CN-122017884-A - Indoor pseudo satellite positioning method based on carrier phase multipath error modeling

CN122017884ACN 122017884 ACN122017884 ACN 122017884ACN-122017884-A

Abstract

The invention relates to an indoor pseudo-satellite positioning method based on carrier phase multipath error modeling, which comprises the steps of collecting pseudo-satellite observation data, calculating carrier phase observation values, eliminating systematic errors, fixing whole-cycle ambiguity, deducing carrier phase multipath errors, mapping the actual space layout of the pseudo-satellite into a two-dimensional grid structure, extracting carrier phase residual errors and carrier-to-noise ratios, constructing a two-channel input tensor, constructing and training a space topology convolutional neural network model, taking the two-channel input tensor as input characteristics, generating carrier phase multipath error prediction, designing a double-thread unscented Kalman filtering positioning framework, sequentially correcting the carrier phase observation values and resolving two-dimensional positions of users in real time according to two threads, and realizing accurate positioning under an indoor scene, thereby relieving multipath effects of an indoor pseudo-satellite system and improving indoor positioning accuracy.

Inventors

XIA YAN
Cui Chenkang
WANG BIN

Assignees

南京工业大学

Dates

Publication Date: 20260512
Application Date: 20260318

Claims (10)

1. An indoor pseudo satellite positioning method based on carrier phase multipath error modeling is characterized by comprising the following steps: Deploying pseudolites and receivers in an indoor experimental field, calibrating coordinates of the pseudolites and the receivers by using a total station, and collecting pseudolites observation data by the receivers; The method comprises the steps of obtaining carrier phase observation values of a receiver and a pseudolite based on pseudolite observation data, selecting a reference pseudolite to perform inter-satellite single difference processing to eliminate systematic errors in the carrier phase observation values, fixing whole-cycle ambiguity based on a known point initialization method, deriving carrier phase multipath errors as actual carrier phase multipath errors for training a space topology convolutional neural network model, wherein the systematic errors comprise clock differences and hardware delays of the receiver and the pseudolite; mapping the actual space layout of the pseudolite into a two-dimensional grid structure, combining the two-dimensional position of the current user and the carrier phase observation value processed by the inter-satellite single difference, extracting carrier phase residual error from the two-dimensional grid structure, acquiring carrier-to-noise ratio, and constructing a dual-channel input tensor; The method comprises the steps of constructing and training a space topology convolutional neural network model comprising a convolutional layer, a pooling layer and a full-connection layer, generating carrier phase multipath error prediction by taking a double-channel input tensor as an input characteristic, and optimizing model parameters by combining a cross-validation algorithm; The method comprises the steps of designing a double-thread unscented Kalman filtering positioning framework, for a first thread of each observation epoch, collecting pseudolite observation data in real time under an indoor scene, constructing a double-channel input tensor, inputting the double-channel input tensor into a trained space topology convolutional neural network model to generate carrier phase multipath error prediction of the observation epoch, correcting carrier phase observation values after inter-satellite single-difference processing by using the carrier phase multipath error prediction generated by the first thread, and carrying out real-time calculation on a user two-dimensional position by using the corrected carrier phase observation values based on unscented Kalman filtering to obtain user position estimation, so as to realize accurate positioning under the indoor scene.
2. The method for indoor pseudolite positioning based on carrier phase multipath error modeling of claim 1, wherein the method comprises the steps of: the pseudolites are intensively deployed in a circular area with the radius not smaller than a preset value and are connected to the same pseudolite signal transmitter to serve as a unified clock source; the receivers are sequentially arranged at the static sampling points which are uniformly distributed, and pseudolite observation data of each static sampling point is collected.
3. The method for indoor pseudolite positioning based on carrier phase multipath error modeling of claim 1, wherein the method comprises the steps of: The method for acquiring carrier phase observations of a receiver and a pseudolite based on pseudolite observation data comprises the following steps: Constructing a carrier phase observation equation to obtain carrier phase observation values of a receiver and a pseudolite: ; Wherein, the For receivers and pseudolites Carrier phase observations between; For receivers and pseudolites The geometrical distance between the two is calculated based on the coordinates of the pseudolite from the total station and the receiver; is the speed of light; And Receiver and pseudolite respectively Is a clock difference of (2); And Pseudolites respectively Hardware delay with the receiver; Is the pseudolite signal wavelength; 、、 Pseudolites respectively Whole-cycle ambiguity of (a), observation noise, carrier phase multipath error; the selecting the reference pseudolite to perform inter-planet single difference processing to eliminate systematic errors in carrier phase observed values comprises the following steps: will currently pseudolite As target pseudolites, one pseudolite is selected To reference pseudolites, the systematic error is eliminated using the inter-satellite single difference method: ; Wherein, the Representing carrier phase observed values after single difference of the target pseudolite and the reference pseudolite based on measurement data of the same receiver, namely carrier phase observed values after single difference processing among satellites; the method comprises the steps that a geometric distance value after single difference is made for a target pseudolite and a reference pseudolite based on measurement data of the same receiver, namely, the geometric distance value after single difference processing between satellites is calculated based on pseudolite coordinates from a total station; the whole-cycle ambiguity is processed by single difference between stars; Is the carrier phase multipath error after the inter-satellite single difference processing, The observation noise after the inter-satellite single difference processing is represented; The fixed integer ambiguity based on the known point initialization method comprises the following steps: the fixing of the integer ambiguity is performed using a known point initialization method, which takes the form: ; Wherein, the Is the fixed integer ambiguity; In order to round the function of the rounding, Representing pseudolite signal wavelength Is the reciprocal of (2); The deriving carrier phase multipath error includes: Using fixed integer ambiguity Carrier phase observed value after inter-satellite single difference processing Geometric distance value after single difference processing between stars Combining pseudolite signal wavelengths Deriving carrier phase multipath error : 。
4. The method for indoor pseudolite positioning based on carrier phase multipath error modeling of claim 1, wherein the method comprises the steps of: The two-dimensional grid structure is constructed based on the position of the pseudolites in the actual space, and each pixel in the grid corresponds to one pseudolite; The dual-channel input tensor comprises a carrier phase residual characteristic channel and a carrier-to-noise ratio characteristic channel; the method comprises the steps of extracting carrier phase residual errors through unscented Kalman filtering and obtaining corresponding carrier-to-noise ratios through pseudolite observation data.
5. The method for indoor pseudolite positioning based on carrier phase multipath error modeling of claim 4, wherein said extracting carrier phase residual by unscented Kalman filtering comprises: And calculating the difference between the two-dimensional position of the user and the actual inter-satellite single-difference processed carrier phase observed value based on the predicted value of the unscented Kalman filtering derived observed quantity serving as the predicted value of the inter-satellite single-difference processed carrier phase observed value by taking the two-dimensional position of the user as a state quantity and the inter-satellite single-difference processed carrier phase observed value as an observed quantity, so as to obtain the carrier phase residual error.
6. The method for indoor pseudolite positioning based on carrier phase multipath error modeling of claim 4, wherein: The carrier phase residual error does not exist in the reference pseudolite, and a carrier phase residual error characteristic channel of a pixel where the reference pseudolite is positioned is shielded by using a masking technology so as to be excluded from the training process; And eliminating the carrier phase residual characteristic channel of the central pixel of the two-dimensional grid structure by using the same mask technology.
7. The method for indoor pseudolite positioning based on carrier phase multipath error modeling of claim 1, wherein the method comprises the steps of: The spatial topological convolutional neural network model applies normalization processing to input features, relieves interference caused by scale differences of carrier phase residual errors and carrier-to-noise ratios, applies LeakyReLU activation functions to avoid gradient vanishing problems, trains by a mean square error loss function, and optimizes network parameters based on 10-fold cross validation results; and the mean square error loss function calculates the mean square error loss between the carrier phase multipath error prediction generated by the space topology convolutional neural network model and the actual carrier phase multipath error.
8. The method for indoor pseudolite positioning based on carrier phase multipath error modeling of claim 1, wherein the method comprises the steps of: the carrier phase observed value after the inter-satellite single difference correction is carried out in the following form: ; Wherein, the Representing the integer ambiguity after fixing based on the known point initialization method; Representing the corrected carrier phase observations; representing the carrier phase observed value after inter-satellite single difference processing; Is the pseudolite signal wavelength; carrier phase multipath error prediction for a first thread; The method for real-time calculation of the two-dimensional position of the user based on unscented Kalman filtering by using the corrected carrier phase observation value comprises the steps of taking the two-dimensional position of the user as a state quantity, taking the corrected carrier phase observation value as an observation quantity, and deriving an updated value of the state quantity based on unscented Kalman filtering to serve as the user position estimation.
9. An electronic device comprising at least one processor and a memory communicatively coupled to the at least one processor, wherein the memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the carrier phase multipath error modeling based indoor pseudolite positioning method of any of claims 1-8.
10. A computer readable storage medium storing computer instructions for causing a processor to implement the carrier phase multipath error modeling based indoor pseudolite positioning method of any one of claims 1-8 when executed.

Description

Indoor pseudo satellite positioning method based on carrier phase multipath error modeling Technical Field The invention relates to the technical field of indoor pseudo satellite system positioning, in particular to an indoor pseudo satellite positioning method based on carrier phase multipath error modeling. Background High-precision indoor positioning is a key direction in the navigation and position service field, and pseudolite technology is an important way for solving the problem, and signals compatible with a Global Navigation Satellite System (GNSS) are broadcast by arranging ground transmitting stations indoors, so that the coverage defect of the GNSS signals in a closed or blocked serious area (such as indoors) is effectively overcome. The technology has the core advantages that similar signals of on-orbit satellites are broadcast, the positioning principle is consistent with that of GNSS, and the technology can be combined with GNSS for positioning and can also be independently networked to provide indoor location service. And the pseudolite has theoretical potential for realizing centimeter-level high-precision positioning by utilizing the carrier phase observation value. Meanwhile, the existing commercial GNSS receiver can be compatible with pseudolite signal reception by only firmware upgrade, so that the application threshold is greatly reduced, and the pseudolite becomes an ideal choice for constructing an indoor and outdoor integrated high-precision navigation system. Pseudolites are also challenging to apply to indoor high-precision positioning, mainly due to the complex indoor signal propagation environment. Unlike open outdoor environments, indoor space structures (walls, ceilings, furniture, etc.) can significantly affect signal propagation, resulting in severe multipath effects in the pseudolite positioning system, which in turn affects the accuracy of user positioning. Compared with the outdoor, the indoor positioning is more complex in signal propagation, space layout and the like, and the pseudolite signal power is stronger, so that the multipath error is larger, the characteristics are more complex, and the multipath error is also more difficult to effectively eliminate by the traditional method. Therefore, how to accurately model and suppress severe multipath errors caused by complex spatial topologies becomes a core challenge for high-precision indoor pseudolite positioning. The existing multipath error suppression method comprises the steps of traditional data processing, antenna hardware improvement and the like, and relieves multipath effects to a certain extent, but has the limitations of strong hardware dependence, insufficient utilization of pseudolite physical layout information, poor adaptability and the like. In recent years, deep learning, especially Convolutional Neural Network (CNN), has been applied to multipath error suppression because of its strong spatial feature extraction capability, but the existing methods are mostly based on direct input of data matrix, and do not map pseudolite physical space layout directly into network structure. Therefore, a positioning method capable of explicitly modeling multipath errors of a pseudolite system based on spatial topology information is needed to improve the positioning accuracy of users. Disclosure of Invention Aiming at the defects of the prior art, the invention provides an indoor pseudo-satellite positioning method based on carrier phase multipath error modeling, which solves the problems that the accuracy is limited, the propagation error is difficult to eliminate in an indoor scene due to the influence of an indoor space structure on signal propagation in the traditional indoor positioning method, the utilization of pseudo-satellite space layout information is insufficient and the like in the traditional CNN-based indoor positioning method, and establishes a model for predicting carrier phase multipath error by explicitly fusing a pseudo-satellite space topological relation with multichannel characteristics, thereby relieving the multipath effect of an indoor pseudo-satellite system and improving the indoor positioning precision. In order to achieve the technical aim, the invention provides the technical scheme that the indoor pseudo satellite positioning method based on carrier phase multipath error modeling comprises the following steps: Deploying pseudolites and receivers in an indoor experimental field, calibrating coordinates of the pseudolites and the receivers by using a total station, and collecting pseudolites observation data by the receivers; The method comprises the steps of obtaining carrier phase observation values of a receiver and a pseudolite based on pseudolite observation data, selecting a reference pseudolite to perform inter-satellite single difference processing to eliminate systematic errors in the carrier phase observation values, fixing whole-cycle ambiguity based on a known point initializ