High-speed microcavity soliton optical frequency comb generation control system based on FPGA chip

Abstract

The application belongs to the technical field of laser, and particularly relates to a high-speed microcavity soliton optical frequency comb generation control system based on an FPGA chip. The application utilizes the high programmability and real-time performance of the FPGA chip to realize real-time monitoring and collection of parameters of each optoelectronic device of an optical frequency comb generation link, and to process monitoring data, and introduces a machine learning scheme to classify and process spectrometer data. Through the combination of an adaptive control algorithm and a real-time feedback system, intelligent adjustment and optimization of the parameters of each optoelectronic device of the optical frequency comb generation link are realized, so as to reduce the complexity of operation and improve the efficiency and stability of the microcavity soliton optical frequency comb generation process. The system has wide application value in the fields of optical frequency measurement, spectrum analysis, fiber communication and the like, and provides an efficient and stable optoelectronic technology solution for corresponding industries.

Description

Technical Field

[0001] This invention belongs to the field of laser technology, specifically relating to a high-speed microcavity soliton optical frequency comb generation and control system based on an FPGA chip. It can be widely used in optical fiber communication, optoelectronic sensing technology, laser research and development, optical frequency measurement, and spectral analysis and other related fields. Background Technology

[0002] An optical frequency comb is a special type of laser spectrum in which the frequency components are arranged at equal intervals like the teeth of a comb. It is used in fields such as precision spectral measurement, laser frequency control, and fiber optic communication.

[0003] However, commonly used optical frequency comb generation methods, such as nonlinear optical processes, such as the Kerr effect or the four-wave mixing process of microcavities, require precise control of laser parameters, resulting in limited stability and real-time performance.

[0004] Microcavity soliton optical frequency combs, as an emerging optical frequency comb source, can provide equally spaced frequencies with high precision. However, their generation and stability heavily depend on the precise control of microcavity physical conditions, such as the thermal balance within the microcavity. Implementing these adjustments requires sophisticated equipment and techniques, and also places high demands on the operator's professional skills. Furthermore, the generation and stability of microcavity soliton optical frequency combs are extremely sensitive to the physical conditions of the microcavity and the parameters of the input laser. This means that in actual operation, multiple rounds of trials and adjustments may be required to achieve the desired results. Therefore, to improve the stability and real-time performance of microcavity soliton optical frequency combs, we need a system capable of monitoring and adjusting these key parameters in real time. Summary of the Invention

[0005] To address the aforementioned problems and shortcomings, and to resolve the issue of poor stability in microcavity soliton optical frequency combs, this invention provides a high-speed microcavity soliton optical frequency comb generation and control system based on an FPGA chip. The system of this invention can detect and control various conditions required for microcavity soliton optical frequency comb generation in real time, including the wavelength and power of the pump laser, the physical conditions of the microcavity, etc., to ensure stable generation of the microcavity soliton optical frequency comb, and possesses extremely high real-time performance and ease of use.

[0006] A high-speed microcavity soliton optical frequency comb generation and control system based on FPGA chip includes: an optical frequency comb generation device and a modulation link.

[0007] The optical frequency comb generating device is divided into two links:

[0008] The first link is the pump optical path, which includes: a narrow-linewidth laser, a first electronic polarization controller, a first fiber amplifier, a first coupler, and a micro-ring cavity. During operation, the narrow-linewidth laser sequentially passes the ultra-short-linewidth laser light through the first electronic polarization controller, the first fiber amplifier, and the first coupler before entering the micro-ring cavity, providing the light source conditions for the generation of the micro-cavity soliton frequency comb.

[0009] The second link is the auxiliary optical path, which includes an auxiliary laser, a second electronic polarization controller, a second fiber amplifier, a second coupler, and a micro-ring cavity. During operation, the laser emitted from the auxiliary laser passes sequentially through the second electronic polarization controller, the second fiber amplifier, and the second coupler before entering the micro-ring cavity. Its function is to maintain thermal equilibrium within the micro-ring cavity and improve the stability of the optical frequency comb. By utilizing the thermal effect of the auxiliary laser, the micro-cavity soliton optical frequency comb can be controlled and manipulated more effectively, thereby achieving more stable and predictable state transitions.

[0010] The modulation link consists of an FPGA chip, an environmental sensor and parameter control device, a signal generator, a spectrometer, an optical power meter, and information processing equipment (such as a computer).

[0011] The spectrometer is connected to a second coupler, a second driver module of the FPGA chip, and a signal detection module.

[0012] The optical power meter is connected to the first and second couplers and the signal detection module of the FPGA chip.

[0013] The environmental sensing and parameter control device (independent of the FPGA chip) is used to detect and adjust the temperature conditions of the micro-annular cavity to the expected temperature to ensure its repetition rate stability.

[0014] The signal generator produces various waveform signals, which are connected to the first driving module of the narrow linewidth laser and the FPGA chip. By controlling the generated waveforms, the wavelength of the pump laser is indirectly and precisely adjusted, thereby improving the stability of the microcavity soliton optical frequency comb.

[0015] The FPGA chip is divided into a signal detection module, a signal processing module, and three drive modules, which are connected to the information processing equipment to process all control signals and acquired data. These five modules are connected to the spectrometer, optical power meter, narrow linewidth laser, auxiliary laser, and two electronic polarization controllers, respectively.

[0016] The signal detection module monitors information from the spectrometer and optical power meter in real time, converts it into digital signals, and sends it to the signal processing module. By monitoring changes in the spectrum and optical power in real time, it provides data support for adjusting the parameters of the electronic polarization controller and laser.

[0017] The signal processing module filters and amplifies the received signal, extracts key features, performs relevant information calculations, and then sends control signals to the three drive modules based on the expected calculation results.

[0018] The first drive module connects the narrow linewidth laser and the auxiliary laser, and adjusts their switching state and operating wavelength according to the received control signal to maintain the stability of the optical frequency comb; at the same time, it is also connected to the signal generator to further finely adjust its bias voltage, signal waveform and other operating parameters.

[0019] The second drive module connects to the spectrometer, adjusts its resolution and scanning speed, and automatically stores the spectrum when a specific state is reached; the specific state refers to the primary comb, secondary comb, multi-soliton state, or single-soliton state in the soliton generation division.

[0020] The third drive module connects to the first and second electronic polarization controllers and adjusts the polarization according to the received control signals to meet the optimal conditions for soliton generation.

[0021] The system architecture and linkage method of the present invention achieve high-speed and precise control over the microcavity soliton optical frequency comb generation process, thereby improving the ability to generate stable and reliable microcavity soliton optical frequency combs.

[0022] Furthermore, the key features include the number of comb teeth, the standard deviation of the comb tooth spacing, the difference in comb tooth power, the number of comb teeth near the primary comb, and the presence or absence of non-periodic comb teeth.

[0023] Furthermore, the signal processing module automatically adjusts the electronic polarization controller through the third drive module based on the power signals received from both ends of the optical power meter using a proportional-integral-derivative PID control strategy, so as to achieve the relative maximum power of the two optical paths.

[0024] Furthermore, in the process of generating the microcavity soliton optical frequency comb, this invention also applies a machine learning scheme to process the spectrometer data. Based on the data characteristics, the generation of solitons is divided into four states: primary comb, secondary comb, multi-soliton state, and single soliton state; then, the typical characteristics of these four states are programmed into the FPGA chip, enabling it to quickly identify and provide real-time feedback on the state of the optical cavity.

[0025] Data preprocessing: Smoothing and normalizing the raw data acquired by the spectrometer.

[0026] Key features were extracted, including: number of comb teeth, standard deviation of comb tooth spacing, difference in comb tooth power, number of primary comb teeth in the 30-50nm wavelength range, and presence of non-periodic comb teeth.

[0027] Furthermore, regarding the key feature selection: these five features can be directly used for classification. In practical applications, feature selection algorithms can be used to further optimize and determine the subset of features to be used.

[0028] Model Training: A multi-class classifier is trained using a labeled training dataset (containing samples of single solitons, multiple solitons, primary comb states, and secondary comb states). A decision tree machine learning algorithm is used in this setup. The optimal model is selected through cross-validation.

[0029] Model Testing and Application: The trained multi-class classifier is used to predict new spectrometer data to determine whether it is a single soliton, multiple soliton, primary comb, or secondary comb state. The performance of the classifier is improved by continuously updating the training dataset and adjusting the model parameters.

[0030] Furthermore, based on the obtained microcavity soliton state information, the output wavelength of the narrow-linewidth laser is initially adjusted in steps of 0.5 nm to lay the foundation for subsequent precise adjustment. Next, the FPGA chip activates a state machine-based switching strategy corresponding to the soliton state. When the soliton state changes, such as from a secondary comb to a multi-soliton state, the FPGA chip automatically switches the adjustment target from the pump laser to the signal generator. For the signal generator, its output bias voltage is adjusted with various step sizes, such as 0.5V, 0.1V, 0.05V, and 0.01V, to meet our needs for indirect adjustment of the narrow-linewidth laser wavelength. If, during wavelength adjustment, the microcavity soliton transitions from its initial state (which could be a primary comb, secondary comb, or multi-soliton) to the next state (which could be a secondary comb, multi-soliton, or single-soliton state), the adjustment process stops. Conversely, if the state regresses, such as returning from a multi-soliton state to a secondary comb state during wavelength adjustment, the adjustment process stops. The real-time feedback system will backtrack to the laser parameters of the previous step, so that the microcavity soliton state returns to the previous state.

[0031] The core of this invention lies in the adaptive control algorithm and the real-time feedback system. Their combination optimizes the performance of the spectrometer, erbium-doped fiber amplifier (EDFA), and laser, providing high flexibility and accuracy. A photoelectric detection device provides the optical signal to be processed, and this data is sent to the FPGA chip for real-time adaptive adjustment to improve efficiency and stability.

[0032] The adaptive control algorithm is based on a multi-class classifier using SVM, with training data sourced from a spectrometer. After preprocessing, feature extraction, and selection, features reflecting soliton states (single soliton, multiple solitons, primary comb, and secondary comb) are fed into the classifier for training. The trained classifier can predict soliton states based on new spectrometer data, and its performance is improved by continuously updating the training data and adjusting the model parameters.

[0033] The real-time feedback system is based on Q-Learning, a reinforcement learning algorithm. Through continuous experimentation and learning, the Q-Learning algorithm gradually learns how to select the optimal laser parameter adjustment scheme under different soliton states, thereby achieving real-time control of the microcavity soliton optical frequency comb generation process. This control considers both the current state (i.e., the data provided by the photoelectric detection device) and past learning experience (i.e., the Q-value table), and can adaptively adjust as the experiment progresses.

[0034] By combining SVM and Q-Learning, the control system of this invention reduces operational complexity, improves efficiency and stability, and provides efficient and stable solutions for fields such as optical frequency measurement, spectral analysis, and fiber optic communication.

[0035] In summary, this invention provides a high-speed microcavity soliton optical frequency comb generation and control system based on an FPGA chip. Utilizing the high programmability and real-time performance of the FPGA, and through the combined application of adaptive control algorithms and a real-time feedback system, it achieves intelligent adjustment and optimization of the parameters of the spectrometer, optical power meter, laser, and electronic polarization controller, thereby improving the efficiency and stability of the microcavity soliton optical frequency comb generation process. This system has wide application value in fields such as optical frequency measurement, spectral analysis, and fiber optic communication. Attached Figure Description

[0036] Figure 1 This is a schematic block diagram of a microcavity soliton optical frequency comb generation device;

[0037] Figure 2 The multi-soliton spectral state diagram is shown in the example.

[0038] Figure 3 The single soliton spectral state diagram is shown in the example.

[0039] Figure reference numerals: 1-Narrow linewidth laser, 2-First electronic polarization controller, 3-First fiber optic amplifier, 4-Auxiliary laser, 5-Second electronic polarization controller, 6-Second fiber optic amplifier, 7-Spectrometer, 8-First coupler, 9-Optical power meter, 10-Second coupler, 11-Micro-ring cavity, 12-Signal generator. Detailed Implementation

[0040] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments.

[0041] In the control system of this embodiment, we used Xilinx's Spartan-7 series FPGA chip, which plays a crucial role in the automatic excitation process of the microcavity soliton optical frequency comb. The FPGA chip reads the spectrometer data in real time, processes the frequency distribution information of the optical signal, and then determines the adjustment direction of the laser output wavelength. The adaptive control algorithm inside the FPGA automatically adjusts the laser wavelength according to the real-time spectral data, realizing a precise and stable microcavity soliton optical frequency comb generation process. At the same time, the real-time feedback system between the FPGA chip and the laser can adjust the laser output wavelength in real time according to the algorithm calculation results.

[0042] like Figure 1 The microcavity soliton optical frequency comb generating device used in this invention includes an optoelectronic device that can be automatically and rapidly controlled, specifically including: 1-narrow linewidth laser, 2-first electronic polarization controller, 3-first fiber optic amplifier, 4-auxiliary laser, 5-second electronic polarization controller, 6-second fiber optic amplifier, 7-spectrometer, 8-first coupler (9:1), 9-optical power meter, 10-second coupler (9:1), 11-micro-ring cavity, 12-signal generator.

[0043] The FPGA chip is divided into a signal detection module, a signal processing module, a first driver module, a second driver module, and a third driver module. In actual use, each module of the FPGA chip is connected to the corresponding optoelectronic device to receive and process the data transmitted from the spectrometer in real time.

[0044] The spectrometer parameters include a measurable power range (-70dBm to +20dBm), a wide measurable range (1200nm-2400nm), wavelength accuracy ±0.05nm, and a dynamic range of 55dB. The narrow-linewidth laser operates at approximately 1550nm. The auxiliary laser operates at approximately 1535.75nm in ACC mode. The first fiber optic amplifier (Erbium-doped fiber amplifier EDFA) operates at 1550nm with a power of 300mW in ACC mode. The second fiber optic amplifier (Erbium-doped fiber amplifier EDFA) operates at 1550nm with a power of 500mW. The electronic polarization controller operates in the 1260-1620nm wavelength range, with an insertion loss <0.5dB and a PDL of 0.05dB (including connectors).

[0045] During implementation, an environmental sensor and parameter control device is connected to the micro-annular cavity to monitor and adjust its temperature in real time. If the environmental parameters exceed the preset range, the device will make adjustments to maintain a stable repetition rate within the micro-annular cavity.

[0046] Before the frequency comb is generated, the FPGA chip receives the power signals from both ends of the optical power meter and adjusts the electronic polarization controller. The chip sets the target as the relative maximum value of the two powers and uses a PID control strategy to adjust until the power reaches the target.

[0047] During the frequency comb generation process, the FPGA chip uses machine learning to set feature parameters, including comb power and comb spacing, and generates a training dataset. After training the model, it is embedded into the FPGA chip to monitor and quickly identify the current state in real time.

[0048] In the process of soliton generation, this invention designs a signal processing algorithm. This algorithm determines the current soliton state by judging parameters such as comb power and spectral characteristics. First, in the primary comb state, the comb tooth spacing is set to multiple complete FSRs. If it is observed that the spacing between two comb teeth is less than the FSR range determined during model training, it is determined that the system has transitioned to the secondary comb state. We can understand and control the physical system by controlling and observing the comb tooth spacing.

[0049] Furthermore, we determine whether a multi-soliton state has been entered through spectral characteristics. When the secondary comb state exhibits periodic peaks and the envelope of the spectral signal does not have a sech function shape, we consider it to have entered a multi-soliton state (e.g., ...). Figure 2 As shown). After adjustment, the multi-soliton successfully transformed into a single-soliton (as shown). Figure 3 (As shown).

[0050] During the determination of characteristic parameters, the wavelength of the frequency comb generating device (mainly the pump laser) needs to be adjusted. This requires connecting the signal processing module of the FPGA chip to the pump laser. Furthermore, adjustments need to be made in real-time and at high speed based on the signal received by the spectrometer. Based on the transmission rate of the spectrometer signal, we can set it to 115200 bit rate and adjust the signal processing module accordingly. Specifically, this involves simultaneously connecting the pump laser and the signal generator to the signal processing module.

[0051] Furthermore, by adjusting the pump laser wavelength over a wide range, the first step is to find a wavelength range that allows the spectrometer to transition from the primary comb to the secondary comb. The FPGA chip's state machine strategy responds to soliton state changes, such as transitioning from the secondary comb to a multi-soliton state; it automatically switches the adjustment target from the pump laser to the signal generator.

[0052] Furthermore, the FPGA chip implements a state machine switching strategy corresponding to the soliton state. When the soliton state changes, such as transitioning from a secondary comb state to a multi-soliton state, the FPGA chip automatically changes its adjustment target, shifting from the pump laser to the signal generator. The signal generator is then used to make small adjustments to the wavelength of the pump laser. When adjusting the pump laser wavelength, we can indirectly adjust the pump laser wavelength by changing the signal waveform generated by the signal generator.

[0053] Specifically, this involves adjusting the waveform of the signal generator, transforming it from a periodic triangular wave signal (100% symmetry) into a simple DC signal, or generating a DC signal with positive and negative voltages, which is then input into the pump laser. The voltage of the signal generator is adjusted according to the actual situation. By changing the output voltage of the signal generator, minute adjustments to the pump laser wavelength are achieved. Simultaneously, the maximum voltage is adjusted to avoid over- or under-adjustment of the wavelength. During this process, by collecting data and using machine learning algorithms, the optimal wavelength adjustment path from the secondary comb to multiple solitons, and then to single solitons, is identified, enabling high-speed automatic adjustment of the pump laser. During the adjustment process, the operating wavelength and output power of the auxiliary laser are also adjusted in a timely manner to help the microcavity soliton frequency comb smoothly enter the single soliton state.

[0054] Generally, when the microcavity soliton frequency comb enters the multi-soliton state, we intelligently adjust the parameters based on data from the signal processing module. If the auxiliary light is determined to be under-detuned (or over-detuned), the FPGA chip will adjust in the opposite direction. Upon successfully entering the single-soliton state, the second drive module begins operation, connecting to the spectrometer, adjusting its resolution and scanning speed, and automatically storing the spectrum when a specific state is reached. We provide control software to monitor the state transition process, monitoring the laser's wavelength, power, and state, and also allows for manual adjustments to optimize the overall device performance.

[0055] As can be seen from the above embodiments, this invention utilizes the high programmability and real-time performance of FPGA chips to monitor, collect, and process the parameters of each optoelectronic device in the optical frequency comb generation link in real time. It also introduces a machine learning scheme to classify and process the spectrometer data. Through the combined application of adaptive control algorithms and a real-time feedback system, it achieves intelligent adjustment and optimization of the parameters of each optoelectronic device in the optical frequency comb generation link, thereby reducing operational complexity and improving the efficiency and stability of the microcavity soliton optical frequency comb generation process. This system has wide application value in optical frequency measurement, spectral analysis, and fiber optic communication, providing efficient and stable optoelectronic technology solutions for corresponding industries.

Claims

1. A high-speed microcavity soliton optical frequency comb generation and control system based on an FPGA chip, characterized in that: Includes optical frequency comb generation device and modulation link; The optical frequency comb generating device is divided into two links: The first link is the pump optical path, which includes: a narrow linewidth laser, a first electronic polarization controller, a first fiber amplifier, a first coupler, and a micro-ring cavity. During operation, the narrow linewidth laser sequentially passes the ultra-short linewidth laser through the first electronic polarization controller, the first fiber amplifier, and the first coupler before entering the micro-ring cavity, providing the light source conditions for the generation of the micro-cavity soliton optical frequency comb. The second link is the auxiliary optical path, which includes: an auxiliary laser, a second electronic polarization controller, a second fiber optic amplifier, a second coupler, and a micro-ring cavity. During operation, the laser emitted by the auxiliary laser passes through the second electronic polarization controller, the second fiber optic amplifier, and the second coupler in sequence before entering the micro-ring cavity. Its function is to maintain the thermal balance in the micro-ring cavity and improve the stability of the optical frequency comb. The modulation link consists of an FPGA chip, an environmental sensor and parameter control device, a signal generator, a spectrometer, an optical power meter, and information processing equipment. The spectrometer is connected to a second coupler, a second drive module of the FPGA chip, and a signal detection module. The optical power meter is connected to the first and second couplers and the signal detection module of the FPGA chip; The environmental sensing and parameter control device is used to detect and adjust the temperature conditions of the micro-annular cavity to the expected temperature. The signal generator produces various waveform signals, which are connected to the first driving module of the narrow linewidth laser and the FPGA chip. By controlling the generated waveforms, the wavelength of the pump laser is indirectly and precisely adjusted. The FPGA chip is divided into a signal detection module, a signal processing module, and three driver modules, which are connected to the information processing equipment to process all control signals and acquired data. The signal detection module monitors the information from the spectrometer and optical power meter in real time, converts it into digital signals, and sends them to the signal processing module to provide data support for parameter adjustment of the electronic polarization controller and laser. The signal processing module filters and amplifies the received signal, extracts key features, performs relevant information calculations, and sends control signals to the three drive modules based on the expected calculation results. The first drive module connects the narrow linewidth laser and the auxiliary laser, and adjusts their switching state and operating wavelength according to the received control signal to maintain the stability of the optical frequency comb; at the same time, it is also connected to the signal generator to further finely adjust its bias voltage and signal waveform. The second drive module connects to the spectrometer, adjusts its resolution and scanning speed, and automatically stores the spectrum when a specific state is reached. The specific state refers to the primary comb, secondary comb, multi-soliton state, or single-soliton state in the division of soliton generation. The third drive module connects to the first and second electronic polarization controllers and adjusts the polarization according to the received control signals to meet the optimal conditions for soliton generation.

2. The high-speed microcavity soliton optical frequency comb generation and control system based on FPGA chip as described in claim 1, characterized in that: The key features include the number of comb teeth, the standard deviation of the comb tooth spacing, the difference in comb tooth power, the number of comb teeth near the primary comb, and the presence of non-periodic comb teeth.

3. The high-speed microcavity soliton optical frequency comb generation and control system based on FPGA chip as described in claim 1, characterized in that: The signal processing module automatically adjusts the electronic polarization controller through the third drive module based on the power signals received from both ends of the optical power meter using a proportional-integral-derivative PID control strategy, so as to achieve the relative maximum power of the two optical paths.

4. The high-speed microcavity soliton optical frequency comb generation and control system based on FPGA chip as described in claim 1, characterized in that, During the generation of the microcavity soliton optical frequency comb, the processing of the spectrometer data specifically includes: Based on the data characteristics, the generation of solitons is divided into four states: primary comb, secondary comb, multi-soliton state, and single soliton state; then the typical characteristics of these four states are programmed into the FPGA chip, enabling it to quickly identify and provide real-time feedback on the state of the optical cavity. Data preprocessing: Smoothing and normalizing the raw data acquired by the spectrometer; Key features were extracted, including: number of comb teeth, standard deviation of comb tooth spacing, difference in comb tooth power, number of primary comb teeth in the 30-50nm wavelength range, and presence of non-periodic comb teeth. Model training: A multi-class classifier is trained using a labeled training dataset containing state samples of single solitons, multiple solitons, primary combs, and secondary combs; the best model is selected using a machine learning algorithm based on decision trees through cross-validation. Model testing and application: The trained multi-class classifier is used to predict new spectrometer data to determine whether it is a single soliton, multiple soliton, primary comb, or secondary comb state; and the performance of the classifier is improved by continuously updating the training dataset and adjusting the model parameters.

5. The high-speed microcavity soliton optical frequency comb generation and control system based on FPGA chip as described in claim 4, characterized in that: The five key features are further optimized using a feature selection algorithm to determine the subset of features to be used.

6. The high-speed microcavity soliton optical frequency comb generation and control system based on FPGA chip as described in claim 4, characterized in that: Based on the obtained microcavity soliton state information, the output wavelength of the narrow linewidth laser is first initially adjusted in steps of 0.5 nm. Then, the FPGA chip enables a state machine-based switching strategy corresponding to the soliton state. When the soliton state changes from the secondary comb to the multi-soliton state, the FPGA chip automatically switches the adjustment target from the pump laser to the signal generator. For the signal generator, the bias voltage of its output is adjusted in steps of 0.5V, 0.1V, 0.05V, and 0.01V to indirectly adjust the wavelength of the narrow-linewidth laser. If the microcavity soliton moves from the initial state to the next state during the wavelength adjustment process, the adjustment process stops. Conversely, if the state regresses, it will backtrack to the laser parameters of the previous step, causing the microcavity soliton state to return to the previous state.

7. The high-speed microcavity soliton optical frequency comb generation and control system based on FPGA chip as described in claim 1, characterized in that: In the process of determining characteristic parameters, the pump laser and the signal generator are simultaneously connected to the signal processing module. First, the wavelength of the pump laser is adjusted over a wide range to find the wavelength range that allows the spectrometer state to change from the primary comb to the secondary comb. The state machine strategy of the FPGA chip responds to the soliton state change. When the soliton state changes, the FPGA chip automatically changes its adjustment target from the pump laser to the signal generator. The signal generator is controlled to make small-range adjustments to the wavelength of the pump laser. The wavelength of the pump laser is indirectly adjusted by changing the signal waveform generated by the signal generator.

Images