An infrared pair tube signal strength enhancement method and system based on data processing

Abstract

This invention relates to the field of data processing technology and discloses a method and system for enhancing the signal strength of an infrared phototransistor receiver based on data processing. The method includes: acquiring the original signal data of the target object receiver; analyzing the original signal data to obtain a distortion characteristic index representing the degree of refraction interference from the light-transmitting container; obtaining the distortion characteristic index through analysis of the original signal data, which accurately reflects the degree of refraction interference from the light-transmitting container; adaptively dividing the dynamic signal segments according to the distortion characteristic index to ensure that the segment boundaries highly match the signal distortion region; and then calculating the gain determination factor to evaluate the signal strength enhancement effect. This solution can accurately suppress the refraction interference caused by light-transmitting containers such as ampoules and improve the reception strength of the effective signal.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and specifically to a method and system for enhancing the signal strength of infrared photodiode receivers based on data processing. Background Technology

[0002] Currently, the core method for improving the signal strength of infrared photodiodes is to combine optical path optimization and signal modulation. By adjusting the installation angle of the infrared emitting and receiving tubes, the emitted and received light rays are precisely aligned. Furthermore, by modulating the infrared signal, interference from ambient natural light is suppressed, and effective signals are extracted, thereby improving the purity and intensity of the received signal.

[0003] However, the above methods for improving signal strength still have the following drawbacks: For example, when the infrared pair is installed on the ampoule conveyor line and the conveyance of the ampoules is monitored, the refraction and scattering of infrared light by the transparent ampoules will cause the signal data collected by the receiver to contain a large amount of refraction interference data. The existing technology does not perform effective data processing on the refraction interference data, and therefore cannot distinguish between the effective signal and the noise data in the data processing, resulting in the effective signal being masked and reducing the received signal strength of the infrared pair. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a method and system for enhancing the signal strength of infrared photodiode receivers based on data processing, thus solving the aforementioned problems.

[0005] The above-mentioned technical objective of the present invention is achieved through the following technical solution: A method for enhancing the received signal strength of an infrared photodiode based on data processing includes: Step S1: Obtain the original signal data from the receiver of the target object, analyze the original signal data, and obtain the distortion characteristic index representing the degree of refraction interference of the light-transmitting container. Step S2: Based on the distortion feature index, the original signal data is adaptively segmented to obtain several dynamic signal segments with independent distortion features. Step S3: Perform data decomposition on each dynamic signal segment, separate out and remove the interference signal component caused by the refraction of the light-transmitting container, and obtain the pure signal strength value representing the effective signal strength after removing the interference. Step S4: Calculate the original signal data to obtain a reference strength value representing the stable energy level of the original signal. Perform a joint analysis between the pure signal strength value and the reference strength value to obtain a gain determination factor for judging whether the received signal strength of the target object has been improved.

[0006] Furthermore, the original signal data is analyzed to obtain distortion characteristic indices representing the degree of refractive interference from the transparent container, including: Phase space reconstruction is performed on the original signal data to map the one-dimensional time-domain signal to a high-dimensional space, resulting in a high-dimensional trajectory matrix representing the dynamic evolution characteristics of the signal.

[0007] Furthermore, analysis of the original signal data yields a distortion characteristic index representing the degree of refractive interference from the transparent container, which also includes: Matrix decomposition is performed on the high-dimensional spatial trajectory matrix to separate the characteristic components of the light-transmitting container refraction interference, resulting in an interference characteristic subspace representing the spatial distribution of the interference signal; Energy distribution calculations are performed on the interference characteristic subspace to obtain the interference energy concentration coefficient, which represents the intensity of refracted interference energy.

[0008] Furthermore, analysis of the original signal data yields a distortion characteristic index representing the degree of refractive interference from the transparent container, which also includes: Data analysis of the interference energy concentration coefficient yields a distortion characteristic index that represents the degree of refraction interference from the transparent container.

[0009] Furthermore, based on the distortion characteristic index, the original signal data is adaptively segmented to obtain several dynamic signal segments with independent distortion characteristics, including: Time series analysis of the distortion feature index is performed within a sliding window to analyze the data difference between adjacent sliding windows, resulting in a dynamic distortion boundary sequence representing the boundary of the signal distortion region. The original signal data is adaptively divided based on the dynamic distortion boundary sequence to obtain several dynamic signal segments with independent distortion characteristics.

[0010] Furthermore, each dynamic signal segment is decomposed to separate and remove the interference signal component caused by refraction from the transparent container, resulting in a pure signal strength value representing the effective signal strength after interference removal, including: Each dynamic signal segment is decomposed, an adaptive trajectory matrix is ​​constructed, and singular value decomposition is performed. Based on the characteristics of singular value energy distribution, the segments are sieved to obtain an adaptive singular spectral component set representing the principal components of the signal.

[0011] Furthermore, each dynamic signal segment is decomposed to separate and remove the interference signal component caused by refraction from the light-transmitting container, resulting in a pure signal strength value representing the effective signal strength after interference removal. This also includes: Feature extraction is performed on the adaptive singular spectrum component set to identify and label the interference signal components, resulting in an interference component label vector representing the refraction interference component. The interference component marker vector is analyzed and reconstructed to obtain the pure signal strength value representing the effective signal strength after interference removal.

[0012] Furthermore, the original signal data is calculated to obtain a reference strength value representing the stable energy level of the original signal. The pure signal strength value and the reference strength value are then analyzed together to obtain a gain determination factor for judging whether the received signal strength of the target object has improved, including: Data processing is performed on the original signal data to obtain a reference intensity value representing the stable energy level of the original signal.

[0013] Furthermore, the original signal data is calculated to obtain a reference strength value representing the stable energy level of the original signal. The pure signal strength value and the reference strength value are then analyzed together to obtain a gain determination factor for judging whether the received signal strength of the target object has improved. This also includes: The pure signal strength value is compared with the reference strength value to obtain the gain determination factor for judging whether the received signal strength of the target object has been improved.

[0014] Furthermore, an infrared photodiode receiving signal strength enhancement system based on data processing, applied to the above method, includes: The data analysis unit is used to acquire the raw signal data of the target object's receiving end, analyze the raw signal data, and obtain the distortion characteristic index, which represents the degree of refraction interference of the light-transmitting container. The distortion analysis unit is used to adaptively segment the original signal data according to the distortion characteristic index to obtain several dynamic signal segments with independent distortion characteristics. The interference analysis unit is used to decompose each dynamic signal segment into data, separate out and remove the interference signal components caused by the refraction of the light-transmitting container, and obtain the pure signal strength value representing the effective signal strength after the interference is removed. The strength judgment unit is used to calculate the original signal data to obtain a reference strength value representing the stable energy level of the original signal. The pure signal strength value and the reference strength value are analyzed together to obtain a gain judgment factor to determine whether the received signal strength of the target object has been improved.

[0015] In summary, the present invention has the following main beneficial effects: By reconstructing the phase space of the original signal data, the one-dimensional time-domain signal is mapped to a high-dimensional space, resulting in a high-dimensional trajectory matrix. This high-dimensional trajectory matrix is ​​then decomposed to separate the characteristic components of refraction interference from the transparent container. An interference feature subspace is constructed, and the interference energy concentration coefficient is calculated. The resulting distortion feature index reflects the degree of refraction interference from the transparent container. Multiple dynamic signal segments are then divided, and an adaptive trajectory matrix is ​​constructed for each segment, followed by singular value decomposition. Interference signal components are then labeled, ultimately yielding a pure signal strength value. This achieves precise separation and elimination of refraction interference. Finally, the pure signal strength value is co-analyzed with a baseline strength value to obtain a gain determination factor, which can then be used to evaluate the signal strength improvement effect. This solution, through targeted data processing, can accurately suppress refraction interference caused by transparent containers such as ampoules, effectively improving the signal strength received by the infrared photodiode. Attached Figure Description

[0016] Figure 1 This is a step diagram of a method for enhancing the received signal strength of an infrared photodiode based on data processing according to the present invention; Figure 2 This is a schematic diagram of an infrared photodiode receiving signal strength enhancement system based on data processing according to the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] refer to Figure 1 and Figure 2 A method for enhancing the signal strength of an infrared photodiode receiver based on data processing, comprising: Step S1: Obtain the original signal data from the receiver of the target object, analyze the original signal data, and obtain the distortion characteristic index representing the degree of refraction interference of the light-transmitting container. The target object is an infrared pair tube, and the light-transmitting container is an ampoule. Step S2: Based on the distortion feature index, the original signal data is adaptively segmented to obtain several dynamic signal segments with independent distortion features. Step S3: Perform data decomposition on each dynamic signal segment, separate out and remove the interference signal component caused by the refraction of the light-transmitting container, and obtain the pure signal strength value representing the effective signal strength after removing the interference. Step S4: Calculate the original signal data to obtain a reference strength value representing the stable energy level of the original signal. Perform a joint analysis between the pure signal strength value and the reference strength value to obtain a gain determination factor for judging whether the received signal strength of the target object has been improved.

[0019] In one embodiment, the original signal data is analyzed to obtain a distortion characteristic index representing the degree of refractive interference from the light-transmitting container, including: Phase space reconstruction is performed on the original signal data to map the one-dimensional time-domain signal to a high-dimensional space, resulting in a high-dimensional trajectory matrix representing the dynamic evolution characteristics of the signal. Specifically, this includes: preprocessing the original signal data to obtain a numerical sequence; for the numerical sequence, calculating the absolute value of the difference between the current sampling point value and the previous point value, and dividing it by the difference between the maximum and minimum values ​​in the entire numerical sequence to obtain the instantaneous volatility of the current sampling point; and combining the instantaneous volatility of all sampling points to form an instantaneous volatility sequence with a length one less than the numerical sequence. Based on the instantaneous volatility sequence, find all volatility peaks. A volatility peak is a point whose value is greater than the values ​​of the point before and after it. Count the number of sampling intervals between each pair of adjacent volatility peaks and use the median of all such intervals as the embedding delay. Based on the obtained embedding delay, starting from dimension 2, a sequence of trajectory vectors is constructed in the current dimension based on the numerical sequence and the embedding delay. Each trajectory vector is composed of numerical values ​​extracted from the numerical sequence according to the embedding delay. Calculate the Euclidean distance between all pairs of trajectory vectors in the current dimension, and calculate the standard deviation of these distances. When the dimension increases by 1, reconstruct the trajectory vector sequence in the new dimension, use it as the new trajectory vector sequence, calculate the Euclidean distance between all pairs of trajectory vectors in the new trajectory vector sequence, calculate the standard deviation of these distances, and use this standard deviation as the new standard deviation. The dynamic threshold is obtained by calculating the ratio of the median of the differences between adjacent instantaneous volatility values ​​in the instantaneous volatility sequence to the mean of the instantaneous volatility sequence. The rate of change of the standard deviation before the dimension is added and the new standard deviation are calculated. When the rate of change is first less than the dynamic threshold, the dimension addition is stopped and the current dimension is determined as the final embedded dimension. Given a numerical sequence with L sampling points arranged in the order of acquisition, let τ denote the embedding delay and m denote the embedding dimension. Starting from the first sampling point of the numerical sequence, take that point as the first component of the first vector. Then skip τ sampling points and take the value at that position as the second component of the first vector. Skip τ sampling points again and take the value at that position as the third component of the first vector, and so on, until m components are collected to form the first high-dimensional trajectory vector. Then, move the starting point one sampling point backward. Starting from the second sampling point, extract a value every τ sampling points according to the same rule, and take m components in sequence to form the second high-dimensional trajectory vector. In this way, move the starting point backward in sequence and repeat the above extraction process until the index of the m-th component of the formed vector is ≤L. If the last vector cannot take m components, discard the vector and terminate at this time. The index is counted from 1. Arrange all the sequentially constructed high-dimensional trajectory vectors in order, with each row corresponding to one vector, to obtain a high-dimensional spatial trajectory matrix that represents the dynamic evolution characteristics of the signal. The high-dimensional spatial trajectory matrix can fully reflect the dynamic evolution characteristics of the signal under refraction interference.

[0020] In one embodiment, analyzing the original signal data to obtain a distortion characteristic index representing the degree of refractive interference from the light-transmitting container further includes: Matrix decomposition is performed on the high-dimensional spatial trajectory matrix to separate the characteristic components of the light-transmitting container refraction interference, resulting in an interference feature subspace representing the spatial distribution of the interference signal. Specifically, this involves: treating each row of the high-dimensional spatial trajectory matrix as a high-dimensional trajectory vector; for each dimension, taking the difference between the maximum and minimum values ​​of all high-dimensional trajectory vectors in that dimension as the fluctuation amplitude of that dimension; sorting the fluctuation amplitudes of each dimension from largest to smallest; and taking the fluctuation amplitudes of the corresponding proportional dimensions as the high-fluctuation dimension set, where the corresponding proportion is calculated by dividing the number of rows of the high-dimensional spatial trajectory matrix by the number of columns and then multiplying by one-third; if the calculated proportion is less than 1 divided by the number of columns, then one dimension is taken as the high-fluctuation dimension set, ensuring that the set is not empty, thereby ensuring that the size of the set is adaptively adjusted according to the scale of the high-dimensional spatial trajectory matrix. In a set of high-fluctuation dimensions, the medians of all high-dimensional trajectory vectors in each dimension are arranged in dimensional order to form a median vector with the same dimensions as the high-dimensional trajectory vectors. The Euclidean distance between each high-dimensional trajectory vector and this median vector is calculated to obtain a distance sequence. The distance sequence is sorted from smallest to largest. The quantile position is determined according to the number of rows in the high-dimensional spatial trajectory matrix. Specifically, when the number of rows is less than 100, the average of the distance values ​​at the 10% and 90% positions after sorting is taken as the threshold for extracting interference features. When the number of rows is greater than or equal to 100, the average of the distance values ​​at the 5% and 95% positions after sorting is taken as the threshold for extracting interference features. The above quantile positions are set based on the sparsity of ampoule refraction interference relative to the normal signal. Because the proportion of interference samples is relatively high when the number of rows is small, the 10% and 90% quantiles are used more leniently. When the number of rows is large, more accurate interference distribution statistics can be obtained, so the 5% and 95% quantiles are used more strictly to ensure the accuracy of interference feature extraction. All high-dimensional trajectory vectors whose distance from the median vector is greater than the interference feature extraction threshold are marked as interference candidate vectors, and the remaining vectors are marked as normal candidate vectors. The average value of the interference candidate vectors in each dimension is calculated to obtain the interference feature vector; the average value of the normal candidate vectors in each dimension is calculated to obtain the normal feature vector. Finally, the values ​​of the interference feature vector and the normal feature vector in the same dimension are calculated by difference to obtain the interference feature weight in that dimension. All dimensions with non-zero weights are combined, and based on the column vectors corresponding to these dimensions in the high-dimensional space trajectory matrix, an interference feature subspace representing the spatial distribution of the interference signal is constructed. The interference feature subspace consists of several column vectors, each column corresponding to an interference feature dimension, which fully reflects the distribution characteristics of the refraction interference of the light-transmitting container in the phase space.

[0021] The energy distribution of the interference feature subspace is calculated to obtain the interference energy concentration coefficient representing the intensity of the refracted interference energy. Specifically, each column in the interference feature subspace is regarded as an interference feature dimension. The number of values ​​in each column is equal to the number of rows in the high-dimensional space trajectory matrix. For each column, the sum of squares of all values ​​in the column is calculated to obtain the energy value of that dimension. This energy value is mainly used to reflect the total energy contribution of the interference feature in that dimension throughout the entire sampling time. The energy values ​​of all dimensions are sorted from largest to smallest to obtain the energy sequence. The sum of the energy sequences is calculated as the total energy of the interference feature subspace. Starting from the first maximum value in the energy sequence, the energy values ​​are accumulated sequentially. When the accumulated value first exceeds the total energy accumulation threshold, the number of dimensions that have been accumulated is recorded and used as the number of main energy dimensions. The accumulation threshold is dynamically adjusted by the ratio of the number of columns in the interference feature subspace to the number of rows in the high-dimensional space trajectory matrix. For example, when the ratio of the number of columns to the number of rows is less than 0.1, two-thirds of the total energy is taken as the accumulation threshold; otherwise, half of the total energy is taken as the accumulation threshold. Because the interference energy is more concentrated when the number of columns is relatively small, a higher threshold is used to ensure the accuracy of the main dimension selection. Extract the energy values ​​corresponding to the main energy dimensions and calculate the standard deviation of these energy values. At the same time, calculate the standard deviation of the energy values ​​of all dimensions. Divide the standard deviation of the main energy dimensions by the standard deviation of the energy values ​​of all dimensions to obtain the preliminary clustering coefficient. Multiplying the initial clustering coefficient by the ratio of the number of main energy dimensions to the total number of dimensions of the interference feature subspace yields the interference energy clustering coefficient, which represents the intensity of the refracted interference energy. The larger the interference energy clustering coefficient, the more concentrated the refracted interference energy is in a few dimensions, and the more significant the interference features are.

[0022] In one embodiment, analyzing the original signal data to obtain a distortion characteristic index representing the degree of refractive interference from the light-transmitting container further includes: Data analysis of the interference energy concentration coefficient yields a distortion characteristic index representing the degree of refraction interference from the transparent container. Specifically, the instantaneous volatility at the first 30% of the instantaneous volatility sequence is used as a high volatility reference value, and the instantaneous volatility at the last 30% of the instantaneous volatility sequence is used as a low volatility reference value. The 30% ratio is determined by the length of the current signal segment. When the number of sampling points is less than 50, 20% is used, and when the number of sampling points is greater than or equal to 50, 30% is used to ensure that the reference value is adaptively adjusted according to the number of sampling points. The difference between the interference energy concentration coefficient and the low fluctuation reference value is subtracted, and then divided by the difference between the high fluctuation reference value and the low fluctuation reference value. The calculation result is then normalized to the 0-1 range to obtain the distortion characteristic index, which represents the degree of refraction interference of the transparent container. The closer the distortion characteristic index is to 1, the more severe the degree of refraction interference.

[0023] By analyzing the original signal data and adaptively determining the embedding delay and final embedding dimension based on the instantaneous volatility sequence, a high-dimensional spatial trajectory matrix is ​​constructed, which fully preserves the dynamic evolution characteristics of the signal under refraction interference. Then, based on the high volatility dimension set and quantile, the interference candidate vector is accurately separated, and the interference feature vector and normal feature vector are constructed to form an interference feature subspace reflecting the spatial distribution of interference. On this basis, the energy concentration of refraction interference is analyzed, and finally the distortion feature index is obtained. By removing the refraction interference component from the data level, the identifiability of the effective signal is improved, overcoming the defect of refraction interference masking the effective signal in traditional technology, and improving the strength of the infrared photodiode received signal.

[0024] In one embodiment, the original signal data is adaptively segmented according to the distortion characteristic index to obtain several dynamic signal segments with independent distortion characteristics, including: The distortion feature index is analyzed in a time series within a sliding window. The data difference between adjacent sliding windows is analyzed to obtain a dynamic distortion boundary sequence representing the boundary of the signal distortion region. Specifically, the analysis includes: taking the numerical sequence of the original signal data as the analysis object, setting the length of the sliding window to the product of the distortion feature index and the total number of sampling points, rounding the product down, comparing the integer with one-third of the total number of sampling points, taking the smaller value, and then comparing the smaller value with 5, taking the larger value. The minimum value of 5 can initially capture the local trend of signal fluctuations, while the maximum value is one-third of the total number of signal sampling points. This is to avoid the segmentation granularity being too coarse due to an excessively large window. Ampoule refraction interference is sudden and local. If the window length exceeds one-third of the total number of sampling points, a single window may cross multiple distortion regions, mixing signals that should be segmented independently together and destroying the accuracy of subsequent segmentation. The sliding window is moved point by point across the numerical sequence with the specified length. The ratio of the range to the median of the numerical sequence within each sliding window is calculated as the local fluctuation characteristic value of the sliding window. At the same time, the absolute value of the difference between the local fluctuation characteristic values ​​of two adjacent sliding windows is calculated as the data difference between adjacent sliding windows. Sort the data differences from smallest to largest. Take the data difference at the 100% position of the distortion feature index after sorting as the high threshold, and take the data difference at the 50% position of the distortion feature index after sorting as the low threshold. The larger the distortion feature index, the later the selected threshold position is, that is, the stricter the boundary judgment of the distorted region. The intersection of adjacent sliding windows with data difference greater than a high threshold is marked as a strong boundary, the intersection of adjacent sliding windows with data difference less than a low threshold is marked as a weak boundary, and the remaining positions are marked as non-boundaries. Starting from the beginning of the data difference sequence, the positions of all strong and weak boundaries are extracted sequentially and arranged in chronological order to form a dynamic distortion boundary sequence representing the boundary of the signal distortion region. When there is no weak boundary between two adjacent strong boundaries, only the strong boundary is retained; when there is a weak boundary between two adjacent strong boundaries, the weak boundary is also retained.

[0025] The original signal data is adaptively divided according to the dynamic distortion boundary sequence to obtain several dynamic signal segments with independent distortion features. Specifically, each boundary position in the dynamic distortion boundary sequence is used as a candidate segmentation point. Starting from the starting position of the original signal data, each candidate segmentation point is traversed sequentially. For the current candidate segmentation point, its corresponding data difference degree is taken, and the data difference degree is multiplied by the distortion feature index to obtain the merging decision threshold for that point. At the same time, the difference between the local fluctuation feature values ​​of the two adjacent sliding windows before and after the candidate segmentation point is calculated. If the difference is less than the merging decision threshold, the candidate segmentation point is marked as a merging point, indicating that the difference in distortion features on both sides of the position is small and should not be used as the boundary of an independent segment. If the difference is greater than or equal to the merging decision threshold, it is marked as a retention point and used as the actual segmentation position. All candidate segmentation points marked as reserved points are output in sequence to form the actual segmentation boundary sequence. Starting from the beginning position of the original signal data, the signal data between every two adjacent actual segmentation boundaries is extracted in sequence to form a dynamic signal segment. The first segment is formed between the beginning position and the first actual segmentation boundary, and the last segment is formed between the last actual segmentation boundary and the end of the signal. Finally, check the length of each dynamic signal segment. If the segment length is less than the sum of the product of the preset minimum segment length and the median of the instantaneous volatility of all sampling points in the segment and the distortion characteristic index, where the preset minimum segment length is 5 sampling points, then the segment is merged with the adjacent segment. During merging, the data difference degree at the boundary between the current segment and the adjacent segments on the left and right is calculated respectively. The adjacent segments with smaller data difference degree are selected for merging. If the data difference degree of the adjacent segments on the left and right is the same, the segment with the adjacent segment on the left is merged first. Among them, the median of instantaneous volatility within a segment reflects the overall intensity of the signal fluctuation in that segment. The distortion characteristic index serves as a global adjustment factor, and the product of the two is used as a merging premise. This allows segments with severe fluctuations to have a smaller independent length, while segments with gentle fluctuations need to reach a longer length to be considered independent segments, ensuring that the segmentation results match the characteristics of the signal itself. By using the above division method, several dynamic signal segments with independent distortion characteristics, relatively consistent internal distortion characteristics, clear boundaries, and reasonable lengths can be obtained.

[0026] By adaptively segmenting the original signal data using the distortion feature index and conducting dynamic threshold analysis of local fluctuation feature values ​​and data differences using a sliding window, a dynamic distortion boundary sequence that accurately reflects the boundary of the signal distortion region is constructed. Ultimately, the original signal is divided into several dynamic signal segments with relatively consistent internal distortion features, clear boundaries, and reasonable lengths. This enables the accurate capture of the suddenness and locality of ampoule refraction interference, effectively avoiding the overlap between distorted and undistorted regions.

[0027] In one embodiment, each dynamic signal segment is decomposed to separate and remove interference signal components caused by refraction from the light-transmitting container, resulting in a pure signal strength value representing the effective signal strength after interference removal, including: Each dynamic signal segment is decomposed, an adaptive trajectory matrix is ​​constructed, and singular value decomposition is performed. Based on the singular value energy distribution characteristics, the segments are screened to obtain an adaptive singular spectral component set representing the principal components of the signal. Specifically, this includes: taking the value of the current dynamic signal segment as the analysis object, setting the length of the dynamic signal to M, calculating the absolute value of the difference between all adjacent sampling points within the dynamic signal segment, using the median of these differences as the reference step size, calculating the range of values ​​within the dynamic signal segment, dividing the reference step size by the range to obtain a dimensionless fluctuation scaling factor, multiplying the fluctuation scaling factor by M to obtain the window length. If the window length is less than 3, it is set to 3; if the window length is greater than M minus 1, it is set to M minus 1. The window length is determined by the fluctuation characteristics and amplitude range of the segment itself, so that the window size is adaptively adjusted with local changes in the signal. Starting from the first sampling point of the dynamic signal segment, K consecutive sampling points are taken as the first row of the trajectory matrix. Then, the starting point is moved one sampling point backward, and K consecutive sampling points are taken as the second row of the trajectory matrix. This process continues until the starting point moves to the M-K+1th sampling point. At this point, the last K consecutive sampling points are taken as the last row of the trajectory matrix. This results in an adaptive trajectory matrix with M-K+1 rows and K columns. Singular value decomposition is performed on the adaptive trajectory matrix to obtain a sequence of singular values ​​arranged in descending order of singular values, as well as the corresponding left and right singular vectors. Here, K represents the length of the window selected when constructing the adaptive trajectory matrix. The cumulative energy percentage of the singular value sequence is calculated by successively accumulating the squares of the singular values ​​and dividing by the sum of the squares of all singular values ​​to obtain the cumulative energy percentage sequence. Starting from the first singular value, it is determined whether the current cumulative energy percentage reaches the ratio of the standard deviation to the mean of the instantaneous volatility sequence within the dynamic signal segment. The ratio is normalized to the 0-1 interval and used as the dynamic energy threshold. When the dynamic energy threshold is reached or exceeded for the first time, the position of the current singular value is recorded as the dividing point. The singular values ​​before the dividing point correspond to the principal components of the signal, and the singular values ​​after the dividing point correspond to the noise and interference components. Based on the location of the boundary point, several singular values ​​before the boundary point are retained, while the rest are discarded. Each retained singular value corresponds to an independent singular spectral component. All retained components constitute the adaptive singular spectral component set representing the principal components of the signal.

[0028] In one embodiment, each dynamic signal segment is decomposed to separate and remove interference signal components caused by refraction from the light-transmitting container, resulting in a pure signal strength value representing the effective signal strength after interference removal. The method further includes: Feature extraction is performed on the adaptive singular spectrum component set to identify and label the interference signal components, resulting in an interference component label vector representing the refraction interference component. Specifically, each component in the adaptive singular spectrum component set is treated as an independent signal sequence. For each component, its correlation coefficient with the numerical sequence is calculated to obtain a correlation coefficient sequence. At the same time, the median of the absolute values ​​of the differences between all adjacent sampling points within the component is calculated, which is the fluctuation intensity. Multiply the correlation coefficient of each component by the fluctuation intensity of that component to obtain the preliminary discriminant value of that component. Sort all the preliminary discriminant values ​​of the components from largest to smallest to obtain the sorted discriminant value sequence. Multiply the distortion feature index by the total number of components in the adaptive singular spectrum component set to obtain the index value. If the index value is not an integer, round it down. Use the value at that index value in the discrimination value sequence as the interference discrimination threshold. Compare the preliminary discrimination value of each component with the interference discrimination threshold. If the preliminary discrimination value is greater than the interference discrimination threshold, mark the component as a valid signal component; otherwise, mark the component as an interference candidate component. Calculate the mean of the fluctuation intensity of all interference candidate components. Re-mark the interference candidate components with fluctuation intensity greater than the mean as interference signal components, and mark the remaining interference candidate components as valid signal components. Arrange all interference candidate components in sequence to form an interference component marking vector representing the refracted interference component. In the interference component marking vector, the position marked as interference signal component is assigned a value of 1, and the position marked as valid signal component is assigned a value of 0.

[0029] The interference component marker vector is analyzed and reconstructed to obtain the pure signal strength value representing the effective signal strength after interference removal. Specifically, this includes: adding the values ​​of all effective signal components in the interference component marker vector at the same time position to obtain the reconstructed effective signal sequence. This sequence is the same length as the original dynamic signal data and completely preserves the signal components after removing the refraction interference. Calculate the arithmetic mean of the reconstructed effective signal sequence and normalize the result to the 0-1 interval to obtain the pure signal strength value representing the effective signal strength after removing interference.

[0030] By adaptively decomposing each dynamic signal segment and generating an adaptive singular spectrum component set, and then adaptively setting the interference discrimination threshold by combining the distortion feature index, the refraction interference components are accurately identified and marked, and finally the pure signal strength value is obtained. This allows for the precise removal of interference signal components caused by refraction from the transparent container, while retaining the effective signal components. This effectively solves the problem of refraction interference masking the effective signal in existing technologies, and improves the signal purity and received signal strength of the infrared pair in transparent container transport monitoring scenarios.

[0031] In one embodiment, the original signal data is calculated to obtain a reference strength value representing the stable energy level of the original signal. The pure signal strength value and the reference strength value are then analyzed together to obtain a gain determination factor for judging whether the received signal strength of the target object has improved, including: Data processing is performed on the original signal data to obtain a benchmark intensity value representing the stable energy level of the original signal. Specifically, this includes: taking the instantaneous volatility at the median position of the instantaneous volatility sequence as the volatility threshold, marking the sampling points with instantaneous volatility less than the volatility threshold as stable points, and forming a stable data set with all stable points. Calculate the arithmetic mean of all values ​​in the stable dataset as the initial baseline strength value, and calculate the standard deviation of all values ​​in the stable dataset. Use the ratio of the standard deviation to the initial baseline strength value as the coefficient of variation. Finally, the skewness coefficient of the values ​​in the stable dataset is calculated. If the skewness coefficient is greater than 0, the coefficient of variation is compared with the ratio of the interquartile range of the values ​​in the stable dataset to the median. If the coefficient of variation is less than the ratio, the preliminary baseline strength value is used as the baseline strength value; otherwise, the median of the values ​​in the stable dataset is used as the baseline strength value. If the skewness coefficient is less than 0, the coefficient of variation is compared with the ratio of the range of the values ​​in the stable dataset to the mean. If the coefficient of variation is less than the ratio, the preliminary reference strength value is used as the reference strength value. Otherwise, the arithmetic mean and the median of the values ​​in the stable dataset are used as the reference strength value. If the skewness coefficient is equal to 0, the preliminary reference strength value is directly used as the reference strength value, thus obtaining the reference strength value representing the stable energy level of the original signal.

[0032] In one embodiment, the original signal data is calculated to obtain a reference strength value representing the stable energy level of the original signal. The pure signal strength value and the reference strength value are then analyzed together to obtain a gain determination factor for judging whether the received signal strength of the target object has improved. The method further includes: The pure signal strength value is compared with the reference strength value to obtain the gain determination factor for judging whether the received signal strength of the target object has been improved. Specifically, the reference strength value is used as a reference, the pure signal strength value is used as the value to be evaluated, the difference between the value to be evaluated and the reference is calculated, and the difference is divided by the reference to obtain the preliminary gain ratio. The preliminary gain ratio reflects the degree of relative change of the signal strength after interference removal relative to the original stable energy level. Calculate the difference between the maximum and minimum instantaneous volatility of all sampling points in the original signal data, divide the difference by the median of the instantaneous volatility sequence to obtain the volatility range coefficient, and multiply the initial gain ratio by the volatility range coefficient to obtain the corrected gain ratio. The ratio of the number of all stable points in the original signal data to the total number of sampling points is calculated as the stability weight. The corrected gain ratio is multiplied by 1 and summed with the stability weight to obtain the gain determination factor for judging whether the received signal strength of the target object has been improved. If the gain determination factor is greater than 0, it means that the received signal strength of the target object has increased; if the gain determination factor is less than 0, it means that the received signal strength of the target object has decreased; if the gain determination factor is equal to 0, it means that the received signal strength of the target object has not changed. In this way, by using the form of 1 + stability weight, the weight suppresses the core gain when the proportion of stable points is low, ensuring that the determination factor can truly reflect the signal strength improvement effect, while retaining the adjustment effect of stability on the determination result.

[0033] By processing the original signal data, stable points are identified, and a stable data set is constructed. Then, a reference strength value is calculated to ensure that the reference strength value accurately reflects the stable energy level of the original signal under interference-free or low-interference conditions. Finally, the pure signal strength value and the reference strength value are analyzed together. By calculating the preliminary gain ratio, fluctuation range coefficient, and stability weight, the gain determination factor is finally obtained. The gain determination factor not only reflects the relative improvement of the signal strength after removing refraction interference, but also integrates the fluctuation characteristics of the original signal and the proportion of stable points, which can accurately determine the actual gain effect after refraction interference processing.

[0034] In one embodiment, an infrared photodiode receiving signal strength enhancement system based on data processing is applied to the above-described method, comprising: The data analysis unit is used to acquire the raw signal data of the target object's receiving end, analyze the raw signal data, and obtain the distortion characteristic index, which represents the degree of refraction interference of the light-transmitting container. The distortion analysis unit is used to adaptively segment the original signal data according to the distortion characteristic index to obtain several dynamic signal segments with independent distortion characteristics. The interference analysis unit is used to decompose each dynamic signal segment into data, separate out and remove the interference signal components caused by the refraction of the light-transmitting container, and obtain the pure signal strength value representing the effective signal strength after the interference is removed. The strength judgment unit is used to calculate the original signal data to obtain a reference strength value representing the stable energy level of the original signal. The pure signal strength value and the reference strength value are analyzed together to obtain a gain judgment factor to determine whether the received signal strength of the target object has been improved.

[0035] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for enhancing the signal strength of an infrared photodiode receiver based on data processing, characterized in that, include: Step S1: Acquire the raw signal data from the target object's receiving end, analyze the raw signal data, and obtain the distortion characteristic index representing the degree of refraction interference from the light-transmitting container, including: Phase space reconstruction is performed on the original signal data to map the one-dimensional time-domain signal to a high-dimensional space, resulting in a high-dimensional trajectory matrix representing the dynamic evolution characteristics of the signal. Matrix decomposition is performed on the high-dimensional spatial trajectory matrix to separate the characteristic components of the light-transmitting container refraction interference, resulting in an interference characteristic subspace representing the spatial distribution of the interference signal; Energy distribution calculations are performed on the interference characteristic subspace to obtain the interference energy concentration coefficient representing the intensity of refracted interference energy; Data analysis of the interference energy concentration coefficient yields a distortion characteristic index representing the degree of refractive interference from the transparent container; Step S2: Based on the distortion feature index, the original signal data is adaptively segmented to obtain several dynamic signal segments with independent distortion features. Step S3: Decompose each dynamic signal segment into data, separate out and remove the interference signal components caused by the refraction of the light-transmitting container, and obtain the pure signal strength value representing the effective signal strength after removing the interference. Step S4: Calculate the original signal data to obtain a reference strength value representing the stable energy level of the original signal. Perform a joint analysis between the pure signal strength value and the reference strength value to obtain a gain determination factor for judging whether the received signal strength of the target object has been improved.

2. The method for enhancing the signal strength of an infrared photodiode receiver based on data processing according to claim 1, characterized in that, Based on the distortion characteristic index, the original signal data is adaptively segmented to obtain several dynamic signal segments with independent distortion characteristics, including: Time series analysis of the distortion feature index is performed within a sliding window to analyze the data difference between adjacent sliding windows, resulting in a dynamic distortion boundary sequence representing the boundary of the signal distortion region. The original signal data is adaptively divided based on the dynamic distortion boundary sequence to obtain several dynamic signal segments with independent distortion characteristics.

3. The method for enhancing the signal strength of an infrared photodiode receiver based on data processing according to claim 2, characterized in that, For each dynamic signal segment, data decomposition is performed to separate and remove the interference signal components caused by refraction from the transparent container, resulting in a pure signal strength value representing the effective signal strength after interference removal, including: Each dynamic signal segment is decomposed, an adaptive trajectory matrix is ​​constructed, and singular value decomposition is performed. Based on the characteristics of singular value energy distribution, the adaptive singular spectral component set representing the principal components of the signal is obtained.

4. The method for enhancing the signal strength of an infrared photodiode receiver based on data processing according to claim 3, characterized in that, For each dynamic signal segment, data decomposition is performed to separate and remove the interference signal component caused by refraction from the transparent container, resulting in a pure signal strength value representing the effective signal strength after interference removal. This also includes: Feature extraction is performed on the adaptive singular spectrum component set to identify and label the interference signal components, resulting in an interference component label vector representing the refraction interference component. The interference component marker vector is analyzed and reconstructed to obtain the pure signal strength value representing the effective signal strength after interference removal.

5. The method for enhancing the signal strength of an infrared photodiode receiver based on data processing according to claim 4, characterized in that, The raw signal data is calculated to obtain a reference strength value representing the stable energy level of the raw signal. The pure signal strength value and the reference strength value are then analyzed together to obtain a gain determination factor for judging whether the received signal strength of the target object has improved, including: Data processing is performed on the original signal data to obtain a reference intensity value representing the stable energy level of the original signal.

6. The method for enhancing the signal strength of an infrared photodiode receiver based on data processing according to claim 5, characterized in that, The raw signal data is calculated to obtain a reference strength value representing the stable energy level of the raw signal. The pure signal strength value is then co-analyzed with the reference strength value to obtain a gain determination factor for judging whether the received signal strength of the target object has improved. This also includes: The pure signal strength value is compared with the reference strength value to obtain the gain determination factor for judging whether the received signal strength of the target object has been improved.

7. A data processing-based infrared photodiode signal strength enhancement system, applied in the method described in any one of claims 1-6, characterized in that, include: The data analysis unit is used to acquire the raw signal data of the target object's receiving end, analyze the raw signal data, and obtain the distortion characteristic index, which represents the degree of refraction interference of the light-transmitting container. The distortion analysis unit is used to adaptively segment the original signal data according to the distortion characteristic index to obtain several dynamic signal segments with independent distortion characteristics. The interference analysis unit is used to decompose each dynamic signal segment into data, separate out and remove the interference signal components caused by the refraction of the light-transmitting container, and obtain the pure signal strength value representing the effective signal strength after the interference is removed. The strength judgment unit is used to calculate the original signal data to obtain a reference strength value representing the stable energy level of the original signal. The pure signal strength value and the reference strength value are analyzed together to obtain a gain judgment factor to determine whether the received signal strength of the target object has been improved.

Images