1. Field of the Invention
This invention relates generally to technologies for switching and routing optical wavelengths. More particularly, this invention relates to innovative method, structures and processes to manufacture and design improved waveguide grating-based wavelength selective switches.
2. Description of the Related Art
Current state of the art in wavelength-selective optical switching based signal transmission systems are still limited by several performance deficiencies caused by crosstalk, low coupling efficiency, and large size and poor form factors.
Because of the extremely wide transmission bandwidth allowed by optical fibers, all-optical fiber networks are increasingly being used as backbones for global communication systems. To fully exploit the fiber bandwidth in such networks, wavelength-division multiplexing (WDM) and wavelength-division demultiplexing (WDD) technologies are employed so that several independent optical signal streams may share the same fiber simultaneously, wherein the streams are distinguished by their center wavelengths. In the past, the adding, dropping, and cross connecting of individual optical signal in communication systems are done by first converting the optical signal into electrical signals. The electrical signals are manipulated electronically, which are then converted back into optical signals. However, the development of all-optical WDM communication systems has necessitated the need for all-optical wavelength selective devices. It is desirable for such devices to exhibit the properties of low insertion loss, insensitivity to polarization, good spectral selectivity, and ease of manufacturing.
In general, there are three prevailing types of technology commonly implemented in the all-optical Dense WDM (DWDM) networks: (1) Thin Film Filter (TFF), (2) Arrayed Waveguide Gratings (AWG), and (3) Fiber Bragg Grating (FBG). Among these three types of implementations, TFF technology is the predominant choice when the channel spacing is greater than 100 GHz. The advantages of TFF-based devices are that they are relatively insensitive to temperature, have minimal cross talk, and provide good isolation between different wavelengths. However, devices built using current TFF technology have the following disadvantages: they are difficult to manufacture when the channel spacing is below 200 GHz; the packaging cost is very high; and the yield is low. Due to these disadvantages, when the channel spacing is 100 GHz or less, AWG and FBG wavelength selecting devices dominate the market. The advantages of AWG devices are they can support high channel counts, are easy to manufacture, and have a small footprint. Meanwhile, the disadvantages are that AWG devices are prone to cross talk and their packaging is complex. The second dominant technology, i.e., the FBG technology, has the advantages of short development time, low capital investment, and low packaging cost as channel spacing is reduced to 100 GHz or less. However, the FBG products available in the current market have relatively high loss. Moreover, each channel requires a circulator, which increases component costs and possibly increases packaging costs.
Furthermore, there are several optical switching technologies under development today. They are as follows: Micro Electro-Mechanical Systems (MEMS), Liquid Crystals, Thermal Optics, Holograms, Acousto-Optic, etc. Among all these optical switching technologies, MEMS is emerging to be the most promising technology, as benefited from its potential of batch processing and cheap replication, as well as its sound record on reliability in a wide range of applications. All the other technologies are still in the experimental stage and need years to become reliable enough for commercial applications. Different embodiments of MEMS optical switches are made available in the marketplace that are implemented with a de-multiplexing device to first separate the input signals into multiple channels (each having a specific central wavelength) transmitted over a specific waveguide. Optical switching operations are performed for each of these de-multiplexed signals. Then a device is employed to multiplex these switched signals for transmission over optical fibers. Alternately, the wavelength selective optical switches are implemented with a de-multiplexing device to first separate the optical signal into channels of different wavelengths. The optical switching operations are carried out for each channel and these channels are connected to optical output ports. Again, a de-multiplexing operation must be performed first before wavelength selective switching can be carried out.
There are two types of optical MEMS switch architectures under development, or commercially available: mechanical and micro-fluidic. Mechanical-type MEMS-based switches use arrays of miniaturized mirrors fabricated on a single chip. The optical signal is reflected off this tiny mirror in order to change the transmission channel. Micro-fluidic-type MEMS-based switches, on the other hand, have no moving mirrors. Rather, they rely on the movement of bubbles in micro-machined channels.
Mechanical-type MEMS-based switches can be further classified into two categories according to mirror movement: two-dimensional (2-D) switches and three-dimensional (3-D) switches. In 2-D switches, the mirrors are only able to execute a two-position operation—that is, the mirrors can move either up and down or side by side. In 3-D switches, the mirrors can assume a variety of positions by swiveling in multiple angles and directions. These products (2-D switches or 3-D switches) are able to offer such benefits as excellent optical performance, minimal cross-talk, and the promise of improved integration, scalability, and reliability. On the other hand, these products and their methods of use are disadvantageous in the following aspects, First, in these switches, light travels through free space, which causes unbalanced power loss. Secondly, in order to steer each mirror, three to four electrodes need to be connected to each mirror, which is a major challenge to produce large-scale mechanical-type MEMS-based switches. Thirdly, alignment and packaging are difficult tasks particularly for large-scale switches.
While above-mentioned micro-mirror-based approach is widely taken by most major companies to build up their MEMS-based optical switches, Agilent Technology, Inc. has developed micro-fluidic-type MEMS-based switches by combining its micro-fluidics and ink-jet printing technology. In these switches, an index-matching fluid is used to switch wavelengths. This fluid enables transmission in a normal condition. To direct light from an input to another output, a thermal ink-jet element creates a bubble in the fluid in a trench located at the intersection between the input waveguide and the desired output waveguide, reflecting the light by total internal reflection. The advantages of these switches are that they have no moving mechanical parts and are polarization independent. The disadvantages are their questionable reliability and the excessive insertion loss for large-scale switches.
A common drawback of both of these MEMS-based switches is the requirement to work with external de-multiplexing and re-multiplexing devices in order to function properly in an optical networking system. The requirements of implementing de-multiplexing and re-multiplexing functions add tremendous complexities to the system configuration and significantly increase the cost of manufacture, system installation, and maintenance of the optical network systems. Another drawback is that these prior-art switching systems are not wavelength selective switches. In another words, switching systems based on MEMS cannot selectively switch a particular wavelength from an input waveguide to a desired output waveguide. In short, they are not wavelength intelligent devices.
To add wavelength intelligence to optical switches, Bragg grating has been shown to have excellent wavelength selection characteristics. A Bragg grating behaves as a wavelength-selective filter, reflecting a narrow band of wavelengths while transmitting all other wavelengths. The Massachusetts Institute of Technology (MIT) has developed a technology for building Bragg grating devices in planar optical waveguides. These so-called integrated Bragg gratings offer many advantages over the fiber Bragg grating, according to MIT. However, since switching optical signals requires routing of the optical transmission, a wavelength-selective filter alone is not sufficient to build an optical switch. In most waveguide grating device designs, since input signal and its reflective response occur within the same waveguide (see
Specifically, Horita et al. disclosed in U.S. Pat. No. 5,859,941, entitled “Optical Add/Drop Multiplexer Device”, an add/drop multiplexer device provided to extract or inserting optical signals or arbitrary wavelength and having a wavelength selection characteristic with a narrow bandwidth. The multiplexer device is formed on a substrate implemented with a layered structure. In another U.S. Pat. No. 6,567,573, entitled “Switchable Optical Components” issued on May 20, 2003, Domash et al. disclosed electrically switchable Bragg grating devices and device geometries realized by using holographic polymer or dispersed liquid crystal materials.
However, these patented technologies have technical limitations due to the interference of signal transmissions may occur when the wavelength of the reflected waves falls within the optical communication band as will be further explained below in
Therefore, a need still exists in the art to provide an innovative configurations and method of manufacturing and designing the wavelength selective optical devices to overcome such limitations.
It is therefore an object of the present invention to provide new and improved methods and configurations for wavelength selective optical transmission system with coupling waveguides having wavelength selective Bragg gratings wherein the coupling waveguides are coupled based selection rules to reduce cross-talk between channels and to increase signal coupling of the optical transmissions between two waveguides. Specifically, the coupling waveguides are selected to have different optical propagation constant. In specific embodiments the coupling waveguides have different sizes and shape of cross sections. In another different embodiment, the coupling waveguides have different thickness to width ratios, i.e., different aspect ratios, to enhance the coupling of signals between the two adjacent waveguides.
Briefly, the present invention discloses a method for configuring a wavelength selective optical signal transmission system. The method includes a step of providing a set of Bragg gratings on at least one of two coupled waveguides near an interfacing section between the coupled waveguides. The method further includes a step of selecting a second waveguide such that the set of wavelength-selective Bragg gratings are to reflect a reflecting optical signal back to the first waveguide and for transmitting a contra-directional optical signal and a co-directional optical signal having respectively a contra-directional selected wavelength and a co-directional selected wavelength corresponding to the Bragg gratings wherein one of the contra-directional and co-directional wavelengths is chosen as a designated wavelength, and the reflecting optical signal and one of the contra-directional or co-directional optical signals are outside of a predefined range surrounding the designated wavelength.
In accordance with the invention, the optical device includes wavelength-selective optical transmission system. The optical signal transmission system includes a first and a second waveguides. The second waveguide disposed on a vertically stacked position on the first waveguide and at least one of the first and second waveguides have a set of wavelength selective Bragg gratings disposed near a coupling section between the first and second waveguides wherein the first and second waveguides have different optical propagation constants. In a specific embodiment, the first and second waveguides are composed of different optical transmission materials. In another preferred embodiment, the first and second waveguides have different sizes and shape of cross sections.
The present invention can be better understood with reference to the following drawings. The components within the drawings are not necessarily to scale relative to each other, emphasis instead being placed upon clearly illustrating the principles of the present invention.
In the following description, numerous specific details are provided, such as the identification of various system components, to provide a thorough understanding of embodiments of the invention. One skilled in the art will recognize, however, that the invention can be practiced without one or more of the specific details, or with other methods, components, materials, etc. In still other instances, well-known structures, materials, or operations are not shown or described in detail to avoid obscuring aspects of various embodiments of the invention. Reference throughout this specification to “one embodiment” or “an embodiment” means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the present invention. Thus, the appearance of the phrases “in one embodiment” or “in an embodiment” in various places throughout this specification are not necessarily all referring to the same embodiment. Furthermore, the particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.
A typical waveguide grating device is shown in
Referring to
As is well understood in literature (see, for example, Ramaswami, R., Optical Networks: A Practical Perspective, Section 3.3.2, Academic Press, 1998), a Bragg grating refers to any periodic perturbation or variation in the propagating medium. A Bragg grating, therefore, may come about as a periodic variation of either the refractive index of the propagating medium, the structure of the propagating medium, or both.
The three main classes of grating assisted directional coupler device design are illustrated in
Other designs where the Bragg gratings are formed in the cladding, although not explicitly depicted here, are feasible and therefore implied.
Referring to
If the grating assisted directional coupler device is designed such that only contra-directional coupling is to be utilized, wavelength λi would be one of the aforementioned N wavelengths but wavelengths λj and λk would be outside of the set of aforementioned N wavelengths. In other words, signals 304 and 306 would carry no optical power. The remainder optical signal of the wavelengths λ1, λ2, . . . , λi−1, λi+1, . . . , λN is not affected and continues to propagate along waveguide 301.
For the purpose of signal transmission in the telecommunication network, optical signals of certain range of wavelengths are employed, e.g., C-band optical signals are limited within a range of 1529.75˜1569.59 nm. It is a common practice to assume some margin to allow for random fluctuations in process and/or material non-uniformity. In the case when this margin is set at 5 nm, for example, one may consider setting as set forth below:
λmin=1529.75−5=1524.75 nm; and
λmax=1569.59+5=1574.59 nm
The gratings structure of our switches is designed such that contra-directional coupling, co-directional coupling and direct Bragg reflection occur at specific wavelengths. Depending on which coupling mechanism is to be utilized, the specific wavelength corresponding to the particular coupling mechanism falls within the bandwidth of interest, (e.g. the C-band). For example, if only contra-directional coupling is to be utilized, it is desirable that the wavelengths specific to co-directional coupling and direct Bragg reflection are outside of the bandwidth of interest.
Similar considerations are applicable for designs where the Bragg gratings 308 are formed only on waveguide 302 as in
Refer to
Referring to
It should be specified that in (1),
More generally, concerning the two waveguides in a grating assisted directional coupler such as those illustrated in
If the wave number k is given by
Then (1) can be rewritten as
kλR=(β1+β2)Λ (3)
It is also necessary to avoid crosstalk due to (a) response 306 (co-directional coupling); and (b) response 304 (direct Bragg reflection). To be more precise, the wavelength λBC at which co-directional coupling (response 306) occurs is given by
kλBC=(β2−β1)Λ (4)
Similarly, the wavelength λB1 pertains to direct Bragg reflection (response 304) is given by
kλB1=2β1Λ (5)
As mentioned earlier, it is necessary to make sure that these wavelengths are outside of our bandwidth of interest. This is illustrated in
Section Rules
Application of the Contra-Directional Coupling Signal 305
The two special cases that need to be considered are (1) when η1<β2; and (2) when β1<β2
Case 1: Application of the Contra-directional Coupling Signal 305 with β1<β2
It should be noticed that in this case, from (3), (4) and (5) we have
As a result this case can be further subdivided into two sub-cases, as follows.
If the window of wavelength is limited (within the desired range that are interested in) to between λmin and λmax, the grating period should be designed such that
λB1(max)<λmin (6)
In view of (5), this becomes
2β1Λmax<kλmin (7)
On the other hand, according to (3)
kλmax=(β1+β2)Λmax (8)
Combining (7) and (8), one obtains the selection rule for this case
Similar to the previous situation, in this case, it is required that
λBC(max)<λmin (10)
Referring to (4), this means
(β2−β1)Λmax<λmin (11)
Combining (7) and (10), the selection rule for this case is given by
Thus, from (9) and (12) we obtained a general selection rule for Case 1 when β1<β2
Case 2: Application of the Contra-directional Coupling Signal 305 with β1>β2
Given the condition that β1>β2, from (3), (4) and (5), we have in this case λB1>λR>λBC.
Under this situation it is required that
λB1(min)>λmax (14)
From (3) and (5), we have
kλmin=(β1+β2)Λmin (15)
kλB1(min)=2β1Λmin (16)
Putting these together, the selection rule in this case is found to be
This means that even in the extreme case, it is necessary that
λBC(max)<λmin (18)
Recall (3) and (4) such that
kλmax=(β1+β2)Λmax (19)
kλBC(max)=(β1−β2)Λmax (20)
Combining (18), (19) and (20), one obtains the selection rule for this case
To summarize, from (17) and (21), the general selection rule for Case 2 when β1>β2 is found to be
Application of the Co-Directional Coupling Signal 306
Referring to
kλT=|β1−β2|Λ (23)
The wavelengths pertain to (a) response 305 (contra-directional coupling); and (b) response 304 (direct Bragg reflection) are respectively given by
kλBC=(β1+β2)Λ (24)
and
kλB1=2β1Λ (25)
As in the previous case, to ascertain these wavelengths to be outside of our bandwidth of interest several special cases need to be considered.
Case 3: Application of the Co-directional Coupling Signal 306 with β1<β2
From (23), (24) and (25) and given the condition that β1<β2 there are 2 possible cases to consider:
In this case it is demanded that
λB1(min)>λmax (26)
From (23) and (24) it follows with these relationships
kλmin=(β2−β1)Λmin (27)
kλB1(min)2β1Λmin (28)
Combining (26), (27) and (28), the selection rule for this case is found to be
In this case it is demanded that
λB1(max)<λmin (30)
λmax<λBC(min) (31)
By considering (30) first. From (23) and (25), it follows that
kλmax=(β2−β1)Λmax (32)
kλB1(max)=2β1Λmax (33)
By combining (30), (32) and (33), part one of the selection rule is thus obtained
Next, by considering (31) and from (24) and (27), it follows that:
kλBC(min)=(β1+β2)Λmin (35)
kλmin=(β2−β1)Λmin (27)
Combining (31), (27) and (35), part two of the selection rule is found to be
The selection rule for Embodiment 3b when β1<β2 is therefore given by
In summary, the general selection rule for Case 3 is
Case 4: Application of the Co-Directional Coupling Signal 306 with β1>β2
Again, from (23), (24) and (25) and given the condition that β1>β2 there are two possible cases to consider:
It is required that
λBC(min)>λmax (38)
From (27) and (35) we have these relationships
kλmin=(β1−β2)Λmin (27)
kλBC(min)=(β1+β2)Λmin (35)
Combining (38), (27) and (35), the selection rule for this case is found to be
This is similar to Embodiment 3a. The conclusion is therefore similar. The selection rule for this case is found to be
To summarize, the general selection rule for Case 4 is given by
Below is a table to sum up the selection rules derived so far:
Consider the first row of Table 1. It is simple to conclude that if β1<β2<3β1 then
The converse is true if β2>3β1.
Based on the second row of Table 1, it is not difficult to conclude that if (√{square root over (5)}−2)β1<β2<β1 then
For the converse to be true, it is necessary that β2<(√{square root over (5)}−2)β1.
Simplification of the third row of Table 1 is a bit more complicated. We observe that if β1>(√{square root over (5)}−2)β2, then
So the initial conclusions are such that if (√5−2)β2<β1<β2, then
Otherwise, that is if β1<(√5−2)β2, then
Considering (45), it is easy to show that if
As a result, Table 1 is simplified. The simplified selection rules are shown in Table 2.
In particular, selection rules for the case of contra-directional coupling are presented as a graph shown in
and the vertical axis denotes
Notice that the straight line 501 in
for C-band, which corresponds to roughly 0.975.
The selection rules for contra-directional coupling are manifested into such that only the regions (denoted by 504) above the curve 502 and below the straight line 501 are allowed. Note that according to
Similarly, selection rules for the case of co-directional coupling are presented as a graph shown in
and the vertical axis denotes
Again, the straight lines 601 and 611 in, respectively,
for C-band which, as mentioned earlier, corresponds to roughly 0.975.
As in the case of contra-directional coupling, the selection rules for co-directional coupling are translated, in terms of
From the selection rules, it is observed that the propagation constants of waveguide 301 and waveguide 302 must be significantly different, i.e. β1≠β2, in a contra-directional grating assisted waveguide coupler.
In recent years, process technology employed in fabrication of integrated optics circuitry is largely borrowed from silicon IC fabrication technology. If the waveguides shown in
Method (a) is expensive and difficult to realize since extra masks and process steps are required. Method (b) is very limited due to two reasons. First, the thickness of both waveguides is necessarily similar as the process is complicated otherwise. By adjusting waveguide widths alone the difference in propagation constants is not expected to be large enough to satisfy any of the selection rules.
Referring to
Herein a novel vertically stacked waveguide structure is disclosed. It should be mentioned that vertically stacked waveguide structures are common in semiconductor (such as gallium arsenide or indium phosphide based) optical component design. There are several significant differences which distinguish the vertically stacked waveguide structure proposed here from the ones used in semiconductor optical components. First, it is proposed here that the two waveguides be made of different material. This is contrary to the semiconductor optical components in which the vertically stacked waveguides are made of the same material. Second, it is proposed here that the bus waveguides are designed (by using appropriate material and geometry) such that they can be coupled to fibers efficiently. In the case of semiconductor optical components, due to the high refractive index, coupling to fibers is typically not very effective.
By fabricating the input/output waveguides and the coupling waveguide on two different levels, design of each type of waveguides can be independent of each other. For example, the dimensions of the coupling waveguide can be very different from the input/output waveguides, and each type of waveguides can be fabricated using different material. In other words, both β1 and β2 can be chosen freely to satisfy the selection rules in Table 2. Another added advantage is that the gap between the input/output waveguide and the coupling waveguide, which significantly controls the grating assisted directional coupler device characteristics, can be controlled precisely.
In
Referring to
It should be mentioned that in the case when the Bragg grating is in the form of periodic geometrical variation of the propagating medium, it is preferred to have the grating structure be formed on top of the bridge waveguide. This way, undesirable reflection may be avoided.
The grating assisted directional coupler devices formed with the vertically stacked configurations as shown in
As mentioned earlier, both the bus and the bridge waveguides are to be fabricated using different material. In the above embodiments, choices of material suitable for each component are as follows.
As a result there are at least 96 different combinations. They are tabulated in Table 3.
In the case when both the cladding and the bus waveguide are made out of doped SiO2 or polymer, it should be understood that the doping levels or the constituents are different so that the refractive indices of the cladding and the bus waveguide are different.
Finally, choices of material suitable for the substrate include Si and SiO2.
It should be emphasized that even that the vertically stacked waveguide structure has been suggested elsewhere by other patented disclosures, the disclosures of the present invention is novel when these waveguides are made of different kinds of materials and that the gratings are formed on either or both of the waveguides, or in the cladding.
Although the present invention has been described in terms of the presently preferred embodiment, it is to be understood that such disclosure is not to be interpreted as limiting. Various alternations and modifications will no doubt become apparent to those skilled in the art after reading the above disclosure. Accordingly, it is intended that the appended claims be interpreted as covering all alternations and modifications as fall within the true spirit and scope of the invention.
This application claims priority to pending U.S. patent application entitled “IMPROVED GRATING-BASED WAVELENGTH SELECTIVE SWITCH” filed Dec. 9, 2003 by Ling et al. and accorded Ser. No. 10/731,297 the benefit of its filing date being hereby claimed under Title 35 of the United States Code.