical device. At the exit port, the spot images may be all superimposed in space (but not in time), and have the same angle of propagation as the corresponding light beams are all coming from the same direction.

[0075] A binary White cell optical cross-connection device was used by way of example in the above descriptions for the purposes of discussion, but the beam combiner solutions apply equally well to any optical device which combines multiple beams into a single beam, a multi-pass optical cross-connection device being one example. Further, the beam combiner may be applied anywhere were beams arriving from different places and from different angles should be superimposed. A three-dimensional White cell beam combiner arrangement with spherical mirrors and lenses may be used to combine an array of rows of light beams to a single column of spot images, albeit at different times. In addition, the light beams of the array may be pre-delayed in advance (in another White cell-based or other optical delay line, for example) such that when they pass through the beam combiner 110, they may exit at the same time as well. Alternatively, the beam combiner may be embodied in a waveguide approach in which one waveguide for each row of light beams may be used to combine all of the beams.

[0076] While the present embodiment has been described herein above in connection with a plurality of embodiments, it is understood that such presentations were made merely by way of example with no intent of limiting the invention to any single embodiment or a combination of embodiments. Rather, the present invention should be construed in breadth and broad scope in accordance with the recitation of the claims appended hereto.

Claims

1. An optical beam combiner for combining a plurality of light beams, said beam combiner comprising: a plurality of spherical mirrors; and a flat mirror, said plurality of spherical mirrors and said flat mirror configured to form at least one multiple pass light beam optical arrangement for receiving said plurality of light beams and for superimposing spot images of said light beams onto a single location with a single incident angle.

2. The optical beam combiner of claim 1 wherein the single location is at an output of the beam combiner.

3. The optical beam combiner of claim 1 wherein the flat mirror includes an exit port; and wherein the single location is at said exit port of the flat mirror.

4. The optical beam combiner of claim 1 wherein the spot images of each of the plurality of light beams are superimposed onto the single location at different times.

5. The optical beam combiner of claim 1 wherein all of the plurality of spherical mirrors are configured to have their centers of curvature on the surface of the flat mirror, said centers of curvature of the plurality of spherical mirrors being spaced apart predetermined distances in relation to each other.

6. The optical beam combiner of claim 1 wherein the flat mirror comprises at least one fixed angle micro mirror disposed at each of a plurality of predetermined locations at the flat mirror.

7. The optical beam combiner of claim 6 wherein the micro mirrors are disposed on a surface of the flat mirror.

8. The optical beam combiner of claim 6 wherein each of the micro mirrors comprises a micro prism.

9. The optical beam combiner of claim 6 wherein the predetermined locations include locations on a surface of the flat mirror which are illuminated by the plurality of light beams.

10. The optical beam combiner of claim 1 wherein the plurality of spherical mirrors comprises three spherical mirrors, all three spherical mirrors having their centers of curvature on the surface of the flat mirror, said centers of curvature of the three spherical mirrors being spaced apart predetermined distances in relation to each other.

11. An optical beam combiner for combining an array of light beams, said beam combiner comprising: a plurality of spherical mirrors; and a flat mirror, said plurality of spherical mirrors and said flat mirror configured to form at least one multiple pass light beam optical arrangement for receiving simultaneously said array of light beams and for superimposing spot images of each light beam of said array onto a single location with a single incident angle.

12. The optical beam combiner of claim 11 wherein the single location is at an output of the beam combiner.

13. The optical beam combiner of claim 11 wherein the flat mirror includes an exit port; and wherein the single location is at said exit port of the flat mirror.

14. The optical beam combiner of claim 11 wherein the spot images of the array of light beams are superimposed onto the single location at different times.

15. The optical beam combiner of claim 11 wherein the flat mirror comprises an array of fixed angle micro mirrors disposed at each of a plurality of predetermined locations on a surface of the flat mirror; and wherein said arrays of micro mirrors are configured to bounce the array of light beams to pre-designated spherical mirrors of the plurality within the multiple pass light beam optical arrangement.

16. The optical beam combiner of claim 15 wherein the light beams of the array bounce simultaneously among the plurality of spherical mirrors and flat mirror within the multiple pass light beam optical arrangement.

17. The optical beam combiner of claim 16 wherein the single location is at the surface of the flat mirror; wherein the spot images illuminated by the array of light beams on the surface of the flat mirror shift along the surface at certain bounces toward said single location to be positioned thereat; and wherein the light beams of the array continue to bounce among the plurality of spherical mirrors and flat mirror within the multiple pass light beam optical arrangement until all of said spot images of the array of light beams are superimposed at said single location.

18. A waveguide-based optical White cell comprising: a waveguide having front and rear edges, the inside surfaces thereof being coated with a reflective material, wherein said front edge including an input section for the passage of at least one light beam into said waveguide; at least one waveguide lens disposed in front of the inside surface of said rear edge to form a plurality of waveguide spherical mirrors at said rear edge; a plurality of angled micro mirrors disposed at the inside surface of said front edge; and said plurality of waveguide spherical mirrors and said coated front edge configured to form at least one waveguide White cell.

19. The waveguide-based optical White cell of claim 18 wherein the White cell is configured for receiving a plurality of light beams and for superimposing spot images of said light beams onto a single location at the front edge with a single incident angle.

20. The waveguide-based optical White cell of claim 19 wherein the front edge includes an exit port; and wherein the single location is disposed at said exit port at the front edge.

21. The waveguide-based optical White cell of claim 18 wherein the rear edge of the waveguide is angled to locate the centers of curvature of the plurality of waveguide spherical mirrors at the front edge of the waveguide, said centers of curvature of the waveguide spherical mirrors being spaced predetermined distances from each other at the front edge to form the waveguide White cell.