Ship RCS feature extraction method in ship lock based on 5g-a multi-path inversion

Abstract

The application discloses a ship RCS feature extraction method based on 5G-A sensing and multi-path inversion, and the method comprises the following steps: a 5G-A base station transmits a sensing reference signal to a ship lock and receives a multi-path echo signal generated by scattering of a ship in a lock chamber and reflection of a lock chamber wall and a water surface in the lock chamber; after the multi-path echo signal is demodulated by OFDM, the channel frequency response is estimated on each subcarrier and inverse Fourier transform is performed to obtain a channel impulse response, and then a TOA measurement value and a received power of each propagation path are extracted; according to the real position of the 5G-A base station, the mirror source coordinates of the 5G-A base station about each reflection surface of the lock chamber are found; all the mirror source coordinates and the real position of the 5G-A base station jointly constitute a virtual anchor point set; the TOA measurement value of each propagation path is matched with the virtual anchor point set by nearest neighbor correlation, and then the one-way propagation distance of each propagation path is calculated; and the ship RCS feature is obtained based on the monostatic radar equation inversion.

Description

Technical Field

[0001] This invention relates to a method for extracting RCS features of ships within locks based on 5G-A integrated sensing and multipath inversion, belonging to the field of ship monitoring equipment and radio frequency sensing technology. Background Technology

[0002] The radar cross section (RCS) of a ship is a key physical parameter characterizing its electromagnetic scattering properties. Different ships, due to variations in materials (such as steel, aluminum alloy, and fiberglass), geometry, and superstructure, exhibit stable and distinguishable RCS frequency and angular responses under radio frequency electromagnetic wave illumination. This characteristic can serve as a ship's radio frequency fingerprint for ship type identification and lock-passing status assessment, effectively supplementing perception dimensions that optical sensors (such as lidar and visible light cameras) cannot cover. Specifically, in the semi-enclosed space of a lock, the length, width, wall height, and water level are all known engineering parameters or real-time measurable values. When electromagnetic waves propagate in such geometrically known spaces, in addition to the direct path, they undergo multiple reflections through the lock walls and water surface. According to the image source theorem, each reflection path can be mapped to a virtual anchor point in space, thus enabling accurate modeling and separation of multipath signals. This multipath controllability arising from known geometric constraints provides unique conditions for separating path components from complex echoes and inverting the target's physical scattering parameters.

[0003] Currently, the following technical solutions exist for the measurement and identification of ship RCS: Technology 1: Ship RCS Measurement System Based on Dedicated Radar These technologies rely on dedicated radar measurement equipment. For example, a Chinese patent application for a method, apparatus, and measurement system for measuring the RCS value of maritime targets (Publication No. CN118642097A, Publication Date 2024.09.13) uses radar and other devices to form a measurement system to measure the RCS value of maritime ship targets. Another example is a Chinese patent application for a method for measuring the RCS of targets and classifying them under moving conditions using shipborne radar (Publication No. CN1169 73864A, Publication Date 2023.10.31). Under moving conditions of the shipborne radar, it first collects background sea clutter data when there are no targets, and then performs RCS measurement and target classification. Furthermore, traditional RCS assessment often relies on field measurements. However, the common drawbacks of the above solutions are: they all require the deployment of dedicated radar hardware, resulting in complex systems and high deployment costs; background clutter spectrum needs to be collected or electromagnetic simulations performed before measurement, making it difficult to meet the real-time online monitoring requirements in lock scenarios; and none of them involve utilizing the integrated sensing capabilities of 5G-A communication base stations.

[0004] Technique 2: Ship Target Recognition Method Based on RCS Sequence This type of technology focuses on deep learning classification of acquired RCS sequences. For example, the Chinese patent application "Method, Apparatus and Electronic Equipment for Ship Target Recognition Based on RCS Sequence" (Publication No. CN117591922A, Publication Date 2024.02.23) inputs the acquired RCS sequence into an improved OS-CNN model for target recognition. The core assumption of this method is that the RCS value has been pre-obtained by a traditional radar system. Its innovation lies in the classifier design, rather than RCS inversion at the signal level. Its shortcomings are: the RCS data source still relies on the traditional radar system and does not involve 5G-A integrated sensing capabilities; and this method is not suitable for RCS extraction in multipath environments such as ship locks, and cannot solve the technical problem of how to separate the path components and accurately invert the RCS value in complex multipath echoes.

[0005] Technology 3: RCS Sensing in 5G-A Sensory Integration Currently, there are technologies available for verifying the perception capabilities of low-altitude targets (such as drones) based on 5G-A integrated sensing base stations. For example, using a 4.9GHz integrated sensing module, a detection capability of an RCS of approximately 0.01m has been achieved. 2 Real-time positioning and trajectory perception of low-altitude unmanned aerial vehicles (UAVs). Meanwhile, some academic research has estimated the RCS fluctuation characteristics (such as Swerling type) of aerial targets in open spaces using the OFDM-ISAC system. However, the shortcomings of the above technologies are: they only use the RCS value as a threshold reference for target detection, without quantitative inversion and feature extraction; the verification environment is always open space, not involving semi-enclosed environments with dense multipath components such as ship locks; the sensing object is a small aerial target, not a large surface vessel; and they do not utilize multipath components to achieve multi-angle RCS sampling techniques.

[0006] Technique 4: Multipath Assisted Positioning Method Multipath-assisted positioning technology has been extensively studied in indoor positioning, such as indoor multipath-assisted positioning methods based on mirror source models (deriving the Cramer-Rao lower bound CRLB) and the MAMPI-UWB system using channel impulse response to extract multipath component amplitude and phase information for device-free positioning. In the radar field, there are also techniques that utilize multipath to detect non-line-of-sight concealed targets. These techniques share common drawbacks: their core task is spatial target localization, not the extraction of physical scattering characteristics (RCS parameters); they only utilize multipath delay information or its amplitude information to improve positioning accuracy, without inverting the received power into the target's physical RCS value through radar equations; and they do not involve the construction of ship RCS feature vectors and their application in ship type identification.

[0007] In summary, in the semi-enclosed environment of a ship lock, where multiple paths are dense, there is currently no literature addressing the question of how to separate the various multipath components from the 5G-A integrated sensing echo signal, accurately invert the equivalent radar cross section value of the ship at various observation angles, and construct a multidimensional feature vector that can characterize the ship's radio frequency scattering characteristics. Summary of the Invention

[0008] The technical problem to be solved by this invention is to provide a method for extracting the RCS features of ships in locks based on 5G-A integrated sensing and multipath inversion. In the semi-enclosed space of a lock with known geometric parameters, the method utilizes the integrated sensing and multipath capability of the 5G-A communication base station to extract the equivalent RCS value of the ship from the multipath echo and construct a multi-angle RCS feature vector through multipath separation assisted by mirror source and radar equation inversion.

[0009] To solve the above-mentioned technical problems, the present invention adopts the following technical solution: A method for extracting RCS features of ships within locks based on 5G-A integrated sensing multipath inversion includes the following steps: Step 1: Transmit a sensing reference signal to the lock via a 5G-A base station, and receive multipath echo signals generated by scattering from ships inside the lock chamber and by reflection from the lock chamber wall and the water surface inside the lock chamber. Step 2: Demodulate the multipath echo signal using orthogonal frequency division multiplexing (OFDM) technology, estimate the channel frequency response on each subcarrier, perform inverse Fourier transform on the channel frequency response to obtain the channel impulse response, and extract the arrival time measurement and received power of each propagation path from the power-delay spectrum of the channel impulse response. Step 3: Based on the actual location of the 5G-A base station, find the mirror source coordinates of the 5G-A base station with respect to each reflecting surface of the gate chamber in the pre-generated virtual anchor point location lookup table; all mirror source coordinates and the actual location of the 5G-A base station together constitute the virtual anchor point set; Step 4: Perform nearest neighbor association matching between the arrival time measurement value of each propagation path extracted in Step 2 and the set of virtual anchor points. Find the virtual anchor point that best matches each propagation path from the set of virtual anchor points, and then calculate the one-way propagation distance of each propagation path. Step 5: Based on the transmit power, antenna gain, and electromagnetic wave wavelength of the 5G-A base station, as well as the receive power and one-way propagation distance of each propagation path, the equivalent radar cross section value of the ship corresponding to each propagation path is inverted based on the monostatic radar equation; the equivalent radar cross section values ​​of the ship corresponding to all propagation paths are combined to form the ship RCS feature.

[0010] Compared with the prior art, the present invention, employing the above technical solution, has the following technical effects: 1. The radar cross section (RCS) feature vector of ships extracted from the 5G-A integrated sensing echo signal is a type of information that cannot be obtained by optical sensors (such as visible light cameras and lidar) and satellite positioning methods. It can form a complementary sensing system with optical point cloud, image and other data, and significantly improve the comprehensive perception capability of ships.

[0011] 2. This invention fully utilizes the multipath propagation characteristics of the semi-enclosed space of the lock chamber, which has a known geometry. It can obtain the RCS values ​​of ships from multiple observation angles with a single measurement, eliminating the need for mechanical scanning or multiple measurements, thus achieving multi-angle RCS sampling in a single measurement. Taking the No. 3 lock of the Changzhou Lock on the Xijiang River in Guangxi as an example, under typical operating conditions, the number of effective multipath paths that can be identified is 5-8, meaning that 5-8 RCS samples from different observation angles can be obtained in a single measurement, significantly improving measurement efficiency.

[0012] 3. This invention relies entirely on the integrated sensing capabilities of 5G-A communication base stations to extract the RCS of ships. It does not require the additional deployment of dedicated radar measurement equipment or the construction of dedicated sensing networks. The hardware reuse rate is high, and the marginal cost of engineering implementation is close to zero, which has good economic efficiency and scalability.

[0013] 4. Based on measured data verification, the method of this invention shows that for different ship types such as steel cargo ships, aluminum alloy passenger ships, and fiberglass yachts, the Mahalanobis distance between the RCS feature vectors extracted at the same lock location is greater than 2, exhibiting statistically significant differences. This excellent distinguishability can effectively support subsequent upper-level application tasks such as ship type identification and lock passage vessel status determination.

[0014] 5. The sensing method based on radio frequency electromagnetic waves is not affected by weather and environmental conditions such as light intensity, rain, fog, and smoke. It can achieve stable operation in all weather and all time, making up for the lack of applicability of optical sensors and satellite positioning in the complex environment of ship locks, and significantly improving the robustness and practicality of the system. Attached Figure Description

[0015] Figure 1 This is a flowchart of the method for extracting RCS features of ships in locks based on 5G-A integrated sensing multipath inversion according to the present invention; Figure 2 This is a schematic diagram of the channel impulse response and power-delay spectrum in Embodiment 1 of the present invention, wherein (a) is the channel impulse response and (b) is the power-delay spectrum; Figure 3 These are radar images of the RCS feature vectors of three different ship types in Embodiment 1 of the present invention; Figure 4 This is the Mahalanobis distance matrix heatmap in Embodiment 1 of the present invention; Figure 5This is a robustness diagram of RCS feature extraction of ships at different water levels in Embodiment 4 of the present invention. Detailed Implementation

[0016] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0017] This invention proposes a method for extracting the RCS (Radio Frequency Scattering) features of ships within locks based on 5G-A integrated sensing and sensing multipath inversion, applicable to locks with known lock chamber geometry. This method utilizes a mirror source model constructed from the known lock chamber geometry to accurately separate multipath echo signals. Then, based on radar equations, it inverts the equivalent RCS value of the ship from the received power of each path and the known propagation distance, ultimately constructing a multi-angle RCS feature vector. This method requires no dedicated radar equipment, fully utilizing the integrated sensing and sensing capabilities of 5G-A communication base stations to achieve online extraction of ship radio frequency scattering features, providing a novel sensing data source for ship type identification and state determination. Figure 1 As shown, it includes the following steps:

[0018] Step S1 (Sensing Signal Transmission and Multipath Echo Reception): A sensing reference signal (PRS) is transmitted via a 5G-A base station deployed on a tower near the lock, and multipath echo signals generated by scattering from ships inside the lock chamber and reflections from the lock chamber walls and water surface are received. The 5G-A base station operates at 3.5 GHz (n78 band) with a signal bandwidth of 100 MHz, using the Zadoff-Chu sequence as the sensing reference signal to ensure excellent autocorrelation characteristics. The base station transmit power is no less than 40 dBm, and the antenna gain is no less than 15 dBi.

[0019] Existing technologies (5G-A drone RCS sensing, multipath assisted positioning) all operate in open spaces using single antennas or simple antenna configurations, without being optimized for the specific geometric constraints of a ship lock environment. The base station deployment parameters in this step (installation height 35-45 meters, horizontal distance from the lock chamber 450-550 meters, downtilt angle 3°-8°) are specifically designed for the coverage requirements of the ship lock chamber, ensuring that the incident angle of the base station's electromagnetic waves on the sidewall plane of the ship is within the angle range dominated by specular reflection (approximately 30°-60°), so that the specular reflection component with a high signal-to-noise ratio dominates in the echo signal.

[0020] Based on the area of ​​the ship's side walls carrier wavelength calculate, The values ​​are the length and height of the ship's sidewalls, respectively. At the 3.5GHz frequency band, the peak RCS of the specular reflection is: , The typical RCS of diffraction from the hull edges is approximately 30–40 dBsm in the X-band (10 GHz) and even lower in the 3.5 GHz band. Therefore, within the incident angle range of 30°–60°, the power of the specular reflection component is at least 10 dB higher than that of the diffraction component.

[0021] Step S2 (Channel Frequency Response Acquisition and Channel Impulse Response Generation): The echo signal is acquired from the base station's receiving antenna port, demodulated using Orthogonal Frequency Division Multiplexing (OFDM), and the channel frequency response (CFR) is estimated on each subcarrier, i.e., the complex gain value on each subcarrier is recorded. The channel frequency response can be expressed as: (1) in, For the first The complex gain of the path (including propagation loss, reflectivity of the reflector and ship scattering effect). For the first Round-trip delay of the path, It is complex Gaussian noise. The number of valid paths. For the first The frequency of each subcarrier , This refers to the frequency of the first subcarrier obtained after demodulation using orthogonal frequency division multiplexing (OFDM). For subcarrier index, The subcarrier spacing is defined. An inverse Fourier transform (IFFT) is performed on the CFR to obtain the channel impulse response (CIR), which presents multiple peaks in the time delay domain, each peak corresponding to a propagation path. From the power-time delay spectrum of the CIR, the time of arrival (TOA) and received power of each resolvable path are extracted using a peak detection algorithm. CIR peak detection can be achieved using constant false alarm rate (CFAR) detection or compressed sensing super-resolution algorithms instead of conventional threshold detection.

[0022] The key to this step is that the independent extraction of each multipath component is achieved through high-resolution CIR analysis, which provides the basic measurement data for subsequent path matching and RCS inversion.

[0023] Step S3 (Construction of mirror source location lookup table): Establish the gate chamber coordinate system: with the intersection of the left wall of the gate chamber and the upstream gate head end face on the bottom plate plane as the origin O, the X-axis points downstream along the longitudinal direction of the gate chamber, and the Y-axis points to the right wall along the transverse direction of the gate chamber. (The right wall plane), Z-axis vertically upward ( The base plate plane, (Water surface plane).

[0024] Based on the known geometric parameters of the gate chamber (including the gate chamber length L, width B, and wall height) and current water level The mirror source theorem is used to pre-generate a virtual anchor point location lookup table. For the three main reflecting surfaces of the gate chamber—the left wall (plane)—... ), right wall (flat) ), water surface (plane) ), calculate the coordinates of the mirror source of the real base station with respect to each reflecting surface. Let the location of the base station be... The coordinates of the mirror source are as follows: The left wall is mirrored once. The right wall is mirrored once. The water surface reflects the image once. Secondary reflection mirror sources (such as combined reflections from the left wall and water surface, or the right wall and water surface) are obtained through continuous mirroring operations. The real base station location and the coordinates of all mirror sources together constitute a set of virtual anchor points. In actual lock deployments, based on the number of resolvable multipath components, the following six virtual anchor points are typically selected: the real base station, the left wall primary reflection mirror, the right wall primary reflection mirror, the water surface primary reflection mirror, the left wall + water surface secondary reflection mirror, and the right wall + water surface secondary reflection mirror. When channel conditions permit, other secondary combined mirror sources can be added, and the actual number of virtual anchor points used is generally 6 to 8. This selection strategy ensures sufficient RCS multi-angle sampling while avoiding matching ambiguity and computational redundancy introduced by too many weak signal paths.

[0025] The key difference in this step is that the geometric parameters of the lock chamber are completely known in engineering (provided by the lock engineering drawings, and the water level is measured in real time by water level sensors). Therefore, the positions of all mirror sources can be pre-analyzed, calculated, and stored as lookup tables before deployment, eliminating the need for online iterative estimation, which greatly reduces computational complexity and improves reliability.

[0026] Step S4 (Multipath Association Matching and Propagation Distance Calculation): The TOA measurement value of each path extracted in Step S2 is matched with the virtual anchor points in the virtual anchor point lookup table constructed in Step S3 using nearest neighbor association matching. The specific method is as follows:

[0027] For each measurement path, the theoretical round-trip time delay (i.e., theoretical TOA) from the ship to each virtual anchor point in the lookup table is calculated using the ship's coarse position estimate. The ship's coarse position estimate can be provided by LiDAR point cloud positioning, BeiDou / GPS satellite positioning, or radio frequency positioning methods based on multipath delay.

[0028] The virtual anchor point with the smallest absolute difference between the measured TOA and the theoretical TOA is used as the matching result for that path, thereby determining the reflection path type corresponding to each multipath component, i.e., the combination of reflecting surfaces encountered by the path during propagation (including direct path, single reflection from the left wall, single reflection from the right wall, single reflection from the water surface, double reflection from the left wall and water surface, double reflection from the right wall and water surface, etc.). After successful association, the one-way propagation distance corresponding to that path is calculated: (2) in, At the speed of light, For the first Round-trip delay of each propagation path.

[0029] When multipath radars encounter multipath environments such as urban buildings, the geometry of the reflecting surface usually requires assumptions and inferences, leading to ambiguity in data association and matching. In contrast, the geometry of the gate chamber environment in this invention is completely known, with only three types of reflecting surfaces (left wall, right wall, and water surface). The path types are limited and enumerable, thus significantly reducing the ambiguity of data association and resulting in a significantly higher matching accuracy than existing technologies.

[0030] Step S5 (Equivalent RCS Inversion Based on Radar Equations): Using the following known or obtained parameters: Base Station Transmit Power Antenna gain carrier wavelength ( GHz, The one-way distances of each path obtained in step S4 and the received power of each path obtained in step S2 The equivalent radar cross section of the ship for each path is retrieved based on the monostatic radar equations. The monostatic radar equations are: (3) By performing the inverse operation on this equation, we obtain the inversion formulas for the equivalent RCS of each path: (4)

[0031] This invention utilizes the integrated sensing capability of communication base stations to invert the RCS from communication echoes through the inverse operation of radar equations, thus achieving quantitative inversion of RCS values. At the same time, by taking advantage of the natural characteristics of multipath propagation in the gate chamber, RCS samples from multiple observation angles can be obtained in a single measurement.

[0032] The inversion obtained The RCS values ​​of each path form a feature vector: (5) This feature vector serves as a set of parameters characterizing the current radio frequency (RF) scattering characteristics of a ship. Since different paths correspond to different signal incident and scattering angles, this feature vector essentially provides RCS samples from multiple observation angles, constituting the RF fingerprint of the ship at its current lock chamber position and attitude. The output RCS feature vector is used for subsequent ship type identification, ship state determination, or fusion with optical sensor data.

[0033] Example 1 (Xijiang Changzhou Ship Lock No. 3 - Standard Steel Cargo Ship)

[0034] Deployment parameters: The implementation scenario is based on the No. 3 lock of the Changzhou Ship Lock on the Xijiang River. Lock chamber geometric parameters: , , Current water level The 5G-A base station is deployed on a tower at a horizontal distance D=500m from the center of the gate chamber, with an installation height H=40m and an antenna downtilt angle of [missing information]. Base station operating frequency The base station transmit power is GHz, with a signal bandwidth of 100MHz, a subcarrier spacing of 30kHz, and a total number of subcarriers K=3276. The Zadoff-Chu sequence is used as the PRS pilot sequence. dBm (approximately 20 W), antenna gain dBi (approximately 31.6 times).

[0035] Mirror source lookup table construction: Base station actual location coordinates are set to (with the left wall) (plan) and gate chamber floor ( The endpoint of the intersection line (in the plane) at the upstream end face is the origin of the coordinate system. Based on the known geometric parameters of the gate chamber, the mirror source is pre-calculated: the real base station. ; left wall mirrored once ; Mirror image of the right wall ; Water surface mirror image Left wall + water surface secondary mirror Right wall + water surface secondary mirror The real base stations and all mirror sources together constitute a set of virtual anchor points.

[0036] Signal Transmission and CIR Extraction: A standard steel cargo ship (approximately 80 m in length, 12 m in width, and 5 m in freeboard) enters the central position of the lock chamber. The base station continuously transmits PRS signals and receives echoes. The receiver collects CFR data on 3276 subcarriers and obtains the Channel Impulse Response (CIR) via IFFT. Under typical operating conditions, six effective paths can be identified in the power-delay spectrum of the CIR, such as... Figure 2 As shown in (a) and (b) above. The TOA and received power measurements for each path are shown in Table 1:

[0037] Table 1. TOA and Received Power Measurements for Each Path

[0038] RCS Inversion: Taking the above 6 paths as examples, calculate the equivalent RCS value for each. Taking path 1 (direct path) as an example: , , , , Substitute into equation (4) to calculate: , Convert to logarithmic units (dBsm): Similarly, the RCS values ​​for the other 5 paths can be calculated. Finally, a 6-dimensional RCS feature vector is obtained. .

[0039] RCS Discrimination Test: At the same lock chamber location, RCS feature vectors were extracted for three different ship types, such as... Figure 3 As shown, there are steel cargo ships (80 m long, 12 m wide), aluminum alloy passenger ships (50 m long, 8 m wide), and fiberglass yachts (25 m long, 5 m wide). At least 100 frames of data were collected for each ship type. The mean and covariance matrix of the RCS feature vectors for each ship type were calculated, and the Mahalanobis distance was calculated according to equation (6). Figure 4 As shown in Table 2:

[0040] Table 2 Test Results

[0041] The Mahalanobis distance between the three ship types is greater than 2, which verifies that the RCS feature vector extracted by this method has good distinguishability for different ship types.

[0042] Mahalanobis distance The calculation formula is: (6) in, These are the RCS feature vectors for different ship types A and B, respectively.

[0043] For the same steel cargo ship, RCS features were extracted using both a traditional dedicated RCS measurement system and the method of this invention. Results show that this method can obtain RCS samples from six angles in a single measurement, while the traditional dedicated system can only obtain one angle, improving sampling efficiency by six times. In terms of cost, this method fully reuses 5G-A communication base stations, resulting in marginal costs approaching zero, while the deployment cost of a traditional dedicated RCS measurement system is typically over 500,000 yuan.

[0044] Example 2 (Different base station deployment parameters - installation height 35m)

[0045] The base station installation height was adjusted to H=35 m, the horizontal distance from the center of the gate chamber was adjusted to D=450 m, and the downtilt angle was adjusted to... Other parameters and all steps are the same as in Example 1. Experimental results show that although the absolute value of the RCS of each path varies with the angle, the Mahalanobis distance between the three different ship types remains above 2, proving that this method has good robustness to changes in base station deployment parameters.

[0046] Example 3 (Dual Base Station Cooperative RCS Extraction Scheme)

[0047] One 5G-A base station is deployed upstream and downstream of the gate chamber (the upstream base station is 450 m from the center of the gate chamber, and the downstream base station is 550 m from the center of the gate chamber, both with an installation height of 40 m). The two base stations use a unified GPS time reference and independently execute steps S1 to S5 to obtain two sets of RCS feature vectors respectively. and The two sets of feature vectors are concatenated into a higher-dimensional (12-dimensional) joint feature vector. After adopting the dual-base station scheme, the Mahalanobis distance between the three ship types is further increased (the Mahalanobis distance between steel cargo ships and aluminum alloy passenger ships increases from 2.8 to approximately 3.5), and the distinguishability is significantly improved.

[0048] Example 4 (RCS extraction under different water level conditions)

[0049] Based on Example 1, the water level in the gate chamber was changed. The water level drops from 15 m to 12 m (dry season), while other parameters remain unchanged. In step S3, the coordinates of the mirror source regarding the water surface are recalculated, and the lookup table for secondary mirror sources involving water surface reflection is updated. For example... Figure 5 As shown, water level changes mainly affect the RCS value of the water surface reflection correlation path, with a smaller impact on the direct path and wall reflection path. However, the Mahalanobis distance between different ship types remains above 2, demonstrating that this method has good adaptability to water level changes.

[0050] Based on the same inventive concept, this application provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the aforementioned method for extracting RCS features of ships in locks based on 5G-A integrated sensing multipath inversion.

[0051] Based on the same inventive concept, embodiments of this application provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the aforementioned method for extracting ship RCS features within a lock based on 5G-A integrated sensing multipath inversion.

[0052] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0053] This invention is described with reference to flowchart illustrations of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each step in the flowchart, and combinations of steps in the flowchart, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing device, generate instructions for implementing the steps in the flowchart. Figure 1 A device for a function specified in one or more processes.

[0054] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 The function specified in one or more processes.

[0055] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 Steps of a specified function in one or more processes.

[0056] The above embodiments are merely illustrative of the technical concept of the present invention and should not be construed as limiting the scope of protection of the present invention. Any modifications made to the technical solutions based on the technical concept proposed in this invention shall fall within the scope of protection of this invention.

Claims

1. A method for extracting RCS features of ships within locks based on 5G-A integrated sensing multipath inversion, characterized in that, Includes the following steps: Step 1: Transmit a sensing reference signal to the lock via a 5G-A base station, and receive multipath echo signals generated by scattering from ships inside the lock chamber and by reflection from the lock chamber wall and the water surface inside the lock chamber. Step 2: Demodulate the multipath echo signal using orthogonal frequency division multiplexing (OFDM) technology, estimate the channel frequency response on each subcarrier, perform inverse Fourier transform on the channel frequency response to obtain the channel impulse response, and extract the arrival time measurement and received power of each propagation path from the power-delay spectrum of the channel impulse response. Step 3: Based on the actual location of the 5G-A base station, find the mirror source coordinates of the 5G-A base station with respect to each reflecting surface of the gate chamber in the pre-generated virtual anchor point location lookup table; all mirror source coordinates and the actual location of the 5G-A base station together constitute the virtual anchor point set; Step 4: Perform nearest neighbor association matching between the arrival time measurement value of each propagation path extracted in Step 2 and the set of virtual anchor points. Find the virtual anchor point that best matches each propagation path from the set of virtual anchor points, and then calculate the one-way propagation distance of each propagation path. Step 5: Based on the transmit power, antenna gain, and electromagnetic wave wavelength of the 5G-A base station, as well as the receive power and one-way propagation distance of each propagation path, the equivalent radar cross section value of the ship corresponding to each propagation path is inverted based on the monostatic radar equation; the equivalent radar cross section values ​​of the ship corresponding to all propagation paths are combined to form the ship RCS feature.

2. The method for extracting RCS features of ships within locks based on 5G-A integrated sensing multipath inversion according to claim 1, characterized in that, In step 1, the operating center frequency of the 5G-A base station is 3.5GHz, the signal bandwidth is 100MHz, the Zadoff-Chu sequence is used as the sensing reference signal, the transmit power of the 5G-A base station is not less than 40dBm, and the antenna gain is not less than 15dBi. The incident angle of the sensing reference signal on the ship's sidewall plane inside the lock chamber is in the angle range dominated by specular reflection, namely 30°~60°, so that the power of the specular reflection component in the multipath echo signal is at least 10dB higher than that of the diffraction component.

3. The method for extracting RCS features of ships within locks based on 5G-A integrated sensing multipath inversion according to claim 1, characterized in that, In step 2, the channel frequency response is expressed as: ， in, For the first The complex gain of the propagation path includes propagation loss, reflection coefficient of the reflecting surface, and ship scattering effect; To the number of effective transmission paths, The imaginary unit, For the first The frequency of each subcarrier , This refers to the frequency of the first subcarrier obtained after demodulation using orthogonal frequency division multiplexing (OFDM). For subcarrier index, For subcarrier spacing, , For the first Round-trip delay of each propagation path, For the first Complex gain values ​​on each subcarrier It is complex Gaussian noise; In the power-delay spectrum of the channel impulse response, the arrival time and received power of each propagation path are extracted using a peak detection algorithm.

4. The method for extracting RCS features of ships within locks based on 5G-A integrated sensing multipath inversion according to claim 1, characterized in that, In step 3, based on the actual location of the 5G-A base station, the coordinates of the mirror source of the 5G-A base station with respect to each reflecting surface of the gate chamber are looked up in a pre-generated virtual anchor point location lookup table, as follows: With the intersection of the left wall of the gate chamber and the upstream gate head end face on the bottom plate plane as the origin, the X-axis is the longitudinal direction of the gate chamber pointing downstream, the Y-axis is the transverse direction of the gate chamber pointing to the right wall of the gate chamber, and the Z-axis is the vertical upward direction. Establish the gate chamber coordinate system. Based on the actual position of the 5G-A base station in the gate chamber coordinate system, find the mirror source coordinates of the 5G-A base station with respect to the left and right walls of the gate chamber and the water surface inside the gate chamber in the pre-generated virtual anchor point position lookup table. In the lock chamber coordinate system, The right wall plane, The base plate plane, The water surface is horizontal. The width of the gate chamber. This is the current water level.

5. The method for extracting RCS features of ships within locks based on 5G-A integrated sensing multipath inversion according to claim 4, characterized in that, The process of generating the virtual anchor point lookup table is as follows: Based on the known geometric parameters of the lock chamber, including the lock chamber length ,width Wall height and current water level The true location of the 5G-A base station in the gate chamber coordinate system was determined as follows. , Let X, Y, and Z be the values ​​of the 5G-A base station on the X, Y, and Z axes, respectively. Then, the coordinates of the first reflection mirror source of the 5G-A base station about the left wall of the gate chamber are: The coordinates of the primary reflection mirror source of the 5G-A base station about the right wall of the gate chamber are: The coordinates of the first reflection image source of the 5G-A base station with respect to the water surface inside the sluice gate are: The coordinates of the secondary reflection source of the 5G-A base station, which first passes through the left wall and then the water surface, are: The coordinates of the secondary reflection source of the 5G-A base station, which first passes through the right wall and then the water surface, are: ; The coordinates of the above 5 mirror sources and the actual location of the 5G-A base station are selected to form a set of virtual anchor points.

6. The method for extracting RCS features of ships within locks based on 5G-A integrated sensing multipath inversion according to claim 1, characterized in that, The specific process of step 4 is as follows: For the Using a coarse estimation of the ship's position, the theoretical round-trip time (i.e., the theoretical arrival time) from the ship to each virtual anchor point in the set of virtual anchor points is calculated. The virtual anchor point with the smallest absolute value of the difference between the theoretical arrival time and the measured arrival time is selected as the first virtual anchor point. The matching results of the propagation path are used to determine the first propagation path. The reflection path type corresponding to the propagation path, i.e., the first... The reflective surfaces encountered by each propagation path during the propagation process; After a successful match, calculate the first... One-way propagation distance of each propagation path : ， in, At the speed of light, , For the first Round-trip delay of each propagation path.

7. The method for extracting RCS features of ships within locks based on 5G-A integrated sensing multipath inversion according to claim 1, characterized in that, In step 5, the equivalent radar cross section value of the ship corresponding to each propagation path is expressed as follows: ， in, For the first The equivalent radar cross section of the ship corresponding to each propagation path, For the first The received power of each propagation path, For the first The one-way propagation distance of each propagation path, This refers to the transmit power of a 5G-A base station. For antenna gain, For carrier wavelength, , At the speed of light, The operating frequency for 5G-A base stations; The equivalent radar cross section values ​​of the ship corresponding to all propagation paths are used to compose the ship's RCS feature: ， in, Represents the RCS feature vector of a ship. The first The equivalent radar cross section of the ship corresponding to each propagation path, This indicates transpose.

8. A computer device comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method for extracting ship RCS features in locks based on 5G-A integrated sensing multipath inversion as described in any one of claims 1 to 7.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the method for extracting RCS features of ships in locks based on 5G-A integrated sensing multipath inversion as described in any one of claims 1 to 7.

Images