A multi-channel optical amplification system and method based on 1260-1650 nm wavelength

Abstract

The application relates to the technical field of optical amplification, and specifically discloses a multi-channel optical amplification system and method based on 1260-1650 nm wavelengths, which comprises the following steps: constructing an attenuation coefficient set and a gain characteristic set by simultaneously collecting backward Rayleigh scattering and backward Raman scattering data of an optical fiber along a line; identifying a mismatch section based on the statistical distribution of the attenuation slope change rate, and optimizing the pump injection points in a non-equidistant manner; judging the deviation degree of the gain accumulation rate by using the first-order difference and the second-order difference of the gain accumulation curve, and performing local compensation; calculating the attenuation integral value from each pump injection point to the end point of the effective action interval, and uniformly setting the residual power reference in a feedforward manner; and switching the pump wavelength according to the relationship between the tangent slope of each section gain accumulation curve and the target interval, so that the joint adaptive configuration of the pump injection point position, the pump power and the pump wavelength is realized, and the gain flatness of the 1260-1650 nm wide-band multi-channel Raman amplification system is improved.

Description

Technical Field

[0001] This invention relates to the field of optical amplification technology, specifically to a multi-channel optical amplification system and method based on wavelengths of 1260~1650nm. Background Technology

[0002] In optical communication systems, the 1260–1650 nm wavelength band covers multiple communication windows including O, E, S, C, and L, and is the core operating band for wavelength division multiplexing (WDM) systems. Distributed Raman amplifiers based on stimulated Raman scattering are widely used for multi-channel optical amplification in this band due to their advantages such as low noise and wide gain spectrum.

[0003] However, existing Raman amplification techniques have the following shortcomings: the local attenuation coefficient of actual optical fibers varies non-uniformly along their length, and the equally spaced injection points cannot match the rate of attenuation slope change, leading to a mismatch between pump energy distribution and fiber loss characteristics. Current techniques cannot monitor the gain accumulation rate along the fiber in real time, obtaining the total gain value only through the power difference between the input and output ends. This makes it difficult to locate sections with insufficient or excessive gain, and also fails to identify curvature abrupt changes on the gain accumulation curve. The power settings for each pump injection point are independent, failing to consider the consistency of residual power when the pump light reaches the end of its effective range, resulting in uneven gain in the later stages. Furthermore, a single pump wavelength cannot simultaneously meet the differentiated requirements for gain efficiency and transmission distance in different segments across a wide bandwidth. Longer wavelength segments have lower attenuation but lower Raman gain coefficients, while shorter wavelength segments have higher gain coefficients but faster attenuation. A single wavelength is insufficient to balance the gain accumulation rate across all segments of the fiber. Therefore, existing methods lack the ability to adaptively match the actual attenuation and gain distribution of the fiber in the three dimensions of injection point layout, power distribution, and wavelength configuration, which limits the gain flatness performance of multi-channel Raman amplification systems.

[0004] Therefore, the present invention provides a multi-channel optical amplification system and method based on wavelengths of 1260~1650nm. Summary of the Invention

[0005] To overcome the shortcomings of existing technologies and solve at least one of the technical problems mentioned in the background, this invention provides a multi-channel optical amplification system and method based on wavelengths of 1260~1650nm. Addressing the gain unevenness problem caused by the mismatch between the pump injection point location, pump power, and pump wavelength and the actual attenuation and gain distribution of the optical fiber in a multi-channel optical amplification system in the 1260~1650nm band, this invention constructs a set of attenuation coefficients and gain accumulation characteristics. It then sequentially optimizes the pump injection point spacing (non-equidistant), performs local power compensation on the gain accumulation rate, unifies the residual power reference using a feedforward method, and switches the pump wavelength in a closed loop based on the tangent slope of the gain accumulation curve for each segment. This achieves adaptive matching between pump resources and the actual attenuation and gain distribution of the optical fiber.

[0006] The objective of this invention can be achieved through the following technical solution: a multi-channel optical amplification method based on wavelengths of 1260~1650nm, comprising the following steps:

[0007] Simultaneously, backscattering Rayleigh scattering and backscattering Raman scattering data along the optical fiber are collected, and the local attenuation coefficient and cumulative gain along the fiber are calculated at each sampling point to construct an attenuation coefficient set and a gain characteristic set.

[0008] Based on the statistical distribution of the rate of change of attenuation slope in the set of attenuation coefficients, it is determined whether there is a mismatch between the pump injection point location and the actual attenuation slope change of the fiber. If so, the mismatched section is identified and the distribution of pump injection points is optimized by non-equal spacing.

[0009] Based on the gain accumulation curve, determine whether the gain accumulation rate deviates from the target range. If so, perform local compensation by adding pump injection points and adjusting the power of the pump injection points.

[0010] Calculate the attenuation integral value of the pump light transmitted from each pump injection point to the end of the effective range, and determine whether the residual power at each point is consistent. If they are inconsistent, unify the residual power reference of each pump injection point through a feedforward method.

[0011] The optical fiber is divided into multiple segments, and the tangent slope of the gain accumulation curve of each segment is calculated to determine whether a single pump wavelength is suitable for the gain accumulation rate differences of different segments. If not, the pump wavelength is switched based on the comparison analysis between each segment and the target interval.

[0012] Furthermore, the attenuation coefficient set and gain characteristic set are constructed as follows:

[0013] Sampling positions are marked along the optical fiber at fixed sampling intervals. Probe pulses are emitted and the backscattered Rayleigh power at each sampling position is read. The local attenuation coefficient is calculated based on the backscattered Rayleigh power of the current sampling position and the previous adjacent sampling position, forming an attenuation coefficient set.

[0014] Simultaneously, the same detection pulse is emitted to separate the anti-Stokes and Stokes Raman scattered light power, and the fiber temperature at each sampling location is calculated.

[0015] The average gain coefficient is calculated based on the current anti-Stokes power and the reference power at the same temperature, and the cumulative gain value along the sampling position is obtained by accumulating it segment by segment, forming a set of gain characteristics.

[0016] Furthermore, the method for determining whether there is a mismatch between the pump injection point location and the actual attenuation slope change of the optical fiber is as follows:

[0017] The optical fiber between adjacent sampling positions is marked as a segment, and the rate of change of attenuation slope of each segment is calculated based on the local attenuation coefficient of adjacent sampling positions.

[0018] Calculate the arithmetic mean and standard deviation of the attenuation slope change rate for all segments. If the attenuation slope change rate of a segment is greater than the arithmetic mean plus three times the standard deviation, or less than the arithmetic mean minus two times the standard deviation, it is determined that there is a mismatch between the pump injection point location and the actual attenuation slope change of the fiber.

[0019] Furthermore, the process of optimizing the pump injection point distribution with non-uniform spacing is as follows:

[0020] The section where the rate of change of the attenuation slope is greater than the arithmetic mean plus 3 times the standard deviation is marked as a steep change section. A pump injection point is added at the center of the steep change section. If the length of the steep change section is greater than 100 meters, multiple injection points are added at equal intervals within the section, with the interval being one-quarter of the section length.

[0021] The segment with a rate of change of decay slope less than the arithmetic mean minus 2 standard deviations is marked as a slowly changing segment. If there are more than two pump injection points in a slowly changing segment, only the injection point located at the midpoint of the segment is retained, and the rest are removed.

[0022] Furthermore, the method for determining whether the gain accumulation rate deviates from the target range is as follows:

[0023] Extract the cumulative gain values ​​of adjacent sampling positions from the gain characteristic set, and subtract them to obtain the gain value within the segment. Set the lower limit and upper limit of the target rate.

[0024] If the gain value within a segment is less than the lower limit of the target rate, it is determined to be a segment with too slow gain; if the gain value within a segment is greater than the upper limit of the target rate, it is determined to be a segment with too fast gain.

[0025] Calculate the second-order difference of the gain values ​​within adjacent segments. If the absolute value of the second-order difference is greater than twice the arithmetic mean of the absolute values ​​of all second-order differences, then the sampling point is determined to be a point of abrupt change in gain curvature.

[0026] If the gain value within a segment is less than the lower limit of the target rate, or the gain value within a segment is greater than the upper limit of the target rate, or if there is a sampling point that is a point of sudden change in gain curvature, then the cumulative gain rate is determined to deviate from the target range.

[0027] Furthermore, the process of local compensation is as follows:

[0028] For sections with excessively slow gain and no pump injection point, a new injection point is added at the center of the section. The initial power of the new injection point is the arithmetic mean of the power of the existing injection points before and after the section. If there are existing injection points in the section but their power is too low, the current power of the injection point is increased.

[0029] For sections with excessively high gain, reduce the current power of existing injection points within the section;

[0030] For the gain curvature abrupt change point, an injection point is added at half the length before and after the gain curvature abrupt change point. The power of the two new injection points is half of the power of the original injection point near the gain curvature abrupt change point.

[0031] Furthermore, the process for determining whether the residual power at each point is consistent is as follows:

[0032] Determine the effective range endpoint of each pump injection point: Starting from the injection point, calculate the attenuation factor segment by segment along the optical fiber. The current position when the cumulative attenuation factor meets the requirements is the effective range endpoint.

[0033] The sum of the products of the local attenuation coefficient and the segment length of all fiber segments from the pump injection point to the end of the effective range is calculated to obtain the attenuation integral value.

[0034] Set a target residual power benchmark value, calculate the initial power required for each injection point. If the initial power required for the pump injection point is greater than the maximum allowable power of the pump laser, or if the ratio of the maximum to the minimum initial power required for different injection points meets the requirements, then it is determined that the residual power at each point is inconsistent.

[0035] Furthermore, the process of unifying the residual power reference at each pump injection point is as follows:

[0036] Select a target power baseline value that can be achieved at all pump injection points, and calculate the required initial power for each injection point;

[0037] If the required initial power is less than or equal to the maximum allowable power of the pump laser, then set the output power of the pump injection point to the required initial power;

[0038] If the required initial power is greater than the maximum allowable power of the pump laser, the output power of the pump injection point is set to the maximum allowable power of the pump laser, and an auxiliary injection point is added one injection point spacing in front of the current pump injection point. The residual power that the auxiliary injection point needs to supplement is calculated based on the required initial power, the maximum allowable power of the pump laser, and the attenuation integral value.

[0039] Furthermore, the process of wavelength switching based on the comparison analysis between each segment and the target interval is as follows:

[0040] The optical fiber is divided into N consecutive segments, where N is set according to the number of pump injection points. The ratio of the gain difference between the sampling points before and after the midpoint of each segment to the distance difference is calculated to obtain the tangent slope. The lower limit and upper limit of the target slope are set.

[0041] If the tangent slope of any segment exceeds the target slope range, it is determined that a single pump wavelength is not suitable for the gain accumulation rate differences of different segments.

[0042] If the tangent slope of the section is lower than the lower limit of the target slope, the gain is determined to be too slow, and the pump wavelength is switched to the longer wavelength direction. If the tangent slope of the section is higher than the upper limit of the target slope, and the power can no longer be reduced, the pump wavelength is switched to the shorter wavelength direction.

[0043] Set the switching step size, and remeasure the tangent slope of the segment after each switching until it falls into the target interval. After the switching is completed, remeasure the gain accumulation curve of the entire fiber and recalculate the tangent slope of each segment. Repeat until the tangent slope of all segments falls into the target interval, and generate a segment wavelength matching mapping table.

[0044] A multi-channel optical amplification system based on wavelengths of 1260~1650nm includes the following modules:

[0045] Data acquisition module: Simultaneously acquires backscattered Rayleigh and backscattered Raman data along the optical fiber, calculates the local attenuation coefficient and cumulative gain along the path at each sampling point, and constructs an attenuation coefficient set and a gain characteristic set;

[0046] Injection point distribution optimization module: Based on the statistical distribution of the attenuation slope change rate in the attenuation coefficient set, determine whether there is a mismatch between the pump injection point location and the actual attenuation slope change of the fiber. If so, identify the mismatched section and optimize the non-equidistant spacing of the pump injection point distribution.

[0047] Local compensation module: Based on the gain accumulation curve, it determines whether the gain accumulation rate deviates from the target range. If so, it performs local compensation by adding pump injection points and adjusting the power of the pump injection points.

[0048] Power reference unification module: Calculates the attenuation integral value of pump light transmission from each pump injection point to the end of the effective range, determines whether the residual power of each point is consistent, and if not, unifies the residual power reference of each pump injection point through feedforward.

[0049] Wavelength matching module: Divide the optical fiber into multiple segments, calculate the tangent slope of the gain accumulation curve of each segment, and determine whether a single pump wavelength is suitable for the gain accumulation rate difference of different segments. If not, switch the pump wavelength based on the comparison analysis between each segment and the target interval.

[0050] The beneficial effects of this invention are as follows:

[0051] By calculating the rate of change of attenuation slope and identifying steep and gradual change sections, injection points are added in steep change sections and injection points are merged in gradual change sections. This achieves non-equidistant adaptive matching between the distribution of pump injection points and the actual attenuation slope change of the optical fiber, thus solving the problem of injection point location mismatch.

[0052] The first-order difference of the gain accumulation curve is used to determine whether the gain accumulation rate of each segment deviates from the target range. The second-order difference is used to identify curvature change points. For the excessively slow segment, excessively fast segment and curvature change points, local compensation is carried out by adding injection points, adjusting the power of injection points, or symmetrically adding double injection points, respectively, to ensure that the gain accumulates uniformly and smoothly along the fiber length direction.

[0053] Based on the attenuation integral value from each injection point to the end of the effective range, a unified target residual power benchmark is set. The initial power required for each injection point is directly calculated in a feedforward manner. When the power exceeds the limit, an auxiliary injection point is added to ensure that the residual power of all injection points at the end of their range remains consistent, thus solving the problem of inconsistent power cascading.

[0054] The optical fiber is divided into multiple segments, and the tangent slope of the gain accumulation curve of each segment is calculated. When the slope is too low, the fiber is switched to the long wavelength direction to reduce the attenuation rate. When the slope is too high and the power cannot be reduced, the fiber is switched to the short wavelength direction to improve the gain efficiency. Through iteration, the slope of each segment falls into the target range, thus realizing the adaptive matching of the pump wavelength along the longitudinal direction of the optical fiber. Attached Figure Description

[0055] The invention will now be further described with reference to the accompanying drawings.

[0056] Figure 1 This is a flowchart of a multi-channel optical amplification method based on a wavelength of 1260~1650nm as described in Embodiment 1 of the present invention;

[0057] Figure 2 This is a logic judgment diagram of a multi-channel optical amplification method based on a wavelength of 1260~1650nm in Embodiment 1 of the present invention;

[0058] Figure 3 This is a functional block diagram of a multi-channel optical amplification system based on a wavelength of 1260~1650nm in Embodiment 2 of the present invention. Detailed Implementation

[0059] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below in conjunction with specific embodiments.

[0060] Example 1: Please refer to Figures 1-2As shown in the embodiment of the present invention, a multi-channel optical amplification method based on a wavelength of 1260~1650nm specifically includes the following steps:

[0061] Step 1: Simultaneously collect backscattering Rayleigh scattering and backscattering Raman scattering data along the optical fiber, and calculate the local attenuation coefficient and cumulative gain value along the path at each sampling point to construct the attenuation coefficient set and gain characteristic set;

[0062] In step one, the process of constructing the attenuation coefficient set and the gain characteristic set includes:

[0063] The first point to clarify is the power of the light scattered towards Ruili after collection, specifically:

[0064] Along the fiber length, the distance sampling interval of the optical time domain reflectometer is used as the position division reference. The sampling interval is determined by the transmitted pulse width, which is set to ten nanoseconds, corresponding to a spatial resolution of one meter. The fiber start point is set as the zero position, and a sampling position is marked every meter from the start point until the fiber end point. All subsequent data acquisition is carried out at these sampling positions.

[0065] An optical time domain reflectometer (OTDR) is connected to the input end of an optical fiber. A probe pulse with a wavelength of 1550 nm is emitted. The probe pulse propagates along the optical fiber, generating backscattered Rayleigh light at each sampling position. The OTD receives the returned Rayleigh scattered light signal. Based on the conversion relationship between reception time and light speed, the time axis is converted into a distance axis, thereby reading the backscattered Rayleigh light power value corresponding to each sampling position and recording it as the backscattered Rayleigh light power at that position. At the same time, the power value of the probe pulse emitted from the input end is directly read from the OTD and recorded as the input probe light power.

[0066] Secondly, it should be noted that the calculation of the local attenuation coefficient is as follows:

[0067] For each sampling position, the backscattered Rayleigh power at the sampling position is proportionally calculated to the backscattered Rayleigh power at the previous adjacent sampling position to obtain the backscattered Rayleigh power ratio. The natural logarithm of the backscattered Rayleigh power ratio is then divided by twice the distance between the two sampling positions to obtain the average attenuation coefficient between the two sampling positions. The average attenuation coefficient is assigned to the next sampling position as the local attenuation coefficient for that sampling position. For the fiber starting position, the local attenuation coefficient is the result of the calculation of the first sampling interval. After traversing all sampling positions, the local attenuation coefficient corresponding to each sampling position is obtained, forming an attenuation coefficient set.

[0068] It is understandable that the physical meaning of the local attenuation coefficient is: it describes the ratio of power attenuation of an optical signal after passing through a unit length when the optical signal is transmitted in the optical fiber. The physical unit is usually neper per meter or decibel per kilometer. It reflects the inherent loss characteristics of optical fiber for optical power, including absorption loss, Rayleigh scattering loss and bending loss. The term "local" emphasizes that the coefficient is not the average attenuation value of the entire optical fiber, but the attenuation characteristics of a small section of the optical fiber near each sampling position. It is used to characterize the non-uniformity of the attenuation distribution along the length of the optical fiber.

[0069] It should be noted that the calculation of the local attenuation coefficient is based on the backscattering Rayleigh principle of the optical time domain reflectometer. When the probe pulse is transmitted in the optical fiber, backscattered Rayleigh light is generated at each point along the way. The Rayleigh scattered light returns to the input end along the original path. For two sampling points located at position one and position two at a distance from the starting point, there is a definite mathematical relationship between the power of the backscattered Rayleigh light.

[0070] For example, let the distance between position one be z1 and the distance between position two be z2, and z2>z1. The probe pulse propagates from the input end to position two and then scatters back to the input end. The total path length is 2×z2. Similarly, for position one, the total path length is 2×z1. Since the Rayleigh scattering coefficient can be regarded as a constant in the same optical fiber, the backscattered light power is mainly affected by the attenuation on the transmission path.

[0071] Specifically, the backscattered Rayleigh power is proportional to the incident power and related to the attenuation index on the propagation path. For a uniform fiber segment, the ratio of the power Pr(z2) returned at position 2 to the power Pr(z1) returned at position 1 depends only on the attenuation of the fiber segment between the two positions. This ratio satisfies the following relationship: Pr(z2) divided by Pr(z1) equals e^(-2)α multiplied by (z2 minus z1) raised to the power of α, where α is the average attenuation coefficient of the fiber segment between position 1 and position 2, and the constant 2 comes from the round-trip path.

[0072] Taking the natural logarithm of both sides of the above equation, we get: the natural logarithm (Pr(z2) divided by Pr(z1)) is equal to negative 2 times α multiplied by (z2 minus z1), thus solving for: α is equal to negative (the natural logarithm (Pr(z2) divided by Pr(z1))) divided by 2 times (z2 minus z1);

[0073] Since Pr(z2) is usually less than Pr(z1), the ratio in parentheses is less than 1, and its natural logarithm is negative. Therefore, the calculated result of α is positive. The calculated α is assigned to the next sampling position (i.e., position two), which gives the local attenuation coefficient at that position. The local attenuation coefficient represents the average attenuation characteristics of the short fiber segment from position one to position two. When the sampling interval is small enough, it can be approximated as the instantaneous local attenuation value at position two.

[0074] Thirdly, it should be noted that the power of the Raman scattered light after collection is related to the local temperature, specifically:

[0075] A Raman optical time-domain reflectometer is connected to the fiber input end, and the same pulse as that for Rayleigh scattering detection is emitted. The Raman optical time-domain reflectometer is equipped with two narrowband filters to separate the anti-Stokes Raman scattered light and the Stokes Raman scattered light, respectively. At each sampling position, the power values ​​of the anti-Stokes Raman scattered light and the Stokes Raman scattered light are read. The anti-Stokes power value is divided by the Stokes power value to obtain the Raman anti-Stokes-Stokes power ratio. The Raman anti-Stokes-Stokes power ratio is substituted into a pre-calibrated temperature demodulation function to calculate the local temperature value of the fiber at that sampling position.

[0076] The calibration process for the temperature demodulation function includes:

[0077] Place a fiber sample of the same type as the fiber to be tested in a thermostatic bath with precise temperature control. The temperature adjustment range of the thermostatic bath covers the expected operating temperature range of the system, usually from -40 degrees Celsius to +80 degrees Celsius. Select a fixed position on the fiber sample as a reference point to ensure that the temperature at that position is consistent with the set temperature of the thermostatic bath.

[0078] Connect the Raman optical time domain reflectometer to the reference fiber, emit a probe pulse, and when the thermostat is set to the first temperature value, such as zero degrees Celsius, wait for the temperature to stabilize, read the anti-Stokes Raman scattered light power and the Stokes Raman scattered light power at the reference point, calculate the ratio between the two, and record the temperature value and the corresponding power ratio. Set the thermostat to multiple different temperature values ​​in sequence, for example, set a point every ten degrees Celsius, and repeat the acquisition and calculation to obtain a set of corresponding data pairs of temperature and power ratios.

[0079] Theoretical derivation shows that the relationship between the ratio of anti-Stokes power to Stokes power and absolute temperature is as follows: the natural logarithm of this ratio is linearly related to the reciprocal of the absolute temperature. Take the natural logarithm of the collected power ratio, convert the corresponding Celsius temperature to Kelvin absolute temperature and take the reciprocal, and plot a scatter plot with the reciprocal of the absolute temperature as the x-axis and the natural logarithm of the power ratio as the y-axis. The scatter points should be approximately distributed on a straight line.

[0080] The least squares method is used to fit the scattered points to a straight line to obtain the temperature demodulation function. After the power ratio at any location is measured, the natural logarithm is taken first, and then substituted into the temperature demodulation function to solve for the reciprocal of the absolute temperature. Finally, the reciprocal is taken and converted back to Celsius to obtain the local temperature value of the optical fiber at that location.

[0081] Fourthly, it should be noted that the calculation of the cumulative gain along the path is as follows:

[0082] Beforehand, under pump-free conditions, the anti-Stokes Raman scattering power at each sampling location was measured as a reference power, and the corresponding temperature was recorded.

[0083] In actual measurement, the gain ratio is calculated by comparing the anti-Stokes Raman scattered light power at the current sampling position with the reference power at the same temperature. The natural logarithm of the gain ratio is then divided by two to obtain the average gain coefficient from the fiber start point to the current sampling position.

[0084] Subtract the cumulative gain values ​​of two adjacent sampling positions to obtain the intra-segment gain. Accumulate the intra-segment gain segment by segment starting from the starting point to obtain the cumulative gain value along the path at each sampling position. After traversing all sampling positions, obtain the cumulative gain value corresponding to each sampling position to form a gain characteristic set.

[0085] The local attenuation coefficient and the cumulative gain value at the same sampling location are paired to form a record. All records at the sampling locations are arranged in ascending order of distance to form a set of fiber longitudinal attenuation and gain characteristics. The set contains three fields: sampling location distance, local attenuation coefficient, and cumulative gain value along the path.

[0086] It should be noted that the purpose of constructing the attenuation coefficient set and the gain characteristic set is to provide a unified measured data basis for all subsequent judgments and adjustments, so that subsequent steps can be carried out based on the actual attenuation distribution and gain accumulation distribution of the optical fiber.

[0087] Step 2: Based on the statistical distribution of the attenuation slope change rate in the attenuation coefficient set, determine whether there is a mismatch between the pump injection point location and the actual attenuation slope change of the fiber. If so, identify the mismatched section and optimize the non-equidistant spacing of the pump injection point distribution.

[0088] In step two, the process of determining whether there is a mismatch between the pump injection point location and the actual attenuation slope change of the optical fiber includes:

[0089] The fiber between any two adjacent sampling positions in the attenuation coefficient set is marked as a segment. For any segment, the local attenuation coefficient of the segment is calculated, and the difference between the local attenuation coefficient of the next position and the local attenuation coefficient of the previous position in the segment is obtained. The difference is calculated proportionally to the distance between the two sampling positions to obtain the attenuation slope change rate of the segment. The attenuation slope change rate characterizes the rate of change of the attenuation coefficient along the distance, and the unit is per meter or per kilometer.

[0090] It is understandable that the physical meaning of the rate of change of attenuation slope is: the rate of change of the local attenuation coefficient of the optical fiber along the length direction, which characterizes the spatial non-uniformity of the loss characteristics of the optical fiber. The larger the absolute value, the more drastic the change of the loss characteristics of the optical fiber with distance near that location. When it is close to 0, it means that the attenuation coefficient of the optical fiber is basically constant and the loss characteristics are uniform.

[0091] Calculate the rate of change of the attenuation slope for all segments to obtain a sequence consisting of the rate of change of the attenuation slope for each segment. Calculate the arithmetic mean of the sequence and sort the values ​​of each rate of change in the sequence in descending order.

[0092] Calculate the arithmetic mean and standard deviation of the attenuation slope change rate for all segments. For any segment, if the attenuation slope change rate is greater than the arithmetic mean + 3 times the standard deviation, or less than the arithmetic mean - 2 times the standard deviation, it is determined that there is a mismatch between the pump injection point location and the actual attenuation slope change of the fiber.

[0093] In step two, the process of identifying the mismatched segment includes:

[0094] The segments with a rate of change of the decay slope greater than the arithmetic mean plus 3 standard deviations are marked as steeply changing segments, and the segments with a rate of change of the decay slope less than the arithmetic mean minus 2 standard deviations are marked as slowly changing segments.

[0095] Each segment marked as a steep change segment or a gradual change segment is marked as a mismatch segment. Each mismatch segment is defined by its start position and end position, with the start position being the previous sampling position and the end position being the next sampling position.

[0096] For mismatched sections where the rate of change of the attenuation slope is too fast, record the length and center position of the section; for mismatched sections where the rate of change of the attenuation slope is too slow, record the length of the section and the boundaries at both ends.

[0097] All identified mismatched segments are arranged in order of distance to form a list of mismatched segments. If the interval between two adjacent mismatched segments is less than a set threshold (e.g., 20 meters), they are merged into a continuous mismatched segment.

[0098] In step two, the process of optimizing the pump injection point distribution with non-uniform spacing includes:

[0099] For steeply changing sections, a pump injection point is added at the center of the steeply changing section. If the section length is greater than 100 meters, multiple pump injection points are added at equal intervals within the section, with the interval length being one-quarter of the section length. The added pump injection points and the original pump injection points together constitute a new pump injection point distribution.

[0100] It should be noted that the rate of change of attenuation slope exhibits an approximately symmetrical distribution within the steep transition section. The central position is the extreme point or the location of the average rate of change of attenuation slope. Setting the pump injection point at the center can make the attenuation characteristics experienced by the pump light when it propagates to both sides symmetrical, thereby balancing the gain compensation effect of the two segments. If it deviates from the center, the pump injection point will be closer to one side, resulting in overcompensation on the short side and insufficient compensation on the long side.

[0101] In the 1260~1650nm band, the distance at which the pump power attenuates to 1 / e (approximately 36.8%) of its initial value is typically on the order of hundreds of meters. When the length of the abrupt change section exceeds 100 meters, a single central pump injection point cannot provide sufficient pump energy for the entire section. The interval between additional pump injection points is set to one-quarter of the section length. This is based on an engineering analogy derived from the Nyquist sampling theorem. In order to fully respond to the spatial frequency of the attenuation slope change, the sampling interval should be less than half of the change period. Since the attenuation slope change is usually gradual, setting the interval to one-quarter of the section length can ensure that the pump light coverage of adjacent pump injection points overlaps, thus avoiding gain dips.

[0102] For the gradually changing section, check the existing pump injection points in the gradually changing section. If there are more than two pump injection points, only the pump injection point located at the midpoint of the gradually changing section will be retained, and the rest of the pump injection points will be removed. If there is only one pump injection point in the gradually changing section, no operation is required. If there is no pump injection point in the gradually changing section, no operation will be performed.

[0103] It should be noted that, according to the uniform transmission line theory, in an optical fiber with a constant attenuation coefficient, the pump light only needs to be injected once at the beginning of the interval. Its power attenuation with distance is an exponential function, so there is no need for multiple pump injection points. The reason for retaining the pump injection point at the midpoint is that the two ends of the gradually changing section are often connected to the steeply changing section. The midpoint position allows the pump injection point to provide auxiliary pumping for the edges of the steeply changing sections on both sides at the same time, so as to achieve a smooth transition.

[0104] It should be noted that the purpose of optimizing the distribution of pump injection points is to solve the problem of mismatch between the location of pump injection points and the actual attenuation characteristics of the optical fiber, and to make the layout of pump injection points adapt to the spatial variation of optical fiber loss.

[0105] Step 3: Determine whether the gain accumulation rate deviates from the target range based on the gain accumulation curve. If so, perform local compensation by adding pump injection points and adjusting the power of the pump injection points.

[0106] In step three, the process of determining whether the gain accumulation rate deviates from the target range includes:

[0107] The cumulative gain values ​​of two adjacent sampling positions are taken out sequentially from the gain characteristic set. The cumulative gain value of the previous position is subtracted from the cumulative gain value of the latter position to obtain the intra-segment gain value of the fiber. The intra-segment gain value divided by the distance between the two sampling positions is the average gain accumulation rate of the segment. Since the sampling interval is fixed, the intra-segment gain value is directly used as the characterization of the gain accumulation rate.

[0108] Calculate the arithmetic mean of the gain values ​​within all segments, and take 0.8 times the arithmetic mean as the lower limit of the target rate and 1.2 times the arithmetic mean as the upper limit of the target rate. The lower limit of the target rate and the upper limit of the target rate constitute the acceptable target range of the gain accumulation rate.

[0109] The intra-segment gain value of each segment is compared with the target rate lower and upper limits:

[0110] If the gain value within a segment is greater than or equal to the lower limit and less than or equal to the upper limit, the segment gain accumulation rate is determined to be normal and has not deviated from the target range.

[0111] If the gain value within a segment is less than the lower limit, it is determined that the gain accumulation rate of the segment is too slow and deviates from the lower side of the target range.

[0112] If the gain value within a segment is greater than the upper limit, it is determined that the gain accumulation rate of the segment is too fast and deviates from the high side of the target range.

[0113] To further pinpoint the precise location of the abrupt change in gain accumulation rate, the difference between the gain values ​​in two adjacent segments is calculated to obtain the second-order difference. The arithmetic mean of the absolute values ​​of all second-order differences is calculated and multiplied by two to serve as the curvature change threshold. When the absolute value of the second-order difference of a sampling point is greater than this threshold, the sampling point is determined to be a point of abrupt change in gain curvature, which is also regarded as a manifestation of the gain accumulation rate deviating from the target range and is classified as an abnormal site to be processed.

[0114] It should be noted that the curvature change threshold is taken as twice the arithmetic mean of the absolute values ​​of all second-order differences, which is an engineering simplification of the 3σ criterion in statistics: under the assumption of normal distribution, points exceeding twice the standard deviation of the mean are considered to be significantly deviated.

[0115] Understandably, the first-order differential detects the gain level of the current segment to identify segments with insufficient gain and segments with excessive gain, while the second-order differential detects the stability of the gain rate to identify curvature abrupt change points, i.e. the turning points where the gain changes from slow to fast or fast to slow. The first-order differential ensures that the absolute value of the gain in each segment meets the standard, while the second-order differential ensures that the gain changes smoothly along the fiber without abrupt changes, together constituting a complete judgment on the gain accumulation rate.

[0116] In step three, the process of local compensation by adding pump injection points or increasing the power of existing pump injection points includes:

[0117] For sections with slow gain accumulation rate, the priority is to add pump injection points. Check if there are pump injection points in the section. If not, add a new pump injection point at the center of the section. The initial power of the new pump injection point is set to the average power of the two existing pump injection points before and after the section. If there are existing pump injection points in the section but the power is too low, use the power boosting method.

[0118] For sections with excessively fast gain accumulation rate, the existing pump injection point power is reduced without adding new pump injection points. The existing pump injection points in the section are identified, and their current power is reduced by one-tenth. After the reduction, the gain accumulation rate of the section is re-evaluated. If it is still too fast, the power is reduced further until the rate falls back to the target range.

[0119] For curvature abrupt change points, a pump injection point is added at half the length before and after the curvature abrupt change point. The power of the two new pump injection points is set to half the power of the original pump injection point near the point. The inflection of the gain curve is smoothed by the symmetrical compensation of the two pump injection points.

[0120] For overlapping areas where both slow sections and curvature abrupt change points exist, priority is given to adding dual pump injection points at curvature abrupt change points. Then, it is assessed whether the slow section still exists. If it still exists, the power is further added or increased according to the treatment method for the slow section.

[0121] The power enhancement method is as follows: when it is determined that a certain segment needs to increase its power and there is already a pump injection point in the segment, the current power of the pump injection point is multiplied by 1.1 (i.e., increased by 10%). The single enhancement amount shall not exceed 5% of the maximum allowable power. After each enhancement, the gain accumulation rate of the segment is remeasured. If it still does not reach the lower limit, it is enhanced again. The maximum number of consecutive enhancements is three. If it is still insufficient after three times, an auxiliary pump injection point is added in the segment. The power of the auxiliary pump injection point is taken as 0.5 times the power of the original pump injection point in the segment.

[0122] It should be noted that the role of local compensation is as follows: based on the gain accumulation curve, the first-order difference is used to determine whether the gain accumulation rate of each segment deviates from the target range (too slow or too fast), the second-order difference is used to identify curvature change points, and for different types of anomalies, local compensation is carried out by adding pump injection points, increasing or decreasing the power of existing pump injection points, symmetrically adding dual pump injection points, etc., to ensure that the signal light gain accumulates uniformly and smoothly along the fiber length direction.

[0123] Step 4: Calculate the attenuation integral value of the pump light transmitted from each pump injection point to the end of the effective range, and determine whether the residual power at each point is consistent. If not, unify the residual power benchmark of each pump injection point through feedforward.

[0124] In step four, the calculation process for the attenuation integral of the pump light transmitted from each pump injection point to the end of the effective range includes:

[0125] The first point to clarify is that the effective range endpoint of each pump injection point is determined as follows:

[0126] The effective range endpoint is defined as: the position reached when the pump light power propagates along the optical fiber from the pump injection point and the power attenuates to one-tenth of the initial power at the pump injection point;

[0127] It should be noted that the definition of the effective range endpoint comes from the engineering definition of the effective range of Raman amplification: when the pump power decays to less than 10% of the initial value, its contribution to the gain of the signal light is lower than the system noise level and can be ignored. It can be set by the user in the range of 5% to 20% according to the system gain requirements and noise index. The smaller the threshold, the longer the effective range and the greater the required pump power.

[0128] Within the set of local attenuation coefficients, starting from the pump injection point, the attenuation is accumulated segment by segment along the fiber. For each fiber segment, the power attenuation factor is e, which is a negative local attenuation coefficient raised to the power of the segment length. The attenuation factors of all segments are multiplied together to obtain the cumulative attenuation factor. When the cumulative attenuation factor is equal to one-tenth, the current position is the end point of the effective range.

[0129] Secondly, it should be noted that the calculation of the attenuation integral value is as follows:

[0130] The attenuation integral value is defined as the sum of the products of the local attenuation coefficient and the segment length of all fiber segments from the pump injection point to the end of the effective range. Its physical meaning is the total attenuation index of the pump light on the transmission path.

[0131] In step four, the process of determining whether the residual power at each point is consistent includes:

[0132] Residual power refers to the power remaining when the pump light reaches the end of the effective range. The relationship between residual power and the initial power at the pump injection point is: Residual power = Initial power at the pump injection point × e -衰减积分值 To determine whether the residual power of each pump injection point is consistent, the comparison is whether the residual power values ​​of different pump injection points at the end of their respective effective ranges are equal. Since the effective range lengths of each pump injection point are different, direct comparison of residual power requires normalization of the initial power of the pump injection points.

[0133] Specifically, a target residual power baseline value is set. For each pump injection point, the required initial power for the pump injection point is calculated based on the relationship between the residual power and the initial power of the pump injection point. That is, the required initial power for the pump injection point = target residual power baseline value × e-衰减积分值 If the maximum power that the pump injection point can actually provide is less than the initial power required by the pump injection point, or if the difference in the initial power required by different pump injection points exceeds a set threshold (e.g., the ratio of the maximum value to the minimum value is greater than 2), then it is determined that the residual power of each point is inconsistent.

[0134] In step four, the process of unifying the residual power reference at each pump injection point via a feedforward method includes:

[0135] The core idea of ​​the feedforward method is to directly calculate the initial power required to be output at each pump injection point based on the known attenuation integral value and the target residual power reference, without relying on the feedback signal at the output end.

[0136] Specifically, a target residual power value achievable at all pump injection points is selected. This is typically 80% of the minimum residual power calculated from the maximum possible power at all pump injection points, or a fixed value, such as 0.5 milliwatts, can be chosen. This benchmark represents the pump power level the designer wishes to maintain at the end of each effective operating range.

[0137] For any pump injection point, the initial power required at the pump injection point = target residual power baseline value × e -衰减积分值 The calculated initial power required for the pump injection point is compared with the maximum allowable power of the pump laser:

[0138] If the initial power required at the pump injection point is less than or equal to the maximum allowable power of the pump laser, then the initial power required at the pump injection point shall be used as the output power setting value of the pump injection point.

[0139] If the initial power required by the pump injection point exceeds the maximum allowable power of the pump laser, the target residual power cannot be achieved through a single pump injection point. Therefore, the power of the pump injection point is set to the maximum allowable power of the pump laser, and an auxiliary pump injection point is added in front of it at a distance equal to the distance between the two pump injection points. The target residual power of the auxiliary pump injection point remains the same as the original target residual power value, but the portion already contributed by the original pump injection point needs to be deducted. Specifically, the actual residual power generated by the original pump injection point at the end of its effective range is the maximum allowable power of the pump laser × e. -衰减积分值 The residual power that needs to be replenished at the auxiliary pump injection point is the target residual power value minus the maximum allowable power of the pump laser × e. -衰减积分值 The initial power of the auxiliary pump injection point is repeatedly calculated, and multiple auxiliary pump injection points can be cascaded if necessary.

[0140] It should be noted that the purpose of calculating the initial power to be output at each pump injection point by means of feedforward when the residual power is inconsistent is to achieve global feedforward compensation of pump power and ensure that the residual power at the end of the action range of each pump injection point reaches the target reference value in a uniform manner.

[0141] Step 5: Divide the optical fiber into multiple segments, calculate the tangent slope of the gain accumulation curve of each segment, and determine whether a single pump wavelength is suitable for the gain accumulation rate difference of different segments. If not, switch the pump wavelength according to the comparison analysis between each segment and the target interval.

[0142] In step five, the calculation process for the tangent slope of the cumulative gain curve for each segment includes:

[0143] The entire optical fiber is divided into N consecutive segments from the start point to the end point. The length of each segment is the ratio of the total length of the optical fiber to the number of segments. The value of N is taken with reference to the number of pump injection points, and is usually set to 2 to 3 times the number of pump injection points to ensure that each segment contains enough sampling points for slope calculation.

[0144] For any given segment, calculate the location of the segment's center point, which is the midpoint between the segment's start and end points;

[0145] Based on the gain characteristic set, read the cumulative gain value of the sampling position closest to the midpoint of the segment, and at the same time read the distance and cumulative gain value of the sampling position before the sampling position closest to the midpoint of the segment, as well as the distance and cumulative gain value of the sampling position after the sampling position.

[0146] The tangent slope is the instantaneous rate of change of the cumulative gain curve at that point, which is equal to the first derivative of the cumulative gain value with respect to distance. Specifically, take two sampling points near the center position, subtract the cumulative gain value of the previous point from the cumulative gain value of the later point to obtain the gain difference, subtract the distance of the previous point from the distance of the later point to obtain the distance difference, and divide the gain difference by the distance difference to obtain the average rate of change at the midpoint position. Since the sampling interval is small enough, this average rate of change can be approximated as the tangent slope at the midpoint position.

[0147] In step five, the process of determining whether a single pump wavelength is suitable for the gain accumulation rate differences in different segments includes:

[0148] Calculate the arithmetic mean of the tangent slopes of all segments, multiply the arithmetic mean by 0.8 to obtain the lower limit of the target slope, and multiply the arithmetic mean by 1.2 to obtain the upper limit of the target slope. The upper limit of the target slope and the lower limit of the target slope together form the target slope interval. The target interval represents the acceptable range of gain accumulation rate fluctuation under a single pump wavelength.

[0149] It should be noted that the calculation principle of the upper and lower limits of the target slope is derived from the engineering experience standard of optical amplifier gain flatness. In multi-channel Raman amplification systems, it is usually required that the gain fluctuation between channels be controlled within ±1 dB. Converting 1 dB to a linear ratio is about 1.26 times, corresponding to a deviation of ±26%. Considering that there are also margins such as measurement errors and polarization jitter in actual systems, the target is often tightened to ±20% in engineering, that is, 0.8 times and 1.2 times the linear ratio. This range can effectively control gain unevenness without triggering frequent and meaningless adjustments due to excessive stringency.

[0150] For each segment, the tangent slope of the segment is compared with the upper and lower limits of the target slope:

[0151] If the lower limit of the target slope ≤ the tangent slope of the section ≤ the upper limit of the target slope, it is determined that the gain accumulation rate of the section is within an acceptable range, and a single pump wavelength is applicable in this section.

[0152] If the tangent slope of a segment is less than the lower limit of the target slope, it is determined that the gain accumulation in that segment is too slow, the single pump wavelength is not suitable for that segment, the provided gain efficiency is too low, or the pump light decays too quickly.

[0153] If the tangent slope of a segment is greater than the upper limit of the target slope, it is determined that the gain accumulation in that segment is too fast, the single pump wavelength is not suitable for that segment, the provided gain efficiency is too high, or the pump light attenuates too slowly.

[0154] Therefore, if the tangent slope of all segments falls within the target slope range, it is determined that the single pump wavelength is suitable for the gain accumulation rate difference of the entire fiber length, and no wavelength switching is required. If the tangent slope of any segment exceeds the target slope range, it is determined that the single pump wavelength is not suitable, and a wavelength switching operation is required.

[0155] In step five, the process of wavelength switching based on the comparison analysis between each segment and the target interval includes:

[0156] It should be noted that the core principle of wavelength switching is as follows: In the 1260~1650nm band, the shorter the pump wavelength, the greater the attenuation coefficient of the optical fiber (faster attenuation), but the higher the Raman gain coefficient. The longer the pump wavelength, the smaller the attenuation coefficient (longer transmission distance), but the lower the Raman gain coefficient. Therefore, switching to the longer wavelength direction can reduce the attenuation rate and extend the pumping distance, which is suitable for sections where the gain accumulation is too slow. Switching to the shorter wavelength direction can improve the gain efficiency, which is suitable for sections where the gain accumulation is too fast.

[0157] The first point to clarify is that the required switching direction for each segment is determined as follows:

[0158] For segments where the tangent slope is less than the target slope lower limit, i.e., the gain is too slow, it is necessary to switch to the longer wavelength direction to reduce the attenuation rate of the pump light, so that the pump energy can propagate further and thus improve the gain accumulation rate of the segment.

[0159] For segments where the tangent slope is greater than the upper limit of the target slope, i.e., the gain is too fast, it is necessary to switch to a shorter wavelength to improve the Raman gain coefficient, but at the same time, it will accelerate the decay. In actual operation, if the gain is too fast, it usually means that the pump power is excessive. It is preferable to reduce the power rather than switch wavelengths. However, when the power has been reduced to the safe lower limit and is still too fast, switching to a shorter wavelength can further improve the gain efficiency, but it is generally not the first choice. Therefore, for segments where the tangent slope is greater than the upper limit of the target slope, try to reduce the pump power first. If it still cannot solve the problem, then switch to a shorter wavelength.

[0160] Secondly, it should be noted that the wavelength switching step size is determined as follows: a basic step size is set, which is one-fifth of the full width at half maximum (FWHM) of the Raman gain spectrum. In the 1260~1650nm band, the typical full WHM of the Raman gain spectrum is about 100nm, so the basic step size can be 20nm. The step size can be dynamically adjusted according to the response during the switching process.

[0161] Thirdly, it should be noted that the wavelength switching is performed as follows:

[0162] For each segment that needs to be switched, read the current pump wavelength. If it is necessary to switch to a longer wavelength, increase the current pump wavelength by a basic step size to obtain a new wavelength. If it is necessary to switch to a shorter wavelength, subtract a basic step size from the current pump wavelength to obtain a new wavelength.

[0163] Set the pump laser wavelength to the new wavelength and remeasure the tangent slope of the segment. The new tangent slope is then compared with the upper and lower limits of the target slope. If the new tangent slope falls within the target slope range, the switching stops. If it is still outside the target slope range, the process continues to increase or decrease by one basic step in the same direction and repeats.

[0164] The wavelength can be switched up to 5 times consecutively. If it still cannot fall into the target range, the wavelength of the last switch is maintained and the segment is marked as a special segment. Subsequent adjustments can be made to the pump power.

[0165] The switching will affect the gain accumulation rate of multiple segments before and after the switching point. Because the attenuation characteristics and gain efficiency of the pump light will change along the entire fiber after the pump light wavelength changes, after the switching is completed, the gain accumulation curve measurement needs to be performed again and the tangent slope of each segment needs to be recalculated. Check whether all segments have fallen into the target range. If there are still segments that exceed the target range, the wavelength needs to be fine-tuned again for the segments that exceed the target range. Usually, convergence can be achieved after 2 to 3 rounds of global iteration.

[0166] Record the pump wavelength that is finally stabilized in each segment and generate a segment wavelength matching mapping table. The segment wavelength matching mapping table will serve as the direct basis for subsequent pump laser wavelength configuration.

[0167] It should be noted that the function of wavelength switching is to solve the problem of mismatch between pump wavelength and gain requirements of different fiber segments, and to achieve adaptive closed-loop matching of wavelength along the longitudinal direction of the fiber.

[0168] The technical solution and advantages of this application embodiment are as follows: Simultaneously collect backscattering Rayleigh and backscattering data along the optical fiber, and calculate the local attenuation coefficient and cumulative gain value along the path at each sampling point to construct an attenuation coefficient set and a gain characteristic set; based on the statistical distribution of the attenuation slope change rate in the attenuation coefficient set, determine whether there is a mismatch between the pump injection point location and the actual attenuation slope change of the optical fiber. If so, identify the mismatched section and optimize the non-equidistant spacing of the pump injection point distribution; based on the gain accumulation curve, determine whether the gain accumulation rate deviates from the target range. If so, perform local compensation by adding pump injection points and adjusting the power of the pump injection points; calculate the attenuation integral value of the pump light transmitted from each pump injection point to the end of the effective action interval, and determine whether the residual power at each point is consistent. If inconsistent, unify the residual power reference of each pump injection point through a feedforward method; divide the optical fiber into multiple segments, calculate the tangent slope of the gain accumulation curve of each segment, and determine whether a single pump wavelength is suitable for the gain accumulation rate differences of different segments. If not, switch the pump wavelength based on the comparison analysis between each segment and the target interval. This invention constructs attenuation coefficient sets and gain characteristic sets by simultaneously acquiring backscattering Rayleigh scattering and backscattering data along the optical fiber; identifies mismatched sections based on the statistical distribution of the attenuation slope change rate and optimizes the non-equidistant spacing of the pump injection points; uses the first and second differences of the gain accumulation curve to determine the deviation of the gain accumulation rate and performs local compensation; calculates the attenuation integral value from each pump injection point to the end of the effective range and unifies the residual power reference in a feedforward manner; and switches the pump wavelength according to the relationship between the tangent slope of the gain accumulation curve of each section and the target range, realizing the joint adaptive configuration of pump injection point position, pump power, and pump wavelength, thereby improving the gain flatness of the 1260~1650nm wideband multi-channel Raman amplification system.

[0169] Example 2: Please refer to Figure 3 As shown in the embodiment of the present invention, a multi-channel optical amplification system based on a wavelength of 1260~1650nm includes the following modules:

[0170] Data acquisition module: Simultaneously acquires backscattered Rayleigh and backscattered Raman data along the optical fiber, calculates the local attenuation coefficient and cumulative gain along the path at each sampling point, and constructs an attenuation coefficient set and a gain characteristic set;

[0171] Injection point distribution optimization module: Based on the statistical distribution of the attenuation slope change rate in the attenuation coefficient set, determine whether there is a mismatch between the pump injection point location and the actual attenuation slope change of the fiber. If so, identify the mismatched section and optimize the non-equidistant spacing of the pump injection point distribution.

[0172] Local compensation module: Based on the gain accumulation curve, it determines whether the gain accumulation rate deviates from the target range. If so, it performs local compensation by adding pump injection points and adjusting the power of the pump injection points.

[0173] Power reference unification module: Calculates the attenuation integral value of pump light transmission from each pump injection point to the end of the effective range, determines whether the residual power of each point is consistent, and if not, unifies the residual power reference of each pump injection point through feedforward.

[0174] Wavelength matching module: Divide the optical fiber into multiple segments, calculate the tangent slope of the gain accumulation curve of each segment, and determine whether a single pump wavelength is suitable for the gain accumulation rate difference of different segments. If not, switch the pump wavelength based on the comparison analysis between each segment and the target interval.

[0175] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the present invention should still fall within the scope of the present invention.

Claims

1. A multi-channel optical amplification method based on wavelengths of 1260~1650nm, characterized in that: Includes the following steps: Simultaneously, backscattering Rayleigh scattering and backscattering Raman scattering data along the optical fiber are collected, and the local attenuation coefficient and cumulative gain along the fiber are calculated at each sampling point to construct an attenuation coefficient set and a gain characteristic set. Based on the statistical distribution of the rate of change of attenuation slope in the set of attenuation coefficients, it is determined whether there is a mismatch between the pump injection point location and the actual attenuation slope change of the fiber. If so, the mismatched section is identified and the distribution of pump injection points is optimized by non-equal spacing. Based on the gain accumulation curve, determine whether the gain accumulation rate deviates from the target range. If so, perform local compensation by adding pump injection points and adjusting the power of the pump injection points. Calculate the attenuation integral value of the pump light transmitted from each pump injection point to the end of the effective range, and determine whether the residual power at each point is consistent. If they are inconsistent, unify the residual power reference of each pump injection point through a feedforward method. The optical fiber is divided into multiple segments, and the tangent slope of the gain accumulation curve of each segment is calculated to determine whether a single pump wavelength is suitable for the gain accumulation rate differences of different segments. If not, the pump wavelength is switched based on the comparison analysis between each segment and the target interval.

2. The multi-channel optical amplification method based on wavelengths of 1260~1650nm according to claim 1, characterized in that: The attenuation coefficient set and gain characteristic set are constructed as follows: Sampling positions are marked along the optical fiber at fixed sampling intervals. Probe pulses are emitted and the backscattered Rayleigh power at each sampling position is read. The local attenuation coefficient is calculated based on the backscattered Rayleigh power of the current sampling position and the previous adjacent sampling position, forming an attenuation coefficient set. Simultaneously, the same detection pulse is emitted to separate the anti-Stokes and Stokes Raman scattered light power, and the fiber temperature at each sampling location is calculated. The average gain coefficient is calculated based on the current anti-Stokes power and the reference power at the same temperature, and the cumulative gain value along the sampling position is obtained by accumulating it segment by segment, forming a set of gain characteristics.

3. The multi-channel optical amplification method based on wavelengths of 1260~1650nm according to claim 1, characterized in that: The method for determining whether there is a mismatch between the pump injection point location and the actual attenuation slope change of the optical fiber is as follows: The optical fiber between adjacent sampling positions is marked as a segment, and the rate of change of attenuation slope of each segment is calculated based on the local attenuation coefficient of adjacent sampling positions. Calculate the arithmetic mean and standard deviation of the attenuation slope change rate for all segments. If the attenuation slope change rate of a segment is greater than the arithmetic mean plus three times the standard deviation, or less than the arithmetic mean minus two times the standard deviation, it is determined that there is a mismatch between the pump injection point location and the actual attenuation slope change of the fiber.

4. The multi-channel optical amplification method based on wavelengths of 1260~1650nm according to claim 3, characterized in that: The process of optimizing the non-uniform spacing of the pump injection point distribution is as follows: The section where the rate of change of the attenuation slope is greater than the arithmetic mean plus 3 times the standard deviation is marked as a steep change section. A pump injection point is added at the center of the steep change section. If the length of the steep change section is greater than 100 meters, multiple injection points are added at equal intervals within the section, with the interval being one-quarter of the section length. The segment with a rate of change of decay slope less than the arithmetic mean minus 2 standard deviations is marked as a slowly changing segment. If there are more than two pump injection points in a slowly changing segment, only the injection point located at the midpoint of the segment is retained, and the rest are removed.

5. The multi-channel optical amplification method based on wavelengths of 1260~1650nm according to claim 1, characterized in that: The method for determining whether the gain accumulation rate deviates from the target range is as follows: Extract the cumulative gain values ​​of adjacent sampling positions from the gain characteristic set, and subtract them to obtain the gain value within the segment. Set the lower limit and upper limit of the target rate. If the gain value within a segment is less than the lower limit of the target rate, it is determined to be a segment with too slow gain; if the gain value within a segment is greater than the upper limit of the target rate, it is determined to be a segment with too fast gain. Calculate the second-order difference of the gain values ​​within adjacent segments. If the absolute value of the second-order difference is greater than twice the arithmetic mean of the absolute values ​​of all second-order differences, then the sampling point is determined to be a point of abrupt change in gain curvature. If the gain value within a segment is less than the lower limit of the target rate, or the gain value within a segment is greater than the upper limit of the target rate, or if there is a sampling point that is a point of sudden change in gain curvature, then the cumulative gain rate is determined to deviate from the target range.

6. The multi-channel optical amplification method based on wavelengths of 1260~1650nm according to claim 5, characterized in that: The process of local compensation is as follows: For sections with excessively slow gain and no pump injection point, a new injection point is added at the center of the section. The initial power of the new injection point is the arithmetic mean of the power of the existing injection points before and after the section. If there are existing injection points in the section but their power is too low, the current power of the injection point is increased. For sections with excessively high gain, reduce the current power of existing injection points within the section; For the gain curvature abrupt change point, an injection point is added at half the length before and after the gain curvature abrupt change point. The power of the two new injection points is half of the power of the original injection point near the gain curvature abrupt change point.

7. The multi-channel optical amplification method based on wavelengths of 1260~1650nm according to claim 1, characterized in that: The process for determining whether the residual power at each point is consistent is as follows: Determine the effective range endpoint of each pump injection point: Starting from the injection point, calculate the attenuation factor segment by segment along the optical fiber. The current position when the cumulative attenuation factor meets the requirements is the effective range endpoint. The sum of the products of the local attenuation coefficient and the segment length of all fiber segments from the pump injection point to the end of the effective range is calculated to obtain the attenuation integral value. Set a target residual power benchmark value, calculate the initial power required for each injection point. If the initial power required for the pump injection point is greater than the maximum allowable power of the pump laser, or if the ratio of the maximum to the minimum initial power required for different injection points meets the requirements, then it is determined that the residual power at each point is inconsistent.

8. The multi-channel optical amplification method based on wavelengths of 1260~1650nm according to claim 1, characterized in that: The process of unifying the residual power reference at each pump injection point is as follows: Select a target power baseline value that can be achieved at all pump injection points, and calculate the required initial power for each injection point; If the required initial power is less than or equal to the maximum allowable power of the pump laser, then set the output power of the pump injection point to the required initial power; If the required initial power is greater than the maximum allowable power of the pump laser, the output power of the pump injection point is set to the maximum allowable power of the pump laser, and an auxiliary injection point is added one injection point spacing in front of the current pump injection point. The residual power that the auxiliary injection point needs to supplement is calculated based on the required initial power, the maximum allowable power of the pump laser, and the attenuation integral value.

9. The multi-channel optical amplification method based on wavelengths of 1260~1650nm according to claim 1, characterized in that: The process of wavelength switching based on the comparison analysis between each segment and the target segment is as follows: The optical fiber is divided into N consecutive segments, where N is set according to the number of pump injection points. The ratio of the gain difference between the sampling points before and after the midpoint of each segment to the distance difference is calculated to obtain the tangent slope. The lower limit and upper limit of the target slope are set. If the tangent slope of any segment exceeds the target slope range, it is determined that a single pump wavelength is not suitable for the gain accumulation rate differences of different segments. If the tangent slope of the section is lower than the lower limit of the target slope, the gain is determined to be too slow, and the pump wavelength is switched to the longer wavelength direction. If the tangent slope of the section is higher than the upper limit of the target slope, and the power can no longer be reduced, the pump wavelength is switched to the shorter wavelength direction. Set the switching step size, and remeasure the tangent slope of the segment after each switching until it falls into the target interval. After the switching is completed, remeasure the gain accumulation curve of the entire fiber and recalculate the tangent slope of each segment. Repeat until the tangent slope of all segments falls into the target interval, and generate a segment wavelength matching mapping table.

10. A multi-channel optical amplification system based on wavelengths of 1260~1650nm, characterized in that: Includes the following modules: Data acquisition module: Simultaneously acquires backscattered Rayleigh and backscattered Raman data along the optical fiber, calculates the local attenuation coefficient and cumulative gain along the path at each sampling point, and constructs an attenuation coefficient set and a gain characteristic set; Injection point distribution optimization module: Based on the statistical distribution of the attenuation slope change rate in the attenuation coefficient set, determine whether there is a mismatch between the pump injection point location and the actual attenuation slope change of the fiber. If so, identify the mismatched section and optimize the non-equidistant spacing of the pump injection point distribution. Local compensation module: Based on the gain accumulation curve, it determines whether the gain accumulation rate deviates from the target range. If so, it performs local compensation by adding pump injection points and adjusting the power of the pump injection points. Power reference unification module: Calculates the attenuation integral value of pump light transmission from each pump injection point to the end of the effective range, determines whether the residual power of each point is consistent, and if not, unifies the residual power reference of each pump injection point through feedforward. Wavelength matching module: Divide the optical fiber into multiple segments, calculate the tangent slope of the gain accumulation curve of each segment, and determine whether a single pump wavelength is suitable for the gain accumulation rate difference of different segments. If not, switch the pump wavelength according to the comparison analysis between each segment and the target interval.

Images