Communication method of two-channel unmanned aerial vehicle line-of-sight communication system

Abstract

The invention relates to the technical field of unmanned aerial vehicle communication, in particular to a communication method of a dual-channel unmanned aerial vehicle line-of-sight communication system, which comprises the steps of acquiring communication quality, acquiring an optimal transmission frequency point, acquiring task priority of an unmanned aerial vehicle, transmitting video stream data and executing multipath fading compensation. Calculating a frequency point quality index through a signal-to-noise ratio, a signal rate and an error rate, and dynamically selecting an optimal transmission frequency point; allocating communication resources based on task priorities to ensure that key instructions are transmitted preferentially; a differential geometry theory is innovatively adopted to carry out multipath fading compensation, communication signals are mapped to a Riemannian manifold space, and fading features are extracted and optimized; cooperative compensation of high and low frequency channels is realized, and mutual enhancement is realized through receiving spectrum analysis and training sequence correction; and adaptive QAM modulation is supported, modulation parameters are dynamically adjusted according to manifold curvature, and a star-shaped multi-node communication network is constructed.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of unmanned aerial vehicle communication, in particular to a communication method of a dual-channel unmanned aerial vehicle line-of-sight communication system, which is suitable for long-distance and high-reliability communication of fixed-wing unmanned aerial vehicles. BACKGROUND

[0002] With the rapid development of unmanned aerial vehicle technology, unmanned aerial vehicles are increasingly widely used in military reconnaissance, disaster monitoring, agricultural plant protection, power inspection and other fields. In these application scenarios, long-distance and high-reliability communication capability is the key to ensuring the safe and efficient execution of unmanned aerial vehicle tasks. Traditional unmanned aerial vehicle communication systems usually use a single frequency band for information transmission, such as the commonly used 2.4 GHz, 5.8 GHz or 433 MHz frequency bands. However, single-frequency band communication faces many challenges in practical applications: high-frequency bands (such as 5.8 GHz) have higher bandwidth and can support large data volume transmission such as video, but their transmission distance is short and they are easily blocked by obstacles; low-frequency bands (such as 433 MHz) have a longer transmission distance and stronger penetration ability, but the bandwidth is limited and it is difficult to support high-speed data transmission. In addition, during high-speed flight of the unmanned aerial vehicle, the rapid changes of the communication channel, multipath effects and various interference factors will all lead to unstable communication quality, especially in complex terrain and electromagnetic environments. In the prior art, although there are schemes that use dual-frequency band communication, they are mostly simple redundancy backup or fixed division of labor, lack intelligent frequency selection mechanisms and effective channel optimization methods, and are difficult to fully exploit the potential of dual channels.

[0003] For the problem of multipath fading, traditional methods mainly use statistical models for analysis and compensation, such as the Rayleigh fading model and the Rice fading model. These methods are effective in some scenarios, but they are difficult to accurately describe and handle complex multipath effects in high-speed flight environments, especially difficult to capture the nonlinear and time-varying characteristics of the multipath environment.

[0004] Therefore, there is an urgent need to develop an unmanned aerial vehicle communication method that can intelligently utilize the advantages of dual-frequency bands and effectively address the challenges of multipath fading, in order to improve the distance, reliability and data transmission capability of unmanned aerial vehicle communication. SUMMARY

[0005] The purpose of the present application is to provide a communication method of a dual-channel unmanned aerial vehicle line-of-sight communication system, which solves the technical problems of long-distance and high-reliability transmission in existing unmanned aerial vehicle communication through intelligent collaboration of high-frequency and low-frequency channels, multipath fading compensation based on differential geometry theory and other innovative technologies.

[0006] The present application proposes a communication method of a dual-channel unmanned aerial vehicle line-of-sight communication system, which includes:

[0007] Acquiring a communication quality; comprising: setting a high frequency channel and a low frequency channel to transmit data to the target UAV, and acquiring a communication quality index; the high frequency channel is a 5.8GHz frequency band, and the low frequency channel is a 433MHz frequency band; the communication quality index includes a signal-to-noise ratio, a signal rate and a bit error rate;

[0008] Acquiring an optimal transmission frequency point; comprising: judging the frequency point quality of the high frequency channel and the low frequency channel based on the communication quality index, and determining an optimal transmission frequency point;

[0009] Acquiring a task priority of the UAV; comprising: determining a task priority type according to the task priority, and selecting a corresponding transmission frequency point to transmit a control instruction based on the task priority type and the optimal transmission frequency point;

[0010] Transmitting video stream data; comprising: acquiring video stream data transmitted by the UAV, and transmitting the video stream data to the target UAV through the optimal transmission frequency point;

[0011] Performing multipath fading compensation; comprising: collecting communication signals at multiple time points, mapping the communication signals to a differential manifold space, extracting fading features in the differential manifold space, performing optimization processing on the communication signals based on the fading features, and generating optimized communication signals.

[0012] Preferably, the performing multipath fading compensation specifically comprises:

[0013] Acquiring a signal-to-noise ratio of the communication signals at multiple time points;

[0014] Sorting the signal-to-noise ratios and selecting one signal-to-noise ratio as a reference signal-to-noise ratio;

[0015] Based on the reference signal-to-noise ratio, acquiring an average fading depth of the communication signals at multiple time points;

[0016] Multiplying the signal-to-noise ratio of the communication signal at each time point by the average fading depth to obtain a weight value of the communication signal at each time point;

[0017] Based on the weight value, superimposing the communication signals at multiple time points to obtain a superimposed signal;

[0018] Extracting a communication baseband signal from the superimposed signal and acquiring a signal-to-noise ratio of the communication baseband signal;

[0019] When the signal-to-noise ratio of the communication baseband signal is greater than the signal-to-noise ratios of the communication signals at multiple time points, it is determined that the multipath fading compensation is completed.

[0020] Preferably, the acquiring an optimal transmission frequency point specifically comprises:

[0021] when the acquired frequency point quality index of the low frequency channel is greater than a first preset threshold and the acquired frequency point quality index of the high frequency channel is available, the high frequency channel is taken as the best transmission frequency point at the current moment;

[0022] when the acquired frequency point quality index of the low frequency channel is less than the first preset threshold and the acquired frequency point quality index of the high frequency channel is available, the low frequency channel is taken as the best transmission frequency point at the current moment;

[0023] when the acquired frequency point quality index of the low frequency channel is unavailable, both the high frequency channel and the low frequency channel are taken as the best transmission frequency point at the current moment.

[0024] Preferably, the selecting of the corresponding transmission frequency point for transmitting the control instruction based on the task priority type and the best transmission frequency point specifically comprises:

[0025] when the task priority type is the highest priority task, the high frequency channel or the best transmission frequency point at the current moment is taken as the transmission frequency point of the highest priority task, and the control instruction is transmitted to the target UAV;

[0026] when the task priority type is the second highest priority task, the transmission frequency point of the second highest priority task is the one with higher frequency point quality index, and the control instruction is transmitted to the target UAV;

[0027] when the task priority type is the ordinary task, the low frequency channel or the best transmission frequency point at the current moment is taken as the transmission frequency point of the ordinary task, and the control instruction is transmitted to the target UAV.

[0028] Preferably, the mapping of the communication signal to the differential manifold space specifically comprises:

[0029] constructing a parameter space, each dimension of the parameter space corresponding to a signal feature at a moment;

[0030] defining a metric tensor to establish a distance concept in the parameter space;

[0031] mapping the parameter space to a Riemann manifold through local coordinate transformation, keeping the topological structure unchanged;

[0032] establishing an orthogonal basis on the Riemann manifold for subsequent geometric analysis.

[0033] Preferably, the extracting of the fading feature in the differential manifold space specifically comprises:

[0034] calculating a curvature tensor on the Riemann manifold to represent the local space bending degree;

[0035] calculating the propagation characteristics of the signal on the Riemann manifold through parallel transmission equation;

[0036] analyzing the principal direction of the curvature tensor, identifying the main influencing factor of fading;

[0037] establishing a reference plane with reference signal-to-noise ratio, calculating the geodesic distance of other signal points to the reference plane;

[0038] converting the geodesic distance into a fading depth parameter, establishing a mapping relationship between geometric quantities and physical quantities;

[0039] constructing a fading depth distribution map to represent the spatial distribution characteristics of the multipath environment.

[0040] As preferred, the optimization processing of the communication signal based on the fading characteristics specifically includes:

[0041] calculating the geometric weight of the signal at each moment, which is determined by the product of the signal-to-noise ratio and the average fading depth;

[0042] calculating the geodesic line connecting each signal point on the Riemannian manifold to represent the optimal signal change path;

[0043] segmenting and interpolating the signal along the geodesic line to maintain phase continuity;

[0044] constructing a curvature flow equation on the Riemannian manifold to describe the optimal evolution of the Riemannian manifold over time;

[0045] setting the evolution boundary condition to ensure convergence to a physically meaningful solution;

[0046] controlling the evolution rate to balance the calculation efficiency and optimization accuracy;

[0047] ensuring the consistency of the signal phase in the fusion process through parallel transmission theory.

[0048] As preferred, it also includes a high-low frequency channel cooperative compensation step:

[0049] receiving the received spectrum of the low-frequency channel through the high-frequency channel;

[0050] obtaining multipath fading information based on the received spectrum;

[0051] correcting transmission errors in the data transmission process of the low-frequency channel according to the multipath fading information;

[0052] transmitting signals to the low-frequency channel through the high-frequency channel;

[0053] adding a training sequence in the low-frequency channel;

[0054] compensating for the phase noise of the low-frequency channel transmission signal according to the training sequence;

[0055] Constructing high-low frequency joint manifold to describe the coupling characteristics of the dual channels;

[0056] Analyzing the correlation and complementarity of the fading characteristics of different frequency bands;

[0057] Establishing the mapping relationship between frequency bands to achieve complementary enhancement.

[0058] As preferred, it further includes an adaptive QAM modulation step:

[0059] Obtaining communication data, converting the communication data into a plurality of signals;

[0060] Baseband signal modulation is performed on the plurality of signals, and the baseband signals are spread;

[0061] Based on the differential manifold analysis result, a channel state reference is provided for adaptive QAM modulation of the high-frequency channel;

[0062] According to the multipath characteristics, the distribution of the signal constellation diagram is automatically adjusted;

[0063] In the high-curvature area of the Riemann manifold, the modulation order is reduced;

[0064] In the flat area of the Riemann manifold, the modulation order is increased;

[0065] A bidirectional information interaction mechanism is established between the modulation system and the compensation system.

[0066] As preferred, it further includes a multi-node communication network construction step:

[0067] A star network is constructed, including a control end, relay nodes, slave nodes, and target unmanned aerial vehicles;

[0068] The relay nodes, slave nodes, and target unmanned aerial vehicles are respectively connected to the control end;

[0069] Time diversity reception is used between any two adjacent nodes;

[0070] The slave nodes send communication signals to the control end;

[0071] When the communication signals arrive at the control end, there are multiple time communication signals in the control end;

[0072] The multiple time communication signals are respectively QAM modulated to obtain the optimal timing of multiple times;

[0073] The optimal timing of multiple times is used for signal recovery of the communication signals;

[0074] The control end receives the multi-antenna output signal of the relay node;

[0075] Compensate for time delay difference of the multi-antenna output signal;

[0076] Superimpose the compensated signal with the received multi-antenna output signal to obtain communication signals at multiple time points.

[0077] The present application has the following advantages:

[0078] 1. Through intelligent selection and collaborative work of high and low frequency channels, both communication distance and bandwidth requirements are considered, and the adaptability and reliability of the unmanned aerial vehicle communication system are significantly improved;

[0079] 2. The channel allocation strategy based on task priority ensures the reliable transmission of critical control instructions and improves the safety of unmanned aerial vehicle flight;

[0080] 3. The differential geometry theory is innovatively applied to multipath fading compensation, and a geometric characterization model based on signal manifold is established, which can more accurately capture the nonlinear and time-varying characteristics of the multipath environment;

[0081] 4. Through geodesic fusion and curvature flow optimization technology, the optimal synthesis of multipath signals is realized, which effectively improves the signal-to-noise ratio and prolongs the effective communication distance;

[0082] 5. The cross-band collaborative compensation mechanism of high and low frequency channels fully utilizes the complementary advantages of the two frequency bands, further improving the communication reliability;

[0083] 6. Deep integration with adaptive QAM modulation technology realizes dynamic matching of channel state and modulation parameters, and optimizes the spectrum utilization efficiency;

[0084] 7. Support for multi-node communication network construction provides communication guarantee for multi-unmanned aerial vehicle cooperative task.

[0085] Overall, the dual-channel unmanned aerial vehicle line-of-sight communication method provided by the present application has made a significant breakthrough in theoretical innovation and engineering practicability, and provides a new technical path for long-distance and high-reliability communication of unmanned aerial vehicles. BRIEF DESCRIPTION OF DRAWINGS

[0086] Figure 1 The communication method flowchart of the dual-channel unmanned aerial vehicle line-of-sight communication system of the present application;

[0087] Figure 2 The high and low frequency channel frequency point quality judgment flowchart of the present application;

[0088] Figure 3 The channel allocation strategy flowchart based on task priority of the present application;

[0089] Figure 4 The multipath fading compensation method flowchart based on differential manifold of the present application. DETAILED DESCRIPTION

[0090] Reference will now be made to the drawings and specific working examples, which are intended to be illustrative only so that the application can be readily understood. Figures 1-4 The skilled person will understand that these examples are merely illustrative of the application and do not limit the scope of the application.

[0091] Example 1

[0092] As shown in the accompanying drawings and specific working examples, the application provides a dual-channel unmanned aerial vehicle line-of-sight communication method, comprising the following steps: Figure 1

[0093] In a preferred embodiment of the application, a high-frequency channel and a low-frequency channel are first set up to transmit data to the target unmanned aerial vehicle and obtain a communication quality index. Specifically, the high-frequency channel is a 5.8 GHz frequency band, and the low-frequency channel is a 433 MHz frequency band. The selection of these two frequency bands is based on the actual needs of unmanned aerial vehicle communication: the 5.8 GHz frequency band has a higher bandwidth and is suitable for transmitting large amounts of data such as video; the 433 MHz frequency band has a longer transmission distance and stronger penetration ability, and is suitable for transmitting critical information such as control instructions.

[0094] The communication quality index includes signal-to-noise ratio, signal rate, and bit error rate. In actual applications, the communication quality index can be calculated by the following formula:

[0095] ,

[0096] wherein: is the frequency point quality index, representing the comprehensive quality score of the communication link, with a value range of 0-100; is the current signal-to-noise ratio, representing the ratio of signal power to noise power, with a unit of dB; is the reference signal-to-noise ratio, serving as a normalization reference value, and is usually taken as 10 dB; is the current signal rate, representing the data transmission speed, with a unit of Mbps; is the maximum signal rate, representing the maximum data transmission speed supported by the system, and is usually taken as 48 Mbps for the high-frequency channel and 2 Mbps for the low-frequency channel; is the current bit error rate, representing the ratio of received error bits to total transmission bits, and is dimensionless; is the maximum acceptable bit error rate, representing the upper limit of the bit error rate that the system can tolerate, and is usually taken as ; , , is the weight coefficient, used to adjust the importance of each index, and satisfies , and is adjusted according to actual application needs, with a typical value of ,​ 、 .

[0097] This calculation method comprehensively considers the signal-to-noise ratio, signal rate and bit error rate three key indicators, and can comprehensively reflect the quality state of the communication link.

[0098] As shown in Figure 2 , based on the communication quality index, the frequency point quality of the high frequency channel and the low frequency channel is judged, and the best transmission frequency point is determined. The specific judgment strategy is as follows:

[0099] When the acquired frequency point quality index of the low frequency channel is greater than the first preset threshold and the acquired frequency point quality index of the high frequency channel is available, the high frequency channel is used as the best transmission frequency point at the current moment. In the preferred embodiment, the first preset threshold is set to 80, which is an empirical value based on a large amount of experimental data analysis, and can achieve good balance in most application scenarios. "Available" is defined as the frequency point quality index being greater than a third preset threshold (usually set to 30).

[0100] When the acquired frequency point quality index of the low frequency channel is less than the first preset threshold and the acquired frequency point quality index of the high frequency channel is available, the low frequency channel is used as the best transmission frequency point at the current moment. At this time, although the quality of the low frequency channel is not very high, due to its inherent transmission distance advantage, in the case that the high frequency channel is available, the low frequency channel is preferred to ensure the basic reliability of communication.

[0101] When the acquired frequency point quality index of the low frequency channel is not available, the high frequency channel and the low frequency channel are both used as the best transmission frequency point at the current moment. "Not available" usually means that the frequency point quality index is lower than a second preset threshold (usually set to 20), at this time, a dual-channel simultaneous transmission strategy is adopted to maximize the reliability of communication.

[0102] This dynamic frequency point selection mechanism can automatically select the best transmission channel according to the real-time communication environment, and fully exert the respective advantages of the high and low frequency channels.

[0103] As shown in Figure 3 , the present application determines the task priority type according to the task priority of the unmanned aerial vehicle, and selects the corresponding transmission frequency point to transmit the control instruction based on the task priority type and the best transmission frequency point.

[0104] In the preferred embodiment of the present application, the task priority type includes the highest level task, the second highest level task and the ordinary task. The highest level task usually involves the safety control of the unmanned aerial vehicle, such as emergency obstacle avoidance, return instruction, etc.; the second highest level task includes flight path adjustment, attitude control, etc.; the ordinary task includes data acquisition, environment monitoring and other non-critical operations.

[0105] The specific channel selection strategy is as follows:

[0106] When the task priority type is the highest priority task, the high-frequency channel or the best transmission frequency point at the current moment is used as the transmission frequency point of the highest priority task, and a control instruction is transmitted to the target UAV. Preferably, the high-frequency channel is used because of its high data transmission rate, which can ensure the timely transmission of critical instructions.

[0107] When the task priority type is the second highest priority task, the transmission frequency point with the higher frequency point quality index is used as the transmission frequency point of the second highest priority task, and a control instruction is transmitted to the target UAV. This strategy can ensure communication quality while providing sufficient communication resources for the second highest priority task.

[0108] When the task priority type is the ordinary task, the low-frequency channel or the best transmission frequency point at the current moment is used as the transmission frequency point of the ordinary task, and a control instruction is transmitted to the target UAV. For ordinary tasks, the low-frequency channel is preferred to save high-frequency channel resources, and the long-distance transmission advantage of the low-frequency channel is utilized.

[0109] This channel allocation strategy based on task priority ensures the reliable transmission of critical instructions and improves the safety and effectiveness of UAV control.

[0110] After completing the transmission of the control instruction, the application obtains video stream data transmitted by the UAV and transmits the video stream data to the target UAV through the best transmission frequency point.

[0111] In practical applications, video stream data usually has a large amount of data and requires a high transmission bandwidth. Therefore, when the communication conditions are good, video data is preferentially transmitted through the high-frequency channel; when the communication conditions are poor or long-distance transmission is required, the resolution or frame rate can be reduced to transmit critical video information through the low-frequency channel.

[0112] In addition, the application can dynamically adjust the video encoding parameters and transmission strategy according to the importance of the video content. For example, for critical regions containing target recognition results, a higher encoding quality is used; for background regions, a lower encoding quality can be used to save bandwidth resources.

[0113] As shown in Figure 4 An innovative point of the application is a multipath fading compensation method based on differential geometry theory. The method includes: collecting communication signals at multiple moments, mapping the communication signals to a differential manifold space, extracting fading features in the differential manifold space, optimizing the communication signals based on the fading features, and generating optimized communication signals.

[0114] In a preferred embodiment of the present application, n time instants of communication signals are collected, where n is an integer greater than 2, and the recommended value is 5-7. The interval of collection time points is usually set to 1 / 3 of the channel coherence time to ensure the capture of the dynamic changes of the channel. For a fixed-wing unmanned aerial vehicle in the 5.8 GHz frequency band, the coherence time is about 10-20 ms, so it is appropriate to set the sampling interval to 3-7 ms.

[0115] For each sampling point, the complete signal waveform characteristics are recorded, including amplitude, phase, frequency, and timestamp information. The antenna position is kept relatively stable during sampling to reduce additional fading caused by position changes.

[0116] The present application innovatively maps the collected communication signals to a differential manifold space and constructs a geometric representation model of the signals. The specific implementation includes the following steps:

[0117] A parameter space is constructed, and each dimension of the parameter space corresponds to a signal feature at a time instant. For example, for n time instants of collected signals, an n-dimensional parameter space can be constructed, and each dimension represents a signal feature (such as amplitude, phase, etc.) at the corresponding time instant.

[0118] A metric tensor is defined to establish the distance concept in the parameter space. The metric tensor can be represented as:

[0119] ,

[0120] where: is the component of the metric tensor, representing the metric in the i-th and j-th coordinate directions in the parameter space, and is dimensionless; is the k-th component of the signal feature vector, which can be a physical quantity such as amplitude, phase, etc.; , is the coordinate of the parameter space, representing the parameters at different times or different features; is the dimension of the signal feature vector, representing the number of features; represents the feature is the partial derivative of the parameter , representing the degree of influence of parameter changes on the feature.

[0121] The parameter space is mapped to a Riemannian manifold through local coordinate transformation, keeping the topological structure unchanged. The coordinate transformation can be represented as:

[0122] ,

[0123] where: is the transformed coordinate, representing the position in the new coordinate system; is the coordinate transformation function, which is the mapping relationship from the original coordinate to the new coordinate; is the original coordinate, representing the position in the original coordinate system.

[0124] An orthogonal basis is established on the Riemannian manifold to provide a foundation for subsequent geometric analysis. The orthogonal basis can be generated by the Gram-Schmidt orthogonalization process, which ensures that the basis vectors are mutually orthogonal, facilitating subsequent geometric calculations.

[0125] This differential manifold-based signal representation method can convert discrete time-domain signals into continuous geometric structures, providing a mathematical foundation for the extraction and analysis of fading characteristics.

[0126] On the constructed Riemannian manifold, the invention extracts the characteristics of multipath fading through geometric analysis, with the specific steps as follows:

[0127] The curvature tensor is calculated on the Riemannian manifold to represent the degree of local space curvature. The Riemann curvature tensor can be expressed as:

[0128] ,

[0129] where: is the component of the Riemann curvature tensor, representing the curvature of the manifold in different directions, and is dimensionless. , , and are the components of the metric tensor, describing the distance relationship on the manifold. , and represent the second-order partial derivatives of the metric tensor with respect to coordinates, reflecting the rate of change of the metric tensor. is the Christoffel symbol, which is represented as:

[0130] ,

[0131] where: is the Christoffel symbol, describing the rate of change of the curve in parallel transmission on the manifold, and is dimensionless. is the inverse matrix element of the metric tensor, satisfying where is the Kronecker function. , and represent the partial derivatives of the metric tensor with respect to coordinates.

[0132] The propagation characteristics of the signal on the Riemannian manifold are calculated through the parallel transmission equation:

[0133] ,

[0134] where: is the component of the vector field, representing the vector on the manifold. represents the covariant derivative of the vector field along the curve, describing the rate of change of the vector on the curve. represents the ordinary derivative of the vector field. A tangent vector representing the curve, describing the direction of the curve. For a curve parameter, it can be understood as time.

[0135] Analyzing the principal directions of the curvature tensor, identifying the main factors affecting the fading. This can be achieved through eigenvalue decomposition of the Riemann curvature tensor:

[0136] ,

[0137] Where: is the simplified Riemann curvature tensor, is the contraction of ; is the eigenvalue, representing the size of the principal curvature; is the Kronecker function, equal to 1 when , otherwise equal to 0.

[0138] Establish a reference plane with the reference signal-to-noise ratio, and calculate the geodesic distance from other signal points to the reference plane. The geodesic distance can be obtained by solving the geodesic line equation:

[0139] ,

[0140] Where: represents the point on the geodesic line, and is the coordinate on the manifold; is the Christoffel symbol; and are the tangent vector components of the geodesic line; is the acceleration component of the geodesic line; is the parameter of the geodesic line.

[0141] Convert the geodesic distance to the fading depth parameter, establish the mapping relationship between the geometric quantity and the physical quantity:

[0142] ,

[0143] Where: is the fading depth parameter, representing the degree of signal fading, with units of dB; is the geodesic distance, representing the shortest distance between two points on the manifold, dimensionless; is the proportionality coefficient, used to convert the geometric distance to the physical fading quantity, usually calibrated according to actual measurement data, typical value is 0.1-0.5 dB / unit distance.

[0144] Construct the fading depth distribution map to characterize the spatial distribution characteristics of the multipath environment. This can be achieved by mapping the fading depth parameter to the original signal space.

[0145] Through this differential geometry-based analysis method, the nature of multipath fading can be understood in depth from the geometric perspective, and a theoretical basis is provided for subsequent signal optimization.

[0146] Based on the extracted fading features, the present application adopts a geodesic fusion and curvature flow optimization method to process the communication signals and generate optimized communication signals.

[0147] The geometric weight of each signal at each time is calculated, which is determined by the product of the signal-to-noise ratio and the average fading depth:

[0148] ,

[0149] wherein: is the geometric weight of the signal at the i-th time, indicating the importance of the signal at this time in the synthesis process, and is dimensionless; is the signal-to-noise ratio of the signal at the i-th time, with the unit of dB; is the average fading depth of the signal at the i-th time, which can be in dB; is a normalization coefficient to ensure that the total weight is 1, and the calculation method is .

[0150] The geodesic connecting each signal point is calculated on the Riemannian manifold to represent the optimal signal change path. The solution of the geodesic equation can be expressed as:

[0151] ,

[0152] wherein: represents a point on the geodesic, and is the coordinate on the manifold; is the starting point, indicating the initial position; is the starting velocity, indicating the initial direction; is the starting acceleration, which is determined by the Christoffel symbol; represents the high-order terms of the third order and above, which can be ignored when t is small.

[0153] ,

[0154] wherein: is the Christoffel symbol at the starting point; and are the components of the starting velocity. The signal is segmented and interpolated along the geodesic to maintain phase continuity. The interpolation formula can be expressed as:

[0155] ,

[0156] wherein: is the interpolated signal, indicating the signal value at the parameter ; For the first time instance, the signal is a known discrete signal value; For the interpolation basis function, a function that maintains phase continuity is usually chosen, such as a cubic spline function; For the number of time instances.

[0157] Construct the curvature flow equation on a Riemannian manifold, which describes the optimal evolution of the Riemannian manifold over time. The curvature flow equation can be expressed as:

[0158] ,

[0159] Where: The metric tensor describes the distance relationship on the manifold; The Ricci curvature tensor is the contraction of the Riemann curvature tensor; Indicates the rate of change of the metric tensor over time; The evolution parameter can be understood as time.

[0160] Set the evolution boundary condition to ensure convergence to a physically meaningful solution. Typical boundary conditions include maintaining the total volume of the manifold or maintaining the curvature characteristics of a specific region.

[0161] Control the evolution rate to balance the calculation efficiency and optimization accuracy. The evolution step size is usually related to the curvature value:

[0162] ,

[0163] Where: The evolution step size represents the time interval of each iteration; The control parameter controls the stability of the evolution, and the typical value is 0.01-0.1; The absolute value of the maximum component of the Ricci curvature tensor represents the maximum curvature of the manifold. Through the parallel transmission theory, the consistency of the signal phase in the fusion process is ensured. The phase correction matrix can be expressed as:

[0164] ,

[0165] Where: The phase correction matrix is used to adjust the signal phase; Indicates the path from the reference point to the target point; The Christoffel symbol; The infinitesimal of the path; Indicates the matrix exponential function.

[0166] The signal optimization method based on differential geometry can effectively improve signal quality and reduce the negative impact of multipath fading while maintaining key signal characteristics.

[0167] Embodiment 2

[0168] In another embodiment of the present application, as shown in Figure 4 The specific implementation steps of multipath fading compensation are as follows:

[0169] Obtain the signal-to-noise ratios of the communication signals at multiple time instants;

[0170] Sort the signal-to-noise ratios and select one as a reference signal-to-noise ratio;

[0171] Based on the reference signal-to-noise ratio, obtain the average fading depth of the communication signals at multiple time instants;

[0172] Multiply the signal-to-noise ratio of the communication signal at each time instant by the average fading depth to obtain the weight value of the communication signal at each time instant;

[0173] Superimpose the communication signals at multiple time instants based on the weight values to obtain a superimposed signal;

[0174] Extract the communication baseband signal from the superimposed signal and obtain the signal-to-noise ratio of the communication baseband signal;

[0175] When the signal-to-noise ratio of the communication baseband signal is greater than the signal-to-noise ratios of the communication signals at multiple time instants, it is determined that the multipath fading compensation is completed.

[0176] In actual application, the selected reference signal-to-noise ratio is usually the median or weighted average of the signal-to-noise ratios at multiple time instants, which can reduce the influence of extreme values on the system. The average fading depth can be calculated by the following formula:

[0177] ,

[0178] Wherein: is the average fading depth, representing the average degree of signal fading, dimensionless; is the reference signal-to-noise ratio, usually taking the median or weighted average of the signal-to-noise ratios at multiple time instants, unit: dB; is the signal-to-noise ratio at the i-th time instant, unit: dB; is the number of time instants; represents the absolute value of the relative deviation of the i-th time instant signal-to-noise ratio from the reference signal-to-noise ratio.

[0179] The calculation formula of the weight value is:

[0180] ,

[0181] Wherein: is the weight value of the i-th moment communication signal, indicating the importance of the signal at this moment in the superposition process, dimensionless; is the signal-to-noise ratio of the i-th moment, with the unit of dB; is the average fading depth, dimensionless; is the sum of the product of the signal-to-noise ratio and the average fading depth of all moments, used for normalization.

[0182] The formula for superimposing the communication signals of multiple moments based on the weight value is:

[0183] ,

[0184] wherein: is the superimposed signal, indicating the synthesized signal; is the weight value of the i-th moment communication signal, dimensionless; is the i-th moment communication signal; indicates the weighted sum of the signals of all moments.

[0185] The communication baseband signal can be extracted from the superimposed signal by band-pass filtering and demodulation. The condition for judging whether the compensation is completed is:

[0186] ,

[0187] wherein: is the signal-to-noise ratio of the superimposed signal, with the unit of dB; is the maximum value of the signal-to-noise ratio of all moments, with the unit of dB; indicates the greater than relationship, judging whether the signal-to-noise ratio of the superimposed signal exceeds the signal-to-noise ratio of all original signals.

[0188] This weighting and superimposition method based on the signal-to-noise ratio and the fading depth is a simplified implementation of the differential geometry optimization method, and is suitable for scenarios with limited computing resources.

[0189] Embodiment 3

[0190] In another embodiment of the present application, the high-low frequency channel cooperative compensation step comprises:

[0191] receiving the received spectrum of the low frequency channel through the high frequency channel;

[0192] obtaining multipath fading information based on the received spectrum;

[0193] correcting transmission errors in the data transmission process of the low frequency channel according to the multipath fading information;

[0194] transmitting signals to the low frequency channel through the high frequency channel;

[0195] adding a training sequence in the low frequency channel;

[0196] Compensate the phase noise of low-frequency channel transmission signal according to the training sequence;

[0197] Construct a high-low frequency joint manifold to describe the coupling characteristics of the dual channels;

[0198] Analyze the correlation and complementarity of fading characteristics in different frequency bands;

[0199] Establish the mapping relationship between frequency bands to achieve complementary enhancement.

[0200] This high-low frequency channel compensation mechanism fully utilizes the complementary advantages of the two frequency bands, further improving the communication reliability. Specifically, the high-frequency channel (5.8 GHz) has high bandwidth and resolution, which can accurately capture channel characteristics; the low-frequency channel (433 MHz) has strong penetration ability and stability, which is suitable as a basic communication guarantee.

[0201] In this embodiment, after the high-frequency channel receives the received spectrum of the low-frequency channel, the multipath fading information is extracted through spectrum analysis. The multipath fading information can be represented by the power delay profile (PDP):

[0202] ,

[0203] Where: PDP is the power delay profile, representing the power distribution of signals at different delay times, with units of W / s; Pi is the power of the ith path, with units of W; Ti is the delay of the ith path, with units of s; N is the number of multipaths, representing the total number of propagation paths; is the Dirac function, which has an infinite value when the parameter is 0, and is 0 otherwise, and the integral is 1.

[0204] According to the extracted multipath fading information, a channel estimation model can be constructed to correct the transmission errors of the low-frequency channel. The channel estimation model can be represented as:

[0205] ,

[0206] Where: H(f) is the channel frequency response, which describes the influence of the channel on different frequency signals, dimensionless; Ai is the amplitude of the ith path, representing the signal strength of the path, dimensionless; Ti is the delay of the ith path, with units of s; f is the frequency, with units of Hz; j is the imaginary unit, satisfying ; e(j2πft) represents the complex exponential function, which describes the phase delay.

[0207] The high-frequency channel transmits a signal to the low-frequency channel, and a training sequence is added in the low-frequency channel, which can help to accurately estimate the phase noise. The training sequence is usually selected to have good autocorrelation characteristics, such as Zadoff-Chu sequence, PN sequence, etc.

[0208] The phase noise compensation can be achieved by the following steps:

[0209] 1. Estimate the phase noise using the training sequence:

[0210] ,

[0211] Wherein: is the phase noise, which represents the random fluctuation of the signal phase, and the unit is radian; is the received signal, which contains amplitude and phase information; is the known training sequence, which is the reference signal; represents the phase angle, which takes the argument of the complex number, and the unit is radian.

[0212] 2. Compensate for the phase noise:

[0213] ,

[0214] Wherein: is the compensated received signal; is the original received signal; is the compensation factor, which is used to offset the phase noise. The high-low frequency joint manifold can be constructed by tensor product space:

[0215] ,

[0216] Wherein: is the joint manifold, which represents the combined characteristics of the high-low frequency channel; is the high-frequency channel manifold, which describes the characteristics of the high-frequency signal; is the low-frequency channel manifold, which describes the characteristics of the low-frequency signal; represents the tensor product, which is a mathematical operation to combine two manifolds.

[0217] On the joint manifold, the correlation and complementarity of the fading characteristics of different frequency bands can be analyzed, and the mapping relationship between the frequency bands can be established to realize complementary enhancement. This joint analysis method can fully exploit the synergistic advantages of high-low frequency channels and improve the overall communication quality.

[0218] Embodiment 4

[0219] In another embodiment of the present application, the adaptive QAM modulation step comprises:

[0220] Obtain communication data, and convert the communication data into a plurality of signals;

[0221] The baseband signal is modulated for the multi-path signal, and the baseband signal is spread spectrum;

[0222] Based on the analysis result of differential manifold, the channel state reference is provided for adaptive QAM modulation of high frequency channel;

[0223] The distribution of signal constellation is automatically adjusted according to the multi-path characteristics;

[0224] The modulation order is reduced in the high curvature area of Riemann manifold;

[0225] The modulation order is increased in the flat area of Riemann manifold;

[0226] The bidirectional information interaction mechanism of modulation system and compensation system is established.

[0227] The adaptive QAM modulation is another innovation point of the application, which is deeply integrated with the multi-path fading compensation method based on differential geometry, and realizes the dynamic matching of channel state and modulation parameters.

[0228] In actual implementation, the communication data is first converted into 4-path signal (in the preferred embodiment), and then the baseband signal is modulated by the QAM modulator. The constellation of QAM modulation can be dynamically adjusted according to the channel state, from QPSK (4QAM) to 256QAM.

[0229] The baseband signal is spread spectrum, which can enhance the anti-interference ability of the signal. The spread spectrum factor is usually selected between 4-16, which depends on the communication environment and bandwidth requirement.

[0230] Based on the analysis result of differential manifold, the channel state reference is provided for adaptive QAM modulation, which is a key innovation of the application. Specifically, according to the curvature distribution of Riemann manifold, the "flat area" and "high curvature area" of the channel can be identified. The flat area corresponds to the part of stable channel state, which is suitable for using high-order modulation (such as 64QAM, 256QAM); the high curvature area corresponds to the part of rapid change of channel state, which should use low-order modulation (such as QPSK, 16QAM).

[0231] The selection of curvature threshold is usually based on empirical value, for example:

[0232] When the curvature is 256QAM;

[0233] When is 64QAM;

[0234] When is 16QAM;

[0235] When is QPSK.

[0236] wherein, denotes the norm of the Riemann curvature tensor, calculated as:

[0237] ,

[0238] wherein: is the norm of the Riemann curvature tensor, a dimensionless quantity that represents a comprehensive measure of the manifold's degree of curvature; are the components of the Riemann curvature tensor; denotes the summation over all indices.

[0239] The bidirectional information interaction mechanism between the modulation system and the compensation system ensures that the two systems can work cooperatively. The modulation system provides signal feature information, such as modulation order and constellation distribution, to the compensation system; the compensation system feeds back channel quality evaluation, such as curvature distribution and multipath characteristics, to the modulation system. This bidirectional interaction mechanism forms a closed-loop control, which can optimize modulation parameters in real time and improve spectral utilization efficiency.

[0240] Embodiment 5

[0241] In another embodiment of the present application, the step of constructing a multi-node communication network comprises:

[0242] The star network is constructed, including a control end, relay nodes, slave nodes and target unmanned aerial vehicles;

[0243] The relay nodes, slave nodes and target unmanned aerial vehicles are connected to the control end respectively;

[0244] Time diversity reception is adopted between any two adjacent nodes;

[0245] The slave nodes send communication signals to the control end;

[0246] When the communication signals arrive at the control end, there are multiple time instants of communication signals in the control end;

[0247] The multiple time instants of communication signals are respectively subjected to QAM modulation to obtain optimal timing of multiple time instants;

[0248] The optimal timing of multiple time instants is adopted to recover the communication signals;

[0249] The control end receives the multi-antenna output signals of the relay nodes;

[0250] The time delay difference of the multi-antenna output signals is compensated and synchronized;

[0251] The compensated signals are superimposed with the received multi-antenna output signals to obtain the communication signals of multiple time instants.

[0252] This multi-node communication network structure supports multi-UAV cooperative tasks and provides communication guarantee for complex task scenarios. In the star network, the control end serves as the central node, responsible for coordinating communication between nodes; relay nodes are used to expand the communication range; slave nodes can be other UAVs or ground stations; target UAVs are the final task executors.

[0253] Timing diversity reception is an important feature of this network, which uses signals received at different times to improve communication reliability. In implementation, the slave node sends a communication signal to the control end, and when the signal reaches the control end, there are four communication signals at different times (S1, S2, S3, S4) in the control end.

[0254] QAM modulation of communication signals at each time can improve spectral efficiency. The acquisition of the optimal timing is usually based on signal quality indicators such as signal-to-noise ratio, bit error rate, etc. Using the optimal timing to recover the communication signal can minimize channel interference.

[0255] When the control end receives the multi-antenna output signal of the relay node, due to the spatial differences between the antennas, there will be a time delay difference. Synchronizing these time delay differences is a key step to ensure correct signal combination. Time delay difference compensation can be achieved by the following method:

[0256] ,

[0257] Where: is the time delay estimation of the ith antenna, indicating the time difference relative to the reference antenna, with the unit of s; is the received signal of the ith antenna, which is a function of time; is the received signal of the reference antenna, usually the antenna with the best signal quality is chosen as the reference; represents the complex conjugate of the reference signal; represents the value of finding the maximum of the following expression; ; represents the integration of time, which calculates the cross-correlation function of the two signals.

[0258] After time delay compensation, the signals of each antenna are combined:

[0259] ,

[0260] Where: is the combined signal, which is a function of time; is the weight coefficient of the ith antenna, usually related to signal quality, satisfying ; is the ith antenna signal after time delay compensation; is the number of antennas; represents the weighted sum of all antennas.

[0261] The multi-node network structure combined with the time diversity reception technology can significantly improve the robustness and coverage of the communication system, and provides reliable communication support for multi-unmanned aerial vehicle cooperative tasks.

[0262] The dual-channel unmanned aerial vehicle line-of-sight communication method provided by the application comprehensively improves the distance, reliability and data transmission capacity of the unmanned aerial vehicle communication system through the intelligent cooperation of high and low frequency channels, the multipath fading compensation based on the differential geometry theory, the task priority driven channel allocation, the high and low frequency cross-band cooperative compensation, the adaptive QAM modulation and the multi-node communication network and the like. The method has a significant breakthrough in theoretical innovation and engineering practicability, provides a new technical path for long-distance and high-reliability communication of unmanned aerial vehicles, and has important application value and popularization prospect.

[0263] The above-mentioned embodiments only express the specific implementation of the application, and the description is more specific and detailed, but it cannot be understood as the limitation of the patent scope of the application. It should be pointed out that for ordinary skilled persons in the art, without departing from the concept of the application, a number of modifications and improvements can be made, which belong to the protection scope of the application.

Claims

1. A two-channel unmanned vehicle line-of-sight communication method, characterized by, The method comprises the following steps: acquiring communication quality; including: setting high-frequency channels and low-frequency channels for data transmission of the target UAV, and acquiring communication quality indicators; the high-frequency channel is a 5.8 GHz frequency band, and the low-frequency channel is a 433 MHz frequency band; the communication quality indicators include signal-to-noise ratio, signal rate and bit error rate; acquiring the best transmission frequency point; including: judging the frequency point quality of the high-frequency channel and the low-frequency channel based on the communication quality indicators, and determining the best transmission frequency point; acquiring the task priority of the UAV; including: determining the task priority type according to the task priority, and selecting the corresponding transmission frequency point to transmit the control instruction based on the task priority type and the best transmission frequency point; transmitting video stream data; including: acquiring the video stream data transmitted by the UAV, and transmitting the video stream data to the target UAV through the best transmission frequency point; performing multipath fading compensation; including: collecting communication signals at multiple time points, mapping the communication signals to a differential manifold space, extracting fading features in the differential manifold space, performing optimization processing on the communication signals based on the fading features, and generating optimized communication signals.

2. The dual channel unmanned line of sight communication method of claim 1, wherein, The method of performing multipath fading compensation specifically comprises: acquiring the signal-to-noise ratio of the communication signals at multiple time points; sorting the signal-to-noise ratios and selecting one signal-to-noise ratio as a reference signal-to-noise ratio; based on the reference signal-to-noise ratio, acquiring the average fading depth of the communication signals at multiple time points; multiplying the signal-to-noise ratio of the communication signal at each time point by the average fading depth to obtain the weight value of the communication signal at each time point; based on the weight value, superimposing the communication signals at multiple time points to obtain a superimposed signal; extracting a communication baseband signal from the superimposed signal and acquiring the signal-to-noise ratio of the communication baseband signal; when the signal-to-noise ratio of the communication baseband signal is greater than the signal-to-noise ratio of the communication signals at multiple time points, it is determined that the multipath fading compensation is completed.

3. The dual channel unmanned line of sight communication method of claim 1, wherein, The method of acquiring the best transmission frequency point specifically comprises: when the acquired frequency point quality indicator of the low-frequency channel is greater than a first preset threshold and the acquired frequency point quality indicator of the high-frequency channel is available, the high-frequency channel is used as the best transmission frequency point at the current time; when the acquired frequency point quality indicator of the low-frequency channel is less than the first preset threshold and the acquired frequency point quality indicator of the high-frequency channel is available, the low-frequency channel is used as the best transmission frequency point at the current time; when the acquired frequency point quality indicator of the low-frequency channel is unavailable, the high-frequency channel and the low-frequency channel are both used as the best transmission frequency point at the current time.

4. The dual channel unmanned line of sight communication method of claim 1, wherein, The method of selecting the corresponding transmission frequency point to transmit the control instruction based on the task priority type and the best transmission frequency point specifically comprises: when the task priority type is the highest priority task, the high-frequency channel or the best transmission frequency point at the current time is used as the transmission frequency point of the highest priority task, and the control instruction is transmitted to the target UAV; when the task priority type is the second highest priority task, the transmission frequency point of the second highest priority task is the one with the higher frequency point quality indicator, and the control instruction is transmitted to the target UAV; When the task priority type is a normal task, the low-frequency channel or the best transmission frequency point at the current moment is taken as a transmission frequency point of the normal task, and a control instruction is transmitted to the target unmanned aerial vehicle.

5. The dual channel unmanned line of sight communication method of claim 1 wherein, The mapping of the communication signal to the differential manifold space specifically includes: a parameter space is constructed, each dimension of the parameter space corresponding to a signal feature at a moment; a metric tensor is defined to establish a distance concept in the parameter space; the parameter space is mapped to a Riemann manifold through local coordinate transformation, keeping the topological structure unchanged; an orthogonal basis is established on the Riemann manifold for subsequent geometric analysis.

6. The dual channel unmanned line of sight communication method of claim 5, wherein, The extraction of the fading feature in the differential manifold space specifically includes: a curvature tensor is calculated on the Riemann manifold to represent the local space bending degree; the propagation characteristics of the signal on the Riemann manifold are calculated through a parallel transmission equation; the principal direction of the curvature tensor is analyzed to identify the main influencing factor of fading; a reference plane is established with a reference signal-to-noise ratio to calculate the geodesic distance of other signal points to the reference plane; the geodesic distance is converted into a fading depth parameter to establish a mapping relationship between geometric quantities and physical quantities; a fading depth distribution diagram is constructed to represent the spatial distribution characteristics of the multipath environment.

7. The dual channel unmanned line of sight communication method of claim 6, wherein, The optimization processing of the communication signal based on the fading feature specifically includes: a geometric weight of the signal at each moment is calculated, the geometric weight being determined by the product of the signal-to-noise ratio and the average fading depth; a geodesic line connecting the signal points is calculated on the Riemann manifold to represent the optimal signal change path; the signal is segmented and interpolated along the geodesic line to keep the phase continuity; a curvature flow equation on the Riemann manifold is constructed to describe the optimal evolution of the Riemann manifold over time; an evolution boundary condition is set to ensure convergence to a physically meaningful solution; an evolution rate is controlled to balance the calculation efficiency and the optimization accuracy; the consistency of the signal phase in the fusion process is ensured through the parallel transmission theory.

8. The dual channel unmanned line of sight communication method of claim 1, wherein, It also includes a high-low frequency channel cooperative compensation step: the receiving spectrum of the low-frequency channel is received through the high-frequency channel; multipath fading information is obtained based on the receiving spectrum; transmission errors in the data transmission process of the low-frequency channel are corrected according to the multipath fading information; signals are transmitted to the low-frequency channel through the high-frequency channel; a training sequence is added in the low-frequency channel; the phase noise of the low-frequency channel transmission signal is compensated according to the training sequence; a high-low frequency joint manifold is constructed to describe the coupling characteristics of the double channels; the correlation and complementarity of fading characteristics of different frequency bands are analyzed; a mapping relationship between frequency bands is established to realize complementary enhancement.

9. The dual channel unmanned line of sight communication method of claim 5, wherein, It also includes an adaptive QAM modulation step: communication data are obtained and converted into multiple signals; baseband signal modulation is performed on the multiple signals, and the baseband signals are spread; based on the differential manifold analysis result, a channel state reference is provided for adaptive QAM modulation of the high-frequency channel; the distribution of the signal constellation diagram is automatically adjusted according to the multipath characteristics; the modulation order is reduced in the high-curvature area of the Riemann manifold; the modulation order is increased in the flat area of the Riemann manifold; a bidirectional information interaction mechanism between the modulation system and the compensation system is established.

10. The dual channel unmanned line of sight communication method of claim 1 wherein, Also comprising a step of constructing a multi-node communication network: Constructing a star network, including a control end, relay nodes, slave nodes and target drones; The relay nodes, slave nodes and target drones are connected to the control end respectively; Time diversity reception is used between any two adjacent nodes; The slave nodes send communication signals to the control end; When the communication signals arrive at the control end, there are multiple time instants of communication signals in the control end; The multiple time instants of communication signals are respectively QAM modulated to obtain multiple time instants of optimal timing; Signal recovery is performed on the communication signals using the multiple time instants of optimal timing; The control end receives the multi-antenna output signals of the relay nodes; The time delay difference of the multi-antenna output signals is compensated and synchronized; The compensated signals are superimposed with the received multi-antenna output signals to obtain multiple time instants of communication signals.

Images