Distributed fiber optical modulation system based on mutual enhancement circulator

Abstract

The present invention proposes a distributed fiber optical modulation system based on a mutual enhancement circulator, which relates to the field of fiber optic communication technology and includes: multiple modulation nodes distributed along an optical fiber link, each modulation node including a circulator, an electro-optical modulator, a monitoring unit and an optical fiber array, wherein the circulator and the electro-optical modulator are connected in series to form a modulation link, the monitoring unit is used to collect the signal quality parameters of each modulation node and feed the signal quality parameters back to a central control unit, and multiple modulation nodes are connected in sequence through optical fiber links to form a topological structure; a laser is connected to the electro-optical modulator of the first modulation node of the system to provide an optical signal; a central control unit is configured to execute a modulation parameter control algorithm, generate and send a power control signal and a phase control signal for each modulation node; a parameter database is used to store system operating parameters and historical data. The present invention can improve the transmission quality of long-distance optical communications, expand the transmission capacity, and extend the transmission distance.

Description

Technical Field

[0001] The present invention relates to the field of optical fiber communication technology, and in particular to a distributed optical fiber modulation system based on a mutual enhancement circulator. Background Art

[0002] As demand for optical communication network capacity continues to grow, distributed fiber optic systems are gaining attention as a key solution. Traditional optical communication systems typically utilize a point-to-point transmission model, where signals are modulated only once at the transmitting end, resulting in significant signal quality degradation over long-distance transmission. Distributed fiber optic systems deploy multiple optical nodes along the transmission path to relay and process signals, effectively extending transmission distances and improving system reliability. Currently, distributed fiber optic systems are primarily used in long-distance optical communication networks, submarine cable systems, and large-scale optical sensor networks, and have become a key development direction in optical communication technology.

[0003] Optical modulation systems are a core technology in optical communications, responsible for converting electrical signals into optical signals and performing modulation. Common optical modulation techniques include intensity modulation, phase modulation, polarization modulation, and frequency modulation. With the continuous increase in communication speeds, advanced optical modulation techniques such as quadrature phase-shift keying (QPSK), quadrature amplitude modulation (QAM), and pulse amplitude modulation (PAM) are widely used in high-speed optical communication systems. In these modulation systems, electro-optical modulators are key components, primarily including Mach-Zehnder modulators, electro-absorption modulators, and microring resonator modulators. Circulators, as important passive optical components, exhibit nonreciprocal transmission characteristics, enabling directional transmission of optical signals. A typical three-port circulator can directionally transmit an optical signal from its input port to its output port while isolating the reverse signal. It is widely used for signal routing, isolation, and multiplexing in optical communication systems.

[0004] However, existing distributed fiber optic systems and optical modulation technologies still face numerous challenges. First, over long-distance transmission, factors such as dispersion, nonlinear effects, and power attenuation cause signal quality to continuously degrade, making it difficult for existing systems to achieve ultra-long-distance transmission while maintaining high transmission quality. Second, the interoperability between system nodes is insufficient, making it difficult to optimize node parameters based on overall network conditions. Third, existing systems are not well adapted to complex and changing transmission environments, struggling to cope with interference factors such as temperature fluctuations, mechanical vibration, and fluctuations in link characteristics. Finally, system maintenance and fault diagnosis are complex, and there is a lack of efficient automated management mechanisms. Summary of the Invention

[0005] In view of this, the present invention proposes a distributed fiber optical modulation system based on a mutually enhanced circulator. By distributing modulation nodes composed of multiple circulators and electro-optical modulators along the optical fiber link, and using a central control unit to implement intelligent algorithms such as multi-parameter collaborative power balancing, dynamic phase complementary enhancement, and statistical feature error prediction and compensation, a highly collaborative, adaptive, and predictive optical signal modulation system is constructed, thereby improving the transmission quality of long-distance optical communications, expanding transmission capacity, extending transmission distance, and enhancing the stability and reliability of the system in complex environments.

[0006] The technical solution of the present invention is achieved as follows:

[0007] The present invention provides a distributed fiber optical modulation system based on a mutual enhancement circulator, comprising:

[0008] Multiple modulation nodes distributed along the optical fiber link, each modulation node includes a circulator, an electro-optical modulator, a monitoring unit, and an optical fiber array, wherein the circulator and the electro-optical modulator are connected in series to form a modulation link, and the monitoring unit is used to collect signal quality parameters of each modulation node and feed the signal quality parameters back to the central control unit. Multiple modulation nodes are sequentially connected through the optical fiber link to form a topological structure;

[0009] a laser, connected to an electro-optical modulator of the first modulation node of the system, for providing an optical signal;

[0010] a central control unit configured to execute a modulation parameter control algorithm, generate and send a power control signal and a phase control signal for each modulation node;

[0011] Parameter database, used to store system operating parameters and historical data.

[0012] Preferably, the topology is a chain structure or a ring structure; when it is a chain structure, the system has a starting end and an ending end; when it is a ring structure, the last modulation node of the system is connected back to the first modulation node to form a closed loop.

[0013] Preferably, each of the circulators has three ports, the first port of the circulator is connected to the electro-optical modulator, the second port is connected to the downstream optical fiber, and the third port is connected to the monitoring unit; the optical signal enters the circulator after being modulated by the electro-optical modulator, and the circulator outputs the main signal from the second port to the downstream optical fiber, and at the same time outputs part of the signal from the third port to the monitoring unit.

[0014] Preferably, the modulation parameter control algorithm includes:

[0015] A multi-parameter collaborative power balancing algorithm uses signal quality parameters collected by the monitoring unit and historical data in the parameter database to construct a signal quality function. It then calculates and optimizes the output power of each modulation node through a nonlinear model.

[0016] A dynamic phase complementary enhancement algorithm calculates a phase adjustment value based on the signal quality parameters collected by the monitoring unit, and adjusts the phase relationship of each modulation node in real time so that the signals of adjacent nodes form a constructive superposition;

[0017] The statistical feature error prediction and compensation algorithm uses the signal quality parameters collected by the monitoring unit in combination with the system operating parameters stored in the parameter database to analyze the error distribution characteristics, predict the signal degradation trend and generate compensation instructions.

[0018] Preferably, the multi-parameter coordinated power balancing algorithm updates the power of each modulation node through the following iterative formula:

[0019] ;

[0020] Where, represents the power of the i-th node at the t-th iteration; is the adaptive learning rate; The objective function Q is The partial derivative at ; is the historical contribution index of the i-th node; is the nonlinear adjustment factor; It is the environmental adaptability coefficient, which is determined by temperature, vibration, humidity and atmospheric pressure.

[0021] Preferably, the historical contribution index The calculation uses the following formula:

[0022] ;

[0023] Where H is the size of the historical observation window; , represents the change in the overall signal quality of the system between iteration u and iteration u+1; , represents the power adjustment amount of the i-th node between iteration u and iteration u+1.

[0024] Preferably, the dynamic phase complementary enhancement algorithm realizes adaptive phase calibration based on a dynamic phase codebook, the dynamic phase codebook is dynamically updated according to the characteristics of the optical fiber link, and phase features are extracted through fast Fourier transform; the phase adjustment frequency of the dynamic phase complementary enhancement algorithm is 1000 times per second.

[0025] Preferably, the statistical feature error prediction and compensation algorithm uses a fractal interpolation method to construct a nonlinear prediction function, and the nonlinear prediction function uses a statistical distribution model of historical data to predict the signal degradation trend and generate a compensation scheme; the prediction function uses a second-order Markov process to describe the error propagation characteristics, and the prediction time window is 5 milliseconds to 200 milliseconds.

[0026] Preferably, the signal quality parameters include signal-to-noise ratio, bit error rate, polarization mode dispersion, chromatic dispersion, four-wave mixing effect intensity and stimulated Brillouin scattering threshold.

[0027] Preferably, the parameter database stores the following data types: system operating parameters, historical data, optical fiber link characteristic data, environmental factor data and fault record data; the system also includes an exception handling mechanism, which includes a simplified protection switching protocol, and achieves rapid isolation of fault points and service recovery through a pre-calculated backup path table. When a node failure is detected, the system completes fault isolation and service recovery within 30 milliseconds.

[0028] The present invention has the following beneficial effects compared to the prior art:

[0029] (1) The present invention forms a complete distributed optical modulation network by distributing modulation nodes composed of multiple circulators and electro-optical modulators, combined with the coordinated control of a central control unit. This structure fully utilizes the non-reciprocal transmission characteristics of the circulator and the high-speed modulation capability of the electro-optical modulator, so that each modulation node forms a mutual reinforcement effect under the unified coordination of the central control unit, which can effectively extend the effective transmission distance of the traditional single-point modulation system, while reducing the overall bit error rate of the system and improving the transmission performance of long-distance optical communications;

[0030] (2) The multi-parameter collaborative power balancing algorithm of the present invention achieves dynamic optimization of the output power of each modulation node through the historical contribution index and the environmental adaptability coefficient. This algorithm breaks through the limitations of the traditional independent node control mode, establishes a nonlinear optimization model based on global signal quality, and solves the signal attenuation and distortion problems caused by uneven power distribution among multiple nodes. This solution enables the system to maintain relatively stable signal quality in complex environments, and improves the system's anti-interference ability and power utilization efficiency;

[0031] (3) The dynamic phase complementary enhancement algorithm of the present invention is based on fast Fourier transform and dynamic phase codebook, which realizes the precise phase relationship adjustment between each modulation node. The algorithm ensures the constructive superposition of signals of adjacent nodes through high-frequency phase calibration, effectively overcoming the phase mismatch problem caused by fluctuations in transmission medium characteristics. This solution can maintain phase stability when external factors such as temperature change, improve signal enhancement efficiency, and reduce signal distortion in long-distance transmission;

[0032] (4) The statistical feature error prediction and compensation algorithm of the present invention uses fractal interpolation and a second-order Markov process to establish a prediction model for signal degradation trends. By analyzing historical error patterns, the algorithm predicts and actively compensates for signal degradation before it occurs, achieving a shift from passive response to active prevention. This solution can pre-identify potential signal degradation trends, enhance the system's fault tolerance, and improve the stability and reliability of optical communication systems in complex environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0034] Figure 1 It is a system framework diagram of the present invention;

[0035] Figure 2 It is a technical implementation diagram of the present invention;

[0036] Figure 3 This is a flow chart of the multi-parameter collaborative power balancing algorithm of the present invention;

[0037] Figure 4 This is a flow chart of the dynamic phase complementary enhancement algorithm of the present invention;

[0038] Figure 5 This is a flow chart of the statistical feature error prediction and compensation algorithm of the present invention. DETAILED DESCRIPTION

[0039] The following will be combined with the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0040] like Figure 1 As shown, the present invention provides a distributed fiber optical modulation system based on a mutual enhancement circulator, comprising:

[0041] Multiple modulation nodes distributed along the optical fiber link, each modulation node includes a circulator, an electro-optical modulator, a monitoring unit, and an optical fiber array, wherein the circulator and the electro-optical modulator are connected in series to form a modulation link, and the monitoring unit is used to collect signal quality parameters of each modulation node and feed the signal quality parameters back to the central control unit. Multiple modulation nodes are sequentially connected through the optical fiber link to form a topological structure;

[0042] a laser, connected to an electro-optical modulator of the first modulation node of the system, for providing an optical signal;

[0043] a central control unit configured to execute a modulation parameter control algorithm, generate and send a power control signal and a phase control signal for each modulation node;

[0044] Parameter database, used to store system operating parameters and historical data.

[0045] like Figure 2As shown, the technical implementation process of the present invention is as follows: First, multiple modulation nodes are distributed along the optical fiber link according to the network topology requirements (chain or ring structure). Each node consists of an electro-optical modulator, a circulator, a monitoring unit, and an optical fiber array. The electro-optical modulator and circulator are connected in series to form a modulation link. After being emitted by the laser, the optical signal directly enters the electro-optical modulator of the first modulation node for initial modulation. It then flows through each modulation node in the system. In each modulation node, the signal flows as follows: first, it is modulated by the electro-optical modulator. The modulated signal then enters the circulator, which splits the signal into two paths: the main signal is output from the circulator's second port to the downstream optical fiber for transmission to the next node, and a portion of the signal is output from the circulator's third port to the monitoring unit for signal quality monitoring. The monitoring unit performs multi-dimensional signal characteristic analysis through the connected optical fiber array, collecting various signal quality parameters (including signal-to-noise ratio, bit error rate, polarization mode dispersion, chromatic dispersion, four-wave mixing effect intensity, and stimulated Brillouin scattering threshold), and transmits these parameters to the central control unit and stores them in a parameter database. Based on these real-time and historical data, the central control unit executes a multi-parameter coordinated power balancing algorithm, a dynamic phase complementation enhancement algorithm, and a statistical feature error prediction and compensation algorithm. The multi-parameter coordinated power balancing algorithm dynamically optimizes the power of each node using historical contribution metrics and environmental adaptability coefficients. The dynamic phase complementation enhancement algorithm extracts phase features based on a dynamic phase codebook and fast Fourier transform and adjusts phase relationships in real time. The statistical feature error prediction and compensation algorithm uses fractal interpolation and a second-order Markov process to construct a nonlinear prediction function to predict signal degradation trends. The calculation results of each algorithm are converted into specific modulation parameter adjustment instructions and sent to the corresponding modulation node via the communication network. This guides the electro-optical modulator to precisely modulate the optical signal, achieving coordinated optimization and dynamic improvement of the overall system signal quality. The system also features an exception handling mechanism. When a node failure is detected, it rapidly isolates the fault and restores service using a simplified protection switching protocol and a pre-calculated backup path table, ensuring high system reliability and continuous operation.

[0046] Specifically, in one embodiment of the present invention, the distributed fiber optic optical modulation system based on the mutual enhancement circulator of the present invention adopts a distributed architecture, and multiple modulation nodes are set along the optical fiber link according to the network transmission requirements. According to the requirements of different application scenarios, the system can be configured into two topologies: a chain structure or a ring structure. When a chain structure is adopted, the system has a clear starting end and an ending end. After the optical signal is generated by the laser, it directly enters the electro-optical modulator of the first modulation node, enters the circulator after initial modulation, and then passes through the step-by-step processing of each modulation node (electro-optical modulator modulation → circulator shunting → next node), and is finally output from the ending end; when a ring structure is adopted, the last modulation node of the system is connected back to the first modulation node to form a closed loop, realizing the cyclic transmission of the signal and enhancing the redundancy and reliability of the system.

[0047] Each modulation node consists of an electro-optical modulator, a three-port circulator, a monitoring unit, and an optical fiber array. The electro-optical modulator is responsible for modulating the optical signal. The circulator, a key optical routing element, has its first port connected to the electro-optical modulator to receive the modulated optical signal; its second port is connected to the downstream optical fiber, responsible for transmitting the signal to the next node; and its third port is connected to the monitoring unit, which diverts part of the signal for monitoring and analysis. In terms of signal flow, the optical signal first enters the electro-optical modulator for modulation, and the modulated signal enters the circulator for diversion. The main signal is output from the circulator's second port to the downstream optical fiber for further transmission, and part of the signal is output from the circulator's third port to the monitoring unit for signal quality monitoring and multi-dimensional characteristic analysis.

[0048] The system's core control module is the central control unit, which establishes a stable, bidirectional data exchange link with each modulation node via independent auxiliary communication channels. The central control unit receives and processes real-time signal quality parameters collected by monitoring units at each node, including but not limited to bit error rate, optical signal-to-noise ratio, polarization mode dispersion, chromatic dispersion, four-wave mixing (FWM) intensity, and stimulated Brillouin scattering (SBS) threshold. Based on these parameters, the central control unit executes a multi-parameter collaborative power balancing algorithm, a dynamic phase complement enhancement algorithm, and a statistical feature error prediction and compensation algorithm to generate optimized modulation parameter adjustment instructions, which are then sent to the electro-optical modulators at each modulation node for execution.

[0049] The monitoring unit at each modulation node receives a portion of the diverted optical signal via a connection to the circulator's third port. It is responsible for collecting signal quality parameters in real time and transmitting these parameters to the central control unit via an auxiliary communication channel. Working in conjunction with the underlying fiber array, the monitoring unit employs high-precision photodetectors and dedicated signal processing chips to accurately capture changes in the various physical properties of the optical signal. The monitoring unit also features basic self-diagnosis capabilities. When a local device anomaly is detected, it immediately reports it to the central control unit, triggering the system's exception handling mechanism.

[0050] The system's distributed architecture allows for physically dispersed but logically tightly coordinated modulation nodes. A central control unit coordinates the electro-optical modulator parameters of each node by executing a mutual enhancement algorithm, optimizing overall optical signal quality. The optical signal processing process in this system is as follows: After being modulated by the electro-optical modulator, the optical signal enters the circulator, which splits the signal into two paths: the primary signal is transmitted downstream, while a portion is analyzed by a monitoring unit, forming a closed-loop control system.

[0051] Specifically, see Figure 3In one embodiment of the present invention, traditional power equalization methods are typically based on simple linear models and are difficult to adapt to changes in complex fiber environments. The present invention proposes a multi-parameter collaborative power equalization algorithm that accurately describes the power distribution relationship in the circulator link through a nonlinear model.

[0052] The output power of the i-th modulation node is defined as , the signal quality function of the end-to-end link is , then the optimal power allocation problem can be expressed as:

[0053] ;

[0054] ;

[0055] In traditional distributed fiber optical modulation systems, signal quality can be measured using metrics such as signal-to-noise ratio and bit error rate. These metrics are typically weighted summed using linear or simple logarithmic calculations for multi-node coordination, which makes it difficult to fully capture the "mutual reinforcement" effect between nodes. Therefore, this embodiment introduces a nonlinear cross-node coupling term into the signal quality function for an end-to-end link to characterize the positive (or negative) interactions brought about by multi-node joint modulation. The signal quality function is defined as:

[0056] ;

[0057] in: represents the output power of the i-th modulation node at the t-th iteration time; The signal gain weight set for the i-th node is usually determined by its position in the network or its importance (such as a critical link), and is used to measure the contribution of the node to the global signal quality; The relationship between the modulation node output power and the signal-to-noise ratio improvement is reflected in logarithmic form to avoid the problem of excessively large or small gradients caused by simple linearization. is the "mutual reinforcement" coupling coefficient between nodes i and j, which reflects the positive (or negative) influence of the preferences and optical powers of different nodes on each other. It can be set based on experience or test data. A positive value indicates that the greater the power of the two nodes, the more significant the overall gain. The coupling strength between nodes i and j is expressed in square root form, which can reflect the multiplicative interaction between the two and avoid the gradient explosion caused by pure product. is the noise sensitivity adjustment coefficient of node k, which is used to balance the trade-off between power gain and noise amplification; It represents the noise or interference level of the kth node at time t, which can be obtained by real-time monitoring unit or historical data statistics.

[0058] This embodiment uses the following formula to solve the optimal power allocation problem to update the power of each modulation node:

[0059] ;

[0060] Where, represents the power of the i-th node at the t-th iteration; is the adaptive learning rate; The objective function Q is The partial derivative at ; is the nonlinear adjustment factor; is the environmental adaptability coefficient, which is determined by temperature, vibration, humidity and atmospheric pressure:

[0061] ;

[0062] in 、 、 、 represent the temperature, vibration intensity, relative humidity and atmospheric pressure at time t respectively; 、 、 is the optimal working point of each parameter, that is, the ideal environmental conditions; 、 、 、 It is the normalized reference value used to unify the dimensions of each parameter; 、 、 、 is the sensitivity coefficient, which reflects the influence of various environmental factors on the optical transmission performance; is the time-dependent correction factor, expressed as:

[0063] ;

[0064] in, Indicates various environmental parameters, is the corresponding time-varying sensitivity coefficient, is the overall time-varying suppression coefficient, which reflects the system's sensitivity to the rate of change of environmental parameters.

[0065] is the historical contribution index of the i-th node; it is used to measure the contribution of the i-th node to the improvement of global signal quality in the past period of time, so as to amplify or reduce the power adjustment amount of the node in the next iteration. This embodiment improves on the basis of differential drive and defines this index as:

[0066] ;

[0067] Where H is the size of the historical observation window, which can be fixed or adaptively adjusted to balance real-time performance and stability. , represents the change in the overall signal quality of the system between iteration u and iteration u+1; , represents the power adjustment amount of the i-th node between iteration u and iteration u+1.

[0068] In this formula, The denominator is used to limit the power consumption when the power changes too much. The extreme amplification effect of , to avoid gradient explosion; if and If the values ​​are the same and larger, it means that increasing the power of the node can indeed improve the global signal quality, and the contribution of the node in this history will increase accordingly. is the averaging factor, so The value range of is more stable and controllable.

[0069] Specifically, the present invention uses the mutual enhancement effect between nodes as a coupling term The form of is added to Q(t), so that multiple nodes do not need to focus on their own optimal power configuration, but can form positive synergy at the overall level. It can be used as a fixed parameter or updated periodically in combination with the monitoring unit feedback, thus achieving adaptive scheduling between different links or nodes. and , while improving power gain while avoiding the risk of unlimited noise amplification. The optimal power allocation formula and algorithm provided by this invention can more accurately reflect the impact of node power regulation on global quality, thereby naturally approaching the global optimum in the power allocation process.

[0070] Specifically, see Figure 4 In one embodiment of the present invention, phase adjustment methods typically use a fixed phase table or simple feedback control, which is difficult to cope with dynamic phase changes caused by temperature, vibration, stress, and other factors in optical fiber links. The dynamic phase complementary enhancement algorithm proposed in this invention implements adaptive phase calibration based on a dynamic phase codebook. The dynamic phase codebook is dynamically updated based on the characteristics of the optical fiber link and phase features are extracted through fast Fourier transform. The dynamic phase complementary enhancement algorithm has a phase adjustment frequency of 1000 times per second and can respond to changes in link characteristics in real time to maintain an optimal phase relationship.

[0071] For the i-th modulation node in the system, its phase adjustment function is defined as:

[0072] ;

[0073] in is the reference phase, is the dynamic adjustment amount, which is determined by the dynamic phase codebook Sure:

[0074] ;

[0075] S(t) represents the system state vector, which contains the signal quality parameters of each node; L(t) represents the link delay vector, which reflects the transmission characteristics of the optical fiber link; i is the node index.

[0076] Dynamic phase codebook update equation:

[0077] ;

[0078] Where: M(t) is the forgetting matrix, which controls the retention rate of historical information, and its element values ​​are between 0 and 1; N(t) is the new information matrix, which is generated by the current observation; is the nonlinear learning coefficient, which controls the sensitivity of the system to new information; E(t) is the system error function, which reflects the quality of the current phase configuration. The generation process of the new information matrix N(t) involves the analysis of the transmission characteristics of the optical fiber link and can be expressed as:

[0079] ;

[0080] Where H(f) is the system frequency response function, is the estimated link delay function. To obtain accurate link characteristic information, the algorithm uses fast Fourier transform to extract the phase characteristics of the received signal:

[0081] ;

[0082] Where, is the signal received by the i-th node, By comparing the phase characteristics of adjacent nodes, the required phase compensation amount can be calculated.

[0083] Specifically, the specific execution process of the dynamic phase complementary enhancement algorithm is as follows: when the system starts, the dynamic phase codebook is initialized according to the optical fiber network topology and theoretical transmission model. The central control unit obtains the signal quality parameters and original waveform data of each modulation node through the monitoring unit to form the system state vector S(t); Fast Fourier transform is applied to the collected signal data to extract its frequency domain phase characteristics , analyze the phase relationship between each node; based on the phase characteristics and the known network topology, estimate the transmission delay L(t) of each segment of the optical fiber link and construct the link delay vector; calculate the new phase codebook according to the update equation The process is performed once every millisecond, achieving high-frequency phase calibration 1000 times per second; the central control unit calculates the optimal phase adjustment amount for each modulation node based on the updated phase codebook , generating a phase adjustment instruction; the phase adjustment instruction is sent to each modulation node through the communication network, and the electro-optical modulator performs precise phase modulation.

[0084] The dynamic phase complementary enhancement algorithm of the present invention can dynamically adjust the phase relationship according to the link delay change, so that the signals of adjacent circulator nodes form a constructive superposition and achieve a mutual enhancement effect.

[0085] Specifically, see Figure 5 In one embodiment of the present invention, error compensation methods are often based on fixed physical models or simple linear predictions, making it difficult to effectively capture complex error patterns in optical fiber links caused by factors such as dispersion, nonlinear effects, and random polarization changes. The statistical feature error prediction and compensation algorithm proposed in the present invention uses a fractal interpolation method to construct a nonlinear prediction function. The nonlinear prediction function uses a statistical distribution model of historical data to predict signal degradation trends and generate a compensation solution. The prediction function uses a second-order Markov process to describe the error propagation characteristics, with a prediction time window of 5 to 200 milliseconds. By combining fractal theory and Markov random processes, a prediction framework capable of capturing subtle signs of signal degradation has been established.

[0086] Define the signal error vector e(t), whose statistical characteristics can be described by the multidimensional distribution D(e(t)). The error prediction model is:

[0087] ;

[0088] Where: G is a nonlinear prediction function; is the error statistical characteristic in the past T time window; C(t) is the current system configuration parameters, including the power level and phase setting of each modulation node;

[0089] The G function is constructed using the fractal interpolation method:

[0090] ;

[0091] in: is the weight coefficient; is the basis function of the error distribution D; is the basis function about the system configuration C; is the system state variable.

[0092] The core of the fractal interpolation method lies in its basis function The selection of is generated by piecewise iterative function system (IFS):

[0093] ;

[0094] Where, is the iteration operator, is the initial basis function, This method can effectively capture the self-similar characteristics of the error distribution and is suitable for describing fractal noise patterns in optical fiber transmission.

[0095] In order to describe the propagation characteristics of the error, the algorithm introduces a second-order Markov process model:

[0096] ;

[0097] ;

[0098] in: and is the state transfer matrix; is the input matrix; is a random disturbance that obeys a specific distribution.

[0099] Based on the prediction error, the compensation signal Generate by inverse system model:

[0100] ;

[0101] Where F represents the system forward transfer function, Its inverse function is used to convert the prediction error into the required compensation signal.

[0102] Specifically, the specific execution process of the statistical feature error prediction and compensation algorithm is as follows: the central control unit collects signal quality parameters from the monitoring unit of each modulation node, constructs a historical error vector Preprocessing includes denoising, normalization and feature extraction; statistical analysis of the error vector is performed to extract its multidimensional distribution characteristics. , including statistics such as mean, variance, skewness, kurtosis, as well as higher-order moments and correlation structures; based on historical data, use maximum likelihood estimation or Bayesian methods to estimate the parameters of the second-order Markov process 、 and B; nonlinear prediction function G and Markov model constructed using fractal interpolation to predict the error trend within the next 5-200 milliseconds The prediction time window can be dynamically adjusted according to system requirements, with a short time window of 5 milliseconds for rapid response and a long time window of 200 milliseconds for trend prediction. Generating compensation signals , and converts it into specific modulation parameter adjustment instructions; the central control unit distributes the compensation instructions to each modulation node, and the electro-optical modulator performs the corresponding pre-compensation modulation; the system continuously monitors the compensation effect and updates the prediction model parameters based on the actual results to form a closed-loop optimization.

[0103] Specifically, in one embodiment of the present invention, the parameter database stores the following data types: system operating parameters, historical data, optical fiber link characteristic data, environmental factor data and fault record data; the system also includes an exception handling mechanism, which includes a simplified protection switching protocol, which uses a pre-calculated backup path table to achieve rapid isolation of fault points and service recovery. When a node failure is detected, the system completes fault isolation and service recovery within 30 milliseconds.

[0104] Specifically, the data types stored in the parameter database mainly include the following categories:

[0105] System operating parameter data records the system's operating status at different points in time, including core operating parameters such as the power level, phase setting, modulation depth, bandwidth occupancy, and signal modulation format of each modulation node. These parameters are stored with high temporal resolution and provide essential data for system operating status analysis and fault diagnosis. System operating parameters also include real-time configuration information for each algorithm, such as the weight coefficients of the power balancing algorithm, the phase codebook state of the phase complementary enhancement algorithm, and the model parameters of the error prediction algorithm.

[0106] For historical data, the system normalizes and statistically processes operating parameters at different time scales to generate historical trend data. Historical data uses a progressive downsampling storage strategy, maintaining high temporal resolution for recent data and reducing storage pressure through statistical aggregation for long-term data. Historical data also includes labeled data required for algorithm training, recording the correspondence between different system configurations and performance indicators.

[0107] Fiber link characteristic data records the physical characteristics of the fiber transmission link, including the type, length, loss coefficient, dispersion coefficient, nonlinear coefficient of each fiber segment, and the insertion loss of connectors and splices. It also includes dynamic link measurement data, such as real-time attenuation distribution, dispersion compensation effectiveness, four-wave mixing intensity distribution, and Brillouin scattering threshold. This data is acquired through a combination of active detection and passive monitoring, and is updated regularly to reflect the long-term evolution of the link status.

[0108] Environmental factor data records external environmental factors that affect system performance. This includes physical parameters such as temperature, humidity, vibration intensity, and atmospheric pressure at each node, as well as operating environment parameters such as power status, computer room environment, and external electromagnetic interference. Environmental factor data is collected through a network of sensors distributed across each node and stored in association with system performance data at the corresponding point in time.

[0109] Fault log data details various fault events that occurred during system operation, including fault time, location, type, severity, scope of impact, handling measures, and recovery time. Each fault record also stores snapshots of the system status before and after the fault, including trends in key parameters and the response process of the exception handling mechanism. Fault log data is stored in a combination of structured and unstructured formats, supporting intelligent retrieval based on keywords and semantics.

[0110] This invention employs a multi-dimensional anomaly detection strategy, integrating threshold detection, trend analysis, and pattern recognition to monitor different types of anomalies in real time. Anomalies are primarily categorized into three types: functional anomalies (e.g., node response timeouts, command execution failures), performance anomalies (e.g., signal-to-noise ratio degradation, bit error rate increase), and resource anomalies (e.g., insufficient processing power, storage space depletion). Each anomaly type is further categorized by severity: warning, minor fault, and major fault, triggering different levels of response.

[0111] The simplified protection switching protocol employed in this invention is based on a finite state machine model, simplifying the fault handling process into five key states: detection, location, isolation, recovery, and confirmation. State transitions are controlled by predefined trigger conditions, and the processing logic for each state is hardware-accelerated to minimize processing delays. This simplified implementation of the protocol omits complex negotiation mechanisms and redundant confirmation steps, replacing them with pre-calculated decision tables and a fast query mechanism, enabling the entire switching process to be completed within strict time constraints.

[0112] In the present invention, the system automatically calculates and generates a comprehensive backup path table during initialization and when network topology changes occur. This path table utilizes a multi-tiered structure, including: a primary backup path that provides a direct bypass solution for single-node failures; a secondary backup path that provides a wider range of detour solutions for multi-node or regional failures; and a degraded service path that provides a downgraded solution to ensure core business continuity when an equivalent backup path cannot be found. The path table calculation takes into account multi-dimensional constraints such as network resource utilization, transmission delay, and signal quality, and uses the K shortest path algorithm and heuristic optimization methods to generate the optimal recovery strategy for each possible failure scenario.

[0113] When the system detects a node failure, the exception handling mechanism proceeds as follows: 1. Fault confirmation: Confirms the existence of the fault through cross-verification using multiple detection methods, taking no more than 5 milliseconds. 2. Fault location: Accurately locates the faulty node or link based on topology analysis and signal tracing, taking no more than 8 milliseconds. 3. Path query: Rapidly retrieves applicable recovery solutions from a pre-calculated backup path table, taking no more than 3 milliseconds. 4. Service switching: Issues switching instructions to relevant nodes and reconfigures signal routing, taking no more than 10 milliseconds. 5. Switchover confirmation: Verifies the connectivity and signal quality of the new path, taking no more than 4 milliseconds. The entire process, from fault detection to service restoration, takes no more than 30 milliseconds, meeting the requirements of high-availability, carrier-grade services. Furthermore, after service restoration, the system simultaneously initiates isolation and diagnostic procedures for the faulty node, attempting to restore the node through remote rebooting, parameter resetting, and other methods. Alternatively, it generates a detailed repair work order to guide on-site technicians in hardware maintenance.

[0114] In this embodiment, the data generated during the exception handling process is stored in a parameter database in real time. At the same time, the exception handling decision also relies on the historical fault mode analysis in the database. The exception handling priority is higher than the conventional control logic, and the system control can be temporarily taken over to ensure the timeliness of the emergency response. After the fault is recovered, the multi-parameter collaborative power balancing algorithm and the dynamic phase complementary enhancement algorithm will receive special initialization parameters to accelerate the optimization process of the new path.

[0115] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A distributed fiber optical modulation system based on a mutual enhancement circulator, characterized in that: include: Multiple modulation nodes distributed along an optical fiber link, each modulation node comprising a circulator, an electro-optical modulator, a monitoring unit, and an optical fiber array, wherein the circulator and the electro-optical modulator are connected in series to form a modulation link, and the monitoring unit performs multi-dimensional characteristic analysis of the signal through the connected optical fiber array, and is used to collect signal quality parameters of each modulation node and feed the signal quality parameters back to a central control unit. Multiple modulation nodes are sequentially connected through the optical fiber link to form a topological structure; each circulator has three ports, wherein the first port of the circulator is connected to the electro-optical modulator, the second port is connected to the downstream optical fiber, and the third port is connected to the monitoring unit; the optical signal enters the circulator after being modulated by the electro-optical modulator, and the circulator outputs the main signal from the second port to the downstream optical fiber, and simultaneously outputs part of the signal from the third port to the monitoring unit; a laser, connected to an electro-optical modulator of the first modulation node of the system, for providing an optical signal; a central control unit configured to execute a modulation parameter control algorithm, generate and send a power control signal and a phase control signal for each modulation node; Parameter database, used to store system operating parameters and historical data.

2. The distributed fiber optical modulation system based on mutual enhancement circulator according to claim 1, characterized in that: The topology is a chain structure or a ring structure; when it is a chain structure, the system has a starting end and an ending end; when it is a ring structure, the last modulation node of the system is connected back to the first modulation node to form a closed loop.

3. The distributed fiber optical modulation system based on mutual enhancement circulator according to claim 1, characterized in that: The modulation parameter control algorithm includes: A multi-parameter collaborative power balancing algorithm uses signal quality parameters collected by the monitoring unit and historical data in the parameter database to construct a signal quality function. It then calculates and optimizes the output power of each modulation node through a nonlinear model. A dynamic phase complementary enhancement algorithm calculates a phase adjustment value based on the signal quality parameters collected by the monitoring unit, and adjusts the phase relationship of each modulation node in real time so that the signals of adjacent nodes form a constructive superposition; The statistical feature error prediction and compensation algorithm uses the signal quality parameters collected by the monitoring unit in combination with the system operating parameters stored in the parameter database to analyze the error distribution characteristics, predict the signal degradation trend and generate compensation instructions.

4. The distributed fiber optical modulation system based on mutual enhancement circulator according to claim 3, characterized in that: The multi-parameter coordinated power balancing algorithm updates the power of each modulation node through the following iterative formula: ； Where, represents the power of the i-th node at the t-th iteration; is the adaptive learning rate; The signal quality function Q is The partial derivative at ; is the historical contribution index of the i-th node; is the nonlinear adjustment factor; It is the environmental adaptability coefficient, which is determined by temperature, vibration, humidity and atmospheric pressure.

5. The distributed fiber optical modulation system based on mutual enhancement circulator according to claim 4, characterized in that: The historical contribution index The calculation uses the following formula: ； Where H is the size of the historical observation window; , represents the change in the overall signal quality of the system between iteration u and iteration u+1; , represents the power adjustment amount of the i-th node between iteration u and iteration u+1.

6. The distributed fiber optical modulation system based on mutual enhancement circulator according to claim 3, characterized in that: The dynamic phase complementary enhancement algorithm implements adaptive phase calibration based on a dynamic phase codebook. The dynamic phase codebook is dynamically updated according to the characteristics of the optical fiber link and phase features are extracted through fast Fourier transform. The phase adjustment frequency of the dynamic phase complementary enhancement algorithm is 1000 times per second.

7. The distributed fiber optical modulation system based on a mutual enhancement circulator according to claim 3, characterized in that: The statistical feature error prediction and compensation algorithm adopts the fractal interpolation method to construct a nonlinear prediction function. The nonlinear prediction function uses the statistical distribution model of historical data to predict the signal degradation trend and generate a compensation scheme. The prediction function adopts a second-order Markov process to describe the error propagation characteristics, and the prediction time window is 5 milliseconds to 200 milliseconds.

8. The distributed fiber optical modulation system based on mutual enhancement circulator according to claim 1, characterized in that: The signal quality parameters include signal-to-noise ratio, bit error rate, polarization mode dispersion, chromatic dispersion, four-wave mixing effect intensity and stimulated Brillouin scattering threshold.

9. The distributed fiber optical modulation system based on mutual enhancement circulator according to claim 1, characterized in that: The parameter database stores the following data types: system operating parameters, historical data, fiber link characteristic data, environmental factor data, and fault record data; the system also includes an exception handling mechanism, which includes a simplified protection switching protocol. Through a pre-calculated backup path table, the system can quickly isolate the fault point and restore business. When a node failure is detected, the system completes fault isolation and business recovery within 30 milliseconds.

Images