Laser system, Lidar system and method for improving linearity of frequency chirp
By using a tunable laser and optical measurement unit to form an angle diversity signal in the FMCW laser system, the nonlinearity of frequency chirp is estimated and compensated, solving the problem of insufficient linearity of laser frequency chirp. This results in a highly efficient and compact laser system, improving ranging accuracy and robustness.
Patent Information
- Application Number
- CN202510586220.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2024-12-10
- Filing Date
- 2025-05-08
- Publication Date
- 2025-11-11
AI Technical Summary
In existing FMCW laser systems, it is difficult to guarantee the linearity of the laser frequency chirp, which leads to a decrease in ranging accuracy, especially in long-distance applications. Furthermore, existing solutions are complex and costly.
A tunable laser is used in conjunction with an optical measurement unit and a control unit. An angle diversity signal is formed by splitting the FMCW optical signal to estimate and compensate for the nonlinearity of frequency chirp. The frequency chirp is linearized using a simplified optical hybrid coupler and control loop.
It achieves a compact and efficient laser system architecture, reduces optical loss, enhances robustness and reliability, can track and compensate for the nonlinearity of frequency chirp in real time, and improves ranging accuracy and signal-to-noise ratio.
Smart Images

Figure CN120933753A_ABST
Abstract
Description
Technical Field
[0001] This specification relates to the field of frequency modulated continuous wave (FMCW) laser systems. More specifically, it relates to methods and systems for achieving linear chirp in laser frequency modulation. Background Technology
[0002] Frequency-modulated continuous wave (FMCW) laser systems rely on rapid laser frequency chirps to operate effectively. The linearity of the frequency chirp is crucial to determining the performance of the laser system. Specifically, FMCW LIDAR systems calculate the distance between the transmitter and the target by utilizing the beat frequency of linearly chirped signals with varying time delays. If the laser chirp is not linear, the beat frequency spectrum becomes wider, reducing the accuracy of LIDAR ranging. Furthermore, the peak frequency is proportional to the target distance, thus requiring a light source with high phase linearity, especially for long detectable distances (such as hundreds of meters).
[0003] High-quality linear chirping is typically achieved using an external modulator and a fast-sweep RF source. However, this approach is complex, incurs high optical losses, and is costly. An alternative method is to achieve linear chirping using a directly tunable laser (called direct modulation). However, the chirping linearity achieved by direct modulation is not as high as that achieved by external modulation. The nonlinearity in direct modulation is affected by various laser conditions, including temperature, chirping speed, chirping range, and laser drive voltage.
[0004] In applications where environmental conditions may vary significantly, chirp linearization in FMCW LIDAR systems requires a control loop. However, known methods involve large systems with long optical fibers and complex digital signal processing (DSP).
[0005] Therefore, a more efficient solution is needed to achieve high-quality linear chirp in FMCW laser systems. Summary of the Invention
[0006] The purpose of this specification is to provide a method and system for achieving high-quality linear chirp in a laser system configured to generate FMCW optical signals.
[0007] Another objective is to simplify the system architecture of FMCW laser systems, such as FMCW LIDAR systems.
[0008] Another objective is to enhance the robustness and reliability of laser systems used to generate FMCW optical signals by minimizing or at least reducing the impact of varying environmental conditions on chirp linearity.
[0009] These and other objects of the invention are satisfied, at least in part, by the invention as defined in the independent claims. Preferred embodiments are set forth in the dependent claims.
[0010] According to a first aspect, a laser system for generating frequency-modulated continuous wave (FMCW) optical signals is provided. The system includes:
[0011] A tunable laser used to generate FMCW optical signals;
[0012] An optical measurement unit configured to receive a portion of an FMCW optical signal and, based on the difference between a first signal and a second signal formed by splitting the portion of the FMCW optical signal, output at least two angular diversity signals via an optical hybrid coupler, wherein the second signal is delayed relative to the first signal, and wherein a pair of the at least two angular diversity signals has a fixed phase shift relative to each other; and
[0013] Control unit, which is configured to:
[0014] Receive the at least two angle diversity signals.
[0015] Based on these at least two angular diversity signals, estimate the compensation required to adjust the nonlinearity of the frequency chirp of the tunable laser, and
[0016] Output the corresponding control signal to the tunable laser.
[0017] The control signal is configured to improve the linearity of the frequency chirp.
[0018] The laser system includes an optical measurement unit configured to output at least two angle diversity signals. The optical measurement unit determines the angle diversity signals based on a portion of the FMCW signal and a delayed version of that portion of the FMCW signal. This means that the angle diversity signals can represent the beat frequency signal formed by the interference of the first and second signals.
[0019] Because of the measurement of the angle diversity signal, simple processing can be performed to determine the adjustment of the nonlinearity of the frequency chirp. The angle diversity signal can be processed to easily determine the frequency or phase of the beat frequency signal, which in turn represents the frequency variation in the chirp. This makes it possible to determine the linearity of the frequency chirp quickly and with high accuracy.
[0020] Because of the high precision in determining the frequency / phase of the beat frequency signal, a short delay in the second signal can be used. This delay can be provided by physically guiding the second signal along the optical path. Since laser systems allow for short delays, they can also be compact because long optical paths are not required.
[0021] Therefore, a laser system using a simplified system architecture to generate FMCW optical signals was employed. The laser system also features reduced optical losses and minimized impact on chirp linearity attributable to variations in environmental conditions. Furthermore, the system exhibits a high capability to use long chirp periods, typically associated with relatively large nonlinearities, as this system ensures that chirp nonlinearity is mitigated.
[0022] The nonlinearity of frequency chirp can be attributed to factors such as the bandwidth limitation of the driver and the nonlinear characteristics of the tunable laser.
[0023] The term "tunable laser" herein refers to any unit, tunable device, and / or element capable of emitting at least partially coherent light. A tunable laser may include a laser diode, for example configured to generate an optical carrier signal. The optical carrier signal is generated in the form of laser light (i.e., an FMCW optical signal).
[0024] In this disclosure, the term "light" should be interpreted broadly, not limited to visible electromagnetic radiation. Rather, the term "light" may also include, for example, ultraviolet and infrared light.
[0025] A tunable laser may include a frequency modulation module configured to modulate the frequency of an optical carrier signal. This refers to the frequency modulation module being external to the laser source, allowing frequency modulation of the laser output from the laser source. Alternatively, the frequency modulation module may be integrated into the laser source, enabling frequency modulation to be performed within the laser source. For example, frequency modulation can be performed via direct frequency modulation of the laser source.
[0026] Furthermore, the laser system can be integrated onto a single semiconductor chip. This laser system provides a compact (e.g., on-chip integrable) solution for laser linear chirp generation, reducing the complexity of digital signal processing (DSP). For example, estimation of the control signal output required for compensation and configuration to improve frequency chirp linearity can be achieved using a small-size architecture that can be integrated onto a photonic integrated circuit (PIC). In other words, some or all components of the laser system can be integrated onto the PIC.
[0027] For example, DSP complexity is reduced by phase accumulation, which is provided by splitting a portion of the FMCW optical signal so that the second signal is delayed relative to the first signal. Specifically, the use of angle diversity signals and their detection and processing by the control unit makes it possible to avoid complex DSP architectures. Therefore, by avoiding complex Hilbert transform-based methods in the frequency domain, lower DSP complexity can be achieved.
[0028] Furthermore, this laser system enables the realization of a real-time adaptive control loop, thereby achieving chirp linearization of the FMCW optical signal. Specifically, adaptive nonlinearity monitoring and control can handle nonlinearities attributable to changes in frequency chirp due to varying (external) conditions. Real-time chirp tracking enables a fast control loop, which may involve chirp calculation on a sample-by-sample basis, thus allowing the determination of the instantaneous phase of the beat frequency signal.
[0029] Here, "sample" can refer to data collected at specific time intervals, which can be compared with frames that can be associated with one or more chirps. The number of samples within a frame can be related to the frequency at which the time-domain waveform is sampled to derive the chirp (the sampling rate may be much higher than the Nyquist limit). Typically, a chirp can be considered as a frame, for example, an up-chirp represents one frame and a down-chirp represents another frame.
[0030] Therefore, compared to frame-by-frame basis, computation on a sample-by-sample basis may involve sequential processing of samples, where on a frame-by-frame basis, a complete set of samples (frames) is processed together.
[0031] The number of samples in a frame can depend on the sampling rate. For example, in FMCW LIDAR applications, the chirp rate can range from 3 to N x 10 GHz within 50 microseconds, where N can be an integer representing a multiple of 10 GHz. In high-speed data communication applications, the chirp rate can range from 1 to N x 100 GHz within 50 to 100 picoseconds. This enables the implementation of compact, adaptive, and integrated laser linear chirp control loops.
[0032] Furthermore, this laser system enables direct tracking of both positive and negative chirps, and facilitates the tracking of both. This can be achieved by using angle diversity signals to provide information about which of the first and second signals has a higher frequency.
[0033] The control unit can also be configured to estimate the instantaneous frequency of the FMCW optical signal. The instantaneous frequency estimation can be phase-based. In other words, the instantaneous frequency can be estimated based on the phase of the angle diversity signal. Therefore, the frequency at any given moment can be determined by analyzing the phase change of the angle diversity signal over time. It is possible to calculate the frequency change of the FMCW optical signal by monitoring how the phase evolves.
[0034] The optical measurement unit may include an interferometer structure, such as an asymmetric Mach-Zehnder interferometer (AMZI), which is configured to split a portion of the FMCW optical signal into a first signal and a second signal, and to delay the second signal relative to the first signal.
[0035] In other words, an interferometer structure can be configured to split a portion of an FMCW optical signal into two separate signals. One of these signals is delayed compared to the other. An asymmetric Mach-Zehnder interferometer is a specific configuration of this interferometer structure that allows for precise control over the splitting and delay of the optical signal.
[0036] Therefore, accurate measurements can be provided by creating a controlled delay between the first and second signals. Additionally, using AMZI can reduce noise and improve measurement clarity. Finite-difference techniques can be used to estimate the instantaneous frequency of an optical signal. This technique benefits from a small delay between the first and second signals, leading to higher accuracy because the finite-difference approximation corresponds very closely to the derivative of the optical signal (instantaneous frequency). The delay can be adjusted based on the rate of change of the laser frequency. However, for slower frequency changes, extremely small delays (e.g., nanoseconds to microseconds) are not required to maintain good accuracy.
[0037] For high-speed applications, such as frequency modulation in data communication systems (e.g., 50 GBaud / s on-key control), latency can be less than 20 picoseconds (e.g., approximately 10 picoseconds).
[0038] Smaller delays can lead to more accurate instantaneous frequency estimations, but may also reduce the signal-to-noise ratio (SNR). Therefore, choosing an appropriate delay can balance accuracy and performance, thus ensuring optimal results for a variety of applications. Consequently, the delay caused by the interferometer structure can be configured differently depending on whether the optical signal from the chirped laser source involves slow or rapid frequency changes.
[0039] In one example, the second signal can be configured to be delayed with a relatively short delay relative to the first signal. The second signal can be delayed relative to the first signal by a delay line, the length of which is in the range of 1 cm to 150 cm, preferably between 5 cm and 50 cm, and more preferably between 10 cm and 20 cm. Therefore, a relatively short delay line length can be used.
[0040] For example, a 10cm path length difference between the first and second signals can achieve accurate measurement. A 25cm path length difference can provide almost the same accuracy as a 200cm difference.
[0041] For on-chip designs, path length differences of less than 25 cm can be considered to maintain compactness and integration efficiency. This allows for compact form factors, such as delay line lengths in the tens of centimeters range.
[0042] A path length difference of 10 cm corresponds to a time delay of approximately 0.3 nanoseconds in free space (or approximately 0.5 nanoseconds in optical fiber), while a path length difference of 25 cm results in a time delay of approximately 0.8 nanoseconds in free space (or approximately 1.2 nanoseconds in optical fiber).
[0043] Typically, an optical measurement unit can be configured to split a portion of the FMCW optical signal into a first optical signal propagating in a first path and a second optical signal propagating in a second path. The second path may have a preset delay relative to the first path. As an example, the preset delay can be provided by a first path and a second path having different optical path lengths. For example, the preset delay may be a known preset delay. The preset delay can be provided by one of the first and second paths being longer than the other. Longer paths can be achieved by guiding the light in a loop, spiral, or other manner, thus providing a long path length in a compact solution.
[0044] After passing through the first and second paths respectively, the first and second optical signals are combined again. When the first and second optical signals are combined, interference may occur between them. The resulting combined signal can form a beat frequency. The beat frequency is related to a preset delay of the optical measurement unit. As an example, the beat frequency can be proportional to the preset delay.
[0045] The optical measurement unit may also include an optical hybrid coupler. The optical hybrid coupler can therefore be configured to output at least two angle diversity signals based on the interference between the first signal and the second signal.
[0046] Optical hybrid couplers can be, for example, 120-degree optical hybrid couplers or 90-degree optical hybrid couplers.
[0047] A 120-degree optical hybrid coupler is a device that splits an input optical signal into three output signals (i.e., three angular diversity signals), with a fixed phase difference of 120 degrees between each pair of outputs. The 120-degree phase shift ensures that the signals are uniformly separated in phase.
[0048] A 90-degree optical hybrid coupler, also known as an orthogonal coupler, splits an input signal into two output signals (i.e., two angular diversity signals) with a 90-degree phase difference between them.
[0049] In addition, the optical measurement unit may include at least two photodiodes for detecting at least two angle diversity signals.
[0050] A photodiode can typically be any unit or device that includes a photosensitive element, which is configured to detect the intensity of light incident on the photosensitive element, generate an electrical signal in response, and allow the electrical signal to be read out.
[0051] A photodiode can be a single-ended or balanced photodiode (BPD). A balanced photodiode, as used herein, refers to a device comprising two photodiodes connected in series. When the two photodiodes detect the same level of light—that is, when they generate equal electrical signals—their signals cancel each other out. This arrangement allows for the detection of small differences in the light levels on the two photodiodes.
[0052] Specifically, a single-ended photodiode directly detects the intensity of the angle diversity signal from the output of the optical hybrid coupler. In contrast, a balanced photodiode measures the intensity difference between the angle diversity signals of the two outputs (i.e., a pair of outputs), which can help eliminate common-mode noise and improve the signal-to-noise ratio.
[0053] Generally, an optical measurement unit may include three components: an optical AMZI with a short arm length difference (such as 20 cm or tens of centimeters to support on-chip structure), an optical hybrid coupler, and an optical sensor (such as a single-ended photodiode or BPD).
[0054] For the detection of optical hybrid couplers and angle diversity signals, a design based on a 120-degree hybrid coupler can include (but is not limited to) a 3x3 MMI followed by three single-ended photodiodes. Alternatively, a 90-degree hybrid coupler can be used, for example by implementing a 2x4 or 4x4 MMI followed by two BPDs. Typically, an optical hybrid coupler can include at least one MMI (or directional coupler).
[0055] By incorporating at least one MMI or directional coupler into the optical hybrid coupler, precise control over the splitting and combining of optical signals is provided. Furthermore, using such a coupler enables compact and integrable designs (e.g., suitable for on-chip integration), i.e., reducing the size and complexity of the optical measurement unit.
[0056] A 90-degree hybrid coupler structure can reduce phase noise by using balanced detection. On the other hand, a 120-degree hybrid coupler-based approach can have a less complex MMI structure, fewer manufacturing tolerances, and / or a smaller footprint. The 120-degree optical hybrid coupler design can use a single-ended photodiode (compared to a BPD), thus, for example, reducing the number of components required. This reduction in components also means fewer RF connections are needed.
[0057] Furthermore, using a single-ended photodiode allows for a simpler transimpedance amplifier (TIA) structure. The current generated by a single-ended photodiode is proportional to the incident light, and this can be converted into a voltage using a single TIA. Therefore, this conversion process requires fewer components and connections compared to a BPD configuration involving differential signals.
[0058] The fixed phase shift can be the same between any pair of at least two angular diversity signals.
[0059] In other words, the fixed phase shift between any two of the at least two angular diversity signals can be the same, such as 90 degrees or 120 degrees. This means that the phase difference between each pair of angular diversity signals remains constant and equal, whether it is 90 degrees, 120 degrees or any other phase angle.
[0060] A fixed phase relationship ensures that the angle diversity signals maintain uniform phase alignment. A fixed phase shift can be achieved using an optical hybrid coupler. For example, a 90-degree hybrid coupler can provide a fixed 90-degree phase shift between the angle diversity signals output to the control unit, while a 120-degree hybrid coupler can provide a fixed 120-degree phase shift.
[0061] The control unit can be configured to repeatedly output control signals to the tunable laser. Conversely, the control signals can be configured to repeatedly increase the linearity of the frequency chirp. In other words, the control unit can output control signals cyclically based on the estimated compensation required to adjust the nonlinearity of the frequency chirp of the tunable laser.
[0062] Therefore, tunable lasers can be continuously updated, that is, the nonlinearity of the frequency chirp of a tunable laser can be continuously linearized or compensated.
[0063] The repetition rate may be relatively low, for example, once every ten seconds. However, for unstable lasers or lasers susceptible to, for example, variations in chirp patterns and / or ambient temperature, the nonlinearity of the frequency chirp of the tunable laser can be monitored or detected, for example, when the nonlinearity exceeds a threshold. Therefore, the control signal can be configured to increase the linearity of the frequency chirp as needed to optimize the performance of the tunable laser.
[0064] Thresholds can be defined using quality factors (i.e., key performance indicator (KPI) values). For example, the quality factor (r) can be extracted from formulas such as... Where υ nl,rms [Hz] is the root mean square of the nonlinear component of the (ramp-up or ramp-down) frequency chirp, f exc [Hz] is the offset frequency of the tunable laser.
[0065] In one example, if the quality factor exceeds 1e-5, the linearity of the frequency chirp can be considered insufficiently linear. This threshold can indicate that the linearity of the frequency chirp needs to be adjusted or improved.
[0066] Monitoring of nonlinearity can be adaptive. In other words, if nonlinearity is detected, linearity can be improved at a faster rate.
[0067] The control unit can also be configured to calculate the frequency and / or phase corresponding to at least two angular diversity signals by creating a vector and calculating the (phase) angle of the vector. The frequency and / or phase can be either instantaneous frequency or phase (angle).
[0068] This vector can be created by summing at least two angular diversity signals. For example, the angular diversity signals can be rotated relative to each other in the complex plane.
[0069] In the case of a 120-degree optical hybrid coupler, first (S1), second (S2), and third (S3) angular diversity signals can be provided, so the vector can be formed by S1 + S2*exp(j2π / 3) + S3*exp(j4π / 3), where j is a complex number. Therefore, the second angular diversity signal is rotated 120 degrees in the complex plane relative to the first angular diversity signal, and the third angular diversity signal is rotated 120 degrees in the complex plane relative to the second angular diversity signal.
[0070] Similarly, for a 90-degree hybrid coupler, where the angle diversity signal is detected by the BPD, resulting in two detected angle diversity signals (S1, S2), the vector can be represented by S1 + S2 * exp(jπ / 2). Therefore, the second angle diversity signal is rotated 90 degrees relative to the first angle diversity signal in the complex plane.
[0071] For example, a vector angle calculator can be used to extract the angle (i.e., the phase angle) of the resulting vector. Alternatively, a method can be used to expand the phase angle, which involves adding or subtracting multiples of 2π to eliminate discontinuities and maintain a smooth transition between successive phase values. This expansion process can transform phase angles that initially fall outside the standard range of -π to π into a continuous sequence.
[0072] The laser system may also include an analog-to-digital converter (ADC) and a digital-to-analog converter (DAC). The ADC can be configured to convert the analog input signal arriving at the control unit into a digital signal based on at least two angle diversity signals. Furthermore, the control unit can be configured to convert the digital signal to the complex domain, calculate the phase angle, estimate nonlinearity, perform predistortion calculations, and / or output the processed digital signal, wherein the DAC can be configured to convert the processed data signal into a control signal.
[0073] Therefore, compared to the real domain, it enables phase calculation and accumulation in the complex domain. Using complex signals makes it possible to achieve real-time and positive / negative chirp measurements. Furthermore, smaller time delays can be achieved without sacrificing estimation performance accuracy (e.g., at chirp edges between upslopes and downslopes, and vice versa).
[0074] The at least two angle diversity signals can be analog signals. Therefore, the at least two angle diversity signals can be converted into digital signals by an ADC. In other words, the ADC can be configured to convert the detected signals from analog detection signals into digital detection signals and send the digital detection signals to the control unit.
[0075] Furthermore, the laser system may include a TIA configured to amplify the detected signal (e.g., one TIA for each detected signal). The amplified signal can then be sent to the ADC.
[0076] An ADC can be associated with at least two photodiodes, for example, configured to detect at least two angular diversity signals. For example, each diversity signal can be converted into a digital signal by a corresponding ADC.
[0077] The control unit can then sample the digital (i.e., converted) angle diversity signal. Furthermore, the control unit can convert the digital signal to the complex domain.
[0078] In other words, a complex signal can be synthesized using multiple real (digital) signals (e.g., at least two real signals). A complex signal can be represented by a vector, for example, as described above.
[0079] Based on complex signals, the (phase) angle can be calculated.
[0080] It will become clear that the wavelength of an angle diversity signal can be non-linearly mapped to the driving voltage of a tunable laser, and the mapping between voltage and time is a linear curve. Therefore, the relationship between the (phase) angle and time is non-linear. However, linear frequency chirps require a linear angle-time (i.e., angle-voltage) mapping. Therefore, the non-linearity of the frequency chirp can be estimated using the measured non-linear angle-time mapping. In one example, the non-linearity can be estimated by unfolding the phase angle and comparing it to a linear response (i.e., by measuring the deviation of the actual output from the output predicted by the linear model).
[0081] Based on the estimated nonlinearity, a predistorted voltage ramp can be obtained. The predistorted voltage ramp can be obtained from the estimated nonlinearity, for example, through least mean square or linear interpolation. For example, chirped linearization can be implemented iteratively to obtain a higher linear phase. To do this, the calculated voltage ramps (up and down) can be stored and applied to the next iteration.
[0082] The calculated predistortion curve can then be used to update the drive voltage of the tunable laser. Specifically, the processed digital output signal can include the predistortion curve. Therefore, the predistortion curve can be converted into a control signal via a DAC.
[0083] The control unit may also include a reference signal, wherein the control unit and the measurement unit form an opto-phase-locked loop (OEPLL) configured to stabilize the frequency and phase of the FMCW optical signal.
[0084] In other words, an OEPLL is a fast, real-time control loop that can lock the laser chirp to a reference linear signal. The reference signal can be, for example, a reference ramp signal, which changes at a rate that is slower than the feedback rate (e.g., 100 times slower). The estimation error between the measured nonlinear angle-time map and the reference ramp signal is used for further calculations, such as scaling, integration, and nonlinear compensation summation.
[0085] Therefore, OEPLL enables faster feedback control. For example, OEPLL can be used in rapidly changing lasers.
[0086] According to a second aspect, a LIDAR system is provided, comprising a laser system according to a first aspect and a LIDAR detection unit, wherein the LIDAR detection unit is configured to receive a reflected light signal based on an FMCW light signal.
[0087] This aspect can usually present the same or corresponding advantages as the first aspect.
[0088] Therefore, a system for nonlinearity estimation (and predistortion) of direct laser frequency chirp in a LIDAR system has been realized. Specifically, this LIDAR system enables adaptive frequency chirp nonlinearity monitoring and control to manage nonlinearity that can vary in response to changing conditions such as external temperature.
[0089] The LIDAR system can also be implemented using a small-size architecture that can be integrated onto a chip (such as a PIC). The optical (non-linear) measurement unit can be integrated onto the same LIDAR PIC, such as being compactly packaged with the control unit on the same LIDAR motherboard.
[0090] As discussed in the first aspect, for example, angle diversity detection and phase accumulation can simplify the complexity of digital signal processing. Specifically, frequency domain processing methods can be avoided, that is, the Hilbert transform method, which is based on high digital signal processing complexity, can be avoided.
[0091] The tunable laser of this laser system may include an optical beamsplitter arranged in the path of the FMCW optical signal and configured to split the FMCW optical signal into a LIDAR transmission signal that can be emitted from the LIDAR system. This portion of the FMCW signal is provided to the optical measurement unit of the laser system. Therefore, "a portion of the FMCW optical signal" is the portion of the FMCW optical signal that is split off and guided to the optical measurement unit. The remaining portion of the FMCW optical signal then forms the LIDAR transmission signal, which can be emitted from the LIDAR system, for example, toward a target.
[0092] In other words, a LIDAR system can split the FMCW optical signal from a tunable laser into two parts: a local oscillator (LO) signal and a transmission (Tx) signal. The FMCW optical signal can be split, for example, by a beam splitter. Typically, the LO signal may constitute a small portion (e.g., 1%) of the optical signal, while the Tx signal may comprise the majority (e.g., 99%). The Tx signal is used to detect distance and is reflected back from the target to the LIDAR detection unit.
[0093] In the detection unit, the LO signal and the reflected light signal can be coupled together for coherent detection. For example, the process may involve using receiver optics to collect the reflected light signal, using a 2x2 coupler to combine the LO and reflected signals, using a balanced photodetector to detect the combined signal and convert it into an electrical signal, and / or using a signal processing unit to analyze the electrical signal, for example, to determine the target's distance, velocity, and / or other characteristics.
[0094] In addition, a LIDAR system may include an optical phased array to facilitate the dynamic guidance of the Tx signal within the field of view.
[0095] LIDAR systems can be integrated into motor vehicles and configured to provide real-time distance and speed data of the vehicle's surroundings.
[0096] In other words, a LIDAR system can be integrated into a vehicle to continuously monitor and measure the distance and speed of objects around the vehicle in real time. LIDAR systems can be implemented in devices or systems such as robots, drones, and various industrial applications to provide real-time distance and / or speed data.
[0097] Real-time distance and speed data enable vehicles to detect and respond to obstacles and other vehicles in a timely manner. Furthermore, when implemented in other devices such as robots and drones, LIDAR systems enhance their ability to navigate and avoid obstacles in their environment.
[0098] Furthermore, linear frequency chirp in a LiDAR system can improve distance resolution, for example, allowing for more accurate measurement of the distance to objects that are very close together. Additionally, linear frequency chirp can enhance the signal-to-noise ratio and facilitate the detection of weak reflections from distant objects.
[0099] According to a third aspect, a method is provided for improving the linearity of the frequency chirp of a frequency-modulated continuous wave (FMCW) optical signal, the method comprising:
[0100] FMCW optical signals are generated via tunable lasers;
[0101] A portion of the FMCW optical signal is received at the optical measurement unit;
[0102] Based on the difference between the first and second signals formed by splitting the FMCW optical signal, at least two angular diversity signals are output via an optical hybrid coupler, wherein the second signal is delayed relative to the first signal, and one pair of the at least two angular diversity signals has a fixed phase shift relative to each other;
[0103] The at least two angle diversity signals are received at the control unit.
[0104] Based on the at least two angular diversity signals, estimate the compensation required to adjust the nonlinearity of the frequency chirp of the tunable laser, and
[0105] Output the corresponding control signal to the tunable laser.
[0106] The control signal improves the linearity of the frequency chirp.
[0107] This aspect can usually present the same or corresponding advantages as the first and second aspects.
[0108] The steps for estimating the compensation required to adjust the nonlinearity of the frequency chirp of a tunable laser may include: converting an analog signal into a digital signal based on at least two angular diversity signals, converting the digital signal into the complex domain, calculating the phase angle, estimating the nonlinearity, performing predistortion calculations, outputting the processed digital signal, and / or converting the processed data signal into a control signal.
[0109] Analog signals can be converted into digital signals via an ADC. Multiple ADCs can be used. Furthermore, a DAC can convert processed digital signals into control signals.
[0110] Digital signals can be converted to the complex domain by forming vectors in the complex plane based on angular diversity signals.
[0111] The phase angle can be calculated by measuring the angle of the vector in the complex plane.
[0112] The phase angle can be mapped relative to time, allowing the nonlinearity of the frequency chirp to be estimated based on the (nonlinear) curve of the angle-time mapping. For example, the nonlinearity can be estimated by unfolding the phase angle and comparing it to a linear response (i.e., by measuring the deviation between the actual output and the output predicted by a linear model).
[0113] Based on the estimated nonlinearity, predistortion calculations can be performed by forming a predistortion voltage ramp, for example, through least mean square or linear interpolation.
[0114] Furthermore, the calculated predistortion curve can be used to update the drive voltage of the tunable laser, thereby improving the linearity of the frequency chirp.
[0115] The method may also include repeating these steps of the method, as set forth in the second aspect.
[0116] In other words, control signals can be repeatedly output to a tunable laser, such as to repeatedly improve the linearity of the frequency chirp. The control unit can continuously output control signals based on the estimated compensation required to adjust the nonlinearity of the frequency chirp of the tunable laser.
[0117] The method may also include, at the optical measurement unit: splitting a portion of the FMCW optical signal into a first signal and a second signal at the interferometer structure, such that the second signal is delayed relative to the first signal; and detecting the at least two angle diversity signals via at least two photodiodes.
[0118] In other words, the optical measurement unit may include an interferometer structure for splitting the portion of the FMCW optical signal.
[0119] This portion of the FMCW optical signal can, for example, be split such that 50% of this portion of the FMCW optical signal is delayed relative to the remaining 50%. However, other splitting ratios can also be used. In principle, the first and second signals (after the delay line) should be equal. If the delay line has high loss, more power can be allocated to the second signal with the delay line to compensate for that loss. If the power is not split equally, the signal-to-noise ratio (SNR) of the recombined signal will decrease. However, this method may still be effective with unequal splitting, but performance will degrade as the SNR decreases.
[0120] The effects and features of the first, second, and third aspects are largely similar. The examples mentioned with respect to the first, second, and third aspects are largely compatible. It should also be noted that, unless otherwise expressly stated, this disclosure relates to all possible combinations of features. Attached Figure Description
[0121] The foregoing contents, additional objects, features, and advantages of this specification will be better understood through the following illustrative and non-limiting detailed description with reference to the accompanying drawings. Unless otherwise stated, the same reference numerals will be used for the same elements in the drawings.
[0122] Figure 1 A laser system for generating FMCW optical signals is schematically shown, wherein control signals are configured to improve the linearity of the frequency chirp of the FMCW optical signals.
[0123] Figure 2 An optical measurement unit forming a 120-degree optical hybrid coupler is schematically shown.
[0124] Figure 3 An optical measurement unit forming a 90-degree optical hybrid coupler is schematically shown.
[0125] Figure 4 A block diagram of the signal processing steps is shown.
[0126] Figure 5A A LiDAR system including a laser system is schematically shown.
[0127] Figure 5B A schematic diagram of linear and nonlinear frequency chirps is shown. Detailed Implementation
[0128] Figure 1 A laser system 100 for generating a frequency-modulated continuous wave (FMCW) optical signal 112 is shown. The system 100 includes a tunable laser 110 for generating the FMCW optical signal 112; and an optical measurement unit 120 configured to receive a portion 114 of the FMCW optical signal 112 and output at least two angular diversity signals 124 via an optical hybrid coupler 122 based on the difference between a first signal 114a and a second signal 114b formed by splitting the portion 114 of the FMCW optical signal 112, wherein the second signal 114b is delayed relative to the first signal 114a, and wherein a pair of signals in the at least two angular diversity signals 124 have a fixed phase shift relative to each other.
[0129] The laser system 100 also includes a control unit 130 configured to: receive at least two angle diversity signals 124, estimate the compensation required for adjusting the nonlinearity of the frequency chirp of the tunable laser 110 based on the at least two angle diversity signals 124, and output a corresponding control signal 132 to the tunable laser 100.
[0130] The output control signal 132 is configured to improve the linearity of the frequency chirp of the FMCW optical signal 112 generated by the tunable laser 110.
[0131] The tunable laser 110 that generates the FMCW optical signal 112 can be a laser whose output frequency can be adjusted or varied within a certain range. Tunability allows the laser to be tuned to generate the FMCW optical signal 112 with a linearized frequency chirp.
[0132] Compared to the remaining propagated portion 116 of the FMCW optical signal 112, the portion 114 of the FMCW optical signal 112 received by the optical measurement unit 120 can be relatively small. Portion 114 can, for example, be a small fraction of the FMCW optical signal 112 (e.g., about 1%). Similarly, the remaining propagated portion 116 can correspond to approximately 99% of the FMCW optical signal 112. However, these percentages are exemplary and can vary depending on the specific requirements of the laser system 100. Dividing the FMCW optical signal 112 into portions allows a sufficient amount of light to be available for both linearly chirped measurement and propagation to the target. An optical beamsplitter or similar device can be used to separate portion 114 from the FMCW optical signal 112; for example, the device is configured to direct a small portion of the light to the optical measurement unit 120 while allowing the majority of the FMCW optical signal to continue propagating.
[0133] The optical measurement unit 120 can be a nonlinearity measurement module, for example, based on an asymmetric Mach-Zehnder interferometer (AMZI) structure.
[0134] An AMZI structure can be an optical device configured to split an optical signal into two separate paths. These paths have different optical lengths and are eventually recombined. By analyzing the interference pattern produced when the optical paths merge, an AMZI can detect changes in the difference in optical path lengths.
[0135] exist Figure 1 In the diagram, the optical path of the second signal 114b is shown to be longer than that of the first signal 114a (see the jagged dashed lines). However, the optical path of the first signal 114a may be longer than that of the second signal 114b. Alternatively, the paths may comprise different materials. Different materials can affect the propagation speed of signals 114a and 114b, causing one of the first signal 114a and the second signal 114b to be delayed relative to the other signal.
[0136] The optical paths for the first signal 114a and the second signal 114b can be provided by waveguides for guiding the first signal 114a and the second signal 114b, respectively. Therefore, the optical measurement unit 120 may include a first waveguide for guiding the first signal 114a and a second waveguide for guiding the second signal 114b. The optical hybrid coupler 122 is a device that can be used to combine or split optical signals. It typically includes multiple input and output ports, allowing it to mix signals from different sources or distribute a single signal to multiple destinations. Therefore, the optical hybrid coupler 122 operates based on the principle of interference, where the input optical signals are combined as follows: their phase and amplitude are manipulated to achieve the desired output. Figure 1 In the optical hybrid coupler 122, the input signals are the first signal 114a and the second signal 114b, and the output is the angle diversity signal 124.
[0137] Therefore, optical interference occurs within the optical hybrid coupler 122, such as a multimode interferometer coupler. When two optical signals 114a and 114b, propagating at different optical lengths, are recombined in the optical hybrid coupler 122, they interfere with each other. The interference pattern is a result of the phase difference between the two signals 114a and 114b. In one example, the optical hybrid coupler 122 may be a multimode interferometer (MMI). For example, the optical hybrid coupler 122 may include at least one MMI.
[0138] Then, the resulting combined optical signal, i.e., the resulting angle diversity signal 124, can be guided to a photodiode ( Figure 1 (Not shown in the image). Each signal detected by the photodiode can represent the interference result.
[0139] exist Figure 1 In this embodiment, the optical measurement unit 120 includes an optical hybrid coupler 122 configured to output at least two angular diversity signals 124 based on the interference (i.e., difference) between a first signal 114a and a second signal 114b. However, it should be understood that the optical hybrid coupler 122 may be arranged separately, for example, from the waveguides used to guide the first signal 114a and the second signal 114b. Therefore, the first signal 114a and the second signal 114b can be output from the waveguides and received by the optical hybrid coupler 122, which in turn outputs at least two angular diversity signals 124 to the control unit 130.
[0140] The control unit 130 can repeatedly output control signals 132 to the tunable laser 110. Therefore, control signals 132 can repeatedly improve the linearity of the frequency chirp. The repeatedly output control signals 130 can correspond to the estimated compensation required to adjust the nonlinearity of the frequency chirp of the tunable laser 110.
[0141] The control unit 130 may be a microcontroller, such as a simple microcontroller. In other words, the control unit 130 may have low complexity and / or be easy to integrate on-chip.
[0142] It should be understood that the control unit 130 can be implemented as a processing unit, such as a general-purpose processing unit, like a central processing unit (CPU), which can execute instructions of one or more computer programs to implement the functionality of the control unit 130. The control unit 130 can also or alternatively be implemented as firmware arranged in, for example, an embedded system, or as a specially designed processing unit, such as an application-specific integrated circuit (ASIC) or a field-programmable gate array (FPGA). It should be understood that the control unit 130 can be implemented as a combination of hardware and software components.
[0143] although Figure 1 Not shown in detail, but the laser system 100 can be integrated onto a single semiconductor chip. Specifically, all or some components of the laser system 100 can be integrated onto a single chip. For example, the optical measurement unit 120 and the controller 130 can be integrated onto the same chip (such as the PIC, which is also equipped with electronic circuitry). For example, the tunable laser 110 can be disposed outside the chip. However, it should be understood that the tunable laser can also be disposed or integrated onto the same chip as the optical measurement unit 120 and the controller 130.
[0144] The control unit 130 may also include a reference signal ( Figure 1 (not shown in the image), wherein the control unit 130 and the measurement unit 120 form an opto-phase-locked loop (OEPLL), which is configured to stabilize the frequency and phase of the FMCW optical signal 112.
[0145] OEPLL can be used to lock the laser frequency chirp to a specific ramp signal in real time. The detected phase angle of the beat frequency signal can then be compared to a reference ramp signal, which may change at a slower rate (e.g., 1 kHz) than the feedback rate (e.g., 100 kHz). The corresponding difference (i.e., error) based on this comparison can be used for further calculations, such as scaling, integration, and / or nonlinear compensation. Figure 2 The optical measurement unit 120 is schematically shown in the figure.
[0146] A portion 114 (also denoted as λ) of the FMCW optical signal is received by an interferometer structure 121 configured to split this portion 114 into a first signal 114a and a second signal 114b. This portion 114 of the FMCW optical signal is further split into the first signal 114a and the second signal 114b by a beam splitter 121s or a 1x2 coupler. For example, the beam splitter can be a beam splitter. In one example, this portion 114 of the FMCW optical signal is split into 50-50 segments, such that the first signal 114a corresponds to 50% of the portion 114, and the second signal 114b corresponds to 50% of the portion 114 of the FMCW optical signal.
[0147] The interferometer structure 121 also includes an optical path configured to delay the second signal 114b relative to the first signal 114a. Here, the optical path corresponding to the second signal 114b is shown as forming a delay line through a loop optical path (e.g., an optical fiber loop), i.e., an optical path longer than the optical path of the first signal 114a. The longer optical path can, for example, be about 20 cm longer than the shorter optical path. However, it should be understood that the delay can be achieved by any suitable delay line or signal delay method, for example, by having the signals 114a and 114b propagate in materials with different refractive indices.
[0148] A portion 114 of the FMCW optical signal 112 can be represented by an electric field E(t) with unit amplitude, the first signal 114a can be represented by an electric field E1(t), and the second signal 114b can be represented by an electric field E2(t). These signals can be represented as:
[0149] E(t) = cos[φ] FMCW (t)],
[0150]
[0151] Where φ FMCW (t) is the phase of the FMCW optical signal, and τ is the time delay in a longer optical path.
[0152] Figure 2 The optical hybrid coupler 122 is a 120-degree optical hybrid coupler. The 120-degree optical hybrid coupler splits the input optical signal into three output angular diversity signals 124, with a fixed phase difference of 120 degrees between each pair of outputs. Specifically, as... Figure 2 As shown, the fixed phase shift (α) between any pair of angle diversity signals 124 is the same. Specifically, in Figure 2 In this case, the fixed phase shift is 120 degrees (i.e., α = 120). 0 ).
[0153] although Figure 2While not explicitly shown, the 120-degree hybrid coupler 122 can be based on a directional coupler or a 3x3 MMI, such as a 3x3 multimode interference coupler. A 3x3 MMI coupler can utilize multimode interference to split and combine the input optical signals. The 3x3 MMI coupler may include a multimode waveguide segment where the multimodes interfere destructively, resulting in a desired rate distribution between the output ports of the 120-degree hybrid coupler 122.
[0154] In the case of a 3x3 MMI coupler, the first signal 114a and the second signal 114b can be input to the 3x3 MMI coupler through any two of its three input ports. Therefore, one input port of the 3x3 MMI coupler may be unused or may not receive any signal. In one example, a 2x3 MMI coupler can be used.
[0155] exist Figure 2 In the configuration shown, the optical hybrid coupler 122 is designed to achieve a 120-degree phase shift between the output signals (i.e., the angle diversity signals 124). This can be achieved by selecting the dimensions of the multimode waveguide, such as its width and length, to ensure the achievement of the correct / desired interference pattern. Thus, the 120-degree hybrid coupler 122 provides three output angle diversity signals 124 with equal amplitude signals that are phase-shifted relative to each other by 120 degrees.
[0156] In addition, Figure 2 In this configuration, the angle diversity signal 124 is detected by three photodiodes 126. The optical measurement unit 120, which includes a 120-degree optical hybrid coupler 122, can utilize, for example, three single-ended photodiodes 126. Each photodiode 126 is positioned to detect the light intensity of the corresponding output (i.e., the corresponding angle diversity signal 124) from the 120-degree optical hybrid coupler 122.
[0157] In this setup, the 120-degree optical hybrid coupler 122 splits the incoming signals 114a and 114b into three separate paths, each with a 120-degree phase difference. Then, a photodiode 126 is used to generate three detected angle diversity signals S1, S2, and S3 based on the corresponding signals from these three separate paths. The angle diversity signals can be represented as:
[0158]
[0159] Where A pd It refers to the responsivity of photodiode 126.
[0160] In a particular example, the optical measurement unit 120, which includes a 120-degree optical hybrid coupler 122, may utilize three photodiodes and / or three TIAs (not shown).
[0161] Figure 3 Another optical measurement unit 120 is shown. To avoid excessive repetition, it is understood that... Figure 3 It shows the relationship with Figure 2 The optical measurement unit 120 is discussed in a similar manner. However, in Figure 3 In this context, the optical hybrid coupler 122 is a 90-degree optical hybrid coupler 122. This type of hybrid coupler is designed to produce a 90-degree phase shift between the output signals.
[0162] The 90-degree optical hybrid coupler 122 splits the input signals 114a and 114b into four output signals (i.e., four angle diversity signals 124).
[0163] A 90-degree optical hybrid coupler can achieve this separation using the principle of multimode interference. Specifically, although Figure 3 It is not explicitly shown, but the 90-degree hybrid coupler 122 may include a 2x4 MMI, a 4x4 MMI, or a directional coupler.
[0164] Then, the four output signals from the 90-degree optical hybrid coupler 122 are detected by four photodiodes 124. These photodiodes 124 are arranged to form two balanced photodiodes 128. Each balanced photodiode 128 measures the light intensity difference between the two output angle diversity signals 124 (this differential measurement helps to cancel common-mode noise and enhance the accuracy of the detected signal).
[0165] A balanced photodiode 128 processes four output angle diversity signals 124 to generate two detected angle diversity signals S1 and S2. These detected angle diversity signals S1 and S2 have a 90-degree phase difference (i.e., α = 90°). 0 The conversion from four signals to two signals is achieved by a balanced photodiode 128, which combines the intensity information from each pair of output angle diversity signals to generate the detected angle diversity signals S1 and S2.
[0166] In a particular example, the optical measurement unit 120, which includes a 90-degree optical hybrid coupler 122, may utilize four photodiodes forming two balanced photodiodes, and / or two TIAs (not shown).
[0167] Figure 4 A block diagram of the signal processing steps of the laser system 100 is shown.
[0168] from Figure 4 As can be seen, the laser system 100 includes an ADC and a DAC. The ADC can be arranged inside or outside the control unit 130, and the DAC can be arranged inside or outside the control unit 130.
[0169] like Figure 4 As shown, the ADC converts the analog input signal input to the control unit 130 into a digital signal based on at least two angle diversity signals 124.
[0170] Furthermore, the control unit 130 converts the digital signal to the complex domain. The digital signal can be represented as a vector in the complex plane in the complex domain. Based on the detected angle diversity signal for the 120-degree optical hybrid coupler, the complex signal IQ(t) can be expressed as:
[0171]
[0172] The phase component of the beat frequency signal (e.g., S1) can be extracted. Specifically, as described herein, the angle diversity signal (i.e., the phase diversity amplitude signal) is used to calculate the phase signal of the beat frequency signal. This is the opposite of the method that uses Hilbert transform to extract the phase and where the beat frequency signal is a real number signal.
[0173] In other words, angle (i.e., phase) diversity detection is used to obtain angle (i.e., phase) diversity beat frequency signals S1, S2, and S3. Two or more detections providing angle (phase) information are provided to provide the diversity of angle detection. This can be used to simply determine the angle based on these detections.
[0174] Real signal instantaneous phase (φ) FMCW (t)-φ FMCW (t-τ) can be extracted by using mathematical calculations to convert to the complex domain (IQ without DC). IQ represents the in-phase (I) and quadrature (Q) components of the signal, and "without DC" means that the DC component is removed or not considered.
[0175] In a sense, the IQ signal is another type of S1 signal, but in the complex domain (i.e., with the same phase). By creating the IQ signal, the phase of the S1 signal can be easily extracted (e.g., because the phase of a complex signal is easier, more real-time, and more accurate to extract).
[0176] Then, the phase of signal S1 corresponds to the instantaneous frequency of the laser. It can be understood that the instantaneous phase ((φ) FMCW (t)-φ FMCW (t-τ)) corresponds to the finite difference in the laser phase, which in turn corresponds to the estimation of the instantaneous laser frequency.
[0177] Therefore, the phase of the beat frequency signal can be extracted in the following way:
[0178]
[0179] Interference is measured using a fixed delay, denoted by τ, which is small but not zero. If τ approaches zero, there will be no beat frequency signal because the signal will be identical with no delay. Conversely, if τ is relatively large (or too large), the accuracy of the measurement will be affected because the difference between the phase measured at t and t-τ is taken into account.
[0180] Furthermore, if τ is zero, there is no beat frequency signal because the result essentially corresponds to the DC component. Therefore, this method can use the finite difference of phase to estimate the phase derivative or instantaneous frequency. This requires that τ should not be zero. However, τ can be appropriately chosen based on the rate of change of the laser frequency. A smaller τ can enhance the accuracy of the method, but at the cost of a reduced signal-to-noise ratio (SNR).
[0181] For situations where the laser frequency changes rapidly, τ can be small enough (i.e., smaller than the rate of frequency change) to accurately track these changes. Conversely, if the laser frequency changes slowly, a larger τ can be used without compromising accuracy.
[0182] To achieve this, the delay line described in this paper can be a short delay line, for example, for recording instantaneous frequencies. However, this delay cannot be too small, as it would eliminate the beat frequency signal. The length of the delay line can be adapted or optimized based on these constraints. Specifically, τ can be chosen based on the rate of change of the laser frequency.
[0183] In some cases, it may be feasible to derive the instantaneous frequency using a small fixed delay τ in a delay line by analyzing how the angle of a complex signal changes with time τ. For applications where frequency changes are relatively slow, such as LiDAR, the delay can be selected based on the fastest changing frequency. This delay can be applied to slower frequencies while still providing good or sufficient SNR. For example, a 5cm delay can be used for chirp rates ranging from kHz to MHz.
[0184] In one example, the beat frequency signals of the three output angular diversity signals 124 of the method based on a 120-degree optical hybrid coupler have a 120-degree offset, such as Figure 2 As shown. Three detection signals from three single-ended photodiodes are converted from the analog domain to the digital domain by an ADC (e.g., three ADCs). On the other hand, for the method based on a 90-degree optical hybrid coupler, two beat frequency signals with a 90-degree phase shift are retrieved from two balanced photodiodes (e.g., Figure 3 (As shown).
[0185] For the 120-degree case, the control unit 130 samples three signals S1, S2, and S3; for the 90-degree case, it samples two signals S1 and S2.
[0186] Therefore, for the 120-degree case, the resulting three-dimensional vector in the complex domain can be formed by S1 + S2*exp((j2π / 3) + S3*exp(j4π / 3), where j is a complex number. Thus, the second angle diversity signal S2 is rotated 120 degrees (j2π / 3) relative to the first angle diversity signal S1 in the complex plane, and the third angle diversity signal S3 is rotated 120 degrees (j4π / 3) relative to the second angle diversity signal S2 in the complex plane. Similarly, for the 90-degree case, the two detected angle diversity signals S1 and S2 form a two-dimensional vector in the complex domain, represented by S1 + S2*exp(jπ / 2). Therefore, the second angle diversity signal S2 is rotated 90 degrees (jπ / 2) relative to the first angle diversity signal S1 in the complex plane.
[0187] The instantaneous phase angle of the signal in the complex domain can then be calculated as follows. This phase angle corresponds to the instantaneous frequency of the FMCW optical signal.
[0188] The phase angle can be calculated by determining the angle of the resulting vector. For the case of 120 degrees, use equation (1), and for the case of 90 degrees, use equation (2):
[0189] Angle = UNWRAP(ANGLE(S1+S2*exp(j2π / 3)+S3*exp(j4π / 3))(1)
[0190] Angle = UNWRAP(ANGLE(S1+S2*exp(jπ / 2))(2)
[0191] In equations (1) and (2), j is a complex number, ANGLE is a calculator for calculating vector angles, and UNWRAP is a calculator for expanding radian phase angles by adding multiples of ±2π.
[0192] Therefore, the phase angle can be calculated. In other words, the control unit 130 here calculates the frequency / phase (i.e., phase angle) corresponding to at least two angular diversity signals 124 by creating a vector and calculating the angle of the vector.
[0193] In addition, for example, the ANGLE and UNWRAP calculators are calculators with corresponding names in the MATLAB programming and numerical computation program provided by MathWorks Inc. in Natick, Massachusetts, USA.
[0194] Specifically, the UNWRAP calculator adds a multiple of 2π to the successive phase angle of the phase angle ramp when the phase difference between the phase angle of the (phase angle ramp) and the successive phase angle of the (phase angle ramp) is greater than or equal to π.
[0195] Specifically, if the phase difference between the phase angle and the successive phase angles of the phase angle ramp is greater than or equal to π, the operation of adding a multiple of 2π to the successive phase angles of the phase angle ramp is sometimes called an "unwrap" operation. The purpose of performing the unwrap operation on the extracted phase angle ramp is to eliminate 2π or 360° jumps in the extracted phase angle ramp.
[0196] Furthermore, in equations (1) and (2), signals S1, S2, and S3 are real signals. As seen in equations (1) and (2), signals S1, S2, and S3 are combined to form a vector in the complex domain. In other words, the instantaneous frequency / phase of the resulting beat frequency signal can be measured by synthesizing a complex signal using multiple real beat frequency signals.
[0197] Therefore, as in Figure 4 As seen in the lower part (i.e., the extension of the processing steps phase calculation and accumulation), the phase angle ramp of the FMCW optical signal can be extracted over time.
[0198] It will become clear that the wavelength of the FMCW optical signal is mapped non-linearly to the driving voltage (i.e., power) of the tunable laser, and the mapping between driving voltage and time is a linear curve. Therefore, the mapping between phase angle and time (similar to the mapping between angle and voltage) is a non-linear curve, as... Figure 4 As shown.
[0199] On the other hand, a linear angle-time (angle-voltage) mapping will correspond to a linear frequency chirp. Therefore, the degree of nonlinearity can be estimated using the measured nonlinear mapping.
[0200] like Figure 4 As shown, after estimating the nonlinearity, predistortion calculation is performed. The predistortion calculation may include determining a predistortion curve (i.e., a predistortion voltage ramp) based on the nonlinearity estimate. The predistortion curve is then used to generate a control signal 132 via a DAC. In other words, the predistortion curve is used to update the drive voltage of the tunable laser 110 via a DAC.
[0201] The predistortion voltage ramp can be calculated from the estimated phase nonlinearity, for example, through least mean square or linear interpolation. For example, frequency chirp linearization can be implemented iteratively. To do this, the calculated voltage ramps (up and down) can be stored and applied to subsequent iterations. This iterative approach can also be implemented in the control unit 130.
[0202] exist Figure 4 In the diagram shown in control unit 130, the process of retrieving the phase angle curve (solid line) using input angle diversity signals (three in this case) is illustrated, which is then used to form the predistortion curve (dashed line curve).
[0203] Although not shown in the figure, the reference phase angle ramp can be subtracted from the extracted phase angle ramp. Therefore, real-time locking of the frequency chirp to a specific ramp signal can be achieved. In other words, an optoelectronic phase-locked loop (OEPLL) can be provided. The OEPLL can, for example, stabilize the frequency and phase of the FMCW optical signal 112.
[0204] The term "reference phase ramp" here refers to a phase ramp whose rate of change is much slower than the feedback rate (e.g., 100 times slower).
[0205] By subtracting the reference phase angle ramp from the extracted phase angle ramp, the remaining value is the estimation error. This estimation error can be used for further calculations, such as (but not limited to) scaling, integral calculations, and / or summation of nonlinear compensations.
[0206] Figure 5A A LiDAR system 200 is schematically shown, which includes a laser system 100 (e.g., as per [reference]). Figure 1-4 (As discussed) and LIDAR detection unit 202. LIDAR detection unit 202 receives reflected light signal 118 based on FMCW light signal 112.
[0207] Specifically, the LIDAR detection unit 202 of the LIDAR system 200 detects the LIDAR response signal (i.e., the reflected light signal 118). Therefore, the LIDAR response signal can be based on the propagation portion 116 of the FMCW light signal 112 (i.e., the LIDAR emitted signal). The LIDAR detection unit 202 can be, for example, a photodiode such as a balanced and / or unbalanced photodiode, a photomultiplier tube (PMT), and / or an image detector.
[0208] Before transmitting the LIDAR transmit signal from LIDAR system 200, the LIDAR transmit signal is typically separated from the propagation portion 116 of the FMCW optical signal 112. The portion separated from the propagation portion 116 can be referred to as the local oscillator signal. The LIDAR transmit signal can then be transmitted toward target 204. Upon reaching target 204, at least some of the LIDAR transmit signal can be reflected back to LIDAR system 200. The reflected LIDAR transmit signal (i.e., reflected optical signal 118) and the local oscillator signal can then be recombined at LIDAR system 200. When these two signals are combined, interference may occur between the optical signals. The resulting combined signal can form a beat frequency. This beat frequency is related to the distance between LIDAR system 200 and target 204. In this LIDAR system 200, the combined signal can be detected by LIDAR detection unit 202, thereby detecting the reflected optical signal 118 (i.e., the LIDAR response signal).
[0209] In addition, Figure 5A In the block diagram depicting a LIDAR system 200, the LIDAR system 200 includes a tunable laser 110 (e.g., a fast-frequency chirped laser tunable by phase or gain), an optical measurement unit 120 (e.g., for real-time chirped tracking), and a control unit (for complex domain nonlinearity estimation and predistortion calculation). Here, the angle diversity signal 124 output by the optical measurement unit 120 is converted by an ADC before entering the control unit 130. Then, the predistortion curve (as described in the diagram) is used... Figure 4 (As discussed) the inherent nonlinearity in frequency chirp is compensated by control signal 132. Control signal 132 is output from control unit 130 via DAC 136.
[0210] exist Figure 5A In addition, the LIDAR system 200 also includes an optical phased array or optical antenna 206 for dynamically manipulating the propagation portion 116 of the FMCW optical signal 112 within the field of view.
[0211] Ideally, the frequency chirp is linear, and the distance resolution of the LIDAR system 200 (based on the FMCW optical signal 112) is given by equation (3):
[0212]
[0213] Where c[m / s] is the speed of light in a vacuum, and f exc It is the offset frequency of the tunable laser 110.
[0214] However, when the frequency chirp is non-linear, the peak width of the corresponding beat frequency spectrum increases, thereby reducing distance accuracy (see...). Figure 5B(the lower part). The peak broadening or widening caused by the nonlinearity of the frequency chirp can be quantified by equation (4):
[0215]
[0216] Where D[m] is the distance to target 204, and υ nl,rms [Hz] is the root mean square of the nonlinear component of the (ramp-up or ramp-down) frequency chirp. From equation (4), it can be seen that when the laser nonlinearity υ nl,rms When unsuppressed, the range resolution ΔD increases with increasing target distance D. In other words, the effect of chirp nonlinearity is more significant for distant targets 204. Therefore, high chirp linearity is beneficial, for example, in a LIDAR system 200 capable of measuring long target distances (e.g., on the order of hundreds of meters).
[0217] For example, the LIDAR system 200 can be integrated into a motor vehicle. Therefore, the LIDAR system 200 can provide accurate real-time distance and speed data (even over long distances) of the vehicle's surroundings.
[0218] In summary, the methods and systems described in this paper achieve high-quality linear chirping in FMCW laser systems with a low-complexity architecture and enhanced robustness against environmental changes.
[0219] In the foregoing, the inventive concept has been described primarily with reference to a limited number of examples. However, as will be readily understood by those skilled in the art, other examples besides those disclosed above are equally possible within the scope of the inventive concept as defined by the appended claims.
Claims
1. A laser system (100) for generating a frequency-modulated continuous wave (FMCW) optical signal (112), the system (100) comprising: A tunable laser (110) for generating the FMCW optical signal (112); An optical measurement unit (120) is configured to receive a portion (114) of the FMCW optical signal (112) and output at least two angle diversity signals (124) via an optical hybrid coupler (122) based on the difference between a first signal (114a) and a second signal (114b) formed by splitting the portion (114) of the FMCW optical signal (112), wherein the second signal (114b) is delayed relative to the first signal (114a), and wherein a pair of signals in the at least two angle diversity signals (124) have a fixed phase shift relative to each other; as well as Control unit (130), the control unit (130) being configured to: Receive the at least two angle diversity signals (124), Based on the at least two angle diversity signals (124), the compensation required to adjust the nonlinearity of the frequency chirp of the tunable laser (110) is estimated, and Output a corresponding control signal (132) to the tunable laser (110). The control signal (132) is configured to improve the linearity of the frequency chirp.
2. The laser system (100) according to claim 1, characterized in that, The optical measurement unit (120) includes: An interferometer structure configured to split the portion (114) of the FMCW optical signal (112) into a first signal (114a) and a second signal (114b), and to delay the second signal (114b) relative to the first signal (114). The optical hybrid coupler (122) is configured to output the at least two angle diversity signals (124) based on the interference between the first signal (114a) and the second signal (114b), and / or At least two photodiodes for detecting the at least two angle diversity signals (124).
3. The laser system (100) according to claim 1 or 2, characterized in that, The fixed phase shift is the same between any pair of the at least two angle diversity signals (124).
4. The laser system (100) according to any one of the preceding claims, characterized in that, The control unit (130) is configured to repeatedly output a corresponding control signal (132) to the tunable laser (110), wherein the control signal (132s) is configured to repeatedly increase the linearity of the frequency chirp.
5. The laser system (100) according to any one of the preceding claims, characterized in that, The laser system (200) is integrated on a single semiconductor chip.
6. The laser system (100) according to any one of the preceding claims, characterized in that, The optical hybrid coupler (122) includes at least one multimode interferometer (MMI).
7. The laser system (100) according to any one of the preceding claims, characterized in that, The control unit (130) is also configured to calculate the frequency and / or phase corresponding to the at least two angular diversity signals (124) by creating a vector and calculating the angle of the vector.
8. The laser system (100) according to any one of the preceding claims, characterized in that, It also includes an analog-to-digital converter (ADC) and a digital-to-analog converter (DAC), wherein the ADC is configured to convert an analog input signal input to the control unit (130) into a digital signal based on the at least two angle diversity signals (124), wherein the control unit (130) is configured to convert the digital signal to the complex domain, calculate the phase angle, estimate the nonlinearity, perform predistortion calculation, and / or output the processed digital signal, wherein the DAC is configured to convert the processed digital signal into the control signal (132).
9. The laser system (100) according to any one of the preceding claims, characterized in that, The control unit (130) also includes a reference signal, wherein the control unit (130) and the measurement unit (120) form an opto-phase-locked loop (OEPLL) configured to stabilize the frequency and phase of the FMCW optical signal (112).
10. A LIDAR system comprising a laser system (100) according to any one of claims 1-9 and a LIDAR detection unit, wherein the LIDAR detection unit is configured to receive a reflected light signal based on the FMCW light signal (112).
11. The LIDAR system according to claim 10, characterized in that, The LIDAR system is integrated into the motor vehicle and configured to provide real-time distance and speed data of the environment surrounding the motor vehicle.
12. A method for improving the linearity of the frequency chirp of a frequency modulated continuous wave (FMCW) optical signal (112), the method comprising: The FMCW optical signal (112) is generated via a tunable laser (110); A portion (114) of the FMCW optical signal (112) is received at the optical measurement unit (120); Based on the difference between a first signal (114a) and a second signal (114b) formed by splitting the portion (114) of the FMCW optical signal (112), at least two angle diversity signals (124) are output via an optical hybrid coupler (122), wherein the second signal (114b) is delayed relative to the first signal (114a), and one pair of the at least two angle diversity signals (124) has a fixed phase shift relative to each other; The at least two angle diversity signals (124) are received at the control unit (130); Based on the at least two angle diversity signals (124), the compensation required to adjust the nonlinearity of the frequency chirp of the tunable laser (110) is estimated. as well as Output a corresponding control signal (132) to the tunable laser (110). The control signal (132) improves the linearity of the frequency chirp.
13. The method according to claim 12, characterized in that, The steps for estimating the compensation required to adjust the nonlinearity of the frequency chirp of the tunable laser (110) include: The analog signal is converted into a digital signal based on the at least two angle diversity signals (124). Convert the digital signal to the complex domain. Calculate the phase angle. Estimate nonlinearity. Perform predistortion calculation. Output the processed digital signal, and / or The processed digital signal is converted into the control signal (132).
14. The method according to claim 12 or 13, characterized in that, It also includes repeating the steps described in claim 12.
15. The method according to any one of claims 12-14, characterized in that, At the optical measurement unit: At the interferometer structure, the portion of the FMCW optical signal is split into the first signal and the second signal. The second signal is delayed relative to the first signal; as well as The at least two angle diversity signals are detected via at least two photodiodes.