Crosstalk mitigation in lidar systems

Abstract

Disclosed are LIDAR (Light Detecting and Ranging) systems which mitigate static and transient optical crosstalk from unselected pixels of their pixel arrays, by configuration and control of the optical switches of the optical switch trees.

Description

[0001] CROSSTALK MITIGATION IN LIDAR SYSTEMS

[0002] FIELD OF THE INVENTION

[0003] The present disclosure relates to photonics systems and particularly to LIDAR (Light Detecting and Ranging) systems which and include emitter arrays and optical switching trees for selectively coupling each of the emitters to one or more laser sources and receivers.

[0004] BRIEF SUMMARY

[0005] According to a first aspect, there is provided a LIDAR system comprising: a laser source for generating optical signals; a plurality of pixels arranged in an array; a receiver for detecting optical signals scattered back from a target; a plurality of optical switches arranged in an optical switch tree for guiding the optical signals from the laser source sequentially to each pixel of the pixel array for launching the optical signals to the target and for guiding the optical signals scattered back from the target from the pixel to the receiver, a branch of the optical switch tree comprising a first optical switch and one or more second optical switches; and a controller for controlling the switch tree to guide said optical signals with use of control signals wherein, while the branch is unselected, the controller controls the first optical switch and the one or more second optical switches with the same OFF control signal, such that a first optical signal traversing a cross port of the first optical switch and through said one or more second optical switches traverses a same number of closed ports of said first and one or more second optical switches as a second optical signal traversing a through port of the first optical switch and through said one or more second optical switches.

[0006] In some embodiments, the OFF control signal is a control voltage of 0V.

[0007] In some embodiments, the controller begins said control of the first optical switch and the one or more second optical switches with the same OFF control signal once optical signals from the laser source have been routed away from said branch.

[0008] In some embodiments, the controller controls one or more optical switches at a higher level in the optical switch tree than the first optical switch and the one or more second optical switches to route optical signals from the laser source away from said branch until a time after the controller controls the first optical switch and the one or more second optical switches with the same OFF control signal sufficient for the first optical switch and the one or more second optical switches to reach steady state operation.

[0009] In some embodiments, the first optical switch is located within one of a transmit switch tree or a receive switch tree of the optical switch tree, and the one or more second optical switches is located within the other one of the transmit switch tree or the receive switch tree, and wherein the one or more second optical switches comprises a second optical switch which is coordinate to the first optical switch, such that said first optical signal traversing the cross port of the first optical switch is guided through the optical switch tree to traverse a through port of the second optical switch, and said second optical signal traversing the through port of the first optical switch is guided through the optical switch tree to traverse a cross port of the second optical switch.

[0010] In some embodiments, the first optical switch and the second optical switch are last level switches in the transmit and receive switch trees.

[0011] In some embodiments, the one or more second optical switches comprise a first select switch and a second select switch, wherein the first optical switch, first select switch, and the second select switch are comprised within a compound optical switch occupying a switch position within one of a transmit switch tree or a receive switch tree of the optical switch tree, and wherein the first select switch is coupled to the cross port of the first optical switch and the second select switch is coupled to the through port of the first optical switch, such that said first optical signal traversing the cross port of the first optical switch traverses a through port of the first select switch, and said second optical signal traversing the through port of the first optical switch traverses a cross port of the second select switch.

[0012] In some embodiments, the compound optical switch occupies a last level switch position in one of the transmit and receive switch trees.

[0013] In some embodiments, the optical switch tree comprises a plurality of branches including said branch, each branch comprising a respective first optical switch and respective one or more second optical switches, wherein while any branch of the plurality of branches is unselected, the controller controls the respective first optical switch and the respective one or more second optical switches of the unselected branch with the same OFF control signal, such that a respective first optical signal traversing a cross port of the respective first optical switch and through said respective one or more second optical switches traverses a same number of closed ports of said respective first and one or more second optical switches as a respective second optical signal traversing a through port of the respective first optical switch and through said respective one or more second optical switches

[0014] In some embodiments, the respective first optical switches are located within one of a transmit switch tree or a receive switch tree of the optical switch tree, and the respective one or more second optical switches are located within the other one of the transmit switch tree or the receive switch tree, and wherein each of the respective one or more second optical switches comprises a respective second optical switch which is coordinate to one of the respective first optical switches, such that said respective first optical signal traversing the cross port of the respective first optical switch is guided through the optical switch tree to traverse a through port of the respective second optical switch, and said respective second optical signal traversing the through port of the respective first optical switch is guided through the optical switch tree to traverse a cross port of the respective second optical switch.

[0015] In some embodiments, the controller begins said control of each respective first optical switch and the respective one or more second optical switches of a branch with the same OFF control signal once optical signals from the laser source have been routed away from said branch.

[0016] In some embodiments, the controller controls one or more optical switches at a higher level in the optical switch tree than the respective first optical switch and the respective one or more second optical switches of a branch to route optical signals from the laser source away from said branch until a time after the controller controls the respective first optical switch and the respective one or more second optical switches with the same OFF control signal sufficient for the respective first optical switch and the respective one or more second optical switches to reach steady state operation.

[0017] In some embodiments, the respective first optical switches and the respective second optical switches are last level switches in the transmit and receive switch trees.

[0018] In some embodiments, each respective one or more second optical switch comprises a respective first select switch and a respective second select switch, wherein each respective first optical switch along with its respective first select switch and respective second select switch are comprised within a respective compound optical switch occupying a respective switch position within one of a transmit switch tree or a receive switch tree of the optical switch tree, and wherein each respective first select switch is coupled to the cross port of the respective first optical switch and each respective second select switch is coupled to the through port of the respective first optical switch, such that said respective first optical signal traversing the cross port of the respective first optical switch traverses a through port of the respective first select switch, and said respective second optical signal traversing the through port of the respective first optical switch traverses a cross port of the respective second select switch.

[0019] In some embodiments, each respective compound optical switch occupies a respective last level switch position in one of the transmit and receive switch trees.

[0020] The foregoing and additional aspects and embodiments of the present disclosure will be apparent to those of ordinary skill in the art in view of the detailed description of various embodiments and / or aspects, which is made with reference to the drawings, a brief description of which is provided next.

[0021] BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The foregoing and other advantages of the disclosure will become apparent upon reading the following detailed description and upon reference to the drawings.

[0023] FIG. 1 is a schematic block diagram of an integrated photonics LIDAR system according to an embodiment.

[0024] FIG. 2 is a schematic block diagram of an example pixel block implementation for an integrated photonics LIDAR system according to an embodiment.

[0025] FIG. 3 is a schematic block diagram of an unfolded optical switch tree system of an integrated photonics LIDAR system according to an embodiment.

[0026] FIG. 4A is a schematic block diagram illustrating crosstalk contributed by coordinate switch pairs in an integrated LIDAR system.

[0027] FIG. 4B is a schematic block diagram illustrating crosstalk contributed by mismatched coordinate switch pairs in an integrated LIDAR system. FIG. 4C is a graph illustrating the functional relationship between optical switch states of an optical switch and the control signals provided thereto.

[0028] FIG. 5 is a schematic block diagram of an unfolded optical switch tree system of an integrated photonics LIDAR system according to an example embodiment utilizing mismatched coordinate switch pairs for static crosstalk mitigation.

[0029] FIG. 6A is a timing diagram illustrating the control signals provided to the optical switch tree of the LIDAR system of FIG. 3.

[0030] FIG. 6B is a timing diagram illustrating the control signals provided to the optical switch tree of the LIDAR system of FIG. 5.

[0031] FIG. 7A is a schematic block diagram of a transmit compound switch according to an embodiment.

[0032] FIG. 7B is a schematic block diagram of a receive compound switch according to an embodiment.

[0033] FIG. 8 is a schematic block diagram of an unfolded optical switch tree system of an integrated photonics LIDAR system according to an example embodiment utilizing compound switches for static crosstalk mitigation.

[0034] FIG. 9 is a timing diagram illustrating the control signals provided to the optical switch tree of the LIDAR system of FIG. 8.

[0035] FIG. 10 is a schematic block diagram illustrating operation of the embodiment of FIG. 8 in the context of pixels 0-3, while pixel 0 is selected.

[0036] FIG. 11A is a schematic block diagram illustrating crosstalk attenuation provided by a mismatched coordinate switch pair while deselected.

[0037] FIG. 1 IB is a schematic block diagram illustrating transient crosstalk exhibited by a mismatched coordinate switch pair while in transition between setpoints.

[0038] FIG. 11C is a schematic block diagram illustrating crosstalk attenuation provided by a compound switch while deselected.

[0039] FIG. 1 ID is a schematic block diagram illustrating transient crosstalk exhibited by a compound switch while in transition between setpoints.

[0040] FIG. 12A is a timing diagram illustrating example control signals provided to the optical switch tree of the LIDAR system of FIG. 5 which mitigates transient crosstalk. FIG. 12B is a timing diagram illustrating example control signals provided to the optical switch tree of the LIDAR system of FIG. 8 which mitigates transient crosstalk.

[0041] While the present disclosure is susceptible to various modifications and alternative forms, specific embodiments or implementations have been shown by way of example in the drawings and will be described in detail herein. It should be understood, however, that the disclosure is not intended to be limited to the particular forms disclosed. Rather, the disclosure is to cover all modifications, equivalents, and alternatives falling within the scope of an invention as defined by the appended claims.

[0042] DETAILED DESCRIPTION

[0043] Photonics systems utilize a host of structural and functional elements to guide, launch, manipulate, or otherwise utilize photonic signals to their desired application. Many LIDAR systems utilize multiple optical receive / transmit locations at the system boundary to direct, like an optical antenna, optical signals into fiee-space and to receive optical signals scattered back from objects and surroundings of interest. On-chip pixel locations directing optical signals into free-space may also operate to receive optical signals therefrom, and may be referred to as emitters, each often with a unique location. Although reciprocity is a general rule for typical optical systems, in some LIDAR systems, some emitters can be dedicated to either launching LIDAR optical signals into free space or to receiving scattered signals only. In some implementations, the emitters are arranged in an array and used in tandem with optical lenses and / or mirrors which assist in beam scanning or steering.

[0044] Some chip-based LIDAR systems employ one or more on-chip LIDAR transceiver blocks, one per channel, which launch and receive optical signals for the system. In LIDAR systems implementing beam scanning, typically a controllable optical switch fabric is implemented between each of the one or more LIDAR transceiver blocks and multiple emitters, to switch the optical signals to and receive optical signals from different emitters at different locations in sequence. Typically, at any one time, each one of the one or more transceiver blocks is optically coupled via a switch fabric to one emitter or emitter block, for transmitting optical signals or receiving optical signals or both. Operation of the switch fabric in this manner enables each emitter to function as an addressable “pixel”. Some multiple transceiver block LIDAR systems employing beam steering employ multiple LIDAR transceiver blocks and switch fabrics to address multiple respective emitters in a parallel fashion. In such implementations the switch fabrics may be driven by global drive signals to simultaneously and in a parallel fashion address emitters for each transceiver block at various respective locations simultaneously, for example, in order within parallel sections or rows.

[0045] In some embodiments, each LIDAR transceiver block sends light out via a transmit switch fabric and then receives the back-scattered signal over a receive switch fabric, determining primarily, but not limited to, an object’s distance, reflectance, and velocity. Some of the types of LIDAR transceiver blocks contemplated by the embodiments include time-of-flight LIDAR transceivers, amplitude modulation coded LIDAR transceivers, and Frequency-Modulated-Continuous-Wave (FMCW) LIDAR transceivers.

[0046] In many LIDAR systems the polarization states of light are exploited or otherwise put to useful application in whatever manner depending upon the context. One such application is polarization-multiplexed LIDAR, in which the LIDAR system transmits optical signals of one polarization but is configured to receive and process optical signals of the orthogonal polarization. For example, some LIDAR systems transmit TE but are arranged to receive and process TM. This is useful in particular contexts, as many targets are often depolarizing, converting some of the optical signals incident on them into the orthogonal polarization upon reflection or back scattering of the optical signals. Sometimes, being able to detect the orthogonal polarization improves the LIDAR system’s ability to see or detect something of interest in connection with the target.

[0047] With reference to FIG. 1 an integrated photonics LIDAR system 10 according to an embodiment will now be discussed.

[0048] A photonics integrated chip 110 of the LIDAR system 10, includes multiple laser sources 11021 110211 one for each channel, which launches a transmit optical signal having an original linear polarization of TE for example, over waveguide 11081 110811 into respective transmit optical switch trees comprising interconnected optical switches TXS WX 11401 114011. A splitter splits a portion of the optical signals into “local oscillator” optical signals LO for respective receivers 11041 110411. The receivers 11041 110411 are coupled to respective receive optical switch trees via waveguides 11981 119811 for receiving return optical signals as described below, the receive optical switch trees comprising interconnected optical switches RXS WX 11601 116011. The optical switches of each switch tree are controlled by a controller 1106 which provides electrical control signals over control lines 1107 to control each switch in each switch tree to select one of two output ports or one of two input ports and deselect the other of the output ports or input ports. In some embodiments, the controller 1106 controls switches at the same position within each transmit optical switch tree using the same control signal, and controls the switches at the same position within each receive optical switch tree using the same control signal.

[0049] The transmit optical signals are switched via the transmit optical switch trees to selected pixel blocks from which the signal exits into free space 1121 11211. The optical signal then encounters optics such as mirrors and / or lens system etc. 114 for collimating and beam steering etc. and traverses free space 1161 116II to various portions of or directions to the target 118. The return optical signals scattered back from the target 118 include optical signals having the original linear polarization TESIG as well as optical signals having the orthogonal linear polarization TMSIG. These signals traverse back through free space 1161 11611, the mirror and / or lens system 1141 11411, free space 1121 11211, and into the pixel blocks 11501 115011. Optical signals of the original linear polarization TESIG pass through the pixel blocks 11501 115011 and over the transmit optical switch trees and are discarded. Optical signals having orthogonal linear polarization TMSIG are rotated into TE optical signals by the pixel blocks 11501 115011 after which they traverse over the switches 11601 116011 of the receive optical switch trees to the receivers 11041 110411.

[0050] In embodiments utilizing scanning, each transmit optical switch tree is controlled by the controller to switch the optical signal to a single pixel block coupled to the tree for transmission to the target and simultaneously each receive optical switch tree is controlled to switch optical signals received from that same single pixel block to the respective receiver. As such, for each channel, a single pixel block is selected for transmission and reception of optical ranging signals for a specific duration, prior to switching to a next pixel block. Once all pixel blocks have been selected in sequence, i.e. a frame has been completed, the controller switches the optical switch trees to start over and select the first pixel block of the sequence and then repeats the sequence in a periodic cycle. Although illustrated as all being part of the photonics integrated chip 110, one or more of the controller 1106, the laser sources 11021 110211, and receivers 11041 110411 may be separate elements coupled to the photonics integrated chip 110.

[0051] With reference to FIG. 2, an implementation 20 of a pixel block 2150 for use within the LIDAR system 10 of FIG. 1 will now be discussed.

[0052] An optical signal TE after passing through the transmit optical switch tree from the laser source, passes over a waveguide 2192 to a polarization splitter rotator (PSR) 2152 of the pixel block 2150. The PSR 2152 is arranged to pass the TE polarized optical signal over a waveguide 2155 towards a dual polarization emitter 2154 of the pixel block 2150 which launches the TE polarized optical signal into free space 212 towards the mirror and / or lens system, and target. The return optical signals scattered back from the target through the mirror and / or lens system traverse free space 212 and arrive at the pixel block 2150 as a received return optical signal having the orthogonal linear polarization TMSIG and a received return optical signal having the original linear polarization TESIG. These received return optical signals traverse the dual polarization emitter 2154 and the waveguide 2155 back into the PSR 2152. The PSR 2152 converts (rotates) the received return optical signals having the orthogonal linear polarization TMSIG into optical signals having the original linear polarization TE and sends them (TMSIG as TE) over waveguide 2194 through the receive optical switch tree to the receiver as TE optical signals where they are used for detection and ranging along with the LO signals provided directly from the laser source. The PSR 2152 allows the received return optical signals of the original linear polarization TESIG to pass over the waveguide 2192 through the transmit optical switch tree and are discarded and ignored.

[0053] Although the pixel blocks 11501 1150II of FIG. 1 and the implementation of the pixel block 2150 of FIG. 2 output a received return optical signal of the orthogonal linear polarization TMSIG as a TE signal after rotation by, for example a PSR, it should be understood that in some embodiments the pixel block (or simply pixel) 2150 comprises one or more optical elements (e.g. some gratings based dual-polarization emitters) which separate the received return optical signals without rotation of the polarization of the received return optical signals of the orthogonal linear polarization TMSIG, instead only rerouting them over the waveguide 2194, through the receive optical switch tree to a TM receiver arranged to receive a signal having the orthogonal linear polarization TM. It should be noted that although FIG. 1 and the remaining embodiments only depict 8 pixels per channel and associated optical switch trees of only 8-1 7 switches, any number of pixel blocks and any sized switch tree per channel are contemplated by the embodiments described hereinbelow. In some embodiments, each optical switch tree is a binary tree of 2k-1 switches for routing light to or from N=2kpixels, where k is a positive integer.

[0054] As noted above, the optical switches of the transmit and receive optical switch trees are configured and controlled to route light through one specific selected or active pixel at a time, which at any instant in time corresponds to a specific direction and a specific portion of the target field. The optical switches also are configured and controlled to receive light from that specific portion of the target field via the same specific active pixel. Optical switch imperfections, however, can cause a small amount of light to be routed through one or more pixels other than the active pixel and also consequently cause detection of light coming back through one or more pixels other than the active pixel. It follows that during the scanning period for a particular active pixel in use for detection and ranging, some small amount of light is passing out through other pixels, and is being scattered back and received by those other pixels from the wrong direction and hence from incorrect portions of the target field. Furthermore, although those other pixels are not selected by the receive optical switch tree, some portion of the light those pixels receive leaks through the receive optical switch tree and is received at the receiver contaminating the true signal for the active pixel. The optical signals erroneously sent through and received back through the inactive pixels via the optical switch trees and contaminating the true signal for the active pixel is called crosstalk. Advantageous customization of the structure, arrangement, and / or functional control of the one or more switch elements 11401 1140II 11601 116011 of the transmit and receive optical switch trees of the embodiment depicted in FIG. 1 , in comparison to known LIDAR systems, mitigate various types of crosstalk as explained further below.

[0055] Reference now will also be made to FIG. 3, in which the interleaved transmit and receive optical switch trees of the LIDAR system 10 of FIG. 1 have been redrawn as a LIDAR system 30 with an “unfolded” optical switch tree system to illustrate operation of the switch trees and crosstalk. The system 30 illustrates the LIDAR system 10 of FIG. 1 in a simplified form, namely, the controller, laser sources, and receivers although present are not shown, the pixel blocks are referred to simply as “pixels” 3150. Moreover, the entire round trip via free space, mirrors and / or lens systems and the target, although still present in reality, have all been wrapped-up and dispensed with in the illustration of FIG. 3 as well as in the remaining figures.

[0056] The optical signal from the laser source first encounters the transmit optical switch tree of the system 30 via waveguide 3108. Each transmit switch 3140 comprises an input port 3142, a through port 3146 and a cross port 3148, and is controlled by the controller via a control signal input to the control port 3144 of the transmit switch 3140. As described above the transmit switches route the optical signals to the active pixel 3150 while the receive switches select optical signals received at that same active pixel and cause them to traverse waveguide 3198 for detection. Each receive switch 3160 comprises a through port 3161, a cross port 3163, and an output port 3167, and is controlled by the controller via a control signal input to a control port 3164 of the switch 3160. For consistency, in all of the illustrations the through port of every standard (or compound) optical switch is represented as being the upper port of each pair of ports (e.g. on the input side of a receive switch and on the output side of each transmit switch) whereas the cross port of every optical switch is represented as being the lower port of each pair of ports. It should be understood that this convention is for understanding the figures only and that the physical ports regardless of their relation to each other as upper, lower, left, right, clockwise or counterclockwise etc. may be physically configured as through ports and cross ports without any need of symmetry or consistency as long as they are coupled appropriately via waveguides to achieve the topology and functionality as described and illustrated herein.

[0057] It should be understood from the above discussion that optical crosstalk can affect detection efficacy and accuracy of the LIDAR system 10 and hence should be as low as possible. As an example, if an inactive pixel 4 happens to hit (i.e. be aimed at) a close, retro reflective target, while the selected and active pixel 7 hits a far, dark target, it is quite possible that the signal leaking though and the transmit optical switch tree through pixel 4 to the target and back through pixel 4 and leaking through the receive optical switch tree to the receiver is of a sufficient magnitude to contaminate the signal being detected for pixel 7 and cause an erroneous detection result. This can be true also for all other inactive pixels of the optical switch tree of system 30 resulting in a cumulative contamination of the active pixel’s optical signal with crosstalk. Typically, a LIDAR system 30 is manufactured and operated to specifications which set a limit on the maximum accepted amount of crosstalk i.e. signal contamination for the active pixel, received at the receiver via all other inactive pixels expressed as a certain (often proportional) threshold which must not be exceed. In some embodiments this maximum amount of crosstalk is -60dB calculated in reference to the signal strength over the active pixel or main signal path. If a target were such that its target regions were equally close to the LIDAR system 10 and equally reflective for each pixel, then the signal strength from each of the inactive pixels should be at most -60dB in comparison to the signal strength detected over the active pixel.

[0058] A first type of crosstalk which is addressed by the embodiments described herein is static crosstalk. Static crosstalk is exhibited once each switch in the optical switch tree of system has reached steady state operation in their respective switch states set by the respective control signals which cause each switch to route optical signals over either their through port or their cross port.

[0059] Each switch is manufactured to ideally possess the same control setpoint (e.g. voltage) which causes the switch to route light over the through port and not over the cross port, referred to herein as the through port setpoint. The switches also ideally possess the same control setpoints which causes the switch to route light over the cross port and not over the through port, also referred to as cross port setpoints. This allows for driving tied switches, i.e. simultaneous control of switches at the same position within different trees for different channels, using the same control signal. For example, in FIG. 1, the first level transmit switches TXSWO 11401 and 114011 of both channels I and II, as well as any other parallel first level transmit switches TXSWO of other channels (not shown) may be controlled using the same control signal TX SW00 over control line 1107, reducing cost and complexity of the system. In some embodiments the number of channels is 8, and generally can be any number.

[0060] Due to fabrication variations e.g. from the semiconductor foundry, the tied switches in reality have slightly different setpoints. For example, switches manufactured to have a through port setpoint of 4.8V might actually have a setpoint of 4.78V or 4.82V. Consequently, when tied switches are driven by the same common voltage, some of the switches are not biased at their own perfect setpoints. In some embodiments, and given current switch design and fabrication conditions, the statistically average maximum static crosstalk exhibited from a single switch is around -20 dB. In other words, the unselected port of each optical switch can be expected on average to leak at most -20dB of optical signal expected to traverse the selected port. In the context of the optical switch tree of system 30, when pixel 7 is active, all of the switches TXSWO, TXSW2, TXSW6, RXSW6, RXSW2, and RXSWO are controlled at their through setpoints so that light is routed through their through ports. All the remaining switches are deselected or inactive (“Off” as described below) so that light is routed over their cross ports. The result is a detection signal over waveguide 3198 which has a main signal which passes through pixel 7, and seven other contamination signals from the other pixels, namely, 2ndorder crosstalk (-40dB) from pixels 0, 4, and 6, 4thorder crosstalk (-80dB) from pixels 1, 2, and 5, and 6thorder crosstalk (-120dB) from pixel 3 which are cumulative. Since doubling crosstalk only increases the crosstalk by 6dB, the contributions from the resulting lowest order of crosstalk (and how many of those are cumulative) is the most controlling factor to consider in crosstalk mitigation strategies. As noted above in some embodiments an upper limit of -60dB in crosstalk is desired, i.e. at most 3rdorder crosstalk. Here, in this example, it can be seen that 2ndorder crosstalk, which exceeds the desired amount, is contributed from neighboring pixel 6, as well as being contributed from farther-out pixel 4, and even from the pixel farthest from pixel 7, namely, pixel 0.

[0061] It should be understood that reference to these resulting “orders” of crosstalk from any particular pixel contaminating the main signal is only a comparative measure of crosstalk and is not a statement about the actual strength or absolute magnitude of that contaminating signal. This is due to the fact that the actual backscattered signal contributing to contamination from any particular pixel can vary drastically depending upon factors including the distance to and the reflectivity of the target region corresponding to that pixel. It should be understood, that spurious or unwanted light contaminating the main signal after having passed through a single “closed” port of an optical switch is referred to herein as “first order crosstalk” due to its having a magnitude corresponding to its having been attenuated once by a closed port. Similarly, spurious or unwanted light contaminating the main signal after having passed through N closed ports of any number of optical switches is referred to herein as “Nth order crosstalk” due to its having a magnitude corresponding to its having been attenuated N times by the closed ports.

[0062] Although, in the example embodiments described herein, each closed port imposes attenuation of -20dB, in general this value depends upon the particulars of the implementation and fabrication of the actual optical switches utilized.

[0063] In the embodiments described herein, to save power and to reduce cost and complexity, switches upstream or downstream from only unselected pixels are kept in a default “Off’ state, and only switches which are required to route light are activated with either a through port setpoint signal or a cross port setpoint signal to ensure proper routing of optical signals. As noted in the above example, when pixel 7 is selected as active, only six switches are required to be active TXSWO, TXSW2, TXSW6, RXSW6, RXSW2, and RXSWO, while all other switches are biased at a default “Off’ voltage (e.g. OV) and, in order to save driving power, are generally kept at that level until activated. In some embodiments, each switch is designed to route all light over the cross port when in the “Off’ state i.e. when driven with OV bias voltage.

[0064] Referring now to FIG. 4A, the crosstalk contributed by coordinate switches 40A in dependence upon their states will now be discussed.

[0065] In the context of any single optical switch tree of system 30, a coordinate pair of switches consist of the switches at opposite positions “X” in the transmit and receive trees, such as TXSWX 4140A in the transmit optical switch tree and RXSWX 4160 A in the receive optical switch tree. Examples include TXSW2 3140 and RXSW2 3160 of FIG. 3. When a transmit switch TXSWX 4140A is driven with the “Off’ signal OV, optical signals entering the input port 4142A of the switch are routed over the cross port 4148 A to a lower path while the through port 4146A of the switch 4140 A presents crosstalk attenuation in an amount caused by a single closed port (in the example embodiment -20db but could be any actual value depending upon switches implemented) to light traversing over an upper path. Optical signals on the upper and lower paths from the transmit switch 4140 A pass through various other switches of both trees and multiple pixels 4155 A and then are input to the coordinate receive switch RXSWX 4160A. Optical signals traversing the lower path entering the cross port 4163 A of coordinate receive switch RXSWX 4160A controlled by an “Off’ control signal, see no appreciable attenuation and pass over the output 4167A of the switch 4160 A. Optical signals traversing the upper path entering the through port 4161 A of the coordinate receive switch RXSWX 4160A controlled by the “Off’ signal, traverse and are attenuated by a single closed port. As is clear FIG. 4A, when coordinate switches are driven with the same “Off’ signal, optical signals over one set of paths therebetween (here the upper) traverse and are attenuated by two closed ports of the coordinate switches while optical signals over the other set of paths therebetween (here the lower) encounter no attenuation from any closed ports of the coordinate switches. As noted above, the lowest order crosstalk is of prime importance in determining the contamination of the signal detected and accordingly coordinate switches as depicted in FIG. 4A present some optical signal contamination which has not been attenuated by those coordinate switches to any appreciable amount.

[0066] The embodiment 40B depicted in FIG. 4B addresses the problem of non-attenuation by flipping one of the coordinate switches, for example the receive switch RXSWX 4160B. When the coordinate transmit switch TXSWX 4 MOB is driven with the “Off’ signal 0V, optical signals entering the input port 4142B of the switch are routed over the cross port 4148B to a lower path while the through port 4146B of the switch 4 MOB presents crosstalk attenuation of a closed port to light traversing over the upper path. Optical signals on the upper and lower paths from the transmit switch 4 MOB pass through various other switches of both trees and multiple pixels 4155B and then are input to the coordinate receive switch 4160B which has been flipped such that the through port 4161B is coupled to the lower path and the cross port 4163B is coupled to the upper path, or equivalently, the light passing through the through port of the transmit switch TXSWX 4 MOB is coupled to the cross port of the coordinate receive switch RXSWX 4160B, and light passing through the cross port of the transmit switch TXSWX 4 MOB is coupled to the through port of the coordinate receive switch RXSWX 4160B. Optical signals traversing the lower path entering the coordinate receive switch RXSW 4160B controlled by an “Off’ control signal, traverse and are attenuated by a closed port and pass over the output 4167B of the switch 4160B . Optical signals traversing the upper path entering the coordinate receive switch 4160B controlled by the “Off’ signal, also traverse and are attenuated by a closed port. As is clear from FIG. 4A, when coordinate switches are driven with the same “Off’ signal, and where one of the coordinate switches is flipped with respect to the other, optical signals over both the upper and lower paths therebetween each traverse a respective closed port of the coordinate switches encountering the same amount of crosstalk attenuation from the coordinate switches. As noted above, the lowest order crosstalk is of prime importance in determining the contamination of the signal detected and accordingly it can be seen that coordinate switches which are “Off’, where one has been flipped, present crosstalk attenuation to optical signals of both the upper and lower path, each of which is has been attenuated by one respective closed port.

[0067] Although FIG. 4B depicts the receive switch 4160B as flipped and the transmit switch 4 MOB as not flipped, it should be understood that the embodiments contemplate flipping of either one of the transmit and receive switches relative to the other. Generally speaking, as long as the coordinate switches are “mismatched”, i.e. when both are controlled with the default “Off’ voltage, one switch routes signals without appreciable attenuation over one of the upper and lower paths while the other routes optical signals without appreciable attenuation over the other one of the upper and lower paths, the result will be optical crosstalk attenuation of each signal by a respective closed port of the coordinate pair of switches.

[0068] With reference to FIG. 4C the relationship between the switch states of the optical switches and the control signals provided thereto will now be discussed. FIG. 4C shows the switch output through the cross port and the through port in response to control signals, in this case bias voltage, provided to the control port. As can be seen, OV of bias voltage is a low cross port setpoint (also referred to as “LC”) and is the same voltage provided when the switch is turned to its unselected, inactive or off state (also referred to as “0” or “Off’). Although referring to the same voltage, “LC” is used to denote OV when driving a switch along the active path whereas “0” or “Off” is used to denote OV when driving a switch which is off of the active path. A bias voltage of just below 5 V is a through port setpoint (also referred to as “T”) while a higher voltage of about 7V is a second cross port setpoint i.e. the high cross port setpoint, also referred to as “HC”.

[0069] With reference to FIG. 5 an example optical switch tree of a LIDAR system 50 employing flipped receive switches 5160 in the last level of the receive optical tree, will now be discussed. The optical signal from the laser source first encounters the transmit optical switch tree of the system 50 via waveguide 5108. Each transmit switch 5140 comprises an input port 5142, a through port 5146, and a cross port 5148, and is controlled by the controller via a control signal input to the control port of the switch. As described above the transmit switches route the optical signals to the active pixel 5150 while the receive switches select optical signals received at that same active pixel and cause them to traverse waveguide 5198 for detection. Each receive switch 5160 comprises a through port 5161, a cross port 5163, and an output port 5167, and is controlled by the controller via a control signal input to control port of the switch 5160. It should be clear that for any branch of the optical switch tree of system 50 which is inactive, coordinate pairs of switches therein at the last level of the optical trees, i.e. each of switch pairs TXSW3 - RXSW3, TXSW4 - RXSW4, TXSW5 - RXSW5, and TXSW6 - RXSW6, which are turned off, present crosstalk attenuation of a respective closed port to both upper and lower branches. In an example similar to the above, in the context of the optical switch tree of system 50, when pixel 7 is active, all of the switches TXSW0, TXSW2, TXSW6, RXSW2, and RXSW0 are controlled at their through setpoints so that light is routed through their through ports. Because it is flipped, the receive switch RXSW6 is controlled at its high cross point setpoint (HC), while coordinate switch pairs TXSW3 - RXSW3, TXSW4 - RXSW4, and TXSW5 - RXSW5 are all “Off “ as are all the remaining switches. The result is a detection signal over waveguide 5198 which has a main signal which passes through pixel 7, and seven other contamination signals from the other pixels, namely, 2ndorder crosstalk (-40dB) from pixel 6 only, 3rdorder crosstalk (-60dB) from pixels 5, 4, 1, and 0, and 5thorder crosstalk (- lOOdB) from pixel 3.

[0070] Here all of the distant pixels 0-5 meet the crosstalk specification noted above, since they each exhibit at most -60dB (3rdorder) crosstalk. In this embodiment, 2ndorder crosstalk exceeding the desired amount, is contributed only from neighboring pixel 6. As can be seen from FIG. 5, adjacent pairs of pixels, such as pixel 0 and pixel 1 are coupled to the same pair of last level switches, as are pixels 2 and 3, pixels 4 and 5, and pixels 6 and 7. In general, one pixel adjacent to the active pixel, namely the pixel of the pair of pixels coupled to the same last level switches as the active pixel, contributes 2ndorder crosstalk. However, crosstalk from adjacent pixels is less critical than crosstalk from non-adjacent pixels, because even if the detected signal for a pixel is contaminated by a signal from an adjacent pixel, the worst result is some loss of acuity / angular resolution caused by some portion of the adjacent pixel signal being effectively blurred into the selected pixel’s signal. For systems having a high pixel density, this is less of an issue.

[0071] In some embodiments some or all of the remaining receive switches (other than RXSWO which is always active) are also flipped. In some embodiments some or all of the transmit switches are flipped while the receive switches are not. And in other embodiments any combination of flipped transmit or receive switches, only one flipped for each coordinate pair, is utilized to mitigate static crosstalk.

[0072] In an embodiment similar to that depicted in FIG. 5 having an additional two flipped switches, i.e. all of RXSW1-RXSW6 flipped, when a main signal is passed through pixel 7, the seven other contamination signals from the other pixels include, 2ndorder crosstalk (-40dB) from pixels 6 only, 3rdorder crosstalk (-60dB) only from pixels 5 and 4, and 4thorder crosstalk (-80dB) from pixels 0, 1, 2, and 3.

[0073] Timing diagrams 60A 60B for the operation of the systems depicted in FIGs. 3 and 5 are illustrated respectively in FIGs. 6A and 6B. As can be seen in the diagrams, the control signal setpoints of timing diagram 6B for the flipped switches RXSW3-RXSW6, when active, utilize the opposite bias voltage setpoints compared to that used to drive the corresponding unflipped switches of timing diagram 6A. Recall, that the through ports and cross ports of the flipped switches RXSW3-RXSW6 are coupled to the topology of the system 50 of FIG. 5 in the same manner as the opposite ports of the unflipped switches RXSW3-RXSW6 in system 30 of FIG. 3.

[0074] In more stringent use cases, for example where even adjacent pixels are required to contribute no more than -60dB of crosstalk, something different from or additional to the flipping strategy described hereinabove may be utilized. In one example, any optical switch of the system 30 depicted in FIG. 3 may be augmented with two additional select switches to improve performance. With reference also to FIGs. 7A and 7B these augmented compound switches will now be described.

[0075] The transmit compound switch 7110 of FIG. 7 A includes a main transmit switch 7140 whose input port serves as the input port 7112 of the transmit compound switch 7110. A control signal port of the main transmit switch 7140 serves as the control port 7114 of the transmit compound switch 7110. The main transmit switch 7140 is coupled to a first select switch SSX-07120 and a second select switch SSX-1 7130 of the transmit compound switch 7110. The cross port of the main transmit switch 7140 is coupled to the input port 7122 of the first select switch 7120, while the through port of the main transmit switch 7140 is coupled to the input port 7132 of the second select switch 7130. Both the first and second select switches 7120 7130 are controlled by the same control signal via a single select switch control port 7115 of the compound switch 7110. The first select switch 7120 has a through port 7126 which functions as the cross port 7118 of the transmit compound switch 7110. The first select switch 7120 does not provide any output from its cross port 7128 which is simply terminated. The second select switch 7130 has a cross port 7138 which functions as the through port 7116 of the transmit compound switch 7110. The second select switch 7130 does not provide any output from its through port 7136 which is simply terminated.

[0076] The receive compound switch 7190 of FIG. 7B includes a main receive switch 7160 whose output port serves as the output port 7197 of the receive compound switch 7190. A control signal port of the main receive switch 7160 serves as the control port 7194 of the receive compound switch 7190. The main receive switch 7160 is coupled to a first select switch SSX-07170 and a second select switch SSX-1 7180 of the receive compound switch 7190. The cross port of the main receive switch 7160 is coupled to the output port 7177 of the first select switch 7170, while the through port of the main receive switch 7160 is coupled to the output port 7187 of the second select switch 7180. Both the first and second select switches 7170 7180 are controlled by the same control signal via a single select switch control port 7195 of the receive compound switch 7190. The first select switch 7170 has a through port 7171 which functions as the cross port 7193 of the receive compound switch 7190. The first select switch 7170 does not have any input over its cross port 7173 which is simply terminated. The second select switch 7180 has a cross port 7183 which functions as the through port 7191 of the receive compound switch 7190. The second select switch 7180 does not have any input over its through port 7181 which is simply terminated.

[0077] To select or activate the transmit compound switch 7110 for directing light over its through port 7116, a control signal is sent to the control port 7114 to control the main transmit switch 7140 to route light over its through port to the second select switch 7130, while the select switch control signals are sent to control the second select switch 7130 to route light over its cross port 7138. The same select switch control signal causes the first select switch 7120 to route any optical signal away from its through port 7126 and to its cross port 7128 where it is terminated. In embodiments where the switch states of the switches are controlled in accordance with control signals as depicted in FIG. 4C, the control signal input to the control port 7114 corresponds to the through port setpoint biasing voltage “T”, whereas the select switch control signal is either a low cross port setpoint bias voltage “LC” or a high cross port setpoint bias voltage “HC”. In such a state the main transmit switch 7140 presents crosstalk attenuation of a closed port to signals exiting the cross port of the transmit compound switch as does the first select switch which is biased away from its through port 7126, together causing crosstalk attenuation of two closed ports along the lower path through the transmit compound switch 7110 and over its cross port 7118.

[0078] To select or activate the transmit compound switch 7110 for directing light over its cross port 7118, a control signal is sent to the control port 7114 to control the main transmit switch 7140 to route light over its cross port to the first select switch 7120, while the select switch control signals are sent to control the first select switch 7120 to route light over its through port 7126. The same select switch control signal causes the second select switch 7130 to route any optical signal away from its cross port 7138 and to its through port 7136 where it is terminated. In embodiments where the switch states of the switches are controlled in accordance with control signals as depicted in FIG. 4C, the control signal input to the control port 7114 corresponds to either a low cross port setpoint bias voltage “LC” or a high cross port setpoint bias voltage “HC”, whereas the select switch control signal is set to the through port setpoint biasing voltage “T”. In such a state the main transmit switch 7140 presents crosstalk attenuation of a closed port to signals exiting the through port of the transmit compound switch as does the second select switch which is biased away from its cross port 7138, together causing crosstalk attenuation of two closed ports along the upper path through the transmit compound switch 7110 and over its through port 7116.

[0079] To deselect or turn “Off’ the transmit compound switch 7110, an “Off’ or 0V control signal is sent to the control port 7114 to control the main transmit switch 7140 to route light over its cross port to the first select switch 7120, while the select switch control signals are set to “Off’ or 0V to control the second select switch 7130 to route light over its cross port 7138. The same select switch control signal causes the first select switch 7120 to route any optical signal away from its through port 7126 and to its cross port 7128 where it is terminated. In such a state the main transmit switch 7140 presents crosstalk attenuation of one closed port to signals exiting the through port of the transmit compound switch, while the first select switch which is biased away from its through port 7126, causes one closed port of crosstalk attenuation to signals exiting the cross port of the transmit compound switch, resulting in crosstalk attenuation of one closed port along each of the upper and lower paths through the transmit compound switch 7110.

[0080] As such, when selected and activated, the transmit compound switch 7110 provides an additional closed port amount of crosstalk attenuation over the unselected path in comparison with the individual switches described above.

[0081] Although described in association with transmit compound switch 7110 the same principles apply in the inverse direction for the crosstalk performance of the receive compound switch 7190.

[0082] With reference to FIG. 8 an example optical switch tree of a LIDAR system 80 employing transmit compound switches 8110 in the last level of the transmit optical tree, will now be discussed.

[0083] The optical signal from the laser source first encounters the transmit optical switch tree of the system 80 via waveguide 8108. Each transmit compound switch 8110 is coupled in the topology via its input port 8112, its through port 8116, and its cross port 8118, and is controlled by the controller via control signal inputs to the control port and the select switch control port of the switch 8110. As described above the transmit switches and transmit compound switches route the optical signals to the active pixel 8150 while the receive switches select optical signals received at that same active pixel and cause them to traverse waveguide 8198 for detection. Each receive switch 8160 comprises a through port 8161 , a cross port 8163, and an output port 8167, and is controlled by the controller via a control signal input to the control port of the switch 8160. It should be clear that along last level branches of the optical switch tree of system 80 which are inactive, the transmit compound switches which are turned off present crosstalk attenuation of a closed port over each port. In an example similar to the above, in the context of the optical switch tree of system 80, when pixel 7 is active, all of the switches TXSW0, TXSW2, RXSW6, RXSW2, and RXSW0 as well as transmit compound switch TXCSW6 are controlled at their through port setpoints so that light is routed through their through ports. Accordingly, the select switch control signal sent to the transmit compound switch TXCSW6 8110 controls the select switches therein at their cross port setpoints. The remaining transmit compound switches TXCSW3-TXCSW5 are all switched “Off “ as are all the remaining switches. The result is a detection signal over waveguide 8198 which has a main signal which passes through pixel 7, and seven other contamination signals from the other pixels, namely, 3ndorder crosstalk (-60dB) from pixels 6, 4, and 0, 4thorder crosstalk (-80dB) from pixels 5 and 1, 5thorder crosstalk (-100dB) from pixels 2, and 6thorder crosstalk from pixel 3.

[0084] Here all of the distant pixels 0-5 and the neighboring pixel, pixel 6, meet the crosstalk specification noted above, since they each exhibit at most -60dB (3rdorder) crosstalk.

[0085] In some embodiments some or all of the remaining transmit switches are compound switches. In some embodiments some or all of the receive switches are compound switches. In other embodiments any combination of transmit and receive compound switches and standard transmit and receive switches are utilized to mitigate static crosstalk. Including multiple additional compound switches to mitigate crosstalk can come at the cost of additional power to provide the separate select switch control signals, as well as additional complexity both in terms of signal processing and signal lines, however, crosstalk performance is enhanced which may be preferred for certain implementations and contexts.

[0086] In some embodiments, use of compound switches for better inactive neighboring branch performance as well as flipping within coordinate pairs of standard or compound switches may be used in any combination according to the teachings above to mitigate crosstalk.

[0087] A timing diagram 90 for the operation of the system depicted in FIGs. 8 is illustrated in FIG. 9. Timings for the transmit compound switches TXCSW3-TXCSW6 8110 are provided and labelled according to the respective control signals for the respective main switches TXSW3-TXSW6 therein and the respective select switch control signals SS3-SS6 for the respective select switches therein. Both first select switch SSX-0 and second select switch SSX-1 are controlled by the same select switch control signal which is simply denoted SSX in FIG. 9 for the Xth transmit compound switch. As can be seen in the diagram, the setpoints for the select switches SSX of the corresponding transmit compound switch are inverse to the control signals of the main switch TXSWX of that compound switch when the compound switch is active. Moreover, when the compound switch is “off’ both the select switch control signal SSX and the main control signal TXSWX are OV or “Off’.

[0088] Depicted in FIG. 10 is an example embodiment 100 illustrating pixels 0-3 of the system 80 of FIG. 8 as well as branches of the last two levels of the transmit optical tree which are associated with those four pixels, pixel 0 of which, is selected as active. Two transmit compound switches TXCSW3 and TXCSW4 are shown. Since pixel 0 is activated, TXSW1 (not shown) is switched to its cross port and hence its through port presents -20dB of crosstalk attenuation to the input of the upper transmit compound switch TXCS W4 which feeds pixels 2 and 3. Since the upper transmit compound switch TXCSW4 is on a completely inactive branch it is deselected or turned off, and hence “Off’ or 0V control signals are sent to the main transmit switch TXSW4 and also to both select switches SS4- 0, SS4-1. This causes the main transmit switch TXSW4 to switch to its cross port, causing an extra -20dB of crosstalk attenuation on the through port, resulting in -40dB total crosstalk attenuation into pixel 3. The select control signal being “Off’ also causes the select switches SS4-0 SS4-1 to route light away from their through ports to their cross ports causing an additional -20dB crosstalk attenuation on the through port of the first select switch SS4-0. The result is -40dB total crosstalk attenuation into pixel 2. Since the lower transmit compound switch TXCSW3 is on an active branch, it is selected or turned on to route optical signals over its cross port to pixel 0. Accordingly, a cross port setpoint control signal, in this case a low cross point control signal LC, is sent to the main transmit switch TXSW3 while a through port setpoint control signal T is sent to both select switches SS3- 0, SS3-1. This causes the main transmit switch TXSW3 to switch to its cross port, causing -20dB of crosstalk attenuation on the through port. The select control signal being “T” also causes the select switches SS3-0 SS3-1 to route light away from their cross ports to their through ports causing an additional -20dB crosstalk attenuation on the cross port of the second select switch SS3-1. The result is -40dB total crosstalk attenuation into pixel 1. It should be noted that the receive switches also provide additional crosstalk attenuation to signals passing through pixels 1-3 prior to detection. The above embodiments address static crosstalk which is exhibited after a steady state has been achieved in response to the control signals being sent to all the switches within the optical trees. It should be understood that switching between active pixels, during the process of cyclic scanning through each of the pixels e.g. repeatedly scanning pixel 0 through pixel 7, requires bias voltages to be varied from one of the various states to another, e.g. HC to T, T to HC, HC to 0 / LC, 0 / LC to HC, T to 0 / LC, and 0 / LC to T. With reference once again to FIG. 4C, HC and LC setpoints are, in terms of switch performance in routing light over the cross port, nominally identical. The reason for having two different setpoints giving rise to the same switch states is due primarily to temporal considerations of the transitions between states.

[0089] In order to cause faster transitions between states, the switches are driven with preemphasis in which the driver voltage overshoots the target value momentarily causing a faster switch transition than simply switching to the next steady-state voltage. For example, in a transition from LC to T, the switch is driven with a control voltage that switches from OV to a voltage momentarily exceeding the bias voltage for state T, before being lowered to the steady-state bias voltage corresponding to T. Similarly, in a transition from HC to T, the switch is driven with a control voltage that switches from the bias voltage corresponding to HC to a voltage momentarily below the bias voltage for state T, before being raised to the steady-state bias voltage corresponding to T. It is not possible, for many useful switches, to drive the switch at a negative setpoint (e.g. heaters cannot be “cooled” by setting a negative voltage) and in some cases a negative bias may fall outside of the switch device specifications. Consequently, pre-emphasis is impossible for transitions from T to LC / 0. The HC setpoint is utilized in order to provide a quick transition from T to a cross port setpoint. This enables pre-emphasis in both negative and positive directions without any need to go below OV. When transition time is not critical, the LC or 0 setpoint is utilized because it consumes much less power than the HC setpoint.

[0090] As noted above, in a cyclic switching scheme, the light signal iterates through all the pixels in each “frame” and repeats in the next “frame”. Each pixel duration of the embodiments is typically on the order of tens of microseconds, for example, around 3 Ops, hence switches are controlled to change state to route the light from the one pixel to the next pixel every 30ps. In some embodiments thermo-optic phase shifters are utilized to control the optical switches, and due to the finite time required for the process of heat transfer, it takes a finite time for each optical switch to change state. Although the voltage may be switched almost instantaneously between for example OV and ~5V, the optical state of the switch will require at least microseconds and sometimes tens of microseconds to change. With preemphasis driving, the switch undergoes “fast switching” on the order of microseconds for transitions between LC / 0 to T, T to HC, LC / 0 to HC, and HC to T. With regular driving, the switch undergoes “slow switching” on the order of tens of microseconds for transitions from T to 0 / LC and HC to 0 / LC, due to including the time it takes for natural cooling to the 0 state.

[0091] Switches in an embodiment utilizing the mismatched coordinate switch pairs and / or compound switch strategies above, ideally would close their through ports and open their cross ports instantly when turned off to OV. However, due to the slow cooling on the order of tens of microseconds noted above, both ports of each switch may remain partially open for some period of time while the switch transitions to the steady state. This has a negative impact on the active pixel signal detected when the above static crosstalk mitigation strategies are being employed with standard timings. The crosstalk resulting from the momentary substandard functionality of the switches is referred to as transient crosstalk.

[0092] Referring to FIGs. 11 A, 1 IB, 11C, and 1 ID, transient crosstalk in the context of the performance of coordinate switch pairs and compound switches will now be discussed.

[0093] In FIG. 11 A, during steady state inactive operation of a coordinate switch pair 110A including TXSWX and RXSWX which are both “Off’, the pair of switches together are capable of presenting -20dB of crosstalk attenuation over both the upper and lower branches. However, as illustrated in FIG. 1 IB, during a transition between states, operation of a coordinate switch pair HOB including TXSWX and RXSWX which are each in some arbitrary state, is such that the pair of switches together may present as little as -6dB of crosstalk attenuation over both the upper and lower branches. From FIG. 4C, it can be seen that this worst case scenario could occur if both switches of the coordinate pair are driven at around 3.5V or 6V, however it should be noted that that everywhere in between the bias voltage setpoints LC (or 0), T, and HC, both ports of each of the switches are open to some degree. In FIG. 11C, during steady state inactive operation of a transmit compound switch HOC which is controlled to be “Off’, the main transmit switch and the select switches together are capable of presenting -20dB of crosstalk attenuation over both the cross port and the through port. However, as illustrated in FIG. 11D, during a transition between states, operation of a transmit compound switch 11 OD including TXSWX, SSX-0, and SSX- 1 which are each in some arbitrary state, is such that the transmit compound switch may present as little as -6dB of crosstalk attenuation over each of the cross and through ports. From FIG. 4C, it can be seen that this worst case scenario could occur if all switches of the transmit compound switch are driven at around 3.5V or 6V, however everywhere in between the bias voltage setpoints LC (or 0), T, and HC, both ports of each of the switches are open to some degree.

[0094] Referring once again to the timing diagram 60B of FIG. 6B, in an embodiment utilizing flipped switches, scanning from Pixel 1 to Pixel 2 involves a slow transition for both coordinate switches TXSW3 and RXSW3 from T and HC respectively to a state of 0. This means TXSW3 and RXSW3 are still in the transition process and have both ports partially open during P2 (the duration for pixel 2). For the flipping strategy to successfully provide crosstalk attenuation of a closed port to optical signals from non-adjacent pixels, it requires the coordinate TXSWX and RXSWX switches to fully stay at their predetermined opposite states while “Off’. In such a context the partially open ports can cause unexpected transient crosstalk on branches which are Off. Since received power is integrated over the entire pixel duration in FMCW detection, transient crosstalk can contribute to the overall crosstalk and potentially cause detection error.

[0095] In embodiments directed to transient crosstalk mitigation, for slow transitions from T or HC to 0, switches in a compound switch or a coordinate switch pair may be kept in their current state or switched to a different state for one or more extra pixel durations until one or more of their parent switch or switches (e.g. one or more levels higher) have had a chance to change state so that the optical power has been routed away from the switches of the compound switch or coordinate pair. In some embodiments, switches which are not part of a compound switch nor in a coordinate pair can be kept in their current state or switched to a different state to mitigate transient crosstalk exhibited by switches which are part of a compound switch or in a coordinate pair. In a first example embodiment of transient crosstalk mitigation, timing of the switches in the embodiment of FIG. 5 which utilizes mismatching of coordinate pairs, will now be discussed with reference to the timing diagram 120 A of FIG. 12 A.

[0096] Compared to the timing diagram of FIG. 6B, switch TXSW3 makes its slow transition from T to 0 two pixel durations later, at the beginning of P4. This delays its slow transition until after TXSW1 has already fast switched routing light away from TXSW3 at the beginning of P2. Also, the slow transition is during P4 sometime after the top-level switches TXS WO and RXSWO fast switch light from their cross ports to their through ports, further mitigating transient crosstalk through TXSW3. Related switch RXSW3 is fast switched from HC to T at the beginning of P2 and is held there to maintain coordinate mismatching until the end of the T state for TXSW3.

[0097] In a similar manner TXSW5 makes its slow transition from T to 0 two pixel durations later, at the beginning of P0. This delays its slow transition until after TXSW2 has already fast switched routing light away from TXSW5 at the beginning of P6. Also, the slow transition is during P0 sometime after the top-level switches TXSWO and RXSWO fast switch light away from their through ports to their cross ports, further mitigating transient crosstalk through TXSW5. Related switch RXSW5 is fast switched from HC to T at the beginning of P6 and is held there to maintain coordinate mismatching until the end of the T state for TXSW5.

[0098] Switch TXSW4 makes its slow transition from T to 0 one pixel duration later, after the end of P4. This delays its slow transition until after TXSWO has fully switched routing light away from TXSW4 at the beginning of P4 mitigating transient crosstalk through TXSW4. Related switch RXSW4 is fast switched from HC to T at the beginning of P4 and is held there to maintain coordinate mismatching until the end of the T state for TXSW4.

[0099] Switch TXSW6 makes its slow transition from T to 0 one pixel duration later, after the end of P0. This delays its slow transition until after TXSWO has fully switched routing light away from TXSW6 at the beginning of P0 mitigating transient crosstalk through TXSW6. Related switch RXSW6 is fast switched from HC to T at the beginning of P0 and is held there to maintain coordinate mismatching until the end of the T state for TXSW6.

[0100] Also illustrated in FIG. 12 A, sometimes switch state extension by a single switch which is not in any coordinate pair can lead to transient crosstalk mitigation of switches at lower levels. For example, switch TXSW1 has its slow transition delayed until the beginning of P5, shielding the slow transition of TXSW3 during P4 from light it would begin to route towards TXSW3 once TXSW1 turns off. Similarly, TXSW2 has its slow transition delayed until the beginning of Pl, shielding the slow transition of TXSW5 during PO from light it would begin to route towards TXSW5 once TXSW2 turns off.

[0101] In a second example embodiment of transient crosstalk mitigation, timing of the switches in the embodiment of FIG. 8 which utilizes compound switches, will now be discussed with reference to the timing diagram 120B of FIG. 12B.

[0102] Compared to the timing diagram of FIG. 9, the TXSW3 switch of the compound switch TXCSW3 makes its slow transition from T to 0 two pixel durations later at the beginning of P4, while the SS3 switches of the compound switch TXCSW3 are fast switched from HC to T at the beginning of P2 and are held there to maintain the T state for TXCSW3. This delays TXCSW3’s slow transition until after TXSW1 has already fast switched routing light away from TXCSW3 at the beginning of P2. Also, the slow transition is during P4 sometime after the top-level switches TXSWO and RXSWO fast switch light from their cross ports to their through ports, further mitigating transient crosstalk through TXCSW3.

[0103] Similarly the TXSW5 switch of the compound switch TXCSW5 makes its slow transition from T to 0 two pixel durations later at the beginning of PO, while the SS5 switches of the compound switch TXCS W5 are fast switched from HC to T at the beginning of P6 and Eire held there to maintain the T state for TXCSW5. This delays TXCSW5’s slow transition until after TXSW2 has already fast switched routing light away from TXCSW5 at the beginning of P6. Also, the slow transition is during PO sometime after the top-level switches TXSWO and RXSWO fast switch light from their through ports to their cross ports, further mitigating transient crosstalk through TXCSW5.

[0104] Switch TXSW4 of the compound switch TXCSW4 makes its slow transition from T to 0 one pixel duration later, after the end of P4, while the SS4 switches of the compound switch TXCSW4 are fast switched from HC to T at the beginning of P4 and held to maintain the T state for TXCSW4. This delays TXCSW4’s slow transition until after TXSWO has fully switched routing light away from TXCSW4 at the beginning of P4 mitigating transient crosstalk through TXCSW4. Switch TXSW6 of the compound switch TXCSW6 makes its slow transition from T to 0 one pixel duration later, after the end of PO, while the SS6 switches of the compound switch TXCSW6 are fast switched from HC to T at the beginning of PO and held to maintain the T state for TXCSW6. This delays TXCSW6’s slow transition until after TXSWO has fully switched routing light away from TXCSW6 at the beginning of PO mitigating transient crosstalk through TXCSW6.

[0105] Also illustrated in FIG. 12B, sometimes switch state extension by a single switch which is not in any compound switch can lead to transient crosstalk mitigation of switches at lower levels. For example, switch TXSW1 has its slow transition delayed until the beginning of P5, shielding the slow transition of TXCSW3 during P4 from light TXSW1 would begin to route towards TXCSW3 once TXSW1 turns off. Similarly, TXSW2 has its slow transition delayed until the beginning of Pl , shielding the slow transition of TXCSW5 during PO from light TXSW2 would begin to route towards TXCSW5 once TXSW2 turns off.

[0106] It is to be understood, that the component parts of the LiDAR systems described hereinabove, the receivers, laser sources, controllers, switch trees etc., may operate as part of a single instrument or device or may operate as part of a multiplicity of interconnected devices working together in proximity or remotely, or any combination thereof.

[0107] The above described control of the various optical switches of the optical switch tree by a controller may be performed by a processing device such as a micro- processor / FPGA based controller or any one or more other similar device, which may be implemented using one or more application specific integrated circuits (ASIC), microcontrollers, general purpose computer systems, digital signal processors, programmable logic devices (PLD), field programmable logic devices (FPLD), and the like, programmed according to the teachings as illustrated and described herein, as will be appreciated by those skilled in the optical, networking, software and computing arts.

[0108] In addition, two or more computing systems or devices may be substituted for any one of the processors or controllers described herein. Accordingly, principles and advantages of distributed processing, such as redundancy, replication, and the like, e.g. a separate controller per channel of the LIDAR system, also can be implemented, as desired, to increase the robustness and performance of processors or controllers described herein. The operation of the example control of the switches in accordance with predetermined timings such as that illustrated in the timing diagrams may be performed in accordance with machine readable instructions. In these examples, the machine readable instructions comprise an algorithm for execution by: (a) a processor, (b) a controller, and / or (c) one or more other suitable processing device(s). The algorithm may be embodied in software stored on tangible media such as, for example, a flash memory, a CD-ROM, a floppy disk, a hard drive, a digital video (versatile) disk (DVD), or other memory devices, but persons of ordinary skill in the art will readily appreciate that the entire algorithm and / or parts thereof could alternatively be executed by a device other than a processor and / or embodied in firmware or dedicated hardware in a well-known manner (e.g., it may be implemented by an application specific integrated circuit (ASIC), a programmable logic device (PLD), a field programmable logic device (FPLD), discrete logic, etc.). For example, any or all of the component processes to control the timing of the switch trees could be implemented by software, hardware, and / or firmware.

[0109] While particular implementations and applications of the present disclosure have been illustrated and described, it is to be understood that the present disclosure is not limited to the precise construction and compositions disclosed herein and that various modifications, changes, and variations can be apparent from the foregoing descriptions without departing from the spirit and scope of an invention as defined in the appended claims.

Claims

WHAT IS CLAIMED IS:

1. A LIDAR system comprising: a laser source for generating optical signals; a plurality of pixels arranged in an array; a receiver for detecting optical signals scattered back from a target; a plurality of optical switches arranged in an optical switch tree for guiding the optical signals from the laser source sequentially to each pixel of the pixel array for launching the optical signals to the target and for guiding the optical signals scattered back from the target from the pixel to the receiver, a branch of the optical switch tree comprising a first optical switch and one or more second optical switches; and a controller for controlling the switch tree to guide said optical signals with use of control signals wherein, while the branch is unselected, the controller controls the first optical switch and the one or more second optical switches with the same OFF control signal, such that a first optical signal traversing a cross port of the first optical switch and through said one or more second optical switches traverses a same number of closed ports of said first and one or more second optical switches as a second optical signal traversing a through port of the first optical switch and through said one or more second optical switches.

2. The LIDAR system of claim 1, wherein the OFF control signal is a control voltage of OV.

3. The LIDAR system of claim 1, wherein the controller begins said control of the first optical switch and the one or more second optical switches with the same OFF control signal once optical signals from the laser source have been routed away from said branch.

4. The LIDAR system of claim 1, wherein the controller controls one or more optical switches at a higher level in the optical switch tree than the first optical switch and the oneor more second optical switches to route optical signals from the laser source away from said branch until a time after the controller controls the first optical switch and the one or more second optical switches with the same OFF control signal sufficient for the first optical switch and the one or more second optical switches to reach steady state operation.

5. The LIDAR system of claim 1, wherein the first optical switch is located within one of a transmit switch tree or a receive switch tree of the optical switch tree, and the one or more second optical switches is located within the other one of the transmit switch tree or the receive switch tree, and wherein the one or more second optical switches comprises a second optical switch which is coordinate to the first optical switch, such that said first optical signal traversing the cross port of the first optical switch is guided through the optical switch tree to traverse a through port of the second optical switch, and said second optical signal traversing the through port of the first optical switch is guided through the optical switch tree to traverse a cross port of the second optical switch.

6. The LIDAR system of claim 5, wherein the first optical switch and the second optical switch are last level switches in the transmit and receive switch trees.

7. The LIDAR system of claim 5, wherein the controller begins said control of the first optical switch and the one or more second optical switches with the same OFF control signal once optical signals from the laser source have been routed away from said branch.

8. The LIDAR system of claim 5, wherein the controller controls one or more optical switches at a higher level in the optical switch tree than the first optical switch and the one or more second optical switches to route optical signals from the laser source away from said branch until a time after the controller controls the first optical switch and the one or more second optical switches with the same OFF control signal sufficient for the first optical switch and the one or more second optical switches to reach steady state operation.

9. The LIDAR system of claim 1, wherein the one or more second optical switches comprise a first select switch and a second select switch, wherein the first optical switch, first select switch, and the second select switch are comprised within a compound optical switch occupying a switch position within one of a transmit switch tree or a receive switch tree of the optical switch tree, and wherein the first select switch is coupled to the cross port of the first optical switch and the second select switch is coupled to the through port of the first optical switch, such that said first optical signal traversing the cross port of the first optical switch traverses a through port of the first select switch, and said second optical signal traversing the through port of the first optical switch traverses a cross port of the second select switch.

10. The LIDAR system of claim 9, wherein the controller begins said control of the first optical switch and the one or more second optical switches with the same OFF control signal once optical signals from the laser source have been routed away from said branch.

11. The LIDAR system of claim 9, wherein the controller controls one or more optical switches at a higher level in the optical switch tree than the first optical switch and the one or more second optical switches to route optical signals from the laser source away from said branch until a time after the controller controls the first optical switch and the one or more second optical switches with the same OFF control signal sufficient for the first optical switch and the one or more second optical switches to reach steady state operation.

12. The LIDAR system of claim 9, wherein the compound optical switch occupies a last level switch position in one of the transmit and receive switch trees.

13. The LIDAR system of claim 1, the optical switch tree comprises a plurality of branches including said branch, each branch comprising a respective first optical switch and respective one or more second optical switches, wherein while any branch of the plurality of branches is unselected, the controller controls the respective first optical switch and therespective one or more second optical switches of the unselected branch with the same OFF control signal, such that a respective first optical signal traversing a cross port of the respective first optical switch and through said respective one or more second optical switches traverses a same number of closed ports of said respective first and one or more second optical switches as a respective second optical signal traversing a through port of the respective first optical switch and through said respective one or more second optical switches14. The LIDAR system of claim 13, wherein the respective first optical switches are located within one of a transmit switch tree or a receive switch tree of the optical switch tree, and the respective one or more second optical switches are located within the other one of the transmit switch tree or the receive switch tree, and wherein each of the respective one or more second optical switches comprises a respective second optical switch which is coordinate to one of the respective first optical switches, such that said respective first optical signal traversing the cross port of the respective first optical switch is guided through the optical switch tree to traverse a through port of the respective second optical switch, and said respective second optical signal traversing the through port of the respective first optical switch is guided through the optical switch tree to traverse a cross port of the respective second optical switch.

15. The LIDAR system of claim 14, wherein the controller begins said control of each respective first optical switch and the respective one or more second optical switches of a branch with the same OFF control signal once optical signals from the laser source have been routed away from said branch.

16. The LIDAR system of claim 14, wherein the controller controls one or more optical switches at a higher level in the optical switch tree than the respective first optical switch and the respective one or more second optical switches of a branch to route optical signals from the laser source away from said branch until a time after the controller controls therespective first optical switch and the respective one or more second optical switches with the same OFF control signal sufficient for the respective first optical switch and the respective one or more second optical switches to reach steady state operation.

17. The LIDAR system of claim 14, wherein the respective first optical switches and the respective second optical switches are last level switches in the transmit and receive switch trees.

18. The LIDAR system of claim 13, wherein each respective one or more second optical switch comprises a respective first select switch and a respective second select switch, wherein each respective first optical switch along with its respective first select switch and respective second select switch are comprised within a respective compound optical switch occupying a respective switch position within one of a transmit switch tree or a receive switch tree of the optical switch tree, and wherein each respective first select switch is coupled to the cross port of the respective first optical switch and each respective second select switch is coupled to the through port of the respective first optical switch, such that said respective first optical signal traversing the cross port of the respective first optical switch traverses a through port of the respective first select switch, and said respective second optical signal traversing the through port of the respective first optical switch traverses a cross port of the respective second select switch.

19. The LIDAR system of claim 18, wherein the controller begins said control of each respective first optical switch and the respective one or more second optical switches of a branch with the same OFF control signal once optical signals from the laser source have been routed away from said branch.

20. The LIDAR system of claim 18, wherein the controller controls one or more optical switches at a higher level in the optical switch tree than the respective first optical switch and the respective one or more second optical switches of a branch to route optical signalsfrorn the laser source away from said branch until a time after the controller controls the respective first optical switch and the respective one or more second optical switches with the same OFF control signal sufficient for the respective first optical switch and the respective one or more second optical switches to reach steady state operation.

21. The LIDAR system of claim 18, wherein each respective compound optical switch occupies a respective last level switch position in one of the transmit and receive switch trees.

Images