Systems and methods for mitigating position sensitive errors in photonic integrated circuit testing

Abstract

Methods and systems for mitigating position sensitive optical measurement errors during photonic integrated circuit (PIC) testing are disclosed. The system includes a light source, an optical probe, an optical detector, and an alignment stage. The method includes adjusting a gap between the probe and a PIC to N ≥2 different gap settings with the alignment stage, the gap settings corresponding to probe-to-PIC distances selected to produce N number of round-trip phase shifts within an optical cavity formed in the gap, the phase shifts being substantially evenly spaced within a full 2 π cycles. For each gap setting, a test measurement is performed, the test measurement including injecting test light from the light source into the PIC via the probe, collecting light output from the PIC via the probe, and detecting the collected light with the detector to generate a measurement result. N The number of measurement results are combined into a corrected measurement result having reduced cavity-induced errors.

Description

Technical Field

[0001] The technical field relates to the testing of photonic integrated circuits (PICs), and more specifically to methods and systems for mitigating position-sensitive optical measurement errors in PIC testing. Background Technology

[0002] Photonic integrated circuits (PICs) are microchips that integrate multiple photonic components onto a common substrate for processing and transmitting information. As PICs play an increasingly important role in optical communication systems, the demand for accurate and high-throughput testing methods is also growing. Automated test stations are typically used to characterize PIC performance by injecting light into and collecting light from the PIC under test using optical probes (e.g., fiber arrays) and integrated optical couplers. Two main coupling methods are used in PIC testing: surface coupling, where light is introduced out-of-plane through the top surface of the PIC; and edge coupling, where light is introduced in-plane through the sides of the PIC. Edge coupling is generally preferred for industrial and high-performance applications due to its lower insertion loss and reduced sensitivity to polarization and wavelength variations. However, edge coupling presents alignment challenges, especially in wafer-level or high-throughput environments where consistent probe positioning is difficult. Summary of the Invention

[0003] This specification relates to systems and methods designed to mitigate measurement errors caused by position-sensitive cavity-induced interference effects during PIC testing. The disclosed techniques involve performing multiple test measurements at different probe-to-PIC distances, which are selected to produce phase-opposite interference patterns. These interference patterns are then combined to sufficiently attenuate cavity-induced artifacts in the measurement signal, thereby reducing the sensitivity of the test results to variations in the probe-to-PIC gap. The disclosed techniques allow for repeatable and accurate characterization of PIC parameters, such as insertion loss, return loss, and wavelength-dependent transmission or reflection spectra, including under high-throughput or wafer-level testing conditions.

[0004] According to one aspect, a method for mitigating position-sensitive errors during testing of photonic integrated circuits (PICs) is provided, the method comprising: We provide test systems that include light sources, optical probes, optical detectors, and alignment stages. Use the alignment stage to adjust the gap between the optical probe and the PIC. N There are different gap settings, among which N ≥2, the gap setting corresponds to the corresponding probe-to-PIC distance, which is selected to produce light propagation within the optical cavity formed in the gap. NOne round-trip optical phase shift, the round-trip optical phase shift in the complete 2 π The cyclic internal phases are basically evenly spaced; for N For each of the different gap settings, perform test measurements, which include: Test light from the light source is injected into the PIC via an optical probe; The measurement light output from the PIC is collected via an optical probe; and An optical detector is used to detect the collected measurement light to generate measurement results; and The group comes from N The measurement results of each test measurement are used to generate corrected measurement results with reduced position-sensitive errors induced by the optical cavity.

[0005] In some implementations, the method includes coupling the optical probe to the PIC in an edge-coupled configuration.

[0006] In some implementations, the method includes providing an optical probe as a fiber optic array.

[0007] In some implementation schemes, N =2, and adjusting the gap between the optical probe and the PIC includes setting the distance between the two probes and the PIC to approximately one-quarter of the wavelength of the test light at the design wavelength.

[0008] In some implementations, the measurement results correspond to... N The insertion loss spectrum of the PIC was measured under various gap settings. In some embodiments, the insertion loss spectrum was obtained in a band within the spectral range of approximately 1260 nm to approximately 1680 nm.

[0009] In some implementations, the light source is a tunable laser.

[0010] In some implementations, the method utilizes a PIC provided in a wafer-level or multi-die test configuration.

[0011] In some implementations, adjusting the gap between the optical probe and the PIC involves moving the optical probe relative to the PIC.

[0012] In some implementations, adjusting the gap between the optical probe and the PIC involves moving the PIC relative to the optical probe.

[0013] In some implementations, the combined measurement results include calculations from... N The average of the measurement results of each test.

[0014] According to another aspect, a test system for testing photonic integrated circuits (PICs) with reduced position-sensitive errors is provided, the test system comprising: A light source configured to generate test light; An optical probe configured to couple test light from a light source into the PIC and collect measurement light output from the PIC; An optical detector configured to detect the collected measurement light; Alignment stage, which is configured to be in a series of N In each test measurement, the gap between the optical probe and the PIC was adjusted to... N There are different gap settings, among which N ≥2, the gap setting corresponds to the corresponding probe-to-PIC distance, which is selected to produce light propagation within the optical cavity formed in the gap. N One round-trip optical phase shift, the round-trip optical phase shift in the complete 2 π The cyclic intrinsic phases are basically evenly spaced; and A control and processing unit, coupled to an optical detector, is configured to: for N For each of the test measurements, the corresponding measurement result is received from the optical detector; and combination N The measurement results are used to generate corrected measurement results with reduced position-sensitive errors induced by the optical cavity.

[0015] In some implementations, the optical probe is configured to be coupled to the PIC in an edge-coupled configuration.

[0016] In some implementations, the optical probe includes a fiber optic array.

[0017] In some implementation schemes, N =2, and the alignment stage is configured to set the distance between the two probes and the PIC to approximately one-quarter of the wavelength of the test light at the design wavelength.

[0018] In some implementations, the measurement results correspond to... N The insertion loss spectrum of the PIC is measured under a gap setting. In some embodiments, the optical detector is configured to acquire the insertion loss spectrum in a band within the spectral range of about 1260 nm to about 1680 nm.

[0019] In some implementations, the light source includes a tunable laser.

[0020] In some implementations, the alignment stage is configured to adjust the gap between the optical probe and the PIC by moving the optical probe relative to the PIC.

[0021] In some implementations, the alignment stage is configured to adjust the gap between the optical probe and the PIC by moving the PIC relative to the optical probe.

[0022] In some implementations, the control and processing unit is configured to... N The operation of the light source and alignment stage is automatically controlled during each test and measurement.

[0023] In some implementations, the control and processing unit is configured to calculate from N The average of the measurement results of each test measurement is used to generate the corrected measurement result.

[0024] In some implementations, the test system is configured to test PICs in a wafer-level or multi-die test configuration.

[0025] Other method and process steps may be performed before, during, or after the steps described herein. The order of one or more steps may also differ, and some steps may be omitted, repeated, or combined, as appropriate. It will also be noted that some steps may be performed using various analysis and processing techniques, which may be implemented in hardware, software, firmware, or any combination thereof.

[0026] Other objects, features, and advantages of the present technology will become more apparent after reading the following non-limiting description of specific embodiments of the technology given by way of example only with reference to the accompanying drawings. Although the specific features described in the above summary of the invention and the following detailed description may be described with respect to specific embodiments or aspects, it should be noted that these specific features may be combined with each other unless otherwise stated. Attached Figure Description

[0027] Figure 1 , Figure 2 and Figures 3A to 3C This document describes various aspects, features, and implementations of the technology disclosed herein, or various aspects, features, and implementations related to the technology disclosed herein.

[0028] Figure 1 This is a schematic diagram of a test system for testing photonic integrated circuits (PICs) according to one implementation scheme.

[0029] Figure 2 This is a flowchart of a method for mitigating position-sensitive optical measurement errors during PIC testing, according to another embodiment.

[0030] Figure 3AFour insertion loss spectra were obtained using conventional single-point measurements at different probe-to-PIC distances. Figure 3B Four insertion loss spectra are shown at different probe-to-PIC distances. Each insertion loss spectrum was obtained by averaging a pair of single-point spectra measured using complementary phase shifts. Figure 3C It shows the corresponding Figure 3A and Figure 3B The spectral change envelope of the spectrum. Detailed Implementation

[0031] This disclosure describes methods and systems for mitigating position-sensitive errors during testing of photonic integrated circuits (PICs). The disclosed techniques are designed to reduce the effects of Fabry-Perot (FP) cavities formed between the test probe and the PIC during edge-coupled measurements.

[0032] In a typical test configuration, an optical probe (such as a fiber array) is aligned with the edge coupler of the photoelectric sensor (PIC) to facilitate bidirectional optical access. A light source coupled to the optical probe injects test light into the PIC via the input edge coupler. The measurement light exiting the PIC via the output edge coupler is collected by the optical probe and guided to an optical detector for analysis. Due to the separation between the optical probe and the PIC, and the partially reflective nature of the interface at both ends of the resulting gap, an FP cavity is formed. This cavity induces interference artifacts in the measured optical signal, such as periodic spectral ripples, which introduce systematic errors into the test results.

[0033] Traditional testing methods, such as loopback measurements, offer limited mitigation of these cavity-induced effects. Loopback techniques rely on a reference path and assume consistent insertion loss across the edge coupler, but they do not address the potential sensitivity to the probe-to-PIC gap. While refractive index matching materials (e.g., gels, glues, or oils) can reduce these interference effects, their use is impractical in high-throughput settings such as wafer-level or multi-die testing. Despite the application of such conventional mitigation techniques, experimental results obtained by the inventors indicate a persistent residual measurement error of approximately a few tenths of a dB. These errors have been observed to be highly sensitive to the alignment of the individual fibers within the probe. Because the errors are caused by interference effects, even subwavelength variations in the probe-to-PIC gap (e.g., approximately 100 nm) can produce measurable fluctuations in coupling efficiency, leading to undesirable sensitivity to probe positioning and reduced measurement repeatability.

[0034] To address these issues, this disclosure introduces complementary methods. N Step measurement technology, designed to mitigate measurement errors caused by FP interference in PIC testing. This technology involves a series of... NThe gap between the probe and the PIC is systematically adjusted at different values ​​to reduce the sensitivity of optical test results to minute changes in the probe-PIC spacing. N It is an integer equal to or greater than two. These phase shifts are chosen to be in full 2. π The intervals within the loop are substantially uniform. When appropriately combined, for example by averaging or weighted summation, these measurements help mitigate the periodic modulation effects introduced by the cavity, thereby reducing spectral ripple artifacts in the resulting test signal.

[0035] In the simplest case, when N When the phase difference is 2, the gap is set to two values ​​that are approximately one-quarter or odd multiples of the free-space wavelength of the phase difference test light. This introduces a back-and-forth interference signal between the two measurements. π Phase difference. Since the resulting cavity-induced oscillations are almost opposite in phase, their combination essentially cancels out the oscillatory components, resulting in a more accurate signal that better represents the optical response of the PIC.

[0036] Each test cycle is performed by injecting test light from the light source into the PIC via an optical probe, allowing the test light to propagate through the photonic circuitry system, and collecting the resulting output light for detection and analysis. With all other test conditions kept constant, [the test cycle is performed] for [specific purposes]. N Repeat this process for each gap setting in the individual gap settings. The result is... N The individual measurements (each exhibiting a different interference pattern due to cavity effects) are then combined to produce a corrected measurement output with reduced probe spacing artifacts.

[0037] The disclosed techniques offer several benefits. These techniques enable accurate PIC characterization without requiring a refractive index matching medium and are well-suited for wafer-level, multi-die, and other high-throughput testing environments. They can be integrated into various test platforms and are suitable for a wide range of optical testing scenarios. By reducing measurement variability associated with small variations in probe-to-PIC spacing, the disclosed techniques improve repeatability and reliability. In some implementations, these techniques can also aid in optical alignment by improving consistency in coupling conditions. Measurements may include optical insertion loss, return loss, and transmission or reflection spectra. In some implementations, single-valued parameters such as average transmitted power or peak wavelength response can also be measured.

[0038] The following description, with reference to the accompanying drawings, outlines additional aspects and example implementations.

[0039] refer to Figure 1An embodiment of a test system 100 for characterizing a PIC 102 is illustrated. The test system 100 includes a light source 104, an optical probe 106, an optical detector 108, an alignment stage 110, and a control and processing unit 112. In some implementations, the test system 100 may be provided as a fully integrated, standalone PIC test instrument. In other cases, the individual components may be manufactured and sold separately, allowing for modular assembly into a complete system. Hybrid configurations are also possible, where only some components are integrated while others are supplied separately.

[0040] Further details regarding the structure, configuration, and operation of these components are provided below. It should be noted that... Figure 1 This is a schematic diagram intended to illustrate the functional and structural aspects of the test system 100. For clarity, certain auxiliary features or actual implementation details, such as those found in a complete test station, may be simplified or omitted.

[0041] In the illustrated embodiment, test system 100 is configured for edge coupling to PIC 102. While the disclosed techniques may be particularly suitable for edge coupling configurations, they can also be applied to certain surface coupling configurations where cavity-induced interference effects may occur. One example includes surface coupling via a normal-incident surface coupler (e.g., a grating coupler). Figure 1 In the illustrated edge coupling configuration, PIC 102 includes: an input edge coupler 114 that receives test light from light source 104 via optical probe 106; and an output edge coupler 116 that also transmits measurement light from PIC 102 to optical detector 108 via optical probe 106. Edge couplers 114 and 116 can take various forms, including tapered edge couplers, inverted tapered edge couplers, metamaterial-based edge couplers, multilayer edge couplers, suspended edge couplers, or vertical edge couplers, each offering specific benefits. These structures are generally known in the art and do not require detailed description unless necessary for understanding the present technology.

[0042] PIC 102 also includes a set of interconnected photonic components 118 forming a functional circuit system for the device. These components 118 are optically coupled between input edge coupler 114 and output edge coupler 116, and may include passive and active optical elements, such as waveguides, beam splitters, combiners, modulators, interferometers, switches, and on-chip sources and detectors. Together, these components define the behavior and function of PIC 102 during testing. PIC 102 can be implemented on various material platforms, including silicon-on-insulator, indium phosphide, silicon nitride, lithium niobate, and gallium arsenide. Depending on the manufacturing stage or testing purpose, test system 100 can be configured to characterize various PIC form factors, including complete wafers, wafer strips, and single-die or multi-die configurations. Because the general structure, materials, and manufacturing methods of PICs are well-established, a detailed description of the internal layout of PIC 102 is omitted unless necessary for understanding this technology.

[0043] The light source 104 can be embodied by any suitable device or combination of devices (including both narrowband and broadband sources) capable of generating test light suitable for PIC characterization. In some embodiments, the test light may have a bandwidth in the band ranging from about 1260 nm to about 1680 nm (e.g., covering the O, E, S, C, L, and U telecommunication wavelength bands in the short-wavelength infrared). However, this spectral range is exemplary and not limiting, as the disclosed techniques are also applicable to the visible, near-infrared, and mid-infrared spectra.

[0044] In some embodiments, light source 104 may be a laser source, such as a high-power continuous-wave tunable laser source with narrow linewidth and high-resolution tunability. A tunable laser source allows for continuous adjustment of the laser wavelength within a range, thereby enabling testing of PIC 102 at various operating wavelengths. Non-limiting examples of laser sources include solid-state lasers (e.g., bulk crystal lasers and fiber lasers) and semiconductor lasers (e.g., laser diodes). However, non-laser sources are also possible, including light-emitting diode (LED) sources, amplified spontaneous emission (ASE) sources, and semiconductor optical amplifier (SOA) sources. Light source 104 may also include or be used with adjustment optics (e.g., filters, lenses, or amplifiers) configured to adjust the light from the source before it is delivered to PIC 102.

[0045] The optical probe 106 may include any suitable coupling device or combination of devices capable of establishing bidirectional optical communication with the PIC 102. In some embodiments, the probe 106 includes an array of optical fibers configured to align with the edge couplers 114, 116 of the PIC. Depending on the PIC configuration or test setup, the probe's fiber array may include any suitable number of individual optical fibers 1361, 1362 (e.g., 2, 4, 8, or 16 fibers). For example, in Figure 1 In this configuration, one optical fiber 1361 can be aligned with the input edge coupler 114 of the PIC 102 to transmit test light 138 from the light source 104, while another optical fiber 1362 can be aligned with the output edge coupler 116 to collect measurement light 140 and guide it to the optical detector 108. Each optical fiber 1361, 1362 can be single-mode or multimode. Parameters such as fiber composition and geometry, refractive index distribution, numerical aperture, polarization control, and spectral compatibility can be customized based on specific application requirements.

[0046] While fiber optic arrays are typically used in PIC testing, other types of optical waveguides or couplers (e.g., lithographic waveguides or freeform waveguides) can also be employed in the optical probe 106, depending on the test platform and coupling interface. In some setups, probe 106 may be implemented as a single unit for both injection and collection. In other cases, probe 106 may include separate injection and collection components, with the former coupled between the light source 104 and the input edge coupler 114, and the latter coupled between the output edge coupler 116 and the optical detector 108.

[0047] Optical detector 108 may include any suitable photodetector or combination of photodetectors capable of measuring optical power over the spectral range of measurement light 140 received from PIC 102. Suitable technologies include PIN photodiodes, avalanche photodiodes, and Schottky photodiodes. Optical detector 108 converts the received light 140 into an electrical signal representing its optical properties. Detector selection may be based on parameters such as power level, spectral range, sensitivity, and response time. In some embodiments, optical detector 108 may be a non-spectrally resolved detector (e.g., an optical power meter) that provides a single-value measurement averaged over either the spectral bandwidth of the measurement light 140 or the sensitivity range of the detector (whichever is narrower). In such cases, the spectral response of PIC 102 can still be reconstructed by changing the wavelength of the test light (e.g., using a tunable narrowband light source) and recording the resulting optical power at each wavelength. In other embodiments, optical detector 108 may be a spectrally resolved detector capable of directly measuring the wavelength-dependent response. Such detectors may include an OSA spectrometer, a scanning wavelength system, or other wavelength-selective devices. These detectors can be used in conjunction with broadband sources, such as superradiative LEDs (SLEDs) or ASE sources, to obtain the spectral-resolved response of the PIC 102.

[0048] Alignment stage 110 is configured to enable precise relative positioning between PIC 102 and optical probe 106, thereby ensuring accurate optical coupling and alignment during testing. Depending on the application, alignment stage 110 includes: PIC positioning unit 120 configured to adjust the position of PIC 102 while probe 106 is stationary; probe positioning unit 122 configured to adjust the position of probe 106 while PIC 102 is stationary; or both PIC positioning unit 120 and probe positioning unit 122 to allow coordinated adjustment. Each unit 120, 122 can be independently actuated (e.g., using a precision motor or other actuator) and can provide both translational positioning (e.g., along one, two, or three axes) and rotational positioning (e.g., about one, two, or three axes). For example, PIC positioning unit 120 can provide four degrees of freedom: three translations and one rotation about an axis perpendicular to the plane of the PIC, while probe positioning unit 122 can provide six degrees of freedom. Step size resolution can vary depending on the stage design, but is typically submicron level (e.g., 100 nm or less). Common actuation technologies include piezoelectric actuators, stepper motors, and voice coil drivers. The alignment stage 110 can be adapted to various PIC form factors and can support manual or automatic alignment procedures. Alignment stages suitable for PIC testing are well known in the art and do not require further detailed description unless necessary for understanding this technology.

[0049] The control and processing unit 112 is operatively coupled to various components of the test system 100, including the light source 104, the optical detector 108, and the alignment stage 110. The control and processing unit 112 includes one or more processors 124 and one or more memories 126 implemented in hardware, software, firmware, or any combination thereof. The control and processing unit 112 may be fully or partially integrated with other system components (e.g., the optical detector 108, in its primary signal processing configuration) or implemented as a physically independent module.

[0050] In addition to signal processing (e.g., converting electrical signals from optical detector 108 into test results), control and processing unit 112 can coordinate hardware operations, including the positioning of stage 110 and the control of light source 104. This allows for full or partial automation of test operations. Control and processing unit 112 can support wired or wireless communication with system components and can be operated via direct user input and / or commands. In some implementations, control and processing unit 112 can interact with a user interface including input / output devices (e.g., touchscreen, buttons, displays) and can optionally utilize distributed or cloud-based computing resources for data processing or storage.

[0051] Still referencing Figure 1 During edge-coupled testing, the optical configuration creates an FP cavity 128 between the optical probe 106 and the PIC 102. This cavity 128 is defined by two partially reflective interfaces: a probe-side interface 130 (e.g., an fiber endface) and a PIC-side interface 132 (e.g., a PIC endface). The intermediate region (typically air) between these interfaces 130, 132 defines a cavity length 134, also referred to herein as the measurement gap or alignment gap. Multiple internal reflections within this region produce a resonant interference effect, thereby modulating the effective transmission through the cavity 128 as a function of both wavelength and cavity length. The resulting spectral response, as measured by the optical detector 108, exhibits a periodic constructive and destructive interference pattern dominated by the round-trip optical phase delay within the cavity. Therefore, insertion loss measurements can be affected by such cavity-induced effects, leading to system position-sensitive errors during PIC testing.

[0052] Transmission function of the FP cavity formed by two parallel interfaces T It can be described by the following equation:

[0053] in T 1. R 1 represents the transmission and reflection coefficients at 130° on the probe side interface. T 2. R 2 represents the transmittance and reflectance at the PIC side interface 132, and δ This is the round-trip optical phase shift experienced by the light traveling through cavity 128. For normal incidence, this phase shift is given by the following equation:

[0054] in n It is the refractive index of the medium in the gap (for air, it is usually...). n ≈1), d It is the cavity length, and λ It is the wavelength in free space.

[0055] To assess the impact of cavity-induced error, the transmission function in equation (1) can be expanded to revolve around cos δ Taylor series with =0:

[0056] For illustration, assume that both interfaces 130 and 132 are air-silica boundaries, where... n 空气 ≈1 and n 二氧化硅 ≈1.44, the reflection coefficient at normal incidence becomes:

[0057] This leads to T 1= T 2≈0.967. Substituting these values ​​into equation (3), the transmission function becomes:

[0058] The first-order term introduces periodic modulation into the transmission function, with an amplitude as high as 10log(1+0.065) ≈ 0.27 dB. This modulation is not negligible because it can significantly affect insertion loss measurements under typical test conditions. Furthermore, quantifying or predicting this error is challenging without precise knowledge of the probe-to-PIC gap at both the injection and collection ports on a fraction of half the wavelength. Meanwhile, the second-order term contributes as high as 10log(1+0.004) ≈ 0.017 dB, which is generally negligible in comparison. Therefore, this technique focuses on suppressing this dominant first-order term.

[0059] This inhibition uses the method described earlier. N This is achieved through step measurement technology. For illustration, consider... N The case is =2. Perform two test measurements such that the associated round-trip optical phase shift is equal to 2. δ and δ 'difference π an odd multiple of, that is to say

[0060] According to equation (2), this phase shift condition is transformed into the gap change Δ given by the following equation. d :

[0061] in λ 设计 The wavelength selected is a representative design wavelength within the operating spectrum of the PIC 102 under test (e.g., near the center of the operating spectrum for balanced suppression across the band). In this case, the transmission functions of the two measurements are:

[0062] The average of these two results is as follows:

[0063] Therefore, the first-order terms are canceled out, and only the higher-order terms remain, which are much smaller. This averaging essentially suppresses the dominant cavity-induced modulation in the transmission function.

[0064] By selecting the generation in the complete 2π The loop is basically evenly spaced. N This method, which measures the distance from the probe to the PIC by measuring the round-trip optical phase shift, can be extended to... N The case is greater than 2. For example, in... N When the phase shift is 3, the phase shift will be spaced by 2π / 3 to provide complementary interference conditions, while... N When =4, the phase shift will be spaced π / 2. When N When test measurements are appropriately combined, first-order error terms can be effectively suppressed. As used herein, "substantially uniformly spaced" means that the phase spacing is sufficiently uniform to ensure that the first-order error term is effectively suppressed. N The contribution of each test measurement cancels out or averages the cavity-induced modulation, making any residual error negligible, practically insignificant, or reduced to a level sufficient to produce a significant improvement in measurement accuracy.

[0065] Note that the chosen distance value does not require a minimum interval, because the round-trip optical phase is 2... π The modulus is periodic, thus allowing equivalent phase shifts to be obtained at multiple absolute gap values. This flexibility enables the crossing of... N The measurement uses a large gap difference. For example, for N A minimum quarter-wavelength gap difference of 2 may not be achievable in some setups, but larger odd multiples of the gap (such as 3) are more likely to be successful. λ / 4 or 5 λ / 4) will still meet the phase requirements.

[0066] It should also be noted that the first-order term only applies at the design wavelength. λ 设计 At this wavelength, the error is completely suppressed. However, even at wavelengths deviating from this range, residual error still exists, and for... N =2. Features air gap and quarter-wavelength gap variations λ 设计 In the case of / 4, the residual error can be expressed as:

[0067] The residual first-order error is in λ 设计 = λ It disappears over time, but usually remains small for representative values. For example, for a spectrum extending from 1505 nm to 1675 nm, where λ 设计 =1600nm (as in the experimental results reported below), the maximum residual error reached at the spectral endpoint is approximately 0.05dB.

[0068] Furthermore, by combining multiple measurements with phase-complementary conditions, cavity effects can be reduced not only in the PIC-probe gap but also in other sections of the optical test path that may form unintended FP cavities. This includes reflections between fiber connections, fiber ends, and photonic waveguide elements. Because these parasitic cavities also produce interference patterns sensitive to minute positional changes, applying the phase-shift measurement techniques disclosed herein can help reduce their effects.

[0069] Now refer to Figure 2 To describe the operational aspects of this technology in more detail, Figure 2 A flowchart is depicted for a method 200 for mitigating position-sensitive errors during PIC testing.

[0070] Method 200 includes step 202: providing a test system comprising a light source, an optical probe, an optical detector, and an alignment stage. For example, these components may correspond to... Figure 1 The test system 100 shown may be a component of any other suitable system configured for optical testing of a PIC. The test system is configured to enable optical coupling between the light source and the PIC via an optical probe and to collect measurement light from the PIC for detection by an optical detector.

[0071] The method further includes step 204: adjusting the measurement gap between the optical probe and the PIC using an alignment stage. N Several different gap settings. These gap settings correspond to the respective probe-to-PIC distances, which are selected to produce light propagation within the FP cavity formed in the gaps. N Two round-trip optical phase shifts, where the round-trip optical phase shifts are in the complete 2 π The cycles are essentially uniformly spaced in terms of phase. Depending on the implementation, the probe-to-PIC distance between measurements can be adjusted by moving the probe relative to the PIC, by moving the PIC relative to the probe, or by moving both. N =2, which can be achieved by designing the wavelength. λ 设计 The cavity length is adjusted to an odd multiple of one-quarter, resulting in phase shift and phase difference. π Based on settings or time constraints, it can be used in other implementation schemes. N The higher values ​​(e.g., 3 or 4) and the corresponding values ​​for cavity length offset (e.g., for air gaps, λ 设计 / 6 or λ 设计 / 8). Cavity length adjustment can be automated and performed under software control to ensure consistency and repeatability.

[0072] In each gap setting, method 200 includes a step 206 of performing a test measurement comprising: injecting test light from a light source 208 into the PIC via an optical probe; collecting measurement light output from the PIC via an optical probe 210; and detecting the collected measurement light using an optical detector 212 to generate a measurement result. In practice, this involves the light source emitting test light, which is then passed to the input edge coupler of the PIC. The measurement light leaving the PIC is collected by the probe and guided to the optical detector. In addition to the probe-to-PIC gap, for... N Each test measurement in the test series maintains the same test conditions. This ensures that variations in the measurement results are primarily attributable to phase-sensitive effects within the cavity, rather than changes in the equipment or test setup.

[0073] Method 200 continues to step 214: Combining from N The measurement results of each test measurement are used to generate corrected measurement results with reduced position-sensitive errors induced by the optical cavity. N When the result is 2, the results from a pair of measurements are averaged to eliminate the first-order FP interferometry term. For larger values... N The measurement results are combined using a suitable averaging or weighted summation procedure, which enables destructive interference of cavity-induced modulation for cross-phase-interval measurements.

[0074] Experimental studies were conducted to evaluate and validate different aspects of this technology. It should be understood that the following description presents specific implementations of the basic principles and is not intended to be limiting, as the disclosed technology allows for many variations in implementation and application.

[0075] A series of insertion loss measurements were performed on the PIC by scanning the tunable laser within a 170 nm spectral range (from 1505 nm to 1675 nm). Between each measurement, the fiber array probe was shifted 100 nm along its optical axis to change the FP cavity length between the probe and the PIC. This resulted in a set of 20 measurements spanning a total range of 2 µm. Figure 3A Four representative insertion loss spectra obtained under these conditions are presented. T 0、 T 1. T 2. T 3). Despite the application of a Gaussian filter to suppress high-frequency noise, significant fluctuations remain in the raw data, primarily attributed to polarization-dependent variations in the automatic optical switches feeding the test light into the fiber array. Despite this residual noise, modulation induced by the FP cavity is clearly visible across the measurements (see inset, where the spectrum is vertically shifted from top to bottom). T 0、 T 1.T 2 and T 3) The spectral variation of insertion loss reached as high as 0.3 dB. The fact that these four spectra, acquired under nominally identical conditions except for the probe-to-PIC interval, exhibited significantly different ripple characteristics highlights the sensitivity of conventional single-point measurements to cavity effects and positional variations.

[0076] Figure 3B Presented in N Four insertion loss spectra (Avg() were obtained using the disclosed phase shift measurement technique when the value was 2. T 0, T 0′), Avg( T 1, T 1′), Avg( T 2, T 2′), Avg( T 3, T 3′)). Each spectrum is generated by measuring a pair of single distances (selected from those used to generate the spectrum). Figure 3A The measurements were constructed by averaging the same 20 measurements. Two measurements in each pair were offset by a 400nm probe-to-PIC gap (corresponding to the design wavelength for 1600nm). λ 设计 round trip phase difference π These paired measurements effectively eliminate the dominant cavity-induced ripple because within each pair... π The shifted interference pattern exhibits destructive interference. As shown in the inset, the four corrected spectra almost overlap, with... Figure 3A Compared to individual measurement spectra, it exhibits a significantly reduced position-dependent modulation. These results demonstrate that the disclosed technique can significantly reduce cavity-induced measurement artifacts while preserving the fundamental spectral characteristics of the PIC response.

[0077] Figure 3C This improvement is quantified by plotting the spectral variation envelope around the mean for each case. Uncorrected single measurement data (light-colored data; from...) Figure 3A The corrected data (dark data; from) exhibits a wide error band, while the corrected data (dark data; from) shows a wide error band. Figure 3B This results in a significantly narrower envelope. On average, the standard deviation of the insertion loss measurement is reduced from approximately 0.09 dB to 0.01 dB, leading to an almost order of magnitude improvement in measurement accuracy using the disclosed dual-measurement, phase-complementary method.

[0078] In this specification, similar features in the drawings have been given similar reference numerals. To avoid confusion in some drawings, if some elements have already been identified in the preceding drawings, they may not be indicated. Elements in the drawings are not necessarily depicted to scale, as the focus is on clearly illustrating the elements and structures of this embodiment. For ease and clarity of description, positional descriptors indicating the position and / or orientation of one element relative to another element are used herein. Unless otherwise indicated, these positional descriptors should be understood in the context of the drawings and should not be considered limiting. Specifically, in addition to the orientations illustrated in the drawings, the positional descriptors are intended to cover different orientations in the use or operation of this embodiment. Furthermore, when a first element is referred to as “above,” “over,” “below,” “on top of,” or “under” a second element, the first element may be directly or indirectly above, above, below, on top of, or under the second element, such that one or more intermediate elements may be disposed between the first and second elements.

[0079] The terms “a,” “an,” and “an” are defined herein as meaning “at least one,” meaning that, unless otherwise stated, these terms do not exclude multiple elements.

[0080] Unless otherwise stated, the term "or" is defined as "and / or".

[0081] Terms modifying values, conditions, or characteristics (such as “substantially,” “usually,” and “about”) should be understood to mean that the value, condition, or characteristic falls within acceptable tolerances for normal operation of the described embodiment, or within acceptable limits of experimental error. Specifically, the term “about” generally refers to a range of values ​​that a person skilled in the art would consider equivalent to the stated value (e.g., having the same or equivalent function or result). In some instances, the term “about” means a variation of ±10% of the stated value. Note that, unless otherwise stated, all numerical values ​​used herein are assumed to be modified by the term “about.” Unless otherwise stated, the term “between” refers to a range defined by endpoints, including both endpoints.

[0082] As used herein, the term "based on" is intended to mean "at least partially based on," whether directly or indirectly, and encompasses both "based on only" and "partially based on." Specifically, the term "based on" can also be understood to mean "from," "depending on," "representing," "instructing," "associated with," "involving," etc.

[0083] The terms “match (matching and matched)” in this document refer to the situation where two elements are identical or within a predetermined tolerance of each other. These terms encompass not only exact matching, but also substantive matching, approximate matching, or subjective matching, as well as the best or highest matching among various matching possibilities.

[0084] Unless otherwise stated, the terms “connection” and “coupling”, and their derivatives and variations, refer herein to any form of connection or coupling between two or more elements, whether direct or indirect. Such connection or coupling can take various forms, including but not limited to mechanical, optical, electrical, magnetic, thermal, chemical, logical, fluid, operational, or any combination thereof.

[0085] The term "concurrently" in this document refers to the simultaneous or overlapping occurrence of two or more processes. The term "concurrently" does not necessarily imply complete synchronization, but rather encompasses a variety of scenarios. These scenarios include: the simultaneous occurrence of two processes; a first process that begins and ends during the duration of a second process; and a first process that begins during the duration of a second process but ends after the second process has completed.

[0086] When referring to a quantity or parameter, the term "measurement" is intended to mean that the quantity or parameter can be measured directly or indirectly. In the case of indirect measurement, the quantity or parameter can be derived, retrieved, inferred, or otherwise determined from data of direct measurement.

[0087] The terms “light” and “optical,” as well as their variants and derivatives, encompass radiation across any appropriate region of the electromagnetic spectrum. This includes not only visible light but also extends to invisible regions such as the terahertz (THz), infrared (IR), and ultraviolet (UV) spectral bands. Furthermore, all descriptions provided herein as functions of wavelength can also be formulated as functions of frequency, wavenumber, energy, or other relevant spectral parameters.

[0088] As used herein, the term "processor" refers to any electronic device, circuit system, or component capable of processing, receiving, or transmitting data or instructions (such as computer programs, commands, functions, processes, software code, executable files, applications, and similar entities). The term "processor" implies encompassing a single processor or processing unit, multiple processors or processing units, or other appropriately configured processing elements. When a processor comprises multiple processing elements, these elements may be located at a single site or distributed across multiple sites interconnected by a communication network. Examples of communication networks include local area networks (LANs) and wide area networks (WANs), such as the Internet. Non-limiting examples of processors include general-purpose single-core or multi-core processors; central processing units (CPUs); microprocessors; controllers; microcontrollers; digital signal processors (DSPs); programmable logic devices; field-programmable gate arrays (FPGAs); application-specific integrated circuits (ASICs); digital processors or circuits; analog processors or circuits; state machines; and / or any other device capable of processing information.

[0089] As used herein, the term "memory" refers to any electronic device, circuit system, or component capable of storing electronic data or information. In some instances, the term "memory" may be used interchangeably with the term "computer-readable storage medium." The term "memory" means encompassing a single memory or memory cell, multiple memory or memory cells, or other appropriately configured memory elements. When memory comprises multiple memory elements, these elements may be located at a single site or distributed across multiple sites interconnected by a communication network. Non-limiting examples of memory include any type of random access memory (RAM); any type of read-only memory (ROM); magnetic storage devices; optical storage devices; solid-state drive (SSD) devices, such as flash memory; and any other tangible and / or non-transitory computer-readable medium capable of storing electronic data or information.

[0090] Many modifications may be made to the embodiments described above without departing from the scope of the appended claims.

Claims

1. A method (200) for mitigating position-sensitive errors during testing of a photonic integrated circuit (PIC) (102), the method (200) comprising: Provide (202) a test system (100) including a light source (104), an optical probe (106), an optical detector (108) and an alignment stage (110); The gap between the optical probe (106) and the PIC (102) is adjusted (204) using the alignment stage (110). N There are different gap settings, among which N ≥2, the gap setting corresponds to the corresponding probe-to-PIC distance, which is selected to produce light propagation within the optical cavity (128) formed in the gap. N One round-trip optical phase shift, the round-trip optical phase shift in a complete 2 π The cyclic internal phases are basically evenly spaced; For the above N For each of the different gap settings, perform (206) test measurements, the test measurements including: Test light (138) from the light source (104) is injected (208) into the PIC (102) via the optical probe (106); The measurement light (140) output from the PIC (102) is collected (210) via the optical probe (106); and The optical detector (108) is used to detect (212) the collected measurement light (140) to generate a measurement result; as well as Combination (214) from N The measurement results of each test measurement are used to generate corrected measurement results with reduced position-sensitive errors induced by the optical cavity (128).

2. The method (200) of claim 1, comprising coupling the optical probe (106) to the PIC (102) in an edge-coupled configuration.

3. The method (200) of claim 1 or 2, comprising providing the optical probe (106) as an optical fiber array (1361, 1362).

4. The method (200) as claimed in claim 1 or 2, wherein N =2, and adjusting (204) the gap between the optical probe (106) and the PIC (102) includes: The distances from the two probes to the PIC are set to be approximately one-quarter of the wavelength of the test light (138) at the design wavelength.

5. The method (200) as claimed in claim 1 or 2, wherein the measurement result corresponds to in N Insertion loss spectrum of the PIC (102) measured under a gap setting.

6. The method (200) of claim 5, wherein the insertion loss spectrum is obtained in a band within the spectral range of about 1260 nm to about 1680 nm.

7. The method (200) of claim 1 or 2, wherein the method (200) is implemented using the PIC (102) provided in a wafer-level or multi-die test configuration.

8. The method (200) of claim 1 or 2, wherein the combination (214) of the measurement results comprises: Calculation from the N The average of the measurement results of each test measurement.

9. A test system (100) for testing a photonic integrated circuit (PIC) (102) with reduced position-sensitive errors, the test system (100) comprising: A light source (104) configured to generate test light (138); An optical probe (106) is configured to couple the test light (138) from the light source (104) into the PIC (102) and collect the measurement light (140) output from the PIC (102); An optical detector (108) is configured to detect the collected measurement light (140); Alignment stage (110), the alignment stage being configured to be in a series of N In each test measurement, the gap between the optical probe (106) and the PIC (102) was adjusted to... N There are different gap settings, among which N ≥2, the gap setting corresponds to the corresponding probe-to-PIC distance, which is selected to produce light propagation within the optical cavity (128) formed in the gap. N One round-trip optical phase shift, the round-trip optical phase shift in a complete 2 π The cyclic internal phases are basically evenly spaced; and A control and processing unit (112), coupled to the optical detector (108) and configured to: For the above N Each of the test measurements receives a corresponding measurement result from the optical detector (108); and combination N The measurement results are used to generate corrected measurement results with reduced position-sensitive errors induced by the optical cavity (128).

10. The test system (100) of claim 9, wherein the optical probe (106) is configured to be coupled to the PIC (102) in an edge-coupled configuration.

11. The test system (100) of claim 9 or 10, wherein the optical probe (106) comprises an optical fiber array (1361, 1362).

12. The test system (100) as claimed in claim 9 or 10, wherein N =2, and the alignment stage (110) is configured to set the distances from the two probes to the PIC to be approximately one-quarter of the wavelength of the test light (138) at the design wavelength.

13. The test system (100) as claimed in claim 9 or 10, wherein the measurement result corresponds to... N Insertion loss spectrum of the PIC (102) measured under a gap setting.

14. The test system (100) of claim 9 or 10, wherein the light source (104) comprises a tunable laser.

15. The test system (100) of claim 9 or 10, wherein the control and processing unit (112) is configured to calculate from the N The average of the measurement results from each test measurement is used to generate the corrected measurement result.

Images