Cross-band TDOA error suppression positioning method based on virtual base station mapping
By using virtual base station mapping and cross-band signal processing, combined with QR decomposition and Kalman filtering, the high hardware cost and insufficient accuracy of 5G positioning technology in complex indoor environments are solved, achieving high-precision and low-cost dynamic positioning.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- BEIHANG UNIV
- Filing Date
- 2026-02-10
- Publication Date
- 2026-04-21
AI Technical Summary
Existing 5G positioning technology suffers from high hardware costs, insufficient positioning accuracy and stability in complex indoor environments, especially with severe error accumulation in dynamic scenarios. Traditional methods introduce errors when dealing with nonlinear observation models, and single-band positioning is susceptible to systematic errors.
By synchronously acquiring signals from different frequency bands, identifying non-line-of-sight propagation links and creating virtual base stations, and combining QR decomposition and square root unscented Kalman filtering frameworks, a set of positioning equations is constructed, linearized processing and state prediction are performed, and virtual base station parameters are dynamically managed.
With limited physical base station deployment, it significantly improves positioning accuracy and stability, enhances system adaptability and robustness, reduces hardware costs, and achieves high-precision dynamic positioning.
Smart Images
Figure CN121899747A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of wireless communication and positioning technology, and particularly relates to a cross-band TDOA error suppression positioning method based on virtual base station mapping. Background Technology
[0002] With the commercialization of 5G mobile communication technology, 5G positioning technology is playing an increasingly important role in scenarios requiring high accuracy, such as autonomous driving, IoT device tracking, and robot navigation. Traditional high-precision positioning methods mainly rely on time difference of arrival (TDOA) measurement technology between multiple 5G base stations and terminal devices. However, in complex indoor application environments, such as large shopping malls, factories, or underground parking lots, achieving ideal positioning accuracy often requires deploying dense arrays of physical base stations, leading to expensive hardware and maintenance costs. Meanwhile, walls, furniture, and personnel movement in indoor environments can trigger strong multipath effects and non-line-of-sight propagation, causing unpredictable errors in signal propagation time, further significantly reducing the accuracy and reliability of the positioning system. Especially in dynamic scenarios with frequent personnel movement, errors accumulate over time. Traditional extended Kalman filtering algorithms, when dealing with such highly nonlinear TDOA observation models, must linearize the model, a process that itself introduces significant errors, ultimately weakening the stability and accuracy of the entire positioning system.
[0003] Existing technologies for improving performance mainly focus on two directions: first, increasing the number of physical base stations to improve geometric layout, but this directly increases deployment costs; second, using the Extended Kalman Filter (EKF) algorithm to smooth motion trajectories. However, when facing strongly nonlinear time-of-arrival (TOA) observation equations, the linearization process of the Jacobian matrix upon which the EKF relies is an inherent source of error. More seriously, when the spatial geometry of base stations is poor, the observation equations are prone to ill-conditioned characteristics, causing the state covariance matrix to lose its mathematical positive definiteness during iterative updates, ultimately leading to filter divergence and complete failure of the positioning results. Furthermore, some methods for non-line-of-sight propagation link identification based on channel state information, while theoretically feasible, have extremely high computational complexity, making it difficult to meet the stringent requirements of low latency and real-time performance for applications such as autonomous vehicles and real-time asset tracking.
[0004] On the other hand, most existing positioning methods rely solely on signals from a single frequency band, such as the 5G Sub-6GHz band or millimeter-wave band. However, in real and variable radio propagation environments, signals from different frequency bands exhibit inherent differences in non-line-of-sight propagation sensitivity, obstacle penetration capability, and multipath reflection characteristics. Positioning based on a single frequency band is susceptible to the systematic errors inherent in that band; for example, certain frequency bands may experience severe attenuation under the influence of specific materials, leading to complete signal loss or instability. This limitation renders existing systems inadequately robust and adaptable in complex indoor-outdoor mixed environments with numerous obstructions, making it difficult to guarantee continuous and stable high-precision positioning services. Therefore, the industry urgently needs a high-precision dynamic positioning solution that can effectively suppress non-line-of-sight propagation errors, possess strong numerical stability, and integrate the advantages of multi-frequency band signals while ensuring economic viability. Summary of the Invention
[0005] To address the aforementioned technical problems, this invention proposes a cross-band TDOA error suppression positioning method based on virtual base station mapping, thereby resolving the issues present in the prior art.
[0006] In a first aspect, to achieve the above objective, the present invention provides a cross-band TDOA error suppression positioning method based on virtual base station mapping, comprising the following steps: Simultaneously acquire signals from at least two base stations in different frequency bands, and extract the time difference of arrival and received signal strength characteristics; Based on the arrival time difference and received signal strength characteristics, non-line-of-sight propagation links are jointly identified, and a virtual base station is created or updated for each identified non-line-of-sight propagation link. Construct a set of positioning equations that include all real and virtual base stations; The positioning equations are linearized and solved iteratively using QR decomposition to obtain the initial position estimate. The initial position estimate and the time difference of arrival measurement are input into the square root unscented Kalman filter framework for state prediction and update, and the final positioning result is output. The parameters of the virtual base station are updated based on the final positioning result.
[0007] Optionally, the process of acquiring signals from at least two different frequency band base stations includes: simultaneously receiving and processing signals from 5G-A band base stations and signals from WIFI 6E band base stations.
[0008] Optionally, the process of jointly identifying non-line-of-sight propagation links includes: making a comprehensive judgment based on whether the received signal strength is lower than a threshold based on distance estimation, and combining whether the difference between the time difference of arrival measurement and the delay estimated based on geometric distance exceeds a preset threshold.
[0009] Optionally, the process of creating or updating a virtual base station for each identified non-line-of-sight propagation link includes: inverting the non-line-of-sight additional path length based on the signal attenuation model, and translating the position of the real base station along the direction of the signal angle of arrival from the terminal to the real base station to determine the position coordinates of the virtual base station.
[0010] Optionally, the process of lifecycle management of the virtual base station includes: maintaining a confidence parameter and a lifetime parameter for each virtual base station, updating the confidence parameter based on the final positioning result, and determining whether to remove the virtual base station from the system based on the updated parameter and a preset threshold.
[0011] Optionally, the process of linearizing the positioning equations includes: using the previous time-time position estimate output by the square root unscented Kalman filter framework as the linearization point, performing a first-order Taylor expansion on the nonlinear time difference of arrival observation equations containing real base stations and virtual base stations to form a linearized system.
[0012] Optionally, the state vector of the square root unscented Kalman filter framework includes the position and velocity state of the terminal, as well as the position coordinates of all active virtual base stations.
[0013] Optionally, the observation vector of the square root unscented Kalman filter framework is the original time difference of arrival measurement, and its observation equation is a geometric distance difference function that includes the terminal location, the real base station location, and the virtual base station location.
[0014] Secondly, the present invention also provides a computer terminal device, comprising: One or more processors; A memory, coupled to the processor, for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the steps of the cross-band TDOA error suppression positioning method based on virtual base station mapping in the first aspect above.
[0015] Thirdly, the present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, it implements the steps of the cross-band TDOA error suppression positioning method based on virtual base station mapping in the first aspect described above.
[0016] Compared with the prior art, the present invention has the following advantages and technical effects: This invention provides a cross-band TDOA error suppression positioning method based on virtual base station mapping. By creating a virtual base station for each identified non-line-of-sight (NLOS) propagation link, the NLOS propagation error, which is difficult to process directly, is transformed into an estimable geometric parameter, thereby effectively increasing the number of signal paths available for positioning at the algorithm level. This method combines QR decomposition iterative solution with a square root unscented Kalman filter framework. The former effectively suppresses numerical instability caused by poor base station geometry or ill-conditioned observation equations, while the latter directly processes the nonlinear time difference of arrival (TDOA) observation model through unscented transformation, avoiding the errors introduced by the linearization of traditional extended Kalman filters, significantly improving positioning accuracy and trajectory smoothness. Simultaneously, by fusing signal features from different frequency bands such as 5G-A and WIFI 6E for joint decision-making and positioning calculation, the system's adaptability and robustness in complex and variable propagation environments are enhanced. Furthermore, a dynamic virtual base station lifecycle management mechanism ensures the effective utilization of system resources. Ultimately, this invention achieves high-precision, high-stability, and more cost-effective dynamic positioning under the condition of a limited number of physical base stations. Attached Figure Description
[0017] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings: Figure 1 This is a flowchart illustrating a cross-band TDOA error suppression and positioning method based on virtual base station mapping according to an embodiment of the present invention. Figure 2 This is a flowchart of the SR-UKF fusion algorithm according to an embodiment of the present invention; Figure 3 This is a performance comparison chart between the algorithm of this invention and existing algorithms. Detailed Implementation
[0018] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.
[0019] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.
[0020] Example 1 like Figure 1As shown, this embodiment provides a cross-band TDOA error suppression and positioning method based on virtual base station mapping, including: synchronously acquiring signals from at least two base stations in different frequency bands, and extracting the time difference of arrival and received signal strength features of the signals; Based on the arrival time difference and received signal strength characteristics, non-line-of-sight propagation links are jointly identified, and a virtual base station is created or updated for each identified non-line-of-sight propagation link. Construct a set of positioning equations that include all real and virtual base stations; The positioning equations are linearized and solved iteratively using QR decomposition to obtain the initial position estimate. The initial position estimate and the time difference of arrival measurement are input into the square root unscented Kalman filter framework for state prediction and update, and the final positioning result is output. The parameters of the virtual base station are updated based on the final positioning result.
[0021] Furthermore, the process of acquiring signals from at least two different frequency band base stations includes: simultaneously receiving and processing signals from 5G-A band base stations and signals from WIFI 6E band base stations.
[0022] Specifically, the implementation process of this embodiment includes: Step S1: Select a commercial module that supports the 5G NR protocol or an open-source software radio platform that is compatible with the 5G frequency band to build a 5G signal acquisition system adapted to the China Unicom network. At the same time, use the developed mobile APP to receive WIFI signals from each router.
[0023] The signal acquisition frequency is 10Hz, and the synchronization method is based on the synchronization and time synchronization between base stations.
[0024] The terminal receives signals from M base stations and extracts a set of TDOA measurement values. and the corresponding set of RSS values The data is preprocessed, with initial noise reduction achieved through moving average filtering. During the signal acquisition phase, time synchronization and inter-band delay calibration are performed on signals from different frequency bands to eliminate systematic delay differences introduced by hardware and propagation paths.
[0025] Furthermore, the process of jointly identifying non-line-of-sight propagation links includes: making a comprehensive judgment based on whether the received signal strength is lower than a threshold based on distance estimation, and combining whether the difference between the time difference of arrival measurement and the delay estimated based on geometric distance exceeds a preset threshold.
[0026] Specifically, the implementation process of this embodiment includes: Step S2, NLOS identification and virtual base station mapping (multi-band joint decision), includes the following steps: S2-1: For NLOS identification, a joint decision based on RSS attenuation and TDOA consistency is adopted. For the i-th link, if the following conditions are met... and If the signal is strong, it is identified as an NLOS link and assigned a higher confidence level. If only a specific frequency band is abnormal, it may be frequency-specific interference and will be downweighted.
[0027] in: For the first The first frequency band Received signal strength (RSS) of the link. For the first The first frequency band The strength threshold of the link, To measure TDOA, For time delays estimated based on geometric distance, This is the threshold for time delay difference.
[0028] Furthermore, the process of creating or updating a virtual base station for each identified non-line-of-sight propagation link includes: inverting the non-line-of-sight additional path length based on the signal attenuation model, and translating the position of the real base station along the signal angle of arrival direction from the terminal to the real base station to determine the position coordinates of the virtual base station.
[0029] Specifically, the implementation process of this embodiment includes: S2-2. Generate a virtual base station for each NLOS link. Assume the coordinates of the real base station are... The terminal's initial estimated location is The additional path length introduced by signal reflection or diffraction is Then the virtual base station coordinates We obtain the following by translating along the direction of the signal arrival angle: Based on signal attenuation model Reverse the additional path Along the direction of the signal angle of arrival, from the actual base station location Mapping the location of virtual base stations : ; ; in, , . RSS attenuation model In reverse estimation, β is the path loss exponent. Path loss exponents for signals in different frequency bands. Unlike other methods, the β value of the corresponding frequency band should be used in virtual base station location estimation. estimate.
[0030] Furthermore, the process of lifecycle management for the virtual base station includes: maintaining a confidence parameter and a lifetime parameter for each virtual base station, updating the confidence parameter based on the final positioning result, and determining whether to remove the virtual base station from the system based on the updated parameter and a preset threshold.
[0031] Specifically, the implementation process of this embodiment includes: S2-3. For the lifecycle management of virtual base stations, assign confidence levels to each virtual base station. and lifespan The confidence level is updated based on signal quality, and the lifetime decreases with each iteration. ; when or When that happens, remove the virtual base station.
[0032] S4-3. Virtual Base Station Parameter Update and Lifecycle Management: High-precision position estimation using SR-UKF output The confidence level, lifetime, and location coordinates of each virtual base station are reassessed and updated (using a separate filter to smooth the location of the virtual base station), thus completing the dynamic management closed loop.
[0033] S4-3-1, Confidence Update: For each virtual base station Calculate the residuals from this positioning: ; In a multi-band fusion system, the confidence update of a virtual base station is weighted based on the signal quality of each frequency band. Let the... A virtual base station in the frequency band The residuals below are The signal quality weight is The overall residual is then expressed as: ; Then, the confidence level is updated based on the residuals: ; in: The forgetting factor (taken as 0.8 here); The standard deviation of the residuals can be estimated based on historical residuals, with an initial confidence level. Set it to 0.5; S4-3-2, Lifetime Replacement and Elimination Mechanism: Lifetime of each virtual base station Decrease per cycle: ; The virtual base station will be removed from the system if any of the following conditions are met: 1. ; 2. (like ); 3. The distance between the virtual base station location and the estimated location of the current terminal exceeds the effective range (set to 50 meters here); S4-3-3, Virtual Base Station Location Smoothing (Optional): To improve the location stability of virtual base stations, their coordinates can also be smoothed using Kalman filtering. A location state vector is maintained for each virtual base station. A constant position model is used for filtering, and the observed values are the virtual base station positions inferred from the signals each time. This step can further improve the stability of the system during continuous tracking.
[0034] Furthermore, the process of linearizing the positioning equations includes: using the previous time-time position estimate output by the square root unscented Kalman filter framework as the linearization point, performing a first-order Taylor expansion on the nonlinear time difference of arrival observation equations containing real base stations and virtual base stations to form a linearized system.
[0035] Specifically, the implementation process of this embodiment includes: SR-UKF fusion algorithm such as Figure 2 As shown, step S4, SR-UKF fusion solution (multi-band state and observation extension), includes the following steps: S4-1, Optimal estimate of SR-UKF at the previous moment For the linearization point, for the nonlinear system of equations: ; set up At the initial estimation point Perform a first-order Taylor expansion at (the previous filtering result): ; ; ; The linearized system is obtained: ,in This is the position increment.
[0036] Let be a Jacobian matrix, and its th... The row element is: ; in, Let be a constant vector, and its th... The elements are: ; This is the noise vector.
[0037] Due to potentially poor base station geometry, the matrix It is prone to ill-conditioned problems, and direct inversion will amplify the error. Therefore, QR decomposition is used for numerically stable solutions.
[0038] For the coefficient matrix Perform QR decomposition: .
[0039] in, It is an orthogonal matrix. It is an upper triangular matrix. (The rest is missing.) The blocks are as follows: ; in It is a non-singular upper triangular matrix.
[0040] The least squares solution is: ,in for The first two columns.
[0041] Finally, use the obtained... Updated location estimation: ; Repeat step S4-1 above to... This serves as a new linearization point until the convergence condition is met: ; in For the preset convergence threshold (e.g.) (meters), or reaching the maximum number of iterations (e.g., 10 times). The final position estimate is denoted as , as the observation value for Kalman filtering.
[0042] Furthermore, the state vector of the square root unscented Kalman filter framework includes the position and velocity state of the terminal, as well as the position coordinates of all active virtual base stations.
[0043] Specifically, the implementation process of this embodiment includes: S4-2. State estimation based on square root unscented Kalman filter (SR-UKF): This embodiment uses SR-UKF instead of the traditional EKF to directly process the nonlinear TDOA observation model, smoothing the trajectory in the time domain and improving the continuity and dynamic performance of positioning. Specifically, it includes the following sub-steps: S4-2-1 Definition of State Vector and Motion Model: The state vector, containing the virtual base station location, is obtained using the data set from S4-1: ; in, The target's position at time k includes virtual base station location parameters corresponding to different frequency bands. Let the velocity of the target be at time k; The frequency band information for each virtual base station, which is currently active, is maintained separately in its attributes and used to calculate the corresponding path loss index. And observation noise weights.
[0044] Assuming the target moves at a constant velocity during the sampling period, the state transition equation is: ; in, Here is the state transition matrix: ; The sampling period is 0.1 seconds (used here). The noise is a process noise that follows a zero-mean Gaussian distribution. The covariance matrix is It is usually set as a diagonal matrix to reflect the uncertainty of position and velocity.
[0045] Covariance matrix: ; .
[0046] Furthermore, the observation vector of the square root unscented Kalman filter framework is the original time difference of arrival measurement, and its observation equation is a geometric distance difference function that includes the terminal location, the real base station location, and the virtual base station location.
[0047] Specifically, the implementation process of this embodiment includes: S4-2-2, Observation Model: The observed values are the original TDOA measurements. .
[0048] The observation equation is a nonlinear function: ; in, To determine the target position based on the state vector and the known set of base station locations (Including real and virtual base stations) Functions for calculating TDOA. For zero-mean observation noise, its covariance moment for The diagonal matrix. Considering the different measurement error characteristics of signals in different frequency bands, elements on the diagonal According to the generation of the first TDOA observations (corresponding base stations) ) link frequency band To set: ,in and .
[0049] S4-2-3, SR-UKF Algorithm Flow: 1. Initialization: Set initial state estimates and its error covariance square root factor (satisfy ).
[0050] 2. Prediction Step (Time Update): a.Sigma point generation: based on and Calculate a set of Sigma points by proportionally correcting the unscented transformation rule. .
[0051] b. Sigma point propagation: Propagating each Sigma point through the process model: .
[0052] c. Calculate the predicted state and its square root covariance: Calculate the mean of the predicted state. Furthermore, the square root factor of the predicted state covariance is calculated through QR decomposition and Cholesky factor update iteration. .
[0053] 3. Update step (measurement update): a. Observation and prediction: Regenerate a set of Sigma points (based on...) and ), and through nonlinear observation functions spread: .
[0054] b. Calculate the predicted observation mean Root of covariance of new information and cross-covariance matrix .
[0055] c. Calculate the Kalman gain: ,in This indicates an efficient back-substitution solution, avoiding direct inversion.
[0056] d. Status update: .
[0057] e. Square root covariance update: Calculate the square root factor of the updated state covariance using an efficient rank-1 Cholesky update or a rank-reducing algorithm. .
[0058] Filtered output The positional component in the equation is the final optimized smooth positioning result.
[0059] Figure 3 The diagram shows an error comparison between the algorithm provided in this embodiment and other existing algorithms. With the NLOS link accounting for 61.6% of the cross-band channel, the algorithm provided in this embodiment reduces the average error by 67.9% compared to the traditional least squares method, and by 41.2% compared to the traditional Kalman filter. Furthermore, the cross-band fusion algorithm reduces the average error by 46.2% and 55.7% compared to methods using only 5G signals and only WIFI signals, respectively. Overall, the algorithm provided in this embodiment has a higher overall complexity than traditional methods, but it achieves a significant improvement in accuracy and stability.
[0060] Example 2 In this embodiment, a computer terminal device is provided, including: One or more processors; A memory, coupled to the processor, for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the steps of the cross-band TDOA error suppression positioning method based on virtual base station mapping described above.
[0061] In this embodiment, a computer-readable storage medium is also provided, on which a computer program is stored. When the computer program is executed by a processor, it implements the steps of the above-described cross-band TDOA error suppression positioning method based on virtual base station mapping.
[0062] This invention provides a cross-band TDOA error suppression positioning method based on virtual base station mapping. By creating a virtual base station for each identified non-line-of-sight (NLOS) propagation link, the NLOS propagation error, which is difficult to process directly, is transformed into an estimable geometric parameter, thereby effectively increasing the number of signal paths available for positioning at the algorithm level. This method combines QR decomposition iterative solution with a square root unscented Kalman filter framework. The former effectively suppresses numerical instability caused by poor base station geometry or ill-conditioned observation equations, while the latter directly processes the nonlinear time difference of arrival (TDOA) observation model through unscented transformation, avoiding the errors introduced by the linearization of traditional extended Kalman filters, significantly improving positioning accuracy and trajectory smoothness. Simultaneously, by fusing signal features from different frequency bands such as 5G-A and WIFI 6E for joint decision-making and positioning calculation, the system's adaptability and robustness in complex and variable propagation environments are enhanced. Furthermore, a dynamic virtual base station lifecycle management mechanism ensures the effective utilization of system resources. Ultimately, this invention achieves high-precision, high-stability, and more cost-effective dynamic positioning under the condition of a limited number of physical base stations.
[0063] The above are merely preferred embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.
Claims
1. A cross-band TDOA error suppression positioning method based on virtual base station mapping, characterized in that, Includes the following steps: Simultaneously acquire signals from at least two base stations in different frequency bands, and extract the time difference of arrival and received signal strength characteristics; Based on the arrival time difference and received signal strength characteristics, non-line-of-sight propagation links are jointly identified, and a virtual base station is created or updated for each identified non-line-of-sight propagation link. Construct a set of positioning equations that include all real and virtual base stations; The positioning equations are linearized and solved iteratively using QR decomposition to obtain the initial position estimate. The initial position estimate and the time difference of arrival measurement are input into the square root unscented Kalman filter framework for state prediction and update, and the final positioning result is output. The parameters of the virtual base station are updated based on the final positioning result.
2. The method according to claim 1, characterized in that, The process of acquiring signals from at least two base stations in different frequency bands includes: simultaneously receiving and processing signals from 5G-A band base stations and WIFI 6E band base stations.
3. The method according to claim 1, characterized in that, The process of jointly identifying non-line-of-sight propagation links includes: making a comprehensive judgment based on whether the received signal strength is lower than a threshold based on distance estimation, and whether the difference between the time difference of arrival measurement and the delay estimated based on geometric distance exceeds a preset threshold.
4. The method according to claim 3, characterized in that, The process of creating or updating a virtual base station for each identified non-line-of-sight propagation link includes: inverting the non-line-of-sight additional path length based on the signal attenuation model, and translating the position of the real base station along the direction of the signal angle of arrival from the terminal to the real base station to determine the position coordinates of the virtual base station.
5. The method according to claim 4, characterized in that, The process of lifecycle management of the virtual base station includes: maintaining a confidence parameter and a lifetime parameter for each virtual base station, updating the confidence parameter based on the final positioning result, and determining whether to remove the virtual base station from the system based on the updated parameter and a preset threshold.
6. The method according to claim 1, characterized in that, The process of linearizing the positioning equations includes: taking the previous time position estimate output by the square root unscented Kalman filter framework as the linearization point, performing a first-order Taylor expansion on the nonlinear time difference of arrival observation equations containing real base stations and virtual base stations to form a linearized system.
7. The method according to claim 1, characterized in that, The state vector of the square root unscented Kalman filter framework contains the position and velocity state of the terminal, as well as the position coordinates of all active virtual base stations.
8. The method according to claim 7, characterized in that, The observation vector of the square root unscented Kalman filter framework is the original time difference of arrival measurement, and its observation equation is a geometric distance difference function that includes the terminal location, the real base station location, and the virtual base station location.
9. A computer terminal device, characterized in that, include: One or more processors; A memory, coupled to the processor, for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors perform the steps of the method as described in any one of claims 1-8.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1-8.