TW548232B - MEMS-based wavelength equalizer - Google Patents

Abstract

A wavelength specific optical equalizer for selectively attenuating discrete wavelength signals contained within a wavelength division multiplexed signal without affecting the adjacent signals. The wavelength equalizer includes a demultiplexer adapted to separate a wavelength division multiplexed signal into a plurality of discrete wavelength signals and to direct each of the discrete wavelength signals along a plurality of first optical paths. A micro-mechanical device comprising at least one micro-mirror is optically coupled with each of the first optical paths. A plurality of second optical paths is positioned to receive the discrete wavelength signals reflected from the respective micro-mirrors. At least one actuator is mechanically coupled with each of the micro-mirrors. The actuators are adapted to selectively displace one or more to divert at least a portion of the discrete wavelength away from the corresponding second optical paths. The orientation of the micro-mirror determines a signal strength of the discrete wavelength signal reflected to the corresponding second optical path. A multiplexer is provided to combine the discrete wavelength signals in the plurality of second optical paths into a reconstituted wavelength division multiplexed signal.

Description

548232 A7 B7 V. Description of the invention (1) Generally speaking, the present invention relates to a wavelength-specific equalizer based on MEMS, and particularly to a method and device for using a MEMS-based device to steer and control a beam of light that is equalized by the wavelength equalizer. . In a wavelength division multiplexed (Γ WDM) network, a single fiber often carries multiple independent and unrelated data channels, where each data channel is assigned to a discrete optical wavelength. In fiber optic telecommunication networks, different wavelengths may not have the same amplitude. For example, as the signal propagates through the fiber optic network, the signal encounters point-to-point transmission loss and coupling loss within the network. In order to compensate for transmission loss and coupling loss, the multiplexed signal where the fiber is located is often amplified at various points on the network using a doped fiber amplifier. The hybrid fiber amplifier amplifies each optical wavelength contained in the multiplexed signal. However, the gain provided by the bait-doped fiber amplifier is not uniform with frequency. Instead, some wavelengths are amplified much more than others. When a multiplexed signal experiences multiple periods of transmission loss, coupling loss, and then amplifies, the intensity change between each wavelength increases. Without correction, non-uniform signal strength may cause inter-channel tones and leakage of transmission data. In order to reduce the non-linear effects of these amplifiers and allow the use of common communication loops for all channels, similar power levels are required for each wavelength channel. But when each wavelength is added to or deleted from each node of the network, the power level of each channel changes, which in turn changes the gain spectrum of the amplifier. This kind of network changes slowly when the customer deletes it, or when the data traffic of the network is dynamically rerouted to improve efficiency, the network changes quickly. Frequent power changes are roughly monotonic in terms of wavelength, but can be changed by changing with wavelength -4-

548232 A7 B7 V. Description of the invention (2) Monotonic change in attenuation, in other words, full compensation is obtained by the attenuation slope of the spectrum. Systems and methods have been developed to compensate for the non-uniform wavelength-dependent gains of bait and hybrid fiber amplifiers in fiber optic communications. One of such previous techniques is to pre-compensate the non-uniform gain of the hybrid fiber amplifier by pre-emphasizing the input level of the fiber carrier signal. This prior art technique is exemplified in "End-to-End Equalization Experiments in Amplified Lightwave Systems" by author A · R · Chraplyvy et al. IEEE Photonic Technology Letters, 4 (4), pages 428-429 (April 1993) ). The second prior art technique to compensate the non-uniform gain of a doped fiber amplifier is to use an in-line optical filter. The fiber can screen wavelengths with the greatest amplification, thereby forming a more uniform signal strength across all signal wavelengths. This prior art technique is exemplified in "Tuneable Gain Equalization Using MACH-ZENHDER Optical Filters in Multi-stage Fiber Amplifiers", author R. Inoue et al., IEEE Photonics Technology Letters, 3 (8), 718-720 P. (April 1991). See also US Patent No. 5,430,817 (Vengsarker). U.S. Patent No. 5,500,761 (Goossen et al.) Discloses a mechanical anti-reflection switch modulator (Γ MARS modulator) that can be used for optical power equalization. The MARS modulator is basically a Fabry-Perot cavity containing a gap between the optical film and the substrate. Modulation is based on the vertical voltage control of the film relative to the substrate. MARS modulators provide broadband, high-contrast reflection modulation at rates in excess of millions of bits per second. MARS modulators with low insertion loss and wide operating bandwidth, MARS modulators with low insertion loss and wide operating bandwidth that are particularly advantageous for use in optical communication applications are described in US Patent No. 5,870,221. -5- This paper size applies to China National Standard (CNS) A4 (210X 297 mm) 548232

Referring to the drawings, FIG. 1 is a schematic cross-sectional view of a single-unit MARS modulator of the type described in the Goossen patent. The device 9 includes a substrate 10 and a film 15 which are spaced apart from each other and define a gap 20 therebetween. The substrate 10 is a conductive material such as doped silicon, and the film 15 includes one or more conductive materials such as a silicon nitride top layer 15a and a polycrystalline silicon bottom layer 5b. The top layer has a square root having a refractive index approximately equal to the refractive index of the substrate, and the bottom layer has a refractive index approximately equal to the refractive index of the substrate. i5a and 15b are each less than a quarter of the operating wavelength a. The film μ and the substrate 10 are separated from each other by a peripheral support layer i 2 of an insulating material. The electrodes 3 0 and 3 i allow the film 15 and the substrate 10 to be connected to the bias source 29 terminals, respectively. The air gap 20 can be controlled by the bias voltage between the substrate and the film. The highest peak of relative reflection occurs when the gap 20 is the operating wavelength; an odd multiple of a quarter of I. The lowest chirp occurs when the gap 20 is zero or an even multiple of A / 4. The present invention relates to a wavelength equalizer, which is used for selectively attenuating wavelengths and dividing a plurality of discrete wavelength signals contained in a multiplexed signal without affecting adjacent signals. In a specific embodiment, the wavelength equalizer includes a demultiplexer, which is adapted to separate the wavelength division multiplexed signal into a plurality of discrete wavelength signals, and guide each discrete wavelength signal along a plurality of first optical paths. A microcomputer device including at least one micromirror optically closes each first optical path. Each of the second optical paths is located at a position capable of receiving a discrete wavelength signal reflected by a respective micromirror. At least one actuator is mechanically coupled to each micromirror. The actuator is adapted to selectively replace one or more micromirrors, turning at least partly discrete wavelengths away from the corresponding second optical path. The direction of the micromirror determines the signal intensity of the discrete wavelength signal reflected to the second optical path. Set the multiplexer to be used for multiple second lights -6- This paper size applies to China National Standard (CNS) A4 (210 X 297 mm)

Binding

548232

Light discrete wavelength signals are combined into a restructured wavelength division multiplexed signal. In a specific embodiment, the first optical path includes a second optical path. In other words, the scattered wavelength is reflected off the back side of the micromirror along the radial direction of the first light. Set multiple intensity detectors to detect the optical signal intensity of discrete wavelength signals. In a specific embodiment, the beam splitter is located along a plurality of first optical paths. The signal strength detector is optically coupled to the beam splitter and detects the signal strength of discrete wavelength signals. The controller is coupled to the actuator and signal strength detector as needed. The controller may selectively attenuate one or more discrete wavelength signals in response to the monitored signal strength. The micromirror and the actuator are preferably constructed on the surface of the substrate. The micromirror is attached to the surface of the substrate by a hinge, a universal joint or a torsion spring. The micromirror is adapted to move through at least one degree of freedom. The actuator is preferably a thermal actuator. The thermal actuator includes at least a hot arm having a first end anchored to the surface and a free end located above the surface, and a cold arm having a first end anchored to the surface and a free end. The cold arm is positioned above the hot arm relative to the surface. A component mechanically and electrically couples the free ends of the hot and cold arms. The member is in a position that can engage the mirror when current is applied to at least the hot arm. In a specific embodiment, the thermal actuator comprises at least one ' degree which is adapted to provide controlled bending. In another embodiment, the thermal actuator includes at least one reinforcing member. When the actuator is in an unactivated configuration, the micromirror reflects substantially all of the discrete wavelengths, from one of the optical paths to the corresponding second optical path. When the actuator is in the active configuration, the micromirror only reflects part of the discrete wavelength signals from the first light. The paper size applies the Chinese National Standard (CNS) A4 specification (210X 297 mm) 548232 A7 B7 V. Description of the invention

One of the paths to the corresponding second optical path. In a specific embodiment, when the actuator is in an activated configuration, the micromirror does not substantially reflect any discrete wavelength signal from the first optical path · one to the corresponding second optical path. In another specific embodiment, the wavelength equalizer includes a demultiplexer, which is suitable for separating the wavelength division multiplexed signal into a plurality of discrete wavelength signals, and guiding each discrete wavelength signal along a corresponding plurality of first optical paths. . The plurality of second optical paths are located at positions where discrete wavelength signals are received. The microcomputer device includes a thermal actuator adapted to selectively steer at least a portion of the discrete wavelengths away from the corresponding second optical path by one or more first optical paths. A multiplexer is set up to combine the discrete wavelength signals of multiple first optical paths into a restructured wavelength division multiplexed signal. In a specific embodiment, the microcomputer device includes a plurality of micromirrors which are mechanically coupled to a plurality of thermal actuators. In another specific embodiment, each of the plurality of first lights I extends along the surface of the corresponding thermal actuator. When the thermal actuator is not excited, the second optical path is optically aligned with the corresponding first optical path. When the thermal actuator is in an activated condition, the first optical paths are respectively turned from the corresponding first optical paths. ~ The present invention is also directed to an optical communication system using the wavelength equalizer of the present invention: System: One embodiment of an optical communication system is a switched wavelength division duplex optical communication system, which includes at least one optical fiber carrying a wavelength Divide the multiplexed signal and the wavelength equalizer of the present invention. Other features of the present invention will be more apparent from the following detailed description of specific embodiments and the accompanying drawings. FIG. 1 is a side view of an anti-reflection switching modulator of a prior art machine.

Binding

548232

FIG. 2 is a schematic explanatory diagram of a switched wavelength division multiplexed optical communication system according to the present invention. … Figure 3 is an end view of the optical fiber of the wavelength equalizer of FIG. 2. FIG. 4 is another end view of the optical fiber in the wavelength equalizer of FIG. 2. Fig. 5 is a schematic explanatory diagram of a second embodiment of the wavelength equalizer according to the present invention. Fig. 6 is a schematic explanatory diagram of a third embodiment of the wavelength equalizer according to the present invention. Fig. 7 is a schematic explanatory diagram of a fourth embodiment of the wavelength equalizer according to the present invention. ~ FIG. 8 is a top view of a microcomputer device according to the present invention. Fig. 9 is a top view of a universal joint according to the present invention. FIG. 10 is a perspective view of the microcomputer device of FIG. 8 with a mirror in a non-coplanar configuration. FIG. 11 is a perspective view of another microcomputer device according to the present invention. FIG. 12 is a perspective view of a non-coplanar configuration of the microcomputer device of FIG. 11. FIG. Fig. 13 is a top view of a thermal actuator for a universal joint micromirror according to the present invention. FIG. 14 is a side view of the thermal actuator of FIG. 13. FIG. 15 is a sectional view of the thermal actuator of FIG. 13. FIG. 16 is a sectional view of the thermal actuator of FIG. 13. FIG. 17 is a side view of the thermal actuator of FIG. 14 in an actuated position. 18 is a top view of another thermal actuator including a waveguide according to the present invention. FIG. 19 is a side view of the thermal actuator of FIG. 18. Figure 20 shows a fifth specific embodiment of the wavelength equalizer according to the present invention. -9- This paper size applies the Chinese National Standard (CNS) A4 specification (210X297 public love)

Binding

548232 V. Description of invention A7 B7

Intention illustration. The present invention relates to a wavelength equalizer for an optical communication system. Although the method and device can be used for any application where it is desired to attenuate to a specific wavelength of a multiplexed signal, the present invention is particularly suitable for selectively attenuating the signal number 俾 to compensate for the optical path dependence loss of a multiplexed optical fiber communication system. For example, the method and device will explain the application of the switched wavelength division multiplexed optical communication system. The system contains multiple optical paths and each optical path contains a unique degree of signal loss. As used herein, "wavelength equalizer" means a device that can selectively reduce the signal amplitude of any wavelength by any amount without affecting adjacent wavelengths. Although "equalizer" is an industry term that implies signals of equal strength, equalizers do not need to equalize signal strengths at adjacent wavelengths. For example, for some applications, equalizers can be used to maintain differential signal strength between adjacent wavelengths. Fig. 2 is a schematic illustration of a switched wavelength division multiplexed optical communication system 50 coupled to a wavelength equalizer 52 according to the present invention. The communication system 50 includes an input fiber 54 capable of carrying a multiplexed # number. The optical fiber 54 has a specified transmission loss, coupling loss, and frequency division loss, and may or may not pass any number of bait-doped fiber amplifiers 56. The input fiber 5 4 is optically coupled to the wavelength division demultiplexer 5 8. The demultiplexer 58 converts the multiplexed optical signal 60 carried by the input fiber 54 into a plurality of discrete wavelengths or channels 70. "Discrete wavelength" or "channel" is used here to indicate a segment or slice of electromagnetic frequency. Segments or fragments used for telecommunications purposes at the current state of the art typically have a size of about 0.05 nm or more. The demultiplexer means a beam that receives multiple wavelengths and splits the beam into -10-

Binding

Line This paper size is suitable for Towel ® ® * Standard (CNS) A4 specification (21GX297 public love) 548232 A7 B7 V. Description of the invention () Device with discrete wavelength or channel. A multiplexer means a device whose function is exactly the opposite. The multiplexer accepts multiple optical wavelengths and converges them into a single beam. Passive demultiplexers are based on prisms, diffraction gratings, arrayed waveguide gratings, and spectral chirpers. The active demultiplexer is based on a combination of passive components and a tunable detector. Each detector is tuned to a specific frequency. The discrete wavelengths 70 are transmitted through the discrete optical fibers 62 respectively, and the optical fibers are located at optical coupling positions with a corresponding number of microcomputer mirrors 66 of the microcomputer device 68. In the configuration illustrated in FIG. 2, substantially the entire optical signal 70 transmitted by the optical fiber 62 is reflected away from the microcomputer mirror 66 to the corresponding optical fiber 64. The micromirror 6 6 is preferably in a coplanar configuration relative to the microcomputer device 6 8. The term “coplanar configuration” used here indicates the configuration of roughly parallel substrate surfaces. As illustrated in FIG. 3, the optical signal 70 is substantially reflected to the optical fiber 64. The reflected optical signal 70 passes through the multiplexer 72, where the scattered wavelength 70 is recombined into a restructured multiplexed signal 7 §. One or more of the mirrors 66 can be rotated non-coplanar with respect to the microcomputer device 68. As used herein, the term 'Γ non-coplanar' means an angle greater than 0 to 90 degrees relative to the surface of the substrate. Say 66 is typically moving through one or more degrees of freedom. By changing the angle of the one or more micro-mirrors 66 relative to the microcomputer device 68 and thus the angles corresponding to the corresponding optical fibers 62, 64, part of the optical signal 70 is turned by the optical fiber 64. The term "steering" is used here to indicate that a portion of or all of the optical signal is attenuated as a result of redirecting or misdirecting a portion of the optical signal. As best shown in Fig. 4, only a portion of the optical signal 70 reflected by the non-coplanar mirror 66 is optically coupled to the optical fiber 64. The remaining optical signals 70 are diverted to the area 74 on the shield 64 of the optical fiber 64, and thus the optical signals 70 are attenuated upstream of the multiplexer 72. Typically, only certain wavelengths require attenuation. As a result, part of the optical signal 7 is attenuated -11- This paper size is in accordance with the Chinese National Standard (CNS) A4 specification (21 × 297 mm) 548232 A7 B7 V. Description of the invention (9) minus ’and other optical signals are not. For example, some micromirrors are in a coplanar configuration, while other micromirrors are in a non-coplanar configuration. The attenuated and undecreased optics ## 7 0 is mixed with the multiplexer 7 2 to form a restructured wavelength division multiplexed signal 7.8. As used herein, "restructured wavelength division multiplexed signal" means a multiplexed signal formed by a plurality of discrete wavelengths sent from a wavelength equalizer. Referring back to FIG. 2, the demultiplexer 58 and the multiplexer 72 may optionally include a series of beam splitters or coupling devices, and the beam splitters or coupling devices turn a small part of the discrete wavelength 70 to the detector, the detector Measure signal strength (or optical power) at discrete wavelengths. The measured signal strength is passed to the controller 80. The controller 80 then selectively controls the position of the mirror 66 to achieve a predetermined degree of attenuation of the optical signal 70. In another specific embodiment, the information about the wavelength intensity of the input or output optical signal may be measured upstream of the demultiplexer 58 and / or downstream of the multiplexer 72. The controller 80 adjusts the signal intensity of each wavelength in real time in response to the detected signal intensity. As a result, the present wavelength equalizer 52 is suitable for use in a switched wavelength division multiplexed optical communication system 50 to dynamically change the signal strength of discrete wavelengths. Fig. 5 is a schematic illustration of a second specific embodiment of the wavelength equalizer 100 according to the present invention. The input quasi-chirped multiplexed optical signal 102 is guided to the diffraction grating 104. The quasi-multiplexed optical signal 102 impacts the diffraction grating 104, and thus each wavelength of light is dispersed at an angle proportional to its wavelength. Each wavelength 106 is directed toward a microcomputer device uo a plurality of discrete mirrors 108. In the name configuration shown in Figure 5, virtually all wavelengths 106 are reflected off the mirror 108 and directly reflected back to the grating 104. Here, the wavelength ι〇6 is multiplexed to output -12. This paper scale applies China National Standard (CNS) A4 specification (210X 297 mm) 548232 A7 ______B7 5. Description of the invention (1Q) Optical signal 112. The wavelength 106 is essentially along the same optical path from the micromirror 108 ° 4¾ 108. The angular relationship of the 108 with respect to the grating i04 is adjusted by a small amount to bias the partial beam 106, and avoid recombining the biased portion to the grating ι04 Output optical letter 5 tiger 112. The controller (refer to FIG. 2) is typically used to adjust the position of the mirror 08. FIG. 6 is a schematic diagram of a third embodiment of the wavelength equalizer ι20 according to the present invention. The input quasi-multiplexed optical signal 122 strikes the grating 124, and the grating disperses each wavelength of light t at an angle proportional to its wavelength. Each wavelength 126 is guided to strike the mirror 128 of the microcomputer device 130. The wavelength 126 is reflected off the mirror 128 and guided to the second grating 132, which combines the wavelength ι26 into an output optical signal 134. The wavelength 126 passes through different optical paths before and after the reflection deviates from the micromirror 128. Again, a small adjustment of the angular relationship of the mirror 128 with respect to the grating 132 will bias the partial beam 126 and prevent the deflected portions from being recombined into the grating 132 into an output optical signal 134. Fig. 7 illustrates a fourth specific embodiment of another wavelength equalizer 14o according to the present invention. The prism 142 divides the input pseudo-multiplexed optical signal 144 into discrete component wavelengths 146. The beam splitter 148 is positioned downstream of the prism 142 and redirects the small sub-wavelength 146 to the detector 151 of the controller 150 as needed. As discussed earlier, the wavelength 146 hits the mirror 152, and the wavelength 146 reflects to the prism 154. The prism 154 combines the wavelength 146 into an output signal I%. The beam splitter 158 is positioned downstream of the prism 154 as necessary. The beam splitter diverts a small portion of the output optical signal 156 to the prism i60, and the prism 160 separates the optical signal into its constituent wavelength. The wavelength is then guided to the detector 501 of the controller 501, and the intensity of each wavelength of the output optical signal 156 is measured. The controller 150 can use the intensity measured by the optical signals collected from the beam splitters 148 and 158 to control the mirror-13- This paper size applies the Chinese National Standard (CNS) A4 specification (210X297 mm) 548232 A7 B7 V. Invention Description (11 152 positions. Used here '"microcomputer device" means a micron-sized machine, opto-mechanical, motor, or opto-electric device constructed on a substrate. A variety of techniques for making microcomputer devices can be utilized, such as available from Konos ( Cronos) Multi-user MEMS processes (MUMPs) by Integrated Microsystems Inc. (Scientific Research Triangle Park, North Carolina). One of the assembly procedures is described in rMUMPs Design Surgery, "Rev. 5 (200). (Years) available from Konos Microsystems. Polycrystalline silicon surface is suitable for cutting integrated circuit (IC) industry-known planar process steps for manufacturing micromotors or microcomputer devices. The standard microblock cutting process for polycrystalline silicon surfaces is deposition And lithographic patterning of low-stress polycrystalline debris (called Shao Xixi), and alternate layers of protective materials (such as dream dioxide or dream glass). Provides anchor points anchored to the substrate and mechanical and electrical interconnections between the layers of polycrystalline silicon. The functional elements of the device are deposited using a series of deposition and patterning process steps. Completed in the device structure | The protective material is ejected and moved away, and the etchant is, for example, hydrofluoric acid (HF) which does not substantially attack the polycrystalline silicon layer. The result is a construction system whose outline is provided by the first polycrystalline silicon layer which provides electrical interconnection and / Or voltage reference planes, as well as additional layers of mechanical polysilicon, which can be used to form functional elements, ranging from simple cantilever beams to complex motor systems. The micromirror system is located coplanar with the substrate. Since the entire process system is based on Standard 1 (: manufacturing technology, so a large number of fully assembled devices can be manufactured in batches on a silicon substrate without a fragment assembly. Figure 8 is a top view of a microcomputer device 22o suitable for this wavelength equalizer,- 14- This paper size applies Chinese National Standard (CNS) A4 specification (210X 297 mm) 548232 A7 B7 V. Description of the invention (It includes a The micromirror 221 of the joint and one or more thermal actuators 252A-252L (collectively referred to as "252"). The mirror 222 on the micromirror 221 with the universal joint is formed to make the surface 224 highly reflective. The mirror 222 is fixed to the base plate 226 by a plurality of torsional hinges or universal joints 228A-228D (collectively referred to as "228"). As used herein, the "universal joint" means a mechanically engaged mirror or other structure To the substrate surface, at the same time, it is allowed to move through the microcomputer device with at least two degrees of freedom (typically throwing and rolling) relative to the substrate surface. In this specific embodiment, the mirror 222 is roughly square, and the universal joint 228 is positioned along its four sides. The shape of the mirror, the number of universal joints, and the position of the universal joints can be changed according to the application of the micromirror 221 with the universal joint. For example, the universal joint 228 may be located at the corner of the micromirror 221 with the universal joint. The micromirror 221 of the present invention is preferably shaped to allow a tight packing array, such as a triangle including a triangle, a rectangle or having five or more sides, a hexagon, an octagon, and the like. The micromirror 22 1 with a universal joint may be circular. As best shown in FIG. 9, the universal joints 228 each include a pair of first arms 230, 232 cantilevered from a mirror 222 to members 234, 236. The second arms 23 8, 240 are cantilevered from the members 234, 236 to the anchor 242. Although the arms 232, 240 and 230, 238 are roughly perpendicular to the mirror 222 and roughly parallel in the specific embodiment, this configuration is not necessary. The arms 230, 232, 238, 240 may be angled relative to the mirror 222 and / or angled relative to each other. Further, the arms 230, 232, 238, 240 may be curved. In a specific embodiment, the universal joint 228 suspends the mirror 222 above the surface of the base plate 226. In another specific embodiment, the mirror 222 is docked on the surface of the substrate 226, but is movably fixed to the substrate 226 by a universal joint 228. The spacer 244 can be set under the members 234 and 236 as needed. Support arm -15- This paper size applies to China National Standard (CNS) A4 (210X297 mm) binding

Line 548232 A7 B7 V. Description of the invention (230, 232, 23 8, 240 ° The gasket 244 can also be used as the limit or end stop of the arm 230, 232, 238, 240 and / or mirror 222. Universal joint during operation The resistance and stiffness of 228 are modified by increasing or decreasing the number, length, and cross-sectional area of the arm and various other factors. For example, the anchor 242 can be moved closer to the members 234, 236. In another embodiment, the arms 232, 240.

Referring back to the specific embodiment of FIG. 8, a plurality of supporting members or bearing brackets 246 protrude from the periphery 248 of the mirror 222. In a specific embodiment, the support member 246 includes a spacer 250, and the spacer 250 is suitable for joining the surface of the substrate 226 when the mirror 222 is in a neutral position. In another embodiment, the substrate 250 is located on the surface of the substrate 226. In another embodiment, the spacer 250 also serves as the movement limit or end stop of the mirror 222. In another specific embodiment, when the spacer 250 is in the neutral position of the mirror 222, the fixed and repeatable relationship of the mirror 222 with respect to the substrate 226 can be maintained. As used herein, the term "neutral position" means the relationship of the mirror to the surface of the substrate when the thermal actuator is in an inactive position. In a specific embodiment, the bearing bracket 246 is docked on the neutral position cushion 250. The neutral position can also be coplanar or non-coplanar.

A plurality of thermal actuators 252 are positioned around the periphery of the mirror 222. The number, location, and configuration of the thermal actuators 252 can vary depending on the application. In this embodiment, the thermal actuator 252 may be located at a corner of the square mirror 222. The free end 253 of the thermal actuator 252 is located under the support member 257 of the corner of the mirror 222, but is not attached to the support member 257. The mirror 222 is attached to the substrate 226 by a universal joint 228 independent of the thermal actuator 252. When any thermal actuator 252 is activated, one or more free ends 253 engage adjacent support members 257, and -16- this paper size applies Chinese National Standard (CNS) A4 specifications (210X 297 mm) 548232 A7 B7 5 Explanation of the invention (14 Lifting mirror 222 leaves the plane (refer to FIG. 10). When the thermal actuator 252 is in the inactive position, the mirror 222 returns to the neutral position (refer to FIG. 8). When the actuator 252 is in the non-active position When the position is activated, due to the torsion of the universal joint, the mirror 222 substantially returns to the neutral position. In a specific embodiment, the mirror 222 can be returned to the neutral position by electrostatic force.

Each thermal actuator 252 includes one or more anchors 254, 256. The electrical track 25 8 connects the anchor 254 to the ground track 260. The electrical track 262 connects the anchor 256 to the current source 2 6 4. As exemplified in FIG. 10, the mirror 222 may be moved and / or rolled away from the plane by selectively applying a current to a portion of the thermal actuator 252. When the thermal actuators 252A, 252B, 252C are moved to a non-coplanar configuration, the free end 253 A-253C engages the support member 257 and raises the mirror 222.

In some specific embodiments, the free ends 253A-253C move through the arc at the active position, so the free ends 253 have a number of lateral displacements (parallel to the substrate surface direction) relative to the support 257. As a result, when the mirror 222 is raised, the free end 253 can slide along the lower surface of the support member 257 (or the mirror 222). Part or all of the universal joint 228 is deformed, and the displacement of the mirror 222 is compensated. Since the free ends 253A-253C are not attached to the support member 257, the mirror 222 can move with less force and higher accuracy. Fig. 11 is a perspective view of a second embodiment of a microcomputer device 290 according to the present invention. A pair of thermal actuators 300, 302 are mechanically coupled to a micromirror 304 formed on a substrate 306. The mirror 304 is attached to the substrate 306 by one or more curved portions or hinges 308 formed on the first end 310. The curved part used on the thermal actuator is suitable for this project. Otherwise, the mirror 304 is not attached to the substrate. The hinge 308 provides at least one degree of freedom to the micromirror. -17- This paper size is in accordance with Chinese National Standard (CNS) A4 specification (210X 297 mm) 548232 A7 _______B7 V. Description of the invention (15) The 308 keys of the Syrian keys are flat on the surface of the substrate 306, but can be rotated away from flat. A variety of hinges and plates with different sizes, shapes, and positions on the substrate can be used. The hinge 308 may provide a return force to the twisted hinge or curved portion to move the mirror back to a coplanar position (refer to FIG. 11). In addition, the mirror 304 can restore the coplanar configuration by gravity, a microcomputer spring structure formed on the substrate 306, electrostatic force, or a variety of other mechanisms. The formation of such a hinge plate is disclosed in Pister et al., "Micromechanical Hinge", No. 33, Sensors and Actuators a, pp. 249-256, 1992, and US Patent Nos. 5,923,798 and 5,912,094. The shaft 312 is attached to the mirror 304 near the first end 31. The shaft 312 may be supported by multiple anchors 314 to allow the shaft 3 12 to rotate within a certain limit. One or more flip levers 316 are anchored to the shaft 312. The free ends 318, 320 of the thermal actuators 300, 302 are located below the flip lever 316, respectively. The positions of the free ends 318, 320 are preferably close to the shaft 312 'so as to obtain the greatest mechanical advantage. FIG. 12 is a perspective view of the thermal actuators 30 and 30 of FIG. 11 in an activated configuration. When a current is applied to the thermal actuators 300, 302, the lifting force discussed above raises the flip lever 316, which in turn raises the micromirror 304 to an off-plane position. When the current is removed, the micromirror 30 returns to a substantially coplanar position, as shown in FIG. 11. Although Figures 11 and 12 illustrate a mirror with two thermal actuators, a single thermal actuator or multiple thermal actuators may be used. The microcomputer device of the present invention can use multiple actuators to position the mirror, including: using electrostatic, piezoelectric, thermal, and magnetic actuators. For example, U.S. Patent No. 6,028,689 (Michalicek et al.) Discloses multiple moving micromirrors manipulated by electrostatic potentials. In a specific embodiment, the mirror is positioned by a thermal actuator. Micron-sized thermal actuator can repeat the rapid movement of the micro-mirror off the plane, repeat exactly -18- This paper size applies Chinese National Standard (CNS) A4 specifications (210X 297 mm) 548232 A7 B7 V. Description of the invention (Controlling the beam 13 to 17 illustrate a specific embodiment of a thermal actuator 450 suitable for use in the present invention. The term "thermal actuator" as used herein means a removable optical device, such as the micromirror in a coplanar position and a non- Thermally-activated microcomputer device between coplanar locations. In this specific embodiment, the thermal actuator 450 is designed to provide controlled bending. "Controlled bending" as used herein means that it occurs mainly in discrete locations rather than along the The bending of the actuator beam distribution. The thermal actuator 450 is coplanarly oriented on the surface of a substrate 452, which typically includes a silicon wafer 454 with a layer of silicon nitride 456 deposited thereon. The actuator 450 includes a nitride The first polycrystalline silicon layer 460 on the silicon layer 456. As shown most clearly in FIG. 16, the first layer 460 includes bumps, which form a reinforcement 485 of the cold beam 484. The second polycrystalline silicon layer 462 is configured as Have first and second anchors 464, 466 and A pair of beams 468, 470 are cantilevered by anchors 464, 466. In the specific embodiment shown in FIG. 13, the anchors 464, 466 include electrical contacts 476, 478 formed on the substrate 452, and the electrical contacts are suitable for carrying current to Beams 468, 470. The tracks 476, 478 typically extend to the edge of the substrate 452. In addition, various electrical contact devices and / or packaging methods such as ball grid array (BGA), land grid array (LGA), plastic lead chip carrier (PLCC) ), Pin grid array (PGA), edge card, small outline integrated circuit (SOIC), dual online package (dip), quad flat package (QFP), leadless chip carrier (LCC), chip size The package (CSP) can be used to transmit current to the beams 468, 470. The beams 468, 470 are electrically and mechanically coupled to the respective free ends 471, 473 to form a circuit. In another specific embodiment, the beam 468 , 470 series electric light closing to the ground mark 477. The ground mark 477 is electrically coupled to the beam 468, -19- binding

The size of the paper is applicable to the Chinese National Standard (CNS) A4 specification (210X297 mm) 548232 A7 B7 V. Description of the invention (The contact 479 from 470 to the substrate 452 is in an unactivated configuration (refer to Figure 14) and the activated group (Refer to FIG. 17). The grounding mark 477 may be a flexible or spring piece, and the grounding pin is suitable for maintaining contact with the contact 479. The grounding pin may be used in any of the specific embodiments disclosed herein.

The beams 468, 470 are physically separated from the first layer 460, so the member 472 is located above the base plate 452. One or more dimples 474 may be formed on the member 472 俾 support beams 468, 470 above the base plate 452 as needed. In another embodiment, the dimples or bumps 474 may be formed on the substrate 452. As shown in Fig. 14, the beams 468, 470 are substantially parallel to the surface of the substrate 452 in an unactivated configuration. The "unactivated configuration" used herein means that there is substantially no current flowing through the circuit formed by the beam 468, the member 472, and the beam 470.

The third layer of polycrystalline silicon 480 is provided with anchors 482 which are attached to the anchors 464 and 466 of the substrate 452. The third layer 480 forms an upper beam 484 cantilevered by an anchor 482. The anchor 482 has a free end 483 which is mechanically coupled to a member 472 above the beams 468,470. In several specific embodiments, the reinforcing member 485 is formed at least a part of the length of the upper beam 484; and the curved portion 4 7 7 is formed near the anchor 482 of the upper beam 484 if necessary. In a specific embodiment, a metal layer is applied to the upper beam 484 as needed. As used herein, "reinforcement" means one or more ridges, bumps, grooves, or other structures that increase resistance to bending. Reinforcement is typically formed during the MUMPs process, so it is integrally formed with the upper beam 484. In the illustrated embodiment, the reinforcing member 485 is a curved ridge (refer to FIG. 16) extending along a part of the upper beam 484, but may be rectangular, square, triangular, or various other shapes. In addition, the stiffener 485 may be located at the center of the upper beam 484 or along its edge. Multiple reinforcements can also be used. -20- The size of this paper applies to the Chinese National Standard (CNS) A4 (210 X 297 mm) 548232 Five A7 B7. Description of the invention (18) The "curved part" used here means narrow, thin or fragile materials, Recesses, depressions, holes, slots, cutouts of other materials or other structures or material changes that can provide controlled camber at specific locations. Other materials suitable for use as curved parts include polycrystalline silicon, metal or polymeric materials. As best shown in Figures 13 and 15, the curved portion 487 is a recessed portion 489. The recess 489 contains the most vulnerable section of the upper beam 484, so that this position is most likely to bend during actuation of the thermal actuator 450. The rigidity of the upper beam 4 8 4 relative to the rigidity of the bent portion 487 determines the controlled bending amplitude (position and direction) of the thermal actuator 450 to a great extent. In a specific embodiment, the reinforcing member 485 is used in combination with the curved portion 487. In another embodiment, the reinforcing member 485 extends along a portion of the upper beam 484 without using a curved portion. The portion of the upper beam 484 excluding the reinforcing member 485 is a position where the camber can be controlled. In yet another embodiment, the bent portion 487 is formed on the beam 484 without the reinforcing member 485, so the bent portion 487 has a controlled camber position. The thermal actuator 450 may also be used without the reinforcement 485 or the bent portion 487. A through hole 488 is formed in the member 472 and / or the free end 483 to mechanically couple the free end 483 of the upper beam 484 to the member 472. Other structures may be used to mechanically couple the member 472 of the upper beam 484. The upper beam 484 is substantially parallel to the surface of the base plate 452 in an unactivated configuration. FIG. 17 is a side cross-sectional view of the thermal actuator 450 of FIGS. 13-16 in a non-coplanar or activated configuration. "Active configuration" means the application of current to one or more beams. In the illustrated embodiment, a current is applied to a circuit formed by the beam 468, the member 472, and the beam 470 (refer to FIG. I3). Beams 468, 470 are "hot arms" and beams 484 are cold arms. Used here "hot arm" or "multiple hot arms" means when the voltage is applied

Line -21-This paper size applies to China National Standard (CNS) A4 specification (210X297 mm) 548232 Five A7 B7, In the invention description (u), beams or members with higher current density than cold arms. A "cold arm" or "multiple cold arms" means a beam or member that has a lower current density than a hot arm when a voltage is applied. In some embodiments, the cold arm has a zero current density. As a result, the hot arm has greater thermal expansion than the cold arm. The current heats the hot arms 468, 470, extending the length of the hot arms in the direction 490. The expansion of the beams 468, 470 causes the free ends 471, 473 of the thermal actuator 450 to move in an upward arc 492, generating a lifting force 494 and a displacement 495. However, the cold arm 484 is fixed to the anchor 482 and is electrically isolated, so the current is completely or substantially completely passed through the circuit formed by the hot arms 468 and 470 and the member 472. Due to the height difference between the cold arm 484 and the hot arms 468, 470, a moment is applied near the anchor 482 of the cold arm 484. The cold arm 484 is bent near the bent portion 487, and as a result, the displacement near the free end 483 is larger than the displacement without the bent portion 487. The hot arms 468 and 470 are also easily bent, and there is little resistance to the movement 492 of the cold arms 484. The reinforcing member 485 resists bending along the cold arm 484. When the load is placed at the free end 483, bending along the cold arm 484 near the member 472 often occurs. In the illustrated embodiment, the displacement 495 is from 0.5 micrometers to 4 micrometers. When the current is over, the thermal actuator 450 returns to its previously unactivated configuration, as shown in Figure 14. In another embodiment, the anchor 482 and the cold arm 484 are electrically coupled to the member 472. The current flowing at least partially through the hot arms 468, 470 flows along the cold arms 484 to the anchor 482. It is also possible that all of the current flowing through the hot arms 468, 470 flows out of the thermal actuator 450 through the cold arms 484. The material and / or geometry of the cold arm 484 is suitable for having a lower current density than the hot arms 468, 470, and lower current density than the hot arms even when the same voltage is applied. In a specific embodiment, the cold arm 484 is made of a paper with a size of -22-. This paper applies the Chinese National Standard (CNS) A4 specification (210X297 mm).

Binding

548232 five A7 B7, the description of the invention (%) is made of a material with a linear thermal expansion coefficient smaller than that of the thermal arms 468, 470. In yet another embodiment, the cold arm 484 provides a lower resistivity by having a larger cross-sectional area. In another embodiment, a conductive layer is disposed on the cold arm 484. Suitable conductive materials include metals such as aluminum, copper, tungsten, gold or silver, semiconductors, and doped organic conductive polymers such as polyacetylene, polyaniline, polypyrrole, polyfluorene, polyEDOT, and derivatives or combinations thereof. As a result, the net expansion of the hot arms 468, 470 is greater than the net expansion of the cold arms 484. In yet another embodiment, all or part of the current flowing through the hot arms 468, 470 flows through the ground mark 477 to the contact 479 of the substrate 452. When the thermal actuator 450 is moved from an unactivated position to an activated position as shown in FIG. 17, the ground mark 477 can maintain electrical contact and physical contact 479. Another thermal actuator is disclosed in commonly assigned U.S. Patent Application No. 09 / 659,572, entitled "Direct Action Vertical Thermal Actuator", application dated September 12, 2000; U.S. Patent Application No. 09 / No. 659,798, name Γ Direct-acting vertical thermal actuator with controlled camber ", application dated September 12, 2000; and U.S. Patent Application No. 09 / 659,282, name" Combined Horizontal and Vertical Thermal Actuator ", application date is September 12, 2000. 18 and 19 illustrate another thermal actuator 500 suitable for use with the wavelength equalizer of the present invention. The thermal actuator 500 is schematically illustrated in FIGS. 1 to 17 and includes a waveguide 502 attached to a cold arm 504. The waveguide 502 is formed as a part of the manufacturing process or added as a separate component. The waveguide 50 2 is typically an optical fiber. Since the waveguide has little or no thermal expansion, the waveguide 502 can be effectively positioned on the cold arm 504. The cold arm 504 is preferably electrically isolated from the hot arms 506, 508. -23- This paper size applies to China National Standard (CNS) A4 (210X297 mm)

Binding

548232 Five A7 B7, description of the invention (21) As shown most clearly in Figure 19, when the thermal actuator 500 is in an unexcited configuration or a coplanar configuration, the waveguide 502 is set to be optically coupled to the adjacent waveguide. 5 10 (refer to Figure 3). Upon activation or non-coplanar configuration, the waveguide 502 can be steered by the adjacent waveguide 510 (refer to Figure 4). By changing the current applied to the thermal actuator 500, the discrete wavelength signal 512 can be attenuated as necessary. Fig. 20 illustrates a fifth specific embodiment of another wavelength equalizer 520 using the thermal actuator shown schematically in Figs. The input fiber 522 carries a multiplexed signal 526. The input fiber 522 is optically coupled to a wavelength division demultiplexer 524. The demultiplexer 524 converts the multiplexed optical signal 526 carried by the input optical fiber 522 into a plurality of discrete wavelengths or channels 528. The discrete wavelengths 528 are each transmitted via a discrete optical fiber 530, and the discrete optical fibers are each mounted on a thermal actuator 532. When the thermal actuator 532 is in the name configuration shown in FIG. 19, substantially the entire optical signal 528 is transmitted from the optical fiber 530 to the corresponding optical fiber 534. In the name configuration, the thermal actuator 532 is preferably a coplanar and inactive configuration. The optical fiber 534 is coupled to the multiplexer 536. The multiplexer 536 combines discrete wavelengths 528 into an output signal 538. The current can be applied to one or more thermal actuators 532. Therefore, the optical fibers 530 and 534 are not aligned. Partial or complete attenuation (refer to Figure 4). In a specific embodiment, the controller 540 monitors the upstream and / or downstream signal strength of the optical fibers 530, 534, and controls the position of the thermal actuator 532 to equalize the signal strength. Although specific specific embodiments of the present invention have been shown and described herein, it should be understood that these specific embodiments are merely examples to illustrate a number of possible specific configurations in the application design of the principles of the present invention. Those skilled in the art have not deviated from this hair -24- This paper size applies to China National Standard (CNS) A4 (210X 297 mm) binding

548232 Five A7 B7 Description of the Invention (22) The scope and essence of the invention can make countless other configurations based on these principles. For example, any of the curved sections, reinforcement structures, anchor locations, and beam configurations disclosed herein can be combined to produce a variety of thermal actuators.

-25- This paper size applies to China National Standard (CNS) A4 (210X297 mm)

Claims (1)

548232 No. 090125548 patent application A8 B8 C8 Chinese application for deafness, repair] too much "91 cars 11 ^ D8 month correction | patent application scope a wavelength equalizer, including: a demultiplexer, which Suitable for separating wavelength division multiplexed signals into a plurality of discrete wavelength signals, and guiding each discrete wavelength signal along multiple first optical paths; a microcomputer device including at least one micromirror optically coupling each first light Diameter; multiple second optical paths, which are located to receive discrete wavelength signals reflected by respective micromirrors; at least one actuator, which is mechanically coupled to each micromirror, the actuator is suitable for selective Displace one or more micromirrors to turn at least part of the discrete wavelength away from the corresponding second optical path; and 2. a multiplexer suitable for combining discrete wavelength signals combined at multiple second optical paths to a restructured wavelength Divides the multiplexed signal. The number is equal to the wavelength equalizer of item No. of the patent application, in which 5 the direction of the micromirror is reflected to the discrete wavelength signal corresponding to the second optical path. Intensity. ^ Includes the wavelength equalizer of item 1 in the patent application range, where the first light beam contains the second optical path. Intensity, such as the wavelength equalizer of item 丨 in the patent application range, includes multiple signal detection The device is suitable for detecting optical signals of discrete wavelength signals such as the wavelength equalizer of the first patent application range, and includes: a beam splitter arranged along a plurality of first optical paths; and a signal intensity detection n, It is an optical method of Hanhe to Beamsplitters. The paper size is applicable to China National Standard (CNS) A4 (210X297 mm) A BCD 548232 6. Scope of patent application 1. Measure the signal strength of discrete wavelength signals. 6 • The wavelength equalizer as described in the first patent application, including a controller operatively coupled to the actuator. 7. The wavelength equalizer according to item 1 of the patent application scope, including: multiple k-number intensity detectors, which are suitable for detecting the signal strength of discrete wavelength signals; and a controller, which is suitable for monitoring the strength of the borrowed signal strength detection The discrete-wavelength signal strength measured by the detector, and the activation actuator 选择性 selectively attenuates one or more discrete-wavelength signals in response to the monitored signal strength. 8. The wavelength equalizer as described in the first patent application range, wherein the micromirror and the actuator are constructed on the surface of the substrate. Among them, each of the micromirror systems, each of the micromirror systems, each of the micromirror systems, and the actuator package 9 The wavelength equalizer such as the eighth item of the patent application scope is attached to the substrate surface by at least one hinge. 1 0. The wavelength equalizer such as the item No. 8 in the scope of the patent application is attached to the substrate surface by at least one universal joint. 1 1. The wavelength equalizer such as the item No. 8 in the patent scope is suitable for movement through at least one degree of freedom. 1 2 · The wavelength equalizer as in item 1 of the patent application scope includes a thermal actuator. 13 • The wavelength equalizer according to item 12 of the patent application scope, wherein the thermal actuator comprises: at least one thermal arm having a first end anchored to the surface and a free end located above the surface; a cold Arm 'having a first end anchored to the surface and a free end The cold arm is located above the hot arm and opposite to the surface; and: the component, which mechanically and electrically couples the free end of the hot arm and the cold arm, when the current is applied to at least the hot arm, the component is positioned to Attach the mirror. The wavelength equalizer of item 12 of the patent, wherein the thermal actuator includes at least one curved portion adapted to provide a controlled camber. 15. As the scope of patent application. A wavelength equalizer, wherein the thermal actuator includes at least one stiffener. 16 · The wavelength equalizer as described in item 1 of the patent scope, when the actuator is in a non-activated configuration, a micromirror reflects substantially from one of the first optical paths to the corresponding second optical path. All discrete wavelength signals. 17 · If the scope of patent application is the first! In terms of the wavelength equalizer, when the actuator system is in an activated configuration, a micromirror reflects only a part of discrete wavelength signals from one of the first optical paths to the corresponding second optical path. 1 8 · The wavelength equalizer as described in item 1 of the patent application scope, in which when the actuator is the starter of the t group, a micromirror does not reflect any discrete wavelength from one of the first optical paths to the corresponding first optical path. signal. 19 · An exchangeable wavelength division multiplexed optical communication system, comprising at least one optical fiber carrying a wavelength division multiplexed signal and a wavelength equalizer as described in the first patent application. 20. A wavelength equalizer comprising: a demultiplexer adapted to separate a wavelength division multiplexed signal into a plurality of discrete wavelength signals, and to guide each discrete wavelength signal along a plurality of first optical paths ; Multiple second optical paths, set to receive discrete wavelength signals; -3- This paper size applies to China National Standard (CNS) A4 (210 X 297 mm) AB c D 548232 Sixth, the scope of the patent application is a microcomputer device that includes a thermal actuator, which is suitable for selectively turning at least the wavelength of the divergent divergence from one or more first optical paths away from the corresponding second optical path: The discrete wavelength signals used to combine the multiple second optical paths become restructured wavelength division multiplexed signals. 2 1 The wavelength equalizer as described in the patent application No. 20, wherein the microcomputer device includes a plurality of micromirrors mechanically coupled to a plurality of thermal actuators. 2 2 · The wavelength equalizer according to the scope of patent application No. 20, wherein each of the plurality of first optical paths extends along the surface of the corresponding thermal actuator. 2 3. The wavelength equalizer according to item 22 of the scope of patent application, wherein when the thermal actuator is in an unactivated condition, the second optical path is optically aligned with each corresponding first optical path, respectively. 24. The wavelength equalizer according to item 22 of the scope of patent application, wherein when the thermal actuator system is in an activated condition, the first optical path is respectively optically aligned with each corresponding first optical path. 2 5 · — An exchangeable wavelength division multiplexed optical communication system comprising: at least one optical fiber carrying a wavelength division multiplexed signal; a wavelength equalizer comprising; a demultiplexer, which is suitable for The wavelength division multiplexed signal is separated into a plurality of discrete wavelength signals, and each discrete wavelength signal is guided along a plurality of first optical paths; a microcomputer device including at least one micromirror optically coupled to each first optical path; a plurality of The second light path is located at -4 which can receive the reflection from the respective micromirrors.-This paper size is applicable to the Chinese national standard (c ^ n ^ i (210X 297 male 17 548232 AB c D 6. The position of the patent application Fan Yuan scattered wavelength signal; at least one actuator, which is mechanically coupled to each micromirror, the actuator is suitable for selectively displacing one or more micromirrors, turning at least partly discrete wavelengths away from Corresponding to the second optical path; and a multiplexer, which is suitable for combining the discrete wavelength signals of the plurality of second optical paths into a restructured wavelength division multiplexed signal. -5- This paper size applies to China National Standard (CNS) A4 (210 X 297 mm)