Improved Resonant Cavity Dielectric Filter Dense Wavelength Division Multiplexer/Demultiplexer
Field of the Invention The present invention relates to an optical wavelength demultiplexing system. The present application is a continuation-in-part of U.S. Patent Application Number 09/086,360, filed May 29, 1998 to Kapany, et al. the disclosure of which is specifically incorporated herein by reference. Background of the Invention
The optical uses to which coatings of thin dielectric films have been put recently are varied and diverse. Coatings to eliminate unwanted reflection from surfaces are often used in a variety of applications.
Multi-layer, non-absorbing beams splitters and dichroic mirrors can be purchased commercially for various applications. One commonly used example of a multilayer periodic system based on dielectric thin films is the quarter-wave stack. The mathematical treatment of a quarter-wave stack follows relatively straight forwardly from the consideration of a linearly polarized wave impingent on a thin dielectric film between two semi- infinite transparent media. In practice, this might correspond to a dielectric layer a fraction of a wavelength thick, deposited on a glass surface or other material. Through relatively straight forward boundary value analysis, the relation between the electric and
magnetic field vectors at boundaries I and II are given by the following expression:
where k0h = k0 (nιdcosθiIι:) / cosk0h (iύnkj γ is the
Hi γjύnkji cosk i Hπj optical path length; k0 is the
propagation vector in free space; θiττ is the angle of
incidence with respect to the normal at boundary II;
___
Α = (ni/cosθ n) ; n-j. is the index of refraction of
the film and d is the film thickness.
This can be expressed as :
JII
Hτ = M, H ( 2 )
I I
and Mi is the characteristic matrix which relates the field vectors at the two adjacent boundaries. For two films, straight forward analysis yields
Ei ■■--iii Hz M M2 HIII
and so forth with each successive film multiplying by the appropriate characteristic matrix for the additional film to the relation. Accordingly, it can be shown that the characteristic matrix of a stock of films is a 2x2
matrix. Thereafter, straight forward analysis will render the transmission and reflection coefficient.
From the foregoing analysis, it can be shown that to determine the transmission and reflection coefficients for any configuration of thin films, it is necessary only to compute the characteristic matrices for each film, multiply them, and through straight forward analysis determine the reflection and transmission coefficients. One special case of dielectric thin film analysis is that in which the angle of incidence is normal to the plane of incidence for the angle of incidence is parallel to the normal. In this case, the reflectance can be determined by multiplying the reflection coefficient by its complex conjugate. In the case where
k0h = - π 2 the reflectance is
(npn2 - nM
Ri = (n0n2 - nx 2)
where n0 is the index of refraction of air; ni is the index of refraction for the first film and n is the index of refraction of the second film. It follows that this expression will equal zero when n0n2 = ni2.
This special case where the reflectance is zero means that all of the incident light is transmitted, and a narrow band pass filter is realized. This condition of choosing the film thickness so that the optical thickness h is a multiple of λo /4 is the basis of quarter wave filters of varying types. By choosing h to be λo/4, then d from the above expression is λf /4, where
/ni at normal incidence.
As stated previously, the simplest kind of periodic system is a quarter wavelength stack, which is made up of a number of quarter wavelength layers with alternating layers of higher and lower indices of refraction and is based on the analysis discussed above. While the discussion above was clearly for the case of normal incidence relative to the quarter length stack, at other than normal incidence, up to approximately 30 degrees, there is very little degradation in the response of the thin film coatings. In general, the effect of increasing the angle of incidence relative to the normal is a shift in the whole reflectance curve down to slightly shorter wavelengths. Further, the while the analysis discussed above is for linearly polarized light, the polarization of the light passed by a quarter wavelength stack filter can be an issue. To this end, the polarization of the light can be of some concern in filter design, as the filter will behave quite differently for the two planes of polarization (S and P polarization) . However, for angles of incidence up to
approximately 20 degrees from the normal, these effects are minimal.
Quarter wavelength stacks for a filter can be put to a variety of uses in applied optical systems. As discussed above, the anti-reflection coating is often a multi-layer quarter wavelength stack for a particular wavelength. Another application of a quarter wavelength stack of thin dielectric films is in the demultiplexing of an optical signal. To this end, as can be appreciated, if a multiplexed signal of a number of wavelengths is incident on a multi-layer stack of dielectric films with the thickness chosen to be transmissive of a desired wavelength, the desired component of the incident light (desired wavelength) will be transmitted while all other wavelengths will be reflected. Further analysis of the basic principals of quarter wavelength dielectric thin films leads to a number of observations. An increase in the ratio of the values of the of the indices of refraction between adjacent layers results in an increase in the width of the band of wavelengths that are reflected. Additionally, the total reflectance increases as the number of layers increases. Accordingly, for typical commercial filters, the number of dielectric layers in the thin film stack (with three to four resonant cavities) is on the order of 70 - 100 layers. Further details of the use of dielectric films for filters and other applications can be found in text books on optics,
for example Optics , by Hecht and Zajac, copyright 1974, 4th printing, 1979, pages 311 - 316, the disclosure of which is specifically incorporated herein by reference. One example of the use of multi-layer thin film filter for demultiplexing and multiplexing an optical signal is disclosed in US Patent 5,583,683 to Scobey, the disclosure of which is specifically incorporated by reference herein. The disclosure of Scobey relies upon an optical block which has an optical port for passing multiple wavelength collimated light in multiple ports arrayed in a spaced relation relative to one another along a multi-port surface of the optical block. At each of the multiple ports, the continuous interference filter transmits a different wavelength sub-range of the multiple wavelength collimated light passed by the optical port and reflects the other wavelengths. The light not transmitted through the first port is reflected to strike a second port, at which a second wavelength is transmitted and all other light again reflected. The reflected optical signals thus cascade in a multiple bounce sequence down the optical block of the device, sequentially removing each channel of the multiplexed signal. The reference to Scobey relies upon a continuous, variable thickness interference filter, with the preferable structure being a multi-cavity interference filter (Fabry Perot filter) carried on the multi-port surface of the optical block to provide multiple ports. Because this continuous interference
filter extending over the multi-port surface has a different optical path length at each of the multiple ports, the wavelength or range thereof passed by the filter at each such port will differ from that passed at other ports.
While the structure to Scobey does provide for a relatively straight forward technique for multiplexing and demultiplexing to form a dense channel wavelength division multiplexer (DWDM) , there are certain disadvantages to the reference to Scobey. To this end, such a structure does in fact require tuning at each of the ports. This tuning is effected by varying the thickness of the layers of the film stack coating with Fabry-Perot etalon therein. Accordingly, it is necessary to have more than one multi-layer stack for carrying out the multiplexing and demultiplexing as set forth in the Scobey reference. Such a structure requires relatively expensive technology for the multi- cavity interferometric filter fabrication in precise increments, typically one filter per wavelength or wavelength band.
The application to Kapany, et al. from which the present application claims priority discloses a novel technique for using the same dielectric stack to effect multiplexing/demultiplexing in optical system. While the reference to Kapany et al. is quite advantageous in reducing the cost of the overall multiplexer/demultiplexer for dense wavelength division
multiplexing/demultiplexing systems, there are yet further improvements that can be made. To this end, the system to Kapany, et al. is advantageous in its use of the same dielectric stack for each filter element. However, for each wavelength, a filter is required.
Accordingly, for a system which is used to demultiplex for wavelength channels with center wavelengths λl t %2 ,
2 ι ^- ror individual filter are required. Accordingly, while the system achieves an overall reduction in complexity and the cost through the use of the same dielectric filter oriented at a particular angle for each wavelength channel with a particular center wavelength, it would be even more desirable to have a system which is less complex and achieve even further efficiencies.
Accordingly, what is needed is an improved structure for effecting multiplexing and demultiplexing in a dense channel wavelength division multiplexer which reduces the complexity of the multiplexer/demultiplexer system.
Summary of the Invention
The present invention is drawn to a multi-channel dense WDM demultiplexer having a single segmented filter element, the filter element being one of the filter elements described in the above captioned patent application to Kapany, et. al. For purposes of discussion, and exemplary embodiment of the invention shown in Figure 1, conveys the salient aspects of the
invention of the present disclosure. A multi-channel dense WDM add-drop arrangement has an input array with a pre-determined number of input fibers. Each input fiber has a predetermined number of wavelength channels with a center wavelength for each channel. For example, in the embodiment shown in Figure 2, there are 5 input fibers for the input fiber array. Fiber 1 has 5 input wavelength channels with center wavelengths λi, λ2 f
λ3,λ4f and λ5 respectively. Light from Fiber 1 is incident upon the collimating lens and is thereby collimated. This light is incident on the dielectric stack filter at a pre-determined angle of incidence. The angle of incidence is chosen so that the wavelength channel having center wavelengths λ is transmitted through the filter and is therefore demultiplexed. The remaining wavelengths λ2 f λ3,λ f λ5 are reflected and are focused by a second lens as shown. This reflected light is input to Fiber 2 , and this light is thereafter incident upon the collimating lens, which collimates the light so that it is incident upon the filter at a second predetermined angle. This second predetermined angle is chosen so that the filter transmits the wavelength channel having center wavelengths λ2 and reflects the
wavelengths channels having center wavelengths λ3,λ4f λ5. The process continues with the next input Fiber 3 serving as the input fiber and demultiplexing of the signal is achieved. As can be appreciated by one of
ordinary skill in the art, by basic physical principles, a multiplexer is achieved by a reversal of the sequence shown in Figure 1. To this end, in this multiplexing scheme, that which is labeled input fiber array in Figure 1 would it in fact be the output fiber array, and the output fiber array in Figure 1 would be the input fiber array in the multiplexer. Brief Description of the Drawings
Figure 1 is a perspective view of the preferred embodiment of the invention of the present disclosure showing five input fibers for multiplexing/demultiplexing five wavelength channels. Figure 2 is a an enlarged perspective view of a segment of the basic system shown in Figure 1 with five input fibers in the input fiber array.
Figure 3 is a perspective view showing the invention of the present disclosure with an embodiment using a optical relay system in lieu of loops of optical fiber. Figure 4 is a perspective view of a single fiber collimator in which a point source (optical fiber) emitter is the source of diverging light which is collimated by the lens thereby resulting in parallel light rays. Figure 5 is a perspective view of a fiber multicollimator in which a graded refractive index (GRIN) lens with a second curved surface is used.
Figure 6 show tabular representations of the input fiber array correspondence to the reflect output fiber array, showing the wavelength channels in Examples I. Figure 7 is a perspective view of the orthoscopic optics used in the embodiment of Example II. Figure 8 is a perspective view of the multiplexer/demultiplexer of the embodiment of Example II.
Figure 9a is a cross-sectional view of the input and output reflect fiber array coordinate system.
Figure 9b is a cross-sectional view along the line 9-9 of Figure 8.
Figure 10 is a tabular representation for the 8 channel DWDM showing coordinates of the input fibers, the center wavelength for each channel and the correspondence between input and output reflect fibers. Detailed Description of the Invention
The foundation of the invention of the present disclosure is the dielectric stack filter as is discussed above and as disclosed in the above captioned application to Kapany, et al. This dielectric stack filter in the preferred embodiment has approximately 70- 100 layers of dielectric thin film, each having a thickness equal to one-quarter wavelength of the wavelength chosen at orthogonal or normal incidence. In the preferred embodiment the filter has three to four resonant cavities. As can be readily appreciated by one of ordinary skill in the art, a dielectric stack filter
having multiple layers has an effective index of refraction denoted, ne. As the angle of incidence changes from the normal to the dielectric stack, the center wavelength transmitted through the filter shifts to shorter wavelengths. The magnitude of this shift in wavelength is a function of this effective index of refraction of the filter. The dependence of the transmitted wavelength λ on the angle of incidence is well known to be: λt (θ) = λn (1 I 1/ ne 2 sin2 B) h (1)
where λn is the transmittal wavelength at normal incidence; ne is the effective index of refraction; and θ is the angle of incidence relative to a vector normal to the dielectric stack.
The basis of the invention of the present disclosure relies upon the physics of the dielectric stack filter as defined above mathematically in the above equation. To this end, the change in the angle of incidence from the normal will result in a change in the transmitted center wavelength. Furthermore, the filter bandwidth and the channel bandwidth are chosen to match as closely as possible for a particular center wavelength. Thus, the center wavelength of the filter at a particular angle of incidence governs the transmitted wavelengths through a particular segment of
the filter. In the invention of the present disclosure, it is envisioned that four to eight wavelength channels be multiplexed/demultiplexed with the wavelength channels being separated by a fixed frequency in the range of approximately 50-200 GHz is used. To this end as the angle of incidence increases beyond a certain point, losses can be too significant for efficient multiplexing/demultiplexing. Accordingly, the dielectric stack filter is chosen so that 4-8 wavelength channels can be accommodated. Increasing the number of channels beyond about eight can be accomplished either by concatenating more than one of the below described embodiments or by using a wavelength band demultiplexer which separates groups of individual wavelength channels onto individual fibers. To this end, such a band demultiplexer with n output fibers with n DWDM demultiplexers, each having m individual output fibers as described above results in a compound demultiplexer having n x m output channels. Such multiplexer/demultiplexer schemes for more than eight wavelengths channels can incorporate the multiplexing/demultiplexing invention of the present disclosure. Accordingly, a multiplexer/demultiplexer that incorporates the invention of the appended claims is considered within the scope of the present invention. Before discussing the examples as set forth below, it is important to mention that the optical systems used
in the invention of the present disclosure must, in most circumstances, take into account optical aberrations and thereby should attempt to correct these aberrations. While it is true that in an optical system it is virtually impossible to design a system which is free from all first order and higher order aberrations, a suitable compromise as to the magnitude of the aberrations must be made. In the invention of the present disclosure, the return loop, or the optical path from the input fiber array to the filter and then to the reflect output fiber array should have an optical system which reduces the Petzval field curvature as well as distortion, and in particular coma, spherical, negative or barrel distortion as well as positive or pin-cushion distortion. Furthermore, as will be appreciated by one of ordinary skill in the art, there is a necessary trade-off in any optical system which attempts to flatten the field to compensate for field curvature, and distortion (for example, pin-cushion or barrel distortion) . Further details of aberrations within the limits of geometrical optics can be found in various text books on optics, for example, OPTICS, by Hecht and Zajac, Copyright 1974, Fourth Printing February 1979, pages 1 75-186, the disclosure of which is incorporated by reference herein. Finally, the optical systems described in Examples I and II are used to correct aberrations and distortion and are merely exemplary.
The most important features of a system for the present invention is that it renders a flat field and corrects positive or negative distortion. Many types of symmetrical systems are possible to achieve this goal, and a further example is a Cooke triplet, known in the art.
The examples discussed hereinbelow are examples of the invention of the present disclosure using a single dielectric multilayer stack filter to effect the multiplexing and demultiplexing. These examples are for the purposes of illustration and while they disclose two embodiments of the invention, certainly they are not exhaustive or limiting. Accordingly, it is clear that modifications and variations of the teaching of the invention of the present disclosure are within the perview one of ordinary skill of the art having had the benefit of the present disclosure.
EXAMPLE I Turning to Figure 1, the input fiber array is shown at 100. This input fiber array has 4 input fibers in the embodiment shown in Figure 1. These are labeled as 1,2, 3, and 4. This is an exemplary embodiment, and is intended in no way to be limiting. In the embodiment shown in Figure 1, input Fiber 1 has four wavelength channels having respective center wavelengths λi, λ2,
λ3 and λ4. As stated previously, and in the parent
application to the present application, each channel has a center wavelength (e.g. λx for channel 1) with a channel bandwidth which is defined by (as closely as possible) the filter bandwidth for the particular center wavelength of the channel. For example, for DWDM systems having 100 GHz center frequency channel separation, a customary bandwidth is 25 GHz. At approximately 1550 nm this corresponds to channels having center-to-center spacing on the order of 0.8nm and bandwidths on the order of 0.2nm. Input fiber 1 emits light as shown which is diverging and impingent upon the collimating lens 102. The collimated light from the lens 102 is impingent upon the filter element 101 as shown. The orientation input optical fiber 1, collimating lens and filter are chosen so that light from input optical fiber 1 is impingent upon the filter 101 at an angle of incidence (with respect to the normal) chosen so that the wavelength channel having center wavelength λi is transmitted. This is shown at 103. This demultiplexed wavelength channel is thereafter impingent upon a second lens 104 which is a converging lens and focuses, the light upon an output fiber in the transmit output fiber array shown as 105. The remaining wavelength channels λ2, λ3, λ4 are reflected by the dielectric stack filter based upon the physical principles of a dielectric stack filter as described
above and in the parent application to the present application. This light is impingent upon a third lens 104 which is also a converging lens in this arrangement and focuses the light at an image plane 106 where the reflect output fiber array (not shown) is placed.
Turning to Figure 2, an enlarged view of the light path from the input fiber array to the reflect output fiber array is shown. For clarity of discussion, the filter elements as well as the converging lens' (104 in Figure 1) and transmit output fiber array 105 in Figure 1) are eliminated. Furthermore, the structure shown in Figure 5 shows a five channel multiplex/demultiplex scheme. The structure shown in Figure 2 shows the input fiber array from input fiber 1 as a cone of light impingent upon the collimating lens. This cone of light is shown at 201 with the collimating lens is shown at 202. The light is collimated by the collimating lens 202 and is impingent upon the filter as discussed above with channel 1 having center wavelength λi transmitted.
The remaining wavelength channels λ2, λ3, and λ4 are reflected and are incident upon the second lens, a converging lens in this arrangement shown at 203. The converging lens focuses the light cone shown generally at 204 with an image point 205 in the image plane. An optical fiber, which is designated as the "reflect output fiber" 1' is placed at this point. This output
fiber 1 ' is connected to the input fiber 2 shown at the input fiber array in figure 2. Accordingly, the light that emerges from the input fiber 2 contains channels 2, 3, 4 and 5 having center wavelengths λ2, λ3, λ4, and λ5. The process continues with the orientation of input fiber 2 chosen so that the light is impingent upon the filter at an angle of incidence which transmits the channel having center wavelength λ2 and reflects all other channels. This reflected channel is impingent upon the lens 203, which focuses upon a second "reflect output fiber" shown 2'. The reflect output fiber 2' is located at a point 206, which is an image point in the image plane. Reflect output fiber 2' is connected into input fiber 3, and the process of demultiplexing continues. To this end, reflect output fiber 2' is connected to input fiber 3 and input channels 3, 4 and 5 having center wavelengths λ3, λ4, and λ5 respectively are emitted from input fiber 3 shown at the input fiber array in figure 2. Light from input fiber 3 is incident upon the filter 202 at an angle of incidence chosen so that channel 3 having a center wavelength λ3 is transmitted to the output fiber array (not shown in Figure 2) and the remaining wavelength channels 4 and 5 are reflected and focused at the image plane at an image point 207. At the image point 207, the reflect output fiber 3' is located. Reflect output fiber 3' is
connected to input fiber 4 in the input fiber array, with the process thereby continuing and with channel 4 having center wavelength λ4 being transmitted through the filter into the output fiber array. The reflected channel 5 have center wavelength λ5 is focused at the image plane at image point 208 where reflect output fiber 4' is located. Reflect output fiber 4' is connected to input fiber 5 which is disposed so that it is incident upon the filter at an angle of incidence which enable the channel 5 having center wavelength λ5 to be transmitted. In the structure shown in Figure 2, if one where to have more than 5 channels. The basic optical system used in the embodiment of Example I is discussed presently. Following from basic geometrical optics in which the small angle approximation applies, the point source emitter on the left in Figure 4 is the source of a diverging beam that is collimated by the lens to form parallel light on the right. The lens shown in Figure 4 will take a collimated light beam and focus it at a focal point shown at F in Figure 4, where F is the focal point of the lens. The array multicolli ator of the invention of the present disclosure uses a parallel fiber array spaced one focal length from a lens, preferably a lens designed for infinity conjugate use in having a clearly flat global surface shown at f in Figure 2. The lens shown in
Figure 5 is a graded refractive index lens with a curved second surface. In practice, a lens optimized for flatness of the focal surface will give minimal diversions for all the collimated beams. The graded refractive index lens (GRIN) as shown in Figure 5 takes the light output from an array of parallel fibers at the focal surface f which give rise to diverging beam shown at 501 in Figure 5. These are approximated by light cones, with the divergent light from the individual optical fibers of the optical fiber array being collimated by the GRIN lens 502 into a collimated light beams 503. Again, from basic principles of geometrical optics, collimated light 503 which is impingent upon the lens 502 will be focused at the focal surface designated in the figure as f with the converging light 501 as shown. These basic lens elements are used in the invention of the present disclosure. For example, in the demultiplexing mode, lens element 102 is a collimating lens taking the diverging beams from the input fiber array and collimating the light therefrom. Lens elements 104 and 105, on the other hand, act as converging lenses taking the collimated light and focusing them at focal points in a focal surface. This is more readily seen, for example in Figure 2 where in the multiplexing mode lens element 203 focuses light at focal surface onto the reflect output fiber arrays at point 205, 206, 207 and 208, for example.
EXAMPLE I I
Turning to Figures 7 and 8, an alternative embodiment of the invention of the present disclosure is shown. This embodiment which makes use of a modified orthoscopic lens system to image both the transmit and receive arrays has certain advantages in keeping with the discussion of aberrations above. To this end, the orthoscopic lens system, in the symmetric arrangement shown in Figures 7 and 8, has a relatively high degree of freedom from distortion, allowing the use of linear fiber arrays offset laterally from the optical center of the system. Systems with more distortion might require the use of curved arrays to compensate for the distortion. To this end, turning to Figure 7, for a single input fiber with the input fiber disposed at 701, the orthoscopic system shown at 702 focuses the light upon the filter 703, which again has the wavelength dependence as discussed above. The wavelength λi in this case is incident upon the filter at the requisite angle so as to be transmitted by the filter at 704. This light from wavelength channel λ, is thereafter focused on the image plane 705 by a second orthoscopic eye piece 706. The other wavelength channels from input fiber 701 are reflected as shown at 707 and are thereafter focused at reflect output fiber array 1 ' shown in the image plane at 708. Turning to Figure 8, a system having four
input fibers 801, 802, 803 and 804 is shown. The reflect output fiber array is shown with optical fibers 805, 806, 807 and 808. Such a system would have, for example, four wavelength channels in input fiber 1, with the first channel being incident upon the filter element 809 so that it is transmitted to output array shown in 801. The orthoscopic eyepiece in the exemplary embodiment as discussed above is shown in 811 for the transmit/reflect channels and at 812 for the receive channel. The demultiplexing process as discussed above with respect to the embodiments shown in Figure 1-5 as well as Figure 7 are the same for the system shown in Figure 8.
Figure 9a shows the coordinates of the input and reflect fiber arrays. To this end, the array structure for the input array and reflect output array are as shown. The angle at which the light is impingent upon the filter is given by: sinθ = (x2 + v2) * (2) f
where x and y are shown in Figure 9a and f is the focal length of the lens elements shown at 811. To this end, the x coordinate is along the horizontal axis (x- axis) while the y coordinate is along the y axis. The coordinate axes are with respect to the center of the input array, as shown. Given the relation as set forth in the equation above, the transmitted wavelength for each of the input fibers shown in Figure 9b can be
readily calculated, and thereby the placement of the input array can be readily determined. To this end, substituting equation (2) into equation (1), an expression is derived relating the x, y coordinates:
which, when solving for (x) , yields
Ixl = (n2f2 (l-(λ/λ0)2 -y2)1/2 (4)
Accordingly, input fiber array 1 has wavelength channels 1, 2,...., 8 having center wavelengths λi, λ2,_.f λs respectively. The wavelength channel having center wavelength λi is transmitted to the output array shown in Figure 8 at 810, while the remaining wavelength channels have center wavelengths λ2,_, λθ are reflected and focused by the orthoscopic optics 811 to the reflect output array, specifically reflect output array fiber 1'. The output array fiber 1' is in optical communication with input fiber 2, shown in Figure 9b. The light traversing input fiber array 2 having wavelength channels 2, 3,...., 8 having center wavelengths λ2, λ2,_.., λ8 is impingent upon the filter 809 in Figure 8, and the channel 2 having center wavelengths λ2 is transmitted to the filter and focused by the orthoscopic optics 812 to the output array 810. The remaining wavelength channels 3,...., 8 having center wavelengths λ3...., λ8 are reflected by the
filter, are focused by the orthroscopic optics 811 to the reflect output array. Specifically, the reflected channels 3,..., 8 having centerwavelength λ3,_..λ8, respectively, are focused at reflect output array fiber 2'. Reflect output array fiber 2' is connected to input fiber 3, and the sequence continues therefrom. Clearly, as is discussed above, various modifications can be made to the structure, and in particular to the number of wavelength channels to be demultiplexed. The limitations of the optical system are clearly the only limitations to the demultiplexing process. To this end, the limitations of the optical elements, in particular 811 and 812, as well as the filter element 809 in the particular example described currently are the limiting factors to the number of wavelength channels that can be demultiplexed. Furthermore, as discussed previously in connection with Example I, a multiplexing scheme could readily be configured with the system above in as shown in Figures 7, 8 and 9.
The invention having been described in detail, it is clear that modifications and variations of the invention of the present disclosure are within the perview of one of ordinary skill in the art. To this end, the multiplexer, demultiplexer having the single dielectric stack filter element with the orientation of
the angle of incidence from a particular fiber being chosen to transmit a particular center wavelength of a particular wavelength channel enable a low cost multiplexer/demultiplexer scheme to be realized. Therefore, this invention should not be limited to the disclosed embodiments, but should be limited only by the spirit and scope of the appended claims.