Search

CN-121997196-A - Ionosphere parameter and short-wave frequency band prediction method and electronic equipment

CN121997196ACN 121997196 ACN121997196 ACN 121997196ACN-121997196-A

Abstract

The invention provides an ionosphere parameter and shortwave frequency band prediction method and electronic equipment, comprising the following steps: acquiring the geographic coordinates of two places of a communication link and the midpoint of a great circle, acquiring ionosphere parameter data in an IRI model corresponding to the upper space of the midpoint of the great circle, acquiring geomagnetic sun activity data of an open source data website, performing ionosphere oblique detection by utilizing a communication two-ionosphere detector, acquiring an ionosphere oblique detection ionization diagram, acquiring short-wave communication available frequency band data from the oblique detection ionization diagram, performing data preprocessing, selecting different tasks, setting corresponding transducer-LSTM model super-parameters, performing model training, storing optimal model parameters, and performing an ionosphere parameter prediction task or a short-wave communication available frequency band prediction task by using a corresponding model. The method disclosed by the invention integrates ionosphere detection data, environment monitoring data and the like in the deep learning field and the short wave communication field, and has important significance and application prospect.

Inventors

  • YANG GUOBIN
  • Dai Chenqing

Assignees

  • 武汉大学

Dates

Publication Date
20260508
Application Date
20260408

Claims (10)

  1. 1. An ionospheric parameter and shortwave frequency band prediction method, comprising: Acquiring geographic coordinates of a transmitting station and a receiving station and ionosphere parameters corresponding to an international reference ionosphere model above a relay point, and acquiring global geomagnetic activity parameters and solar activity parameters; Using ionized layer detectors of a transmitting station and a receiving station to perform oblique sweep frequency detection to obtain an ionized layer oblique detection ionization diagram; Performing post-processing on the ionosphere oblique detection ionization diagram to obtain oblique detection frequency band data; preprocessing global geomagnetic activity parameters, solar activity parameters and oblique detection frequency band data, and selecting data with the same time resolution in continuous time; constructing a transducer-LSTM combined model, and performing model training by adopting specified preprocessed data to obtain an ionospheric parameter prediction model or a shortwave frequency band prediction model; and inputting the historical data of the appointed duration and variety into an ionospheric parameter prediction model or a shortwave frequency band prediction model, and outputting an ionospheric parameter prediction result or a shortwave frequency band prediction result.
  2. 2. The ionospheric parameters and shortwave band prediction method of claim 1, wherein obtaining geographic coordinates of a transmitting station and a receiving station comprises: The method comprises the steps that geographic coordinates are adopted to represent a transmitting station and a receiving station, and longitude and latitude coordinates of a midpoint of a large circle between the two stations are obtained through geographic coordinate calculation; and inputting the time range, the resolution, the longitude and latitude coordinates of the midpoint of the large circle, the ground height range and the resolution into the international reference ionosphere model.
  3. 3. The ionospheric parameters and shortwave band prediction method according to claim 2, wherein acquiring ionospheric parameters corresponding to an international reference ionospheric model of a space above a relay point of a transmitting station and a receiving station comprises: acquiring ionosphere parameter historical data by using an international reference ionosphere model; and calling an international reference ionosphere model, and inputting a time range, resolution, longitude and latitude coordinates of a large circle midpoint, a ground height range and resolution to obtain the critical frequency, peak height, half thickness of the ionosphere of the large circle midpoint of the link between two stations and the distribution condition of electron concentration in space.
  4. 4. The ionosphere parameter and shortwave band prediction method as set forth in claim 1, wherein the acquisition of the ionosphere oblique detection ionization map by oblique sweep detection using ionosphere detectors of a transmitting station and a receiving station comprises: the transmitting station transmits m-sequence sweep frequency modulation signals by using an inverted V antenna, sets the initial frequency, the termination frequency and the stepping frequency of sweep frequency according to different ionosphere states, receives oblique detection signals by using a two-wire dipole antenna in different ways, performs data processing on the received oblique detection signals, and draws to obtain an ionosphere oblique detection ionization diagram.
  5. 5. The ionosphere parameter and shortwave frequency band prediction method as set forth in claim 1, wherein the performing post-processing on the ionosphere oblique detection ionization map to obtain oblique detection frequency band data includes: adopting a projection method, focusing on the maximum signal amplitude value corresponding to each frequency point in the sweep frequency range; And extracting the initial frequency and the termination frequency corresponding to the signal frequency band by using a threshold method and adopting a preset threshold value as oblique detection frequency band data.
  6. 6. The ionosphere parameter and shortwave band prediction method according to claim 1, wherein preprocessing global geomagnetic activity parameter, solar activity parameter and oblique detection band data, selecting data with the same time resolution in continuous time, comprises: and carrying out missing data interpolation and abnormal data rejection by adopting the time range and the time resolution of the unified data to obtain preprocessed data.
  7. 7. The ionospheric parameters and shortwave band prediction method of claim 1, wherein constructing a transducer-LSTM combined model comprises: Determining that the transducer model only comprises an encoder part of the original transducer, and combining an LSTM layer in the LSTM model with a feedforward neural network of a decoder of the original transducer, residual connection and layer normalization to form a transducer-LSTM combined model; And determining to adopt key model parameters comprising the sequence length and the prediction length for model training, and storing optimal model parameters after model training is completed.
  8. 8. The ionospheric parameter and shortwave band prediction method of claim 7, wherein model training is performed using the specified preprocessed data to obtain the ionospheric parameter prediction model or shortwave band prediction model, comprising: Model training is carried out by adopting global geomagnetic activity parameters and solar activity parameters to obtain an ionosphere parameter prediction model; and performing model training by adopting global geomagnetic activity parameters, solar activity parameters and ionosphere oblique detection ionization diagrams to obtain a short-wave frequency band prediction model.
  9. 9. The ionospheric parameter and shortwave band prediction method according to claim 1, wherein inputting the historical data of the specified duration and kind into the ionospheric parameter prediction model or shortwave band prediction model, outputting the ionospheric parameter prediction result or shortwave band prediction result, comprises: Loading optimal parameters of the trained model, and determining historical data of the type corresponding to ionosphere parameter prediction and/or short-wave frequency band prediction; After the historical data of the sequence length is input, outputting predicted data of a predicted length after the end time of the sequence length by a model, and outputting ionosphere parameter predicted results or short-wave frequency band predicted results when the starting time of a time sequence corresponding to the sequence length and the ending time of a time sequence corresponding to the predicted length are in a test set time range corresponding to model training.
  10. 10. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the ionospheric parameters and shortwave band prediction method of any one of claims 1 to 9 when the program is executed by the processor.

Description

Ionosphere parameter and short-wave frequency band prediction method and electronic equipment Technical Field The present invention relates to the field of ionosphere detection technologies, and in particular, to an ionosphere parameter and shortwave frequency band prediction method and an electronic device. Background Short wave communication is an important mode of radio communication, electric waves are transmitted and reach receiving equipment through ionosphere reflection, the communication distance is long, and the short wave communication is the only remote communication means which is not limited by an active relay. The short wave communication can cover the global scope through the repeated reflection of the ionized layer, including special areas such as mountain areas, gobi, ocean and the like which cannot be covered by the ultrashort wave. Short wave communication does not need satellite, and construction and maintenance cost is low, has extremely strong flexibility, survivability and stability. Ionosphere is a highly dependent channel of short-wave communication whose dispersive time-varying properties can have an impact on the effect of short-wave communication. Because the state of the ionized layer is greatly influenced by solar activity and seasonal variation, abnormal states frequently occur, and in the practice of short-wave real-time communication, the quality and stability of the short-wave communication must be ensured by in-situ switching of different communication frequencies through the state of the ionized layer. Therefore, frequency selection in combination with ionosphere parameters is important for short wave communication. Time series prediction is a method for predicting future values by analyzing a history pattern of time series data, and has been widely used in many fields. The prediction of ionospheric parameters can be broadly divided into long-term prediction and short-term prediction, and international reference ionospheric IRI models in long-term prediction models are accepted by international research, and the models have the capability of predicting a plurality of parameters of the ionosphere in the global scope. The estimated value given by the IRI model is more accurate in geomagnetic calm or ionosphere stable change period, but the disadvantage is that due to the ionosphere change characteristic, larger errors can occur only when ionosphere disturbance caused by factors such as solar activity, magnetic storm and the like occurs. The autonomous frequency selection technology used in short wave communication at present integrates the spectrum sensing technology and the historical communication data, and has excellent performance. Unfortunately, this technique does not take into account ionospheric factors that have an important impact on short wave transmissions, but rather merely performs real-time frequency selection based on noise and historical empirical data. Because the influence of ionosphere parameters is ignored, the single historical communication data source is subjected to frequency selection, and the stability is poor when the ionosphere is abnormal. Disclosure of Invention The invention provides an ionosphere parameter and short-wave frequency band prediction method and electronic equipment, which are used for solving the defects in the prior art, acquiring parameter data related to global geomagnetic activity and solar activity by acquiring ionosphere parameter data corresponding to an IRI model above a relay point of two short-wave stations, acquiring historical detection data of the two short-wave stations for oblique detection by using an ionosphere detector, performing data preprocessing, training a deep learning model Transformer-LSTM by using different super parameters and processed data according to different task types such as an ionosphere parameter prediction task or a short-wave communication available frequency band prediction task, inputting historical ionosphere parameter data, historical geomagnetic solar activity index data and historical detection data with specified duration, and obtaining a prediction result of the corresponding ionosphere parameter or short-wave communication available frequency band in the specified duration in the future by combining trained model parameters. In a first aspect, the present invention provides an ionospheric parameter and shortwave frequency band prediction method, including: Acquiring geographic coordinates of a transmitting station and a receiving station and ionosphere parameters corresponding to an international reference ionosphere model above a relay point, and acquiring global geomagnetic activity parameters and solar activity parameters; Using ionized layer detectors of a transmitting station and a receiving station to perform oblique sweep frequency detection to obtain an ionized layer oblique detection ionization diagram; Performing post-processing on the ionosphere oblique detection ioniza