IN-PLANE-MEMS-OPTIKSCHALTER

Abstract

Optical switch comprising: a first bus waveguide (4706, 4406a, 608) supported by a substrate (4702, 602, 802, 902, 1002); an optical antenna (4710, 4408, 806, 906, 1006) suspended by a spring (4712) over the first bus waveguide (4706, 4406a, 608); interlocking electrodes (4708, 4320, 4420) couple the substrate (4702, 602, 802, 902, 1002) to the optical antenna (4710, 4408, 806, 906, 1006) and are configured to control a position of the optical antenna (4710, 4408, 806, 906, 1006) relative to the first bus waveguide (4706, 4406a, 608), wherein, if a voltage difference applied to the interlocking electrodes (4708, 4320, 4420) is less than a lower threshold value, the optical antenna (4710, 4408, 806, 906, 1006) aligns itself with a position relative to the first bus waveguide (4706, 4406a, 608) is in the first offset position when the voltage difference applied to the toothed electrodes (4708, 4320, 4420) is greater than an upper threshold value,the optical antenna (4710, 4408, 806, 906, 1006) is located at a second position offset from the first bus waveguide (4706, 4406a, 608) and the offset at the second position is greater than at the first position, a second bus waveguide (4406b, 4406c) which is supported by the substrate (4702, 602, 802, 902, 1002) and is parallel to the first bus waveguide (4706, 4406a, 608), wherein, if a voltage difference applied to the toothed electrodes (4708a, 4320, 4420) is less than a lower threshold, the optical antenna (4710, 4408, 806, 906, 1006) is located at a first position in a first distance offset to the first bus waveguide (4706, 4406a, 608), if the voltage difference applied to the toothed electrodes (4708a, 4320, 4420) is greater than an upper threshold, the optical antenna (4710, 4408, 806, 906, 1006) is located at a second position at a second distance offset to the second bus waveguide (4406b, 4406c,608).

Description

FIELD OF TECHNOLOGYThis invention relates generally to a system and apparatus for guiding and emitting light signals.PRIOR ARTLidar is believed to be a key role for many new technologies, such as autonomous vehicles. It is believed that chip lidar (also referred to as solid state lidar or lidar on a chip) is widely used in future autonomous vehicles. However, there are still many issues with creating a practical chip lidar that can meet the needs of vehicle first-out equipment. One of the challenges is beam steering. Various approaches have been developed for steering a beam in the chip lidar, such as micromirror arrays, optical phased arrays, wavelength tuning, and photonic crystal waveguides, but these approaches are faced with problems such as limited field of view, complicated control electronics, and high laser requirements.US 2018 / 0 172 918 A1 discloses an optical transmit / receive terminal which is totally solid-state and includes binary optical switches for directing an optical beam without mechanical components, a phased array of emitters / collectors, or a large number of phase shifters. A lens optically couples a surface array of emitters / collectors to the free space, thereby giving each emitter / collector a corresponding direction in the free space. The emitters / collectors are also connected via an "H-tree" or other branched optical fibre network to a common input / output terminal and from there to a receiver and / or transmitter. US 2002 / 0 176 653 A1 discloses an optical switch capable of preventing occurrence of crosstalk. On a waveguide substrate, a first waveguide path and a second waveguide path are provided, which intersect at a specific angle. In the surface of the waveguide substrate, a trench is formed in a straight line crossing the central axes of the first waveguide path and the second waveguide path. The trench is so deep as to expose the entire end face of the first waveguide path and the second waveguide path. The first waveguide path and the second waveguide path are arranged such that their central axes are arranged asymmetrically with respect to the straight line.QUAK Niels et al. (Scalable Row / Column Addressing of Silicon Photonic MEMS Switches. In: IEEE photonics technology letters, Vol. 28, 2016, No. 5) show a scalable addressing scheme for silicon photonics MEMS switches.US 2017 / 0 016 990 A1 discloses an optical phased array antenna consisting of a large number of nanophotonic antenna elements. These nanophotonic phased arrays, including the nanophotonic antenna elements and waveguides, may be fabricated on a single silicon chip using complementary metal oxide semiconductor (CMOS) processes.US 2019 / 0 377 135 A1 discloses a lidar or other optical beam control device that includes an optical switch having a first port and a plurality of second ports.US 2019 / 0 170 946 A1 discloses an optical switch and an optical switching system. The optical switch includes a first waveguide, a second waveguide, and a first movable waveguide. The first waveguide and the second waveguide are immovable relative to a substrate. The first waveguide and the second waveguide are located in a first plane, and the first waveguide and the second waveguide do not intersect. The first movable waveguide is movable relative to the substrate. When the first movable waveguide is at a first location, the first movable waveguide is optically decoupled from the first waveguide and the second waveguide and the optical switch is in a pass-through state. When the first movable waveguide is at a second location, the first movable waveguide is optically coupled to the first waveguide and the second waveguide and the optical switch is in a blocking state.SUMMARY OF THE INVENTIONThe invention relates to an optical switch as defined in claims 1 and 7 and to a beam steering system as defined in claim 12.BRIEF DESCRIPTION OF THE DRAWINGSFIG. 1 is a plan view of an optical switch array system. FIG. 2 is a side view of an optical switch array system. FIG. 3 is a perspective view of a silicon photonic switch array. FIG. 4A is a plan view of an optical switch of an optical switch array. FIG. 4B is a side view of an optical switch of an optical switch array in an off state (OFF). FIG. 4C is a side view of an optical switch of an optical switch array in an on state (ON). FIG. 5A is a perspective view of an optical switch having an optically coupled bending beam (cantilever) and a fixed optical antenna in a switch array. FIG. 5B is a perspective view of a two-dimensional fixed optical antenna. Figure 6 is a perspective view of an optically coupled cantilever with a fixed optical antenna. FIG. 7A is a sectional view of the optically coupled cantilever of the MEMS switch of FIG. 6 in a rest state. FIG. 7B is a cross-sectional view of the optically coupled cantilever of the MEMS switch of FIG. 6 in an actuated state. FIG. 7C is a cross-sectional view of the fixed optical antenna of the MEMS switch of FIG. 6 in a quiescent state. FIG. 7D is a cross-sectional view of the fixed optical antenna of the MEMS switch of FIG. 6 in an actuated state. FIG. 8 is a perspective view of an optical switch array system that transmits an optical signal. FIG. 9 is a perspective view of an optical switch array system receiving an optical signal. FIG. 10 is a perspective view of an optical switch array system that receives and transmits optical signals. FIG. 11 is a perspective view of an optical switch array system including at least two types of optical switches. FIG. 12 is a perspective view of an optical switch array system including a split tree (split tree) and an optical switch. FIG. 13 is a perspective view of an optical switch array system including a splitter tree and at least two types of optical switches. FIG. 14 is a view of a switch array system configured to independently drive rows and columns. FIG. 15 is a diagram of a switch array system configured to simultaneously drive sub-arrays. FIG. 16 is a perspective view of a bus waveguide and a corresponding coupler waveguide. FIG. 17 is a graphical representation of a coupling field profile with respect to distance. FIG. 18 is a series of two-dimensional plots of coupling field profiles with respect to a propagation path. Fig. 19 is a graph of coupling efficiency with respect to the length of the coupling taper. Fig. 20 is a graph of transmission loss with respect to wavelength. FIG. 21 is a graph of energy loss with respect to gap size. FIG. 22 is a graph of a radiation pattern with respect to angle. FIG. 23 is a side view showing a waveguide grating with respect to the clearance angle. FIG. 24 is a perspective view of an electrically biased beam spring. FIG. 25 is a graph of displacement along a Z-axis with respect to shift speed. FIG. 26 is a plan view of a switch array assembly showing a single element in detail. FIG. 27A is a perspective view of a switch array having a movable optical coupler on a suspension layer and a fixed optical antenna on a substrate layer. FIG. 27B is a top view of a switch array with a movable optical coupler on a suspension layer and a fixed optical antenna on a substrate layer. FIG. 28A is a cross-sectional view of the movable optical coupler of FIG. 27 in the OFF position. FIG. 28B is a cross-sectional view of the movable optical coupler of FIG. 27 in the ON position. Figure 29 is a top view of a switch array assembly having a movable optical coupler on a suspension layer and a fixed optical antenna on a substrate layer, showing a single element in detail. FIG. 30 is a top view of a switch array assembly having a movable optical coupler on a suspension layer and a fixed optical antenna on a substrate layer, showing a single element in detail. FIG. 31 is a top view of an element of a switch array assembly having a movable optical coupler on a suspension layer and a fixed optical antenna on a substrate layer. FIG. 32 is a cross-sectional view of the element of FIG. 31. FIG. 33 is a graph of a field profile of a grid antenna with respect to the X and Y coordinates. FIG. 34 is a perspective view of an illustrative deflection of an optical coupler. FIG. 35 is a perspective view of a grid antenna showing radiation angles. FIG. 36 is a graph of intensity with respect to radiation angles. FIG. 37 is a side view illustrating a mirror for compensating for a non-zero antenna radiation angle. Fig. 38 is a side view showing an optical prism for compensating for a non-zero antenna radiation angle. FIG. 39 is a side view illustrating a micro prism array for compensating for a non-zero antenna radiation angle. FIG. 40A is a top view of an antenna array with divider beams and switches. FIG. 40B is a plan view of an antenna array having multiple switch types. FIG. 40C is a plan view of an antenna array including division beams and multiple types of switches. FIG. 41A is a perspective view of a switch array having a movable grid traveling longitudinally with multiple holding positions. FIG. 41B is a detailed perspective view of a movable grating traveling longitudinally of the bus waveguide of FIG. 43A with multiple stop positions. FIG. 41C is a perspective view of a switch array with a movable grating traveling longitudinally of the bus waveguide. FIG. 41D is a detailed perspective view of two movable gratings traveling longitudinally of the bus waveguide of FIG. 43C. Fig. 41E is a plan view of a movable grating traveling longitudinally of the bus waveguide via interdigitated electrodes. FIG. 42A is a perspective view of a switch array with a movable grid traversing in a transverse direction with multiple holding positions to the bus waveguide. FIG. 42B is a detailed perspective view of a movable grid of the switch array of FIG. 43A, which grid moves in the transverse direction to the bus waveguide with a plurality of holding positions. Figure 42C is a top view of a movable grating traversing across interdigitated electrodes in the transverse direction of the bus waveguide. Figure 43 is a perspective view of a switch array with a movable grating traversing transversely with stop positions in and out of the plane of waveguides. FIG. 44 is a perspective view of an electrostatic levitation switch for controlling a coupling distance between a bus waveguide and the gratings. FIG. 45A is a perspective view of a switch having a comb drive for controlling a coupling distance between a bus waveguide and the gratings. FIG. 45B is a cross-sectional view taken along a section of FIG. 47A for illustrating the comb drive. FIG. 46 is a cross-sectional view of a switch array system illustrating encapsulation. Fig. 47 is a perspective view of a double-grating switch traveling transversely with holding positions in the plane of waveguides. Fig. 48 is a perspective view of a dual grating switch traversing transversely with stop positions in and out of the plane of waveguides. Fig. 49 is a perspective view of a bimorph switch in which a grating is an in-plane position. Fig. 50 is a perspective view of a bimorph switch which vertically rotates a grating at an out-of-plane position. Figure 51 is a perspective view of a switch array in which grating elements rotate about an axis perpendicular to the waveguide. FIG. 52 is a perspective view of a switch array in which grating elements rotate about an axis parallel to the waveguide. FIG. 53 is a plan view of a grid array on a turntable that utilizes a rotating mechanism for in-plane and out-of-plane operation. Figure 54A is a cross-sectional view of a coupler waveguide with a bistable membrane in the OFF position. FIG. 54B is a cross-sectional view of a coupler waveguide with a bistable membrane in the ON position.DETAILED DESCRIPTIONDetailed embodiments of the present invention are disclosed herein as needed; however, it is to be understood that the disclosed embodiments are mere examples of the invention that may be embodied in various and alternative forms. The drawings are not necessarily to scale; some features may be exaggerated or minimized to show details of particular components. Therefore, specific structural and functional details disclosed herein are not to be interpreted as limiting, but merely as being representative of one of ordinary skill in the art to obscure the various uses of the present invention.For purposes of describing disclosed or claimed embodiments, the term "substantially" is used herein. The term "substantially" may modify a value or relative characteristic disclosed or claimed in the present disclosure. In such cases, "substantially" may mean that the value or relative characteristic thus modified is in the range of ± 0%, 0.1%, 0.5%, 1%, 2%, 3%, 4%, 5% or 10% of the value or relative characteristic.Although various embodiments using silicon MEMS structures are illustrated in this application, the MEMS structures may be another MEMS material such as SiC, SiN.This disclosure includes microelectromechanical (MEMS) switches, switch arrays, and systems that include MEMS switches with coupler cantilevers and multiple fixed optical antenna configurations. Such MEMS switches may be used to couple light from a waveguide on a photonic integrated circuit (PIC) chip to free space in various applications.Lidar plays a key role in many new systems such as autonomous driving. It is believed that chip lidar (also referred to as solid state lidar lidar lidar on a chip) will find wide use in future autonomous vehicles due to the compact size and low cost. Nevertheless, there are still numerous issues with creating a practical chip lidar that satisfies the needs of vehicle first-out equipment. One of the challenges is beam steering. Various approaches have been developed for beam steering in the chip lidar, such as micromirror array, optical phased array, wavelength tuning, photonic crystal waveguides, etc. However, these approaches are faced with one or more challenges, such as limited field of view, complicated control electronics, high laser requirement, low technological maturity, etc.In addition to the above-mentioned approaches, an optical amplitude array can also direct an optical beam in the chip lidar. FIG. 1 is a plan view of an optical switch array system 100. The system 100 includes a photonic integrated circuit (PIC) chip 102, multiple optical sources / antennas 104, and an objective 106. FIG. 2 is a side view of an optical switch array system 200. The system 200 includes a photonic integrated circuit (PIC) chip 202, a plurality of optical sources / antennas 204, and an objective 206.As shown in FIG. 1 (plan view) and FIG. 2 (side view), an optical amplitude array is comprised of an array of optical sources or optical antennas (e.g., 104, 204) (each optical source or optical antenna may be considered a "pixel") and an objective or lens system (e.g., 106, 206) ("objective" is used for simplicity in the following description), the optical pixel array being located on the focal plane (or in close proximity to the focal plane) of the objective. The amplitude of light emitted by each optical pixel may be either 0 (no light emission) or 1 (full light emission). The light emitted from each optical pixel is refracted when passing through the objective and deflected in a certain direction according to physical-optical laws. By turning on an optical pixel at different positions, light can be controlled and directed in two different directions in free space, and thus beam steering is achieved. A two-dimensionally (2D) arranged optical pixel array can realize 2D beam steering.There are various ways to create an optical pixel array. For example, such an array may be made from an array of optical sources, such as vertical external cavity surface emitting lasers (VECSELs). Such an array may also consist of a waveguide array with an optical antenna array controlled by an optical switch array. In this disclosure, methods are being explored that include MEMS optical switch arrays configured to control light emission of optical antennas.The present disclosure includes beam steering enabled by a MEMS switch-based amplitude array. This MEMS switch-based amplitude array consists of a waveguide array in which MEMS switchable antennas are arranged on or in the immediate vicinity of the focal plane of an objective. Turning on optical antennas at different positions may direct the beam in different directions, as shown in FIGS. 1 and 2.Compared to micromirror array-based beam steering, amplitude array beam steering can achieve a larger field of view. Compared to beam steering by an optical phased array, the control electronics required for amplitude array beam steering is much simpler. In comparison with wavelength-tuning-based beam steering, no great wavelength-tunability of the laser source is necessary in the amplitude array beam steering. Compared to beam steering based on photonic crystal waveguides, amplitude array beam steering is more uncomplicated and technically more stable.There are many options for optical switches. The most widely used switches that can be used in the amplitude array application include micromirror-based switches, liquid crystal switches, thermo-optical switches, etc. The micromirror-based switches and liquid crystal switches are relatively large, such as several hundred microns, and require a switching time in the millisecond range, so that each pixel is too large and too slow for the chip lidar application. The thermo-optical switches, on the other hand, have a low extinction ratio and a high insertion loss, which is not good for this application either. A MEMS structure-based optical switch may be compact and fast switching and have a high extinction ratio, making this technology a good candidate for the amplitude array application.FIG. 3 is a perspective view of a silicon photonic integrated (PIC) switch array 300. The array has input ports 310, pass-through ports 314, and exit ports 312 such that when a low-loss crossover 316, such as an optical switch, is activated, light passes from the input port 310 through the low-loss crossover 316 to the exit port 312. In order for the light to pass from the input port 310 to the low loss crossover 316, a MEMS actuated adiabatic coupler must be at a certain distance from the bus waveguide. If a low-loss crossing 316 is not activated, the light travels from the input port 310 to the pass-through port 314.A close-up view of an optical switching element of array 304 includes a MEMS actuated adiabatic coupler 318 and a bus waveguide 320. An illustration shows the optical switch in a switched-off state 306, in which light passes from the input 310 to the output 314 while the cantilever coupler is at a distance 316 from the bus waveguide, such that the coupling efficiency is less than 1%. An illustration shows the optical switch in a switched-on state 308 in which light passes from the input 310 to the output 312 while the cantilever coupler 318 is activated so that light from the bus waveguide couples to the output 312 with a coupling efficiency of over 50%.As shown in FIG. 3, such a silicon photonic MEMS switch network utilizes two orthogonal sets of bus waveguides and MEMS actuated vertical adiabatic couplers. The vertical gap distance may be controlled by electrostatic MEMS actuators and the mechanical stops. When in the off (OFF) state, the adiabatic couplers are located far above the waveguides, light continues to propagate without interruption towards the pass-through port. In the on (ON) state, the adiabatic couplers are moved toward the bus waveguides, and light is coupled into the adiabatic couplers and then coupled out to the outgoing port through another adiabatic coupler.With the aid of MEMS switches, light can be coupled selectively into and out of a waveguide in an optical transmission / reception terminal. In one embodiment, each optical switch may be implemented by a physically displaceable optical grating. In the OFF state, the displaceable optical grating is situated far above the bus waveguide, and in the ON state, MEMS actuators can move the displaceable optical grating downward toward the bus waveguide for efficient light coupling between the grating and the bus waveguide.FIG. 4A is a plan view of an optical switch 400. The optical switch 400 has a grating 402 comprising a plurality of optical coupler elements 404, 406, 408 aligned with a bus waveguide 410. FIG. 4B is a side view of the optical switch 400 in an OFF state. A shaft 412 travels in a direction 414 in the bus waveguide 410. The bus waveguide is supported on a substrate 416, the grating 402 is situated at a distance 418 above the bus waveguide 410, so that the coupling efficiency between the grating 402 and the bus waveguide 410 is low and the wave propagates further in the bus waveguide 410. FIG. 4C is a side view of the optical switch 400 in an ON state. In this figure, the grating 402 is located a distance 420 above the bus waveguide 410 so that the coupling efficiency between the grating 402 and the bus waveguide 410 is high and the wave breaks 422 from the bus waveguide 410 into the free space.In the present disclosure, several MEMS switch embodiments include a coupler cantilever and optical antenna MEMS switch. FIG. 5A illustrates an array of such MEMS switches used in optical terminals. As an example to illustrate the concept, gratings are used here as the optical antenna, other optical antennas (for example, FIG. 5B ) also function in this concept.FIG. 5A is a perspective view of an optical switch having an optically coupled cantilever (cantilever) and a fixed optical antenna in a switch array 500. The switch array 500 may be located on a single substrate 502 or configured as a multichip module in which multiple optical switches are monolithically integrated on a photonic integrated circuit (PIC) chip, which is then combined with other PIC chips to form the array system 500. The single substrate photonic integrated circuit (PIC) chip 502 includes a bus waveguide configured to introduce light 504 and distribute the light to multiple rows 510 and columns of optical switches. Here, three rows are provided, as a first row 510 a, a second row 510 band a third row 510 c, where each row has three columns of optical switches 506.FIG. 5B is a perspective view of a two-dimensional fixed optical antenna 550. The optical antenna includes a coupler cantilever 552 and an optical antenna having a length 554 and width 556. The ratio of length 554 to width 556 may be substantially 3:1, 5:2, 2:1, or the like. With a gap 558 between the optical antenna and the side that may be substantially 0.2, 0.3, 0.4, 0.5, 0.6 of width 556. The optical antenna may be configured to output or receive light substantially at an angle in an x-axis 560 and at an angle in a y-axis 562.FIG. 6 is a perspective view of a MEMS switch 600 having an optically coupled cantilever 614 and a fixed optical antenna 616. The MEMS switch 600 includes a substrate 602, such as a silicon substrate, with an insulating layer 604, such as silicon dioxide. On top of the insulating layer 604 are pattern layers 606 such as silicon layers. The structure layers 606 include structures such as a bus waveguide 608, a MEMS actuation electrode 610, a MEMS spring 612, the coupling beam 614, and the optical antenna 616. In this embodiment, the movement or deflection of the coupling waveguide 614 is performed according to MEMS principles. For example, the MEMS actuation electrode 610 may be configured to cooperate with a reaction electrode disposed adjacent to the coupling waveguide and between the MEMS spring 612 and the coupling waveguide 614. If a voltage difference is applied between the reaction electrode and the actuation electrode, which voltage difference is smaller than a lower threshold value, the coupling waveguide is held at a first distance from the bus waveguide by electrostatic forces. The lower threshold may be a low differential voltage such as 0 volts or any low voltage around zero volts such as -5, -4, -3, -2, -1, 1, 2, 3, 4, or 5 volts. At a voltage below this lower threshold, the coupling cantilever may either maintain a rest position or may move somewhat such that the coupling efficiency between the bus waveguide 608 and the optical antenna 616 via the coupling waveguide 614 is less than 1% (e.g., turned off). When the voltage difference between the reaction electrode and the actuation electrode 610 is greater than an upper threshold, the coupling waveguide is maintained at a second distance to the bus waveguide 608 via electrostatic forces, the second distance being less than the first distance. At the second distance, the coupling efficiency between the bus waveguide 608 and the optical antenna 616 via the coupling waveguide 614 may be greater than 50% (e.g., turned on), for example.In other words, FIG. 6 is a schematic illustration of a single MEMS switch 600 that consists of a movable optical coupler 614 and a fixed optical antenna 616. The optical coupler 614 can be moved sufficiently close to the bus waveguide so that a sufficient amount of light is coupled out of the bus waveguide 608 into the coupler waveguide 614. An example of such an optical coupler may be a tapered taper waveguide, and an electrostatically actuated bending beam (cantilever) may move the optical coupler up and down. The optical antenna 616 is on the same layer plane with the optical coupler, but is fixed. The light from the optical coupler may be radiated to the free space via the optical antenna, and conversely, the light from the free space may be coupled back to the optical coupler via the optical antenna, then further coupled back to the bus waveguide.The MEMS switch includes a substrate 700 such as a silicon substrate with an insulating layer 702 such as silicon dioxide. On top of the insulating layer 702 is a silicon layer 704. The first three layers (700, 702, 704) may be a silicon-on-insulator (SOI) substrate. The next layer may be an oxide 706, such as a low temperature oxide (LTO), used to form anchor points. The top layer may be a silicon layer 708, such as a polysilicon layer, which may be deposited via plasma enhanced chemical vapor deposition (PECVD) or low pressure chemical vapor deposition (LPCVD). In this embodiment, the structures such as bus waveguide 608 and MEMS actuation electrode 610 are in the silicon layer 704, while the MEMS spring 612, coupling cantilever 614, reaction electrode and optical antenna 616 are in the polysilicon layer 708.FIG. 7A is a sectional view of the optically coupled cantilever of the MEMS switch of FIG. 6 in a rest state. In the idle or off state, the cantilever of the polysilicon layer 708 may be at a first distance from the bus waveguide of the silicon layer 704 when a voltage difference between the reaction electrode and the actuation electrode is less than the lower threshold. And the coupling efficiency between the bus waveguide 608 and the optical antenna 616 via the coupling waveguide 614 is less than 1% (e.g., turned off).FIG. 7B is a cross-sectional view of the optically coupled cantilever of the MEMS switch of FIG. 6 in an actuated state. In the on-state, the cantilever of the polysilicon layer 708 may be at a second distance from the bus waveguide of the silicon layer 704 when a voltage difference between the reaction electrode and the actuation electrode is greater than the upper threshold. And the coupling efficiency between the bus waveguide 608 and the optical antenna 616 via the coupling waveguide 614 is greater than 50% (e.g., turned on).FIG. 7C is a cross-sectional view of the fixed optical antenna of the MEMS switch of FIG. 6 in a quiescent state. In the off state, the optical antenna of the polysilicon layer 708 may be substantially at a first distance from the bus waveguide of the silicon layer 704.FIG. 7D is a cross-sectional view of the fixed optical antenna of the MEMS switch of FIG. 6 in an actuated state. In the ON state, the optical antenna of the polysilicon layer 708 is maintained substantially at the first distance from the bus waveguide of the silicon layer 704.FIGS. 7A-D illustrate the cross-section of the optical coupler and the optical antenna in an on (ON) and off (OFF) state, respectively. The advantage of this concept is that fewer design constraints are imposed on the MEMS switch than in the prior art. The operating frequency of the lidar system requires very high switching / moving speeds. This manifests itself in a limitation of the displacement mass, which requires a design compromise in the design of antenna and coupler. In the present disclosure, however, the antenna remains static and therefore offers a much greater freedom of design, for example it can have a greater extent in the plane vertical to the bus waveguide, which is reflected in a favourable, slenderr profile of the emitted light beam. For example, for the configuration in Figure 6, the aperture angle of the optical antenna may be larger, the antenna area may be larger along both the bus waveguide and across the waveguide, so that the output beam is more closely collimated. The narrowed beam is favorable for achieving a better angular resolution in beam steering.In the case of the antenna not being fixed, the area of the antenna is preferably limited to 30 × 30 μm 2 or less, and there is no such limitation with the antenna fixed. Moreover, the optical antenna can have more freedom in puncto periods and aspect ratio, which can increase its radiation efficiency. Since there is no hard limit to the surface area and the mass of the fixed antenna, it can have, for example, a greater extent along the waveguide, i.e. have more periods. The aspect ratio is limited by the mass of the antenna. Without limitation in the case of mass, the aspect ratio can be designed predominantly on the basis of the radiation efficiency and instead of a compromise solution between radiation efficiency and antenna mass. And without a movable antenna, the optical coupler can be made smaller and lighter in weight, whereby in this MEMS switch configuration, a fast switching operation and an optimized quality of the output beam can be achieved.In another embodiment, more than one coupling cantilever may be coupled to each antenna. They can be connected to opposite end sides of the antenna and provide a separate transmit and receive channel. The transmit cantilever would be directed toward the laser source while the receive cantilever would be directed toward the light detector, either using the same waveguide for light coupling or using separate waveguides. Separate transmit and receive channels could be operated independently, thereby allowing additional functionalities. For example, the light could be received from multiple pixels surrounding the current transmit pixel, thereby increasing the overall efficiency of light collection.An array of MEMS switches may be organized into a rectangular pattern (FIG. 8 ), a round pattern, or another pattern in an optical terminal. In combination with an objective, this terminal can operate as an optical transmitter, wherein the transmission beam is directed in different directions (for example FIG. 8 ). FIG. 8 is a perspective view of an optical switch array system 800 that transmits an optical signal. Here, a substrate 802 may be a monolithic chip such as a silicon chip, silicon-on-insulator (SoI) chip, silicon carbide chip, silicon nitride chip, or other monolithic MEMS chip, or the substrate may be a multichip module on a substrate. Each optical antenna 806 is arranged to align with a bus waveguide 810, in this embodiment there are, for example, a first bus waveguide 810a, a second bus waveguide 810b, and a third bus waveguide 810c. In this embodiment, each bus waveguide 810 is coupled to a main bus waveguide via a type I switch 818. This is coupled to a light source 814 (e.g., laser, LED, etc.) such that radiated light from the light source 814 travels along a direction 812 and is distributed to the optical antennas 806 via the type I switches 818. Positioned near the chip is an objective 804 configured to generate a collimated beam of light 816 from the optical antennas 806 via the lens 804. In this embodiment, the light beam 816 may be collimated by translating the lens 804 along an axis 808 perpendicular to the substrate 802 and the optical antennas 806, or by translating the substrate 802 and the optical antennas 806 along an axis 809 perpendicular to the lens 804.Such a terminal may also operate as an optical receiver which can receive the beam returning from the various directions and feed the beam back to the PIC (Fig. 9). FIG. 9 is a perspective view of an optical switch array system 900 receiving an optical signal. Here, a substrate 902 may be a monolithic chip such as a silicon chip, silicon-on-insulator (SoI) chip, silicon carbide chip, silicon nitride chip, or other monolithic MEMS chip, or the substrate may be a multichip module on a substrate. Each optical antenna 906 is aligned to align with a bus waveguide 910, in this embodiment, there are, for example, a first bus waveguide 910a, a second bus waveguide 910b, and a third bus waveguide 910c. In this embodiment, each bus waveguide 910 is coupled to a main bus waveguide via a type I switch 918. This is coupled to a light detector 924 (e.g., photodiode, phototransistor, CCD, etc.) such that light is collected from the light detector 924 as the light travels along a direction 922 and collected via the type I switches 918 that have detected via the optical antennas 906. Positioned near the chip is an objective 904 configured to collect a beam of light 920 to optical antennas 906 via the lens 904. In this embodiment, the light beam 920 may be focused by translating the lens 904 along an axis 908 perpendicular to the substrate 902 and the optical antennas 906, or by translating the substrate 902 and the optical antennas 906 along an axis 909 perpendicular to the lens 904.Such a terminal may also operate as a transceiver that can both transmit and receive a beam (FIG. 10 ). FIG. 10 is a perspective view of an optical switch array system 1000 that receives and transmits an optical signal. Here, a substrate 1002 may be a monolithic chip such as a silicon chip, silicon-on-insulator (SoI) chip, silicon carbide chip, silicon nitride chip, or other monolithic MEMS chip, or the substrate may be a multichip module on a substrate. Each optical antenna 1006 is aligned to align with a bus waveguide 1010, in this embodiment, there are, for example, a first bus waveguide 1010 a, a second bus waveguide 1010 b, and a third bus waveguide 1010 c. In this embodiment, each bus waveguide 1010 is coupled to a main bus waveguide via a type I switch 1018. This is coupled to a light detector 1024 (e.g., photodiode, phototransistor, CCD, etc.) such that light is collected from the light detector 1024 as the light travels along a direction 1022 and collected via the type I switches 1018 that have detected via the optical antennas 1006. Positioned near the chip is an objective 1004 configured to capture a focused beam of light 1020 from the optical antennas 1006 via the lens 1004. In this embodiment, the light beam 1020 may be focused by translating the lens 1004 along an axis 1008 perpendicular to the substrate 1002 and the optical antennas 1006, or by translating the substrate 1002 and the optical antennas 1006 along an axis 1009 perpendicular to the lens 1004. And there is coupling to a light source 1014 (e.g., laser, LED, etc.) such that light emitted from the light source 1014 travels along a direction 1012 and is distributed to the optical antennas 1006 via the type I switches 1018. Positioned near the chip is an objective 1004 configured to generate a collimated beam of light 1016 from the optical antennas 1006 via the lens 1004. In this embodiment, the light beam 1016 may be collimated by translating the lens 1004 along an axis 1008 perpendicular to the substrate 1002 and the optical antennas 1006, or by translating the substrate 1002 and the optical antennas 1006 along an axis 1009 perpendicular to the lens 1004. If such a terminal functions only as a transmitter, an independent optical photodetector or photodetector array or a receiver such as the receiver shown in Fig. 9 may be used as the receiver. Likewise, when such a terminal is used only as a receiver, another optical radiator, a radiator array, or a transmitter similar to the illustration in FIG. 8 may be used as the independent transmitter.Although not limited to the presented embodiments, in the above-mentioned design embodiments, the lens and / or the MEMS switch array chip may be integrated into the mechanical structure so that one or both of the elements may move along the z-axis, as shown in FIGS. 8-10. The advantage of having this freedom is that it is possible to adjust the distance between the lens and the MEMS switch array chip to maximize the transmit power and receive power from / into each pixel.The proposed system may be used in a chip lidar system that operates with time-of-flight (ToF) and frequency modulated continuous wave (FMCW) detection. In a chip lidar system, light is coupled onto a waveguide of the PIC and then distributed into sub-waveguides. The proposed MEMS switch may be combined with other types of binary switches (FIG. 11 ) or optical splitter trees (FIG. 12 ) or switches and optical splitter trees (FIG. 13 ) for light distribution. FIG. 11 is a perspective view of an optical switch array system 1100 including at least two types of optical switches. Here, a substrate 1102 is illustrated with a cantilever coupling switch coupled to an optical antenna 1106 configured to align with a bus waveguide 1110, in this embodiment there is, for example, a first bus waveguide 1110 a, a second bus waveguide 1110 b, and a third bus waveguide 1110 c. In this embodiment, each bus waveguide 1110 is coupled to a main bus waveguide via a type I switch 1118.Fig. 11 shows an arrangement plan with two kinds of binary switches. Taking the transmitting terminal as an example, light propagates in the main waveguide. At the intersection of the main waveguide and the row waveguide, the type I switch is located to selectively guide the light into the selected row (in this example, the row of the waveguide 1110 bis selected). And then, the light propagates in the selected row waveguide until it reaches and radiates the MEMS switch (switch type II) which is in the ON state. The type I switch may be either a MEMS switch or other switches such as a thermo-optic switch, electro-optic switch, etc. FIG. 12 shows an arrangement plan with divider trees and binary MEMS switches. FIG. 12 is a perspective view of an optical switch array system 1200 including a splitter tree and an optical switch. Here, a substrate 1202 is shown with a cantilever coupling switch coupled to an optical antenna 1206 arranged to align with a bus waveguide 1210, in this embodiment there is for example a first bus waveguide 1210a, a second bus waveguide 1210b, a third bus waveguide 1210c and a fourth bus waveguide 1210d. In this embodiment, each bus waveguide 1210 is coupled to a main bus waveguide via a splitter tree 1218. Figure 12 also shows a cantilever coupling and optical antenna 1206a in a turned off state and a cantilever coupling and optical antenna 1206b in a turned on state, such that light 1216 is radiated from the optical antenna 1206b via the coupling cantilever coupled to the bus waveguide 1210c when the coupling cantilever and optical antenna 1206b are turned on. The difference between these two arrangements is that in the split tree arrangement the optical power from the laser is distributed evenly to the waveguides, while in the combined binary switch arrangement the light from the laser is selectively guided to a desired waveguide.These two arrangements can be combined into a third arrangement, as shown in Fig. 13. FIG. 13 is a perspective view of an optical switch array system 1300 including a divider tree and at least two types of optical switches. Here, a substrate 1302 is illustrated with a cantilever coupling switch coupled to an optical antenna 1306 configured to align with a bus waveguide 1310, in this embodiment, for example, there are a first bus waveguide 1310 a, a second bus waveguide 1310 b, and a third bus waveguide 1310 c. In this embodiment, each bus waveguide 1310 is coupled to a main bus waveguide via a splitter tree 1318 and a type I switch 1328. Fig. 13 also shows a cantilever coupling and optical antenna 1306a in a turned-off state and a cantilever coupling and optical antenna 1306b, 1306c, 1306d, and 1306e in a turned-on state, so that light is radiated from the optical antenna 1306b, 1306c, 1306d, and 1306e via the coupling cantilever coupled to the bus waveguide 1310b, 1310e, 1310h, and 1310j when the coupling cantilever and the optical antenna 1306b, 1306c, 1306d, and 1306e are turned on. In this arrangement, light from the laser is distributed to multiple portions of waveguide sub-arrays, then binary switches are used in each waveguide sub-array to selectively guide the light to desired waveguides. With this arrangement, the MEMS switches in different sub-arrays can be controlled independently and simultaneously. All of these arrangements can function in MEMS switch array-based transmit, receive and transmit / receive terminals.An advantage of the MEMS switch array is the relatively simple control electronics. FIGS. 14 and 15 show two examples of the electronic controllers for electrostatically actuating the MEMS switch array.FIG. 14 is a view of a switch array system 1400 configured to independently drive rows and columns. This system 1400 drives columns via a column contact controller 1404 and drives rows via a row contact controller 1406. In this illustration, column contact controller 1404c is activated thereby turning on selected or all switches associated with these columns, and row contact controller 1406.3 is activated thereby turning on the waveguide associated with this switch. The result includes turning on the optical switch 1408 to enable emission of light 1410 from a single optical switch of the array. FIG. 15 is a view of a switch array system 1500 configured to simultaneously drive sub-arrays. This system 1500 drives columns via a column contact controller 1504 and drives rows via a row contact controller 1506. In this figure, column contact controller 1504b is activated thereby turning on selected or all switches associated with that column and row contact controllers 1506.11, 1506.8, 1506.5, and 1506.3 are activated thereby turning on the waveguide associated with those switches. The result includes turning on the optical switches 1512a, 1512b, 1512c, and 1512d and allowing light 1514a, 1514b, 1514c, and 1514d to radiate from a single optical switch of the array.In other words, one way of operating the MEMS switches in the array is, of course, to have each switch individually driven, that is, when MxN switches are in the array, where M is the number of rows and N is the number of columns, MxN controllers are needed. An example method approach to simplify control is to drive rows and columns (as shown in FIG. 14 ) so that MxN switches need only M+N controls. In Figure 14, the illuminating switchable pixel is activated by applying the proper amount of voltage to row 1406 and column 1404. Another example approach is to split the MEMS switch array into sub-arrays and drive the switch simultaneously into multiple sub-arrays, as shown in FIG. 15. In FIG. 15, four switchable pixels are illuminated simultaneously and are activated by applying appropriate voltages to column-1504, row-3, row-5, row-8, and row-11 to actuate the corresponding MEMS switches. And the sub-arrays can of course also be controlled individually. All of these electronic control methods can function in MEMS switch array-based transceiver terminals. The advantages of the control method shown in FIGS. 14 and 15 lie in the simplicity. The number of controls is substantially reduced compared to the individually controlled switches.FIG. 16 is a perspective view of a bus waveguide and a corresponding coupler waveguide at an optical switch 1600. A substrate 1602 supports a bus waveguide 1604 with bus waveguide 1604 aligned with a cantilever coupler 1610 separated from bus waveguide 1604 by a vertical gap 1606. The cantilever coupler 1610 can be tapered (taper) to have a narrow tip 1608 and a wider base 1612. For example, the tip 1608 may be a point tip, a rounded tip, or a blunt end, wherein a ratio of the tip 1608 to the base 1612 is 1:3, 1:4, 1.5, etc. For example, a peak width 1608 may be 0.08, 0.1, 0.15, 0.2 μm, while the base 1612 may be 0.2, 0.3, 0.4, 0.5, etc. The length of the cantilever coupler 1614 is designed such that the cantilever coupler can be deflected to reduce the gap 1606 such that a coupling efficiency between the waveguide 1604 and the cantilever coupler 1610 exceeds a threshold value such as 50%, 60% or higher. It should be noted that on the substrate 1602 parallel to the bus waveguide 1604 there are typically two actuation electrodes, one electrode on each side of the bus waveguide 1604, the length of the electrodes being approximately equal to the length of the cantilever coupler 1610. Similarly, on the cantilever coupler 1610, substantially parallel to the bus waveguide 1604, typically two reaction electrodes, one electrode on each side of the cantilever coupler waveguide, are located in the contour of the narrow tip 1608 and wide base 1612. The length of the reaction electrodes is approximately equal to the length of the cantilever coupler 1610.FIG. 17 is a graphical representation of a coupling field profile 1700 with respect to distance. A field profile is shown when a coupling cantilever is switched on and energy is transferred from the bus waveguide to the cantilever waveguide. This illustration is in connection with the cantilever coupler of Figure 16 having a length of 7.5 μm and shows the energy transfer along the length of the cantilever coupler. Similarly, FIG. 18 is a series of two-dimensional plots of coupling field profiles 1800 with respect to the propagation distance. A field profile is shown when a coupling cantilever is turned on and power is transferred from a bus waveguide. This illustration is in connection with the cantilever coupler of Figure 16 having a length of 7.5 μm and shows the energy transfer along the length of the cantilever coupler.FIG. 19 is a graphical representation 1900 of coupling efficiency 1902 with respect to the length 1904 of the coupling taper. Coupling efficiency 1906 is shown when a coupling cantilever having different taper lengths is turned on and power is transferred from a bus waveguide. This illustration is in connection with the cantilever coupler of Figure 16 having different taper lengths and shows the energy transfer along the length of the cantilever coupler. FIG. 20 is a graphical representation 2000 of transmission loss 2002 with respect to wavelength 2004. Shown is a transmission loss 2002 when a coupling cantilever is turned on and energy is transferred from a waveguide. This illustration is in connection with the cantilever coupler of Figure 16 having a length of 7.5 μm and shows transmission loss 2002 along the length of the cantilever coupler. As stated in this disclosure, the cantilever coupler is not limited to silicon, but in this example, the cantilever coupler is a silicon cantilever coupler.FIG. 21 is a graph 2100 of energy loss 2102 with respect to gap dimension 2104. In this illustration, the energy to bus 2106 is at a minimum at about 180 nm, at which point the energy to MEMS 2108 begins to deviate and drop. Based on this, during the "ON" state, the gap is maintained substantially at 180 nm to ensure maximum coupling efficiency to the MEMS waveguide. In an "OFF" state, the gap returns to a distance above a distance of, for example, 650 nm, so that a resulting energy to the MEMS is less than -30 dB. FIG. 22 is a graphical representation 2200 of a radiation pattern 2206 with respect to an angle β 2202 (angle 562 as shown in FIG. 5B ) and an angle θ 2204 (angle 560 as shown in FIG. 5B ). In Figure 22, the energy intensity is normalized to a 1 by gray scale and the contour lines are shown as radiation patterns 2206.FIG. 23 is a side view illustrating an optical system 2300 including a waveguide grating 2302 in a waveguide 2304. The waveguide may be a material such as silicon, polysilicon, silicon nitride, silicon carbide, or another material that can be configured to conduct energy (e.g., a waveguide). The grating 2302 has gap spacings 2310, which may vary from 100 nm to 500 nm or more, with a pitch 2312, which may be a factor of 1.5, 2.0 or 2.5 times gap spacing 2310. The gap spacing may vary across a plane perpendicular to the underlying waveguide. The performance of the waveguide can be evaluated along a plane 2314 substantially parallel to the substrate and a distance 2316 above the substrate. The energy is substantially diffracted into free space at an angle of incidence θ 2318 to the plane 2314. FIG. 24 is a perspective view of a beam spring 2400 showing displacement due to an electric bias between a reaction electrode and an operation electrode. A voltage difference of 40 V between the reaction electrode and the actuation electrode represents, for example, a displacement of substantially 620 nm. FIG. 25 is a graphical representation 2500 of displacement along a Z-axis 2502 with respect to the shift speed 2504. This graph shows the displacement along the Z-axis of Fig. 24 at different positions of the cantilever waveguide. In this example, three profiles are shown, a displacement near the tip of the cantilever waveguide 2506, a displacement in the middle toward the cantilever waveguide 2508, and a displacement near the end where it is connected to the optical antenna 2510.FIG. 26 is a top view of a switch array assembly 2600 having multiple optical switches on a substrate 2602 and a detailed representation of a single element 2604. The optical switch includes a waveguide 2606 coupled to the substrate 2602, to which substrate is also coupled an anchor point 2608. A MEMS spring 2610 coupled to one anchor point 2608 is shown, while an optical grating 2612 is coupled to another anchor point 2608. The MEMS spring 2610 includes a first end coupled to the anchor point 2608 and a second end coupled to a cantilever coupler 2614 suspended above the bus waveguide 2606. Mounted on the substrate 2602 on each of the two sides of the bus waveguide 2606 is an actuation electrode, and mounted on each of the two sides of the cantilever coupler 2614 is a reaction electrode configured to cooperate with the actuation electrode. In this figure, the optical grating / antenna 2612 is on the same layer as the cantilever coupler 2614. In an optical switch array system for beam steering applications, one embodiment includes a system in which each optical switch consists of a movable MEMS optical coupler on a suspension layer and an optical antenna on the substrate layer. The switch array activated by the MEMS optical coupler may be used to launch light from the waveguide on the photonic integrated circuit (PIC) chip into free space (or vice versa) in various applications including lidar.FIG. 27A is a perspective view of a switch array 2700 having a movable optical coupler 2714 on a suspension layer 2716 and a fixed optical antenna 2712 on a substrate layer 2702. FIG. 27B is a top view of the switch array 2700 with a movable optical coupler on a suspension layer and a fixed optical antenna on a substrate layer. The optical switch includes a waveguide 2706 coupled to the substrate 2702, to which substrate is also coupled an anchor point 2708. MEMS spring 2710 coupled to anchor point 2708 is shown. The MEMS spring 2710 includes a first end coupled to the anchor point 2708, and a second end coupled via the suspension layer 2716 to a cantilever coupler 2714 suspended above the bus waveguide 2706. Mounted on the substrate 2702 on each of the two sides of the bus waveguide 2706 is an actuation electrode 2718, and mounted on each of the two sides of the cantilever coupler 2714 is a reaction electrode 2716 configured to cooperate with the actuation electrode 2718. In this figure, the optical grating / antenna 2712 is on the same layer as the bus waveguide 2706.The MEMS switch array may be activated by movable MEMS optical couplers 2714 on a suspension layer 2716 and fixed optical antennas 2712 on the substrate layer 2702. FIGS. 27A and 27B show a perspective view and a plan view, respectively, of such a MEMS switch. Each MEMS switch consists of a movable MEMS optical coupler 2714 on the suspension layer 2716 (called "MEMS layer") and an optical antenna 2712 on the substrate layer 2702. Here, an example of a grid antenna 2712 is used to illustrate the concept, other embodiments of the optical antenna also work for this concept. In an OFF state, the MEMS layer, reaction electrode 2716, and cantilever coupler 2714 are far from bus waveguide 2706 and actuation electrode 2718, so that light can propagate undisturbed in bus waveguide 2706. In an ON state, the reaction electrode 2716 is actuated and moves the MEMS optical coupler 2714 close to the bus waveguide 2706, so that light is coupled into the coupler waveguide 2714 on the MEMS layer with the reaction electrode 2716, a second cantilever coupler couples out to a remote bus waveguide connected to the optical antenna 2712 that sends the signal into free space. This path of light is reversed when a signal is received.FIG. 28A is a cross-sectional view of the movable optical coupler of FIG. 27 in the OFF position. In an OFF state, the MEMS layer (including the reaction electrode) 2716 is a first distance above the top edge of the bus waveguide 2706. The spacing is caused by a voltage difference between the actuation electrode 2718 and the reaction electrode 2716 being below a voltage threshold such that the cantilever coupler 2714 is far from the bus waveguide 2706. The first spacing is designed such that light propagates undisturbed in bus waveguide 2706.FIG. 28B is a cross-sectional view of the movable optical coupler of FIG. 27 in the ON position. In an ON state, the MEMS layer (including the reaction electrode) 2716 is at a second distance above the top edge of the bus waveguide 2706. The second distance is established by a voltage difference between the actuation electrode 2718 and the reaction electrode 2716 being above the voltage threshold value, so that the cantilever coupler 2714 is pulled toward the bus waveguide 2706. At the second distance, light is coupled into coupler waveguide 2714 on the MEMS layer with reaction electrode 2716 and a second cantilever coupler that couples out to a remote bus waveguide connected to optical antenna 2712 that sends the signal to free space. This path of light is reversed when a signal is received.Figs. 28A and 28B show the operation of the optical coupler in cross section. The advantage of this concept is that the optical antenna is on the substrate layer, which does not add additional weight to the suspension structure and presents less manufacturing problems compared to other structures. The operating frequency of a lidar system requires very high switching / moving speeds. This results in a limitation of the displacement mass. However, in the present disclosure, the optical antenna remains on the substrate layer so that its dimension has no influence on the operation speed. As a result, the antenna may have a larger footprint to form a favorable Gaussian beam profile and radiation efficiency may be optimized by flexible grid periods and / or duty cycle. The coupler waveguide on the suspension layer is small and light, which also promising a high switching speed. The optical coupler may be operated electrostatically, piezoelectrically, or by other mechanisms. It may be a two-sided bending beam (double cantilever) in which the middle part is anchored so that both ends can be operated together or both ends can be operated independently. It may also be a beam in which the entire coupler waveguide moves during operation (as shown in Figure 27B).The concepts described herein may be implemented in alternative embodiments. Another embodiment is shown in Figures 29 and 30. FIG. 29 is a top view of a switch array assembly 2900 having a movable optical coupler 2914 on a suspension layer and a fixed optical antenna 2912 on a substrate layer 2902 showing a single element in detail. The optical switch includes a waveguide 2906 coupled to the substrate 2902, to which substrate is also coupled an anchor point 2908. MEMS spring 2910 coupled to anchor point 2908 is shown. The MEMS spring 2910 includes a first end coupled to the anchor point 2908 and a second end coupled via the suspension layer 2916 to a cantilever coupler 2914 suspended above the bus waveguide 2906. Mounted on the substrate 2902 on each of the two sides of the bus waveguide 2906 is an actuation electrode 2918, and mounted on each of the two sides of the cantilever coupler 2914 is a reaction electrode configured to cooperate with the actuation electrode 2918. In this figure, the optical grating / antenna 2912 lies on the same layer as the bus waveguide 2906.Figure 30 is a top view of a switch array assembly 3000 having a movable optical coupler on a suspension layer and a fixed optical antenna on a substrate layer, showing a single element in detail. The optical switch includes a waveguide 3006 coupled to the substrate 3002, with the substrate also having an anchor 3008 coupled thereto. MEMS spring 3010 coupled to anchor point 3008 is shown. The MEMS spring 3010 includes a first end coupled to the anchor point 3008, and a second end coupled via the suspension layer 3016 to a cantilever coupler 3014 suspended above the bus waveguide 3006. Mounted on the substrate 3002 on each of the two sides of the bus waveguide 3006 is an actuation electrode 3018, and mounted on each of the two sides of the cantilever coupler 3014 is a reaction electrode configured to cooperate with the actuation electrode 3018. In this figure, the optical grating / antenna 3012 lies on the same layer as the bus waveguide 3006.The arrangement in Figure 29 consists of an array of rectilinear waveguides with an array of optical antennas 2912 spaced apart from the waveguides 2906. In this configuration, coupler waveguide 2914 curves to receive light from bus waveguide 2906 and couple it out to antenna 2912. Alternatively, the arrangement in Figure 30 is comprised of an array of curved bus waveguides 3006 having an array of optical antennas 3012 aligned with the rectilinear portions of the bus waveguide 3006. In this configuration, the coupler waveguide is rectilinear.FIG. 31 is a top view of an element of a switch array assembly 3100 including an optical coupler 3126 on a suspension layer 3124 and a fixed optical antenna 3104 on a substrate layer. The optical antenna 3104 covers an angle 3106 such as 45 degrees, 60 degrees or 90 degrees and has a base or taper 3108 after which the grating begins. The optical coupler 3126 may be a single element or may be comprised of multiple segments. Here, the optical coupler 3126 has three segments, the first segment 3112 is the segment that, when activated, is pulled down to couple to the optical antenna 3104. The second segment 3114 connects the first segment 3112 to the third segment 3116. The third segment 3116 is the segment that moves downward to couple to the bus waveguide 3102. The second segment 3114 may be stationary and directly coupled to an anchor point such that the distance from the second segment 3114 to the substrate does not change, or in another embodiment, the second segment 3114 is suspended and moves with the movement of the first 3112 and third segments 3116. The bus waveguide 3102 may have a width 3118 at a point where the tip of the optical coupler 3126 will couple upon activation, and this width may narrow to a width 3120 before a radius curvature 3122, such that the curvature is configured to be substantially adiabatic. Transmission through the optical coupler 3126 may include a first loss 3128 via the third segment 3116, a second loss 3130 via the second segment 3114, and a third loss 3132 via the first segment 3112. These losses may be a fraction of a dB, the first 3128 and third 3132 may be, for example, 0.1 dB, while the loss in the second segment 3130 may be less than 0.02 dB. FIG. 32 is a cross-sectional view of element 3200 of FIG. 31 Here, optical coupler 3126 is coupled to suspension layer 3124 and aligned above bus waveguide 3102. The bus optical waveguide 3102 has a height 3210, and the suspension layer 3124 overlies a distance 3212 above the bus waveguide 3102, this distance varies with the state, for example, in the on state, the distance 3212 may be less than 0.2 μm, and in the off state, the distance 3212 may be greater than 0.6 μm. The suspension layer 3124 may have a thickness 3214 such that it may provide support, but still provide optical coupling between the bus waveguide 3102 and the optical coupler 3126. The optical coupler 3126 may have a thickness of 3216.Figs. 31 and 32 show an example of a single MEMS optical coupler and a fixed grating antenna in the arrangement of Fig. 30. the dimensions indicated in Fig. 31 are an example, Figs. 17 and 18 show the coupling efficiency with the MEMS optical coupler at a distance of 180 nm from the bus waveguide, and Fig. 33 shows the electric field distribution at the grating. FIG. 33 is a graphical representation of a field profile 3300 of a grid antenna with respect to the X coordinates 3304 and Y coordinates 3302. FIG. 34 is a perspective view of an illustrative displacement of an optical coupler 3400 such as the optical coupler of FIG. 31, FIG. 34 is a simulation of the displacement of a MEMS optical coupler. Depending on different designs, the optical coupling length generally varies from a few microns to hundreds of microns, and the distance between the coupler waveguide and the bus waveguide / antenna waveguide is a few nanometers to a few microns in the ON state.Some optical antennas have a non-zero radiation angle θ as shown in FIGS. 35 and 36, which may become a problem in the amplitude array terminal. FIG. 35 is a perspective view of a grid antenna 3500 of a MEMS structure 3502 and shows the radiation angles β 3504 and θ 3506. FIG. 36 is a graph of intensity 3600 with respect to the radiation angles β 3602 and θ 3604. These figures show 10 periods, illustrating intensity over a gray scale.One solution to this problem is to sacrifice optical antenna radiation efficiency and far field beam quality to obtain a zero radiation angle. The second solution is to add an additional structure for forcing vertical light emission, such as in particular a bottom reflector, an additional upper dielectric layer, an extra fine etching groove. The third solution consists in designing the objective correspondingly so that the radiation angle is compensated for by not zero. However, due to manufacturing variations, this approach is not flexible enough to compensate for different radiation angles. Here, we propose using a mirror or prism between the antenna array and the objective to compensate for the non-zero radiation angle, as shown in FIGS. 39, 40, 41. FIG. 39 is a side view 3900 illustrating a mirror for non-zero antenna radiation angle compensation. Here, a telecentric objective 3902 is used to focus (transmit) the light from the antenna array or photonic integrated circuit via a mirror toward the free space or reflect (receive) from a mirror 3904 toward an antenna array or photonic integrated circuit (PIC) 3910. Here, a virtual image 3906 of the PIC 3910 is displayed on a virtual plane 3908. FIG. 40 is a side view illustrating a prism optical system 4000 for non-zero antenna radiation angle compensation. Here, a telecentric objective 4002 is used to focus (transmit) the light from the antenna array toward the free space via a prism or focus (receive) toward an antenna array 4004 (PIC) via a prism. FIG. 41 is a side view 4100 illustrating a micro prism array for non-zero antenna radiation angle compensation. Here, a telecentric objective 4102 is used to focus (transmit) the light from an antenna array via micro-prisms toward the free space or focus (receive) light via micro-prisms toward an antenna array 4104 (PIC).The mirror or prism may diffract the light so that the reflected or refracted rays are incident on the lens system at an incidence angle of zero. Since the mirror or prism are rotatable, they can be adjusted in the calibration phase in order to compensate for the manufacturing errors. At the same time, a telecentric objective can be used to fit the tilted radiation array. The advantage of this method is that it does not require a complicated design of the objective.The proposed system may be used in a chip lidar system that operates with time-of-flight (ToF) and frequency modulated continuous wave (FMCW) detection. In a chip lidar system, light is coupled onto a waveguide of the PIC and then distributed into sub-waveguides. FIG. 42A is a top view of an antenna array 4200 including splitter trees 4202 and optical switches 4204. FIG. 42B is a plan view of an antenna array 4230 including multiple types of switches, in the present example, there are type I switches 4206, and optical switches 4204. FIG. 42C is a top view of an antenna array 4360 having divider trees 4202 and multiple types of switches such as type I switches 4206 and optical switches 4204. The proposed MEMS switch may be combined with divider trees (FIG. 42A ) or other types of binary switches (FIG. 42B ) or switches and divider trees (FIG. 42C ) for light distribution. Fig. 42B shows an arrangement plan with two kinds of binary switches. As can be seen by way of example with reference to the transmission terminal, light propagates in the main waveguide. At the intersection of the main waveguide and the row waveguide, the type I switch can selectively guide the light into the selected row (in the drawing, the second row is selected from above). Then, the light propagates in the selected row waveguide until it reaches and radiates the turned-on MEMS switch (switch type II). The type I switch may be either a MEMS switch or other switches such as a thermo-optic switch, electro-optic switch, etc. Fig. 42A shows an arrangement plan with division trees. The difference between these two arrangements is that in the split tree arrangement the optical power from the laser is evenly distributed into the waveguides, while in the combined binary switch arrangement the light from the laser is selectively guided to an optical waveguide at a time. These two arrangements can be combined into a third arrangement as shown in Fig. 42C. In this arrangement, light from the laser is distributed into multiple sections of waveguide sub-arrays, then binary switches are used in each waveguide sub-array to selectively guide the light to desired waveguides. With this arrangement, the MEMS switches in different sub-arrays can be controlled independently and simultaneously. All of these arrangements may function in the MEMS switch array-based transmit / receive terminals. In the arrangement of Figures 42B and 42C, the active row of the type I switch is selected, while the active column is selected by the type II switches. As a result, the control complexity for an array M×N is linearly scaled (O(M+N)) with the array size rather than quadratic (O(M×N)).This section discloses MEMS switches and arrays that move in-plane for optical coupling. The proposed MEMS switch may be used to couple light from waveguides on a photonic integrated circuit (PIC) chip to free space in various applications.FIG. 43A is a perspective view of a switch array having a movable grid traveling longitudinally with multiple holding positions. In this embodiment, the grating 4308 travels longitudinally such that there are multiple stop positions, this illustration shows 3 stop positions, 4308a, 4308b and 4308c. In this embodiment, 4308 amay be the neutral position such that a voltage applied to the interdigitated electrodes will move the grid to the other positions 4308 band 4308 c. The grid may also be configured such that the neutral position is in the middle 4308 bsuch that a positive voltage across the interdigitated electrodes moves the grid to one position (e.g., 4308 a) and a negative voltage across the interdigitated electrodes moves the grid to the other position (e.g., 4308 c). In this figure, the light travels along the waveguide 4304 and to three bus waveguides 4306a, 4306b, and 4306c. As described above, the light may travel via a beam splitter (splitter) or an optical switch, and the light may also travel in the opposite direction of 4304. FIG. 43B is a detailed perspective view of a movable grating 4314 traveling longitudinally 4312 with multiple stop positions to bus waveguide 4316 of FIG. 43A. The bus waveguide 4316 is supported by the substrate. The substrate supports an armature 4318 coupled to the interdigitated electrodes 4320.FIG. 43C is a perspective view of a switch array having a movable grid 4328 traveling longitudinally 4322 to the bus waveguide 4326. This illustration shows 2 holding positions, 4328a and 4328b. In this embodiment, 4328 amay be the neutral position such that a voltage applied to the interdigitated electrodes will move the grid to the other position 4328 b. In this figure, the light travels along waveguide 4324 and to three bus waveguides 4326a, 4326b, and 4326c. As described above, the light may travel via a beam splitter (splitter) or an optical switch, and the light may also travel in the opposite direction from 4324. FIG. 43D is a detailed perspective view of two movable gratings 4342 moving longitudinally toward the bus waveguide 4316 of FIG. 43C. The bus waveguide 4316 is supported by the substrate. The substrate supports an armature 4338 coupled to the interdigitated electrodes 4334. FIG. 43E is a top view of a movable grating 4342 moving to the bus waveguide 4336 via interdigitated electrodes 4334. The bus waveguide 4336 is supported by the substrate. The substrate supports an armature 4338 coupled to the interdigitated electrodes 4334. A spring 4340 is coupled to the grid 4342 on one side and to an armature 4338.FIG. 44A is a perspective view of a switch array having a movable grid 4808 moving in the transverse direction 4402 to the bus waveguide 4406 having multiple holding positions. In this embodiment, the grid 4408 travels in the transverse direction such that there are multiple stop positions, this figure shows 3 stop positions, 4408a, 4408b and 4408c. In this embodiment, 4408 amay be the neutral position such that a voltage applied to the interdigitated electrodes will move the grid to the other positions 4408 band 4408 c. The grid may also be configured such that the neutral position is at the center 4408 bsuch that a positive voltage across the interdigitated electrodes moves the grid to one position (e.g., 4408 a) and a negative voltage across the interdigitated electrodes moves the grid to the other position (e.g., 4408 c). In this figure, the light travels along the waveguide 4404 and to three bus waveguides 4406a, 4406b, and 4406c. As described above, the light may travel via a beam splitter (splitter) or an optical switch, and the light may also travel in the opposite direction of 4404. FIG. 44B is a detailed perspective view of a movable grating 4808 moving in the transverse direction 4402 to the bus waveguide 4406 with multiple holding positions of the switch array of FIG. 44A. Bus waveguide 4406 is supported by the substrate. The substrate supports an armature 4418 that is coupled to the interdigitated electrodes 4420. FIG. 44C is a top view of a movable grid 4442 moving across interdigitated electrodes 4434 in a transverse direction to bus waveguide 4436. In this figure, a spring 4440 is coupled to an armature 4438 that is coupled to the substrate. The other end of the spring 4440 is coupled to the grid 4442 such that a force applied to the interdigitated electrodes 4434 moves the grid 4442 to a position above a bus waveguide 4436 as described above.FIG. 45 is a perspective view of a switch array with a movable grating traveling in the transverse direction 4502 with stop positions in and out of the plane of waveguides 4506. In this embodiment, light travels through the bus waveguide in a direction 4504 and is distributed to three bus waveguides 4506a, 4506b, and 4506c. In this embodiment, out-of-plane gratings are shown as 4510 and are not aligned with any bus waveguide 4506. In-plane gratings are shown as 4508 and aligned with a bus waveguide 4506 so that there is optical coupling between the grating 4508 and the bus waveguide 4506b.FIGS. 43, 44 and 45 show three examples of in-plane movable MEMS switches in the amplitude array lidar application. In all of these three figures, grid antennas are used as examples for illustrating the switch concepts; other antenna configurations would also work for these switch concepts. In all of these three configurations, the MEMS switch layer is suspended above the waveguide layer. The gap distance between the MEMS switch layer and the bus waveguide layer is a few nanometers to a few hundred nanometers, so that the coupling efficiency between the MEMS switch and bus waveguide is sufficient to couple more than 50% of the light from the bus waveguide layer to the MEMS switch layer. In FIG. 43, the MEMS switches may move along the waveguide, and the pixels are virtual pixels defined by the holding position of the MEMS switch. As soon as a MEMS switch stops, the light is decoupled from this position and the corresponding virtual pixel is switched on. Depending on the size of the waveguide and the range of motion of the MEMS switch, one (FIGS. 43A and 43B ) or more than one (FIGS. 43C and 43D) MEMS switch may cover a waveguide length. In FIG. 44, the MEMS switches may move across multiple waveguides in the plane along the x-direction. Similar to the layout in FIG. 43, the pixels are again virtual pixels defined by the MEMS switch's hold position. Depending on the dimension of the waveguide array and the range of motion of the MEMS switch, a MEMS switch may cover all of the waveguides in the array or a portion of the waveguides in the array. An advantage of the MEMS switch shown in FIGS. 43 and 44 is that the switches move in the plane, which is easier to implement than the switches that are movable out of the plane. The virtual switch stops also allow switches to stop flexibly at any position along the travel path.FIG. 45 illustrates another planar design of MEMS switches. Unlike the layout in FIG. 44, the pixels here are real pixels defined by the relative overlap of MEMS switches and bus waveguides. In the OFF state, the switches are offset from the waveguides and in the ON state, the switch moves to overlap the waveguide. The advantages are that the travel distance for each switch is so small that the switches can be implemented more easily.The proposed in-plane MEMS switches can also be organized in an array for beam steering in optical transmitting, receiving or transmitting / receiving terminals and combined with objectives. All concepts of the switch combination and divider tree combination and row / column wiring also work with the proposed in-plane MEMS switches.The concepts of a vertically movable MEMS switch with cantilever coupler and fixed optical antenna as presented in the present disclosure require a stop of the cantilever coupler. In the vertically movable MEMS switches (for example FIGS. 4A-4C, 6 ), the stopping of the MEMS switch or optical cantilever at a certain distance above the bus waveguide must be controlled. Mechanical stops (buffers) may be used to mechanically stop the MEMS switches. However, since the suspension layer will approach the mechanical stoppers in each operation, the mechanical stoppers may wear and thereby degrade the light coupling efficiency of the system. Here, as the MEMS switch stopper in the MEMS switch light coupling application, an electrostatic levitation system and method is used.FIG. 46 is a perspective view of an electrostatic levitation switch 4600 for controlling a coupling distance between a bus waveguide and the gratings. The electrostatic levitation switch 4600 includes a substrate 4602 supporting lower lateral or repulsion electrodes 4604, a bus waveguide 4606, lower central or activation electrodes 4608 and anchors 4614. Mounted above bus waveguide 4606 is an optical grating 4610 suspended via a spring 4612 coupled to armature 4614. On top of grid 4610, an upper or reaction electrode 4616 is disposed. The methods for controlling the holding position of vertically movable MEMS switches with non-mechanical stoppers will now be described. A method for utilizing electrostatic levitation is illustrated in FIG. 46 (the concept is exemplified by a switchable MEMS grating). In such a configuration, the MEMS structure is electrostatically actuated to the pull-in state or beyond the pull-in state by the lower central electrodes and the upper electrodes, then the lower central electrodes and the upper electrodes are maintained at the same voltage level (e.g., both at ground), and a high positive voltage is applied to the lower lateral electrodes. The large electric field generated by the lower lateral electrode thereby pushes the MEMS structure upwards and finally keeps the structure at a certain distance above the bus waveguide. The advantage of the electrostatic levitation stopper is that the suspension layer does not physically contact the lower bus waveguide layer, so that during MEMS operation there are no problems with stiction or physical wear. However, this method requires a relatively high stress to keep the suspension layer at a certain position.Another method is to use a comb drive, an example is shown in Figs. 47A and 47B. FIG. 47A is a perspective view of a switch 4700 having a comb drive for controlling a coupling distance between a bus waveguide 4706 and the gratings 4710. Electrostatic comb levitation switch 4700 includes a substrate 4702 supporting lower lateral or activation comb electrodes 4704, a bus waveguide 4706 and anchors 4714. Mounted over bus waveguide 4706 is an optical grating 4710 suspended over a spring 4712 coupled to armature 4714. Coupled to the optical grating 4710 is an upper or reaction comb electrode 4708. FIG. 47B is a cross-sectional view taken along a section of FIG. 47A illustrating the comb drive in which the lower lateral or activation comb electrodes 4704 are interdigitated with the upper or reaction comb electrode 4708 adjacent the bus waveguide 4706. In this embodiment, the blades of the comb drive are perpendicular to the bus waveguide 4706 but the blades may also be arranged parallel to the bus waveguide 4706.Finger electrodes are formed on the movable suspension layer (upper electrodes) and fixed substrate layer (lower electrodes). The electric field direction between upper and lower electrodes may be controlled by the polarity of the voltages applied to the electrodes. When the voltage applied to the upper electrodes is higher than that of the lower electrodes, the electric field has migrated downward and can attract the suspension layer to move toward the bus waveguide. Once it reaches a certain position and the voltage of the upper electrodes is changed to be smaller than that of the lower electrodes, the electric field travels upward and may help overcome gravity and maintain the suspension layer at a certain position. The advantage of the comb drive is that it does not require as high a voltage as electrostatic levitation, and it is also a nonmechanical stop.Various designs of MEMS switches and switch arrays have been disclosed in this disclosure. These various designs will be used in the optical transmission, reception and transmission / reception terminals. Methods for integrating the MEMS switches and switch arrays into a chip lidar system and for packaging (packaging) such a system will now be described.MEMS switch array-based chip lidar systems integrate MEMS structures on a silicon photonic platform. This combination of MEMS structure and silicon photonic platform leads to new challenges that are not present in pure MEMS systems and in pure silicon photonic systems. Therefore, the system integration and the packaging must be performed properly in design and processing. FIG. 48 is a cross-sectional view of a switch array system for illustrating integration and packaging of a MEMS switch array-based chip lidar module 4800.MEMS structures may have many failure mechanisms (e.g., dust, moisture), therefore all the components of the chip lidar must be packaged together and hermetically packaged. A hermetically packaged LIDAR module 4800 includes a heat sink / substrate platform 4802, a light source 4804 (e.g., laser, LED, or other commonly used light source), a light coupling component 4806, a photonic integrated circuit (PIC) chip with MEMS switch and ASIC controller 4808, an objective 4810, an inner anti-reflective coating 4812, an outer anti-reflective coating 4814, one or more photodetector(s) 4816 (e.g., photodiode, phototransistor, charge coupled device, or other commonly used photodetector), and other necessary coupling components. The light source 4804 may be coherent and polarized.An example of system integration and package installation is illustrated in FIG. 48. All components are integrated onto a substrate platform with heat dissipation mechanisms and hermetically packaged. The cover of the package includes a window area that is transparent to the wavelength used in the LIDAR chip so that light can exit the package and return to the PIC chip. The window portion may constitute a part or the entire surface of the lid. Anti-reflective coatings on the inner and outer surfaces of the window may help reduce reflection as it passes through the window. Different parts of the system may be packed at different pressure levels and with different gas compositions. For example, the MEMS switch array / PIC 4808 requires a certain pressure level (e.g., above 1 Torr) to induce attenuation so that MEMS switches may stop at the switch-on position. To achieve this purpose, the chamber is filled with heavy gases such as nitrogen, argon, etc. until the required pressure level is reached. Outside this range (i.e., within the hermetically sealed module 4800), the pressure may have a vacuum level (e.g., below 1 mTorr) to minimize the impact of particles and moisture. The hermetic sealing temperature of 4800 should be lower than the temperature limits that may affect the performance of the other components (e.g., PIC 4808, light source 4804, photodetector 4816) in the system. For example, it is recommended that the sealing temperature remain below 300°C so as not to degrade the performance of the laser. In the integration of an ASIC, die bonding via silicon flow rates can minimize the influence of electrical paths on switching speed and switching power.The present section proposes several designs and concepts of MEMS switches and switch arrays for optical coupling. These proposed structures can be used in optical terminals for optical beam steering. Various design and placement methods for MEMS switches and MEMS switch array are described herein. All of these methods can be used in the construction of optical beam steering terminals.An example of the MEMS switch array design is illustrated in Figure 49. Figure 49 is a perspective view of a dual grating switch 4900 traveling in the transverse direction 4906 with in-plane holding positions with waveguides 4912. The substrate 4902 includes a plurality of type I switches (e.g., 4910a, 4910b) to direct light to bus waveguides 4912a, 4912b, 4912c, and 4912d. In this figure, the dual grating 4914a and 4914b is either aligned so that grating 4914a is aligned with bus waveguide 4912a or so that grating 4914b is aligned with bus waveguide 4912b so that light 4904 is coupled.To illustrate the concept, the gratings are used here as optical couplers / antennas integrated with the MEMS switch. Other optical coupler / antenna designs also work with this concept. In the MEMS switch array shown in Figure 49, each MEMS switch carries two optical antennas and is responsible for two pixels on top of two adjacent bus waveguides. Each MEMS switch can be actuated horizontally and move the two grating couplers / antennas along the x-direction across the two waveguides. The MEMS switches have two positions (4914a and 4914b): a grating overlaps one of the two waveguides (one pixel is ON, position 4914a and position 4914b). For example, to turn on the second pixel above the waveguide 4912b, the series switch 4910b is turned on and the MEMS switch is actuated to move the two grating couplers / antennas to the -x direction, so that one / e grating coupler / antenna may have sufficient overlap with the waveguide 4912b and sufficient light may couple out from the waveguide 4912b to that grating and then radiate into the free space. Depending on which pixel is switched on, the switch is deflected up to a certain stop against a mechanical spring (as shown in the single-switch diagram of FIG. 49). An actuator-type comb drive can be used to actuate the switch movement in the plane. Since the grating is larger than the waveguide, reciprocation has very little effect on performance. The deflection can be electrostatic, inductive, piezoelectric or similar. The distance between the grating layer and the bus waveguide layer should be small enough to ensure that sufficient light couples out from the bus waveguide to the grating when the switch is in either position 4914a or position 4914b. The advantage of this design is that the number of MEMS switches is only half the number of pixels.Another configuration of the MEMS switch array is illustrated in FIG. 50, where the MEMS switches are also actuated to move horizontally along the x-axis and one MEMS switch carries more than one / n optical / n couplers / antenna (here again, a grating is used as coupler / antenna, by way of example, to illustrate the concept).Figure 50 is a perspective view of a dual grating switch 5000 moving transversely 5006 with stop positions in and out of the plane of waveguides 5012. The substrate includes a plurality of type I switches (e.g., 5010a, 5010b) to direct light to bus waveguides 5012a, 5012b, 5012c, and 5012d. In this illustration, the dual grating 5014a and 5014b is either aligned such that the grating 5014a is aligned with a bus waveguide 5012a, the grating 5014a and 5014b is centered between the adjacent bus waveguides 5012a and 5012b so that no light is coupled, or such that the grating 5014b is aligned with the bus waveguide 5012b so that light is coupled. These MEMS switches 5000 have three positions, as shown in FIG. 50: neutral position in which two gratings are arranged between two waveguides and do not overlap either (both pixels are OFF, position 5014), one grating overlaps one waveguide (one pixel is ON, position 1 and position 5014a). Depending on which pixel is switched on, the switch is likewise deflected up to a specific stop against a mechanical spring (as shown in the sketch with the individual switch in FIG. 50). Since the grating is larger than the waveguide, reciprocation has very little effect on performance. The deflection can be electrostatic, inductive, piezoelectric or similar. In this design, however, the jumpers are not necessary because position 5014 is a complete OFF position (neither grid overlaps either bus waveguide). The waveguides may be connected by splitter trees and by setting the switch to position 5014a and 5014b to turn on different pixels. In this case, the MEMS switch is a three-way switch. And, of course, the distance between the grating layer and the bus waveguide layer should be small enough to ensure that sufficient light couples out from the bus waveguide to the grating when the switch is in either position 5014a or position 5014b.Fig. 51 is a perspective view of a bimorph switch 5100 in which a grating 5108 has an in-plane position. Bimorph switch 5100 includes a substrate 5102, and above substrate 5102, a grating 5108 is suspended on a suspension layer 5104 that includes bimorph piezoelectric actuator 5106. When bimorph piezoelectric actuator 5106 is activated 5110, bimorph switch 5100 moves to Fig. 52. Fig. 52 is a perspective view of bimorph switch 5200 which vertically rotates grating 5208 at an out-of-plane position. Bimorph switch 5200 comprises a substrate 5202, and above substrate 5202, a grating 5208 is suspended on a suspension layer 5204 comprising bimorph piezoelectric actuator 5206.FIG. 53 is a perspective view of a switch array 5300 in which grating elements rotate about an axis perpendicular to the waveguide (e.g., rotate about the x-axis). A substrate 5302 supports bus waveguides 5312 so that light moving in a direction 5310 can be distributed to each bus waveguide 5312a 5312b, 5312c, and 5312d. Here gratings 5316, 534 5314 are configured to rotate 5306 about the x-axis and, when active, as grating 5314, light 5308 is directed out of bus waveguide 5312b. Although illustrated as transmitting light, use is also possible for receiving light. Similarly, FIG. 54 is a perspective view of a switch array 5400 in which grating elements rotate about an axis parallel to the waveguide (e.g., rotate about the y-axis). A substrate 5402 supports bus waveguides 5412 so that light moving in a direction 5410 can be distributed to each bus waveguide 5412a 5412b, 5412c, and 5412d. Here gratings 5416, 544, 5414 are configured to rotate 5406 about the y-axis and, when active, as grating 5414, light 5408 is directed out of bus waveguide 5412b. Although illustrated as transmitting light, use for receiving light is also possible.MEMS switches may also be configured to be tiltable or foldable, as illustrated in FIGS. 53 and 54. In the configuration of FIG. 53, the MEMS switches are actuated to rotate in the y-z plane about the x-axis. When the MEMS switch is parallel to the y-axis, there is an overlap between the grating and bus waveguide, and if the distance between the grating and the waveguide is kept small enough, light can couple out from the bus waveguide to the grating and radiate further to the free space, thus turning on the MEMS switch (FIG. 51 ). If the MEMS switch is tilted to a sufficiently large angle so that light coupling is very weak, the MEMS switch is OFF (FIG. 52 ). Similarly, as shown in FIG. 54, the MEMS switches are actuated to rotate in the x-z plane about the y-axis. The MEMS switch may be driven by bimorph piezoelectric actuators as shown in Figures 51 and 52, or by electroactive polymers.The switches can also be rotated in the plane and slide over the waveguide (such as a gyro disc) not only to a 90 degree angle. Similar to Fig. 55, only on a different scale. FIG. 55 is a top view of a grid array 5500 on a turntable 5502 that utilizes a rotation mechanism for tuning in-plane and off-plane operation. Turntable 5502 includes gratings 5504 arranged in an array such that rotation 5508 of the disc causes gratings 5504 to align with bus waveguides 5506 to transmit or receive light. An advantage is that the antennas are controlled by only one motor, rather than each antenna being controlled by a switch. The antennas may be designed with a certain pattern so that polarization is not impaired during rotation.In addition, the MEMS switchable gratings may use a bistable membrane to reduce power consumption during operation. In this case, the MEMS switch actuation may be a combination of two methods to achieve the ON or OFF state. FIG. 56A is a cross-sectional view of a coupler waveguide 5600 with a bistable membrane 5612 in the OFF position. A substrate 5602 supports a bus waveguide 5604, an actuation electrode 5606, and a mechanical stop 5608. On a suspension layer is the bistable membrane 5612 comprising a reaction electrode 5607, a mechanical stop 5608 and a coupler waveguide 5610. FIG. 56B is a cross-sectional view of a coupler waveguide with a bistable membrane 5612 in the ON position. The switch shown in Figure 56A includes the coupler waveguide 5610 (with gratings) on a flexible top membrane 5612. In the OFF state, coupler waveguide 5610 is separated from bus waveguide 5604. To turn the switch to the ON state, the coupler waveguide 5610 must be pulled down to the bus waveguide 5604 in Figure 56B. One method of actuation is electrostatic attractive force applied by electrodes 5606 and 5607, but here constant current flow is necessary to keep the switch on. This would increase power consumption for device operation. In order to avoid the need for a constant current to be applied to maintain the device on-state, a fine adjustment of the diaphragm voltage (stress tuning) may be made so as to produce a bistable diaphragm. To do so, a compression-strained film 5612 would be applied to the top surface of the membrane coupler waveguide 5610. The tension must be adjusted so that the membrane bulges either parallel or quite slightly relative to the underside surface. This membrane has bistable properties, that is it may be flat or slightly curved when no actuation force is applied and when it is pulled downwards with electrostatic actuation it remains downwards even if the current applied by electrode 5606 and 5607 is interrupted. This approach allows the switch to be turned on and pulled down with only a short power pulse, and no current is required during the ON state of the switch, since the switch is independently held in the on state by the bistable membrane. To return the membrane to the separated off state, a pulsed electrostatic repulsive force is used to urge the membrane into the spaced-off mode.These proposed in-plane MEMS switches can also be organized in an array for beam steering in optical transmitting, receiving or transmitting / receiving terminals and combined with objectives. All concepts of the switch combination and divider tree combination and row / column wiring also work with the proposed in-plane MEMS switches.While the entire invention has been illustrated by a description of various embodiments and these embodiments have been described in considerable detail, it is not intended by the applicant to limit the scope of the appended claims in any way to this level of detail. Additional advantages and modifications will be readily apparent to those skilled in the art. Therefore, in its broader aspects, the invention is not limited to the specific details, representative devices and methods illustrated and described. Accordingly, these details may be deviated without departing from the spirit or scope of the general inventive concept.

Claims

An optical switch comprising: a first bus waveguide (4706, 4406a, 608) supported by a substrate (4702, 602, 802, 902, 1002); an optical antenna (4710, 4408, 806, 906, 1006) suspended over the first bus waveguide (4706, 4406a, 608) via a spring (4712); and interdigitated electrodes (4708, 4320, 4420) coupling the substrate (4702, 602, 802, 902, 1002) to the optical antenna (4710, 4408, 806, 906, 1006) and configured to control a position of the optical antenna (4710, 4408, 806, 906, 1006) relative to the first bus waveguide (4706, 4406a, 608), wherein when a voltage difference applied to the interdigitated electrodes (4708, 4320, 4420) is less than a lower threshold, the optical antenna (4710, 4408, 806, 906, 1006) is at a first position offset from the first bus waveguide (4706, 4406a, 608) when the voltage difference applied to the interdigitated electrodes (4708, 4320, 4420) is greater than an upper threshold, the optical antenna (4710, 4408, 806, 906, 1006) is at a second position offset from the first bus waveguide (4706, 4406a, 608) and the offset is greater at the second position than at the first position, a second bus waveguide (4406b, 4406c) supported by the substrate (4702, 602, 802, 902, 1002) and parallel to the first bus waveguide (4706, 4406a, 608), wherein when one is applied to the interdigitated electrodes (4708a, 4320, 4420) is less than a lower threshold, the optical antenna (4710, 4408, 806, 906, 1006) is at a first position offset from the first bus waveguide (4706, 4406a, 608) when the voltage difference applied to the interdigitated electrodes (4708a, 4320, 4420) is greater than an upper threshold, the optical antenna (4710, 4408, 806, 906, 1006) is at a second position offset from the second bus waveguide (4406b, 4406c, 608).The optical switch of claim 1, wherein a first coupling efficiency between the first bus waveguide (4706, 4406a, 608) and the optical antenna (4710, 4408, 806, 906, 1006) is greater than 50 percent and a second coupling efficiency between the second bus waveguide (4406b, 4406c) and the optical antenna (4710, 4408, 806, 906, 1006) is less than 1% when a voltage difference below the lower threshold is applied to the interdigitated electrodes (4708a, 4320, 4420), and the first coupling efficiency between the first bus waveguide (4706, 4406a, 608) and the optical antenna (4710, 4408, 806, 906, 1006) is less than 1 percent and the second coupling efficiency between the second bus waveguide (4406b, 4406c) and the optical antenna (4710, 4408, 806, 906, 1006) is greater than 50 percent when a voltage difference exceeding the upper threshold is applied to the interdigitated electrodes (4708a, 4320, 4420).The optical switch of claim 1, wherein the lower voltage difference threshold is 5 volts and the upper voltage difference threshold is 20 volts.The optical switch of claim 1, wherein the substrate (4702, 602, 802, 902, 1002) is a silicon-on-insulator (SoI) substrate and the bus waveguide (4706, 4406a, 608) is etched in a silicon layer (606) of the substrate (4702, 602, 802, 902, 1002) coupled to an insulating layer (604).The optical switch of claim 1, arranged in an array of optical switches (500, 800, 900, 1000), wherein the optical antenna (4710, 4408, 806, 906, 1006) is an array of optical antennas (500, 800, 900, 1000) arranged in at least two columns, the two columns being spaced apart by an interval, wherein the array of optical antennas (500, 800, 900, 1000) is arranged in at least one row (510) such that the first position is separated from the second position by the interval.The optical switch of claim 5, further comprising an optical tree configured to distribute light to the array of optical antennas (1100, 1200, 1300), the optical tree configured to selectively activate one of the at least one row (510).An optical switch comprising: a bus waveguide (4706, 4406a, 608) supported by a substrate (4702, 602, 802, 902, 1002); an optical antenna (4710, 4408, 806, 906, 1006) suspended over the bus waveguide (4706, 4406a, 608) via a spring (4712); interdigitated electrodes (4708, 4320, 4420) coupling the substrate (4702, 602, 802, 902, 1002) to the optical antenna (4710, 4408, 806, 906, 1006) and configured to control a position of the optical antenna (4710, 4408, 806, 906, 1006) relative to the bus waveguide (4706, 4406a, 608), wherein when a voltage difference applied to the interdigitated electrodes (4708, 4320, 4420) is less than a lower threshold, the optical antenna (4710, 4408, 806, 906, 1006) is controlled, 4408, 806, 906, 1006) is disposed at a first position relative to the bus waveguide (4706, 4406a, 608) when the voltage difference applied to the interdigitated electrodes (4708, 4320, 4420) is greater than an upper threshold, the optical antenna (4710, 4408, 806, 906, 1006) is disposed at a second position relative to the bus waveguide (4706, 4406a, 608), and the first position is separated from the second position by a predetermined distance; wherein a coupling efficiency between the bus waveguide (4706, 4406a, 608) and the optical antenna (4710, 4408, 806, 906, 1006) is greater than 50 percent when the optical antenna (4710, 4408, 806, 906, 1006) is at the first position and at the second position; an array of optical switches (500, 800, 900, 1000), wherein the optical antenna (4710, 4408, 806, 906, 1006) is an array of optical antennas (500, 800, 900, 1000) arranged in at least two rows (510) spaced apart by an interval and at least one column such that the first position is separated from the second position by the interval; An objective (804, 904, 1004) spaced apart from the substrate (4702, 602, 802, 902, 1002) and configured to diffract light onto the optical antenna (4710, 4408, 806, 906, 1006), wherein the objective (804, 904, 1004) is configured to shift along a vector perpendicular to the substrate (4702, 602, 802, 902, 1002), and a distance of the objective (804, 904, 1004) to the substrate (4702, 602, 802, 902, 1002) changes as the objective (804, 904, 1004) shifts along the vector.The optical switch of claim 7, further comprising an optical tree configured to distribute light to the optical antenna array (500, 800, 900, 1000), wherein the optical antenna array (500, 800, 900, 1000) is arranged in at least one row (510) and in at least one column, and the optical tree is configured to selectively activate one of the at least one row (510).The optical switch of claim 7, further comprising an optical tree configured to collect light from the optical antenna array (500, 800, 900, 1000), wherein the optical antenna array (500, 800, 900, 1000) is arranged in at least two rows (510) and in at least one column, and the optical tree is configured to selectively activate one of the at least one row (510).The optical switch of claim 7, further comprising an objective (804, 904, 1004) spaced apart from the substrate (4702, 602, 802, 902, 1002) and configured to diffract light onto the optical antenna (4710, 4408, 806, 906, 1006), wherein the substrate (4702, 602, 802, 902, 1002) is configured to shift along a vector perpendicular to the objective (804, 904, 1004), and a distance of the substrate (4702, 602, 802, 902, 1002) to the objective (804, 904, 1004) changes as the substrate (4702, 602, 802, 902, 1002) shifts along the vector.The optical switch of claim 7, wherein when a voltage difference applied to the interdigitated electrodes (4708, 4320, 4420) is between the lower threshold and the upper threshold, the coupling efficiency is greater than 50 percent.A beam steering system comprising: a bus waveguide (4706, 4406a, 608) supported by a substrate (4702, 602, 802, 902, 1002); an optical antenna (4710, 4408, 806, 906, 1006) supported via a spring (4712) at a distance above the bus waveguide (4706, 4406a, 608); interdigitated electrodes (4708, 4320, 4420) coupling the substrate (4702, 602, 802, 902, 1002) to the optical antenna (4710, 4408, 806, 906, 1006) and configured to control a position of the optical antenna (4710, 4408, 806, 906, 1006) relative to the bus waveguide (4706, 4406a, 608); an optical tree comprising at least one type of optical switch, the optical tree configured to collect light from the optical antenna (4710, 4408, 806, 906, 1006); and an objective (804, 904, 1004) spaced from the substrate (4702, 602, 802, 902, 1002) and configured to diffract the light onto the optical antenna (4710, 4408, 806, 906, 1006), wherein when a voltage difference applied to the interdigitated electrodes (4708, 4320, 4420) is less than a lower threshold, the optical antenna (4710, 4408, 806, 906, 1006) is positioned offset from the bus waveguide (4706, 4406a, 608) a first distance when the voltage difference applied to the interdigitated electrodes (4708, 4320, 4420) is greater than an upper threshold, the optical antenna (4710, 4408, 806, 906, 1006) is positioned offset from the bus waveguide (4706, 4406a, 608) by a second distance and the second distance is less than the first distance; wherein the objective (804, 904, 1004) is configured to shift along a vector perpendicular to the substrate (4702, 602, 802, 902, 1002), and the distance of the objective (804, 904, 1004) to the substrate (4702, 602, 802, 902, 1002) changes as the objective (804, 904, 1004) shifts along the vector.The beam steering system of claim 12, wherein the optical antenna (4710, 4408, 806, 906, 1006) is an array of optical antennas (500, 800, 900, 1000) arranged in at least one row (510) and in at least one column, and the optical tree is configured to selectively activate one of the at least one row (510).The beam steering system of claim 12, wherein the optical antenna (4710, 4408, 806, 906, 1006) is an array of optical antennas (500, 800, 900, 1000) arranged in at least one row (510) and in at least one column, and the optical tree is configured to selectively activate one of the at least one column.The beam steering system of claim 12, wherein the optical antenna (4710, 4408, 806, 906, 1006) is a two-dimensional grating that provides a Gaussian beam spot in a far field, and the Gaussian beam spot is focused on the optical antenna (4710, 4408, 806, 906, 1006) as the objective (804, 904, 1004) shifts along the vector.

Images