Electromagnetic spectrum situation prediction method and system based on time series model

Abstract

The present invention discloses a method and system for electromagnetic spectrum situation prediction based on a time series model, including: collecting electromagnetic spectrum time series data and preprocessing the collected data; training a pre-built time series model and estimating model parameters based on the electromagnetic spectrum time series data obtained by preprocessing; using the trained time series model to predict the electromagnetic spectrum situation to obtain prediction results, and evaluating the prediction results. The present invention selects a suitable time series model and accurately estimates parameters based on the periodicity and trend characteristics of electromagnetic spectrum data, can accurately capture the change pattern, and is more accurate in both short-term and long-term predictions, providing a reliable basis for decisions such as spectrum resource allocation.

Description

Technical Field

[0001] The present invention relates to the technical field of electromagnetic spectrum situation prediction, and in particular to an electromagnetic spectrum situation prediction method and system based on a time series model. Background Art

[0002] With the increasing use of modern communications, radar, electronic countermeasures and other electromagnetic-related applications, electromagnetic spectrum resources are becoming increasingly scarce and the electromagnetic environment is becoming increasingly complex. Accurately predicting the electromagnetic spectrum situation is crucial for the rational allocation of spectrum resources, ensuring communication security, and improving the performance of electronic systems. Traditional electromagnetic spectrum monitoring methods can only obtain the current spectrum usage, but cannot effectively predict the future spectrum situation. Although there are some prediction methods, they have problems such as low prediction accuracy and poor adaptability to complex electromagnetic environments. Therefore, there is an urgent need for a more accurate and efficient electromagnetic spectrum situation prediction method. Summary of the invention

[0003] The present application provides an electromagnetic spectrum situation prediction method and system based on a time series model to solve the problems that traditional electromagnetic situation analysis methods are difficult to process multi-source electromagnetic data, have low data processing efficiency, inaccurate fusion, etc.

[0004] According to the first aspect, an embodiment provides an electromagnetic spectrum situation prediction method based on a time series model, the method comprising:

[0005] Collect electromagnetic spectrum time series data and pre-process the collected data;

[0006] Based on the electromagnetic spectrum time series data obtained by preprocessing, the pre-built time series model is trained and the model parameters are estimated;

[0007] The trained time series model is used to predict the electromagnetic spectrum situation, obtain the prediction results, and evaluate the prediction results.

[0008] Furthermore, electromagnetic spectrum time series data is collected, including:

[0009] Electromagnetic spectrum monitoring sites are deployed in typical electromagnetic environment areas. Each monitoring site is equipped with a high-performance spectrum analyzer. The spectrum analyzer collects electromagnetic spectrum time series data at preset time intervals. The collected data includes signal strength, frequency occupancy, and signal type information.

[0010] Furthermore, the collected data is preprocessed, specifically including:

[0011] Data cleaning: Clean the collected raw electromagnetic spectrum data to remove outliers; identify and correct data points that are significantly deviated from the normal range by setting reasonable data thresholds;

[0012] Data normalization: Normalize the cleaned electromagnetic spectrum data.

[0013] Furthermore, based on the electromagnetic spectrum time series data obtained by preprocessing, the pre-built time series model is trained and the model parameters are estimated, which specifically includes:

[0014] Select appropriate time series models based on the characteristics of electromagnetic spectrum data, including:

[0015] For electromagnetic spectrum data with obvious periodicity and trend, the seasonal autoregressive integrated moving average model is selected; for data showing short-term autocorrelation and moving average characteristics, the autoregressive moving average model is selected.

[0016] Furthermore, based on the electromagnetic spectrum time series data obtained by preprocessing, the pre-built time series model is trained and the model parameters are estimated, which specifically includes:

[0017] Using the preprocessed electromagnetic spectrum time series data, the parameters of the selected time series model are estimated by methods including maximum likelihood estimation.

[0018] Furthermore, based on the electromagnetic spectrum time series data obtained by preprocessing, the pre-built time series model is trained and the model parameters are estimated, which specifically includes:

[0019] For the seasonal autoregressive integrated moving average model SARIMA(p,d,q)(P,D,Q) s Model, determine the autoregressive order p, difference order d, moving average order q of the non-seasonal part and the corresponding parameters P, D, Q and seasonal period s of the seasonal part;

[0020] According to the autocorrelation function ACF and partial autocorrelation function PACF diagrams of electromagnetic spectrum data, the optimal parameter combination is determined in combination with the information criterion to ensure that the model can accurately fit the inherent laws of electromagnetic spectrum data.

[0021] Furthermore, the trained time series model is used to predict the electromagnetic spectrum situation and obtain prediction results, including:

[0022] Short-term prediction: Input the latest pre-processed electromagnetic spectrum data into the constructed time series model to perform short-term electromagnetic spectrum situation prediction;

[0023] Long-term prediction: Through long-term trend analysis of historical data and models, long-term prediction of the electromagnetic spectrum situation can be achieved.

[0024] Furthermore, the prediction results are evaluated, including:

[0025] The mean square error (MSE) and mean absolute error (MAE) of the prediction results are calculated. If the MSE or MAE exceeds the preset threshold, the data preprocessing process and model parameters are rechecked.

[0026] Furthermore, the method further comprises:

[0027] Using the latest electromagnetic spectrum data collected regularly, the time series model is regularly optimized and trained, and the model parameters are adjusted.

[0028] According to the second aspect, an embodiment provides an electromagnetic spectrum situation prediction system based on a time series model, the system comprising:

[0029] The data acquisition and processing module is used to acquire electromagnetic spectrum time series data and pre-process the acquired data;

[0030] A model building module is used to train a pre-built time series model and estimate model parameters based on the electromagnetic spectrum time series data obtained by preprocessing;

[0031] The prediction module is used to use the trained time series model to predict the electromagnetic spectrum situation, obtain prediction results, and evaluate the prediction results.

[0032] According to a third aspect, an embodiment provides an electronic device, the device comprising: a processor and a memory;

[0033] The memory is used to store one or more program instructions;

[0034] The processor is used to run one or more program instructions to execute the steps of an electromagnetic spectrum situation prediction method based on a time series model as described in any of the above items.

[0035] According to the fourth aspect, an embodiment provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of a method for electromagnetic spectrum situation prediction based on a time series model as described in any one of the above items are implemented.

[0036] The present application provides an electromagnetic spectrum situation prediction method and system based on a time series model, which has the following beneficial effects:

[0037] (1) Improve prediction accuracy. According to the periodicity and trend characteristics of electromagnetic spectrum data, appropriate time series models are selected and parameters are accurately estimated, which can accurately capture the change pattern and make both short-term and long-term predictions more accurate, providing a reliable basis for decision-making such as spectrum resource allocation.

[0038] (2) Optimize data quality. Comprehensive data collection and scientific preprocessing processes, including removal of outliers and normalization, ensure high data quality, enhance consistency and comparability, and lay the foundation for model construction and prediction.

[0039] (3) Enhanced adaptability and stability. The evaluation and feedback module uses a variety of indicators to monitor the prediction results, adjusts the model according to the situation, and updates and optimizes it in combination with new data. It can adapt to changes in the electromagnetic environment such as the deployment of new communication systems and maintain stable prediction performance.

[0040] (4) Assist in electromagnetic spectrum management decision-making. Short-term predictions assist in real-time spectrum allocation, and long-term predictions provide a reference for spectrum planning and long-term management of the electromagnetic environment, improve spectrum utilization efficiency, and ensure a stable and orderly electromagnetic environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 A flowchart of an electromagnetic spectrum situation prediction method based on a time series model provided by an embodiment of the present invention;

[0042] Figure 2 A data preprocessing flow chart of an electromagnetic spectrum situation prediction method based on a time series model provided by an embodiment of the present invention;

[0043] Figure 3 A schematic diagram of time series model parameter estimation in an electromagnetic spectrum situation prediction method based on a time series model provided by an embodiment of the present invention;

[0044] Figure 4 An overall architecture diagram of an electromagnetic spectrum situation prediction system based on a time series model provided for one embodiment of the present invention. DETAILED DESCRIPTION

[0045] The present invention is further described in detail below by specific embodiments in conjunction with the accompanying drawings. Wherein similar elements in different embodiments adopt associated similar element numbers. In the following embodiments, many detailed descriptions are for making the present application better understood. However, those skilled in the art can easily recognize that some features can be omitted in different situations, or can be replaced by other elements, materials, methods. In some cases, some operations related to the present application are not shown or described in the specification, this is to avoid the core part of the present application being overwhelmed by too much description, and for those skilled in the art, it is not necessary to describe these related operations in detail, and they can fully understand the related operations according to the description in the specification and the general technical knowledge in the art.

[0046] In addition, the features, operations or characteristics described in the specification can be combined in any appropriate manner to form various implementations. At the same time, the steps or actions in the method description can also be interchanged or adjusted in a manner that is obvious to those skilled in the art. Therefore, the various sequences in the specification and the drawings are only for the purpose of clearly describing a certain embodiment and are not meant to be a required sequence, unless otherwise specified that a certain sequence must be followed.

[0047] The first embodiment of the present invention provides an electromagnetic spectrum situation prediction method based on a time series model. First, a high-quality data acquisition module is constructed to obtain multi-dimensional data covering different frequency bands and times from multiple electromagnetic spectrum monitoring points, including signal strength, frequency occupancy, and signal type, and the acquisition time interval can be flexibly adjusted. Next is the data preprocessing step, which cleans up abnormal values ​​caused by equipment failure and electromagnetic interference by setting scientific thresholds, and uses normalization methods to make data from different sources comparable. Then, according to the periodicity, trend and other characteristics of the electromagnetic spectrum data, a suitable time series model, such as SARIMA or ARMA, is selected, and the model parameters are determined using advanced parameter estimation methods. Finally, the processed data is input into the model to achieve short-term and long-term predictions. Short-term predictions provide a basis for real-time spectrum resource allocation, and long-term predictions assist in spectrum planning and long-term management of the electromagnetic environment. At the same time, an evaluation and feedback mechanism is provided, and the prediction results are evaluated using indicators such as mean square error and mean absolute error. The model is adjusted according to the results to ensure that the model is dynamically optimized to adapt to the ever-changing electromagnetic environment. The following is combined with Figure 1 Provide detailed explanation.

[0048] like Figure 1 As shown, in step S100, electromagnetic spectrum time series data is collected and preprocessed.

[0049] Data collection: Design a dedicated data collection system that can obtain electromagnetic spectrum data of different frequency bands and at different times from multiple electromagnetic spectrum monitoring sites. These monitoring sites can be distributed in different geographical areas to ensure that the collected data is widely representative. The collected data includes but is not limited to information such as signal strength, frequency occupancy, and signal type. The time interval for data collection can be adjusted according to actual conditions. For example, it can be set to every hour, every minute, or even shorter time intervals to obtain high-resolution electromagnetic spectrum time series data.

[0050] In this embodiment, electromagnetic spectrum monitoring stations are deployed in multiple typical electromagnetic environment areas such as urban centers and communication base station dense areas. Each monitoring station is equipped with a high-performance spectrum analyzer, whose working principle is based on the law of electromagnetic induction and Fourier transform. The spectrum analyzer can collect electromagnetic spectrum data in real time, and the collected frequency band ranges from f min =30MHz to f max =3000MHz, according to the formula:

[0051]

[0052] The frequency band is divided into N equally spaced sub-bands, where Δf represents the sub-band bandwidth.

[0053] The data collection time interval Δt = 10 minutes is set, and the collected data includes signal strength I(f, t) (indicating signal strength at frequency and time), frequency occupancy O(f, t) (O(f, t) = 1 indicates that the frequency is occupied at time, O(f, t) = 0 indicates that it is not occupied), signal type T(f, t) (determined by signal feature recognition algorithm, such as based on signal modulation mode, etc.) and other information. The collected data is transmitted to the data processing center through the wireless communication network.

[0054] Data preprocessing, such as Figure 2 As shown:

[0055] Data cleaning: Clean the collected raw electromagnetic spectrum data to remove abnormal values ​​caused by monitoring equipment failure, electromagnetic interference, etc. By setting reasonable data thresholds, identify and correct data points that are significantly deviated from the normal range. For example, if the signal strength of a certain frequency band suddenly exceeds the historical data mean by several times and does not conform to the normal fluctuation pattern of the frequency band, it is judged as an abnormal value and processed.

[0056] In this embodiment, the collected raw data is cleaned in the data processing center. Let μ(f) be the historical mean of the signal strength at frequency f, and σ(f) be its standard deviation. For each data point I(f,t), if |I(f,t)-μ(f)|>kσ(f) (where k=3), the data point is judged to be an outlier. For the processing of outliers, the mean of the data at adjacent time points in the frequency band is used for replacement, that is:

[0057]

[0058] Wherein m is the number of selected adjacent time points, and m=2 is taken here.

[0059] Data normalization: In order to make the data of different frequency bands and different monitoring sites comparable, the cleaned electromagnetic spectrum data is normalized. Common normalization methods, such as minimum-maximum normalization, can be used to map the data to the [0,1] interval. The formula is:

[0060]

[0061] Where x is the original data, x min and x max are the minimum and maximum values ​​in the data sequence, respectively, new is the normalized data.

[0062] In this embodiment, taking the signal strength data I(f, t) of a certain frequency band collected by a monitoring station as an example, the minimum value I in the data sequence is calculated. min (f) and maximum value I max (f), according to the minimum-maximum normalization formula:

[0063]

[0064] Each data point is normalized so that the normalized data I normalized (f,t) is in the interval [0,1].

[0065] like Figure 1 As shown, in step S200, the pre-constructed time series model is trained and model parameters are estimated based on the electromagnetic spectrum time series data obtained by preprocessing.

[0066] Model selection: Select an appropriate time series model based on the characteristics of electromagnetic spectrum data. For electromagnetic spectrum data with obvious periodicity (such as periodic changes in the use of electromagnetic spectrum due to factors such as day and night, working days and rest days) and trend (such as the trend of the use frequency of certain frequency bands gradually increasing or decreasing with the application of new technologies), the seasonal autoregressive integrated moving average model (SARIMA) can be selected. For data that exhibit short-term autocorrelation and moving average characteristics, the autoregressive moving average model (ARMA) can be considered.

[0067] In this embodiment, based on the previous analysis of electromagnetic spectrum data, it was found that there is an obvious periodicity with s = 24 hours as a period (due to the diurnal pattern of human activities) and a long-term trend (with the development of urban communication services Changes), and the seasonal autoregressive integrated moving average model (SARIMA) is selected.

[0068] Model parameter estimation: Using the collected and preprocessed electromagnetic spectrum time series data, the parameters of the selected time series model are estimated by methods such as maximum likelihood estimation. For example, for SARIMA(p,d,q)(P,D,Q) s Model, determine the autoregressive order p, difference order d, moving average order q of the non-seasonal part and the corresponding parameters P, D, Q and seasonal period s of the seasonal part. According to the autocorrelation function (ACF) and partial autocorrelation function (PACF) diagrams of electromagnetic spectrum data, combined with information criteria (such as AIC, BIC), determine the optimal parameter combination to ensure that the model can accurately fit the inherent laws of electromagnetic spectrum data.

[0069] In this embodiment, Figure 3 As shown, the specific calculation process is as follows:

[0070] Let y t is the preprocessed electromagnetic spectrum time series data (such as normalized signal intensity), for SARIMA(p,d,q)(P,D,Q) s The model parameters are determined by analyzing the autocorrelation function (ACF) and partial autocorrelation function (PACF) diagrams of the electromagnetic spectrum data and combining them with the Akaike information criterion (AIC).

[0071] Autocorrelation function:

[0072]

[0073] Where n is the sample size, Yes t , k is the lag order.

[0074] The calculation of the partial autocorrelation function PACF is relatively complicated and can be solved by methods such as the Yule-Walker equation.

[0075] The AIC calculation formula is AIC = -2ln(L) + 2(p + q + P + Q + d + D + 1), where L is the likelihood function value. By constantly trying different parameter combinations, find the parameters that minimize AIC.

[0076] After analysis, the parameters of the SARIMA model are determined to be (p=2, d=1, q=2) (P=1, D=1, Q=1) 24 .

[0077] The maximum likelihood estimation method is used to estimate the model parameters. The goal of maximum likelihood estimation is to find the parameters θ (including all autoregressive, moving average and seasonal parameters) so that the likelihood function Maximum, where f(y t |θ,y t-1 ,y t-2 ,…) is the conditional probability density function.

[0078] like Figure 1 As shown, in step S300, the trained time series model is used to perform electromagnetic spectrum situation forecasting to obtain forecast results, and the forecast results are evaluated.

[0079] Short-term prediction: The latest pre-processed electromagnetic spectrum data is input into the constructed time series model to perform short-term electromagnetic spectrum situation prediction. Short-term prediction can predict the electromagnetic spectrum situation in the next few minutes, tens of minutes or even hours. For example, it can predict the signal strength change trend and frequency occupancy of a key frequency band in the next hour, providing a basis for real-time spectrum resource allocation.

[0080] In this embodiment, the latest normalized electromagnetic spectrum data y is tnew Input into the constructed SARIMA model. For SARIMA(p,d,q)(P,D,Q) s Model, the prediction formula is:

[0081]

[0082] in is the predicted value, φ i and θ j are the non-seasonal autoregressive and moving average parameters, Φ i and θ j are seasonal autoregressive and moving average parameters, ∈ tis a white noise sequence, and L is the number of steps ahead of the prediction. Here, the signal strength changes and frequency occupancy of each frequency band are predicted within the next L = 30 minutes (converted into the number of steps of the time series). The prediction results are presented to spectrum managers through a visual interface so that they can adjust the spectrum allocation strategy in a timely manner.

[0083] Long-term prediction: Through the long-term trend analysis of historical data and models, long-term prediction of the electromagnetic spectrum situation can be achieved. Long-term prediction can predict the overall usage trend of different frequency bands in the next few days, weeks or even months, and provide reference for spectrum planning and long-term electromagnetic environment management. In the long-term prediction process, considering the possible changes in the electromagnetic environment (such as the deployment of new communication systems, the development of large-scale electronic activities, etc.), the prediction results can be appropriately adjusted in combination with expert experience and other relevant information.

[0084] In this embodiment, the forecast is conducted once a week to predict the overall usage trend of different frequency bands in the next two weeks (T = 14 days, converted into the number of steps of the time series). In the forecasting process, considering the information that new 5G base stations will be deployed in some areas in the near future (assuming the impact factor is α), the forecast results are appropriately adjusted in combination with expert experience. The adjustment formula is:

[0085]

[0086] in is the long-term adjusted forecast value, is the model prediction value, △I 5G (t) is the expected change in the electromagnetic spectrum caused by base station deployment (estimated based on expert experience and related models), which provides a basis for long-term spectrum planning.

[0087] Forecast result evaluation: The mean square error (MSE) and mean absolute error (MAE) of the forecast results are calculated every day. The formula for the mean square error is: The mean absolute error formula is: in is the predicted value, y t is the actual value, and n is the number of samples for prediction. If MSE exceeds 0.05 or exceeds 0.03, recheck the data preprocessing process and model parameters.

[0088] like Figure 1 As shown, in step S400, the time series model is regularly optimized and trained using the latest electromagnetic spectrum data collected regularly, and the model parameters are adjusted.

[0089] Model adjustment feedback: The model is retrained and parameter optimized based on the newly collected electromagnetic spectrum data every month. The retraining process is similar to the previous model parameter estimation process. ACF, PACF, and AIC are recalculated with new data to determine new optimal parameters to ensure that the model can adapt to changes in the electromagnetic environment and improve the accuracy of prediction. At the same time, the newly collected data is added to the historical data set, and the data used for model training is continuously updated to reflect the dynamic changes in the electromagnetic environment.

[0090] Corresponding to the above-disclosed electromagnetic spectrum situation prediction method based on a time series model, the embodiment of the present invention further discloses an electromagnetic spectrum situation prediction system based on a time series model, referring to Figure 4 , which specifically include:

[0091] The data acquisition and processing module is used to acquire electromagnetic spectrum time series data and pre-process the acquired data;

[0092] A model building module is used to train a pre-built time series model and estimate model parameters based on the electromagnetic spectrum time series data obtained by preprocessing;

[0093] The prediction module is used to use the trained time series model to predict the electromagnetic spectrum situation, obtain prediction results, and evaluate the prediction results;

[0094] The feedback module is used to utilize the latest electromagnetic spectrum data collected regularly to regularly optimize the time series model and adjust the model parameters.

[0095] It should be noted that for the detailed description of an electromagnetic spectrum situation prediction system based on a time series model provided in an embodiment of the present invention, reference can be made to the relevant description of an electromagnetic spectrum situation prediction method based on a time series model provided in an embodiment of the present application, which will not be repeated here.

[0096] In addition, an embodiment of the present invention also provides an electronic device, comprising: a processor and a memory; the memory is used to store one or more program instructions; the processor is used to run one or more program instructions to execute the steps of an electromagnetic spectrum situation prediction method based on a time series model as described in any of the above items.

[0097] It should be noted that, for the detailed description of an electronic device provided in an embodiment of the present invention, reference can be made to the relevant description of an electromagnetic spectrum situation prediction method based on a time series model provided in an embodiment of the present application, which will not be repeated here.

[0098] In addition, an embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of an electromagnetic spectrum situation prediction method based on a time series model as described in any of the above items are implemented.

[0099] It should be noted that, for a detailed description of a computer-readable storage medium provided in an embodiment of the present invention, reference can be made to the relevant description of an electromagnetic spectrum situation prediction method based on a time series model provided in an embodiment of the present application, which will not be repeated here.

[0100] Those skilled in the art will appreciate that all or part of the functions of the various methods in the above-mentioned embodiments can be implemented by hardware or by computer programs. When all or part of the functions in the above-mentioned embodiments are implemented by computer programs, the program can be stored in a computer-readable storage medium, and the storage medium can include: read-only memory, random access memory, disk, optical disk, hard disk, etc., and the program is executed by a computer to implement the above-mentioned functions. For example, the program is stored in the memory of the device, and when the program in the memory is executed by the processor, all or part of the above-mentioned functions can be implemented. In addition, when all or part of the functions in the above-mentioned embodiments are implemented by computer programs, the program can also be stored in a storage medium such as a server, another computer, disk, optical disk, flash disk or mobile hard disk, and can be downloaded or copied and saved in the memory of the local device, or the system of the local device is updated, and when the program in the memory is executed by the processor, all or part of the functions in the above-mentioned embodiments can be implemented.

[0101] The above specific examples are used to illustrate the present invention, which is only used to help understand the present invention and is not intended to limit the present invention. For those skilled in the art, according to the concept of the present invention, some simple deductions, modifications or substitutions can be made.

Claims

1. A method for predicting electromagnetic spectrum situation based on a time series model, characterized in that: The method comprises: Collect electromagnetic spectrum time series data and pre-process the collected data; Based on the electromagnetic spectrum time series data obtained by preprocessing, the pre-built time series model is trained and the model parameters are estimated; The trained time series model is used to predict the electromagnetic spectrum situation, obtain the prediction results, and evaluate the prediction results.

2. The electromagnetic spectrum situation prediction method based on a time series model as claimed in claim 1, characterized in that: Collect electromagnetic spectrum time series data, including: Electromagnetic spectrum monitoring sites are deployed in typical electromagnetic environment areas. Each monitoring site is equipped with a high-performance spectrum analyzer. The spectrum analyzer collects electromagnetic spectrum time series data at preset time intervals. The collected data includes signal strength, frequency occupancy, and signal type information.

3. The electromagnetic spectrum situation prediction method based on a time series model as claimed in claim 1, characterized in that: Preprocess the collected data, including: Data cleaning: Clean the collected raw electromagnetic spectrum data to remove outliers; identify and correct data points that are significantly deviated from the normal range by setting reasonable data thresholds; Data normalization: Normalize the cleaned electromagnetic spectrum data.

4. The electromagnetic spectrum situation prediction method based on a time series model as claimed in claim 1, characterized in that: Based on the electromagnetic spectrum time series data obtained by preprocessing, the pre-built time series model is trained and the model parameters are estimated, including: Select appropriate time series models based on the characteristics of electromagnetic spectrum data, including: For electromagnetic spectrum data with obvious periodicity and trend, the seasonal autoregressive integrated moving average model is selected; for data showing short-term autocorrelation and moving average characteristics, the autoregressive moving average model is selected.

5. The electromagnetic spectrum situation prediction method based on a time series model as claimed in claim 4, characterized in that: Based on the electromagnetic spectrum time series data obtained by preprocessing, the pre-built time series model is trained and the model parameters are estimated, including: Using the preprocessed electromagnetic spectrum time series data, the parameters of the selected time series model are estimated by methods including maximum likelihood estimation.

6. The electromagnetic spectrum situation prediction method based on a time series model as claimed in claim 5, characterized in that: Based on the electromagnetic spectrum time series data obtained by preprocessing, the pre-built time series model is trained and the model parameters are estimated, including: For the seasonal autoregressive integrated moving average model SARIMA(p,d,q)(P,D,Q) s Model, determine the autoregressive order p, difference order d, moving average order q of the non-seasonal part and the corresponding parameters P, D, Q and seasonal period s of the seasonal part; According to the autocorrelation function ACF and partial autocorrelation function PACF diagrams of electromagnetic spectrum data, the optimal parameter combination is determined in combination with the information criterion to ensure that the model can accurately fit the inherent laws of electromagnetic spectrum data.

7. The electromagnetic spectrum situation prediction method based on a time series model as claimed in claim 1, characterized in that: The trained time series model is used to predict the electromagnetic spectrum situation and obtain the prediction results, including: Short-term prediction: Input the latest pre-processed electromagnetic spectrum data into the constructed time series model to perform short-term electromagnetic spectrum situation prediction; Long-term prediction: Through long-term trend analysis of historical data and models, long-term prediction of the electromagnetic spectrum situation can be achieved.

8. The electromagnetic spectrum situation prediction method based on a time series model as claimed in claim 1, characterized in that: Evaluate the prediction results, including: The mean square error (MSE) and mean absolute error (MAE) of the prediction results are calculated. If the MSE or MAE exceeds the preset threshold, the data preprocessing process and model parameters are rechecked.

9. The electromagnetic spectrum situation prediction method based on a time series model as claimed in claim 1, characterized in that: The method further comprises: Using the latest electromagnetic spectrum data collected regularly, the time series model is regularly optimized and trained, and the model parameters are adjusted.

10. An electromagnetic spectrum situation prediction system based on a time series model, characterized in that: The system comprises: The data acquisition and processing module is used to acquire electromagnetic spectrum time series data and pre-process the acquired data; A model building module is used to train a pre-built time series model and estimate model parameters based on the electromagnetic spectrum time series data obtained by preprocessing; The prediction module is used to use the trained time series model to predict the electromagnetic spectrum situation, obtain prediction results, and evaluate the prediction results.

Images