Low-altitude helicopter communication navigation signal enhancement method and system based on intelligent scheduling

Abstract

The invention discloses a low-altitude helicopter communication navigation signal enhancement method and system based on intelligent scheduling, and the method comprises the steps: obtaining time-frequency characteristic data collected by a sensor array, extracting interference source parameters through employing a fast Fourier transform algorithm, and generating an interference source distribution diagram; if a dynamic prediction result of the deep learning model on the interference source distribution map exceeds a preset threshold value, generating a spatio-temporal variation trend map; calculating a signal path blocking point by adopting a digital elevation model according to the spatio-temporal variation trend chart, and determining a shielding influence area range; extracting electromagnetic interference intensity data from the shielding influence area range, and decomposing the electromagnetic interference intensity data through a spectrum analysis algorithm to obtain an interference frequency principal component; if the electromagnetic interference intensity corresponding to the principal component is higher than a preset threshold value, adjusting parameters of an adaptive filter to generate an optimized filtering signal; and performing space-time coding processing on the optimized filtering signal, and eliminating weather influence factors by adopting a two-dimensional convolution algorithm to obtain an anti-interference enhanced signal.

Description

Technical Field

[0001] The present invention relates to the field of information technology, and in particular, to a method and system for enhancing communication and navigation signals of low-altitude helicopters based on intelligent scheduling. Background Art

[0002] The technology for enhancing communication and navigation signals of low-altitude helicopters is an indispensable research direction in the modern aviation field. Its core lies in ensuring the safety and efficient operation of helicopters in complex low-altitude environments. With the gradual opening of low-altitude airspace and the expansion of helicopter application scenarios, the stability and reliability of communication and navigation systems have become key factors determining flight safety. Especially in diverse terrains such as cities and mountains, signal transmission faces severe challenges, and researching targeted enhancement technologies has important strategic significance.

[0003] Currently, traditional signal enhancement methods have obvious deficiencies in low-altitude environments. Many solutions rely on static spectrum allocation or simple filtering techniques, and it is difficult to cope with dynamically changing interference sources. Especially in scenarios with frequent terrain occlusion and dense electromagnetic interference, these methods often cannot be adjusted in real time, resulting in a decline in signal quality or even signal interruption. In addition, the adaptability of existing technologies in adverse weather conditions is poor, which limits the operating ability of helicopters under all-weather conditions.

[0004] Focusing on specific challenges, the diversity and dynamics of interference sources in low-altitude environments are the primary problems. Terrain occlusion causes signal paths to be blocked, electromagnetic interference is exacerbated due to spectrum congestion, and weather factors such as rain and fog further weaken signal strength. These three core technical factors - the real-time detection of interference sources, the effective filtering of interference signals, and the anti-interference ability of signal transmission - directly determine the system performance. Due to these factors not being fully resolved, signal distortion or transmission failure is likely to occur in the helicopter communication and navigation system in complex environments, thereby triggering a unique technical problem, that is, how to accurately respond to variable interference under limited resources.

[0005] Therefore, how to use intelligent means to real-time sense interference sources in low-altitude environments and effectively improve the anti-interference ability of signals by combining adaptive filtering and space-time coding technologies has become a key issue in ensuring the stable operation of helicopter communication and navigation systems. The solution to this problem will directly promote the development of low-altitude flight technology and lay a foundation for broader applications. Summary of the Invention

[0006] The technical problem to be solved by the present invention is to provide a method and system for enhancing communication and navigation signals of low-altitude helicopters based on intelligent scheduling.

[0007] To achieve the above object, the present invention adopts the following technical solutions:

[0008] A method for enhancing communication and navigation signals of low-altitude helicopters based on intelligent scheduling, comprising:

[0009] Obtain the time-frequency feature data collected by the sensor array, extract the interference source parameters using the fast Fourier transform algorithm, and generate an interference source distribution map;

[0010] If the dynamic prediction result of the deep learning model for the interference source distribution map exceeds the preset threshold, generate a spatio-temporal change trend map;

[0011] According to the spatio-temporal change trend map, calculate the signal path blocking points using the digital elevation model, and determine the range of the occlusion influence area;

[0012] Extract the electromagnetic interference intensity data from the occlusion influence area range, and decompose it through the spectrum analysis algorithm to obtain the main component components of the interference frequency;

[0013] If the electromagnetic interference intensity corresponding to the main component component is higher than the preset threshold, adjust the parameters of the adaptive filter to generate an optimized filtering signal;

[0014] Perform space-time coding processing on the optimized filtering signal, and use the two-dimensional convolution algorithm to eliminate the weather influence factors to obtain an anti-interference enhanced signal;

[0015] Extract the transmission signal from the adjusted coded signal, optimize the bandwidth and power parameters using the dynamic resource allocation algorithm, and output a stable communication signal.

[0016] Preferably, the obtaining the time-frequency feature data collected by the sensor array, extracting the interference source parameters using the fast Fourier transform algorithm, and generating an interference source distribution map includes:

[0017] Obtain the original signal data collected by the sensor array, filter out the noise through preprocessing to obtain clean time-frequency feature data;

[0018] Process the clean time-frequency feature data through the fast Fourier transform algorithm, extract the interference source parameters, and obtain a parameter set;

[0019] Use the clustering algorithm to classify the parameter set to determine the interference source type distribution;

[0020] If the interference source type distribution exceeds the preset threshold, calculate the interference source location through the weighted average method to obtain a set of position coordinates;

[0021] Generate two-dimensional grid data according to the set of position coordinates to obtain a basic interference distribution map;

[0022] Smooth the two-dimensional grid data through the interpolation algorithm to obtain an optimized interference source distribution map;

[0023] For the optimized interference source distribution map, color mapping technology is adopted to generate a visual interference distribution map.

[0024] Preferably, if the dynamic prediction result of the deep learning model for the interference source distribution map exceeds a preset threshold, a spatio-temporal change trend map is generated, including:

[0025] If the prediction result generated by deep learning exceeds the preset threshold, time series data in the dynamic prediction is extracted through time series analysis to obtain change trend features;

[0026] According to the change trend features, the time series data is processed using a convolutional smoothing method to obtain a smoothed trend sequence;

[0027] Through the smoothed trend sequence, the distribution offset at each time point is calculated to determine the spatio-temporal distribution features;

[0028] For the spatio-temporal distribution features, grid division technology is adopted to generate two-dimensional spatio-temporal grid data to obtain a preliminary trend distribution;

[0029] According to the preliminary trend distribution, color coding technology is applied to process the grid data to generate a visual trend map;

[0030] If the distribution offset in the visual trend map exceeds the preset range, the grid data is adjusted through a clustering algorithm to obtain an optimized spatio-temporal change trend map;

[0031] Through the optimized spatio-temporal change trend map, key change regions are extracted to determine the final trend distribution.

[0032] Preferably, according to the spatio-temporal change trend map, a digital elevation model is used to calculate signal path blocking points to determine the range of the occlusion influence area, including:

[0033] Elevation data is obtained through the digital elevation model, and the distribution of the signal path in the spatio-temporal change trend map is calculated to obtain path change features;

[0034] The height differences on the signal path are analyzed using the path change features to judge the positions of the blocking points and determine the set of potential occlusion points;

[0035] The occlusion influence data in the set of blocking points is obtained, and the boundary of the occlusion area is calculated through the elevation data to obtain the range of the occlusion area;

[0036] Terrain features within the range of the occlusion area are extracted to judge whether the signal path is completely blocked to determine the degree of blockage;

[0037] The distribution law of the occlusion influence is analyzed through the degree of blockage to obtain the precise boundary of the influence range;

[0038] Use the random forest algorithm to classify the occluded areas within the affected range, judge the occlusion severity of different areas, and determine the final range of the occlusion affected area;

[0039] According to the final range of the occlusion affected area, adjust the optimized layout of the signal path through elevation data to obtain the adjusted path distribution.

[0040] Preferably, extracting the electromagnetic interference intensity data from the range of the occlusion affected area and decomposing it through the spectrum analysis algorithm to obtain the main component components of the interference frequency, including:

[0041] Determine the range of the area affected by occlusion, and use a measuring tool to obtain the intensity data corresponding to the electromagnetic interference;

[0042] Extract the interference frequency characteristics from the intensity data, and use the spectrum analysis algorithm to decompose to obtain the initial component results;

[0043] Perform frequency analysis on the initial component results to judge the distribution characteristics of the main component components;

[0044] If the main component component exceeds the preset threshold, obtain the independent characteristics of the interference frequency through component decomposition;

[0045] Determine the source direction of the electromagnetic interference according to the independent characteristics, and use the positioning algorithm to obtain the coordinates of the interference source;

[0046] Compare the coordinates of the interference source with the area range to judge the boundary range of the occlusion effect;

[0047] Obtain the intensity data within the boundary range, and use a filtering tool to obtain the optimized interference frequency components.

[0048] Preferably, if the electromagnetic interference intensity corresponding to the main component component is higher than the preset threshold, adjust the parameters of the adaptive filter to generate an optimized filtering signal, including:

[0049] If the main component component of the electromagnetic interference exceeds the preset threshold, determine the component analysis result by analyzing the interference intensity;

[0050] Through the component analysis result, obtain the characteristic data of the interference intensity, and judge whether the filter needs to be adjusted;

[0051] If the characteristic data exceeds the threshold judgment standard, adjust the parameters of the adaptive filter to obtain a preliminary filtering signal;

[0052] According to the preliminary filtering signal, use the signal generation technology to generate an optimized signal;

[0053] Extract the residual components of the electromagnetic interference from the optimized signal, and judge whether the residual components are lower than the preset threshold;

[0054] If the residual component is higher than a preset threshold, the filter parameters are iteratively adjusted to obtain a final optimized signal;

[0055] According to the final optimized signal, the suppression degree of electromagnetic interference is determined to generate a stable output signal.

[0056] Preferably, the space-time coding process is performed on the optimized filtering signal, and a two-dimensional convolution algorithm is used to eliminate weather influence factors to obtain an anti-interference enhanced signal, including:

[0057] Perform optimized filtering on the input signal, and use preset filter parameters to obtain a filtered signal;

[0058] Perform space-time coding on the filtered signal, and generate a coded signal through coding technology;

[0059] Perform convolution operation on the coded signal using a two-dimensional convolution algorithm to obtain a convolution result;

[0060] Remove weather influence factors through the convolution result to obtain a preliminary enhanced signal;

[0061] If there is residual interference in the preliminary enhanced signal, an anti-interference detection method is used to judge the interference degree to obtain a detection result;

[0062] Adjust the preliminary enhanced signal according to the detection result, and use weighted processing to obtain a final enhanced signal;

[0063] Verify the effectiveness of the processing flow through the final enhanced signal to determine the anti-interference enhanced signal.

[0064] Preferably, the transmission signal is extracted from the adjusted coded signal, and the dynamic resource allocation algorithm is used to optimize the bandwidth and power parameters to output a stable communication signal, including:

[0065] Obtain the transmission signal from the coded signal through the decoder to get the initial transmission data;

[0066] Use a preset threshold to judge the integrity of the initial transmission data to obtain a valid transmission signal;

[0067] Adjust the bandwidth parameter of the valid transmission signal through the dynamic resource allocation algorithm to obtain an optimized bandwidth signal;

[0068] Adjust the power parameter of the optimized bandwidth signal through the dynamic resource allocation algorithm to obtain an optimized power signal;

[0069] According to the characteristics of the optimized power signal, use signal processing technology to smooth the noise to obtain a smooth communication signal;

[0070] Detect the stability of the smooth communication signal through a channel simulation tool to obtain a stable communication signal.

[0071] The present invention also provides a communication navigation signal enhancement system for low-altitude helicopters based on intelligent scheduling, including:

[0072] A time-frequency feature extraction module, configured to obtain time-frequency feature data collected by a sensor array;

[0073] An interference source parameter extraction module, configured to extract interference source parameters by using a fast Fourier transform algorithm;

[0074] An interference source distribution map generation module, configured to generate an interference source distribution map;

[0075] A spatio-temporal change trend map generation module, configured to generate a spatio-temporal change trend map if the dynamic prediction result of the deep learning model for the interference source distribution map exceeds a preset threshold;

[0076] A signal path blocking point calculation module, configured to calculate signal path blocking points by using a digital elevation model according to the spatio-temporal change trend map;

[0077] An electromagnetic interference intensity extraction module, configured to determine the range of the occlusion influence area and extract electromagnetic interference intensity data from the range of the occlusion influence area;

[0078] An optimized filtering signal generation module, configured to decompose to obtain the main component components of interference frequencies through a spectrum analysis algorithm, and adjust the parameters of an adaptive filter to generate an optimized filtering signal if the electromagnetic interference intensity corresponding to the main component components is higher than a preset threshold;

[0079] An anti-interference enhanced signal generation module, configured to perform space-time coding processing on the optimized filtering signal and use a two-dimensional convolution algorithm to eliminate weather influence factors to obtain an anti-interference enhanced signal;

[0080] A stable communication signal output module, configured to extract a transmission signal from the adjusted coded signal, optimize bandwidth and power parameters by using a dynamic resource allocation algorithm, and output a stable communication signal.

[0081] The present invention first collects time-frequency feature data through a sensor array, extracts interference source parameters by using a fast Fourier transform and generates a distribution map. When the prediction result exceeds the threshold, a spatio-temporal change trend map is generated, and the occlusion influence area is determined in combination with a digital elevation model. Subsequently, electromagnetic interference intensity data is extracted from this area, and the main component components are obtained through spectrum analysis. If the interference intensity is too high, the parameters of the adaptive filter are adjusted, and space-time coding processing is performed to eliminate the weather influence. Finally, the present invention extracts a transmission signal from the optimized coded signal, optimizes bandwidth and power parameters by using a dynamic resource allocation algorithm, and outputs a stable anti-interference communication signal. This method can effectively cope with a complex electromagnetic environment and improve the anti-interference ability and signal quality of the communication system. BRIEF DESCRIPTION OF THE DRAWINGS

[0082] Figure 1 This is a flowchart of a method for enhancing communication and navigation signals of a low-altitude helicopter based on intelligent scheduling according to the present invention. Specific embodiments

[0083] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0084] As Figure 1 shown, an embodiment of the present invention provides a method for enhancing communication and navigation signals of a low-altitude helicopter based on intelligent scheduling, including:

[0085] S101. Obtain the time-frequency characteristic data collected by the sensor array, extract the interference source parameters by using the fast Fourier transform algorithm, and generate an interference source distribution map.

[0086] Obtain the original signal data collected by the sensor array, filter out the noise through preprocessing to obtain clean time-frequency characteristic data. Process the clean time-frequency characteristic data by using the fast Fourier transform algorithm to extract the interference source parameters and obtain a parameter set. Use a clustering algorithm to classify the parameter set to determine the interference source type distribution. If the interference source type distribution exceeds a preset threshold, calculate the interference source position by using the weighted average method to obtain a position coordinate set. Generate two-dimensional grid data according to the position coordinate set to obtain a basic interference distribution map. Smooth the two-dimensional grid data by using an interpolation algorithm to obtain an optimized interference source distribution map. For the optimized interference source distribution map, use a color mapping technique to generate a visual interference distribution map.

[0087] Specifically, when obtaining the original signal data collected by the sensor array, it can be understood that an array composed of multiple sensors is used to monitor the electromagnetic signals in the environment in real time.

[0088] For example, near a wireless communication base station, the sensor array may capture radio frequency signals mixed with noise, and the data is presented as a waveform of voltage varying with time, with an amplitude range between -5V and 5V. When preprocessing to filter out the noise.

[0089] In a possible implementation manner, a low-pass filter can be used to remove high-frequency noise with a frequency higher than 1000Hz and retain the main signal components.

[0090] Exemplarily, after filtering, the jitter of the signal waveform is significantly reduced, the peak value is more clearly visible, effectively improving the accuracy of subsequent analysis. When processing the clean time-frequency characteristic data by using the fast Fourier transform algorithm.

[0091] Preferably, convert the time-domain signal to the frequency domain and extract the spectral characteristics.

[0092] For example, assume that the signal sampling rate is 2000 Hz. After transformation, it can be observed that the main frequency components are concentrated at 200 Hz and 500 Hz, corresponding to the characteristic frequencies of two potential interference sources respectively. When extracting the parameters of the interference sources.

[0093] Specifically, the peak frequency, amplitude, and bandwidth can be identified from the spectrum to form a parameter set, such as {200 Hz, 0.8 V, 10 Hz} and {500 Hz, 1.2 V, 15 Hz}. This provides a reliable basis for subsequent classification. When using a clustering algorithm to classify the parameter set.

[0094] In one embodiment, the K-means algorithm can be used to divide the parameters into two categories, corresponding to low-frequency and high-frequency interference sources respectively.

[0095] It should be noted that if the number of sample points is 100, it may be found after clustering that 80% belongs to the low-frequency interference near 200 Hz and 20% belongs to the high-frequency interference near 500 Hz, reflecting the distribution of the types of interference sources. If the preset threshold is 70%, then the low-frequency interference exceeds the threshold, triggering the calculation of the position. When calculating the position of the interference source.

[0096] For example, through the weighted average method, combined with the spatial layout of the sensor array, assuming that three sensors are located at (0, 0), (5, 0), and (0, 5) respectively, according to the signal intensity weighting, the position coordinates of the low-frequency interference source can be obtained to be approximately (2, 1). When generating two-dimensional grid data based on the position coordinate set, it can be understood that the coordinates are mapped into a 10×10 grid to form a basic map of the interference distribution. When performing smoothing processing.

[0097] In one possible implementation, a linear interpolation algorithm is used to make the values in the blank areas of the grid smoothly transition based on neighboring points.

[0098] For example, the value near (2, 1) smoothly changes from 1.0 to 0.8 at (3, 2). The optimized interference source distribution map is more continuous. When using color mapping technology for the optimized distribution map.

[0099] Specifically, the areas with an intensity greater than 1.0 can be represented in red, and the areas between 0.5 and 1.0 can be represented in green to generate a visualization graph. This visualization effect intuitively reflects the interference intensity distribution and helps to quickly locate the problem area.

[0100] For example, the distribution map generated by the above method can significantly improve the efficiency of interference source detection. Especially in a dense signal environment, it can clearly distinguish multi-source interference and reduce misjudgment.

[0101] Preferably, this technology can also provide data support for subsequent optimization of the sensor layout and enhance the robustness of the system.

[0102] It can be understood that the implementation of each step is closely connected, jointly constructing a complete link from signal acquisition to visualization, with strict logic and strong practicality.

[0103] S102. If the dynamic prediction result of the deep learning model for the interference source distribution map exceeds a preset threshold, then generate a spatio-temporal change trend map.

[0104] If the prediction result generated by deep learning exceeds the preset threshold, then extract the time series data in the dynamic prediction through time series analysis to obtain the change trend characteristics. According to the change trend characteristics, use the convolution smoothing method to process the time series data to obtain the smoothed trend sequence. Through the smoothed trend sequence, calculate the distribution offset at each time point to determine the spatio-temporal distribution characteristics. For the spatio-temporal distribution characteristics, use the grid division technology to generate two-dimensional spatio-temporal grid data to obtain the preliminary trend distribution. According to the preliminary trend distribution, apply the color coding technology to process the grid data to generate a visualized trend map. If the distribution offset in the visualized trend map exceeds the preset range, then adjust the grid data through the clustering algorithm to obtain the optimized spatio-temporal change trend map. Through the optimized spatio-temporal change trend map, extract the key change regions to determine the final trend distribution.

[0105] Specifically, if the prediction result generated by deep learning exceeds the preset threshold, the data characteristics can be further explored through time series analysis.

[0106] For example, in the scenario of wireless communication signal monitoring, assume that the deep learning model predicts that the interference intensity in a certain area is 0.9, exceeding the threshold of 0.7. During time series analysis.

[0107] Specifically, the time series can be extracted from the monitoring data of 24 consecutive hours, and the data is represented by the interference intensity value per minute, such as 0.85, 0.89, 0.92, etc., showing a gradually increasing trend.

[0108] Exemplarily, this increase may be related to the increase in base station load, reflecting the dynamic change characteristics. When using the convolution smoothing method to process the time series.

[0109] In a possible implementation, a convolution kernel with a length of 5 can be used, and the sliding window covers the data of adjacent 5 time points.

[0110] For example, the original sequence 0.85, 0.89, 0.92, 0.90, 0.87 may become 0.87, 0.89, 0.90, 0.89, 0.88 after smoothing, with reduced fluctuations and a smoother trend.

[0111] It can be understood that this helps to eliminate short-term noise interference and highlight the long-term change rules. When calculating the distribution offset through the smoothed trend sequence.

[0112] Preferably, the deviation of the value at each time point from the average value can be compared.

[0113] For example, the average value is 0.89, the value at a certain point is 0.92, and the offset is 0.03. If the offsets at consecutive multiple points increase, such as 0.03, 0.05, 0.07, it indicates that the interference distribution shifts in space and time.

[0114] It should be noted that this kind of shift may be related to the movement of the signal source or environmental changes, providing a basis for subsequent analysis. When generating two-dimensional grid data using the grid division technique for the spatio-temporal distribution characteristics.

[0115] Specifically, the monitoring area can be divided into a 10×10 grid, and each grid corresponds to the average value of the offset within one hour.

[0116] For example, the value of a certain grid is 0.05, and the adjacent grid is 0.03, forming a preliminary trend distribution.

[0117] In one embodiment, the grid division can also combine the sensor positions to further refine the spatial accuracy. When processing the grid data using the color coding technique.

[0118] For example, it can be set that the offset greater than 0.06 is red, 0.03 to 0.06 is yellow, and lower than 0.03 is green. In the generated trend chart, the red area is concentrated in the center of the grid, indicating an increase in interference.

[0119] It can be understood that this kind of visualization facilitates the quick identification of abnormal areas. If the distribution offset exceeds the preset range, such as being greater than 0.08, the grid data is adjusted through the clustering algorithm.

[0120] Exemplarily, the K-means algorithm is used to divide the grid into two categories: areas with high and low offsets. After adjustment, the boundary of the red area is clearer, and the optimized trend chart reflects that the interference is concentrated in a certain specific direction.

[0121] In one embodiment, this kind of optimization can improve the positioning accuracy. When extracting the key change areas from the optimized trend chart.

[0122] Specifically, the coordinates corresponding to the red area can be locked, such as (3,4) to (5,6).

[0123] For example, from multiple aspects of analysis, the offset at the coordinate (4,5) is 0.09 for three consecutive hours, and the adjacent areas also show high values, indicating that this is the main change source.

[0124] Preferably, this provides a clear direction for tracking the interference source. When determining the final trend distribution.

[0125] In one possible implementation, the time and space characteristics can be combined to generate a panoramic view including dynamic changes.

[0126] For example, the trend chart shows that the interference spreads from (4, 5) to (6, 7) at a speed of about 1 grid per hour.

[0127] It should be noted that this distribution not only reveals the change path, but also provides support for predicting future trends, enhancing the practicality of the system.

[0128] S103. According to the spatio-temporal change trend chart, use a digital elevation model to calculate the signal path blocking points and determine the range of the occlusion influence area.

[0129] Obtain elevation data through the digital elevation model, calculate the distribution of the signal path in the spatio-temporal change trend chart, and obtain the path change characteristics. Analyze the height differences on the signal path using the path change characteristics, judge the positions of the blocking points, and determine the set of potential occlusion points. Obtain the occlusion influence data in the set of blocking points, calculate the boundary of the occlusion area through the elevation data, and obtain the range of the occlusion area. Extract the terrain characteristics within the range of the occlusion area, judge whether the signal path is completely blocked, and determine the degree of blockage. Analyze the distribution law of the occlusion influence through the degree of blockage to obtain the precise boundary of the influence range. Use the random forest algorithm to classify the occlusion areas within the influence range, judge the severity of occlusion in different areas, and determine the final range of the occlusion influence area. According to the final range of the occlusion influence area, adjust the optimal layout of the signal path through the elevation data to obtain the adjusted path distribution.

[0130] Specifically, when obtaining elevation data through the digital elevation model.

[0131] It can be understood that this method relies on the elevation information of the terrain to construct the basic data of the signal path.

[0132] For example, in a mountainous area scene, assuming that the elevation data range of a certain area is between 500 meters and 1200 meters above sea level, after extraction through the elevation model, the elevation value of each grid point can be obtained for subsequent path analysis. When calculating the distribution of the signal path in the spatio-temporal change trend chart.

[0133] Exemplarily, the propagation trajectory of the signal from the emission point to the reception point can be drawn based on the elevation data, and the change characteristics of the path with the terrain undulation can be judged by combining the time dimension.

[0134] Specifically, if the signal path passes through a peak with an altitude of 800 meters, the path change characteristics may be manifested as a sudden rise and then fall in height, and this characteristic directly affects the subsequent analysis. When analyzing the height differences on the signal path using the path change characteristics.

[0135] In a possible implementation, significant changes can be identified by comparing the elevation values of each point on the path.

[0136] For example, the elevation at the starting point of the path is 600 meters. After a certain section, it rises to 900 meters and then drops to 700 meters. Such height differences may indicate the existence of blocking points. When determining the positions of the blocking points.

[0137] Preferably, a height difference threshold can be set, such as 50 meters. If it exceeds this value, it is marked as a potential occlusion point, and then an occlusion point set is formed.

[0138] It should be noted that the determination of blocking points not only depends on height, but also factors such as terrain slope need to be considered, which can more accurately reflect the actual situation. When obtaining the occlusion influence data in the occlusion point set.

[0139] For example, the terrain shadow range near the occlusion point can be analyzed through elevation data. Assuming the elevation of an occlusion point is 1000 meters, its influence range may cover the area within 200 meters below. When calculating the boundary of the occlusion area.

[0140] In one embodiment, the gradient change of elevation data can be utilized to determine the shadow boundary. For example, the area with a slope less than 10 degrees is used as the boundary line. This method helps to accurately divide the occlusion range and is convenient for subsequent terrain feature extraction. When extracting terrain features for the occlusion area range.

[0141] It can be understood that terrain features may include information such as slope direction and slope.

[0142] For example, the slope of an occlusion area is 30 degrees and the slope direction is north, which may cause the signal to be completely blocked. When analyzing the distribution law of the occlusion influence through the blocking degree.

[0143] Specifically, the blocking degree can be divided into three categories: slight, medium, and severe. If 80% of the points on the path are occluded, it is determined as a severe block. This classification helps to quantify the precise boundary of the influence range. When classifying the occlusion area using the random forest algorithm.

[0144] In one embodiment, feature data such as slope and elevation difference can be input, and the occlusion severity of different regions can be output.

[0145] For example, a certain area with an elevation difference of 100 meters and a steep slope is classified as a highly occluded area. This classification method can effectively distinguish the strength of the occlusion influence. After determining the final occlusion influence area range, when adjusting the optimal layout of the signal path through elevation data.

[0146] Exemplarily, the path can be adjusted from the occluded area at an elevation of 900 meters to the open area at 700 meters to avoid the main occlusion points. This adjustment can significantly improve the stability of signal transmission.

[0147] In one possible implementation, the effect of the adjusted path distribution can be further verified.

[0148] For example, the elevation change of the new path is controlled within 50 meters, and the number of occlusion points is reduced to 20% of the original, thus ensuring a wider signal coverage range. The realization of this optimized layout not only improves the rationality of path planning but also provides more favorable terrain conditions for signal transmission.

[0149] S104. Extract the electromagnetic interference intensity data from the range of the occlusion influence area, and decompose it through the spectrum analysis algorithm to obtain the main component of the interference frequency.

[0150] Determine the area range through occlusion influence, and use a measuring tool to obtain the intensity data corresponding to the electromagnetic interference. Extract the interference frequency characteristics from the intensity data, and decompose it through the spectrum analysis algorithm to obtain the initial component result. Conduct frequency analysis on the initial component result to judge the distribution characteristics of the main component. If the main component exceeds the preset threshold, obtain the independent characteristics of the interference frequency through component decomposition. Determine the source direction of the electromagnetic interference according to the independent characteristics, and use the positioning algorithm to obtain the coordinates of the interference source point. Compare the coordinates of the interference source point with the area range to judge the boundary range of the occlusion influence. Obtain the intensity data within the boundary range, and use a filtering tool to obtain the optimized interference frequency component.

[0151] Specifically, after determining the area range through occlusion influence, using a measuring tool to obtain the intensity data corresponding to the electromagnetic interference is a key step.

[0152] For example, in a mountain communication scenario, an electromagnetic wave intensity meter can be used to collect signal intensities at different locations to obtain a set of data reflecting the interference distribution.

[0153] Exemplarily, assume that the measured intensity values in a certain area are from -70 dBm to -50 dBm, which indicates that there may be an influence from an interference source. When extracting the interference frequency characteristics from the intensity data.

[0154] It can be understood that interference often appears as abnormal fluctuations at specific frequencies.

[0155] In a possible implementation, the fast Fourier transform is used to convert the time-domain signal into the frequency domain and decompose the frequency components. For example, if an obvious peak appears in the 2.4 GHz frequency band, it indicates that there may be Wi-Fi interference. Conducting frequency analysis on the initial component result to judge the distribution characteristics of the main component is the basis for subsequent analysis.

[0156] Specifically, it can be observed from the spectrogram whether the main interference frequency is concentrated in a certain range. Assume that the set threshold is -60 dBm. If the intensity of a certain frequency component reaches -55 dBm, it exceeds the threshold and needs further decomposition.

[0157] Preferably, independent features are extracted by principal component analysis, such as separating two independent interference sources of 2.4 GHz and 5 GHz. This method can clearly distinguish the contributions of multi-source interference. When determining the source direction of electromagnetic interference based on independent features, the positioning algorithm plays an important role.

[0158] In one embodiment, the triangulation method can be adopted to calculate the coordinates of the interference source by measuring the intensity difference at three points in the area.

[0159] For example, measuring points A, B, and C record the intensity and angle respectively, and finally it is estimated that the interference source is located about 500 meters northeast of the area. This not only clarifies the interference position but also facilitates subsequent investigation. By comparing the coordinates of the interference source with the range of the occlusion area, the boundary range can be judged.

[0160] For example, if the interference source is close to the ridge, the occlusion boundary may extend along the mountain body, and the influence range is adjusted accordingly. After obtaining the intensity data within the boundary range, it is crucial to use a filtering tool to optimize the interference frequency components.

[0161] It should be noted that the low-pass filter can remove high-frequency noise and retain the main interference characteristics.

[0162] For example, it is found that the 2.4 GHz component is still prominent after filtering, proving that this is the core interference frequency. This processing can improve the data quality and provide a reliable basis for subsequent analysis.

[0163] In one embodiment, the optimized frequency components can be used to draw an interference heat map to visually display the interference distribution in the area and help technicians quickly locate the problem area.

[0164] Exemplarily, when examining the positioning algorithm from multiple aspects, the accuracy of the coordinates can be verified by combining a directional antenna. If the signal is the strongest when the antenna points northeast, which is consistent with the triangulation result, the positioning credibility is higher.

[0165] Preferably, a time dimension analysis can also be introduced to observe the change of interference intensity over time to ensure that the result is not affected by instantaneous fluctuations.

[0166] For example, the data measured in the morning and evening are consistent, indicating that the interference source is stable. This multi-angle verification enhances the rigor of the scheme and at the same time provides a precise basis for interference mitigation, which helps to improve communication stability.

[0167] S105. If the electromagnetic interference intensity corresponding to the principal component is higher than the preset threshold, adjust the parameters of the adaptive filter to generate an optimized filtering signal.

[0168] If the main component of electromagnetic interference exceeds a preset threshold, the interference intensity is analyzed to determine the component analysis result. Based on the component analysis result, characteristic data of the interference intensity is obtained to judge whether the filter needs to be adjusted. If the characteristic data exceeds the threshold judgment criterion, the parameters of the adaptive filter are adjusted to obtain a preliminary filtered signal. According to the preliminary filtered signal, signal generation technology is used to generate an optimized signal. The residual component of electromagnetic interference is extracted from the optimized signal to judge whether the residual component is lower than the preset threshold. If the residual component is higher than the preset threshold, the filter parameters are iteratively adjusted to obtain the final optimized signal. According to the final optimized signal, the suppression degree of electromagnetic interference is determined to generate a stable output signal.

[0169] In a possible implementation, when the main component of electromagnetic interference exceeds the preset threshold, analyzing the interference intensity is crucial.

[0170] For example, assume that the interference intensity reaches 100 units in a certain frequency band, and the threshold is set at 80 units, indicating that further processing is required. The peak characteristics of the interference can be extracted from the spectrum data, and its distribution law on the time axis can be observed, such as three obvious intensity peaks appearing per second. This analysis helps to judge whether the interference has periodic characteristics and provides a basis for subsequent filtering.

[0171] Specifically, when obtaining the characteristic data of the interference intensity through the component analysis result, the frequency range and amplitude change of the interference can be concerned.

[0172] Exemplarily, if the characteristic data shows that the interference is concentrated in the range of 10 kHz to 15 kHz, and the amplitude fluctuates between 50 and 120 units, exceeding the threshold standard by 20 units, the filter needs to be adjusted. At this time, the cut-off frequency of the adaptive filter can be dynamically adjusted to below 15 kHz according to the interference range to initially weaken the interference signal. This adjustment method can quickly respond to the change of interference characteristics and maintain the stability of the system.

[0173] It should be noted that after the preliminary filtered signal is generated, the process of optimizing the signal using signal generation technology is crucial.

[0174] For example, by superimposing a reverse compensation signal, the interference peak is reduced from 100 units to 30 units, approaching the threshold requirement. This technology can effectively smooth the signal waveform and avoid the influence of interference on subsequent processing. The optimized signal can also provide clearer basic data for the analysis of residual components.

[0175] In one embodiment, when extracting the residual component of electromagnetic interference from the optimized signal, it can be judged by spectrum comparison.

[0176] For example, in the optimized signal, the residual component still has an intensity of 20 units at 12 kHz, which is higher than the preset threshold of 10 units. At this time, it is necessary to iteratively adjust the filter parameters. A possible implementation is to reduce the gain of the filter by 0.5 times and narrow the bandwidth to 2 kHz simultaneously to gradually reduce the residual interference. This iterative process can significantly improve the signal quality and ensure effective suppression of interference.

[0177] For example, after the final optimized signal is generated, when determining the degree of electromagnetic interference suppression, the stability of the signal can be observed. Suppose the fluctuation range of the original interference signal is between 80 and 120 units and it drops to 5 to 10 units after processing, indicating that the suppression degree reaches more than 90%. At this time, the generated stable output signal can be directly used for downstream devices to avoid misjudgment or failure caused by interference. This improvement in stability significantly helps the overall performance of the system.

[0178] Preferably, in an actual scenario, the parameter adjustment of the adaptive filter can be combined with real-time monitoring data.

[0179] For example, when the ambient noise suddenly increases, causing the interference intensity to rise from 100 units to 150 units, the filter can automatically increase the cut-off frequency to 18 kHz and enhance the attenuation effect. This flexibility enables the system to still operate efficiently in a complex environment and reduces the need for manual intervention.

[0180] It can be understood that after the residual component is lower than the preset threshold, the reliability of the signal will be significantly improved.

[0181] For example, in a communication device, after the interference drops from 50 units to 8 units, the bit error rate of data transmission drops from 5% to 0.1%, directly improving the communication quality. This effect is particularly important in high-precision application scenarios.

[0182] In one embodiment, after generating the stable output signal, its long-term stability can be further verified.

[0183] For example, after continuous operation for 24 hours, the signal intensity fluctuation remains within 5 units, proving the persistence of the interference suppression scheme. This reliability provides guarantee for the continuous operation of the device and reduces the maintenance cost at the same time.

[0184] S106. Perform space-time coding processing on the optimized filtered signal, and use a two-dimensional convolution algorithm to eliminate weather influence factors to obtain an anti-interference enhanced signal.

[0185] Optimize and filter the input signal using preset filter parameters to obtain a filtered signal. Perform space-time coding on the filtered signal to generate a coded signal through coding techniques. Use a two-dimensional convolution algorithm to perform convolution operations on the coded signal to obtain a convolution result. Remove weather influence factors through the convolution result to obtain a preliminary enhanced signal. If there is residual interference in the preliminary enhanced signal, use an anti-interference detection method to determine the interference degree and obtain a detection result. Adjust the preliminary enhanced signal according to the detection result using weighted processing to obtain a final enhanced signal. Verify the effectiveness of the processing flow through the final enhanced signal to determine the anti-interference enhanced signal.

[0186] Specifically, when optimizing and filtering the input signal, using preset filter parameters is a key step.

[0187] For example, in a wireless communication system, the input signal may be affected by the superposition of multipath effects and noise.

[0188] It can be understood that the preset filter parameters are usually designed based on historical data or channel characteristics. For example, a low-pass filter with a cut-off frequency of 5 kHz is set to initially weaken high-frequency noise and obtain a filtered signal. This method can retain the main features of the signal while reducing the ambiguity caused by interference.

[0189] In a possible implementation, performing space-time coding on the filtered signal is to further improve the spatial and temporal resolution capabilities of the signal.

[0190] Specifically, space-time block coding technology can be used to divide the filtered signal into multiple time slots and allocate them to different antenna elements.

[0191] For example, a 4-antenna system may divide the signal into 4 time slots, each time slot with an interval of 2 milliseconds, and generate a coded signal through coding techniques. This coding helps to distinguish signal components on different paths in subsequent processing.

[0192] Preferably, when using a two-dimensional convolution algorithm to perform convolution operations on the coded signal, it aims to extract the spatial and temporal features of the signal.

[0193] For example, in radar signal processing, a 3×3 convolution kernel can be used to perform a sliding window operation on the coded signal to obtain a convolution result. This method can highlight the local patterns of the signal and remove the influence of random perturbations.

[0194] It should be noted that the size and weight of the convolution kernel can be adjusted according to the signal characteristics. For example, in a strong interference scenario, the kernel size can be increased to 5×5. Removing weather influence factors through the convolution result is an important link in practical applications.

[0195] In one embodiment, if the signal is affected by rain attenuation, the attenuation characteristics caused by the weather can be identified by analyzing the low-frequency components in the convolution result.

[0196] For example, when it is detected that the components with frequencies lower than 1 kHz are abnormally enhanced, they can be classified as weather interference and filtered out to obtain a preliminary enhanced signal. This processing can effectively reduce the shielding of the signal by environmental factors.

[0197] Exemplarily, if there is residual interference in the preliminary enhanced signal, an anti-interference detection method is used to determine the degree of interference.

[0198] Specifically, the ratio of the peak value to the mean value of the signal can be calculated. For example, if the peak value reaches 10 and the mean value is 2, it is concluded that the interference ratio is relatively high. When the detection result shows that the degree of interference exceeds the standard, the signal needs to be further adjusted. This detection method is intuitive and easy to implement, and can quickly locate the problem. When adjusting the preliminary enhanced signal according to the detection result, weighted processing is a commonly used means.

[0199] In one embodiment, different weights can be assigned to different frequency bands of the signal. For example, a weight of 0.8 is assigned to the low-frequency band and a weight of 0.3 is assigned to the high-frequency band to suppress the residual interference and obtain the final enhanced signal. This weighting method can balance the clarity and integrity of the signal.

[0200] It can be understood that to verify the effectiveness of the processing flow through the final enhanced signal, it is usually necessary to compare the differences between the input signal and the output signal.

[0201] For example, if the signal-to-noise ratio of the input signal is 5 dB and the final enhanced signal is increased to 15 dB, it indicates that the anti-interference ability is significantly enhanced. This verification method can intuitively reflect the practical value of the process and provide a basis for subsequent optimization.

[0202] S107. Extract the transmission signal from the adjusted coded signal, optimize the bandwidth and power parameters using a dynamic resource allocation algorithm, and output a stable communication signal.

[0203] The transmission signal is obtained from the coded signal through a decoder to obtain the initial transmission data. The integrity of the initial transmission data is judged using a preset threshold to obtain a valid transmission signal. The bandwidth parameter of the valid transmission signal is adjusted using a dynamic resource allocation algorithm to obtain an optimized bandwidth signal. The power parameter of the optimized bandwidth signal is adjusted using a dynamic resource allocation algorithm to obtain an optimized power signal. According to the characteristics of the optimized power signal, signal processing techniques are used to smooth the noise to obtain a smoothed communication signal. The stability of the smoothed communication signal is detected using a channel simulation tool to obtain a stable communication signal.

[0204] Specifically, obtaining the transmission signal from the coded signal through a decoder to obtain the initial transmission data is a key link in the communication system.

[0205] Exemplarily, in wireless communication, a decoder can extract a bitstream from a received encoded signal based on a preset modulation method, such as QPSK or 16QAM. Assuming a transmission rate of 10 Mbps, the decoder recovers the initial transmitted data containing data frames by identifying the amplitude and phase of the signal. The integrity of the initial transmitted data is judged using a preset threshold to obtain a valid transmission signal, and this method ensures data quality.

[0206] In one possible implementation, the bit error rate threshold can be set to 10^-5. If the bit error rate of the initial transmitted data is lower than this value, the data is considered complete.

[0207] For example, in a scenario where the packet size is 1024 bit, if no more than 1 bit of error is detected, it is determined as a valid transmission signal. This method can effectively screen out reliable data and improve the accuracy of subsequent processing. The bandwidth parameters of the valid transmission signal are adjusted through a dynamic resource allocation algorithm to obtain an optimized bandwidth signal, and this technology is applicable to environments with tight resources.

[0208] Specifically, the dynamic resource allocation algorithm can adjust the bandwidth in real time according to the channel state information CSI. Assuming the current channel capacity is 20 MHz, when the algorithm detects a high-load period, the bandwidth is extended from 5 MHz to 10 MHz to ensure the transmission efficiency. This flexibility can significantly improve the throughput of the system. The power parameters of the optimized bandwidth signal are adjusted through the dynamic resource allocation algorithm to obtain an optimized power signal, and this step further optimizes resource utilization.

[0209] Preferably, the algorithm adjusts the transmission power according to the signal strength and interference level.

[0210] For example, at a distance of 500 meters from the base station where the signal attenuation is large, the power is increased from 10 dBm to 15 dBm to ensure the signal coverage range. This method can not only save energy consumption but also maintain the communication quality. According to the characteristics of the optimized power signal, signal processing technology is used to smooth the noise to obtain a smooth communication signal, and this processing enhances the clarity of the signal.

[0211] In one embodiment, a low-pass filter can be used, and the cut-off frequency is set to 1.2 times the signal bandwidth, such as 12 MHz, to filter out high-frequency noise. For example, if the noise power spectral density is -174 dBm / Hz, after smoothing, the signal-to-noise ratio is increased from 10 dB to 15 dB, which helps to reduce bit errors. The stability of the smooth communication signal is detected through a channel simulation tool to obtain a stable communication signal, and this verification ensures the reliability of the system.

[0212] It is understandable that the channel simulation tool simulates scenarios such as multipath fading and Doppler frequency shift.

[0213] For example, in a mobile environment with a speed of 60 km / h, the simulation results show that the signal amplitude fluctuation is less than 2 dB, proving its stability. This detection provides a reliable basis for actual deployment.

[0214] It should be noted that the implementation methods of each of the above links are closely connected, forming a complete signal processing chain from decoding to stability detection.

[0215] For example, in bandwidth and power adjustment, the dynamic algorithm takes into account both user density and channel congestion. Assuming the number of users increases from 50 to 100, the bandwidth and power will be optimized collaboratively. This design supported by multiple aspects not only ensures the comprehensiveness of the solution but also enriches the feasibility of implementation through specific scenarios.

[0216] Preferably, this method can also adapt to different communication environments and improve the robustness of the system.

[0217] Embodiment 2:

[0218] The present invention also provides an enhanced system for low-altitude helicopter communication navigation signals based on intelligent scheduling, including:

[0219] A time-frequency feature extraction module for obtaining time-frequency feature data collected by a sensor array;

[0220] An interference source parameter extraction module for extracting interference source parameters using the fast Fourier transform algorithm;

[0221] An interference source distribution map generation module for generating an interference source distribution map;

[0222] A spatio-temporal change trend map generation module for generating a spatio-temporal change trend map if the dynamic prediction result of the deep learning model for the interference source distribution map exceeds a preset threshold;

[0223] A signal path blocking point calculation module for calculating signal path blocking points using a digital elevation model according to the spatio-temporal change trend map;

[0224] An electromagnetic interference intensity extraction module for determining the range of the occlusion influence area and extracting electromagnetic interference intensity data from the range of the occlusion influence area;

[0225] An optimized filtering signal generation module for decomposing to obtain the main component components of the interference frequency through a spectrum analysis algorithm, and adjusting the parameters of the adaptive filter to generate an optimized filtering signal if the electromagnetic interference intensity corresponding to the main component components is higher than a preset threshold;

[0226] an anti-interference enhancement signal generation module, configured to perform space-time coding processing on the optimized filtered signal, and adopt a two-dimensional convolution algorithm to eliminate weather influence factors to obtain an anti-interference enhancement signal;

[0227] The stable communication signal output module is used to extract the transmission signal from the adjusted coded signal, optimize the bandwidth and power parameters using a dynamic resource allocation algorithm, and output a stable communication signal.

[0228] It should be noted that the above examples are merely specific embodiments of the present invention. Obviously, the present invention is not limited to the above examples and is subject to numerous variations. All variations that can be directly derived or conceived by a person skilled in the art from the disclosure of the present invention should be considered within the scope of protection of the present invention.

Claims

1. A method for enhancing communication and navigation signals of low-altitude helicopters based on intelligent scheduling, characterized in that Including: Obtain the time-frequency feature data collected by the sensor array, extract the interference source parameters, and generate an interference source distribution map; If the dynamic prediction result of the interference source distribution map exceeds the preset threshold, generate a spatio-temporal change trend map; According to the spatio-temporal change trend map, calculate the signal path blocking point and determine the range of the occlusion influence area; Extract the electromagnetic interference intensity data from the range of the occlusion influence area and decompose it to obtain the main component components of the interference frequency; If the electromagnetic interference intensity corresponding to the main component component is higher than the preset threshold, adjust the adaptive filter parameters to generate an optimized filtering signal; Perform space-time coding processing on the optimized filtering signal to obtain an anti-interference enhanced signal; Extract the transmission signal from the adjusted coded signal, optimize the bandwidth and power parameters, and output a stable communication signal.

2. The method for enhancing communication and navigation signals of a low-altitude helicopter based on intelligent scheduling according to claim 1, wherein Obtain the time-frequency feature data collected by the sensor array, use the fast Fourier transform algorithm to extract the interference source parameters, and generate an interference source distribution map.

3. The method for enhancing communication and navigation signals of a low-altitude helicopter based on intelligent scheduling according to claim 1, wherein According to the spatio-temporal change trend map, use the digital elevation model to calculate the signal path blocking point and determine the range of the occlusion influence area.

4. The method for enhancing communication and navigation signals of a low-altitude helicopter based on intelligent scheduling according to claim 1, wherein Extract the electromagnetic interference intensity data from the range of the occlusion influence area and decompose it through the spectrum analysis algorithm to obtain the main component components of the interference frequency.

5. The method for enhancing communication and navigation signals of a low-altitude helicopter based on intelligent scheduling according to claim 1, wherein Perform space-time coding processing on the optimized filtering signal, use the two-dimensional convolution algorithm to eliminate the weather influence factors, and obtain an anti-interference enhanced signal.

6. The method for enhancing communication and navigation signals of a low-altitude helicopter based on intelligent scheduling according to claim 1, wherein Extract the transmission signal from the adjusted coded signal, use the dynamic resource allocation algorithm to optimize the bandwidth and power parameters, and output a stable communication signal.

7. The method for enhancing communication and navigation signals of a low-altitude helicopter based on intelligent scheduling according to claim 1, characterized in that If the dynamic prediction result of the interference source distribution map by using the deep learning model exceeds the preset threshold, generate a spatio-temporal change trend map.

8. A low-altitude helicopter communication and navigation signal enhancement system based on intelligent scheduling, characterized in that: Including: A time-frequency feature extraction module for obtaining the time-frequency feature data collected by the sensor array; An interference source parameter extraction module for using the fast Fourier transform algorithm to extract the interference source parameters; An interference source distribution map generation module for generating an interference source distribution map; A spatio-temporal change trend map generation module for generating a spatio-temporal change trend map if the dynamic prediction result of the interference source distribution map by the deep learning model exceeds the preset threshold; A signal path blocking point calculation module for calculating the signal path blocking point according to the spatio-temporal change trend map by using the digital elevation model; An electromagnetic interference intensity extraction module for determining the range of the occlusion influence area and extracting the electromagnetic interference intensity data from the range of the occlusion influence area; An optimized filtering signal generation module for decomposing to obtain the main component components of the interference frequency through the spectrum analysis algorithm, and if the electromagnetic interference intensity corresponding to the main component component is higher than the preset threshold, adjusting the adaptive filter parameters to generate an optimized filtering signal; An anti-interference enhanced signal generation module for performing space-time coding processing on the optimized filtering signal, using the two-dimensional convolution algorithm to eliminate the weather influence factors, and obtaining an anti-interference enhanced signal; A stable communication signal output module for extracting the transmission signal from the adjusted coded signal, using the dynamic resource allocation algorithm to optimize the bandwidth and power parameters, and outputting a stable communication signal.

Images