Wireless positioning method based on hybrid communication mode

Abstract

The invention relates to the technical field of wireless positioning, and discloses a wireless positioning method based on a hybrid communication mode, which comprises the following steps of: in an environment modeling stage, enabling a plurality of anchor nodes to perform full-grid dual-mode signal interaction, and obtaining a flight time difference and a path quality score based on channel characteristic evaluation; the method comprises the following steps: generating a three-dimensional sound slowness field representing an environment by using a weighted tomography reconstruction algorithm, in a target positioning stage, communicating a moving target with anchor nodes to obtain real-time measurement data, generating a shortest flight time field model for each anchor node based on the sound slowness field, constructing a nonlinear optimization target function taking a path mass fraction as a weight, and calculating a target position. The method further comprises a dynamic updating mechanism of a sound slowness field, chromatography modeling is carried out on a space acoustic environment in advance, path mass weighting is introduced, limitation of a traditional sound velocity constant assumption is overcome, influences of environment non-uniformity and time-varying characteristics caused by factors such as temperature and airflow can be compensated, and positioning accuracy and environment adaptive capacity are improved.

Description

Technical Field

[0001] This invention relates to the field of wireless positioning technology, specifically a wireless positioning method based on a hybrid communication mode. Background Technology

[0002] In numerous application scenarios such as warehousing and logistics, industrial automation, virtual reality, and personnel monitoring, rapid and accurate location tracking of moving targets has become a critical requirement. Currently, mainstream indoor positioning technologies, such as Wi-Fi fingerprint positioning or Bluetooth angle-of-arrival positioning based on radio frequency signals, are convenient to deploy, but their positioning accuracy is generally limited by factors such as multipath effects, signal attenuation, and non-line-of-sight propagation, making it difficult to reliably meet the high accuracy requirements of sub-meter or even centimeter levels.

[0003] In pursuit of higher accuracy, acoustic signal-based (especially ultrasonic) positioning technology has attracted considerable attention due to its physical characteristics of slow propagation speed and ease of accurate time-of-flight measurement. Traditional acoustic positioning methods typically rely on a core prerequisite: that sound waves propagate at a constant and uniform speed in the air. The system calculates the target coordinates by measuring the time of flight of the signal between the target and multiple known anchor points, and then using multi-point positioning algorithms (such as polygonal measurement).

[0004] However, in real-world physical environments, the ideal assumption of a constant speed of sound often deviates significantly from reality. The propagation speed of sound waves is extremely sensitive to environmental parameters such as temperature, humidity, air pressure, and even airflow. In a large warehouse or production workshop, the combined effects of lighting, ventilation systems, equipment heat dissipation, and personnel activity create a complex and non-uniform temperature and airflow field. This environmental non-uniformity causes the propagation path of sound waves to no longer be a straight line, and their propagation speed varies throughout the space. If the positioning algorithm stubbornly uses a globally uniform sound speed value for calculation, it will introduce systematic errors that contradict reality. These errors worsen with increasing positioning distance and environmental complexity, thus preventing the inherent high-precision potential of acoustic positioning technology from being effectively realized.

[0005] Furthermore, environmental parameters are not static but dynamically changing over time. For example, diurnal temperature variations, seasonal changes, or changes in workshop production conditions can all cause overall drift in the sound velocity field. Existing technologies generally lack the ability to adapt to such spatiotemporal environmental changes, and a one-time calibration cannot guarantee the accuracy and stability of the system during long-term operation, resulting in a significant reduction in the reliability and consistency of positioning results. Therefore, overcoming the changes in the sound velocity field caused by environmental non-uniformity and dynamism is a key bottleneck that must be addressed for high-precision acoustic positioning technology to achieve practical application. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides a wireless positioning method based on a hybrid communication mode. This method solves the problem that existing wireless positioning methods based on hybrid communication modes suffer from significant errors in measuring time difference of flight when facing complex indoor environments. These errors are caused by factors such as temperature, humidity, and airflow leading to uneven distribution of sound speed in space, as well as non-line-of-sight propagation due to obstacles. This results in reduced positioning accuracy.

[0007] To address the aforementioned technical problems, this invention provides a wireless positioning method based on a hybrid communication mode, comprising the following steps: S1. Acquire the radio frequency signal and ultrasonic signal synchronously transmitted by the anchor node or moving target as the transmitter; S2. Calculate the time difference between the radio frequency signal and the ultrasonic signal; S3. Extract channel response features from the radio frequency signal and the ultrasonic signal; S4. Based on the channel response characteristics, generate a path quality score that characterizes the quality of the current signal propagation path; S5. Apply steps S1 to S4 to signal propagation between the multiple fixed anchor nodes to generate a first set of flight time differences and a first set of path quality scores. S6. Based on the first set of flight time differences and the first set of path quality scores, the acoustic slowness of the preset grid cells in the positioning space is calculated by the tomographic reconstruction algorithm, and an acoustic slowness field is generated. S7. Apply steps S1 to S4 to signal propagation between the moving target and the anchor node to generate a second set of flight time difference and a second set of path quality scores. S8. Based on the acoustic slowness field, generate a field representing the shortest flight time required to reach various locations within the positioning space from the anchor node; S9. By combining the shortest flight time fields generated by multiple anchor nodes, the three-dimensional coordinates that best match the second set of flight time differences are calculated as the position of the moving target.

[0008] In one specific embodiment, the channel response features in step S3 include: channel impulse response features extracted based on the complete waveform of the ultrasonic signal, and features extracted based on the channel state information of the radio frequency signal.

[0009] Preferably, the process of calculating the acoustic slowness of the preset grid unit within the positioning space in step S6 specifically includes: First, the positioning space is discretized into multiple preset grid cells, and a set of linear equations describing the relationship between ultrasonic flight time and acoustic slowness is established: ; in, The measurement vector is composed of the first set of flight time differences; The unknown vector to be solved is composed of the acoustic slowness of each preset grid cell; The geometric path matrix has the following elements. Representing the Measurement path between anchor nodes in the first... The length within a preset grid cell.

[0010] Then, a diagonal weight matrix is ​​defined based on the path quality scores of the first group. The slowness is iteratively corrected using an algebraic reconstruction algorithm. Its iterative formula is: ; in, For the first The slowness vector of the next iteration; For the first The updated slowness vector after the next iteration; It is a relaxation factor; Geometric path matrix The transpose of the matrix; This is a diagonal weight matrix, with its diagonal elements... By the The path quality score is composed of the corresponding measurement paths; For measurement vectors; The final update vector is obtained by applying a relaxation factor to the adjustment obtained from the back projection. Scale it, then add it to the previous slowness vector. This allows us to obtain a new estimate of the acoustic slowness field that more closely approximates the actual situation. .

[0011] The iterative process reduces the weighted residuals of the equations until a convergent sonic slowness field is obtained.

[0012] In one specific embodiment, the step of generating the shortest flight time field in step S8 is specifically as follows: The shortest flight time field is generated by solving the Eikonal equation, using the location of a single anchor node as the source point. The Eikonal equation describes the shortest flight time field. The gradient and the sonic slowness field Relationship: ; in, To locate the coordinates of any position within the positioning space; The acoustic slowness field at position The value; The shortest flight time field is located at The value; For gradient operators; The modulus representing the orientation quantity.

[0013] The fast travel algorithm is used to solve the Eikonal equation numerically.

[0014] Preferably, calculating the position of the moving target in step S9 specifically involves: Construct a weighted least squares objective function And by iteratively searching, the three-dimensional spatial coordinates that minimize the value of the objective function are found. The coordinates represent the position of the moving target. The objective function is defined as follows: ; in, For the objective function at the candidate position The function value; The number of anchor nodes participating in this positioning; For the index of the anchor node, traverse from 1 to... ; is the three-dimensional coordinate vector of the candidate moving target's position; By the first The shortest flight time field is obtained by searching or interpolating from the first... Each anchor node propagates to the candidate position. The theoretical shortest flight time; The moving target in the second set of measurement data is related to the first... The time difference of flight measured between each anchor node; Is with the first The weights corresponding to each measurement path are determined by the path quality score of that path. Sure; Indicates all The weighted squared errors generated by each anchor node involved in the positioning are summed.

[0015] The iterative search is performed using a nonlinear optimization algorithm.

[0016] Furthermore, the sonic slowness field is a discretized three-dimensional data field corresponding to the preset grid cell, wherein each data value is the sonic slowness result calculated by the tomographic reconstruction algorithm for a preset grid cell based on the first set of time-of-flight differences and the first set of path quality scores.

[0017] Furthermore, the shortest flight time field is a three-dimensional data field, the value of which represents the shortest time required for a sound wave to travel from a single anchor node to each preset grid cell within the positioning space.

[0018] Preferably, the method further includes: periodically executing steps S5 and S6 to dynamically update the sound slowness field.

[0019] This invention provides a wireless positioning method based on hybrid communication modes, which has the following advantages: 1. This invention uses multiple fixed anchor nodes to perform mutual measurements to obtain the first set of flight time differences and the first set of path quality scores. It then uses a tomographic reconstruction algorithm to calculate the sound velocity field that reflects the acoustic characteristics of the real-time environment. This enables the positioning system to perceive and quantify the uneven distribution of spatial sound velocity caused by factors such as temperature and airflow. This overcomes the model error caused by using fixed sound velocity values ​​in traditional methods and improves the adaptability and positioning accuracy of the positioning method to environmental changes.

[0020] 2. This invention generates a path quality score that can quantitatively evaluate the reliability of the signal propagation path by simultaneously transmitting and receiving radio frequency signals and ultrasonic signals and extracting the channel response characteristics of the two signals. This path quality score is used as a key weighting factor in the subsequent reconstruction of the acoustic slowness field and the calculation of the moving target position, which amplifies the contribution of high-quality measurements such as line-of-sight paths and suppresses the influence of low-quality measurements such as non-line-of-sight paths, thereby enhancing the anti-interference ability of the positioning algorithm in complex propagation environments.

[0021] 3. Based on the reconstructed non-uniform acoustic slowness field, this invention generates a shorter flight time field with more accurate physical meaning by solving the Eikonal equation, replacing the distance calculation model based on the assumption of straight-line propagation. The shorter flight time field can accurately describe the actual propagation path and time of sound waves in non-uniform media. When performing the final position calculation, it eliminates the ranging error caused by the curvature of the sound ray, making the positioning result closer to the real physical process, thereby improving the final positioning accuracy. Attached Figure Description

[0022] Figure 1 This is a schematic diagram of the wireless positioning system of the present invention; Figure 2 This is a flowchart illustrating the wireless positioning method of the present invention; Figure 3 This is a flowchart of the dual-mode signal processing and path quality assessment of the present invention; Figure 4 This is a flowchart of the environmental sound slowness field tomography reconstruction process of the present invention; Figure 5 This is a flowchart illustrating the target positioning process of the present invention. Figure 6 This is a flowchart of the dynamic update process of the sound slowness field of the present invention.

[0023] Among them, 10 is the anchor node; 20 is the moving target; 30 is the central processing server; 11 is the first radio frequency transceiver unit; 12 is the first ultrasonic transceiver unit; 13 is the first synchronization control unit; 14 is the first data processing unit; 15 is the first communication interface unit; 21 is the second radio frequency transceiver unit; 22 is the second ultrasonic transceiver unit; 23 is the second synchronization control unit; 24 is the second data processing unit; and 25 is the second communication interface unit. Detailed Implementation

[0024] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0025] See attached document Figure 1 , Figure 1 This is a schematic diagram of a wireless positioning system according to an embodiment of the present invention. The present invention provides a wireless positioning system based on a hybrid communication mode, which is used to execute the positioning method of the present invention. The system includes multiple fixed anchor nodes 10, one or more mobile targets 20, and a central processing server 30.

[0026] Multiple anchor nodes 10 are deployed within the positioning space, and the three-dimensional coordinates of the position of each anchor node 10 are pre-calibrated and known. The anchor nodes 10 are used to transmit and receive signals bidirectionally with other anchor nodes 10 or moving targets 20 to obtain the measurement data required for positioning.

[0027] In one specific embodiment, each anchor node 10 includes: a first radio frequency transceiver unit 11, a first ultrasonic transceiver unit 12, a first synchronization control unit 13, a first data processing unit 14, and a first communication interface unit 15.

[0028] The first radio frequency transceiver unit 11 is used to transmit and receive radio frequency signals. The first ultrasonic transceiver unit 12 is used to transmit and receive ultrasonic signals. The first synchronization control unit 13 is connected to the first radio frequency transceiver unit 11 and the first ultrasonic transceiver unit 12, and is used to generate a synchronization trigger signal to control the first radio frequency transceiver unit 11 and the first ultrasonic transceiver unit 12 to transmit radio frequency signals and ultrasonic signals at the same time.

[0029] The first data processing unit 14 is used to calculate the time-of-flight difference (TOF) of the received radio frequency signal and ultrasonic signal, extract channel response features from the two signals, and then generate a path quality score characterizing the quality of the signal propagation path. The first communication interface unit 15 is used to send the measurement data generated by the first data processing unit 14, such as the TOF and path quality score, to other anchor nodes 10 or the central processing server 30.

[0030] The moving target 20 is the object whose position within the positioning space needs to be determined. In one specific embodiment, the internal structure of the moving target 20 corresponds to the structure of the anchor node 10, including: a second radio frequency transceiver unit 21, a second ultrasonic transceiver unit 22, a second synchronization control unit 23, a second data processing unit 24, and a second communication interface unit 25. The functions of each unit are consistent with the functions of the corresponding units in the anchor node 10.

[0031] A central processing server 30 is connected to multiple anchor nodes 10 via a communication network. The central processing server 30 receives measurement data collected and transmitted by each anchor node 10 and executes computationally intensive algorithms. Specifically, the central processing server 30 is configured to perform the following operations: based on a first set of time-of-flight differences and a first set of path quality scores measured between the anchor nodes 10, it calculates the sonic slowness of a preset grid cell within the positioning space using a tomographic reconstruction algorithm, generating and storing a sonic slowness field; based on this sonic slowness field, it generates one or more shortest-time-of-flight fields; finally, based on a second set of time-of-flight differences measured between the moving target 20 and the anchor nodes 10, it combines the shortest-time-of-flight fields to calculate the position of the moving target 20. In another embodiment, the above calculations can also be performed collaboratively by one or more anchor nodes 10.

[0032] See attached document Figure 2 , Figure 2 This is a flowchart illustrating a wireless positioning method according to an embodiment of the present invention. The positioning method of the present invention is executed by the system described in the first part, and its overall process is divided into two main stages: an ambient sound slow-motion field modeling stage and a moving target positioning stage.

[0033] The ambient sound desiccation field modeling stage is a preparatory step before locating a moving target. Its purpose is to generate a three-dimensional sound desiccation field that accurately reflects the acoustic characteristics within the current location space. This stage is collaboratively completed by multiple anchor nodes 10 in the system under the scheduling of the central processing server 30. First, the system controls multiple fixed anchor nodes 10 to perform mutual measurements across the entire mesh.

[0034] That is, one anchor node 10 is designated as the transmitter in sequence, and the remaining anchor nodes 10 are designated as receivers, simultaneously transmitting radio frequency signals and ultrasonic signals. The receiving anchor nodes 10 process the received dual-mode signals, calculate the time difference of flight, extract channel response features, and generate path quality scores. This process is repeated until the first set of time differences of flight and the first set of path quality scores covering all or most of the measurement paths in the anchor node network are obtained, and this set of data is sent to the central processing server 30.

[0035] After receiving the first set of data, the central processing server 30 executes the tomographic reconstruction algorithm. The tomographic reconstruction algorithm mathematically discretizes the positioning space into multiple three-dimensional grid cells and establishes a large system of linear equations based on the geometric relationship between the first set of time-of-flight differences and paths.

[0036] The first set of path quality scores serves as a weighting factor in this process, adjusting the contribution of different measurement paths to the reconstruction. By iteratively solving this weighted linear equation system, the slowness value of each grid cell is finally calculated. The slowness values ​​of all grid cells together constitute a three-dimensional slowness field, which is stored in the central processing server 30 for use in subsequent localization stages.

[0037] The moving target localization phase is executed after the acoustic slowness field is generated. When a moving target 20 needs to be localized, the moving target 20 interacts with multiple anchor nodes 10 in the localization space. Similar to the aforementioned modeling phase, a second set of time-of-flight differences and a second set of path quality scores are generated by transmitting and receiving dual-mode signals to connect the moving target 20 with each anchor node 10.

[0038] Subsequently, the central processing server 30 uses the known location of each anchor node 10 participating in this positioning as the source point. Based on the stored sound deslowness field, it generates an independent shortest flight time field for each anchor node 10 by solving the Eikonal equation. Each shortest flight time field is a three-dimensional data field, and the value at any point in the field represents the shortest time required for a sound wave to propagate from the corresponding source anchor node 10, following the path defined by the sound deslowness field, to that point.

[0039] Finally, the central processing server 30 combines the shortest flight time fields generated by multiple anchor nodes 10 to calculate the position. The calculation process is a nonlinear optimization process whose goal is to search for a coordinate point in the three-dimensional positioning space that minimizes the weighted error between the calculated flight time value found from each of the shortest flight time fields and the measured value of the second set of flight time differences. The second set of path quality scores is used as a weighting factor here. The coordinate point that minimizes this weighted error is determined as the final position of the moving target 20.

[0040] See attached document Figure 3 , Figure 3 This is a flowchart of dual-mode signal processing and path quality assessment according to an embodiment of the present invention. The flowchart details the specific implementation of steps S1 to S4 in the method of the present invention.

[0041] In this embodiment, the signal transmitting end, such as an anchor node 10, under the control of its first synchronization control unit 13, transmits signals through its first radio frequency transceiver unit 11 and first ultrasonic transceiver unit 12 at a given time. Simultaneous transmission of radio frequency signals and ultrasonic signals.

[0042] The signal receiver, for example, another anchor node 10, at time... Received radio frequency signal, at time An ultrasonic signal was received. Since the propagation speed of radio frequency signals is close to the speed of light, their propagation time over the indoor positioning distance is negligible; therefore, it can be considered... The first data processing unit 14 calculates the time difference between the two signals. : ; The flight time difference Numerically equivalent to the flight time of an ultrasonic signal from the transmitter to the receiver.

[0043] In calculating the time difference of flight Simultaneously, the first data processing unit 14 at the receiving end performs channel response feature extraction. For the complete waveform of the received ultrasonic signal, it first performs matched filtering with the local transmitted waveform template to obtain the channel impulse response of the ultrasonic propagation path. Subsequently, it extracts a set of numerical features that characterize the ultrasonic propagation path from the channel impulse response. These features include, but are not limited to: the peak amplitude of the first-path component, the Rice K-factor (defined as the ratio of the power of the first-path component to the total power of subsequent multipath components), the root mean square delay spread, and the signal-to-noise ratio.

[0044] For a received radio frequency signal, if it employs orthogonal frequency division multiplexing modulation, channel state information can be obtained from its physical layer. The channel state information obtained by the physical layer describes the amplitude and phase response of the signal on each orthogonal frequency division multiplexing subcarrier. The first data processing unit 14 extracts a set of numerical features from the channel state information data obtained by the physical layer. This set of numerical features includes, but is not limited to: the variance of the amplitude of the channel state information obtained by the physical layer of all subcarriers, and the standard deviation of the phase difference of the channel state information obtained by the physical layer of adjacent subcarriers.

[0045] The first data processing unit 14 combines the extracted ultrasonic channel features and radio frequency channel features into a multi-dimensional feature vector. Multidimensional feature vectors The input is fed into a pre-trained classification or regression model to generate a standardized path quality score. This model can be a support vector machine, logistic regression model, or gradient boosting decision tree; its role is to establish a mapping between multidimensional feature vectors and the quality of the signal propagation path.

[0046] The final path quality score is a scalar value between 0 and 1. A path quality score close to 1 indicates that the current signal propagation path is line-of-sight or near-line-of-sight, indicating high measurement quality; a path quality score close to 0 indicates that the current signal propagation path is non-line-of-sight or subject to severe multipath interference, indicating low measurement quality. This path quality score is related to the corresponding time-of-flight difference. Together with the output of this measurement, it is used for subsequent acoustic slowness field reconstruction or moving target localization.

[0047] See attached document Figure 4 , Figure 4 This is a flowchart of environmental sound slowness field tomography reconstruction according to an embodiment of the present invention. The flowchart details the specific implementation of steps S5 and S6 in the method of the present invention, the core of which lies in calculating the sound slowness distribution in the positioning space through acoustic tomography technology.

[0048] The process is initiated by the central processing server 30. First, the central processing server 30 schedules N fixed anchor nodes 10 within the positioning space to perform a mutual measurement across the entire or a portion of the grid. This measurement process systematically traverses preset anchor node pairs, with one anchor node acting as the transmitter and one or more other anchor nodes acting as the receiver, executing the dual-mode signal processing and path quality assessment process in the third part. Upon completion of this process, the first set of data is generated, which includes the time-of-flight differences corresponding to the measurement paths between M anchor nodes. and path quality score set .

[0049] Before executing the reconstruction algorithm, the central processing server 30 first mathematically discretizes the three-dimensional positioning space, dividing it into... Each voxel is a non-overlapping three-dimensional mesh element, also known as a voxel. It is assumed to have a uniform slowness value. The ultrasonic signal in the first... Flight time on the measurement path This relates to the line integral of the acoustic slowness along the path. In the discretized model, this relationship can be approximated by a linear equation: ; in, It is the first The individual element's slowness value to be solved. It is the first Measurement path between anchor nodes in the first... The length within a preset grid cell. Establishing this equation for all M measurement paths constitutes a large system of linear equations. .

[0050] This embodiment employs a weighted algebraic reconstruction algorithm to iteratively solve the system of linear equations. The first set of path quality scores... Used to construct a diagonal weight matrix Diagonal weight matrix diagonal elements With the corresponding path Value correlation makes high-quality measurement paths have a greater impact during iterative correction.

[0051] The initial condition for iterative solution is the initial sonic slowness of all voxels. A uniform reference value is set, for example, the reciprocal of the speed of sound in air at standard atmospheric pressure and room temperature. The sound manifold field is then updated using the following iterative formula: ; in, For the first The slowness vector of the next iteration; For the first The updated slowness vector after the next iteration; It is a relaxation factor; Geometric path matrix The transpose of .

[0052] The iterative process continues until a preset convergence condition is met, such as the change in the acoustic slowness vector between two iterations being less than a threshold, or the weighted residual... The norm is less than a threshold. The vector obtained after iterative convergence. This represents the slowness calculation result for J grid cells. This vector... It is reconstructed into a three-dimensional data field, forming the final acoustic slowness field that can reflect the current acoustic characteristics of the environment, and stored in the central processing server 30.

[0053] See attached document Figure 5 , Figure 5 This is a flowchart of target localization according to an embodiment of the present invention. The flowchart details the specific implementation of steps S7 to S9 in the method of the present invention, which are used to calculate the precise location of the moving target based on the constructed ambient sound slowness field.

[0054] When a moving target 20 enters the positioning space and needs to be positioned, the moving target 20 performs bidirectional signal interaction with m anchor nodes 10 within its communication range. This process follows the dual-mode signal processing and path quality assessment procedure in Part Three, generating a time difference of flight for each propagation path between the moving target 20 and the anchor node 10. and the corresponding path quality score (in These data together constitute the second set of measurement data used in this positioning calculation.

[0055] ; in, It is a vector representing any position within the location space. It is the gradient operator. Is the slowness field at position The value, It is to be solved, starting from the source point Arrival Location The shortest flight time. This embodiment uses a fast travel algorithm to solve this equation. This algorithm calculates and expands outward layer by layer on a discretized three-dimensional grid, starting from the source point, until the shortest flight time field covering the entire positioning space is obtained. This process will generate m different shortest flight time fields for m anchor nodes.

[0056] After obtaining all m shortest flight time fields, the central processing server 30 performs a nonlinear optimization process to calculate the position of the moving target 20. This process involves constructing and minimizing a weighted least squares objective function. This is achieved by [implementing the objective function]. The objective function is defined as follows: ; in, is the three-dimensional coordinate vector of the candidate moving target's position; The moving target in the second set of measurement data is related to the first... The time difference of flight measured between each anchor node; By the first The shortest flight time field is obtained by searching or interpolating from the first... Each anchor node propagates to the candidate position. The theoretical shortest flight time; Is with the first The weights corresponding to each measurement path are determined by the path quality score of that path. Sure; Indicates all The weighted squared errors generated by each anchor node involved in the positioning are summed.

[0057] The central processing server 30 uses either the Levenberg-Marquardt algorithm or the Gauss-Newton algorithm, starting with an initial position estimate, to process the objective function. Perform an iterative minimization search. At each iteration, the algorithm adjusts the candidate positions. To reduce the value of the objective function. When the iteration converges, i.e., the value of the objective function no longer decreases significantly or the change in the candidate position is less than a preset threshold, the candidate position is... This is the final position calculation result that has been determined as the moving target 20.

[0058] See attached document Figure 6 , Figure 6 This is a flowchart illustrating the dynamic update of a sound slowness field according to an embodiment of the present invention. Because environmental parameters of the positioning space (e.g., temperature, humidity, airflow) change over time, a static sound slowness field gradually loses its accuracy. This embodiment provides a dynamic update mechanism to maintain the consistency between the sound slowness field and the actual physical environment.

[0059] This dynamic update mechanism is managed and executed by the central processing server 30. The update triggering condition can be based on a time period or system performance monitoring. In one embodiment, a periodic triggering mechanism is used. The central processing server 30 has an internal timer set to a preset time period, such as 30 minutes. When the timer expires, an update process for the sonic slowness field is automatically triggered.

[0060] In another embodiment, a condition-triggered mechanism based on performance monitoring is employed. When performing the moving target position calculation in Part Five, the central processing server 30 calculates the final weighted least squares objective function. The server continuously monitors the moving average of this residual value. When this moving average continuously exceeds a preset residual threshold, it indicates that the current sonic field model has deviated significantly from the actual environment, at which point the system triggers an update process.

[0061] In another specific implementation, the conditional triggering mechanism can also be based on monitoring fixed paths between anchor nodes. The central processing server 30 periodically instructs one or more fixed anchor node pairs to exchange signals and obtain their latest time-of-flight difference. The server compares this latest measurement with the theoretical time of flight calculated using the current acoustic slowness field. If the absolute value of the difference consistently exceeds a preset path deviation threshold, an update process is also triggered.

[0062] Once the update process is triggered, the central processing server 30 will instruct the system to completely re-execute all steps in the fourth part of the environmental sound slowness field tomography reconstruction module. This includes: rescheduling all anchor nodes 10 to perform full-grid mutual measurements to obtain a completely new set of first-set time-of-flight differences and path quality scores, and based on this new set of data, recalculating a new sound slowness field using a weighted algebraic reconstruction algorithm. This newly generated sound slowness field will overwrite and replace the old sound slowness field stored in the central processing server 30 for all subsequent moving target localization calculations.

Claims

1. A wireless positioning method based on hybrid communication modes, characterized in that, Includes the following steps: S1. Acquire the radio frequency signal and ultrasonic signal synchronously transmitted by the anchor node or moving target as the transmitter; S2. Calculate the time difference between the radio frequency signal and the ultrasonic signal; S3. Extract channel response features from the radio frequency signal and the ultrasonic signal; S4. Based on the channel response characteristics, generate a path quality score that characterizes the quality of the current signal propagation path; S5. Apply steps S1 to S4 to signal propagation between multiple fixed anchor nodes to generate a first set of flight time differences and a first set of path quality scores. S6. Based on the first set of flight time differences and the first set of path quality scores, the acoustic slowness of the preset grid cells in the positioning space is calculated by the tomographic reconstruction algorithm, and an acoustic slowness field is generated. S7. Apply steps S1 to S4 to signal propagation between the moving target and the anchor node to generate a second set of flight time difference and a second set of path quality scores. S8. Based on the acoustic slowness field, generate a field representing the shortest flight time required to reach various locations within the positioning space from the anchor node; S9. By combining the shortest flight time fields generated by multiple anchor nodes, the three-dimensional coordinates that best match the second set of flight time differences are calculated as the position of the moving target.

2. The wireless positioning method based on hybrid communication mode according to claim 1, characterized in that, In step S3, the channel response features include channel impulse response features extracted based on the complete waveform of the ultrasonic signal and features extracted based on the channel state information of the radio frequency signal.

3. The wireless positioning method based on hybrid communication mode according to claim 1, characterized in that, In step S6, calculating the slowness of the preset grid cells within the positioning space specifically involves: A system of linear equations is constructed between the sound slowness and the first set of flight time differences. Based on the weights defined by the first set of path quality scores, the sound slowness is iteratively corrected using the tomographic reconstruction algorithm to reduce the residual between the predicted value of the system of equations and the first set of flight time differences.

4. The wireless positioning method based on hybrid communication mode according to claim 3, characterized in that, In step S6, the tomographic reconstruction algorithm is an algebraic reconstruction technique algorithm.

5. The wireless positioning method based on hybrid communication mode according to claim 1, characterized in that, In step S8, the step of generating the shortest flight time field specifically includes: The anchor node is set as the source point, and the shortest flight time field is generated by solving an Eikonal equation. The Eikonal equation describes the relationship between the gradient of the shortest flight time field and the acoustic slowness field, and the solution is completed using a fast travel algorithm.

6. The wireless positioning method based on hybrid communication mode according to claim 5, characterized in that, The S9 step is specifically as follows: A weighted least squares objective function is constructed to measure the degree of matching between the value of the shortest flight time field at the candidate position and the second set of flight time differences; and through iterative search, the three-dimensional spatial coordinates that minimize the value of the weighted least squares objective function are found, and the three-dimensional spatial coordinates are determined as the position of the moving target.

7. The wireless positioning method based on hybrid communication mode according to claim 6, characterized in that, The weighted least squares objective function uses the second set of path quality scores as weights.

8. The wireless positioning method based on hybrid communication mode according to claim 1, characterized in that, In step S6, the acoustic slowness field is a discretized three-dimensional data field corresponding to the preset grid cell, wherein each data value is the acoustic slowness result calculated by the tomographic reconstruction algorithm for a preset grid cell based on the first set of flight time differences and the first set of path quality scores.

9. The wireless positioning method based on hybrid communication mode according to claim 5, characterized in that, The shortest flight time field is a three-dimensional data field. The value of the shortest flight time field represents the shortest time required for a sound wave to travel from a single anchor node to each preset grid cell in the positioning space.

10. The wireless positioning method based on hybrid communication mode according to claim 1, characterized in that, The method further includes: periodically executing steps S5 and S6 to dynamically update the sound slowness field.

Images