Passive non-line-of-sight direct positioning method based on channel model

Abstract

The invention discloses a passive non-line-of-sight direct positioning method based on a channel model, and the method comprises the steps: firstly building a passive non-line-of-sight direct positioning model based on a maximum likelihood criterion, then carrying out the equivalent transformation of a target function, and enabling a transformed optimization model to have the physical significance of reconstructing an original signal for the data of each receiving station through the information of the channel model, and carrying out the calculation of the original signal. Enabling the reconstructed signals to be the same; and on the basis, performing gridding search on a to-be-detected region, reconstructing an original signal by utilizing a channel model from each grid point to a receiving station, and finding a point with the maximum target function, namely a target position. According to the invention, a multipath channel can be converted into available information, the problem that a radiation source target cannot be accurately positioned in a non-line-of-sight environment in an existing passive positioning method is solved, even positioning in a single-station or double-station scene can be realized, and the limitation that a traditional positioning model needs more than three positioning stations is broken through.

Description

Technical Field

[0001] This invention belongs to the field of signal processing, specifically relating to a passive non-line-of-sight direct localization method based on a channel model. Background Technology

[0002] In radiation source localization technology, if the propagation path between the target and the receiving station is unobstructed, and the signal propagates at line-of-sight, the target can be located by measuring the time difference of the signal's arrival at the receiving station and constructing a hyperbolic equation system. Solving this equation system allows for target localization; this technique is called time difference-of-sight (TDOS) localization. However, if the propagation path between the target and the receiving station is obstructed, and the signal propagates non-line-of-sight, then traditional TDOS localization fails and cannot accurately locate the target. Current non-line-of-sight (NLS) localization technologies are mainly applied to navigation scenarios or cooperative localization based on radar information, requiring prior building geometry information and multiple cooperative transmitting anchors or radars. However, research on passive NLS localization in non-cooperative scenarios is scarce. Summary of the Invention

[0003] The purpose of this invention is to provide a passive non-line-of-sight direct positioning method based on a channel model, which solves the problem that existing passive positioning methods cannot accurately locate radiation source targets in non-line-of-sight environments. This method can convert multipath channels into usable information, improve positioning accuracy in complex environments, and even achieve positioning in single-station or dual-station scenarios, breaking through the limitation of traditional positioning models that require more than three positioning stations.

[0004] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0005] A non-line-of-sight direct localization method based on a channel model is described below:

[0006] Step 1: Establish the objective function for passive non-line-of-sight direct localization based on the maximum likelihood criterion;

[0007] Step 2: Reconstruct the signal transmitted at the target point based on the channel model;

[0008] Step 3: Perform a gridded search on the area to be tested. Calculate a corresponding reconstructed signal for each grid point and substitute it into the passive non-line-of-sight direct localization objective function. Record the grid point corresponding to the maximum value of the passive non-line-of-sight direct localization objective function as the target point location, thus completing the non-line-of-sight direct localization based on the channel model.

[0009] Furthermore, the specific method of step 1 is as follows:

[0010]

[0011] Where M represents the number of receiving stations. Indicates taking the real part; This indicates the signal transmitted from the target point corresponding to the m-th receiving station; This represents the received signal from the m-th receiving station; This represents the total number of propagation paths corresponding to the m-th receiving station. This represents the attenuation of each propagation path at the m-th receiving station; , This represents the k-th frequency point of the signal received by the receiving station, where K is the number of frequency points of the signal received by the receiving station. Let j represent the time delay of the l-th propagation path at the m-th receiving station, where j represents the imaginary unit. For the preset threshold, .

[0012] Furthermore, the specific method for step 2 is as follows:

[0013]

[0014]

[0015] in, This indicates the signal transmitted from the target point corresponding to the m-th receiving station; This represents the received signal from the m-th receiving station; M represents the number of receiving stations. This is an L x N matrix, where L is the number of sampling points for the signal received at the receiving station, and N is the number of sampling points for the signal transmitted at the target point. The letters at the top of the matrix are used to indicate the number of sampling points. Columns, non-matrix element values , , This represents the rounding function, where the first row of the matrix, the second row, and the third row are rounded up. All elements in the column are 1, starting from the first row and the second row. The elements at the bottom right diagonal (45° angle) of the column are all 1, and the values ​​of all other elements in the matrix are 0. The sampling rate of the received signal at the receiving station; This represents the time delay of the l-th propagation path at the m-th receiving station.

[0016] The beneficial effects of this invention are:

[0017] This invention can convert multipath channels into usable information, solving the problem that existing passive positioning methods cannot accurately locate radiation source targets in non-line-of-sight environments. It can even achieve positioning in single-station or dual-station scenarios, breaking through the limitation of traditional positioning models that require more than three positioning stations. Attached Figure Description

[0018] Figure 1This is a simulation scenario diagram for non-line-of-sight passive direct positioning based on a channel model;

[0019] Figure 2 The results are simulation results for non-line-of-sight passive direct positioning based on the channel model. Detailed Implementation

[0020] The present invention will now be further described in conjunction with the accompanying drawings and specific embodiments.

[0021] A non-line-of-sight direct localization method based on a channel model:

[0022] The technical solution adopted by the passive non-line-of-sight direct positioning method based on the channel model of this invention is as follows: First, a passive non-line-of-sight direct positioning model is established based on the maximum likelihood criterion. Since the signal variables in this model are unknown, the optimal value of the objective function cannot be directly obtained through grid search to obtain the target position. Therefore, an equivalent transformation is performed on the objective function. The physical meaning of the optimized model after the transformation is to reconstruct the original signal from the data of each receiving station using the channel model information, so that the reconstructed signals are the same. Based on this, a grid search is performed on the area to be measured, and the original signal is reconstructed using the channel model from each grid point to the receiving station. The point where the objective function is maximized is the target position. Specifically, as follows:

[0023] Step 1: Establish a passive non-line-of-sight direct localization model based on the maximum likelihood criterion.

[0024] In non-line-of-sight positioning scenarios, the received data from the receiving station can be represented as:

[0025] (1)

[0026] In the formula This represents the attenuation of each propagation path at the m-th receiving station. This represents the time delay of each propagation path at the m-th receiving station. This represents the total number of propagation paths corresponding to the m-th receiving station. Transmit a signal to the target point. Let the noise be at the m-th receiving station; by performing a Fourier transform on the data from each receiving station, we can obtain:

[0027] (2)

[0028] This represents the k-th frequency point of the signal received by the receiving station, where K is the number of frequency points of the signal received by the receiving station. Let j represent the time delay of the l-th propagation path at the m-th receiving station, where j represents the imaginary unit.

[0029] The above formula, based on K frequency points, can be rearranged into a matrix form:

[0030] (3)

[0031] For the reason The diagonal matrix formed ;

[0032] According to the maximum likelihood criterion, the following likelihood function can be obtained:

[0033] (4)

[0034] in, express The covariance matrix, express The variance;

[0035] Maximum likelihood estimation of the above equation is equivalent to

[0036] (5)

[0037] Step 2, Equivalent transformation of the problem

[0038] Because in equation (5) Since the location is unknown and cannot be directly obtained through gridded search, an equivalent transformation of the above equation is required to eliminate... The effect of. The above formula is equivalent to

[0039] (6)

[0040] Further introduce auxiliary variables The above formula is equivalent to

[0041] (7)

[0042] By interchangeing the constraints and the objective function, the above equation becomes further equivalent to:

[0043] (8)

[0044] In this embodiment, the preset threshold is used. ;

[0045] And because

[0046] (9)

[0047] And in fact, the power of the transmitted signal is fixed, that is... , The value of is definite, therefore equation (8) is equivalent to

[0048] (10)

[0049] Step 3: Solve the above problem using a grid search.

[0050] The physical meaning of Equation (10) is to reconstruct the original signal using the channel model information of each receiving station so that the reconstructed signal is the same. At this time, the reference point position corresponding to the channel model is the target position.

[0051] Signal reconstruction can be achieved in the following way. According to equation (1), there is the following relationship between the received data and the original signal:

[0052] (11)

[0053] in This represents the number of sampling points with delay. The above formula can be rewritten in matrix form:

[0054] (12)

[0055] L represents the number of sampling points for the signal received at the receiving station, and N represents the number of sampling points for the signal transmitted at the target point. The letters above the matrix are used to indicate the sampling points. Column, non-matrix element value, the first row, the second row, and the third row of the matrix. All elements in the column are 1, starting from the first row and the second row. The elements at the bottom right diagonal (45° angle) of the column are all 1, and the values ​​of all other elements in the matrix are 0. This is the sampling rate of the received signal at the receiving station.

[0056] Recorded as

[0057] (13)

[0058] The reconstructed signal can then be represented as

[0059] (14)

[0060] Therefore, according to equation (10), a gridded search is performed on the area to be tested, and the original signal is reconstructed using the channel model from each grid point to the receiving station. The point that maximizes the objective function in equation (10) is the target location.

[0061] The following is a more specific example:

[0062] The experiment of this invention is in a Within the area, the target is located at (0 km, 2 km), two receiving stations are located at (-3 km, -4 km), and (3 km, -3 km), and three obstacles are located at (-1 km, 0 km), (0.5 km, 0 km), and (2 km, 1 km). Figure 1As shown. Due to the presence of obstacles, the signal propagation path from the target to the receiving station is no longer a straight line. Target to receiving station 1 has only one reflected path, with a length three times that of the direct path. Target to receiving station 2 has one direct path and one reflected path, with the reflected path being three times the length of the direct path. The direct positioning spectrum obtained by the method of this invention in this scenario is as follows. Figure 2 As shown, there is a clear peak at the target location, proving that the method can effectively achieve target localization in non-line-of-sight scenarios.

[0063] This invention uses a gridded search to reconstruct the target signal by utilizing the channel model from the grid points to different receiving stations. It finds grid points with the same reconstructed signal to locate the target. This method can convert multipath channels into usable information, solving the problem that existing passive positioning methods cannot accurately locate radiation source targets in non-line-of-sight environments. It can even achieve positioning in single-station or dual-station scenarios, breaking through the limitation of traditional positioning models that require more than three positioning stations.

Claims

1. A passive non-line-of-sight direct localization method based on a channel model, characterized in that, The specific process is as follows: Step 1: Establish the objective function for passive non-line-of-sight direct localization based on the maximum likelihood criterion; Step 2: Reconstruct the signal transmitted at the target point based on the channel model; Step 3: Perform a gridded search on the area to be tested. Calculate a corresponding reconstructed signal for each grid point and substitute it into the passive non-line-of-sight direct localization objective function. Record the grid point corresponding to the maximum value of the passive non-line-of-sight direct localization objective function as the target point location, thus completing the non-line-of-sight direct localization based on the channel model.

2. The passive non-line-of-sight direct positioning method based on a channel model according to claim 1, characterized in that, The specific method for step 1 is as follows: ； Where M represents the number of receiving stations. Indicates taking the real part; This indicates the signal transmitted from the target point corresponding to the m-th receiving station; This represents the received signal from the m-th receiving station; This represents the total number of propagation paths corresponding to the m-th receiving station. This represents the attenuation of each propagation path at the m-th receiving station; For the reason The diagonal matrix formed , This represents the k-th frequency point of the signal received by the receiving station, where K is the number of frequency points of the signal received by the receiving station. Let j represent the time delay of the l-th propagation path at the m-th receiving station, where j represents the imaginary unit. For the preset threshold, .

3. The passive non-line-of-sight direct positioning method based on a channel model according to claim 1, characterized in that, The specific method for step 2 is as follows: ； ； in, This indicates the signal transmitted from the target point corresponding to the m-th receiving station; This represents the received signal from the m-th receiving station; M represents the number of receiving stations. This is an L x N matrix, where L is the number of sampling points for the signal received at the receiving station, and N is the number of sampling points for the signal transmitted at the target point. The letters at the top of the matrix are used to indicate the number of sampling points. Columns, non-matrix element values , , This represents the rounding function, where the first row of the matrix, the second row, and the third row are rounded up. All elements in the column are 1, starting from the first row and the second row. The elements at the bottom right diagonal (45° angle) of the column are all 1, and the values ​​of all other elements in the matrix are 0. The sampling rate of the received signal at the receiving station; This represents the time delay of the l-th propagation path at the m-th receiving station.

Images