Communication analysis processing method and device based on dual-mode heterogeneous communication network fusion

Abstract

The application discloses a communication analysis processing method and device based on a dual-mode heterogeneous communication network fusion, relates to the technical field of communication processing, and comprises the following steps: acquiring real-time running data of HPLC and HRF in a fusion communication system, generating a coupling feature vector, and predicting disturbed frequency points and time periods of the future HRF by using a short-time prediction model; screening a clean frequency band candidate set based on the prediction result, calculating the coupling strength index and the adjacent interference index of each candidate frequency band, and obtaining a candidate front value; selecting a frequency band with the maximum candidate front value as a target communication frequency band in a future preset time window, and concentrating HRF links to the frequency band before interference occurs, which can predict the cross-medium interference that may occur in the future in advance, thereby actively avoiding potential influences before the interference occurs, effectively reducing the risk of data error codes, link interruption and switching delay caused by HPLC harmonic radiation to the HRF communication frequency band, and improving the continuity and reliability of communication.

Description

Technical Field

[0001] This invention relates to the field of communication processing technology, and more specifically to a communication analysis and processing method and apparatus based on the fusion of dual-mode heterogeneous communication networks. Background Technology

[0002] Communication based on the fusion of dual-mode heterogeneous communication networks using power line carrier communication (HPLC) and high-frequency wireless communication (HRF) refers to the simultaneous introduction of two complementary communication methods within the same system, ensuring the stability and reliability of data transmission through intelligent scheduling and switching. In complex scenarios such as the Internet of Things (IoT) for power systems, industrial control, and smart cities, communication systems often need to possess both wide-area coverage and the ability to provide high-speed, low-latency data transmission in localized areas. For example, in a smart grid environment, data needs to be exchanged in real time between substations, distribution rooms, and end users. Power line carrier communication (HPLC) can fully utilize existing power line resources, providing relatively economical and wide-area connectivity. However, its disadvantage lies in its sensitivity to external electromagnetic interference, which can easily lead to decreased transmission rates or increased bit error rates. On the other hand, high-frequency wireless communication (HRF) can provide high-speed and low-latency links over short distances, making it ideal for transmitting data with high real-time requirements, such as power equipment status information and smart terminal control commands. However, its drawbacks include limited spectrum resources and susceptibility to frequency congestion caused by ambient noise, equipment interference, or multi-user contention.

[0003] To meet the demands of these complex scenarios, dual-mode heterogeneous converged communication deploys both HPLC and HRF links simultaneously, utilizing a fusion scheduling mechanism to dynamically allocate resources during data transmission: when HPLC encounters strong interference leading to performance degradation, it can quickly switch to the HRF compensation link to ensure real-time performance; when HRF experiences frequency band interference or coverage limitations, HPLC ensures continuous transmission, thus forming a complementary communication architecture. This approach has wide application value in the interconnection of power dispatch centers and distributed energy nodes, in the data backhaul of sensor clusters in smart cities, and in the collaborative control of multiple devices in industrial plants.

[0004] However, in a dual-mode heterogeneous communication environment integrating HPLC and HRF, although the two types of links belong to different media (electric power lines and radio waves) at the physical level, the harmonics generated by HPLC during high-frequency transmission can couple across the medium into the operating frequency band of HRF, forming a kind of "cross-medium covert interference." This type of interference is often random and dynamic, and conventional noise suppression methods are ineffective. Existing research focuses more on rate and link utilization, but neglects cross-medium interference and security risks, lacking a mechanism that combines prediction and suppression. This can lead to errors in link switching or even exploitation by attackers. Therefore, how to predict the frequency points that may be affected by interference in the future and select appropriate HRF communication bands in advance to suppress cross-medium interference remains an unsolved problem. Summary of the Invention

[0005] The purpose of this invention is to solve the problems mentioned above and to provide a communication analysis and processing method and apparatus based on the fusion of dual-mode heterogeneous communication networks.

[0006] In a first aspect of this invention, a communication analysis and processing method based on the fusion of dual-mode heterogeneous communication networks is first proposed, the method comprising:

[0007] S1: Acquire real-time operating data of HPLC and HRF in the converged communication system, and generate a coupling feature vector of HPLC and HRF based on the real-time operating data;

[0008] S2: Based on the coupled feature vector, a short-time prediction model is used to predict the future HRF frequency points that will be interfered with and the corresponding time periods, so as to obtain the prediction set of the future interference frequency points;

[0009] S3: Based on the predicted set of future interference frequencies, select a clean frequency band candidate set for HRF;

[0010] S4: For each frequency band in the clean frequency band candidate set, calculate the coupling strength index and adjacent interference index corresponding to the interfered frequency point to obtain the candidate top value;

[0011] S5: The clean frequency band with the largest candidate value in the clean frequency band candidate set is recorded as the optimal clean frequency band and used as the target communication frequency band for HRF within the future preset time window;

[0012] S6: Before the time period corresponding to the future interference frequency point arrives, concentrate the HRF transmission link to the target communication frequency band to achieve early suppression of cross-medium covert interference.

[0013] Optionally, the step of generating a coupled feature vector of HPLC and HRF based on real-time operational data includes:

[0014] HPLC and HRF signals within a preset time period were collected from the fusion communication system, and short-time fast Fourier transforms were performed on the HPLC and HRF signals respectively to obtain the power spectra of the HPLC signals and the power spectra of the HRF signals.

[0015] The power spectra of the HPLC signal and the power spectra of the HRF signal are aligned by time to form a joint spectrum matrix;

[0016] The Pearson correlation coefficient was calculated for the power sequence of each HPLC frequency point and the power sequence of the HRF frequency point in the joint spectrum matrix, and the Pearson correlation coefficient was denoted as the coupling strength.

[0017] The coupling strength between each frequency point of HPLC and each frequency point of HRF is used as the coupling feature vector of HPLC and HRF for a preset time period.

[0018] Optionally, the steps of using a short-time prediction model based on coupled feature vectors to predict future HRF frequencies and corresponding time periods, and obtaining a predicted set of future interfered frequencies, include:

[0019] The coupling feature vectors of HPLC and HRF over the past N preset time periods are arranged in chronological order as time series data. The time series data is then input into a short-term prediction model to output the coupling degree of each HRF frequency point in each time period in the future. The time period is a discrete time slice within the preset prediction window in the future.

[0020] The coupling degree is compared with a preset interference judgment threshold. If the coupling degree of a certain HRF frequency point in a certain future time period is not less than the preset interference judgment threshold, then the corresponding HRF frequency point is determined to be an interfered frequency point in the corresponding future time period. The greater the coupling degree, the more likely the HRF frequency point is to be affected by cross-media interference from HPLC.

[0021] By combining all the identified interference frequencies with their corresponding time periods, a predicted set of future interference frequencies is obtained.

[0022] Optionally, the step of selecting a clean frequency band candidate set for HRF based on the predicted set of future interference frequencies includes:

[0023] The available spectrum range of the HRF's operating frequency band is divided into several frequency bands according to a preset bandwidth granularity, forming a complete set of frequency bands;

[0024] Based on the predicted frequency points of future interference, find the frequency bands to which all the interference points belong, and remove the interference frequency bands from the full set of frequency bands.

[0025] The remaining frequency bands in the full set of frequency bands are used as the clean frequency band candidate set for HRF.

[0026] Optionally, the calculation steps of the coupling strength index include:

[0027] For each clean frequency band in the clean frequency band candidate set, calculate the average of the start frequency and the end frequency, and use it as the center frequency of the corresponding clean frequency band;

[0028] For each interference frequency in the predicted set of future interference frequency points, calculate the absolute difference between the frequency of each interference frequency point and the center frequency of the clean frequency band, divide the absolute difference by the center frequency of the clean frequency band to obtain the difference ratio, and normalize the reciprocal of the difference ratio to obtain the frequency normalization difference.

[0029] Obtain the coupling strength between the interfered frequency point and the center of the clean frequency band, and divide the coupling strength by the corresponding frequency normalization difference to obtain the coupling asymmetry between the center of the clean frequency band and the interfered frequency point;

[0030] The geometric cumulative risk value corresponding to the clean frequency band is obtained by accumulating the geometric mean of the coupling asymmetry between the center of all clean frequency bands and the frequency points affected by interference.

[0031] Optionally, the calculation step of the coupling strength index further includes:

[0032] Divide each coupling asymmetry by the sum of all coupling asymmetry to obtain the probability distribution of the corresponding coupling asymmetry; and calculate the difference entropy based on the probability distribution of all coupling asymmetry, which is used as the difference entropy corresponding to the clean frequency band.

[0033] Multiply the difference entropy corresponding to the clean frequency band by the corresponding geometric cumulative risk value to obtain the coupling strength index of the corresponding clean frequency band.

[0034] Optionally, the calculation steps of the adjacent interference index include:

[0035] For all frequency bands in the HRF band set, they are numbered in order. For each clean frequency band in the set, it is called the target frequency band. The number of interference points contained in the adjacent target frequency bands is counted and recorded as the first interference point and the second interference point respectively. The first interference point and the second interference point are added together to obtain the total number of adjacent interference points.

[0036] Calculate the absolute difference between the number of the first interference point and the number of the second interference point, and divide the absolute difference by the sum of the total number of adjacent interference points and the value of 1. The quotient of the division is taken as the adjacent interference difference rate of the target frequency band.

[0037] For each target frequency band Traverse all frequency bands sequentially to the left and right, and for each frequency band... The number of frequency points affected by the interference is recorded as follows: The extended adjacent interference intensity value of each target frequency band is calculated using the following formula: In the formula, This represents the extended adjacent interference intensity value of the target frequency band. Indicates frequency band The number of future interference frequencies Indicates frequency band With target frequency band The number of frequency band intervals between them; This represents the total number of all frequency bands in the HRF band set;

[0038] Divide the value corresponding to the total number of adjacent interference points by the value corresponding to the extended adjacent interference intensity value to obtain the adjacency concentration of the target frequency band.

[0039] Optionally, the calculation step of the adjacent interference index further includes:

[0040] Identify whether both adjacent frequency bands to the left and right of the target frequency band have interfering frequencies. If both adjacent frequency bands have interfering frequencies, the target frequency band is marked as a high-impact band, and the impact variable of the high-impact band is marked as 2. If only one adjacent frequency band has an interfering frequency, the target frequency band is marked as a medium-impact band, and the impact variable of the medium-impact band is marked as 1. If neither adjacent frequency band has an interfering frequency, the target frequency band is marked as a low-impact band, and the impact variable of the low-impact band is marked as 0.

[0041] The adjacent interference difference rate, adjacent concentration and pincer attack variables of the target frequency band are added together, and the sum is divided by 3 to obtain the adjacent interference index of the target frequency band.

[0042] Optionally, the step of calculating the coupling strength index and adjacent interference index corresponding to the interfered frequency point to obtain the candidate top value includes:

[0043] For each clean frequency band in the clean frequency band candidate set, a two-dimensional coordinate system is established for the coupling strength index and the adjacent interference index of each clean frequency band, and the coupling strength index and the adjacent interference index of each clean frequency band are mapped to the two-dimensional coordinate system; the coordinate points correspond to the coupling strength index and the adjacent interference index, respectively.

[0044] Calculate the distance from the coordinate point in the two-dimensional coordinate system to the centroid, and use the reciprocal of the distance as the candidate leading value for the corresponding clean frequency band.

[0045] In a second aspect of the invention, a communication analysis and processing apparatus based on the fusion of dual-mode heterogeneous communication networks is proposed, the apparatus comprising:

[0046] Coupling Feature Module: Acquires real-time operating data of HPLC and HRF in the converged communication system, and generates coupling feature vectors of HPLC and HRF based on the real-time operating data;

[0047] Prediction module: Based on the coupled feature vector, a short-time prediction model is used to predict the future HRF frequency points to be interfered with and the corresponding time periods, so as to obtain the prediction set of the future interference frequency points;

[0048] Filtering module: Based on the predicted set of future interference frequencies, filter out a clean frequency band candidate set for HRF;

[0049] Candidate leading value module: For each frequency band in the clean frequency band candidate set, calculate the coupling strength index and adjacent interference index corresponding to the interfered frequency point to obtain the candidate leading value;

[0050] Target communication frequency band module: The clean frequency band with the largest candidate value in the clean frequency band candidate set is recorded as the optimal clean frequency band, and is used as the target communication frequency band for HRF within the future preset time window;

[0051] Analysis and processing module: Before the time period corresponding to the future interference frequency point arrives, the transmission link of HRF is concentrated to the target communication frequency band to achieve early suppression of cross-medium covert interference.

[0052] The beneficial effects of this invention are:

[0053] This invention proposes a communication analysis and processing method and apparatus based on the fusion of dual-mode heterogeneous communication networks. By acquiring real-time operational data from HPLC and HRF in the fused communication system, a coupling feature vector is generated, and a short-time prediction model is used to predict the interference frequencies and time periods of the HRF in the future. Based on the prediction results, a candidate set of clean frequency bands is selected, and the coupling strength index and adjacent interference index of each candidate frequency band are calculated to obtain the leading candidate value. The frequency band with the largest leading candidate value is selected as the target communication frequency band within a preset time window, and the HRF link is concentrated to this frequency band before interference occurs, achieving early suppression of cross-medium covert interference. Through this method, the system can predict potential cross-medium interference in advance and intelligently select the optimal clean frequency band for dynamic switching based on the prediction results, thereby proactively avoiding potential impacts before interference occurs. This not only effectively reduces the risk of data errors, link interruptions, and switching delays caused by HPLC harmonic radiation to the HRF communication frequency band, but also improves the continuity and reliability of communication, while enhancing the system's protection against sudden interference and potential security threats, enabling the fused network to maintain efficient and stable operation in complex power IoT, industrial control, or smart city environments. Attached Figure Description

[0054] The invention will now be further described with reference to the accompanying drawings.

[0055] Figure 1 A flowchart of a communication analysis and processing method based on the fusion of dual-mode heterogeneous communication networks;

[0056] Figure 2 This is a framework diagram of a communication analysis and processing device based on the fusion of dual-mode heterogeneous communication networks. Detailed Implementation

[0057] 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.

[0058] This invention provides a communication analysis and processing method based on the fusion of dual-mode heterogeneous communication networks. See also... Figure 1 , Figure 1 A flowchart illustrating a communication analysis and processing method based on the fusion of dual-mode heterogeneous communication networks provided in an embodiment of the present invention. The method includes the following steps:

[0059] S1: Acquire real-time operating data of HPLC and HRF in the converged communication system, and generate a coupling feature vector of HPLC and HRF based on the real-time operating data;

[0060] S2: Based on the coupled feature vector, a short-time prediction model is used to predict the future HRF frequency points that will be interfered with and the corresponding time periods, so as to obtain the prediction set of the future interference frequency points;

[0061] S3: Based on the predicted set of future interference frequencies, select a clean frequency band candidate set for HRF;

[0062] S4: For each frequency band in the clean frequency band candidate set, calculate the coupling strength index and adjacent interference index corresponding to the interfered frequency point to obtain the candidate top value;

[0063] S5: The clean frequency band with the largest candidate value in the clean frequency band candidate set is recorded as the optimal clean frequency band and used as the target communication frequency band for HRF within the future preset time window;

[0064] S6: Before the time period corresponding to the future interference frequency point arrives, concentrate the HRF transmission link to the target communication frequency band to achieve early suppression of cross-medium covert interference.

[0065] Based on the communication analysis and processing method for dual-mode heterogeneous communication network fusion provided in this invention, the system can predict potential cross-medium interference in advance and intelligently select the optimal clean frequency band for dynamic switching based on the prediction results, thereby proactively avoiding potential impacts before interference occurs. This not only effectively reduces the risk of data errors, link interruptions, and switching delays caused by HPLC harmonic radiation to the HRF communication band, but also improves the continuity and reliability of communication. Simultaneously, it enhances the system's protection against sudden interference and potential security threats, enabling the fused network to maintain efficient and stable operation in complex power IoT, industrial control, or smart city environments.

[0066] In one embodiment, S1: The step of acquiring real-time operating data of HPLC and HRF in the converged communication system and generating a coupling feature vector of HPLC and HRF based on the real-time operating data includes:

[0067] HPLC and HRF signals were acquired from the fused communication system within a preset time period, and short-time fast Fourier transforms were performed on the HPLC and HRF signals respectively to obtain the power spectrum P of the HPLC signal. HPLC Power spectrum P of (t, f) and HRF signal HRF (t, f); where t represents the time point (sampling time), usually in seconds (s); f represents the signal frequency point (the frequency value corresponding to the sampling frequency), in Hertz (Hz), P HPLC (t, f) and P HRF (t, f) represent the power of the HPLC signal at time t and frequency f, and the power of the HRF signal at time t and frequency f, respectively;

[0068] The power spectra of the HPLC signal and the power spectra of the HRF signal are aligned by time to form a joint spectral matrix M(t,f). HPLC ,f HRF ), each element M in the matrix ijk This indicates at time point t i HPLC frequency points f HPLC,j With HRF frequency point f HRF,k The combined power pair;

[0069] The Pearson correlation coefficient was calculated between the power sequences of each HPLC frequency point and the power sequences of the HRF frequency points in the joint spectrum matrix. The formula for the calculation is as follows:

[0070] ,

[0071] In the formula, This represents the Pearson correlation coefficient between frequency pairs, which is also the coupling strength. For HPLC at frequency point f HPLC,j average power, For HRF at frequency point f HRF,k The average power; finally, the coupling strength matrix R is obtained, whose rows correspond to HPLC frequency points, columns correspond to HRF frequency points, and matrix elements represent the coupling strength between the two; the correlation coefficient is the statistical correlation calculated over the entire preset time period, used to quantify the coupling strength between HPLC and HRF at each frequency point within that time period;

[0072] The coupling strength between each frequency point of HPLC and each frequency point of HRF is used as the coupling feature vector of HPLC and HRF for a preset time period. In the formula, m represents the number of HPLC frequency points; n represents the number of HRF frequency points; and the dimension of matrix R is m×n.

[0073] It should be noted that the preset time period usually refers to an observation window that can fully reflect the statistical characteristics of HPLC and HRF signals, and can generally be set between several hundred milliseconds and several seconds, depending on the actual situation. HPLC signal refers to the data carrier signal transmitted through power lines, which usually comes from the output of power line carrier modulation chip or power line coupler, and can be directly obtained from the carrier communication interface of power distribution line or terminal equipment. HRF signal refers to the radio frequency signal transmitted through high-frequency wireless link, which is usually collected by radio frequency transceiver module, such as Wi-Fi, Sub-GHz or customized high-frequency wireless module, which can directly output intermediate frequency or baseband signal for processing. In actual systems, software-defined radio (SDR) devices can be used to sample power line signals and radio frequency signals simultaneously, thereby ensuring the synchronous acquisition of the two types of signals. To achieve time alignment, a unified clock source and timestamp mechanism are typically employed. This can be achieved using GPS time synchronization, IEEE 1588 Precision Clock Synchronization (PTP), or a local high-precision clock. Each data segment is then assigned a strictly consistent timestamp during sampling. This ensures that even if HPLC and HRF are in different acquisition channels, their power spectra can be aligned to a unified time axis during subsequent processing. After performing short-time Fourier transforms on both, the sampling time points are aligned using the PTP protocol, allowing the power characteristics of each frequency point in HPLC and HRF at the same moment to be obtained. This guarantees that each element in the subsequent joint spectrum matrix truly corresponds to a power pair at the "same time." This approach ensures that the signals from two heterogeneous media correspond precisely in the time dimension, allowing the calculated coupling strength to truly reflect the physical relationship across the media.

[0074] In one embodiment, S2: The step of using a short-time prediction model based on coupled feature vectors to predict future HRF frequencies and corresponding time periods to obtain a predicted set of future interfered frequencies includes:

[0075] The coupling feature vectors of HPLC and HRF over the past N preset time periods are arranged in chronological order and used as time series data. The time series data is then input into the short-term prediction model to output the coupling degree of each HRF frequency point in each time period in the future. The time period is a discrete time slice within the preset prediction window in the future.

[0076] The coupling degree is compared with a preset interference judgment threshold. If the coupling degree of a certain HRF frequency point in a certain future time period is not less than the preset interference judgment threshold, then the corresponding HRF frequency point is determined to be an interfered frequency point in the corresponding future time period. The greater the coupling degree, the more likely the HRF frequency point is to be affected by cross-media interference from HPLC.

[0077] By combining all the identified interference frequencies with their corresponding time periods, a predicted set of future interference frequencies is obtained.

[0078] It should be noted that in step S2, a short-time prediction model is used based on the coupling feature vectors to predict the future HRF frequencies affected by interference and their corresponding time periods, resulting in a prediction set of future interference frequencies. Specifically, the coupling feature vectors of HPLC and HRF obtained from the past N preset time periods are arranged sequentially to form a time-series data sequence. This sequence reflects how the coupling strength between HPLC and HRF frequencies has evolved over time. Then, this time-series sequence is input into a preset prediction model. The so-called short-time prediction model refers to a model capable of processing time-series data and predicting future trends. It can be a model based on recurrent neural networks (such as LSTM or GRU), a time-series prediction model based on convolution, or a lightweight autoregressive model. Its core function is to learn the "changing patterns of coupling features over several past time periods" and the "possible patterns of future HRF frequencies." The mapping relationship between "coupling degree" and "coupling degree" is established. After receiving the input, the short-term prediction model outputs the result of a future prediction window. This result is in discrete time slices, giving the coupling degree value of each HRF frequency point in the corresponding time slice. The larger the coupling degree value, the more likely the frequency point is to be affected by cross-media interference from HPLC in the future. In order to identify specific interfered frequency points from the model output, the predicted coupling degree needs to be compared with the preset interference judgment threshold. When the coupling degree value of a certain HRF frequency point in a certain future time slice is greater than or equal to the threshold, the HRF frequency point is determined to be a "future interfered frequency point" in that time slice. By traversing all frequency points and time slices in the entire prediction window, a "prediction set of future interfered frequency points" can be finally obtained. This set consists of several "frequency point-time period" pairs, which fully characterize which HRF frequency points may be interfered with in the future and how long the interference will last. For example, assuming the coupling feature vectors of the past N=5 time periods are used as input, the prediction model outputs the coupling degree results for the next 15 seconds. Among them, the coupling degree of HRF frequency point 2.445GHz is 0.82 between the 1st and 8th seconds of the future, and the preset interference judgment threshold is 0.7. Then the system will determine that 2.445GHz is an interference frequency point in this time period. Similarly, if the coupling degree of 2.556GHz is 0.75 between the 8th and 15th seconds of the future, it will also be judged as an interference frequency point. The final prediction set can be represented as {(2.445GHz,[1s,8s]),(2.556GHz,[8s,15s])}, which provides a direct input basis for the subsequent screening of clean frequency band candidate sets.

[0079] In one embodiment, S3: The step of selecting a clean frequency band candidate set for HRF based on the predicted set of future interference frequencies includes:

[0080] The available spectrum range of the HRF operating frequency band is divided into several frequency bands according to a preset bandwidth granularity, forming a complete set of frequency bands; for example, if the HRF operating frequency band is 2.400GHz–2.480GHz, with a total bandwidth of 80MHz, dividing it into 5MHz segments will yield 16 channels, namely Ch1=2.400–2.405GHz, Ch2=2.405–2.410GHz all the way up to Ch16=2.475–2.480GHz;

[0081] Based on the predicted frequency points of future interference, find the frequency bands to which all the interference points belong, and remove the interference frequency bands from the full set of frequency bands. For example, if the prediction set shows that 2.445GHz will be interfered with within the next 1–8 seconds, then it belongs to Ch10 = 2.445–2.450GHz, and Ch10 needs to be removed from the full set.

[0082] The remaining frequency bands in the full set of frequency bands are used as the clean frequency band candidate set for HRF.

[0083] It should be noted that the available spectrum range of the HRF operating frequency band refers to the frequency range that has been determined during the system design and is available for HRF communication. It is usually determined based on communication protocols, regulatory requirements, and actual equipment capabilities. In this embodiment, it is assumed that the HRF operating frequency band is 2.400GHz–2.480GHz, which is a total of 80MHz of continuous spectrum resources. This means that within this range, the system can complete data transmission tasks through different frequency points. The preset bandwidth granularity refers to the strategy of dividing the entire spectrum into equal-width intervals according to fixed bandwidth intervals. This interval value is usually consistent with the channel bandwidth used in HRF communication, such as 5MHz or 10MHz. The purpose of the division is to quantize the continuous spectrum into several manageable frequency band units, which facilitates subsequent interference labeling, elimination, and selection management. For example, dividing an 80MHz spectrum into 5MHz units can yield 16 channels, with each channel serving as a frequency band unit for subsequent analysis. The advantage of this approach is that it transforms the complex continuous spectrum structure into a discrete structure, making it easier to map the predicted set of future interference frequencies onto specific frequency band numbers, thereby achieving precise location and elimination. Moreover, since the bandwidth of each frequency band is fixed, the remaining frequency bands can be uniformly evaluated to determine whether they meet the transmission requirements without the need for re-division or parameter adjustment.

[0084] The advantages of using this method to screen clean frequency band candidate sets are high operational efficiency, clear computational logic, and strong real-time performance. It is particularly suitable for application scenarios where dynamic interference changes frequently and scheduling systems need to respond quickly. For example, in a typical smart factory, if a certain frequency band of HRF is predicted to pose a risk in the next 8 seconds due to electromagnetic interference from nearby equipment, and this frequency band corresponds to the Ch10 channel, the system can immediately remove Ch10 to avoid scheduling tasks to this potentially conflicting frequency band. At the same time, it can retain the remaining risk-free frequency bands for subsequent scheduling, significantly improving the reliability, stability, and anti-interference capability of the communication system.

[0085] In one embodiment, S4: For each frequency band in the clean frequency band candidate set, calculate the coupling strength index and adjacent interference index corresponding to the interfered frequency point to obtain the candidate top value;

[0086] In one implementation, the steps for calculating the coupling strength index include:

[0087] For each clean frequency band in the clean frequency band candidate set, calculate the average of the start frequency and the end frequency, and use it as the center frequency of the corresponding clean frequency band;

[0088] For each affected frequency in the predicted set of future interference frequencies, calculate the absolute difference between the frequency of each affected frequency and the center frequency of the clean frequency band. Divide the absolute difference by the center frequency of the clean frequency band to obtain the difference ratio. Then, normalize the reciprocal of the difference ratio to obtain the frequency normalization difference. The calculation formula is as follows: In the formula, Indicates the first The center frequency and interference frequency of the clean frequency band The frequency normalization difference, Indicates the first The center frequency of a clean frequency band Indicates the frequency point affected by interference The frequency; The closer the value is to 1, the closer the frequency is to 1.

[0089] Obtain the frequency points affected by interference The coupling strength between the center of the clean frequency band and the center of the clean frequency band is calculated by dividing the coupling strength by the normalized frequency difference between the two. The center of the clean frequency band and the frequency point affected by interference are obtained. The degree of coupling asymmetry;

[0090] The geometric mean of the coupling asymmetry between the center of all clean frequency bands and the interfered frequency points is accumulated to obtain the geometric cumulative risk value corresponding to the clean frequency bands.

[0091] Divide each coupling asymmetry by the sum of all coupling asymmetry degrees to obtain the probability distribution of the corresponding coupling asymmetry degree; then calculate the difference entropy based on the probability distribution of all coupling asymmetry degrees, which is used as the difference entropy corresponding to the clean frequency band; the calculation formula is: In the formula, Indicates the first The difference entropy of a clean frequency band Indicates the first The center and interference points of the clean frequency band The coupling asymmetry, This represents the sum of all coupling asymmetry degrees. Indicates the first The center and interference points of the clean frequency band The probability distribution of the coupling asymmetry. This represents the total number of future interfered frequency points in the predicted set of future interfered frequency points; the sum of the probability distributions corresponding to the coupling asymmetry is 1, which can be understood as the distribution of risk between the candidate frequency band and different interfered frequency points. Substituting these probability distributions into the concept of information entropy: if a certain interference point occupies the majority, the value of the difference entropy will be relatively small, indicating that the risk is relatively concentrated; if the proportions of multiple interference points are similar, the value of the difference entropy will be relatively large, indicating that the risk is relatively dispersed. The final difference entropy factor is used to describe the complexity of the risks faced by the candidate frequency band.

[0092] Multiply the difference entropy corresponding to the clean frequency band by the corresponding geometric cumulative risk value to obtain the coupling strength index of the corresponding clean frequency band.

[0093] It should be noted that in the calculation of the coupling strength index, the center frequency of the clean band comes from the division of the HRF working frequency band in the previous step S3. The start frequency and end frequency of each sub-band are known, and the center frequency of the band can be obtained by calculating the average of the two. Secondly, the frequency information of the future interference frequency points all come from the prediction output of step S2. S2 is based on the coupling feature vector of several past time periods input into the prediction model to obtain the coupling degree of each frequency point in the discrete time slice, and determines which frequency points belong to the interference frequency points in a specific time period by combining the preset threshold. The coupling strength between the interference frequency points and the center of the clean band can be queried from the coupling strength matrix, or it can be obtained by other methods, which are not limited or elaborated.

[0094] It should be noted that the coupling strength index reflects the degree of correlation and risk complexity between the candidate clean frequency band and the predicted interfered frequency point in the future. A higher coupling strength index indicates that the candidate clean frequency band is not only close to the frequency of the interfered frequency point, but also has a high degree of coupling within the predicted time period. Furthermore, this coupling strength is often not concentrated at a single point, but rather dispersed across multiple different interfered frequency points. This means that the overall anti-interference performance of the candidate frequency band is poor, and it is likely to experience sudden or continuous interference risks in the future, thus its priority is naturally reduced. Conversely, a lower coupling strength index indicates that the candidate frequency band is farther away from the interfered frequency point in terms of frequency or time, or even if coupling strength exists, it is concentrated at only a few points, without a large-scale and dispersed risk. In this case, the candidate frequency band is more stable and suitable as a new communication frequency band for HRF. In other words, a higher coupling strength index indicates a stronger coupling between the frequency band and the interfered frequency point, and it may be affected by multiple interfered points simultaneously. This means that it faces a greater risk of interference and has lower stability during future communication, thus the system will lower its priority as a new HRF communication frequency band. Conversely, a lower coupling strength index indicates that the frequency band is farther away from the interference point or has a weaker coupling degree. Even if there is an impact, it is mostly concentrated and controllable. Such a frequency band is safer and more reliable in actual use, and naturally has a higher priority in the candidate set. For example, if the coupling strength index of candidate frequency band A is 0.2, it means that its association with the interference point is limited and the risk is concentrated, so it is more likely to be selected as a new communication frequency band. On the other hand, if the coupling strength index of candidate frequency band B is 0.7, it indicates that it has a significant association with the interference point in multiple future periods, and the risk is complex and unpredictable. Therefore, its priority is significantly lower than that of frequency band A.

[0095] It should be noted that the reason for calculating the coupling strength index in the above manner, rather than using a simple linear weighting or mean method, is that this method can more comprehensively and meticulously capture the multidimensional characteristics of the risks that candidate frequency bands may face in future communications, and avoid misjudgments due to a single factor. Specifically, the advantages of the above approach are as follows: First, by calculating the normalized frequency difference between the center frequency and the interfered frequency, the proximity in frequency space can be accurately characterized, reflecting the potential interference possibility from the source. Second, by introducing "coupling asymmetry" instead of directly using coupling strength, both the predicted strength and frequency proximity can be considered simultaneously, ensuring that the result does not unilaterally amplify one factor. Third, by using the geometric mean accumulation instead of the arithmetic mean to handle the impact of multiple interference points, the most severe interference effect can be highlighted and the masking effect of occasional values ​​on the overall situation can be weakened, thus allowing the overall risk of the candidate frequency band to be presented more realistically. Then, by introducing difference entropy as a measure of risk distribution, the different situations of "risk concentrated at a single point" and "risk dispersed at multiple points" can be effectively distinguished, which is crucial for communication frequency band selection because dispersed risks are often more difficult to predict and avoid than concentrated risks. Finally, multiplying the geometric cumulative risk value by the difference entropy to obtain the final degree not only maintains the dimensionless quantity and the normalized range of 0 to 1, but also naturally integrates strength and dispersion into a unified index. The advantage of this calculation method is that it avoids the subjectivity of manually setting weights and enhances the sensitivity and discriminative power of the index to real risks through a series of nonlinear and information-theoretic calculations. This allows the final coupling strength index to more scientifically reflect the reliability priority of different candidate frequency bands. For example, if candidate frequency band A is strongly coupled to only one interference point, then its differential entropy is small, the risk is concentrated, and its finality will not be too high, so the system may still select it. On the other hand, even if the individual interference points of candidate frequency band B do not have a significant impact, they are distributed across multiple points. The combined effect of geometric accumulation and differential entropy increases the finality, and the system will automatically lower its priority. In this way, a more robust and reasonable determination of frequency band priority can be achieved under complex interference environments.

[0096] In one embodiment, the calculation steps for the neighbor interference index include:

[0097] For all frequency bands in the HRF band set, they are numbered in order. For each clean frequency band in the set, it is called the target frequency band. The number of interference points contained in the adjacent target frequency bands is counted and recorded as the first interference point and the second interference point respectively. The first interference point and the second interference point are added together to obtain the total number of adjacent interference points.

[0098] Calculate the absolute difference between the number of the first interference point and the number of the second interference point, and divide the absolute difference by the sum of the total number of adjacent interference points and the value of 1. The quotient of the division is taken as the adjacent interference difference rate of the target frequency band.

[0099] For each target frequency band Traverse all frequency bands sequentially to the left and right. and For each frequency band The number of frequency points affected by interference is recorded as follows: The extended adjacent interference intensity value of each target frequency band is calculated using the following formula: In the formula, This represents the extended adjacent interference intensity value of the target frequency band. Indicates frequency band The number of future interference frequencies Indicates frequency band With target frequency band The number of frequency band intervals between them; It represents the total number of all frequency bands in the HRF band set; Extended Adjacent Interference Strength, by considering the interference of all frequency bands, represents the overall interference trend and intensity of the neighborhood;

[0100] Divide the value corresponding to the total number of adjacent interference points by the value corresponding to the extended adjacent interference intensity value to obtain the adjacency concentration of the target frequency band.

[0101] Identify whether both adjacent frequency bands to the left and right of the target frequency band have interfering frequencies. If both adjacent frequency bands have interfering frequencies, the target frequency band is marked as a high-impact band, and the impact variable of the high-impact band is marked as 2. If only one adjacent frequency band has an interfering frequency, the target frequency band is marked as a medium-impact band, and the impact variable of the medium-impact band is marked as 1. If neither adjacent frequency band has an interfering frequency, the target frequency band is marked as a low-impact band, and the impact variable of the low-impact band is marked as 0.

[0102] The adjacent interference difference rate, adjacent concentration and pincer attack variables of the target frequency band are added together, and the sum is divided by 3 to obtain the adjacent interference index of the target frequency band.

[0103] It should be noted that the Adjacent Interference Index (AII) is a comprehensive indicator used to measure the structural risk of interference in neighboring frequency bands for a clean frequency band. This index focuses on three spatial trends that suggest the frequency band may be subject to interference in the future: whether the interference distribution is uneven (i.e., adjacent interference difference rate), whether the interference is concentrated near the target frequency band (i.e., adjacent concentration), and whether the frequency band is in a pincer movement (i.e., pincer variable). These three factors together reflect whether a clean frequency band is surrounded by interference paths, subjected to concentrated high-density interference pressure, or asymmetrically affected by interference points in future communication deployments. The risk level is as follows: The adjacent interference difference rate indicates the degree of difference in the number of interference points on the left and right sides of the target frequency band. A higher difference rate indicates that the interference is concentrated on one side, potentially leading to diffraction or backlash interference. Adjacency concentration characterizes the proximity of interference points within the extended adjacent area to the target frequency band. A higher concentration indicates denser interference points close to the target frequency band, forming a high-voltage, strong interference encirclement trend. The pincer attack variable indicates whether the frequency band is in a situation where interference points are simultaneously approaching from both sides. A value of 2 indicates interference on both sides, a high pincer attack state, which is the most dangerous area. A value of 1 indicates a medium pincer attack, with the pincer attack direction biased to one side. A value of 0 indicates that the interference points have not yet formed a squeeze-and-encircle. The adjacent interference index integrates these three trend quantification results to form an intuitive criterion for the future security of each clean frequency band. The larger the value, the more likely the frequency band, although currently clean, is to become an intrusion hotspot in the future due to interference transfer, diffusion, or aggregation. Therefore, it should not be prioritized as an HRF communication frequency band. Otherwise, it may encounter interference intrusion shortly after deployment, resulting in a sharp decline in communication quality. Furthermore, since the frequency band was originally marked as clean, the system may not trigger fault tolerance or frequency hopping mechanisms in a timely manner, causing the communication link to operate in a state of hidden degradation for a long time. It is impossible to proactively avoid and strategically suppress cross-medium interference. For example, if a frequency band has multiple highly concentrated interference points on both sides, forming a pincer attack, but currently has no recorded interference points, and the system fails to identify the high adjacent interference index and selects it as the primary communication frequency, then when interference propagates to that frequency band in the future, it will experience a "concentrated attack without defense," ultimately losing the robustness and foresight of the HRF spectrum deployment strategy. Therefore, the system should sort candidate frequency bands according to the adjacent interference index, prioritizing frequency bands with lower indices and lower spatial interference structure distribution risks to ensure the stability and anti-interference capability of the communication channel. It should be noted that...

[0104] It should be noted that the advantage of calculating the Adjacent Interference Index (AIOI) using the above method is that it can comprehensively assess the risk of future interference erosion of clean frequency bands from three key dimensions, providing a comprehensive judgment capability in terms of structure, space, and trend. The Adjacent Interference Difference Rate reflects whether the left-right interference distribution is unbalanced, helping to identify potential diffraction or biased interference paths; the Adjacent Concentration quantifies the proximity of interference by extending the interference density distribution trend within the neighborhood, revealing whether interference pressure is focusing on the target frequency band; the Pincer Attack Variable qualitatively captures the pincer attack situation where interference approaches from both sides simultaneously, used to determine whether the target frequency band is in a high-risk area. The AIOI synthesized from these three factors considers both the symmetry of the number of interference points and assesses the spatial distribution trend and pressure changes of interference, avoiding one-sided judgments based solely on the current clean state or a single-sided neighborhood. Compared to traditional statistical indicators or methods that only consider the static number of adjacent interferences, this method more accurately screens out frequency bands with true long-term stability, preventing short-term impact from interference spread, thereby significantly improving the anti-interference robustness and stability of HRF communication deployments.

[0105] In one embodiment, the step of calculating the coupling strength index and the adjacent interference index corresponding to the interfered frequency point to obtain the candidate top value includes:

[0106] For each clean frequency band in the clean frequency band candidate set, a two-dimensional coordinate system is established for the coupling strength index and the adjacent interference index of each clean frequency band, and the coupling strength index and the adjacent interference index of each clean frequency band are mapped to the two-dimensional coordinate system; the coordinate points correspond to the coupling strength index and the adjacent interference index, respectively.

[0107] Calculate the distance from the coordinate point in the two-dimensional coordinate system to the centroid, and use the reciprocal of the distance as the candidate leading value for the corresponding clean frequency band.

[0108] It should be noted that the advantage of calculating the candidate top value using the above method is that it unifies two different dimensions of risk factors into a comparable spatial framework, achieving intuitive selection of superiority and inferiority through a geometric approach. This more objectively reflects the dual stability of the frequency band in terms of both "current interference coupling strength" and "future adjacent structure threat," ensuring that the selected communication frequency band is not easily affected by historical interference coupling after deployment, nor will it fail in the future due to the accumulation of adjacent interference. The final step uses the reciprocal of the distance as the top value, further amplifying the advantages of low-risk frequency bands, making the system more inclined to select the ideal frequency band with "low interference + low structural risk."

[0109] In one embodiment, S5: The clean frequency band with the largest candidate value in the clean frequency band candidate set is recorded as the optimal clean frequency band and used as the target communication frequency band for HRF within a future preset time window.

[0110] It should be noted that after completing the dual evaluation of the coupling strength index and the adjacent interference index, the system can further accurately identify the most suitable frequency band for future communication by calculating the candidate leading value of each clean frequency band. Specifically, the system identifies the frequency band with the largest candidate leading value in the clean frequency band candidate set as the current optimal clean frequency band and directly designates it as the target communication frequency band for the HRF system within a preset future time window. This process ensures that the selected frequency band has a low risk in terms of historical interference coupling relationships and also has stronger anti-erosion capabilities in terms of spatial interference structure. For example, in a certain communication scenario, there are frequency bands F1, F2, and F3 in the candidate set, where F1 has a coupling strength index of 0.3 and an adjacent interference index of 0.2, F2 has 0.1 and 0.5, and F3 has 0.2 and 0.2. After two-dimensional coordinate calculation, F3 has the largest leading value, so the system will automatically select F3 as the communication frequency band, thereby achieving a more robust cross-media communication deployment. This mechanism not only improves the scientific nature of communication frequency band selection but also enhances the robustness and foresight of the system in dynamic environments with multiple interferences.

[0111] In one embodiment, S6: Before the time period corresponding to the future interference frequency point arrives, the transmission link of HRF is concentrated to the target communication frequency band to achieve early suppression of cross-medium covert interference.

[0112] Before the time period corresponding to the future interference frequency arrives, the system pre-schedules and concentrates the HRF transmission links to the currently evaluated optimal target communication frequency band, achieving proactive avoidance and suppression of cross-medium covert interference. Because cross-medium interference often has suddenness, unpredictable paths, and penetrability, waiting until the interference actually falls into the frequency band before frequency hopping or switching is often too late, resulting in link performance degradation or information loss. However, by pre-deploying communication links in target frequency bands selected based on coupling strength index and adjacent interference index, upcoming interference hotspots can be avoided, ensuring the continuity and resilience of communication quality. For example, if the system predicts that frequency band F3 will be subject to electromagnetic interference from high-power frequency converters in 10 seconds, and the system identifies frequency band F1 as structurally far from the interference source with low adjacent interference and coupling strength, the system will pre-concentrate the HRF links to F1 before the interference actually propagates, thus achieving proactive suppression of covert interference without relying on real-time frequency hopping.

[0113] Based on the same inventive concept, embodiments of the present invention also provide a communication analysis and processing device based on the fusion of dual-mode heterogeneous communication networks. See also Figure 2 , Figure 2 This is a framework diagram of a communication analysis and processing device based on the fusion of dual-mode heterogeneous communication networks provided in an embodiment of the present invention. The device includes:

[0114] Coupling Feature Module: Acquires real-time operating data of HPLC and HRF in the converged communication system, and generates coupling feature vectors of HPLC and HRF based on the real-time operating data;

[0115] Prediction module: Based on the coupled feature vector, a short-time prediction model is used to predict the future HRF frequency points to be interfered with and the corresponding time periods, so as to obtain the prediction set of the future interference frequency points;

[0116] Filtering module: Based on the predicted set of future interference frequencies, filter out a clean frequency band candidate set for HRF;

[0117] Candidate leading value module: For each frequency band in the clean frequency band candidate set, calculate the coupling strength index and adjacent interference index corresponding to the interfered frequency point to obtain the candidate leading value;

[0118] Target communication frequency band module: The clean frequency band with the largest candidate value in the clean frequency band candidate set is recorded as the optimal clean frequency band, and is used as the target communication frequency band for HRF within the future preset time window;

[0119] Analysis and processing module: Before the time period corresponding to the future interference frequency point arrives, the transmission link of HRF is concentrated to the target communication frequency band to achieve early suppression of cross-medium covert interference.

[0120] Based on the communication analysis and processing device based on the fusion of dual-mode heterogeneous communication networks provided in this embodiment of the invention, the system can predict potential cross-medium interference in advance through the above-mentioned method, and intelligently select the optimal clean frequency band for dynamic switching based on the prediction results, thereby proactively avoiding potential impacts before interference occurs. This not only effectively reduces the risk of data errors, link interruptions, and switching delays caused by HPLC harmonic radiation to the HRF communication frequency band, but also improves the continuity and reliability of communication, while enhancing the system's ability to protect against sudden interference and potential security threats, enabling the fused network to maintain efficient and stable operation in complex power Internet of Things, industrial control, or smart city environments.

[0121] 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 claims of this invention should still fall within the patent coverage of this invention.

Claims

1. A communication analysis processing method based on fusion of dual-mode heterogeneous communication networks, characterized by, The method comprises the following steps: obtaining real-time running data of HPLC and HRF in a converged communication system, and generating a coupling feature vector of HPLC and HRF according to the real-time running data; using a short-term prediction model based on the coupling feature vector to predict future disturbed HRF frequency points and corresponding time periods, and obtaining a prediction set of future disturbed frequency points; on the basis of the prediction set of future disturbed frequency points, screening out a candidate set of clean frequency bands of HRF; for each frequency band in the candidate set of clean frequency bands, calculating a coupling strength index and a neighboring interference index corresponding to the disturbed frequency points, and obtaining a candidate front value; recording a clean frequency band with the maximum candidate front value in the candidate set of clean frequency bands as an optimal clean frequency band, and taking the optimal clean frequency band as a target communication frequency band of HRF in a future preset time window; before the time period corresponding to the future disturbed frequency points arrives, concentrating transmission links of HRF to the target communication frequency band, so as to realize early suppression of cross-medium covert jamming; the coupling strength index calculation step comprises: for each clean frequency band in the candidate set of clean frequency bands, calculating the mean value of the start frequency and the end frequency as the center frequency of the corresponding clean frequency band; for each disturbed frequency point in the prediction set of future disturbed frequency points, calculating the absolute difference value between the frequency of the disturbed frequency point and the center frequency of the clean frequency band, dividing the absolute difference value by the center frequency of the clean frequency band to obtain a difference ratio, and normalizing the inverse of the difference ratio to obtain a frequency normalized difference; obtaining the coupling strength between the center of the disturbed frequency point and the clean frequency band, and dividing the coupling strength by the corresponding frequency normalized difference to obtain the coupling asymmetry degree of the center of the clean frequency band and the disturbed frequency point; geometrically accumulating all coupling asymmetry degrees of the center of the clean frequency band and the disturbed frequency point to obtain the geometric accumulation risk value corresponding to the clean frequency band; the coupling strength index calculation step further comprises: dividing each coupling asymmetry degree by the sum of all coupling asymmetry degrees to obtain the distribution probability of the corresponding coupling asymmetry degree; and calculating the difference entropy according to the distribution probability of all coupling asymmetry degrees as the difference entropy corresponding to the clean frequency band; multiplying the difference entropy corresponding to the clean frequency band by the corresponding geometric accumulation risk value to obtain the coupling strength index of the corresponding clean frequency band; the neighboring interference index calculation step comprises: for all frequency bands in the HRF frequency band set, labeling in sequence, for each clean frequency band in the frequency band set, recording as a target frequency band, counting the number of disturbed frequency points corresponding to the adjacent target frequency bands, respectively recording as the first interference point number and the second interference point number, adding the first interference point number and the second interference point number to obtain the total number of adjacent interference points; calculating the absolute difference value of the first interference point number and the second interference point number, and dividing the absolute difference value by the sum of the total number of adjacent interference points and the value 1, and taking the quotient as the adjacent interference difference rate of the target frequency band; For each target frequency band , all the frequency bands are traversed in turn from left to right and right to left respectively, and for each frequency band , the number of interfered frequency points is counted and recorded as , the extended adjacent interference intensity value of each target frequency band is calculated, and the formula for calculation is: , wherein represents the extended adjacent interference intensity value of the target frequency band, represents the number of future interfered frequency points in the frequency band , and represents the number of frequency band intervals between the frequency band and the target frequency band ; represents the total number of all frequency bands in the HRF frequency band set. dividing the value corresponding to the total number of adjacent interference points by the value corresponding to the extended adjacent interference strength value to obtain the adjacent concentration degree of the target frequency band; the neighboring interference index calculation step further comprises: If the target frequency band is adjacent to the left and right frequency bands, and the left and right adjacent frequency bands both exist interference frequency points, the target frequency band is recorded as a high jamming frequency band, and the jamming variable of the high jamming frequency band is recorded as 2; if the target frequency band is adjacent to the left and right frequency bands, and only one adjacent frequency band exists interference frequency points, the target frequency band is recorded as a medium jamming frequency band, and the jamming variable of the medium jamming frequency band is recorded as 1; if the target frequency band is adjacent to the left and right frequency bands, and the left and right adjacent frequency bands both do not exist interference frequency points, the target frequency band is recorded as a low jamming frequency band, and the jamming variable of the low jamming frequency band is recorded as 0; The adjacent interference difference rate, the adjacent concentration degree and the jamming variable of the target frequency band are added, and the sum is divided by 3 to obtain the adjacent interference index of the target frequency band.

2. The communication analysis processing method based on fusion of dual-mode heterogeneous communication networks according to claim 1, characterized in that, The step of generating the coupling feature vector of the HPLC and the HRF according to the real-time running data comprises: respectively collecting the HPLC signal and the HRF signal in a preset time period in the fusion communication system, and respectively performing short-time fast Fourier transform on the HPLC signal and the HRF signal to obtain the power spectrum of the HPLC signal and the power spectrum of the HRF signal; aligning the power spectrum of the HPLC signal and the power spectrum of the HRF signal by time to form a joint spectrum matrix; calculating the Pearson correlation coefficient of the power sequence of each HPLC frequency point and the power sequence of the HRF frequency point in the joint spectrum matrix, and recording the Pearson correlation coefficient as the coupling strength; taking the coupling strength between each frequency point of the HPLC and each frequency point of the HRF as the coupling feature vector of the HPLC and the HRF in the preset time period.

3. The communication analysis processing method based on fusion of dual-mode heterogeneous communication networks according to claim 1, characterized in that, The step of predicting the future interfered HRF frequency points and the corresponding time period based on the coupling feature vector using a short-term prediction model to obtain a prediction set of future interfered frequency points comprises: arranging the coupling feature vectors of the HPLC and the HRF in the past N preset time periods in time sequence as time series data, and inputting the time series data into a short-term prediction model to output the coupling degree of each HRF frequency point in each time period in the future time period; the time period is a discrete time slice in the future preset prediction window; comparing the coupling degree with a preset interference judgment threshold, if the coupling degree of a certain HRF frequency point in a certain future time period is not less than the preset interference judgment threshold, it is determined that the corresponding HRF frequency point is an interfered frequency point in the corresponding future time period; combining all the determined interfered frequency points and the corresponding time period to obtain a prediction set of future interfered frequency points.

4. The communication analysis processing method based on fusion of dual-mode heterogeneous communication networks according to claim 1, characterized by, On the basis of the prediction set of future interfered frequency points, the step of screening out a clean frequency band candidate set of the HRF comprises: dividing the available frequency spectrum range of the working frequency band of the HRF into a plurality of frequency bands according to a preset bandwidth granularity to form a frequency band universe; finding the frequency bands to which all the interfered frequency points belong as interfered frequency bands according to the interfered frequency points contained in the prediction set of future interfered frequency points, and removing the interfered frequency bands from the frequency band universe; taking the remaining frequency bands in the frequency band universe as the clean frequency band candidate set of the HRF.

5. The communication analysis processing method based on fusion of dual-mode heterogeneous communication networks according to claim 1, characterized by, The step of calculating the coupling strength index and the adjacent interference index corresponding to the interfered frequency points to obtain a candidate front value comprises: For each clean frequency band in the clean frequency band candidate set, a two-dimensional coordinate system is established for the coupling strength index and the adjacent interference index of each clean frequency band, and the coupling strength index and the adjacent interference index of each clean frequency band are mapped into the two-dimensional coordinate; the coordinate points correspond to the coupling strength index and the adjacent interference index respectively; The distance of the coordinate point in the two-dimensional coordinate system to the origin is calculated, and the reciprocal of the distance is taken as the candidate front value of the corresponding clean frequency band.

6. The communication analysis processing apparatus based on dual-mode heterogeneous communication network fusion, for implementing the communication analysis processing method based on dual-mode heterogeneous communication network fusion according to any one of claims 1-5, characterized in that, The device comprises: A coupling feature module: real-time running data of HPLC and HRF in the converged communication system is acquired, and a coupling feature vector of HPLC and HRF is generated according to the real-time running data; A prediction module: based on the coupling feature vector, a short-term prediction model is used to predict the future interfered HRF frequency points and the corresponding time period, and a prediction set of future interfered frequency points is obtained; A screening module: on the basis of the prediction set of future interfered frequency points, a clean frequency band candidate set of HRF is screened out; A candidate front value module: for each frequency band in the clean frequency band candidate set, the coupling strength index and the adjacent interference index corresponding to the interfered frequency points are calculated to obtain the candidate front value; A target communication frequency band module: the clean frequency band with the maximum candidate front value in the clean frequency band candidate set is recorded as the optimal clean frequency band, and is taken as the target communication frequency band of HRF in the future preset time window; An analysis and processing module: before the time period corresponding to the future interfered frequency points arrives, the transmission link of HRF is concentrated to the target communication frequency band, so as to realize the advance suppression of cross-medium covert jamming.

Images