A ship positioning method and storage medium based on Fresnel zone

Abstract

The present invention relates to the field of positioning technology, and in particular to a ship positioning method and storage medium based on Fresnel zones. The present invention arranges wireless receiving devices and wireless transmitting devices in an interlaced radio frequency distribution mode, constructs N layers of Fresnel zones for each wireless receiving device, analyzes radio frequency signals received by the wireless receiving devices to determine radio frequency signal features, and inputs the radio frequency signal features into a pre-trained GRU gated recurrent neural network to determine the first position information of the ship, thereby avoiding the cumulative error problem of a SINS system and helping to improve the accuracy and reliability of ship positioning.

Description

Technical Field

[0001] The present invention relates to the field of positioning technology, and in particular to a ship positioning method and storage medium based on Fresnel zone. Background Art

[0002] In modern shipping, accurate positioning is essential to ensure safe navigation of ships. The Global Positioning System (GNSS) has become the standard technology for ship navigation and positioning. However, GNSS radio frequency signals may be interfered with or completely lost in certain specific environments, such as river canyons, densely populated urban areas, tunnels, or in severe weather conditions. In these cases, the ship's navigation system may cause navigation deviations due to the lack of accurate position information, or even dangerous situations such as collisions or groundings.

[0003] In order to solve the positioning problem when GNSS radio frequency signals are missing or limited, the inertial navigation system (SINS) has become an important auxiliary means. SINS uses accelerometers and gyroscopes to measure the motion state of the ship and obtains position information by integration. However, due to the cumulative error problem of the SINS system, its long-term use will lead to a significant decrease in positioning accuracy. Summary of the invention

[0004] In order to solve the technical problems existing in the above-mentioned background technology, the present invention provides a positioning method and a storage medium based on Fresnel zones, which are helpful to improve the accuracy and reliability of ship positioning.

[0005] In order to implement the above technical solution, in a first aspect, the present invention provides a ship positioning method based on Fresnel zone, comprising the following steps:

[0006] Step 1: Arrange wireless receiving equipment and wireless transmitting equipment on both sides of the river using a staggered RF point layout method;

[0007] Step 2: Construct N layers of Fresnel zones for each wireless receiving device;

[0008] Step 3: monitor each wireless receiving device in real time to analyze the radio frequency signal received by each wireless receiving device to determine the radio frequency signal characteristics of each Fresnel zone;

[0009] Step 4: Input each Fresnel zone RF signal feature into the pre-trained GRU gated recurrent neural network to determine the first position information of the ship.

[0010] Furthermore, the method further comprises the following steps:

[0011] The determined first position information and the second position information determined by the SINS system are input into the EKF model for fusion to obtain the final position information.

[0012] Furthermore, before step 4, the method further includes: pre-training the GRU gated recurrent neural network.

[0013] Furthermore, the step three comprises:

[0014] The characteristics of the radio frequency signal in the Fresnel zone are determined based on the sinusoidal wave characteristics of the radio frequency signal when the ship crosses each layer of the Fresnel zone.

[0015] In a second aspect, the present invention provides a ship positioning system based on Fresnel zone, comprising an inertial navigation system installed on a ship, a wireless transmitting device, a wireless receiving device and a central server;

[0016] An inertial navigation system for determining a second position information of the vessel;

[0017] Wireless transmitting equipment, used to transmit radio frequency signals;

[0018] The wireless receiving device is connected to the wireless transmitting device for receiving the radio frequency signal transmitted by the wireless transmitting device;

[0019] The central server is connected to the wireless receiving device and is used to construct N layers of Fresnel zones of the wireless receiving device and obtain radio frequency signals received by the wireless receiving device, so as to determine the first position information of the ship based on the constructed N layers of Fresnel zones and the obtained radio frequency signals.

[0020] Furthermore, the system further comprises:

[0021] The Kalman filter model is extended to fuse the first position information and the second position information to obtain the final position information.

[0022] Furthermore, the system further comprises:

[0023] Training module, used to pre-train the GRU gated recurrent neural network.

[0024] Furthermore, the central server comprises:

[0025] A construction module for constructing N layers of Fresnel zones for each wireless receiving device;

[0026] A monitoring device, used for monitoring each wireless receiving device to obtain a radio frequency signal received by each wireless receiving device;

[0027] The signal analysis module is used to analyze the radio frequency signal received by the wireless receiving device, and determine the radio frequency signal characteristics of each Fresnel zone based on the N layers of Fresnel zones constructed for each wireless receiving device and the sinusoidal wave characteristics of the radio frequency signal when the ship crosses each layer of the Fresnel zone.

[0028] In a third aspect, the present invention provides a computer-readable storage medium, the computer-readable storage medium including a stored program, wherein when the program is running, the device where the computer-readable storage medium is located is controlled to execute the above-mentioned ship positioning method based on Fresnel zone.

[0029] The beneficial effects of the present invention are:

[0030] (1) The present invention arranges wireless receiving devices and wireless transmitting devices in an interlaced RF distribution method, constructs N-layer Fresnel zones for each wireless receiving device, analyzes the RF signal received by the wireless receiving device to obtain RF signal characteristics, and inputs the RF signal characteristics into a pre-trained GRU gated recurrent neural network to determine the first position information of the ship, which helps to avoid the cumulative error problem of the SINS system, thereby improving the accuracy and reliability of ship positioning.

[0031] (2) The first position information determined by the pre-trained GRU gated recurrent neural network and the second position information determined by SINS are fused through the EKF model, which helps to reduce the impact of noise and thus helps to provide more accurate and stable position information. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0033] Figure 1 The figure is a flow chart of a ship positioning method based on Fresnel zone of the present invention.

[0034] Figure 2 It is a layout diagram of the wireless receiving device and the wireless transmitting device of the present invention.

[0035] Figure 3 This is a schematic diagram of the principle of analyzing radio frequency signals according to the present invention. DETAILED DESCRIPTION

[0036] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0037] It should be noted that the following detailed descriptions are all illustrative and are intended to provide further explanation of the present invention. Unless otherwise specified, each technical and scientific term used in this embodiment has the same meaning as commonly understood by those skilled in the art to which the present invention belongs.

[0038] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprising" and / or "including" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0039] In the present invention, terms such as "upper", "lower", "left", "right", "front", "back", "vertical", "horizontal", "side", "bottom", etc. indicate directions or positional relationships based on the directions or positional relationships shown in the accompanying drawings. They are relational words determined only for the convenience of describing the structural relationships of the various parts or elements of the present invention, and do not specifically refer to any part or element in the present invention and should not be understood as limitations on the present invention.

[0040] In the present invention, terms such as "fixed connection", "connected", "connection", etc. should be understood in a broad sense, indicating that it can be fixedly connected, integrally connected or detachably connected; it can be directly connected or indirectly connected through an intermediate medium. Relevant scientific research or technical personnel in this field can determine the specific meanings of the above terms in the present invention according to specific circumstances, and they should not be understood as limiting the present invention.

[0041] GRU (Gated Recurrent Unit) is a gated recurrent neural network, a variant of LSTM (Long Short-Term Memory), which is designed to solve the gradient vanishing and gradient exploding problems that may occur in traditional RNN (Recurrent Neural Network) when processing long sequence data.

[0042] EKF (Extended Kalman Filter) Extended Kalman Filter.

[0043] Embodiment 1:

[0044] like Figure 1 As shown, this embodiment provides a ship positioning method based on Fresnel zone, comprising the following steps:

[0045] S1: Wireless receiving equipment and wireless transmitting equipment are arranged in an interlaced RF distribution method on both sides of the river.

[0046] The number of wireless transmitting devices and wireless receiving devices can be set according to actual needs.

[0047] S2: Construct N layers of Fresnel zones for each wireless receiving device.

[0048] The N-layer Fresnel zone is constructed with the wireless receiving device and the wireless transmitting device that sends the radio frequency signal to the wireless receiving device as the ellipse focus. There may be multiple N-layer Fresnel zones, and specifically, the number of Fresnel zones is determined by the number of wireless transmitting devices that send the radio frequency signal to the Fresnel zone.

[0049] For example, Figure 2 As shown, there are wireless transmitting modules A, B, C, D, and wireless receiving devices E, F, G. For the wireless receiving device F, four N-layer Fresnel zones are constructed.

[0050] Each Fresnel-region represents the RF signal propagation path between two wireless communication modules. Solid lines and dashed lines represent different RF signal propagation paths, where the solid line represents the main propagation path and the dashed line represents the secondary path.

[0051] S3: Monitor each wireless receiving device in real time to analyze the radio frequency signal received by each wireless receiving device to determine the radio frequency signal characteristics of each Fresnel zone.

[0052] Specifically, based on the sinusoidal wave characteristics of the radio frequency signal when the ship crosses each layer of the Fresnel zone, the radio frequency signal characteristics of each Fresnel zone are determined.

[0053] More specifically, if Figure 3 As shown, assuming that P1 and P2 are a pair of wireless receiving devices and wireless transmitting modules, the N-layer Fresnel region refers to a circle of concentric ellipses corresponding to the wireless receiving device and the wireless transmitting module as the focus. For a radio frequency signal with a given wavelength γ, the Fresnel region containing N layers of concentric ellipses can be expressed by the following formula, where Qn is a point on the trajectory of the Nth layer of ellipse, and:

[0054]

[0055] The innermost ellipse is the first Fresnel-region, and the length of its reflection path P1Q1P2 is half a wavelength (γ / 2) longer than the straight propagation path P1P2, and N is 1. The elliptical ring between the first and second layers of ellipses is defined as the second Fresnel-region, and so on to define the nth Fresnel-region, and each layer of ellipse itself is defined as the boundary of the corresponding Fresnel-region.

[0056] Assume that the ship appears at position Q1 on the boundary of the first Fresnel-region (as shown in the figure above). Since the path length of the reflected RF signal P1Q1P2 is longer than the RF signal of the straight-line propagation path P1P2 (γ / 2), the phase difference between them when they arrive at the receiving end P2 is π; during the reflection process, the RF signal will also shift in phase due to the properties of the medium (such as dielectric constant, conductivity, etc.), and the phase difference is π. Since the phase difference is periodic (every 2π is a cycle), the phase of the reflected RF signal is exactly the same as the phase of the direct RF signal, so they will enhance each other, resulting in a greater strength of the received RF signal, resulting in the reception of an enhanced RF signal.

[0057] Similarly, if the ship is on the boundary of the second Fresnel-region, the phase difference of the reflected RF signal with a path difference of γ relative to the direct RF signal is 2π, and the reflection itself usually introduces an additional phase shift π, so the total phase difference of the reflected RF signal relative to the direct RF signal is 3π. However, since the phase difference is periodic (every 2π is a period), 3π can be regarded as a phase π (3π - 2π = π). At this time, the phase of the reflected RF signal is exactly opposite to the phase of the direct RF signal (that is, the phase difference is π), so they will cancel each other out, resulting in a destructive RF signal when the received RF signal strength is reduced. Therefore, odd-numbered signals are strong and even-numbered signals are cancelled, and the concentric elliptical rings of the Fresnel-region alternate between strong and weak. (Objects on the boundary of odd / even Fresnel-regions will have an enhancement / destruction effect on the strength of the received RF signal, which appears as an alternating enhancement / destruction pattern on the concentric elliptical rings of the Fresnel-region)

[0058] According to the principle of interference, when the ship crosses each Fresnel-region boundary, the received RF signal appears as a peak or a trough. When the object moves along the ellipse, the received RF signal will also remain stable because the length of the reflected RF signal path remains unchanged.

[0059] Furthermore, when the ship crosses multiple Fresnel-region boundaries continuously, the received radio frequency signal is similar to a sine waveform, and the peaks and troughs of this sine-like waveform correspond exactly to the boundaries of the odd / even Fresnel-regions respectively.

[0060] S4: Input the RF signal features of each Fresnel zone into the pre-trained GRU gated recurrent neural network to determine the first position information of the vessel.

[0061] S5: Input the determined first position information and the second position information determined by the SINS system into the EKF model, perform fusion, and obtain final position information.

[0062] The specific steps include:

[0063] By means of Kalman filtering, a balance is found between the first position information obtained by the pre-trained GRU gated recurrent neural network and the second position information obtained by the SINS system on the vessel. The balance point is determined based on the accuracy and noise level of the SINS system and the Fresnel zone.

[0064] This embodiment fuses the first position information determined by the pre-trained GRU gated recurrent neural network and the second position information determined by the SINS system through the EKF model, which helps to reduce the impact of noise and thus helps to provide more accurate and stable second position information.

[0065] It should be noted that the SINS system is a system used for positioning on a ship, which is well known to those skilled in the art, and therefore its functions and uses will not be described in detail.

[0066] Furthermore, the method further comprises: pre-training the GRU gated recurrent neural network. Specifically, the following steps are included:

[0067] Q1: In a specified area, collect the RF signal characteristics at each layer of each Fresnel zone in the specified area.

[0068] Q2: Collect location information of each location in a specified area; the location information is usually represented by two-dimensional coordinates such as (x, y) or three-dimensional coordinates such as (x, y, z).

[0069] Q3: Use the position information to mark the RF signal features at each layer of each Fresnel zone, so that the RF signal features at each layer of each Fresnel zone are marked with a position information to obtain a data set of RF signal features and the corresponding position information.

[0070] Q4: Construct a data set based on the obtained location information and the RF signal features that annotate the location information.

[0071] Specifically, each piece of data in the data set can be expressed as: Input: [radio frequency signal feature] Output: [location information corresponding to the radio frequency signal feature].

[0072] It should be noted that the radio frequency signal feature is a radio frequency signal in the form of a time series.

[0073] Q5: Train the GRU gated recurrent neural network based on the constructed dataset.

[0074] Q4.1: Data preprocessing

[0075] (A) Normalization / Standardization: Normalize or standardize the data in the dataset to ensure that the value of each feature is on the same scale to avoid the influence of larger feature values ​​on the training process.

[0076] (B) Serialized data: Slice each piece of data in the data set to generate a fixed-length sequence. For example, assuming that the length of each input sequence is T, each input data sample can contain T time steps, and each time step contains RF signal features. The input data is:

[0077] Inputshape=(batchsize,sequencelength,featuredimension)

[0078] Q4.2: Input the preprocessed data into the GRU gated recurrent neural network in batches.

[0079] Q4.3: The GRU layer of the GRU gated recurrent neural network processes the data of each time step step by step, maintains an internal hidden state, and selects the data of the last time step.

[0080] Specifically, each time the data of the current time step and the hidden state of the previous time step are input, GRU decides which information to keep and which to discard through its internal structure (update gate, reset gate, etc.) to determine the final output state.

[0081] Among them, there is also a set of hidden units in the GRU layer to capture the temporal correlation of the sequence through time steps.

[0082] It should be noted that the output of the GRU layer can be the state of each time step, and the final output state can also be determined by comparison.

[0083] Q4.4: Map the data of the last selected time step to the target output space to obtain the position prediction value.

[0084] Q4.5: Output the predicted position value.

[0085] Q4.6: Use the mean square error (MSE) to measure the loss error between the output position prediction value and the marked position information. The specific calculation formula is as follows:

[0086]

[0087] Where K represents the number of location information; and is the location information predicted by the model; and It is the corresponding marked location information.

[0088] Q4.7: Optimize the GRU gated recurrent neural network parameters based on the calculated loss error.

[0089] Specifically, based on the calculated loss error, the back-propagation algorithm is used to calculate the gradient of the loss relative to the parameters of the GRU gated recurrent neural network. This step will propagate the gradient of the loss from the fully connected layer to the GRU layer through the chain rule, and then to the parameters of each layer (including the weights and biases of the GRU units, and the weights and biases of the fully connected layers).

[0090] Optimizer update parameters: Use an optimizer (such as Adam or SGD) to update the parameters of the GRU gated recurrent neural network according to the gradient. The optimizer adjusts the value of each parameter so that the loss function gradually decreases.

[0091] Embodiment 2:

[0092] This embodiment provides a ship positioning system based on Fresnel zone, including: a wireless receiving device, a wireless transmitting device and a central server;

[0093] SINS system is used to determine the second position information of the vessel.

[0094] Wireless transmitting equipment, used to transmit radio frequency signals.

[0095] The wireless receiving device is connected to the wireless transmitting device for receiving the radio frequency signal transmitted by the wireless transmitting device.

[0096] The central server is connected to the wireless receiving device and is used to construct N layers of Fresnel zones of the wireless receiving device and obtain radio frequency signals received by the wireless receiving device, so as to determine the first position information of the ship based on the constructed N layers of Fresnel zones and the obtained radio frequency signals.

[0097] The extended Kalman filter model is used to fuse the first position information of the ship determined by the central server and the second position information of the ship obtained by SINS to obtain the final position information.

[0098] A training device is used to train the extended Kalman filter model.

[0099] The central server includes:

[0100] A building block is used to build N-layer Fresnel zones for each wireless receiving device.

[0101] The monitoring device is used to monitor each wireless receiving device to obtain the radio frequency signal received by each wireless receiving device.

[0102] The signal analysis module is used to analyze the radio frequency signal received by the wireless receiving device, and determine the radio frequency signal characteristics of each Fresnel zone based on the N layers of Fresnel zones constructed for each wireless receiving device and the sinusoidal wave characteristics of the radio frequency signal when the ship crosses each layer of the Fresnel zone.

[0103] A GRU gated recurrent neural network is used to determine the first position information of the vessel based on the determined radio frequency signal characteristics of each Fresnel zone.

[0104] It should be noted that the extended Kalman filter model structure includes:

[0105] (A) Input layer, used to input data dimensions.

[0106] The preprocessed data is a time series, and each time step contains the characteristics of the RF signal. Assume that the input data format is: Inputshape=(batchsize, sequencelength, featuredimension)

[0107] For example, batch size = 32, sequence length = 10, feature dimension = 5 (representing 6 RF signal features, such as A\B\C\E\F, etc.).

[0108] These data are input into the model in batches.

[0109] (B) GRU layer

[0110] The preprocessed time series data will be input to the GRU layer. GRU will gradually process the data of each time step and maintain an internal hidden state. Each time the data of the current time step and the hidden state of the previous time step are input, GRU uses its internal structure (update gate, reset gate, etc.) to decide which information to keep and which to discard. After GRU processing, the output of each time step and the final hidden state are obtained.

[0111] It should be noted that there is a set of hidden units in the GRU layer, such as 128 hidden units. These hidden units capture the temporal correlation of the sequence through time steps.

[0112] GRU layer output: The output of the GRU layer can be the state of each time step or the state of the last time step, depending on the task requirements. For position reasoning tasks, we usually only need the output of the last time step as the input of the subsequent layer.

[0113] GRUOutputShape=(batchsize,hiddenunits)

[0114] Assuming the number of hidden units in the GRU layer is 128, the output shape is (32, 128) (corresponding to 32 batches, each with 128 hidden unit outputs).

[0115] (C) Fully connected layer (FC layer)

[0116] Function: The fully connected layer receives the output of the last time step of the GRU layer and maps it to the target output space. Since we need to predict the position (usually 2D or 3D coordinates), the output size of the fully connected layer should be consistent with the target dimension.

[0117] Input to FC layer: The shape of the output of the GRU layer is (32,128), which represents the hidden state of 32 samples. These hidden states are input to the fully connected layer.

[0118] Output layer dimension: Assuming we need to predict 2D position coordinates (x, y), the output dimension of the fully connected layer is 2.

[0119] FCOutputShape=(batchsize,2)

[0120] After the fully connected layer, the hidden state of the GRU layer is mapped to two-dimensional coordinates.

[0121] (D) Activation function

[0122] Linear activation: Because position prediction is a regression problem, the final output layer usually does not use a nonlinear activation function, but directly adopts linear output, so that the output value can be an arbitrary real number, which is suitable for representing coordinates.

[0123] Embodiment 3:

[0124] This embodiment provides a computer-readable storage medium, which includes a stored program. When the program is running, the device where the computer-readable storage medium is located is controlled to execute the ship positioning method based on Fresnel zone described in Embodiment 1.

[0125] In this specification, the same or similar parts between the various embodiments can be referred to each other. In particular, for the terminal embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the description in the method embodiment.

[0126] In the several embodiments provided by the present invention, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of systems or units, which can be electrical, mechanical or other forms.

[0127] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0128] In addition, it should be noted that the flowchart in the accompanying drawings shows the method of the embodiment of the present disclosure. In the corresponding description in the flowchart or block diagram in the accompanying drawings, the operations or steps corresponding to different blocks may also occur in an order different from that disclosed in the description, and sometimes there is no specific order between different operations or steps. For example, two consecutive operations or steps can actually be performed substantially in parallel, and they can sometimes be performed in the opposite order, which can depend on the functions involved. Each block in the block diagram and / or flow chart, and the combination of blocks in the block diagram and / or flow chart, can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0129] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A ship positioning method based on Fresnel zone, characterized in that: The following steps are involved: Step 1: Arrange wireless receiving equipment and wireless transmitting equipment on both sides of the river using a staggered RF point layout method; Step 2: Construct N layers of Fresnel zones for each wireless receiving device; Step 3: monitor each wireless receiving device in real time to analyze the radio frequency signal received by each wireless receiving device to determine the radio frequency signal characteristics of each Fresnel zone; Step 4: Input each Fresnel zone RF signal feature into the pre-trained GRU gated recurrent neural network to determine the first position information of the ship; Before step 4, the method further includes: pre-training the GRU gated recurrent neural network; The pre-training of the GRU gated recurrent neural network includes: In the specified area, collecting radio frequency signal characteristics at each layer of each Fresnel zone of the radio frequency signal in the specified area; Collecting location information of each location in a specified area; wherein the location information is usually represented by two-dimensional coordinates or three-dimensional coordinates; Using the position information, the radio frequency signal features at each layer of each Fresnel zone are annotated, so that the radio frequency signal features at each layer of each Fresnel zone are annotated with a position information to obtain a data set of the radio frequency signal features and the position information corresponding thereto; Building a data set based on the obtained location information and the radio frequency signal features annotated with the location information; Normalize or standardize and serialize the data in the dataset; Input the preprocessed data into the GRU gated recurrent neural network in batches; The GRU layer of the GRU gated recurrent neural network processes the data of each time step step by step, maintains an internal hidden state, and selects the data of the last time step; Map the data of the last selected time step to the target output space to obtain the position prediction value; Output position prediction value; The mean square error is used to measure the loss error between the output position prediction value and the marked position information. The specific calculation formula is as follows: Where K represents the number of location information; and is the location information predicted by the model; and is the corresponding marked location information; According to the calculated loss error, the GRU gated recurrent neural network parameters are optimized.

2. The ship positioning method based on Fresnel zone according to claim 1, characterized in that: The method further comprises the following steps: The determined first position information and the second position information determined by the SINS system are input into the EKF model for fusion to obtain the final position information.

3. The ship positioning method based on Fresnel zone according to claim 1, characterized in that: The step three comprises: The characteristics of the radio frequency signal in the Fresnel zone are determined based on the sinusoidal wave characteristics of the radio frequency signal when the ship crosses each layer of the Fresnel zone.

4. A ship positioning system based on Fresnel zone, comprising an inertial navigation system installed on a ship, characterized in that: It also includes wireless transmitting equipment, wireless receiving equipment and a central server; An inertial navigation system for determining a second position information of the vessel; Wireless transmitting equipment, used to transmit radio frequency signals; The wireless receiving device is communicatively connected with the wireless transmitting device and is used to receive the radio frequency signal transmitted by the wireless transmitting device, wherein the wireless receiving device and the wireless transmitting device are arranged on both sides of the river channel in a staggered radio frequency distribution mode; The central server is connected to the wireless receiving device, and is used to construct N layers of Fresnel zones of each wireless receiving device and obtain the radio frequency signal received by each wireless receiving device, so as to determine the first position information of the vessel based on the constructed N layers of Fresnel zones and the obtained radio frequency signal; The system further comprises: Training module, used to pre-train the GRU gated recurrent neural network; The pre-training of the GRU gated recurrent neural network includes: In the specified area, collecting radio frequency signal characteristics at each layer of each Fresnel zone of the radio frequency signal in the specified area; Collecting location information of each location in a specified area; wherein the location information is usually represented by two-dimensional coordinates or three-dimensional coordinates; Using the position information, the radio frequency signal features at each layer of each Fresnel zone are annotated, so that the radio frequency signal features at each layer of each Fresnel zone are annotated with a position information to obtain a data set of the radio frequency signal features and the position information corresponding thereto; Building a data set based on the obtained location information and the radio frequency signal features annotated with the location information; Normalize or standardize and serialize the data in the dataset; Input the preprocessed data into the GRU gated recurrent neural network in batches; The GRU layer of the GRU gated recurrent neural network processes the data of each time step step by step, maintains an internal hidden state, and selects the data of the last time step; Map the data of the last selected time step to the target output space to obtain the position prediction value; Output position prediction value; The mean square error is used to measure the loss error between the output position prediction value and the marked position information. The specific calculation formula is as follows: Where K represents the number of location information; and is the location information predicted by the model; and is the corresponding marked location information; According to the calculated loss error, the GRU gated recurrent neural network parameters are optimized.

5. The ship positioning system based on Fresnel zone according to claim 4, characterized in that: The system further comprises: The Kalman filter model is extended to fuse the first position information and the second position information to obtain the final position information.

6. The ship positioning system based on Fresnel zone according to claim 4, characterized in that: The central server comprises: A construction module for constructing N layers of Fresnel zones for each wireless receiving device; A monitoring device, used for monitoring each wireless receiving device to obtain a radio frequency signal received by each wireless receiving device; The signal analysis module is used to analyze the radio frequency signal received by the wireless receiving device, and determine the radio frequency signal characteristics of each Fresnel zone based on the N layers of Fresnel zones constructed for each wireless receiving device and the sinusoidal wave characteristics of the radio frequency signal when the ship crosses each layer of the Fresnel zone.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein when the program is running, the device where the computer-readable storage medium is located is controlled to execute the ship positioning method based on Fresnel zone according to any one of claims 1 to 3.

Images