Channel state estimation method based on back scattering communication, and noise source positioning method and system

Abstract

The invention relates to a channel state estimation method based on back scattering communication, and a noise source positioning method and system. The method comprises the following steps: identifying and extracting a single scattering path signal from various scattering paths from a radio frequency source sending a signal to a receiver receiving the signal; sampling a single scattering path signal, and constructing a discrete time function model of the single scattering path receiving signal; and solving the model to obtain the amplitude and phase of the signal, reconstructing the received signal, and analyzing the frequency domain through Fourier transform to realize channel state estimation. Performing covariance matrix characteristic decomposition on the single scattering path signal to determine a noise subspace; and constructing a spectrum peak search function, searching a spectrum peak, estimating a direction of arrival, and calculating the position of a noise source. According to the method, the interference of the multipath effect on channel estimation can be reduced by separating the single scattering signal, a more accurate channel model can be established and accurate noise source positioning can be carried out without a complex algorithm, and the communication quality of an AmBC system can be improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of environmental scattering communication, and more specifically, relates to a channel state estimation method based on scattering communication, a noise source positioning method and a system. Background Art

[0002] In order to meet the needs of efficient perception and communication of the Internet of Things, substations are equipped with ambient backscatter communication (AmBC) systems to ensure the digital and intelligent transformation of traditional substations. However, it is not easy to build an AmBC communication system in a complex substation environment. Factors such as reflection and shielding of obstacles in the substation, electromagnetic interference and special noise generated by high-voltage equipment have greatly affected the operation of the AmBC communication system, posing challenges to the construction of the AmBC communication system in the substation. The channel characteristics in the substation are complex and changeable, and it is difficult to establish an accurate channel model. Multiple scattering causes signal distortion, which causes large errors in subsequent signal processing and causes the receiving end to repeatedly recognize the signal, reducing the effectiveness of the AmBC communication system; the special noise generated by the operation of high-voltage equipment such as switch operation and transformers interferes with the normal operation of the AmBC communication system, and abnormal phenomena such as equipment failure are also accompanied by special noise. These abnormal phenomena may cause power outages, equipment damage and other consequences, and even affect the normal operation of key industries and facilities. Therefore, in order to improve the effectiveness and reliability of the AmBC communication system in the substation scenario, it is urgent to perform channel state estimation to obtain a more accurate channel model and locate the noise source that generates special noise.

[0003] In terms of channel state estimation, a patent with publication number CN110099017A has proposed a channel estimation method based on a hybrid quantization system of a deep neural network. The channel state estimation is performed with the help of random pilot combined with reinforcement learning, so as to achieve accurate mapping of antennas with different precisions. However, the random pilot sequence needs to rely on the synchronization algorithm to ensure the accurate identification and demodulation of the sequence, and it is necessary to ensure that the energy of the pilot sequence is large enough to be received and estimated in the channel, which will lead to complex synchronization algorithms and high power consumption problems. Reinforcement learning algorithms usually require a large number of training samples and iterations to obtain ideal results. In the substation scenario, the channel state changes frequently, and there are problems such as long training time and high consumption of computing resources.

[0004] Existing research on substation wireless channels focuses on measuring channel characteristic parameters and modeling the channels. The methods are based on random geometric modeling, artificial intelligence algorithms, etc. One of the above two methods requires a synchronization algorithm between the transmitter and the receiver, which is highly complex and consumes a lot of power; the other requires a long training time and consumes a lot of computing resources in the frequent channel state changes. The existing channel estimation technology is not suitable for substation environments.

[0005] In terms of noise source positioning, the patent with publication number CN117647774A proposes a noise source positioning method, which locates the noise source by installing collection devices such as microphones and cameras and performing analysis and processing. This patent processes the array input signal and estimates the direction of arrival (DOA) of the noise wave. However, the noise in the substation is more complex. The power signals in the substation usually belong to the low-frequency signal range, such as 50Hz or 60Hz AC signals. These signals are mainly used for power transmission and control. There are multiple non-ideal scattering points in the substation scene, which leads to multiple scattering. The scattered signal interference is serious and distorted, which wastes the calculation amount and is not conducive to the measurement results. In addition, the current positioning method based on physical equipment may be interfered with in complex substation scenes and requires high installation and maintenance costs. Summary of the invention

[0006] In order to solve the problems in the prior art of complex metal equipment and structure, large path loss attenuation, serious interference of multiple scattered signals, difficult channel modeling, complex algorithm, long training time, large consumption of computing resources, serious interference and distortion of multiple scattered signals in noise source positioning, waste of computing power, and unsuitability for substation scenarios, the present invention provides a channel state estimation method and a noise source positioning method and system based on post-scatter communication, which identify and extract single scattering path signals from various complex scattering paths; model and solve the single scattering path based on a set algorithm to obtain the amplitude and phase of the received signal, reconstruct the received signal, and realize channel state estimation by Fourier transform analysis of the frequency domain representation to deduce the delay, assist in channel modeling, and improve the effectiveness of calculation, and determine the noise subspace, calculate the position of the noise source by searching the spectrum peak and estimating the wave arrival direction.

[0007] A first aspect of the present invention provides a method for estimating an environmental backscatter communication channel state. The environmental backscatter communication system includes a radio frequency source, a stationary scatterer and a receiver. The channel state estimation method includes the following steps:

[0008] Step 1, identifying and extracting a single scattering path signal from various scattering paths from when a radio frequency source sends a signal to when a receiver receives a signal;

[0009] Step 2, sampling the single scattering path signal and constructing a discrete time function model of the single scattering path received signal;

[0010] Step 3, solve the model in step 2 to obtain the amplitude and phase of the received signal, reconstruct the received signal, and use Fourier transform analysis to deduce the delay in the frequency domain to achieve channel state estimation.

[0011] Preferably, in step 1, various scattering paths from when a radio frequency source sends a signal to when a receiver receives the signal include multiple scattering paths and single scattering paths;

[0012] Among them, multiple scattering paths refer to: after the signal is emitted from the RF source, it reaches the receiver through any strong scatterer outside the scatterer, or in addition to being scattered by the scatterer, it also reaches the receiver after being reflected / scattered by the strong scatterer;

[0013] A single scattering path means that the signal is sent from the RF source directly through the scatterer to the receiver.

[0014] Preferably, in step 1, the single scattering path signal is identified and extracted in the following manner:

[0015] 1.1 Calculate the possible mobile station position line LPMD for various scattering paths;

[0016] 1.2 Calculate the central reference point of all paths;

[0017] 1.3 Calculate the distance from the central reference point to each LPMD line, and select a decision threshold. Identify the path corresponding to the path below the decision threshold as a single scattering path, otherwise, identify it as a multiple scattering path.

[0018] Preferably, the possible mobile station position line LPMD means: the multiple scattering path is equivalent to a single scattering path scattered by a virtual scatterer. In the single scattering, the transmission angle of the signal transmitted by the RF source is known, the arrival time TOA and the arrival angle AOA of the signal are measured, and the path length is calculated from the arrival time. Then the scatterer position must be at a position with the receiver as the starting point, the arrival angle as the argument and the length less than the path length. If the distance between the scatterer and the receiver is determined, the position of the RF source can be determined; while the position of the virtual scatterer is uncertain, but the TOA values ​​of the signal arrival time under different paths and the positions of the RF source, scatterer and receiver satisfy the formula

[0019]

[0020] in, Indicates the distance between the RF source and the scatterer; represents the distance between the scatterer and the receiver; R represents the length of the path, c0 is the speed of light, and t represents the TOA value measured by the RF source. From this formula, as well as the measured signal arrival time TOA and signal arrival angle AOA, all possible positions of the RF source are obtained by geometric methods. The possible positions are connected together to form a straight line. This line of the mth scattering path is called the possible mobile station position line of the mth signal path. The straight line expression can be recorded as ax m +by m +c=0.

[0021] Preferably, in 1.2, the central reference point of all paths is calculated in the following manner:

[0022] 1.2.1 Use all paths to perform a preliminary fit on the position coordinates of the RF source, and the fit is in represents the final estimated position of the RF source; X = [x, y] T Indicates the actual location coordinates of the RF source; X k represents the RF source position coordinates corresponding to k paths; r k Indicates the length of the kth path, which is equal to the TOA measurement value multiplied by the propagation speed of radio waves; ||XX k || represents the Euclidean distance between the real RF source and the RF source corresponding to the i-th path; Q represents the set of all paths in the current positioning;

[0023] 1.2.2 Calculate the weight of each path signal

[0024] Compute RF source position estimates for each path And calculate the residual variable for each path Divide it by the kth signal Q k The length of is normalized, and the normalized inverse is used as the weight λ of the kth path signal k ;

[0025] 1.2.3 Recalculate the final estimated position of the RF source after weighting And calculate the residual variable Final estimated position Has the smallest positioning residual;

[0026]

[0027] in, The RF source position estimation of the kth path, Q k is the kth path signal;

[0028] Determine the final estimated position If there is no convergence, return to 1.2.2 to adjust the weights again. If there is convergence, the final estimated position is As the middle reference point T(x T ,y T ).

[0029] Preferably, the specific process of 1.3 is: calculating the distance from the intermediate reference point T to each LPMD line, the formula is:

[0030]

[0031] In the above formula, dis m represents the distance from the intermediate reference point T to the LPMD line corresponding to the mth path, a m , b m 、c m Represents the LPMD line equation ax corresponding to the mth path m +by m +c=0 in the parameter, x T ,y T Respectively represent the horizontal and vertical coordinates of the middle reference point T;

[0032] Then the distance weight d from the middle reference point T to the mth LPMD line is calculated m :

[0033]

[0034] In the above formula, M represents the total number of paths, dis k It represents the distance from the intermediate reference point T to the LPMD line corresponding to the kth path. The single scattering path signal is finally identified and extracted by judging the weight and the preset threshold d. The specific judgment expression is:

[0035]

[0036] Wherein, d is the preset threshold.

[0037] Preferably, in step 2, the received signal is modeled as a set of p exponential functions with amplitude, phase, frequency and attenuation factor, and its discrete time function model is expressed as follows:

[0038]

[0039] Among them, b i =A i exp(jθ i ), z i =exp[(α i +j2πf i )Δt], As an approximation of y(n), its error is defined as y(n) is the expression of the nth received signal actually received, A i is the amplitude of the i-th exponential function in the set of functions in the model, θ i is the phase of the i-th exponential function in the set of functions in the model, α i is the attenuation factor of the i-th exponential function in the set of functions in the model, f i is the oscillation frequency of the i-th exponential function in the group of functions in the model, Δt is the sampling interval, p is the order, n represents the n-th discrete signal, and N is the total number of discrete signals.

[0040] Preferably, in step 3, solving the amplitude and phase of the received signal and reconstructing the received signal includes the following:

[0041] 3.1 Construct the discrete time function recursive difference equation of the single scattering path signal and solve the equation coefficients;

[0042] 3.2 Use the coefficients of the difference equation to construct the system function and transform the discrete time function form into a linear equation about the unknown parameters;

[0043] 3.3 Solve the linear equation parameters according to the matrix characteristics, solve the amplitude and phase according to the parameters, and reconstruct the signal.

[0044] Preferably, in 3.1, the discrete-time differential equation of the single scattering path signal is constructed in the following manner, and the coefficients of the equation are solved:

[0045] Construct the cost function ε to represent the sum of squared errors: When the sampling interval Δt = 1, Satisfies the recursive difference equation: n=0,1,...,N-1, at this time, the actual received single scattering path received signal can be expressed as The estimated error is Among them, a i Represent the coefficients of the difference equation; reconstruct the cost function:

[0046]

[0047] In order to minimize the cost function, find the cost function with respect to a i The first-order partial derivative of , and set the result to zero, solve the above equation to get the difference equation coefficient a i and an estimate of the minimum error energy ε(p).

[0048] Preferably, in 3.2, the following method is used to transform the discrete time function form into a linear equation about unknown parameters:

[0049] The coefficient a of the difference equation i Substitute the characteristic polynomial 1+a1z -1 +…+a p z -p = 0, solve its root z i , i=1,…,p; the discrete time function form is simplified to i The linear equation of

[0050] Preferably, in 3.3, the following method is used to solve the linear equation parameters, solve the amplitude and phase according to the parameters, and reconstruct the signal:

[0051] Z is an N×p-dimensional Vandermonde matrix, and z i Different from each other, the least squares solution can be obtained to obtain the unknown parameter b i The vector b is composed of in Substituting the following formulas, the amplitude, phase, frequency and attenuation factor of the received signal can be solved:

[0052]

[0053] Among them, A i is the amplitude, θ i is the phase, f i is the frequency and α i is the attenuation factor, amplitude A i Reflects the change in signal strength after passing through the channel; Phase θ i Reflects the phase shift of the signal relative to the transmitted signal after passing through the channel; frequency f i Reflects the carrier frequency and frequency deviation of the signal; attenuation factor α i It reflects the energy loss of the signal during propagation.

[0054] The second aspect of the present invention provides an environmental backscatter communication channel state estimation system, using the method described in the first aspect of the present invention; comprising:

[0055] Signal extraction module: identifies and extracts single scattering path signals from various scattering paths of signals received by the receiver;

[0056] Signal discrete time function model construction module: Considering the non-line-of-sight transmission in the system, a discrete time function model is constructed for the extracted single scattering path signal;

[0057] Solution module: solves the discrete time function model and calculates the amplitude and phase of the signal;

[0058] Frequency domain analysis module: reconstructs the signal and calculates the amplitude, phase, frequency and attenuation factor of the received signal through Fourier transform analysis in the frequency domain to achieve channel state estimation.

[0059] The third aspect of the present invention provides a method for estimating an environmental backscatter communication noise source. The environmental backscatter communication system includes a radio frequency source, a stationary scatterer, and a receiver. The receiver has a uniform linear array composed of A antenna units. When the high-voltage equipment is running or fails, it will emit special noise. The noise source is a radio frequency source. The noise source estimation method includes the following steps:

[0060] Step 1: The RF source sends out X signals in various directions, which are incident on a uniform linear array composed of Y antenna elements, and a single scattered path signal is identified and extracted from various scattered paths transmitted to the receiver;

[0061] Step 2, constructing the covariance matrix of the single scattering path signal vector x(t) received by Y antennas;

[0062] Preferably, in step 2, the covariance matrix is ​​specifically:

[0063] The covariance matrix R of the array output signal x(t) can be written as:

[0064] R = E[x(t)x(t) H ]=Rs+σ 2 I

[0065] Where Rs = SAS H is the signal covariance matrix, σ 2 I is the noise covariance matrix, σ 2 is the noise power, I is the identity matrix of X×X order, A is the directional response vector, and s is the array steering vector corresponding to X various directions.

[0066] Step 3, eigendecomposition of the covariance matrix to determine the signal subspace and the noise subspace;

[0067] Preferably, in step 3, the covariance matrix is ​​eigen-decomposed to determine the signal subspace and the noise subspace, specifically,

[0068] Perform eigenvalue decomposition on the covariance matrix of the output signal R matrix to obtain (YX) equal eigenvalues; the eigenvalues ​​and eigenvectors describe the signal subspace and the noise subspace. The eigenvectors are sorted according to the size of the eigenvalues. The eigenvectors corresponding to the largest several equal eigenvalues ​​constitute the signal subspace, and the remaining eigenvectors constitute the noise subspace.

[0069] Step 4: construct a spectrum peak search function through the noise subspace, search for the spectrum peak and estimate the wave arrival direction, and calculate the position of the noise source.

[0070] Preferably, in step 4, searching for the spectrum peak and estimating the direction of arrival is specifically performed by using the eigenvector of the noise subspace to construct a spectrum peak search function, the formula of which is:

[0071]

[0072] Among them, Q n Represents the matrix composed of the eigenvectors corresponding to the eigenvalue of the covariance matrix being 0, s(φ) = e jkzsinφ , φ represents the direction of signal incidence; this function is used to calculate the spatial spectrum value for each signal incidence angle and search for spectrum peaks. The number of spectrum peaks is the same as the number of noise sources, and the direction corresponding to the spectrum peak is the direction of the noise source; based on the arrival time difference method, the relative distance from the noise source to the receiver is calculated through the time delay of the noise signal reaching each antenna. Finally, the position of the noise source is calculated by combining the DOA and distance information through the geometric relationship of the antenna array, thereby realizing the noise source position estimation.

[0073] A fourth aspect of the present invention is an environmental backscatter communication channel state estimation system, using the method described in the third aspect of the present invention; comprising:

[0074] Signal extraction module: identifies and extracts single scattering path signals from various scattering paths of signals received by the receiver;

[0075] Noise subspace solution module: constructs the covariance matrix of the single scattering path signal vector x(t) received by Y antennas, and performs eigendecomposition of the covariance matrix to determine the noise subspace;

[0076] Spectral peak construction and search module: constructs a spectral peak search function through the noise subspace, searches for spectral peaks and estimates the direction of arrival;

[0077] Noise source location calculation module: Based on the arrival time difference method, the location of the noise source is calculated by combining the time delay of the noise signal reaching each antenna with the geometric relationship between DOA and the antenna array.

[0078] Compared with the prior art, the present invention has the following beneficial technical effects:

[0079] By identifying the scattering path, the single scattered signal and the multiple scattered signal can be separated, thereby reducing the interference of the multipath effect on the channel estimation, avoiding repeated reception of the same signal while ensuring the signal quality, thereby improving the effectiveness of the communication system. The method fully considers the challenges of the substation as a special application scenario and has good adaptability.

[0080] By sampling the single scattering path signal, building its discrete time function model and solving it to obtain the amplitude and phase of the received signal, reconstructing the received signal, and analyzing the frequency domain representation through Fourier transform to deduce the delay, thus realizing channel state estimation. This can establish a more accurate channel model, which helps to improve the communication quality of the AmBC system.

[0081] The noise source is located by determining the noise subspace and constructing a spectral peak search function for the extracted single scattering path signal. The equipment status can be understood in real time without a microphone or camera, and maintenance and fault repair decisions can be made. This reduces additional hardware equipment and maintenance costs, and supports substation managers in real-time monitoring and decision-making.

[0082] It does not require complex algorithms and extensive training, and helps improve the communication quality and reliability of the AmBC system. BRIEF DESCRIPTION OF THE DRAWINGS

[0083] Figure 1 It is a schematic diagram of the AmBC scenario model of the ambient backscatter communication system;

[0084] Figure 2 It is a flow chart of a method for estimating the state of an environmental backscatter communication channel;

[0085] Figure 3 is a flow chart of a preferred embodiment of calculating a central reference point in a channel state estimation method;

[0086] Figure 4 It is a schematic diagram of the process of performing channel estimation on the extracted single scattering path signal in the channel state estimation method;

[0087] Figure 5 It is the RMSE error between the estimated value and the true value in the channel state estimation in the preferred embodiment.

[0088] Figure 6 It is a flow chart of a method for locating an environmental backscatter communication noise source; DETAILED DESCRIPTION

[0089] In order to make the purpose, technical scheme and advantages of the present invention clearer, the technical scheme of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. The embodiments described in this application are only embodiments of a part of the present invention, rather than all embodiments. Based on the spirit of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work belong to the protection scope of the present invention.

[0090] Embodiment 1 of the present invention provides a method for estimating the state of an environmental backscatter communication channel. Figure 1As shown, it includes a radio frequency source, a stationary scatterer and a receiver; the channel state estimation method is as follows Figure 2 As shown, the following steps are included:

[0091] Step 1, identifying and extracting a single scattering path signal from various scattering paths from when a radio frequency source sends a signal to when a receiver receives a signal;

[0092] Preferably, in step 1, various scattering paths from when a radio frequency source sends a signal to when a receiver receives the signal include multiple scattering paths and single scattering paths;

[0093] Among them, multiple scattering paths refer to: after the signal is emitted from the RF source, it reaches the receiver through any strong scatterer outside the scatterer, or in addition to being scattered by the scatterer, it also reaches the receiver after being reflected / scattered by the strong scatterer;

[0094] A single scattering path means that the signal is sent from the RF source directly through the scatterer to the receiver.

[0095] Preferably, in step 1, the single scattering path signal is identified and extracted in the following manner:

[0096] 1.1 Calculate the possible mobile station position line LPMD for various scattering paths;

[0097] Preferably, the possible mobile station position line LPMD means: the multiple scattering path is equivalent to a single scattering path scattered by a virtual scatterer. In the single scattering, the transmission angle of the signal transmitted by the RF source is known, the arrival time TOA and the arrival angle AOA of the signal are measured, and the path length is calculated from the arrival time. Then the scatterer position must be at a position with the receiver as the starting point, the arrival angle as the argument and the length less than the path length. If the distance between the scatterer and the receiver is determined, the position of the RF source can be determined; while the position of the virtual scatterer is uncertain, but the TOA values ​​of the signal arrival time under different paths and the positions of the RF source, scatterer and receiver satisfy the formula

[0098]

[0099] in, Indicates the distance between the RF source and the scatterer; represents the distance between the scatterer and the receiver; R represents the length of the path, c0 is the speed of light, and t represents the TOA value measured by the RF source. From this formula, as well as the measured signal arrival time TOA and signal arrival angle AOA, all possible positions of the RF source are obtained by geometric methods. The possible positions are connected together to form a straight line. This line of the mth scattering path is called the possible mobile station position line of the mth signal path. The straight line expression can be recorded as ax m +by m+c=0.

[0100] 1.2 Calculate the central reference point of all paths.

[0101] Preferably, if Figure 3 As shown, in 1.2, the central reference point of all paths is calculated in the following way:

[0102] 1.2.1 Use all paths to perform a preliminary fit on the position coordinates of the RF source, and the fit is in represents the final estimated position of the RF source; X = [x, y] T Indicates the actual location coordinates of the RF source; X k represents the RF source position coordinates corresponding to k paths; r k Indicates the length of the kth path, which is equal to the TOA measurement value multiplied by the propagation speed of radio waves; ||XX k || represents the Euclidean distance between the real RF source and the RF source corresponding to the i-th path; Q represents the set of all paths in the current positioning;

[0103] 1.2.2 Calculate the weight of each path signal:

[0104] Use each path to estimate the RF source position, and calculate the inverse of the normalized residual variable of each path as the weight of the k-th path signal;

[0105] The formula is:

[0106]

[0107] in, The RF source position estimation of the kth path, Q k is the kth path signal.

[0108] 1.2.3 Recalculate the final estimated position of the RF source after weighting And calculate the residual variable Final estimated position Has the smallest positioning residual;

[0109]

[0110] in, The RF source position estimation of the kth path, Q k is the kth path signal;

[0111] Determine the final estimated position If there is no convergence, return to 1.2.2 to adjust the weights again. If there is convergence, the final estimated position is As the middle reference point T(x T,y T ).

[0112] 1.3 Calculate the distance from the central reference point to each LPMD line, and select a decision threshold. Identify the path corresponding to the path below the decision threshold as a single scattering path, otherwise, identify it as a multiple scattering path.

[0113] Preferably, the specific process of 1.3 is: calculating the distance from the intermediate reference point T to each LPMD line, the formula is:

[0114]

[0115] In the above formula, dis m represents the distance from the intermediate reference point T to the LPMD line corresponding to the mth path, a m , b m 、c m Represents the LPMD line equation ax corresponding to the mth path m +by m +c=0 in the parameter, x T ,y T Respectively represent the horizontal and vertical coordinates of the middle reference point T;

[0116] Then the distance weight d from the middle reference point T to the mth LPMD line is calculated m :

[0117]

[0118] In the above formula, M represents the total number of paths, dis k It represents the distance from the intermediate reference point T to the LPMD line corresponding to the kth path. The single scattering path signal is finally identified and extracted by judging the weight and the preset threshold d. The specific judgment expression is:

[0119]

[0120] Wherein, d is the preset threshold.

[0121] Step 2, sampling the single scattering path signal and constructing a discrete time function model of the single scattering path received signal;

[0122] Preferably, in step 2, the received signal is modeled as a set of p exponential functions with amplitude, phase, frequency and attenuation factor, and its discrete time function model is expressed as follows:

[0123]

[0124] Among them, b i =A i exp(jθ i ), zi =exp[(α i +j2πf i )Δt], As an approximation of y(n), its error is defined as y(n) is the expression of the nth received signal actually received, A i is the amplitude of the i-th exponential function in the set of functions in the model, θ i is the phase of the i-th exponential function in the set of functions in the model, α i is the attenuation factor of the i-th exponential function in the set of functions in the model, f i is the oscillation frequency of the i-th exponential function in the group of functions in the model, Δt is the sampling interval, p is the order, n represents the n-th discrete signal, and N is the total number of discrete signals.

[0125] Step 3, such as Figure 4 As shown, the model in step 2 is solved to obtain the amplitude and phase of the received signal, reconstruct the received signal, and use Fourier transform to analyze the frequency domain representation to deduce the delay to achieve channel state estimation.

[0126] Preferably, in step 3, solving the amplitude and phase of the received signal and reconstructing the received signal includes the following:

[0127] 3.1 Construct the discrete time function recursive difference equation of the single scattering path signal and solve the equation coefficients;

[0128] Preferably, in 3.1, the discrete-time differential equation of the single scattering path signal is constructed in the following manner, and the coefficients of the equation are solved:

[0129] Construct the cost function ε to represent the sum of squared errors: When the sampling interval Δt=1, Satisfies the recursive difference equation: At this time, the actual received single scattering path received signal can be expressed as The estimated error is Among them, a i Represent the coefficients of the difference equation; reconstruct the cost function:

[0130]

[0131] In order to minimize the cost function, find the cost function with respect to a i The first-order partial derivative of , and set the result to zero, solve the above equation to get the difference equation coefficient a i and an estimate of the minimum error energy ε(p).

[0132] 3.2 Use the coefficients of the difference equation to construct the system function and transform the discrete time function form into a linear equation about the unknown parameters;

[0133] Preferably, in 3.2, the following method is used to transform the discrete time function form into a linear equation about unknown parameters:

[0134] The coefficient a of the difference equation i Substitute the characteristic polynomial 1+a1z -1 +…+a p z -p = 0, solve its root z i , i=1,…,p; the discrete time function form is simplified to i The linear equation of

[0135] 3.3 Solve the linear equation parameters according to the matrix characteristics, solve the amplitude and phase according to the parameters, and reconstruct the signal.

[0136] Preferably, in 3.3, the following method is used to solve the linear equation parameters, solve the amplitude and phase according to the parameters, and reconstruct the signal:

[0137] Z is an N×p-dimensional Vandermonde matrix, and z i Different from each other, the least squares solution can be obtained to obtain the unknown parameter b i The vector b is composed of in Substituting the following formulas, the amplitude, phase, frequency and attenuation factor of the received signal can be solved:

[0138]

[0139] Among them, A i is the amplitude, θ i is the phase, f i is the frequency and α i is the attenuation factor, amplitude A i Reflects the change in signal strength after passing through the channel; Phase θ i Reflects the phase shift of the signal relative to the transmitted signal after passing through the channel; frequency f i Reflects the carrier frequency and frequency deviation of the signal; attenuation factor α i It reflects the energy loss of the signal during propagation.

[0140] Specifically, the signal amplitude A i , phase θ i , frequency f i With the attenuation factor α i The scheme for channel state estimation is:

[0141] Amplitude A i It reflects the change in signal strength after passing through the channel. It is affected by many factors, including path loss, multipath fading and shadowing effect. By analyzing the amplitude of the received signal, the gain or attenuation characteristics of the channel can be obtained.

[0142] Phase θ i It reflects the phase shift of the signal relative to the transmitted signal after passing through the channel. The phase change is mainly caused by the path length of the signal propagation, the characteristics of the propagation medium and the multipath effect. By analyzing the phase of the received signal, the channel delay characteristics and phase shift information can be obtained.

[0143] Frequency f i The Doppler effect in the channel is related to the carrier frequency and frequency deviation of the signal, which can cause the frequency deviation of the signal. This frequency deviation can be used to estimate dynamic information such as the speed and direction of movement. Different frequency components have different attenuation. The frequency selectivity of the channel can be understood through the frequency characteristics.

[0144] Attenuation factor α i It reflects the energy loss of the signal during propagation, including free space propagation loss, penetration loss and scattering loss. By analyzing the attenuation factor, we can understand the overall attenuation characteristics and loss mechanism of the channel.

[0145] In summary, the amplitude and attenuation factor reflect the change of signal strength with increasing distance, which can characterize the path loss characteristics; the change of phase and frequency reveals the interference and delay spread caused by multipath propagation. The amplitude is also affected by the multipath effect, and these parameters can be used to analyze the multipath characteristics; the frequency offset and phase change over time can reflect the time-varying characteristics of the channel. For mobile communication environments, the frequency Doppler effect provides information on the dynamic changes of the channel.

[0146] like Figure 5 It represents the RMSE error between the estimated value and the true value in the channel state estimation. When the number of paths is less than 5, the estimation accuracy is very good. As the number of multipaths increases, the RMSE error gradually increases.

[0147] Embodiment 3 of the present invention proposes a method for estimating an environmental backscatter communication noise source. Figure 1 As shown, it includes a radio frequency source, a stationary scatterer and a receiver; the receiver has a uniform linear array composed of A antenna units. Due to the operation and failure of high-voltage equipment, special noise will be emitted. The noise source is a radio frequency source. The noise source estimation method is as follows: Figure 6 As shown, the following steps are included:

[0148] Step 1: The RF source sends out X signals in various directions, which are incident on a uniform linear array composed of Y antenna elements, and a single scattered path signal is identified and extracted from various scattered paths transmitted to the receiver;

[0149] Preferably, in step 1, various scattering paths from when a radio frequency source sends a signal to when a receiver receives the signal include multiple scattering paths and single scattering paths;

[0150] Among them, multiple scattering paths refer to: after the signal is emitted from the RF source, it reaches the receiver through any strong scatterer outside the scatterer, or in addition to being scattered by the scatterer, it also reaches the receiver after being reflected / scattered by the strong scatterer;

[0151] A single scattering path means that the signal is sent from the RF source directly through the scatterer to the receiver.

[0152] Preferably, in step 1, the single scattering path signal is identified and extracted in the following manner:

[0153] 1.1 Calculate the possible mobile station position line LPMD for various scattering paths;

[0154] Preferably, the possible mobile station position line LPMD means: the multiple scattering path is equivalent to a single scattering path scattered by a virtual scatterer. In the single scattering, the transmission angle of the signal transmitted by the RF source is known, the arrival time TOA and the arrival angle AOA of the signal are measured, and the path length is calculated from the arrival time. Then the scatterer position must be at a position with the receiver as the starting point, the arrival angle as the argument and the length less than the path length. If the distance between the scatterer and the receiver is determined, the position of the RF source can be determined; while the position of the virtual scatterer is uncertain, but the TOA values ​​of the signal arrival time under different paths and the positions of the RF source, scatterer and receiver satisfy the formula

[0155]

[0156] in, Indicates the distance between the RF source and the scatterer; represents the distance between the scatterer and the receiver; R represents the length of the path, c0 is the speed of light, and t represents the TOA value measured by the RF source. From this formula, as well as the measured signal arrival time TOA and signal arrival angle AOA, all possible positions of the RF source are obtained by geometric methods. The possible positions are connected together to form a straight line. This line of the mth scattering path is called the possible mobile station position line of the mth signal path. The straight line expression can be recorded as ax m +by m +c=0.

[0157] 1.2 Calculate the central reference point of all paths.

[0158] Preferably, if Figure 3 As shown, in 1.2, the central reference point of all paths is calculated in the following way:

[0159] 1.2.1 Use all paths to perform a preliminary fit on the position coordinates of the RF source, and the fit is in represents the final estimated position of the RF source; X = [x, y] T Indicates the actual location coordinates of the RF source; X k represents the RF source position coordinates corresponding to k paths; r k Indicates the length of the kth path, which is equal to the TOA measurement value multiplied by the propagation speed of radio waves; ||XX k || represents the Euclidean distance between the real RF source and the RF source corresponding to the i-th path; Q represents the set of all paths in the current positioning;

[0160] 1.2.2 Calculate the weight of each path signal:

[0161] Use each path to estimate the RF source position, and calculate the inverse of the normalized residual variable of each path as the weight of the k-th path signal;

[0162] The formula is:

[0163]

[0164]

[0165] in, The RF source position estimation of the kth path, Q k is the kth path signal.

[0166] 1.2.3 Recalculate the final estimated position of the RF source after weighting And calculate the residual variable Final estimated position Has the smallest positioning residual;

[0167]

[0168] in, The RF source position estimation of the kth path, Q k is the kth path signal;

[0169] Determine the final estimated position If there is no convergence, return to 1.2.2 to adjust the weights again. If there is convergence, the final estimated position is As the middle reference point T(x T ,y T ).

[0170] 1.3 Calculate the distance from the central reference point to each LPMD line, and select a decision threshold. Identify the path corresponding to the path below the decision threshold as a single scattering path, otherwise, identify it as a multiple scattering path.

[0171] Preferably, the specific process of 1.3 is: calculating the distance from the intermediate reference point T to each LPMD line, the formula is:

[0172]

[0173] In the above formula, dis m represents the distance from the intermediate reference point T to the LPMD line corresponding to the mth path, a m , b m 、c m Represents the LPMD line equation ax corresponding to the mth path m +by m +c=0 in the parameter, x T ,y T Respectively represent the horizontal and vertical coordinates of the middle reference point T;

[0174] Then the distance weight d from the middle reference point T to the mth LPMD line is calculated m :

[0175]

[0176] In the above formula, M represents the total number of paths, dis k It represents the distance from the intermediate reference point T to the LPMD line corresponding to the kth path. The single scattering path signal is finally identified and extracted by judging the weight and the preset threshold d. The specific judgment expression is:

[0177]

[0178] Wherein, d is the preset threshold.

[0179] The second aspect of the present invention provides an environmental backscatter communication channel state estimation system, using the method described in the first aspect of the present invention; comprising:

[0180] Signal extraction module: identifies and extracts single scattering path signals from various scattering paths of signals received by the receiver;

[0181] Signal discrete time function model construction module: Considering the non-line-of-sight transmission in the system, a discrete time function model is constructed for the extracted single scattering path signal;

[0182] Solution module: solves the discrete time function model and calculates the amplitude and phase of the signal;

[0183] Frequency domain analysis module: reconstructs the signal and calculates the amplitude, phase, frequency and attenuation factor of the received signal through Fourier transform analysis in the frequency domain to achieve channel state estimation.

[0184] Step 2, constructing the covariance matrix of the single scattering path signal vector x(t) received by Y antennas;

[0185] Preferably, in step 2, the covariance matrix is ​​specifically:

[0186] Since the signals from different signal sources and the noise and signal are orthogonal to each other, the covariance matrix R of the array output signal x(t) can be written as:

[0187] R = E[x(t)x(t) H ]=Rs+σ 2 I

[0188] Where Rs = SAS H is the signal covariance matrix, σ 2 I is the noise covariance matrix, σ 2 is the noise power, I is the unit matrix of X×X order, A is the direction response vector, and S is the array steering vector corresponding to X various directions.

[0189] Step 3, eigendecomposition of the covariance matrix to determine the signal subspace and the noise subspace;

[0190] Preferably, in step 3, the covariance matrix is ​​eigen-decomposed to determine the signal subspace and the noise subspace, specifically,

[0191] Perform eigenvalue decomposition on the covariance matrix of the output signal R matrix to obtain (YX) equal eigenvalues; the eigenvalues ​​and eigenvectors describe the signal subspace and the noise subspace. The eigenvectors are sorted according to the size of the eigenvalues. The eigenvectors corresponding to the largest several equal eigenvalues ​​constitute the signal subspace, and the remaining eigenvectors constitute the noise subspace.

[0192] Step 4: construct a spectrum peak search function through the noise subspace, search for the spectrum peak and estimate the wave arrival direction, and calculate the position of the noise source.

[0193] Preferably, in step 4, searching for the spectrum peak and estimating the direction of arrival is specifically performed by using the eigenvector of the noise subspace to construct a spectrum peak search function, the formula of which is:

[0194]

[0195] Among them, Q n Represents the matrix composed of the eigenvectors corresponding to the eigenvalue of the covariance matrix being 0, s(φ) = e jkzsinφ, φ represents the direction of signal incidence; this function is used to calculate the spatial spectrum value for each signal incidence angle and search for spectrum peaks. The number of spectrum peaks is the same as the number of noise sources, and the direction corresponding to the spectrum peak is the direction of the noise source; based on the arrival time difference method, the relative distance from the noise source to the receiver is calculated through the time delay of the noise signal reaching each antenna. Finally, the position of the noise source is calculated by combining the DOA and distance information through the geometric relationship of the antenna array, thereby realizing the noise source position estimation.

[0196] Embodiment 4 of the present invention provides an environmental backscatter communication channel state estimation system, using the method described in Embodiment 3 of the present invention; comprising:

[0197] Signal extraction module: identifies and extracts single scattering path signals from various scattering paths of signals received by the receiver;

[0198] Noise subspace solution module: constructs the covariance matrix of the single scattering path signal vector x(t) received by Y antennas, and performs eigendecomposition of the covariance matrix to determine the noise subspace;

[0199] Compared with the prior art, the present invention has the following beneficial technical effects:

[0200] By identifying the scattering path, the single scattering signal and the multiple scattering signal can be separated, thereby reducing the interference of the multipath effect on the channel estimation, avoiding the repeated reception of the same signal while ensuring the signal quality, and improving the effectiveness of the communication system. The challenge of the substation as a special application scenario is fully considered, so it has good adaptability. By sampling the single scattering path signal, constructing its discrete time function model and solving it to obtain the amplitude and phase of the received signal, reconstructing the received signal, and analyzing the frequency domain representation through Fourier transform to deduce the delay, that is, to achieve channel state estimation, a more accurate channel model can be established, which helps to improve the communication quality of the AmBC system. The noise source is located by determining the noise subspace and constructing the spectrum peak search function for the extracted single scattering path signal. The device status can be understood in real time without a microphone or camera, and maintenance and fault repair decisions can be made, reducing additional hardware equipment and maintenance costs, and supporting substation managers to conduct real-time monitoring and decision-making. Without complex algorithms and a lot of training, it helps to improve the communication quality and reliability of the AmBC system.

[0201] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents, and any modifications or equivalent replacements that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for estimating the state of an environmental backscatter communication channel, wherein the environmental backscatter communication system comprises a radio frequency source, a stationary scatterer and a receiver; characterized in that: The channel state estimation method comprises the following steps: Step 1, identifying and extracting a single scattering path signal from various scattering paths from when a radio frequency source sends a signal to when a receiver receives a signal; Step 2, sampling the single scattering path signal and constructing a discrete time function model of the single scattering path received signal; Step 3, solve the model in step 2 to obtain the amplitude and phase of the received signal, reconstruct the received signal, and use Fourier transform analysis to deduce the delay in the frequency domain to achieve channel state estimation.

2. The channel state estimation method according to claim 1, characterized in that: In step 1, various scattering paths from when a radio frequency source sends a signal to when a receiver receives the signal include multiple scattering paths and single scattering paths; Among them, multiple scattering paths refer to: after the signal is emitted from the RF source, it reaches the receiver through any strong scatterer outside the scatterer, or in addition to being scattered by the scatterer, it also reaches the receiver after being reflected / scattered by the strong scatterer; A single scattering path means that the signal is sent from the RF source directly through the scatterer to the receiver.

3. The channel state estimation method according to claim 2, characterized in that: In step 1, the single scattering path signal is identified and extracted in the following way: 1.1 Calculate the possible mobile station position line LPMD for various scattering paths; 1.2 Calculate the central reference point of all paths; 1.3 Calculate the distance from the central reference point to each LPMD line, and select a decision threshold. Identify the path corresponding to the path below the decision threshold as a single scattering path, otherwise, identify it as a multiple scattering path.

4. The channel state estimation method according to claim 3, characterized in that: The line of possible mobile station positions LPMD means: the multiple scattering path is equivalent to a single scattering path scattered by a virtual scatterer. In single scattering, the transmission angle of the signal transmitted by the RF source is known, the arrival time TOA and the arrival angle AOA of the signal are measured, and the path length is calculated from the arrival time. The position of the scatterer must be at a position with the receiver as the starting point, the arrival angle as the amplitude angle and the length less than the path length. If the distance between the scatterer and the receiver is determined, the position of the RF source can be determined; while the position of the virtual scatterer is uncertain, but the TOA value of the signal arrival time under different paths and the positions of the RF source, scatterer and receiver satisfy the formula in, Indicates the distance between the RF source and the scatterer; represents the distance between the scatterer and the receiver; R represents the length of the path, c0 is the speed of light, and t represents the TOA value measured by the RF source. From this formula, as well as the measured signal arrival time TOA and signal arrival angle AOA, all possible positions of the RF source are obtained by geometric methods. The possible positions are connected together to form a straight line. This line of the mth scattering path is called the possible mobile station position line of the mth signal path. The straight line expression can be recorded as ax m +by m +c=0.

5. The channel state estimation method according to claim 4, characterized in that: In 1.2, the central reference point of all paths is calculated in the following way: 1.2.1 Use all paths to perform a preliminary fit on the position coordinates of the RF source, and the fit is in represents the final estimated position of the RF source; X = [x, y] T Indicates the actual location coordinates of the RF source; X k represents the RF source position coordinates corresponding to k paths; r k Indicates the length of the kth path, which is equal to the TOA measurement value multiplied by the propagation speed of radio waves; ||XX k || represents the Euclidean distance between the real RF source and the RF source corresponding to the i-th path; Q represents the set of all paths in the current positioning; 1.2.2 Calculate the weight of each path signal Calculate the RF source position estimate for each path And calculate the residual variable for each path Divide it by the kth signal Q k The length of is normalized, and the normalized inverse is used as the weight λ of the kth path signal k ; 1.2.3 Recalculate the final estimated position of the RF source after weighting And calculate the residual variable Final estimated position Has the smallest positioning residual; in, The RF source position estimation of the kth path, Q k is the kth path signal; Determine the final estimated position If there is no convergence, return to 1.2.2 to adjust the weights again. If there is convergence, the final estimated position is As the middle reference point T(x T ,y T ).

6. The channel state estimation method according to claim 5, characterized in that: The specific process of 1.3 is: calculate the distance from the intermediate reference point T to each LPMD line, the formula is: In the above formula, dis m represents the distance from the intermediate reference point T to the LPMD line corresponding to the mth path, a m , b m 、c m Represents the LPMD line equation ax corresponding to the mth path m +by m +c=0 in the parameter, x T ,y T Respectively represent the horizontal and vertical coordinates of the middle reference point T; Then the distance weight d from the middle reference point T to the mth LPMD line is calculated m : In the above formula, M represents the total number of paths, dis k It represents the distance from the intermediate reference point T to the LPMD line corresponding to the kth path. The single scattering path signal is finally identified and extracted by judging the weight and the preset threshold d. The specific judgment expression is: Wherein, d is the preset threshold.

7. The method for estimating the state of an environmental backscatter communication channel according to claim 1, characterized in that: In step 2, the received signal is modeled as a set of p exponential functions with amplitude, phase, frequency and attenuation factor, and its discrete time function model is expressed as follows: Among them, b i =A i exp(jθ i ), z i =exp[(α i +j2πf i )Δt], As an approximation of y(n), its error is defined as y(n) is the expression of the nth received signal actually received, A i is the amplitude of the i-th exponential function in the set of functions in the model, θ i is the phase of the i-th exponential function in the set of functions in the model, α i is the attenuation factor of the i-th exponential function in the set of functions in the model, f i is the oscillation frequency of the i-th exponential function in the group of functions in the model, Δt is the sampling interval, p is the order, n represents the n-th discrete signal, and N is the total number of discrete signals.

8. The method for estimating the state of an environmental backscatter communication channel according to claim 7, characterized in that: In step 3, the amplitude and phase of the received signal are solved and the received signal is reconstructed, including the following: 3.1 Construct the discrete time function recursive difference equation of the single scattering path signal and solve the equation coefficients; 3.2 Use the coefficients of the difference equation to construct the system function and transform the discrete time function form into a linear equation about the unknown parameters; 3.3 Solve the linear equation parameters according to the matrix characteristics, solve the amplitude and phase according to the parameters, and reconstruct the signal.

9. The method for estimating the state of an environmental backscatter communication channel according to claim 8, characterized in that: In 3.1, the discrete-time difference equation of the single scattering path signal is constructed in the following way, and the coefficients of the equation are solved: Construct the cost function ε to represent the sum of squared errors: When the sampling interval Δt=1, Satisfies the recursive difference equation: At this time, the actual received single scattering path received signal can be expressed as The estimated error is Among them, a i Represents the coefficients of the difference equation; reconstructs the cost function J(a): In order to minimize the cost function, find the cost function with respect to a i The first-order partial derivative of , and set the result to zero, solve the above equation to get the difference equation coefficient a i and an estimate of the minimum error energy ε(p).

10. The method for estimating the state of an environmental backscatter communication channel according to claim 9, characterized in that: In 3.2, the following method is used to transform the discrete time function form into a linear equation with respect to the unknown parameters: The coefficient α of the difference equation i Substitute the characteristic polynomial 1+a1z -1 +…+a p z -p = 0, solve its root z i , i=1,…,p; the discrete time function form is simplified to i The linear equation of 11. The method for estimating the state of an environmental backscatter communication channel according to claim 10, characterized in that: In 3.3, the following method is used to solve the linear equation parameters, solve the amplitude and phase according to the parameters, and reconstruct the signal. Z is an N×p-dimensional Vandermonde matrix, and z i Different from each other, the least squares solution can be obtained to obtain the unknown parameter b i The vector b is composed of in Substituting the following formulas, the amplitude, phase, frequency and attenuation factor of the received signal can be solved: Among them, A i is the amplitude, θ i is the phase, f i is the frequency and α i is the attenuation factor, amplitude A i Reflects the change in signal strength after passing through the channel; Phase θ i Reflects the phase shift of the signal relative to the transmitted signal after passing through the channel; frequency f i Reflects the carrier frequency and frequency deviation of the signal; attenuation factor α i It reflects the energy loss of the signal during propagation.

12. An environmental backscatter communication channel state estimation system, using any one of the methods of claims 1-11; characterized in that: include: Signal extraction module: identifies and extracts single scattering path signals from various scattering paths of signals received by the receiver; Signal discrete time function model construction module: Considering the non-line-of-sight transmission in the system, a discrete time function model is constructed for the extracted single scattering path signal; Solution module: solves the discrete time function model and calculates the amplitude and phase of the signal; Frequency domain analysis module: reconstructs the signal and calculates the amplitude, phase, frequency and attenuation factor of the received signal through Fourier transform analysis in the frequency domain to achieve channel state estimation.

13. A method for estimating an environmental backscatter communication noise source, wherein the environmental backscatter communication system includes a radio frequency source, a stationary scatterer, and a receiver; the receiver has a uniform linear array consisting of A antenna units, and a high-voltage device emits special noise when it is running or fails. The noise source is a radio frequency source, and the method is characterized in that: The noise source estimation method comprises the following steps: Step 1: The RF source sends out X signals in various directions, which are incident on a uniform linear array composed of Y antenna elements, and a single scattered path signal is identified and extracted from various scattered paths transmitted to the receiver; Step 2, constructing the covariance matrix of the single scattering path signal vector x(t) received by Y antennas; Step 3, eigendecomposition of the covariance matrix to determine the signal subspace and the noise subspace; Step 4: construct a spectrum peak search function through the noise subspace, search for the spectrum peak and estimate the wave arrival direction, and calculate the position of the noise source.

14. The method for estimating environmental backscatter communication noise sources according to claim 13, characterized in that: In step 2, the covariance matrix is ​​specifically: The covariance matrix R of the array output signal x(t) is written as: R=E[x(t)x(t) H ]=Rs+σ 2 I Where Rs = SAS H is the signal covariance matrix, σ 2 I is the noise covariance matrix, σ 2 is the noise power, I is the unit matrix of X×X order, A is the direction response vector, and S is the array steering vector corresponding to X various directions.

15. The method for estimating noise sources of an environmental backscatter communication channel according to claim 14, characterized in that: In step 3, the covariance matrix is ​​eigendecomposed to determine the signal subspace and the noise subspace, specifically, Perform eigenvalue decomposition on the covariance matrix of the output signal R matrix to obtain (YX) equal eigenvalues; the eigenvalues ​​and eigenvectors describe the signal subspace and the noise subspace. The eigenvectors are sorted according to the size of the eigenvalues. The eigenvectors corresponding to the largest several equal eigenvalues ​​constitute the signal subspace, and the remaining eigenvectors constitute the noise subspace.

16. The method for estimating environmental backscatter communication noise sources according to claim 15, characterized in that: In step 4, searching for the spectrum peak and estimating the direction of arrival are specifically as follows: constructing a spectrum peak search function using the eigenvector of the noise subspace, and its formula is: Among them, Q n Represents the matrix composed of the eigenvectors corresponding to the eigenvalue of the covariance matrix being 0, s(φ) = e jkzsinφ , φ represents the direction of signal incidence; this function is used to calculate the spatial spectrum value for each signal incidence angle and search for spectrum peaks. The number of spectrum peaks is the same as the number of noise sources, and the direction corresponding to the spectrum peak is the direction of the noise source; based on the arrival time difference method, the relative distance from the noise source to the receiver is calculated through the time delay of the noise signal reaching each antenna. Finally, the position of the noise source is calculated by combining the DOA and distance information through the geometric relationship of the antenna array, thereby realizing the noise source position estimation.

17. An environmental backscatter communication channel state estimation system, using any one of the methods of claims 13-16; characterized in that: include: Signal extraction module: identifies and extracts single scattering path signals from various scattering paths of signals received by the receiver; Noise subspace solution module: constructs the covariance matrix of the single scattering path signal vector x(t) received by Y antennas, and performs eigendecomposition of the covariance matrix to determine the noise subspace; Spectral peak construction and search module: constructs a spectral peak search function through the noise subspace, searches for spectral peaks and estimates the direction of arrival; Noise source location calculation module: Based on the arrival time difference method, the location of the noise source is calculated by combining the time delay of the noise signal reaching each antenna with the geometric relationship between DOA and the antenna array.

Images