DETAILED DESCRIPTION OF THE INVENTION

[0028] Referring to FIG. 2 , shown is an exemplary computer system 30 , such as a personal computer that includes a power supply 32 , a processor 34 , input/output device controller and associated memory (I/O controller) 36 , main memory 38 , expansion slots 40 and expansion cards 40 a , 40 b , 40 c and 40 d . The expansion slots 40 are in electrical communication with the power supply 32 over a power bus 42 . The power bus 42 includes multiple lines, each of which is dedicated to carrying a single voltage level. A main system data bus 44 is in data communication with processor 34 , expansion slots 40 and main memory 38 . Main data bus 44 includes eight to sixty-four different lines, depending upon the data transfer protocol supported by the system 30 , e.g., ISA, EISA, or MCA protocols and the like. Main data bus 44 carries data transferred between processor 34 , main memory 38 and expansion slots 40 . An address bus 46 comprising, for example, twenty lines is in data communication with main memory 38 , processor 34 and expansion slots 40 . Address bus 46 carries information that specifies the address from, or to, data is to be moved. To facilitate data transfers, a control bus 48 is included that has a plurality of lines placing main memory 38 and expansion slots 40 in data communication with I/O controller 36 .

[0029] Referring to both FIGS. 2 and 3 , as mentioned above, each of the expansion slots 40 is adapted to receive an expansion card 40 a , 40 b , 40 c and 40 d . One or more optical channels facilitate communication between two or more of the expansion cards 40 a , 40 b , 40 c and 40 d . One optical channel includes one or more sources of optical energy 48 a mounted to expansion card 40 a , and one or more optical detectors 50 a mounted to expansion card 40 b and in data communication with the source of optical energy 48 a . A HOE 52 a is disposed between the source of optical energy 48 a and the detector 50 a . A second optical channel includes one or more sources of optical energy 48 b mounted to expansion card 40 b , and one or more optical detectors 50 b mounted to expansion card 40 a and in data communication with the source of optical energy 48 b . A HOE 52 b is disposed between the source of optical energy 48 b and the detector 50 b.

[0030] Source of optical energy 48 a directs optical energy 54 a along a path 56 a in which the detector 50 a lies. The HOE 52 a is disposed in the optical path 56 a . Source of optical energy 48 b directs optical energy 54 b along a path 56 b in which the detector 50 b lies. The HOE 52 b is disposed in the optical path 56 b . each of the HOEs 52 a and 52 b has both a refractory function and a holographic transform function enabling the HOEs 52 a and 52 b to concurrently filter and refract the optical energy propagating therethrough. In this manner, the HOEs 52 a and 52 b filters the optical energy 54 a and 54 b , respectively so that the optical energy passing therethrough to impinge upon the optical detectors 50 a and 50 b , respectively, having desired characteristics.

[0031] HOE 52 a and 52 b are identical in construction and, therefore, only HOE 52 a will be discussed, but it should be borne in mind that the discussion with respect to HOE 52 a applies with equal weight to HOE 52 b . HOE 52 a is a refractory lens having a bulk hologram recorded therein that defines a holographic transform function. The bulk hologram facilitates characterizing the optical energy 54 a to have desired characteristics that may improve detection, by the optical detector 50 a , of information contained in the optical energy 54 a . For example, the transform function may allow a specific wavelength to pass through the lens, diffracting all other wavelengths to deflect away from the optical detector 50 a . Alternatively, the transform function may allow only a certain polarization of the optical energy 54 a to propagate therethrough, diffracting all other polarizations away from the optical detector 50 a.

[0032] The refractory function of the HOE 52 a facilitates impingement of the optical energy 54 a onto the optical detector 50 a . In this manner, the precise alignment of the optical detector 50 a with respect to the source 48 a and, therefore, the path 56 a may be relaxed. This is beneficial when facilitating communication between expansion cards, such as 40 a and 40 b , because the mechanical coupling of the expansion cards 40 a and 40 b to the respective slots 40 would typically make difficult precisely aligning source 48 a with the detector 50 a.

[0033] Referring to FIG. 4 , the HOE 52 a is a lens 58 having an arcuate surface 60 , e.g., cylindrical, spherical and the like with a bulk holographic transform function formed therein. The bulk holographic transform function is shown graphically as periodic lines 62 for simplicity. The bulk holographic transform function 62 is recorded in substantially the entire volume of the lens 58 through which optical energy will propagate. The transform function 62 is a periodic arrangement of the space-charge field of the material from which the lens 58 is fabricated. To that end, the lens 58 may be formed from any suitable photo-responsive material, such as silver halide or other photopolymers. In this manner, the lens 58 and the bulk holographic transform function 62 are integrally formed in a manner described more fully below. Although the surface 64 of the lens 58 disposed opposite to the spherical arcuate surface 60 is shown as being planar, the surface 64 may also be arcuate as shown in surface 164 of lens 158 in FIG. 5 .

[0034] Referring to both FIGS. 3 and 4 , were it desired to further control the shape of optical energy propagating through lens 58 , a Fresnel lens 258 may be formed opposite to the spherical surface 260 . To that end, the Fresnel lens 258 includes a plurality of concentric grooves, shown as recesses 258 a 258 b and 258 c that are radially symmetrically disposed about a common axis 256 . Thus, the lens 258 may have three optical functions integrally formed in a common element, when providing the bulk holographic transform function 262 therein, which facilitates creation of a well defined optical channels between expansion cards 40 a and 40 b.

[0035] Facilitating communication between expansion cards 40 a and 40 b over optical channels increase the bandwidth of the computer system 30 's bus systems. Specifically, the transfer of power and data between the expansion cards 40 a and 40 b and the computer system 30 is bifurcated. The power to the expansion cards 40 a and 40 b is transferred over power bus 42 and the data transfer between two or more expansion cards may be achieved over one or more optical channels. To that end, the expansion cards 40 a and 40 b are made backwards compatible with existing technology. This is shown by the implementation of standard expansion cards 40 c and 40 d along with expansion cards 40 a and 40 b , as well as the compatibility of expansion cards 40 a and 40 b with standard expansion slots 40 . The presence of the optical channels, however, reduces the need to transfer information between the expansion cards 40 a and 40 b over the main data bus 44 , as well the need to transfer information over the address bus 46 or the control bus 48 , were appropriate control circuitry included on the expansion cards 40 a and 40 b . Thus, employing the optical channels as described above, the computer system 30 bus bandwidth may be increased.

[0036] Referring to both FIGS. 2 and 7 , as mentioned above the expansion cards 40 a and 40 b may each include multiple sources of optical energy 48 a and multiple detectors 50 a . To that end provided are an array of sources of optical energy 348 , shown generally as optical emitters 348 a - 348 p , and an array of optical detectors 350 , shown generally as optical receivers 350 a - 350 p . The optical emitters 348 a - 348 p generate optical energy to propagate along a plurality of axes, and the optical receivers 350 a - 350 p are positioned to sense optical energy propagating along one of the plurality of optical axes. Specifically, the array 348 is an (XxY) array of semiconductor lasers that produce a beam that may be modulated to contain information. The array 350 may comprise of virtually any optical detector known, such a charged coupled devices (CCD) or charge injection detectors (CID). In the present example, the array 350 comprises of CIDs arranged in an (MxN) array of discrete elements. The optical beam from the each of the individual emitters 348 a - 348 p may expand to impinge upon each of the detectors 350 a - 350 p of the array 350 if desired. Alternatively, the optical beam from each of the individual emitters 348 a - 348 p may be focused to impinge upon any subportion of the detectors 350 a - 350 p of the array 350 . In this fashion, a beam sensed by one of the detectors 350 a - 350 p of the array 350 may differ from the beam sensed upon the remaining detectors 350 a - 350 p of the array 350 . To control the wavefront of the optical energy produced by the transmitters 348 a - 348 p , the HOE 52 , discussed above with respect to FIGS. 3 - 6 may be employed as an array of the lenses 252 , shown more clearly in FIG. 8 as array 400 .

[0037] Specifically, referring to FIGS. 5 and 7 , the individual lenses 458 of the array are arranged to be at the same pitch and sizing of the array 348 . The numerical aperture of each of the lenses 458 of the array 400 is of sufficient size to collect substantially all of the optical energy produced by the emitters 348 a - 348 p corresponding thereto. In one example, the array 400 is attached to the array 348 with each lens resting adjacent to one of the emitters 348 a - 348 p . To provide the necessary functions, each of the lenses of the array 400 may be fabricated to include the features mentioned above in FIGS. 2 - 4 . As a result, each of the lenses 458 of the array may be formed to having functional characteristics that differ from the remaining lenses 458 of the array. In this manner, each beam produced by the array 348 may be provided with a unique wavelength, polarization or both. This facilitates reducing cross-talk and improving signal-to-noise ratio in the optical communication system 310 .

[0038] Specifically, an additional array of lenses 400 b that match the pitch of the individual receivers 350 a - 350 p of the array 350 , shown more clearly in FIG. 10 . The lenses may be fabricated to provide the same features as discussed above with respect to array 400 , shown in FIG. 8 .

[0039] Referring to FIGS. 7, 8 and 10 each of the emitters 348 a - 348 p of the array 348 would then be uniquely associated to communicate with only one of the detectors 350 a - 350 p of the array 350 . In this manner, the emitter 348 a - 348 p of the array 348 that is in data communication with one of the one of the receivers 350 a - 350 p of the array 350 would differ from the emitters 348 a - 348 p in data communication with remaining receivers 350 a - 350 p of the array 350 . This emitter/receiver pair that were in optical communication is achieved by having the properties of the lens 458 in array 400 match the properties of the lens 458 b in array 400 b . It should be understood, however that one of the emitters 348 a - 348 p may be in data communication with any number of the receivers 350 a - 350 p by multiple lenses 458 b matching the properties of one of the lenses 458 . Similarly, one of the multiple emitters 348 a - 348 p may be in optical communication with one or more of the detectors 350 a - 350 p by appropriately matching the lenses 458 to the lenses 458 b.

[0040] In one example, superior performance was found by having the array 350 sectioned into (mxn) bins, with each bin corresponding to a particular polarization and/or wavelength that matched a particular polarization and/or wavelength corresponding to a emitter 348 a - 348 p . Thus, were a beam from one or more of the emitters 348 a - 348 p to flood the entire (MxN) array 350 or multiple (mxn) bins, only the appropriate detectors 350 a - 350 p sense information with a very high signal-to-noise ratio and discrimination capability. It will be noted that the (mxn) bins can also be effectively comprised of a single sensing pixel (element) to exactly match the (XxY) array.

[0041] Additional beam-sensor discrimination may be achieved by employing emitters 348 a - 348 p having different wavelengths or by incorporating up-conversion processes that include optical coatings applied to the individual emitters 348 a - 348 p or made integral therewith. One such up-conversion process is described by F. E. Auzel in “Materials and Devices Using Double-Pumped Phosphors With Energy Transfer”, Proc. of IEEE, vol. 61. no. 6, June 1973.

[0042] Referring to FIGS. 3, 10 and 11 , the system 500 employed to fabricate the lens 58 and the lens arrays 400 a and 400 b includes a beam source 502 that directs a beam 504 a into wave manipulation optics 506 such as a ¼ waveplate 508 so that the beam 504 b is circularly polarized. The beam 504 b impinges upon polarizer 510 so that the beam 504 c propagating therethrough is linearly polarized. The beam 504 c impinges upon a Faraday rotator 512 that changes birefringence properties to selectively filter unwanted polarizations from the beam 504 c . In this manner, the beam 504 d egressing from the rotator 512 is linearly polarized. The beam 504 d impinges upon a beam splitter 514 that directs a first subportion 504 e of beam 504 d onto a planar mirror 516 . A second subportion 504 f of the beam 504 d pass through the splitter 514 . The first and second subportions 504 e and 504 f intersect at region 520 forming an optical interference pattern that is unique in both time and space. A photosensitive sheet 558 is disposed in the region so as to be exposed to the optical interference pattern. The interference pattern permeates the photosensitive sheet 558 and modulates the refractive index and charge distribution throughout the volume thereof. The modulation that is induced throughout the volume of the photosensitive sheet 558 is in strict accordance with the modulation properties of the first and second subportions 504 e and 504 f . Depending upon the photosensitive material employed, the holographic transform function may be set via thermal baking.

[0043] Referring to FIGS. 11 and 12 , an arcuate surface is formed in the photosensitive sheet 558 by adhering a photosensitive layer 600 to a sacrificial support 602 , such as glass, plastic and the like to form a photosensitive substrate 604 . Typically, the photosensitive layer 600 is tens of microns thick. As shown in FIG. 13, a photo resist layer 606 is deposited onto the photosensitive layer 600 and then is patterned to leave predetermined areas exposed, shown as 608 in FIG. 14 , defining a patterned substrate 610 . Located between the exposed areas 608 are photo resist islands 612 . The patterned substrate 610 is exposed to a light source, such as ultraviolet light. This ultraviolet light darkens the volume of the photo resist layer 600 that is coextensive with the exposed areas 608 being darkened, i.e., become opaque to optical energy. The volume of the photosensitive layer 600 that are coextensive with the photo resist islands 612 are not darkened by the ultraviolet light, i.e., remaining transparent to optical energy. Thereafter, the photo resist islands 612 are removed using standard etch techniques, leaving etched substrate 614 , shown in FIG. 15 .

[0044] The etched substrate 614 has two arcuate regions 616 that are located in areas of the photosensitive layer 600 disposed adjacent to the islands 612 , shown in FIG. 14 . The arcuate regions 616 of FIG. 15 result from the difference in exposure time to the etch process of the differing regions of the photosensitive layer 600 .

[0045] Referring to FIGS. 10 and 16 , a subsequent etch process is performed to form array 400 . During this etch process the support is removed as well as nearly 50% of the photosensitive layer 600 to form a very thin array. The array 400 is then placed in the apparatus 500 and the bulk holographic transform functions are recorded in the arcuate regions 616 that define the lenses 458 , as discussed above. A Fresnel lens may also be formed on the lenses 458 of the array using conventional semiconductor techniques. Thereafter, the lenses may be segmented from the photo resistive sheet or MxN subarrays of lenses may be segmented therefrom.

[0046] Lenses with differing transform functions are formed on differing photosensitive sheets 558 . Specifically, the transform function is defined by the interference pattern formed by the first and second subportions 504 e and 504 f intersecting at region 520 . This interference pattern is unique in both time and space. As a result, each of the lenses formed on the sheet 558 would have substantially identical holographic transform functions. To create lenses with differing transform functions, an additional photosensitive sheet 558 would be employed and, for example, the Faraday rotator 512 may be rotated to provide the lenses formed on the photosensitive sheet 558 with a holographic transform function that differs from the holographic transform function associated with the lenses formed on a previous photosensitive sheet 558 . Therefore, lenses 458 a associated with the first array 458 would come from differing sheets 558 if the lenses were to have differing holographic transform functions.

[0047] Although the invention has been described in terms of specific embodiments, one skilled in the art will recognize that various changes to the invention may be performed, and are meant to be included herein. For example, instead of forming the arcuate regions 616 using standard etch techniques, the same may be formed by exposing the substrate 610 , shown in FIG. 12 , to thermal energy. In one example, the substrate 610 is convectionally heated, and the photo resist layer 606 is patterned to control the regions of the photosensitive layer 600 that may expand. In another example, the photosensitive layer is heated by conduction employing laser ablation/shaping. Specifically, a laser beam impinges upon areas of the photosensitive layer 600 where lens are to be formed. The thermal energy from the laser beam causes the photosensitive layer 600 to bubble, forming arcuate regions 616 thereon shown in FIG. 13 . Therefore, the scope of the invention should not be based upon the foregoing description. Rather, the scope of the invention should be determined based upon the claims recited herein, including the full scope of equivalents thereof.