A near-field spatial spectrum retrieval method based on damage characteristics
By constructing a multi-dimensional mapping relationship and Trie search tree based on damage features, the problem of vector distortion of spatial spectral features is solved, the recall rate and positioning efficiency are improved, the search space is reduced, and more efficient target positioning is achieved.
Patent Information
- Application Number
- CN202411239904.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-05
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2044-09-05
AI Technical Summary
The feature vectors of the measurement spatial spectrum are easily affected by various factors and cause distortion, resulting in low recall under the traditional search mechanism and huge search space, making it difficult to accurately locate the target.
Using the near-field spatial spectrum search method based on damage features, the search vector is constructed by selecting feature elements from the feature vectors of the template spatial spectrum to form a multi-dimensional mapping relationship and a Trie search tree. The depth-first strategy is used to query the phase in the damage feature search tree. Match search vectors, merge duplicate results to narrow the search subspace.
It improves the recall rate, reduces the search subspace, reduces the amount of complex correlation calculations during the positioning process, and improves the real-timeness of target positioning.
Smart Images

Figure CN118779641B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of data processing, and in particular to a near-field spatial spectrum retrieval method based on damage characteristics. Background Art
[0002] Target positioning can be achieved by utilizing the independence between the Fresnel near-field spatial spectra in different regions. We call the near-field spatial spectrum obtained by actual measurement the measurement spatial spectrum, and the spatial spectrum obtained by simulation for target position comparison the template spatial spectrum. By extracting the query features from the measurement spatial spectrum, the template spatial spectrum related to it can be retrieved to narrow the target position search space. However, the measurement spatial spectrum is easily affected by the Fresnel field superposition effect of multiple targets, which makes the query feature vector of the measurement spatial spectrum easily damaged, and generally only some feature elements can be retained. Using each feature element for retrieval separately will result in too many retrieval results and a huge search space. Using the entire damaged feature vector as the query vector makes it difficult to retrieve the real target. Summary of the invention
[0003] In order to solve the problem that the feature vector is easily distorted by various factors and leads to low retrieval recall rate under the traditional retrieval mechanism using measured spatial spectrum features, a near-field spatial spectrum retrieval method based on damage features is proposed in the present invention.
[0004] In order to achieve the above object, the technical solution adopted by the present invention is as follows:
[0005] A near-field spatial spectrum retrieval method based on damage characteristics comprises the following steps:
[0006] Step 1: From the template space spectrum The characteristic vector i q =[ i 1 q , i 2 q … i N q ] T Select any The combination of characteristic elements constitutes the retrieval vector i c q =[ i 1 q , i 2 q … i j q … i J q ] T , i j q ∈ i q , , , each template space spectrum Total search vectors; among them, , is the feature vector The number of characteristic elements in , is the number of potential target locations, c ∈[1, C N J ] ;
[0007] Step 2: For each search vector , forming a search record ,in c c q =[ x y z ] T It is the template space spectrum The coordinates of each template space spectrum correspond Search records ;
[0008] Step 3: Based on all the search vectors Characteristic elements Different combinations of template space spectrum Search records Classify and construct inverted index data subsets between different feature combinations and template space spectra to obtain and The multi-dimensional mapping relationship between: , forming from 1 feature to A retrieval database for multi-layer classification of features ;
[0009] Step 4: Retrieve vector All characteristic elements in The set is used as the alphabet of the Trie retrieval mechanism to construct a Trie retrieval tree. The root node of the Trie retrieval tree is an empty node. Each node in the nth layer contains n characteristic elements; among them, n ;
[0010] Step 5: According to the different feature combinations corresponding to each node in the Trie retrieval tree constructed in step 4, the inverted index data subset corresponding to the corresponding feature combination in step 3 is loaded to form a damage feature retrieval tree with data set retrieval capability;
[0011] Step 6: Measure the spatial spectrum The characteristic vector i e =[ i 1 e , i 2 e … i N e ] T Select any elements, forming the query vector i c e =[ i 1 e , i 2 e … i j e … i J e ] T , i j e ∈ i e , i j e < i j+1 e , c ∈ 1, C N J , J ≤ N , measuring the spatial spectrum Total query vector;
[0012] Step 7, according to query vector , in the damage feature retrieval tree, using the mapping relationship From the search database The search vector that matches the query , get the query record from the reverse sorted data subset ,Depth-first strategy is adopted in the query process;
[0013] Step 8: Utilize all query records Coordinate data of Get the corresponding template space spectrum , and after merging the repeated template space spectra in all query results, we get the search subspace .
[0014] Due to the adoption of the above technical solution, the beneficial effects of the present invention compared with the prior art are:
[0015] The present invention uses the damage characteristics of the measured spatial spectrum to construct a multidimensional query mechanism. It solves the problem that the feature vector is easily distorted by various factors and leads to low retrieval recall rate under the traditional retrieval mechanism using measured spatial spectrum features. On the basis of ensuring the recall rate, it effectively narrows the search subspace and reduces the complex correlation calculation amount in the positioning process, thereby improving the real-time performance of target positioning. BRIEF DESCRIPTION OF THE DRAWINGS
[0016] Figure 1 Schematic diagram of a radio transceiver array device in an embodiment of the present invention.
[0017] Figure 2 Graph showing the spatial spectrum distribution of the Fresnel near field in an embodiment of the present invention.
[0018] Figure 3 Graph showing the distribution of the template spatial spectrum in the Fresnel near-field spatial spectrum in an embodiment of the present invention.
[0019] Figure 4 The figure is a processing flow chart of a near-field spatial spectrum retrieval method based on damage characteristics in an embodiment of the present invention.
[0020] Figure 5 This is a diagram showing the classification effect of the first-layer nodes in the damage feature retrieval tree in an embodiment of the present invention.
[0021] Figure 6 This is a diagram showing the classification effect of the second-layer nodes in the damage feature retrieval tree in an embodiment of the present invention.
[0022] Figure 7 This is a diagram showing the classification effect of the third-layer nodes of the damage feature retrieval tree in an embodiment of the present invention.
[0023] Figure 8 This is a diagram showing the classification effect of the fourth-level nodes of the damage feature retrieval tree in an embodiment of the present invention.
[0024] Fig. 9 Schematic diagram of damage feature retrieval tree in an embodiment of the present invention. DETAILED DESCRIPTION
[0025] The content of the present invention is further described below in conjunction with the accompanying drawings and specific embodiments. Before data processing, a template library data set of template spatial spectrum is prepared. In this embodiment, the near-field spatial spectrum retrieval method based on damage characteristics is executed by an electronic device such as a computer or the hardware and / or software therein. Figure 1 As shown in the figure, it is a schematic diagram of the radio transceiver array device in this embodiment, assuming that the three-dimensional aperture of the radio transceiver array device is , 100 radio transceivers are randomly arranged inside, working in the L-band, working in the L-band 1.25GHz. All radio transceivers are ideally synchronized in time, frequency and space, and various phase zero values are accurately corrected, detecting near-field targets in a fully coherent state, Figure 1 Where λ is the working wavelength of the radio transceiver, ~ means approximately, and the solid dots represent the radio transceivers randomly distributed in the parallelogram area; Figure 2 As shown in FIG. 1 , it is the measurement spatial spectrum in this embodiment, and its geometric center is the position of the detected target. Figure 3 As shown, in this embodiment, about Q=5.124×10 5 The dotted box represents a template spatial spectrum. The angular spacing between potential target points is 0.002625 degrees, which is greater than 1.5 times the array beam width. This spacing can ensure the independence between template spatial spectra.
[0026] By modeling the radio transceiver array device and placing a unit radiation source at each potential target point , the array receiving signal is obtained by simulation. Sampling matrix It consists of 1024 steering vectors. The angular interval between the steering vectors is greater than 0.002625 degrees to ensure their independence. Weighted, obtain the template space spectrum corresponding to each potential target point position .all Constructing the template library dataset .
[0027] A near-field spatial spectrum retrieval method based on damage characteristics, such as Figure 4 As shown, the following steps are included:
[0028] Step 1: Obtain the spatial spectrum of each template The characteristic vector of the characteristic element i q =[ i 1 q , i 2 q … i 15 q ] T , ; From the feature vector i q =[ i 1 q , i 2 q … i 15 q ] T Select any The combination of characteristic elements constitutes the template space spectrum A search vector i c q =[ i 1 q , i 2 q , i 3 q , i 4 q ] T , i j q ∈ i q , , , c ∈[1, C 15 4 ] , , is the number of potential target locations. ; Each template space spectrum Total search vectors; among them, The constraints combine the search vectors in The elements in the search are arranged in order from small to large, which standardizes the expression of search features and improves search efficiency.
[0029] Step 2: For each search vector , forming a search record ,in c c q =[ x y z ] T , x , y∈ -5000,5000 ,z∈[-25,25] It is the template space spectrum The coordinates of each template space spectrum correspond Search records .
[0030] Step 3: Based on all the search vectors Characteristic elements Different combinations of template space spectrum Search records From 1 feature to Each feature combination forms a subset of the inverted index data for the retrieved records, such as Figure 5 As shown, Each index of the layer corresponds to more than 25,000 template space spectra at most; Figure 6 As shown, Each index of the layer corresponds to a maximum of about 1750 template space spectra; Figure 7 As shown, Each index of the layer corresponds to about 140 template space spectra at most. and The multi-dimensional mapping relationship between: , forming from 1 feature to A retrieval database for multi-layer classification of features .Depend on Figure 8 visible, The characteristics of the layers effectively complete the division of the search subspace, and the number of records in each subspace is less than a dozen.
[0031] Step 4: construct a retrieval mechanism based on damage features based on the Trie retrieval mechanism. All characteristic elements in The set of components is used as the alphabet of the Trie retrieval mechanism to construct The alphabet (characteristic elements) of the Trie retrieval tree is 716. The root node of the Trie retrieval tree is an empty node. Each node in the first layer contains 1 characteristic element. Each node in the second layer contains 2 characteristic elements. And so on. Each node in the layer node contains The schematic diagram of the retrieval mechanism based on damage features is as follows: Fig. 9 shown.
[0032] Step 5, such as Fig. 9 As shown in the box in , the Trie retrieval tree expands the inverted index data subset to form a damage feature retrieval tree. The depth of the damage feature retrieval tree is 4.
[0033] Step 6: The radio transceiver array device transmits a coherent signal to the observation space, performs target detection, obtains array reception data, and processes the array reception data through the same sampling matrix as the template spatial spectrum. Perform weighting to obtain the measured spatial spectrum . For measuring spatial spectrum Get the The characteristic vector of the characteristic element i e =[ i 1 e , i 2 e … i 15 e ] T ; Select any elements, forming the query vector i c e =[ i 1 e , i 2 e , i 3 e , i 4 e ] T , , measuring the spatial spectrum Total query vector.
[0034] Step 7, Step 7, according to the user query vector , in the damage feature retrieval tree, using the mapping relationship From the search database The search vector that matches the query , get the query record from the reverse sorted data subset ,The depth-first strategy is adopted in the query process.
[0035] Step 8: Utilize all query records Coordinate data of Get the corresponding template space spectrum , and after merging the repeated template space spectra in all query results, we get the search subspace .
[0036] In summary, the present invention aims to solve the problem of random false features when the measured spatial spectrum is affected by multiple targets, and constructs a retrieval mechanism based on damage features, which effectively improves the recall rate, narrows the search subspace, reduces the amount of complex correlation calculations in the positioning process, and improves the real-time performance of target positioning.
Claims
1. A near-field spatial spectrum retrieval method based on damage characteristics, characterized in that: The following steps are involved: Step 1: From the template space spectrum The characteristic vector Select any The combination of characteristic elements constitutes the retrieval vector , , , each template space spectrum Total search vectors; among them, , is the feature vector The number of characteristic elements in , is the number of potential target locations, ; Step 2: For each search vector , forming a search record ,in It is the template space spectrum The coordinates of each template space spectrum correspond Search records ; Step 3: Based on all the search vectors Characteristic elements Different combinations of template space spectrum Search records Classify and construct inverted index data subsets between different feature combinations and template space spectra to obtain and The multi-dimensional mapping relationship between: , forming from 1 feature to A retrieval database for multi-layer classification of features ; Step 4: Retrieve vector All characteristic elements in The set is used as the alphabet of the Trie retrieval mechanism to construct a Trie retrieval tree. The root node of the Trie retrieval tree is an empty node. Each node in the nth layer contains n characteristic elements; among them, n ; Step 5: According to the different feature combinations corresponding to each node in the Trie retrieval tree constructed in step 4, the inverted index data subset corresponding to the corresponding feature combination in step 3 is loaded to form a damage feature retrieval tree with data set retrieval capability; Step 6: Measure the spatial spectrum The characteristic vector Select any elements, forming the query vector , measuring the spatial spectrum Total query vector; Step 7, according to query vector , in the damage feature retrieval tree, using the mapping relationship From the search database Search vectors matching the query , get the query record from the reverse sorted data subset ,Depth-first strategy is adopted in the query process; Step 8: Utilize all query records Coordinate data of Get the corresponding template space spectrum , and after merging the repeated template space spectra in all query results, we get the search subspace .
