Space-ground electromagnetic spectrum data fusion processing method and system and storage medium

Abstract

The invention provides a sky-ground electromagnetic spectrum data fusion processing method and system and a storage medium, relates to the technical field of multi-source electromagnetic spectrum data fusion, and aims to solve the problems of different space-time references, data redundancy and stray in the sky-ground electromagnetic spectrum monitoring data fusion process. Comprising the following steps: step 1, carrying out time registration on sky-ground electromagnetic spectrum monitoring data; step 2, space registration is carried out on the sky-ground electromagnetic spectrum monitoring data, system equivalent errors are estimated, and system deviation of a base station measurement spectrum value is compensated; and step 3, carrying out space-time fusion of the multi-source spectrum data based on particle filtering of sequential importance sampling, including avoiding particle weight degradation through resampling, updating the confidence coefficient weight of the spectrum data of each base station in an elastic interactive manner based on the residual posterior probability, and finally generating fused sky-ground electromagnetic spectrum data in combination with the confidence coefficient weight.

Description

Technical Field

[0001] The present invention relates to the technical field of multi-source electromagnetic spectrum data fusion processing, and in particular to a method, system and storage medium for ground-based electromagnetic spectrum data fusion processing. Background Art

[0002] In the current rapidly developing information age, the effective management and utilization of electromagnetic spectrum as an important carrier of information transmission is of vital importance to promoting economic and social development. As an important tool for electromagnetic spectrum management, the ground-based spectrum monitoring system realizes comprehensive monitoring of the space electromagnetic environment through the coordinated work of ground-based and space-based equipment.

[0003] Ground-based electromagnetic spectrum monitoring equipment, because it is deployed on the ground or in low-altitude airspace, can obtain high-precision, small-scale electromagnetic spectrum data. These data play an important role in the analysis of the electromagnetic environment in local areas, but due to geographical restrictions, their coverage is limited and it is difficult to fully reflect the wide-area electromagnetic situation. At the same time, ground-based monitoring data may also be affected by environmental factors such as terrain and buildings, resulting in certain limitations in the data. Space-based electromagnetic spectrum monitoring has the advantages of wide coverage and large scale. Through the monitoring payloads carried by space-based platforms such as satellites, real-time monitoring of the global electromagnetic environment can be achieved. However, space-based monitoring data is often less accurate and cannot meet the needs of high-precision electromagnetic environment analysis. In the fusion process of multi-source electromagnetic spectrum data based on the ground, there are still problems such as inconsistent time and space benchmarks, data redundancy and spuriousness, which leads to a decline in data quality and affects the accuracy of the final results.

[0004] Therefore, in order to give full play to the respective advantages of ground-based and space-based spectrum monitoring data, realize the mutual complementation and coordinated use of data, improve the quality and use value of monitoring data, and generate wide-area and refined space electromagnetic situation products, it is very important to study the fusion processing methods of ground-based and space-based multi-source electromagnetic spectrum data. Summary of the invention

[0005] The technical problems to be solved by the present invention are:

[0006] Existing methods are difficult to solve the problems of inconsistent time and space references, data redundancy and spuriousness in the process of ground-based electromagnetic spectrum monitoring data fusion.

[0007] The present invention adopts the following technical solutions to solve the above technical problems:

[0008] The present invention provides a method for fusion processing of ground-to-earth electromagnetic spectrum data, comprising the following steps:

[0009] Step 1: Time registration of ground-based electromagnetic spectrum monitoring data;

[0010] Step 2: spatially register the ground-to-earth electromagnetic spectrum monitoring data, estimate the system equivalent error, and compensate for the system deviation of the spectrum value measured by the base station;

[0011] Step 3: Perform spatiotemporal fusion of multi-source spectrum data based on sequential importance sampling particle filtering, which includes avoiding particle weight degradation through resampling, and elastically and interactively updating the confidence weight of each base station spectrum data based on the residual posterior probability, and finally generating fused spectrum data based on the confidence weight.

[0012] Furthermore, step 1 includes the following steps:

[0013] 1) Perform nonlinear modeling on the ground-based clock source error, measurement speed deviation and data transmission time, and compensate the time dimension of the spectrum data provided by each space-based and ground-based base station;

[0014] 2) Perform nonlinear fitting based on unscented transformation on the spectrum data to be registered to compensate for the registration error;

[0015] 3) Synchronize the spectrum data of different clock periods and sampling frequencies to a unified system alignment period for alignment.

[0016] Furthermore, step 2 includes the following steps:

[0017] 1) Move the space-ground based spectrum data sensing coordinate system to the Earth-centered Earth-fixed coordinate system;

[0018] 2) The data space registration method based on adaptive Kalman filtering dynamically and time-varyingly estimates the equivalent system deviations of different base stations and calculates the spectrum data of spatial registration;

[0019] 3) Estimate the system equivalent error and compensate for the system deviation of the spectrum value measured by the base station.

[0020] Further, step 22) comprises the following steps:

[0021] ① Assume that the ground-to-ground base stations A and B provide measurement results X at time k pA，k and X pB，k , then:

[0022] R tA，k R lA，k X pA，k +X sA，k =R tB，k R lB，k X pB，k +X sB，k

[0023] Among them, R tA，k , R tB，kis the transformation matrix of base station A and B from the base station Cartesian coordinate system to the base station northeast sky coordinate system at time k, R lA，k , R lB，k is the transformation matrix of base station A and B from the northeast sky coordinate system to the earth-centered earth-fixed coordinate system at time k, X sA，k , X sB，k is the coordinate of base station A and B in the Earth-centered Earth-fixed coordinate system at time k, X pA，k , X pB，k is the measurement value of the sensors on base stations A and B at time k after equivalent system deviation compensation. Its specific expression is as follows:

[0024]

[0025] ② The azimuth and elevation angle equivalent system deviation change rates are used as the state quantities to be estimated for filtering operations, and the state equation is constructed as follows:

[0026] Δ k+1 =ψΔ k +Γw k

[0027] The observation equation is:

[0028] Z k =H k Δ k +v k

[0029] Among them, Δ k =[Δ Ak , Δ Bk ] T , w k is the system state transfer noise, the transfer matrix Noise Matrix

[0030] Get Z k =R tA R lA X pA +X sA -(R tB R lB X pB +X sB );

[0031] Among them, the measurement matrix is ​​H k =[-R tA R lA X pA +X sA , R tB R lB X pB +X sB ], and vk is the observed quantity measurement noise.

[0032] Further, step 23) comprises the following steps:

[0033] 1) Make a one-step prediction for the state quantity to be estimated:

[0034] Δ k+1|k =ΨΔ k|k

[0035] 2) One-step prediction of the error covariance matrix of the state:

[0036] P k+1|k =ψP k|k Ψ T +ΓQ k Γ T

[0037] 3) Calculate the gain matrix of the state quantity:

[0038]

[0039] 4) Update the state quantity in one step:

[0040] Δ k+1|k+1 =Δ k+1|k +K k+1 (Z k+1 -H k+1 Δ k+1|k )

[0041] 5) Update the state error covariance matrix in one step:

[0042]

[0043] 6) Adaptive adjustment of Q and R matrices:

[0044]

[0045] 7) The system equivalent error obtained by filtering is used to compensate the system deviation of the spectrum value measured by the base station to obtain the spectrum data X of base stations A and B after spatial registration at any time k ecef，A and X ecef，B for:

[0046]

[0047] Further, step 3 includes the following steps:

[0048] 1) Initialize the particle swarm;

[0049] At time period K = 0, from the prior distribution p(X 0) to extract a set of samples as the initial particle set Assign each particle a priori confidence weight expressed as probability Then use the same prior filter weights Initialize each particle;

[0050] 2) Sampling is performed sequentially based on the importance of particle sets;

[0051] In the Kth time period, a particle set sequential sampling is performed to estimate the conditional expectation:

[0052] E[g(X k )|Z 1：k ]=∫g(x k )·p(x k |z 1：k )dx k

[0053] From the probability density function q(x k |z 1：k ) to obtain the sample set z k , and through z k Approximate estimate E[g(X k )|Z 1：k ]:

[0054]

[0055] in,

[0056] The recursive form of the distribution is:

[0057]

[0058] Assume that the distribution q has the following recursive form:

[0059] q(x 0：k |z 1：k )=q(x k |x 0：k-1 , z 1：k )q(x 0：k-1 |z 1：k-1 )

[0060] Then the original batch distribution is converted into a sequential sampling form of a single particle:

[0061]

[0062] 3) Update the particle importance weights sequentially;

[0063] Substituting the sequential sampling form of a single particle into the expected calculation formula, we can obtain the recursive update form of the importance weight:

[0064]

[0065] Further,

[0066] q(x k |x 0：k-1 , z 1：k )=q(x k |x k-1 , z k )

[0067] The final recursive update form of the importance weight is:

[0068]

[0069] 4) Sampling the particle set weight

[0070] The effective number of particles is used to measure the degree of degradation of particle weights:

[0071]

[0072] When the number of valid particles reaches the threshold, the particle set Perform resampling;

[0073] 5) Estimate and update the target position parameter state;

[0074] The fusion estimate of the spectrum data at time k is calculated using a weighted method:

[0075]

[0076] in Confidence is assigned to the target multipath distribution;

[0077] 6) Perform interactive weight update to generate fused spectrum data;

[0078] Estimation based on fused spectrum data Calculate the residual probability function for each spectral data particle:

[0079]

[0080] Feed the function back to the particle set, repeat steps 2)-7), update the particle fusion weight; finally, the fusion result of the ground-ground spectrum data is obtained.

[0081] Furthermore, step 3 also includes normalizing the weights updated in step 3):

[0082]

[0083] Furthermore, in step 34), polynomial resampling is used to resample the particle set Resampling includes the following steps:

[0084] First, calculate the cumulative distribution of the normalized weights of the particles:

[0085]

[0086] Then, generate random numbers u~U(0,1) that obey the uniform distribution [0,1], find the position of the random number in the cumulative distribution, and get the index i of the original particle corresponding to the resampled particle, i satisfies:

[0087]

[0088] The present invention provides a ground-to-earth electromagnetic spectrum data fusion processing system, which has a program module corresponding to the steps of the method described in any one of the above technical solutions, and executes the steps in the above ground-to-earth electromagnetic spectrum data fusion processing method when running.

[0089] The present invention provides a computer-readable storage medium, which stores a computer program. The computer program is configured to implement the steps in the ground-based electromagnetic spectrum data fusion processing method described in any one of the above-mentioned technical solutions when called by a processor.

[0090] Compared with the prior art, the present invention has the following beneficial effects:

[0091] The present invention provides a method, system and storage medium for fusion processing of ground-to-earth electromagnetic spectrum data. First, the data is time-aligned and space-aligned according to the characteristics of ground-to-earth electromagnetic spectrum monitoring data. Then, a flexible interactive spatiotemporal spectrum data fusion method based on sequential importance sampling particle filtering is used to adaptively monitor regional environmental changes, and the particle degradation problem is avoided by a resampling method. The confidence weight of each base station spectrum data is elastically and interactively updated based on the residual posterior probability, and finally the consistency of the use of ground-to-earth electromagnetic spectrum data is achieved. The present invention uses the spectrum monitoring data after fusion processing to generate wide-area and refined electromagnetic situation products. BRIEF DESCRIPTION OF THE DRAWINGS

[0092] Figure 1 A schematic diagram of a time registration method in an embodiment of the present invention;

[0093] Figure 2 Schematic diagram of the traceless transformation principle in an embodiment of the present invention;

[0094] Figure 3 is a schematic diagram of a spatial registration method in an embodiment of the present invention;

[0095] Figure 4Schematic diagram of the Earth-centered Earth-fixed coordinate system in an embodiment of the present invention;

[0096] Figure 5 is a flow chart of an adaptive Kalman filter algorithm in an embodiment of the present invention;

[0097] Figure 6 This is a flow chart of a multi-source spectrum data fusion method in an embodiment of the present invention;

[0098] Figure 7 Schematic diagram of the particle filter resampling method in an embodiment of the present invention. DETAILED DESCRIPTION

[0099] In order to enable those skilled in the art to better understand the scheme of the present invention, exemplary implementations or embodiments of the present invention will be described below in conjunction with the accompanying drawings. Obviously, the described implementations or embodiments are only implementations or embodiments of a part of the present invention, not all of them. Based on the implementations or embodiments of the present invention, all other implementations or embodiments obtained by ordinary technicians in the field without creative work should fall within the scope of protection of the present invention.

[0100] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0101] Specific implementation scheme 1: The present invention provides a method for fusion processing of ground-to-earth based electromagnetic spectrum data, comprising the following steps:

[0102] Step 1: Time registration of ground-based electromagnetic spectrum monitoring data;

[0103] Step 2: spatially register the ground-to-earth electromagnetic spectrum monitoring data, estimate the system equivalent error, and compensate for the system deviation of the spectrum value measured by the base station;

[0104] Step 3: Perform spatiotemporal fusion of multi-source spectrum data based on sequential importance sampling particle filtering, in which particle weight degradation is avoided by resampling, and the confidence weight of each base station spectrum data is elastically and interactively updated based on the residual posterior probability, and finally the fused spectrum data is generated by combining the confidence weight.

[0105] Specific implementation plan 2: Figure 1 and Figure 2 As shown, step 1 includes the following steps:

[0106] 1) Perform nonlinear modeling on the ground-based clock source error, measurement speed deviation and data transmission time, and compensate the time dimension of the spectrum data provided by each space-based and ground-based base station;

[0107] 2) Perform nonlinear fitting based on untraceable transformation on the spectrum data to be registered to compensate for the registration error caused by the nonlinear factors of environmental noise;

[0108] 3) A spline interpolation-based registration method is used to synchronize the spectrum data of different clock periods and sampling frequencies to a unified system registration period for registration.

[0109] The rest of this embodiment is the same as the first embodiment.

[0110] Specific implementation plan three: Figure 3 and Figure 4 As shown, step 2 includes the following steps:

[0111] 1) Move the space-ground based spectrum data sensing coordinate system to the Earth-centered Earth-fixed coordinate system;

[0112] 2) The data space registration method based on adaptive Kalman filtering dynamically and time-varyingly estimates the equivalent system deviations of different base stations and calculates the spectrum data for accurate spatial registration;

[0113] 3) Estimate the system equivalent error and compensate for the system deviation of the spectrum value measured by the base station;

[0114] The rest of this embodiment is the same as the first embodiment.

[0115] Specific implementation plan 4: Figure 5 As shown, step 22) includes the following steps:

[0116] ① Assume that the ground-to-ground base stations A and B provide measurement results X at time k pA，k and X pB，k , then:

[0117] R tA，k R lA，k X pA，k +X sA，k =R tB，k R lB，k X pB，k +X sB，k

[0118] Among them, R tA，k , R tB，k is the transformation matrix of base station A and B from the base station Cartesian coordinate system to the base station northeast sky coordinate system at time k, R lA，k , R lB，k is the transformation matrix of base station A and B from the northeast sky coordinate system to the earth-centered earth-fixed coordinate system at time k, X sA，k , X sB，k is the coordinate of base station A and B in the Earth-centered Earth-fixed coordinate system at time k, X pA，k , X pB，kIt is the measurement value of sensors A and B on base stations A and B at time k after equivalent system deviation compensation. Its specific expression is as follows:

[0119]

[0120] ② The azimuth and elevation angle equivalent system deviation change rates are used as the state quantities to be estimated for filtering operations, and the state equation is constructed as follows:

[0121] Δ k+1 =ψΔ k +Γw k

[0122] The observation equation is:

[0123] Z k =H k Δ k +v k

[0124] Among them, Δ k =[Δ Ak , Δ Bk ] T , w k is the system state transfer noise, the transfer matrix Noise Matrix

[0125] Get Z k =R tA R lA X pA +X sA -(R tB R lB X pB +X sB );

[0126] Among them, the measurement matrix is ​​H k =[-R tA R lA X pA +X sA , R tB R lB X pB +X sB ], and v k is the observed quantity measurement noise.

[0127] The rest of this implementation plan is the same as the specific implementation plan three.

[0128] Specific implementation scheme five: Step 23) includes the following steps:

[0129] 1) One-step prediction of the state quantity to be estimated:

[0130] Δk+1|k =ΨΔ k|k

[0131] 2) One-step prediction of the error covariance matrix of the state:

[0132] P k+1|k =ΨP k|k Ψ T +ΓQ k Γ T

[0133] 3) Calculate the gain matrix of the state quantity:

[0134]

[0135] 4) Update the state quantity in one step:

[0136] Δ k+1|k+1 =Δ k+1|k +K k+1 (Z k+1 -H k+1 Δ k+1|k )

[0137] 5) Update the state error covariance matrix in one step:

[0138]

[0139] 6) Adaptive adjustment of Q and R matrices:

[0140]

[0141] 7) The system equivalent error obtained by filtering is used to compensate the system deviation of the spectrum value measured by the base station to obtain the spectrum data X of base stations A and B after spatial registration at any time k ecef，A and X ecef，B for:

[0142]

[0143] The rest of this implementation plan is the same as the specific implementation plan four.

[0144] Specific implementation plan six: Figure 6 As shown, step 3 includes the following steps:

[0145] 1) Initialize the particle swarm;

[0146] At time period K = 0, from the prior distribution p(X 0 ) to extract a set of samples as the initial particle set Each particle has base station coordinates and spectrum intensity state variables, representing a possible distribution state of spectrum data in the system. Each particle is given a priori confidence weight expressed as probability. Indicates how well it matches the actual state of the system. Then use the same prior filter weights Initialize each particle;

[0147] 2) Sampling is performed sequentially based on the importance of particle sets;

[0148] In the Kth time period, a particle set sequential sampling is performed in the current system to estimate the conditional expectation:

[0149] E[g(X k )|Z 1：k ]=∫g(x k )·p(x k |z 1：k )dx k

[0150] In reality, due to the non-cooperative environment, p(x k |z 1：k ) may be difficult to sample directly, or even impossible to sample at all. In this case, we can choose to sample from a probability density function q(x k |z 1：k ) to obtain the sample set z by random Monte Carlo sampling k , and through z k Approximate estimate E[g(X k )|Z 1：k ], the specific calculation method is as follows:

[0151]

[0152] in,

[0153] According to the Bayesian formula, the Markov state independence assumption and the Markov observation independence assumption, the recursive form of the distribution is:

[0154]

[0155] Assume that the distribution q has the following recursive form:

[0156] q(x 0：k |z 1：k )=q(x k |x 0：k-1 , z 1：k )q(x 0：k-1 |z 1：k-1 )

[0157] Then the original batch distribution is converted into a sequential sampling form of a single particle:

[0158]

[0159] 3) Update the particle importance weights sequentially;

[0160] Substituting the sequential sampling form of a single particle into the expected calculation formula, we can obtain the recursive update form of the importance weight:

[0161]

[0162] Since this implementation scheme only cares about the estimation result of the current k time period state, according to the Markov characteristics of the system, further:

[0163] q(x k |x 0：k-1 , z 1：k )=q(x k |x k-1 , z k )

[0164] The final recursive update form of the importance weight is:

[0165]

[0166] 4) Sampling the particle set weight

[0167] In the process of applying sequential particle filtering to the fusion of multi-source spectrum data of ground-based and space-based, there is a problem of particle weight degradation. After several iterations, the weights of most particles will become so small that they can be ignored, and only a few particles have larger weights. This causes the variance of particle weights to grow over time, and the number of valid particles in the state space is small. As the number of invalid sampled particles increases, a large amount of calculation is wasted on particles that have almost no effect on estimating the posterior filter probability distribution, which reduces the estimation performance.

[0168] The effective number of particles is used to measure the degree of degradation of particle weights:

[0169]

[0170] like Figure 7 As shown in the figure, when the number of effective particles reaches the threshold, the particle set Resampling is performed to suppress particle weight degradation. Resampling refers to sampling without replacement from the original sample according to the particle weight. Particles with high weights are more likely to be selected, and particles with low weights are less likely to be selected.

[0171] 5) Estimate and update the target position parameter state;

[0172] Combine the resampled particle set with the corresponding target multipath distribution confidence To update the estimated value of the parameter state at the target position, a weighted method is used to calculate the fusion estimate of the spectrum data at time k:

[0173]

[0174] 6) Perform interactive weight update to generate fused spectrum data;

[0175] Estimation based on fused spectrum data Calculate the residual probability function for each spectral data particle:

[0176]

[0177] Feedback the likelihood function to the particle set, and update the particle fusion weight at time k+1 according to Bayes’ theorem Repeat steps 2)-7) to update the particle fusion weights and finally obtain the fusion result of the ground-ground spectrum data.

[0178] The rest of this implementation plan is the same as the specific implementation plan five.

[0179] Specific implementation scheme seven: Step 3 also includes normalizing the weights updated in step 3):

[0180]

[0181] The rest of this implementation plan is the same as specific implementation plan six.

[0182] Specific implementation scheme eight: In step 34), polynomial resampling is used to resample the particle set Resampling is performed, and the probability of particles with higher weights being resampled is increased through polynomial resampling. First, the cumulative distribution of particle normalization weights is calculated, which includes the following process:

[0183] First, calculate the cumulative distribution of the normalized weights of the particles:

[0184]

[0185] Then, generate random numbers u~U(0,1) that obey the uniform distribution [0,1], use binary search to find the position of the random number in the cumulative distribution, and get the index i of the original particle corresponding to the resampled particle, i satisfies:

[0186]

[0187] The new particle set obtained by polynomial resampling can avoid particle weight degradation and is more concentrated in the area that fits the measured data better.

[0188] The rest of this implementation plan is the same as Specific Implementation Plan Seven.

[0189] A method (algorithm) for fusion processing of ground-to-earth electromagnetic spectrum data proposed in the present invention is the underlying technical core of the present invention, and various products can be derived based on the algorithm.

[0190] Based on the method proposed in the present invention, a ground-to-earth electromagnetic spectrum data fusion processing system is developed using a programming language. The system has program modules corresponding to the steps of the above-mentioned technical solution, and executes the steps in the above-mentioned ground-to-earth electromagnetic spectrum data fusion processing method during operation.

[0191] The computer program of the developed system (software) is stored on a computer-readable storage medium, and the computer program is configured to implement the steps of the above-mentioned method for fusion processing of ground-to-earth electromagnetic spectrum data when called by a processor, that is, the present invention is materialized on a carrier to become a computer program product.

[0192] Various implementations of the systems and techniques described herein can be realized in digital electronic circuit systems, integrated circuit systems, dedicated ASICs (application specific integrated circuits), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0193] The computer programs (also referred to as programs, software, software applications, or codes) of the present invention include machine instructions for programmable processors, and these computer programs can be implemented using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. As used herein, the terms "machine-readable medium" and "computer-readable medium" refer to any computer program product, device, and / or device (e.g., disk, optical disk, memory, programmable logic device PLD) for providing machine instructions and / or data to a programmable processor, including a machine-readable medium that receives machine instructions as machine-readable signals. The term "machine-readable signal" refers to any signal for providing machine instructions and / or data to a programmable processor.

[0194] Although the present invention is disclosed as above, the protection scope of the present invention is not limited thereto. Those skilled in the art may make various changes and modifications without departing from the spirit and scope of the present invention, and these changes and modifications will fall within the protection scope of the present invention.

Claims

1. A method for fusion processing of ground-to-earth electromagnetic spectrum data, characterized in that: The steps include: Step 1: Time registration of ground-based electromagnetic spectrum monitoring data; Step 2: spatially register the ground-to-earth electromagnetic spectrum monitoring data, estimate the system equivalent error, and compensate for the system deviation of the spectrum value measured by the base station; Step 3: Perform spatiotemporal fusion of multi-source spectrum data based on sequential importance sampling particle filtering, which includes avoiding particle weight degradation through resampling, and elastically and interactively updating the confidence weight of each base station spectrum data based on the residual posterior probability, and finally generating fused spectrum data based on the confidence weight.

2. The method for fusion processing of ground-to-earth electromagnetic spectrum data according to claim 1, characterized in that: Step 1 includes the following steps: 1) Perform nonlinear modeling on the ground-based clock source error, measurement speed deviation and data transmission time, and compensate the time dimension of the spectrum data provided by each space-based and ground-based base station; 2) Perform nonlinear fitting based on unscented transformation on the spectrum data to be registered to compensate for the registration error; 3) Synchronize the spectrum data of different clock periods and sampling frequencies to a unified system alignment period for alignment.

3. The method for fusion processing of ground-to-earth electromagnetic spectrum data according to claim 1, characterized in that: Step 2 includes the following steps: 1) Move the space-ground based spectrum data sensing coordinate system to the Earth-centered Earth-fixed coordinate system; 2) The data space registration method based on adaptive Kalman filtering dynamically and time-varyingly estimates the equivalent system deviations of different base stations and calculates the spectrum data of spatial registration; 3) Estimate the system equivalent error and compensate for the system deviation of the spectrum value measured by the base station.

4. The method for fusion processing of ground-to-earth electromagnetic spectrum data according to claim 3, characterized in that: Step 22) comprises the following steps: ① Assume that the ground-to-ground base stations A and B provide measurement results X at time k pA，k and X pB，k , then: R tA，k R lA，k X pA，k +X sA，k =R tB，k R lB，k X pB，k +X sB，k Among them, R tA，k , R tB，k is the transformation matrix of base station A and B from the base station Cartesian coordinate system to the base station northeast sky coordinate system at time k, R lA，k , R lB，k is the transformation matrix of base station A and B from the northeast sky coordinate system to the earth-centered earth-fixed coordinate system at time k, X sA，k , X sB，k is the coordinate of base station A and B in the Earth-centered Earth-fixed coordinate system at time k, X pA，k , X pB，k is the measurement value of the sensors on base stations A and B at time k after equivalent system deviation compensation. Its specific expression is as follows: ② The azimuth and elevation angle equivalent system deviation change rates are used as the state quantities to be estimated for filtering operations, and the state equation is constructed as follows: D k+1 =ψΔ k +Γw k The observation equation is: Z k =H k Δ k +v k Among them, Δ k =[Δ Ak , Δ Bk ] T , w k is the system state transfer noise, the transfer matrix Noise Matrix Get Z k =R tA R lA X pA +X sA -(R tB R lB X pB +X sB ); Among them, the measurement matrix is ​​H k =[-R tA R lA X pA +X sA , R tB R lB X pB +X sB ], and v k is the observed quantity measurement noise.

5. The method for fusion processing of ground-to-earth electromagnetic spectrum data according to claim 4, characterized in that: Step 23) comprises the following steps: 1) One-step prediction of the state quantity to be estimated: D k+1|k =PSD k|k 2) One-step prediction of the error covariance matrix of the state: P k+1|k =ΨP k|k P T +GQ k C T 3) Calculate the gain matrix of the state quantity: 4) Update the state quantity in one step: D k+1|k+1 =D k+1|k +K k+1 (Z k+1 -H k+1 D k+1|k ) 5) Update the state error covariance matrix in one step: 6) Adaptive adjustment of Q and R matrices: 7) The system equivalent error obtained by filtering is used to compensate the system deviation of the spectrum value measured by the base station to obtain the spectrum data X of base stations A and B after spatial registration at any time k ecef，A and X ecef，B for:

6. The method for fusion processing of ground-to-earth electromagnetic spectrum data according to claim 5, characterized in that: Step 3 includes the following steps: 1) Initialize the particle swarm; When the time period K = 0, a set of samples is extracted from the prior distribution p(X0) of the spectrum data of each base station as the initial particle set Assign each particle a priori confidence weight expressed as probability Then use the same prior filter weights Initialize each particle; 2) Sampling is performed sequentially based on the importance of particle sets; In the Kth time period, a particle set sequential sampling is performed to estimate the conditional expectation: E[g(X k )|Z 1：k ]=∫g(x k )·p(x k |z 1：k )dx k From the probability density function q(x k |z 1：k ) to obtain the sample set z k , and through z k Approximate estimate E[g(X k )|Z 1：k ]: in, The recursive form of the distribution is: Assume that the distribution q has the following recursive form: q(x 0：k |z 1：k )=q(x k |x 0：k-1 ,z 1：k )q(x 0：k-1 |z 1：k-1 ) Then the original batch distribution is converted into a sequential sampling form of a single particle: 3) Update the particle importance weights sequentially; Substituting the sequential sampling form of a single particle into the expected calculation formula, we can obtain the recursive update form of the importance weight: Further, q(x k |x 0：k-1 ,z 1：k )=q(x k |x k-1 ,z k ) The final recursive update form of the importance weight is: 4) Sampling the particle set weight The effective number of particles is used to measure the degree of degradation of particle weights: When the number of valid particles reaches the threshold, the particle set Perform resampling; 5) Estimate and update the target position parameter state; The fusion estimate of the spectrum data at time k is calculated using a weighted method: in Confidence is assigned to the target multipath distribution; 6) Perform interactive weight update to generate fused spectrum data; Estimation based on fused spectrum data Calculate the residual probability function for each spectral data particle: Feed the function back to the particle set, repeat steps 2)-7), update the particle fusion weight; finally, the fusion result of the ground-ground spectrum data is obtained.

7. The method for fusion processing of ground-to-earth electromagnetic spectrum data according to claim 6, characterized in that: Step 3 also includes normalizing the weights updated in step 3):

8. The method for fusion processing of ground-to-earth electromagnetic spectrum data according to claim 7, characterized in that: In step 34), polynomial resampling is used to resample the particle set Resampling includes the following steps: First, calculate the cumulative distribution of the normalized weights of the particles: Then, generate random numbers u~U(0,1) that obey the uniform distribution [0,1], find the position of the random number in the cumulative distribution, and get the index i of the original particle corresponding to the resampled particle, i satisfies:

9. A ground-to-earth electromagnetic spectrum data fusion processing system, characterized in that: The system has a program module corresponding to the steps of the method described in any one of claims 1 to 8, and executes the steps in the above-mentioned ground-to-earth based electromagnetic spectrum data fusion processing method when running.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and the computer program is configured to implement the steps in the ground-to-earth based electromagnetic spectrum data fusion processing method according to any one of claims 1 to 8 when called by a processor.

Images