1. A joint iteration tomography method based on dual-frequency narrow-band signals is characterized by comprising the following steps:

s1, scanning an unknown scene from a plurality of angles and positions by using a receiving and transmitting split node pair to acquire a dual-frequency narrow-band signal penetrating the scene;

s2, establishing a signal model according to the relation between the power of the received double-frequency signal and a system matrix;

s3, based on the established signal model, respectively obtaining imaging results of respective frequency bands by using a joint iterative imaging method;

and S4, carrying out incoherent fusion on the imaging result obtained by the dual-frequency narrow-band signal through an arithmetic fusion strategy to obtain a final imaging result.

2. The joint iterative tomography method based on dual-band narrowband signals according to claim 1, wherein in the scanning process, in step S1, the antennas of the transmitting node and the receiving node in the transceiving split node pair are always kept opposite.

3. The joint iterative tomography method based on dual-band narrowband signals according to claim 2, wherein the transceiving split nodes at each angle in step S1 completely cover the unknown scene in the area formed between the sampling paths.

4. The method as claimed in claim 3, wherein the transceiver node scans the unknown region by moving synchronously and equally along the sampling path with corresponding angle.

5. The method of claim 4, wherein the sampling interval is greater than or equal to the imaging grid size.

6. The method according to claim 5, wherein the signal model expression in step S2 is:

△P≈AO

wherein, Δ P is the attenuation power matrix after the received signal power and the received signal power of the null scene are cancelled, and [ [ Δ P ] ]₁,...,△P_M]^TThe superscript T represents transposition, and M is the total number of sampling points; o represents the attenuation rate matrix after the unknown region is dispersed into a plurality of cells, and O is ═ O₁,...,O_N]^TIn which O is_nIs the decay rate of cell N, N is 1,2,3, …, N; a is a system matrix of order M × N.

7. The joint iterative tomography method based on dual-band narrowband signals according to claim 6, wherein the step S3 further comprises: and in each iteration process, optimizing an iteration result by using the improved total variation constraint.

Background

The technical field of radio frequency imaging is mainly to detect an interested unknown area by using radio frequency signals, including the positioning and imaging of physical targets, the imaging of architectural layouts and the like, and has important application value in the fields of anti-terrorism stability maintenance, disaster relief, medical monitoring and the like. In urban environments, targets are often occluded by physical structures such as walls. In order to obtain an accurate target state, it is necessary to obtain information of a building structure or an obstacle, and the research has attracted considerable attention in recent years.

Synthetic Aperture Radar (SAR) imaging is used as a traditional imaging strategy, and projection imaging is performed by using a panoramic reflection echo of a layout structure. The united states army research laboratory scans from both sides of a Building with abandoned barracks using an on-board synthetic aperture Radar system and combines the results of different viewing angles to obtain a complete Building layout (c.le, t.dogaru, l.nguyen and m.a.resler, "ultrawide band (uwb) radio Imaging of Building interfaces: measures and Predictions," IEEE Transactions on Geoscience and remove Sensing, vol.47, No.5, pp.1409-1420, may. 2009.). China national defense science and technology university adopts a form that a Radar system transmits an ultra wide band signal, constructs a distance Doppler frequency spectrum of a Layout structure, and designs a constant false alarm rate detector to extract a Building Layout (Y.Song, J.Hu, N.Chu, T.jin, J.Zhang and Z.ZHou, 'Building Layout Reconstruction in a centralized Human Target Sensing UWB MIMO Through-Wall Imaging Radar,' IEEE Geoscience and Remote Sensing Letters, 15.15, No.8, pp.1199-1203, and aug.2018.). Although the above studies can obtain better layout structure imaging results, the above studies are still limited by the factors of complex system, large volume, high cost, and the like.

In recent years, the work of reconstructing the layout structure using the power of the received signal has been gradually developed. The university of california of the united states utilizes a radar node with a transmitting and receiving split position to scan unknown scenes from different positions and acquire the power of a received signal, and proposes a layout structure reconstruction method Based on image total variation minimization (Y.Mostofi, "Cooperative Wireless-Based Object/Object Mapping and See-Through Capabilities in Wireless Networks," IEEE Transactions on Mobile Computing, vol.12, No.5, pp.817-829, May.2013.). The Chinese academy of sciences proposes a re-weighted Total Variation and Prior Information regularization algorithm (Q.Guo, Y.Li, X.Liang, J.Dong and R.Cheng, "Through-the-Wall Image Reconstruction of the heavy Total Variation and the priority Information in the Radio geographic Imaging," IEEE Access, vol.8, pp.40057-40066,2020), which is based on the minimization of the Total Variation, and considers the Prior Information that the Wall only faces the horizontal direction or the vertical direction, and is used for keeping the direction of the Wall. However, the above studies only consider imaging of a single-band narrow-band signal, and due to factors such as channel jitter, artifacts and blurred edges of a layout structure may occur in the imaging result, thereby affecting the imaging quality. Therefore, the effective layout structure imaging method can solve the problems of artifacts and imaging structure edge blurring, and has important research significance.

Disclosure of Invention

In order to solve the technical problem, the invention provides a joint iteration tomography method based on dual-frequency narrow-band signals, which can accurately and effectively realize the imaging of unknown region layout structures or objects.

The technical scheme adopted by the invention is as follows: a joint iteration tomography method based on dual-frequency narrow-band signals comprises the following steps:

s1, scanning an unknown scene from a plurality of angles and positions by using a receiving and transmitting split node pair to acquire a dual-frequency narrow-band signal penetrating the scene;

s2, establishing a signal model according to the relation between the power of the received double-frequency signal and a system matrix;

s3, based on the established signal model, respectively obtaining imaging results of respective frequency bands by using a joint iterative imaging method;

and S4, carrying out incoherent fusion on the imaging result obtained by the dual-frequency narrow-band signal through an arithmetic fusion strategy to obtain a final imaging result.

Step S1 is to keep the antenna of the transmitting node and the antenna of the receiving node in the transceiving split node pair aligned with each other all the time during the scanning process.

Step S1 shows that the area formed between the sampling paths by the transceiving node at each angle completely covers the unknown scene.

And the receiving and transmitting split node pair synchronously moves at equal intervals along the sampling path with the corresponding angle to scan the unknown area.

The sampling interval is greater than or equal to the imaging grid size.

The signal model expression in step S2 is:

△P≈AO

wherein, Δ P is the attenuation power matrix after the received signal power and the received signal power of the null scene are cancelled, and [ [ Δ P ] ]₁,...,△P_M]^TThe superscript T represents transposition, and M is the total number of sampling points; o represents the attenuation rate matrix after the unknown region is dispersed into a plurality of cells, and O is ═ O₁,...,O_N]^TIn which O is_nIs the decay rate of cell N, N is 1,2,3, …, N; a is a system matrix of order M × N.

Step S3 further includes: and in each iteration process, optimizing an iteration result by using the improved total variation constraint.

The invention has the beneficial effects that: firstly, scanning an unknown area from a multi-angle and multi-position by utilizing a transceiving split wireless node pair according to a planned sampling point position, acquiring a dual-frequency narrow-band signal penetrating a scene, and establishing a signal model; then, according to the relation between the dual-frequency narrow-band signal attenuation power and the established system matrix, acquiring the imaging results of respective frequency bands by respectively utilizing a joint iterative imaging method; secondly, optimizing an iteration result by utilizing improved total variation constraint in each iteration process of the imaging method; finally, carrying out incoherent fusion on the imaging result obtained by the dual-frequency narrow-band signal through an arithmetic fusion strategy to obtain a final imaging result; the method of the invention has the following advantages:

1. the imaging test system has low complexity, low cost, high flexibility and good concealment;

2. the artifact problem and the structure edge blurring problem during single-frequency band narrow-band signal imaging are obviously improved;

3. accurately and effectively realizing the imaging of an unknown region layout structure or an object;

4. the invention can be applied to the fields of anti-terrorism maintenance, disaster relief, medical monitoring and the like.

Drawings

Fig. 1 is a schematic diagram of a wireless node pair operating in a transceiving mode.

FIG. 2 is a schematic diagram of a test scenario and test system in an embodiment;

wherein, fig. 2(a) is a schematic view of a test scene and a test system of a T-shaped brick wall structure; fig. 2(b) is a schematic view of an L-shaped brick wall structure test scene and a test system.

FIG. 3 is a diagram illustrating the received power (unit: dB) of the original dual-frequency signal obtained for a test scenario in an exemplary embodiment;

wherein, fig. 3(a) is the receiving power (unit: dB) of 2.4GHz and 5GHz narrowband signals obtained under the "T" shaped test scene; FIG. 3(b) shows the received power (unit: dB) of 2.4GHz and 5GHz narrow-band signals obtained in the L-shaped test scene.

Fig. 4 is a normalized attenuation power of a dual-frequency signal and an ideal model obtained for a test scenario in an embodiment;

wherein, fig. 4(a) is normalized attenuation power of the dual-frequency signal and the ideal model in a "T" shaped test scene; fig. 4(b) is normalized attenuation power of the dual-frequency signal and the ideal model in the L-shaped experimental scene.

Fig. 5 is a reconstruction result of each of the dual-frequency signals of the test scene in the specific embodiment;

wherein, fig. 5(a) is the imaging result of the T-shaped test scene under the 2.4GHz narrowband signal; FIG. 5(b) is the imaging result of the T-shaped test scene under the 5GHz narrow-band signal; FIG. 5(c) shows the imaging result of the L-shaped test scene under the 2.4GHz narrow-band signal; FIG. 5(d) shows the imaging result of the L-shaped test scene under the 5GHz narrow-band signal.

FIG. 6 is an imaging result of a test scene under dual-frequency signal fusion in an embodiment;

wherein, fig. 6(a) is an imaging result of a T-shaped test scene under dual-frequency signal fusion; fig. 6(b) is an imaging result of an "L" shaped test scene under dual-frequency signal fusion.

Detailed Description

The following description of the embodiments of the present invention is given based on a test in a "T" scenario:

the working diagram of the wireless node pair of the invention is shown in fig. 1, the wireless node pair comprises a group of transmitting and receiving nodes, the wireless node pair synchronously moves along a planned path outside an unknown area, taking a 0-degree sampling path as an example, the wireless node pair comprises a transmitting node moving path (a → b) and a receiving node moving path (c → d) at the front side and the rear side of the unknown area shown in fig. 1, the antennas of the transmitting and receiving nodes are always opposite, and an area formed between the sampling paths needs to cover the unknown area to be imaged.

The corresponding 90-degree sampling paths include the transmit node movement path and the receive node movement path on the left and right sides of the unknown area shown in fig. 1.

In this embodiment, the content of the present invention is described by taking the test scenario shown in fig. 2(a) as an example, and the specific implementation steps are as follows:

step 1: initialization of parameters for unknown regions and wireless nodes

For a T-shaped brick wall structure scene of an unknown area, a test scene is shown in fig. 2(a), the two-dimensional size of the T-shaped structure built by standard red bricks is 2m multiplied by 2m, and the thickness of a wall body is 11.5 cm. The invention considers the two-dimensional tomography of the scene, the size of the imaged scene is selected to be 3m multiplied by 3m, and the size of the imaging grid is set to be 0.02 m. In consideration of the sampling speed and the data complexity, the unknown region is scanned by using two sampling paths of 0 degree and 90 degrees (the sampling paths are as shown in fig. 1). The sampling interval is typically chosen to be greater than or equal to the imaging grid size, with sampling being chosen experimentallyThe interval is 0.04m and the number of samples per path is 79. The total number of grid points in the present embodiment isThe total sampling point number of the 2 sampling paths is 158, namely the actual measurement point number of the invention only accounts for 0.7 percent of the total area point number. By adopting the method, only a small amount of sampling data is needed to obtain the two-dimensional tomography result of the unknown region. The center frequency of the dual-frequency signal of the wireless node is selected to be 2.4GHz and 5GHz with reference to the frequency band of the WiFi signal, and the bandwidth of the dual-frequency signal is selected to be 20 MHz.

Those skilled in the art will note that the method of the present invention is not limited to "T" and "L" shaped architectural scenarios as shown in fig. 2. The rest of the scene may be transparent to electromagnetic waves (and the physical structures contained within the scene may cause attenuation of electromagnetic waves).

Step 2: signal modeling and dual frequency signal propagation analysis

2.1, as shown in fig. 1, with the center of the unknown area as the center point, the transmitting node starts sampling from position a and the receiving node simultaneously receives the signal at position c. The sampling interval is 0.04m until the transmitting and receiving nodes move to b and d respectively, and 0-degree path sampling is completed. Similarly, 90-degree path sampling is completed. The received power (unit: dB) of the original dual frequency signal obtained in the experimental scenario of fig. 2 is shown in fig. 3. The normalized attenuation power of the resulting dual frequency signal and the ideal model is shown in fig. 4.

2.2, according to the Rytov approximate model, the relationship between the actual received signal power and the incident field power at the position r can be expressed as:

P(r)＝Pⁱⁿ(r)+10Im(φ(r))lg(e^-2) (1)

where P (r) represents the actual received signal power at r, Pⁱⁿ(r) is the incident field power at r, and Im () represents the imaginary part, parameterWherein O (r ') represents the decay Rate at position r', Eⁱⁿ(r') represents the incident signal at position rAnd g (r, r') is a green function. The whole area is discretized into 150 × 150 grids and r is used_nTo represent the nth trellis, where n e { 1. Assuming that the total number of samples is M, at the ith sample, the transmitting and receiving nodes are denoted as T respectively_iAnd R_i. Considering the joint M sampling points, a matrix Φ is obtained, which is expressed as:

wherein an element may be expressed as:

where Δ V represents the volume value of a discrete grid, typically taken as 1.

Equation (2) can be expressed as:

Φ＝-jAO (4)

wherein A is system matrix of M × N order, and one element is A_i,n＝g(R_i,r_n)Eⁱⁿ(r_n)△V/Eⁱⁿ(R_i). In summary, when formula (1) is substituted for formula (4), the attenuation power caused by the object can be obtained, and can be expressed as:

△P＝Im(Φ)≈AO (5)

wherein Δ P ═ P ((P-P)_in)/10lg(e^-2) P and P)_inA received signal power matrix and an incident signal power matrix representing positions of all receiving nodes, respectively, where Δ P [ Δ P ]₁,...,△P₁₅₈]^TFor normalized decaying power matrix, O ═ O₁,...,O₂₂₅₀₀]^TFor the decay rate of a discrete grid, a is a system matrix of order 158 × 22500.

And 2.3, three main difference points are obtained by analyzing the electromagnetic propagation difference of the double-frequency signals in the scene, so that the attenuation power curves of the signals in different frequency bands are different, and finally, the imaging result shows that the problems of different artifacts, edge fuzzy extension and the like exist in single-frequency-band signal imaging. Firstly, the penetration attenuation is different when signals of different frequency bands penetrate the same medium, and can be expressed as follows:

where α is an attenuation constant, Re () represents a real part, ω is an angular frequency, μ₀And ε₀Respectively representing the conductivity and relative permittivity, σ, of free space_eAnd ε_r' are the real parts of the effective conductivity and relative permittivity, respectively, of the medium, and gamma denotes the complex transmission constant.

Secondly, multipath fading, which is different when signals of different frequency bands are transmitted, can be expressed as:

wherein h (t) is the equivalent channel of the received signal, C (t) is the number of received signal paths at time t, alpha_c(t) represents the attenuation term, φ, of the c-th signal path at time t_c(t) shows the Doppler shift, τ, of the c-th signal path at time t_c(t) denotes the time delay of the c-th signal path at time t, f_cThe center frequency of the signal.

Finally, the diffraction effect of the edge diffraction effect is related to the frequency when the electromagnetic wave propagates near the wall corner (dihedral angle), and the diffraction field can be expressed as:

wherein R is₀As the position of the observation point, Eⁱⁿ(Q) represents the incident field at diffraction point Q,as the dyadic edge diffraction coefficient, A(s) is the amplitude attenuation coefficient,λ represents the wavelength of the signal and S is the length of the diffraction spot in the direction of the ray.

The above factors all cause different power curves of the received signals under the narrowband signals of different frequency bands. The received power of the dual-frequency signal and the normalized attenuation power of the dual-frequency signal and an ideal model under the two types of scenes given in the invention are shown in fig. 3 and 4. Therefore, in the tomography results of signals of different frequency bands in the prior art, the positions where the problems such as artifacts occur have differences, and the imaging result is shown in fig. 5; the invention can overcome the problem by adopting a dual-frequency signal fusion imaging method, effectively reduces artifacts and retains clear object edge structure information.

It should be noted by those skilled in the art that the attenuation power mentioned in the present invention means that when any group of wireless nodes scans an unknown area, the received signal power at this time is subtracted from the received signal power corresponding to an empty scene, and it is required to ensure that the absolute distances between the transmitting node and the receiving node are the same, and the obtained attenuation power is the power attenuation caused by the object in the unknown area.

And step 3: dual-frequency signal fusion imaging

And 3.1, based on the signal model, providing a joint iterative imaging method considering total variation constraint. Imaging an unknown region, setting an initial imaging result as a 0 matrix, and then combining data values sampled for M times to iteratively update the numerical value of each grid, wherein the updating process can be expressed as:

wherein, O_n ⁽⁰⁾Indicating the initial imaging results. Finishing the updating of all grids, namely finishing a round of iterative updating, adding total variation constraint in each round of iterative updating process, and optimizing an imaging result, which can be expressed as:

min||O||_TV+βf(O,ε),s.t.△P＝AO (10)

wherein, beta represents a weight factor, which is generally taken as 0.5, f (O, epsilon) represents the difference between a certain element in O and the element closest to the element in epsilon, and epsilon represents the theoretical relative dielectric constant vector of the medium.

The optimized imaging result is updated for the previous iteration as the initial value (i.e., O) for the next iteration_n ⁽⁰⁾) Continuing the operations of the formulas (9) to (10) until a termination condition is met, namely the two norms of the difference value between the imaging result and the previous round result are less than or equal to a certain value which is set autonomously, terminating iteration and obtaining a final imaging result; the value to be set autonomously is mainly determined according to the requirement of imaging accuracy, and is generally selected to be about 0.5.

And 3.2, carrying out incoherent fusion on the imaging results respectively obtained by the dual-band narrow-band signals by a multiplication-subtraction-addition fusion strategy to obtain a final imaging result. First, the joint multiplication operation can be expressed as:

wherein, O₁(x, y) and O₂(x, y) respectively indicate pixel values of the imaging result obtained by the dual-frequency signal at (x, y). Operating in tandem with decreasing to obtain O_S(x, y), which can be expressed as:

the final imaging result is obtained by the joint addition operation:

wherein, O_A(x, y) characterize the final imaging result based on the dual frequency signal.

In the exemplary "T" and "L" shaped brick wall structure test scenario of the present invention, fig. 5 shows the imaging results of the narrowband signals with 2.4GHz and 5GHz as the center frequencies, respectively, and fig. 6 shows the imaging results based on the dual-band narrowband signal of the present invention. It can be obviously seen that the tomography result of the single-frequency band signal has more artifacts and edge blurring problems, and whether the object structure is the object structure cannot be judged, which affects the imaging quality. The test result of the algorithm provided by the invention shows that clearer structural edge information is reserved in the imaging result, the problem of artifacts is effectively reduced, and the correctness and the effectiveness of the algorithm are further verified.

It will be appreciated by those of ordinary skill in the art that the embodiments described herein are intended to assist the reader in understanding the principles of the invention and are to be construed as being without limitation to such specifically recited embodiments and examples. Various modifications and alterations to this invention will become apparent to those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the scope of the claims of the present invention.