CN116413758A - Method for satellite positioning in urban complex environment with assistance of radio signals - Google Patents
Method for satellite positioning in urban complex environment with assistance of radio signals Download PDFInfo
- Publication number
- CN116413758A CN116413758A CN202310246306.4A CN202310246306A CN116413758A CN 116413758 A CN116413758 A CN 116413758A CN 202310246306 A CN202310246306 A CN 202310246306A CN 116413758 A CN116413758 A CN 116413758A
- Authority
- CN
- China
- Prior art keywords
- satellite
- positioning
- fingerprint
- pseudo
- value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 64
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 58
- 230000007246 mechanism Effects 0.000 claims abstract description 36
- 238000004364 calculation method Methods 0.000 claims abstract description 21
- 238000012549 training Methods 0.000 claims abstract description 10
- 230000000694 effects Effects 0.000 claims abstract description 6
- 239000011159 matrix material Substances 0.000 claims description 17
- 238000007781 pre-processing Methods 0.000 claims description 10
- 230000006870 function Effects 0.000 claims description 6
- 230000004913 activation Effects 0.000 claims description 3
- 238000012937 correction Methods 0.000 claims description 3
- 238000013178 mathematical model Methods 0.000 claims description 3
- 238000011176 pooling Methods 0.000 claims description 3
- 238000005303 weighing Methods 0.000 claims description 2
- 238000010276 construction Methods 0.000 claims 1
- 238000004590 computer program Methods 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 239000000243 solution Substances 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000010790 dilution Methods 0.000 description 1
- 239000012895 dilution Substances 0.000 description 1
- 230000008034 disappearance Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000004880 explosion Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000011800 void material Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S19/00—Satellite radio beacon positioning systems; Determining position, velocity or attitude using signals transmitted by such systems
- G01S19/38—Determining a navigation solution using signals transmitted by a satellite radio beacon positioning system
- G01S19/39—Determining a navigation solution using signals transmitted by a satellite radio beacon positioning system the satellite radio beacon positioning system transmitting time-stamped messages, e.g. GPS [Global Positioning System], GLONASS [Global Orbiting Navigation Satellite System] or GALILEO
- G01S19/42—Determining position
- G01S19/48—Determining position by combining or switching between position solutions derived from the satellite radio beacon positioning system and position solutions derived from a further system
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S5/00—Position-fixing by co-ordinating two or more direction or position line determinations; Position-fixing by co-ordinating two or more distance determinations
- G01S5/02—Position-fixing by co-ordinating two or more direction or position line determinations; Position-fixing by co-ordinating two or more distance determinations using radio waves
- G01S5/0252—Radio frequency fingerprinting
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S5/00—Position-fixing by co-ordinating two or more direction or position line determinations; Position-fixing by co-ordinating two or more distance determinations
- G01S5/02—Position-fixing by co-ordinating two or more direction or position line determinations; Position-fixing by co-ordinating two or more distance determinations using radio waves
- G01S5/0257—Hybrid positioning
- G01S5/0268—Hybrid positioning by deriving positions from different combinations of signals or of estimated positions in a single positioning system
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/048—Activation functions
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/02—Services making use of location information
- H04W4/029—Location-based management or tracking services
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W84/00—Network topologies
- H04W84/02—Hierarchically pre-organised networks, e.g. paging networks, cellular networks, WLAN [Wireless Local Area Network] or WLL [Wireless Local Loop]
- H04W84/10—Small scale networks; Flat hierarchical networks
- H04W84/12—WLAN [Wireless Local Area Networks]
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Biophysics (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Signal Processing (AREA)
- Position Fixing By Use Of Radio Waves (AREA)
Abstract
The invention provides a method for assisting satellite positioning in urban complex environments by radio signals, which comprises the following steps: step 1, wiFi fingerprint positioning of a convolutional neural network CNN model based on an attention mechanism includes: constructing a convolutional neural network CNN model based on an attention mechanism, and performing offline training; a trained convolutional neural network CNN model based on an attention mechanism is used for solving a positioning result on line to obtain a WiFi fingerprint positioning result; step 2, a wifi fingerprint positioning assists a global navigation satellite system GNSS positioning, including: inverting the pseudo range according to the position obtained by the WiFi fingerprint positioning result; detecting non-line-of-sight received NLOS signal effects; enabling a selected satellite in GNSS positioning to meet a geometric precision factor GDOP; and (3) performing position calculation by using a weighted least square method to finish positioning, namely finishing positioning of the satellite in the urban complex environment by using the radio signal.
Description
Technical Field
The invention relates to a satellite positioning method, in particular to a method for satellite positioning in a complex urban environment assisted by radio signals.
Background
With the popularity of smartphones, radio signals such as WiFi, UWB, and bluetooth are used for positioning. In addition, the WiFi positioning technology has the advantages of low hardware equipment cost, low calculation cost, relatively high positioning precision and the like, and is widely applied to indoor positioning. WiFi positioning methods can be classified into a triangulation positioning method and a fingerprint positioning method, wherein the triangulation positioning method and the fingerprint positioning method are used for carrying out position calculation by matching the acquired RSS data with fingerprints in a fingerprint database. Because of the influence of non-line-of-sight signals and multipath effects, it is difficult to map RSS to distance, so more research is being conducted based on fingerprint database positioning.
Under urban environment, satellite signals are easy to be blocked, and if positioning is performed by using GNSS only, the number of available satellites is affected, so that positioning accuracy is reduced. In addition, the WiFi signal is less affected by NLOS, but is mostly used for indoor positioning, and in the method for detecting and eliminating NLOS, the detection of NLOS assisted satellite positioning by using the WiFi signal is less studied.
When the traditional WiFi fingerprint positioning method is used, the matching algorithm is complex, and a large amount of calculation time and resources are consumed, so that the problem of mismatching often occurs. The existing multi-purpose machine learning method optimizes the WiFi fingerprint matching process, because the data magnitude of a fingerprint database is not large, if the complexity of the adopted model is higher, the data is easy to be over-fitted, the calculation load is increased, and the extraction of fingerprint data characteristics is not facilitated.
Disclosure of Invention
The invention aims to: the invention aims to solve the technical problem of providing a method for assisting satellite positioning in a complex urban environment by using radio signals.
In order to solve the technical problems, the invention discloses a method for assisting satellite positioning in a complex urban environment by using radio signals, which comprises the following steps:
step 1-1, constructing a convolutional neural network CNN model based on an attention mechanism, and performing offline training, wherein the method comprises the following specific steps of:
step 1-1-1, preprocessing data, wherein the specific method is as follows:
the received signal strength RSS is preprocessed, i.e. normalized as follows:
wherein RSS' is data obtained after preprocessing, min (RSS) is the minimum value of the acquired WiFi signal intensity, and max (RSS) is the maximum value of the WiFi signal intensity;
and converting fingerprint data of the WiFi signals into a square matrix, wherein the square matrix is provided with fingerprint points of z observation samples, integrating the data of the z fingerprint points into a matrix, and filling the vacant part of the matrix into 0.
Step 1-1-2, constructing a convolutional neural network CNN model based on an attention mechanism, wherein the method comprises the following steps of:
the convolutional neural network CNN model structure based on the attention mechanism consists of two convolutional layers with the size of 3 multiplied by 3, two pooling layers with the size of 2 multiplied by 2, an attention layer and a full connection layer; adopting a CBAM attention mechanism, wherein the CBAM attention mechanism combines a channel and a spatial attention mechanism module and is integrated into a convolutional neural network CNN structure;
the activation function uses a ReLU function; adding a dropout layer before the full connection layer; setting a dropout parameter to be 0.5, and selecting a learning rate to be 0.0001; the number of convolution kernels of the first convolution layer is set to be 100, and the number of convolution kernels used by the second convolution layer is set to be 300;
performing offline training on the convolutional neural network CNN model based on the attention mechanism by using a pre-collected data set to obtain a fixed characteristic weight corresponding to the data of each fingerprint point; and storing the coordinates of the fingerprint points and the corresponding fixed characteristic weights in a WiFi fingerprint database.
Step 1-2, a trained convolutional neural network CNN model based on an attention mechanism is used for solving a positioning result on line to obtain a WiFi fingerprint positioning result, and the specific method is as follows:
preprocessing the detected WiFi signal intensity data according to the method in the step 1-1-1 to obtain input data X, and inputting the input data X into a pre-trained convolutional neural network CNN model based on an attention mechanism to obtain a characteristic weight; calculating the probability P of the observation position, i.e. the user position, at the jth fingerprint point j The method comprises the following steps:
wherein sigma x Is the variance of the input data X, a x Is the variance parameter of the input data X, Y j Is the output data at fingerprint point j, Y v Output data at fingerprint points v, wherein z is the number of fingerprint points, and v is an index variable used for traversing all fingerprint points;
selecting the first M fingerprint points with the highest probability as adjacent points, calculating the position coordinates S of the user through weighted average, and calculating the weight G of the kth adjacent point k The calculation method comprises the following steps:
wherein P is k Is the probability that the user position is at the kth fingerprint point, P q Is the probability of the user position on the q-th fingerprint point, q is the index variable used for traversing all adjacent points, S is the coordinate of the observed position, namely the positioning result obtained by solving, S k Is the coordinates of the selected kth fingerprint point.
Step 2, the GNSS positioning of the global navigation satellite system is assisted by WiFi fingerprint positioning;
step 2-1, inverting pseudo range according to the position obtained by the WiFi fingerprint positioning result, wherein the specific method comprises the following steps:
recording a user position coordinate (X) obtained by WiFi fingerprint positioning based on an attention mechanism CNN model r ,Y r ,Z r ) The method comprises the steps of carrying out a first treatment on the surface of the Assuming that the received satellite signal is a line-of-sight signal, inverting the pseudo range ρ to the ith receivable satellite si The method comprises the following steps:
wherein, (X si, Y si ,Z si ) I epsilon (1, 2, …, n) is the coordinates of the ith satellite, where the number of received satellite signals is n, τ r For satellite signal receiver clock difference, τ si The clock error of the ith satellite, c is the speed of light in vacuum, V ion Is ionospheric delay, V trop Is the tropospheric delay.
Step 2-2, detecting non line of sight received NLOS signal effects, comprising the steps of:
step 2-2-1, the pseudo-range observation value and the inversion value are subjected to difference, and the specific method comprises the following steps:
in the user position, the visible n satellite pseudo-range observed values and the inversion pseudo-range value are differenced, and the pseudo-range observed value of the ith satelliteAnd inverted pseudo-range value ρ si Taking the absolute value of the difference to obtain the absolute value C of the difference between the pseudo-range observation value and the inversion value of the ith satellite si ;
Step 2-2-2, selecting satellites participating in position resolution by judging that non-line-of-sight received NLOS signals affect the satellites, wherein the specific method comprises the following steps:
setting a threshold T 1 If the absolute value C of the difference between the pseudo-range observation value and the inversion value of the ith satellite si ≤T 1 Selecting the satellite to participate in position calculation; all the visible satellites in the user position are compared and stored in a satellite data set.
Step 2-3, the selected satellite in the GNSS positioning of the global navigation satellite system meets the geometric precision factor GDOP, and the specific method comprises the following steps:
calculating the current geometric precision factor GDOP value according to a pseudo-range observation equation of the selected satellite; setting a threshold T 2 When GDOP < T 2 When the satellite is selected, the satellite is used for position calculation; when GDOP is greater than or equal to T 2 When the T is adjusted upwards 1 And (3) repeating the step 2-2-2 until the geometric precision factor GDOP meets the condition.
And 2-4, performing position calculation by using a weighted least square method to finish positioning, namely finishing positioning of the satellite in the urban complex environment by using a radio signal, wherein the specific method comprises the following steps of:
weight coefficient omega of ith satellite si Expressed as:
wherein sigma 1 Is the mean square error associated with the satellite and is solved by the user ranging factor N.
The weighted least squares mathematical model of satellite pseudo-range single point positioning is:
WGΔX=Wb
where W is a weight matrix, w=diag (ω s1 ,ω s2 ,…,ω sf ) F is the number of satellites selected for storage into the satellite data set; g is a direction cosine matrix of the satellite, deltaX is a calculated three-dimensional position correction value and receiver clock error, and b consists of pseudo-range differences between each satellite pseudo-range observation value and a calculated receiver position inverse; solving equation Δx= (G) T W T WG) - 1 G T W T Wb, repeatedly and iteratively calculating the coordinate position of the receiver to finish positioning.
The beneficial effects are that:
aiming at the problems that GNSS positioning is easily affected by NLOS in urban environment and pseudo-range measurement error is large. The invention provides a method for alleviating the influence of NLOS in GNSS positioning calculation by using WiFi signals, and inverting pseudo ranges of receivable satellites by WiFi fingerprint database positioning results. In addition, the inversion pseudo-range value and the pseudo-range observation value of the GNSS are subjected to difference, satellites which are less influenced by NLOS are compared and selected, and the selected satellites are judged to meet GDOP requirements. And (3) performing position calculation on the difference magnitude weight-reducing NLOS signal of the reference pseudo-range inversion value and the observed value by using a weighted least square method.
The traditional WiFi fingerprint database positioning method often has the problems of mismatching and the like, so that the positioning result is unreliable. To improve the accuracy of the subsequent inverted pseudoranges, an accurate user position should be obtained. The machine learning theory is introduced into the WiFi fingerprint matching problem, the CNN model based on the attention mechanism is adopted to train WiFi fingerprint data, the attention mechanism is added into the CNN network to facilitate parameter training, and important fingerprint information is focused by giving high weight. When more data is input, the calculation amount can be reduced, the running speed can be increased, and the performance of the CNN model is improved.
Drawings
The foregoing and/or other advantages of the invention will become more apparent from the following detailed description of the invention when taken in conjunction with the accompanying drawings and detailed description.
FIG. 1 is a schematic flow chart of the present invention.
Fig. 2 is a schematic diagram of a convolutional neural network CNN model of an attention mechanism according to the present invention.
Detailed Description
The core content of the invention is a method for inverting pseudo-range based on WiFi fingerprint positioning result, assisting GNSS (Global navigation satellite System, global Navigation Satellite System, GNSS) to detect NLOS (non line-of-sight reception, non line of sight, NLOS) influence and weighting and resolving position. Comprising the following 2 stages: (1) The coordinate position of the user is obtained through WiFi fingerprint positioning of a CNN (convolutional neural network, convolutional Neural Network, CNN) model based on an attention mechanism. (2) And inverting the pseudo range at the positioning position, detecting NLOS influence, and then carrying out position calculation by using a weighted least square algorithm to reduce the weight NLOS signal. The general flow is as shown in fig. 1:
(1) WiFi fingerprint positioning method of CNN model based on attention mechanism
1) Offline training of network model
The first step: data preprocessing
To facilitate subsequent data processing, RSS (received signal strength, received Signal Strength, RSS) is standardized:
wherein RSS' is data obtained after preprocessing, min (RSS) is the minimum value of the acquired WiFi signal intensity, and max (RSS) is the maximum value of the WiFi signal intensity.
The fingerprint data should also be converted into a square matrix before model training. Assuming that there are z observation sample fingerprint points, the data of the z fingerprint points can be integrated into one matrix. To meet the matrix row and column agreement, the void should also be filled with 0.
And a second step of: constructing a CNN model based on an attention mechanism
The CNN model structure based on the attention mechanism is shown in fig. 2, and consists of two convolution layers with the size of 3×3, two pooling layers with the size of 2×2, one attention layer and one full connection layer. The invention adopts a CBAM attention mechanism (refer to Cbam: convolutional block attention module [ C ]// Proceedings of the European conference on computer vision (ECCV). 2018:3-19.), combines the attention mechanism modules of the channel and the space, can be seamlessly integrated into a CNN structure as a lightweight universal module, and improves the performance of a CNN model.
The ReLU function is used as an activation function, so that the problems of gradient explosion and gradient disappearance in the gradient descent and counter-propagation process can be effectively solved. To prevent the overfitting phenomenon, a dropout layer is added before the full connection layer. The dropout parameter is set to 0.5 and the learning rate is chosen to be 0.0001. The number of convolution kernels for the first convolution layer is set to 100 and the number of convolution kernels for the second convolution layer is set to 300.
After training is finished, a fixed characteristic weight corresponding to each fingerprint point data can be obtained and stored in a database.
2) Solving the positioning result on line
In the online stage, after preprocessing the observed WiFi signal intensity data, inputting the data into a pre-trained network, and acquiring and outputting the characteristic weight. Calculating the probability of the observation position at the j-th fingerprint point:
wherein sigma x Is the variance of the input data X, a x Is the variance parameter of the input data X, Y j Is the output data at fingerprint point j, Y v Is the output data at fingerprint points v, z being the number of fingerprint points, v being the index variable used to traverse all fingerprint points.
Selecting the first M fingerprint points with the highest probability as adjacent points, calculating the position coordinates S of the user by weighted average, and weighing G of the kth adjacent points k The calculation method comprises the following steps:
wherein P is k Is the probability that the user position is at the kth fingerprint point, P q Is the probability of the user position on the q-th fingerprint point, q is the index variable used for traversing all adjacent points, S is the coordinate of the observed position, namely the positioning result obtained by solving, S k Is the coordinates of the selected kth fingerprint point.
(2) WiFi fingerprint positioning assisted GNSS positioning
1) WiFi fingerprint positioning position inversion pseudo-range
The user position coordinate for recording the positioning of the WiFi fingerprint database is (X) r ,Y r ,Z r ). Assuming that the received satellite signal is an LOS (Line-of-sight) signal, the pseudorange ρ to the ith receivable satellite is inverted si The method comprises the following steps:
wherein (X) si ,Y si ,Z si ) I epsilon (1, 2, …, n) is the coordinates of the ith satellite, the number of satellite signals receivable at this location being n, τ r For receiver clock skew τ si The clock error of the ith satellite, c is the speed of light in vacuum, V ion Is ionospheric delay, V trop Is the tropospheric delay.
2) Detecting NLOS signal effects
The first step: pseudo-range observation value and inversion value difference
In the user position, the visible n satellite pseudo-range observations and the inverted pseudo-range value are differenced, such as the pseudo-range observation of the ith satelliteAnd inverted pseudo-range value ρ si Taking the absolute value of the difference to obtain C si 。
And a second step of: selecting satellites for participation in position resolution by determining NLOS effects
In urban environments, the difference between the pseudorange observations and the inversion values may, to some extent, laterally reflect the extent to which the satellite signals are affected by NLOS. Setting a threshold T 1 If C of the satellite si ≤T 1 The satellite is selected. All of the n satellites in view of the user's location are compared and the selected satellites are stored in the data set. T in the invention 1 Set to 15.
3) To make the selected satellites meet the GDOP (geometric precision factor, geometric Dilution of Precision, GDOP) requirement
From the pseudorange observations equation for the selected satellite, the current GDOP value is calculated (reference: star selecting method [ P ] of multimode GNSS receiver]Beijing: CN103954980a, 2014-07-30.), a satellite that is less affected by NLOS is selected for position resolution by threshold T1, but may degrade satellite geometry. To ensure positioning accuracy, a threshold T is set 2 When GDOP < T 2 In that case, the selected satellite may be used for position resolution; when GDOP is greater than or equal to T 2 When the T is adjusted upwards 1 And (3) repeating the satellite selection process participating in the position calculation until the calculated GDOP meets the condition. The invention sets the threshold T 2 6.
4) Position calculation by weighted least square method
The weight coefficient of the ith satellite can be expressed as:
wherein C is si Is the absolute value of the difference between the pseudo-range value observed by the ith satellite and the inverted simulated pseudo-range value, sigma 1 Is the mean square error associated with the satellite and can be solved by the user ranging factor N.
The weighted least squares mathematical model of satellite pseudo-range single point positioning is:
WGΔX=Wb
where W is a weight matrix, w=diag (ω s1 ,ω s2 ,…,ω sf ) F is the number of satellites selected for storage into the data set; g is a direction cosine matrix of the satellite; Δx is the calculated three-dimensional position correction value and receiver clock error; b consists of the difference between each satellite pseudorange observation and the computed pseudorange inverse of the receiver position. Solving equation Δx= (G) T W T WG) -1 G T W T Wb, and then calculating the coordinate position of the receiver through repeated iteration.
In a specific implementation, the application provides a computer storage medium and a corresponding data processing unit, wherein the computer storage medium can store a computer program, and the computer program can run the invention content and part or all of the steps in each embodiment of a method for assisting urban complex environment satellite positioning by using radio signals provided by the invention when being executed by the data processing unit. The storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM), a random-access memory (random access memory, RAM), or the like.
It will be apparent to those skilled in the art that the technical solutions in the embodiments of the present invention may be implemented by means of a computer program and its corresponding general hardware platform. Based on such understanding, the technical solutions in the embodiments of the present invention may be embodied essentially or in the form of a computer program, i.e. a software product, which may be stored in a storage medium, and include several instructions to cause a device (which may be a personal computer, a server, a single-chip microcomputer, MUU or a network device, etc.) including a data processing unit to perform the methods described in the embodiments or some parts of the embodiments of the present invention.
The invention provides a thought and a method for assisting a satellite positioning method in a complex urban environment by a radio signal, and a method and a way for realizing the technical scheme are numerous, the above description is only a preferred embodiment of the invention, and it should be noted that a plurality of improvements and modifications can be made to those skilled in the art without departing from the principle of the invention, and the improvements and modifications are also considered as the protection scope of the invention. The components not explicitly described in this embodiment can be implemented by using the prior art.
Claims (10)
1. A method for radio signal assisted satellite positioning in a complex urban environment, comprising the steps of:
step 1, wiFi fingerprint positioning of a convolutional neural network CNN model based on an attention mechanism;
step 1-1, constructing a convolutional neural network CNN model based on an attention mechanism, and performing offline training;
step 1-2, a trained convolutional neural network CNN model based on an attention mechanism is used for solving a positioning result on line, and a WiFi fingerprint positioning result is obtained;
step 2, the GNSS positioning of the global navigation satellite system is assisted by WiFi fingerprint positioning;
step 2-1, inverting pseudo-range according to the position obtained by the WiFi fingerprint positioning result;
step 2-2, detecting influence of non-line-of-sight received NLOS signals;
step 2-3, enabling the selected satellites in the GNSS positioning of the global navigation satellite system to meet the geometric precision factor GDOP;
and 2-4, performing position calculation by using a weighted least square method to finish positioning, namely finishing positioning of the satellite in the urban complex environment assisted by the radio signal.
2. The method for assisting satellite positioning in urban complex environment by using radio signals according to claim 1, wherein the constructing of the convolutional neural network CNN model based on the attention mechanism in the step 1-1 and the offline training are performed comprises the following specific steps:
step 1-1-1, preprocessing data;
and step 1-1-2, constructing a convolutional neural network CNN model based on an attention mechanism.
3. A method for assisting satellite positioning in urban complex environment according to claim 2, characterized in that the data preprocessing in step 1-1-1 is as follows:
the received signal strength RSS is preprocessed, i.e. normalized as follows:
wherein RSS' is data obtained after preprocessing, min (RSS) is the minimum value of the acquired WiFi signal intensity, and max (RSS) is the maximum value of the WiFi signal intensity;
and converting fingerprint data of the WiFi signals into a square matrix, wherein the square matrix is provided with fingerprint points of z observation samples, integrating the data of the z fingerprint points into a matrix, and filling the vacant part of the matrix into 0.
4. A method for assisting satellite positioning in urban complex environment by using radio signals according to claim 3, wherein the construction of the convolutional neural network CNN model based on the attention mechanism in the step 1-1-2 is as follows:
the convolutional neural network CNN model structure based on the attention mechanism consists of two convolutional layers with the size of 3 multiplied by 3, two pooling layers with the size of 2 multiplied by 2, an attention layer and a full connection layer; adopting a CBAM attention mechanism, wherein the CBAM attention mechanism combines a channel and a spatial attention mechanism module and is integrated into a convolutional neural network CNN structure;
the activation function uses a ReLU function; adding a dropout layer before the full connection layer; setting a dropout parameter to be 0.5, and selecting a learning rate to be 0.0001; the number of convolution kernels of the first convolution layer is set to be 100, and the number of convolution kernels used by the second convolution layer is set to be 300;
performing offline training on the convolutional neural network CNN model based on the attention mechanism by using a pre-collected data set to obtain a fixed characteristic weight corresponding to the data of each fingerprint point; and storing the coordinates of the fingerprint points and the corresponding fixed characteristic weights in a WiFi fingerprint database.
5. The method for satellite positioning in urban complex environment assisted by radio signals according to claim 4, wherein the on-line solving of the positioning result in step 1-2 is as follows:
preprocessing the detected WiFi signal intensity data according to the method in the step 1-1-1 to obtain input data X, and inputting the input data X into a pre-trained convolutional neural network CNN model based on an attention mechanism to obtain a characteristic weight; calculating the probability P of the observation position, i.e. the user position, at the jth fingerprint point j The method comprises the following steps:
wherein sigma x Is the variance of the input data X, a x Is the variance parameter of the input data X, Y j Is the output data at fingerprint point j, Y v Output data at fingerprint points v, wherein z is the number of fingerprint points, and v is an index variable used for traversing all fingerprint points;
selecting the first M fingerprint points with the highest probability as adjacent points, calculating the position coordinates S of the user by weighted average, and weighing G of the kth adjacent points k The calculation method comprises the following steps:
wherein P is k Is the probability that the user position is at the kth fingerprint point, P q Is the probability of the user position on the q-th fingerprint point, q is the index variable used for traversing all adjacent points, S is the coordinate of the observed position, namely the positioning result obtained by solving, S k Is the coordinates of the selected kth fingerprint point.
6. The method for assisting satellite positioning in urban complex environment according to claim 5, wherein the position inversion pseudo-range obtained according to the WiFi fingerprint positioning result in step 2-1 comprises:
recording a user position coordinate (X) obtained by WiFi fingerprint positioning based on an attention mechanism CNN model r ,Y r ,Z r ) The method comprises the steps of carrying out a first treatment on the surface of the Assuming that the received satellite signal is a line-of-sight signal, inverting the pseudo range ρ to the ith receivable satellite si The method comprises the following steps:
wherein, (X si ,Y si ,Z si ) I epsilon (1, 2, …, n) is the coordinates of the ith satellite, where the number of received satellite signals is n, τ r For satellite signal receiver clock difference, τ si The clock error of the ith satellite, c is the speed of light in vacuum, V ion Is ionospheric delay, V trop Is the tropospheric delay.
7. A method for radio signal assisted urban complex environment satellite positioning according to claim 6, wherein said detecting non-line-of-sight received NLOS signal effects in step 2-2 comprises the steps of:
step 2-2-1, performing difference between the pseudo-range observation value and the inversion value;
and 2-2, selecting satellites participating in position resolution by judging that the non-line-of-sight receiving NLOS signals influence.
8. The method for assisting satellite positioning in urban complex environment according to claim 7, wherein the pseudo-range observation value and the inversion value in step 2-2-1 are differenced, and the specific method comprises:
in the user position, the visible n satellite pseudo-range observed values and the inversion pseudo-range value are differenced, and the pseudo-range observed value of the ith satelliteAnd inverted pseudo-range value ρ si Taking the absolute value of the difference to obtain the absolute value C of the difference between the pseudo-range observation value and the inversion value of the ith satellite si ;
The specific method for selecting satellites participating in position resolution by judging that the non-line-of-sight received NLOS signals influence the satellites in position resolution in the step 2-2-2 comprises the following steps:
setting a threshold T 1 If the difference between the pseudo-range observation value and the inversion value of the ith satelliteAbsolute value C of (2) si ≤T 1 Selecting the satellite to participate in position calculation; all the visible satellites in the user position are compared and stored in a satellite data set.
9. The method as claimed in claim 8, wherein the step 2-3 of enabling the selected satellites in the GNSS positioning to meet the geometric accuracy factor GDOP comprises:
calculating the current geometric precision factor GDOP value according to a pseudo-range observation equation of the selected satellite; setting a threshold T 2 When GDOP<T 2 When the satellite is selected, the satellite is used for position calculation; when GDOP is greater than or equal to T 2 When the T is adjusted upwards 1 And (3) repeating the step 2-2-2 until the geometric precision factor GDOP meets the condition.
10. The method for assisting satellite positioning in urban complex environment according to claim 9, wherein the position calculation in step 2-4 is performed by using weighted least square method, and the specific method is as follows:
weight coefficient omega of ith satellite si Expressed as:
wherein sigma 1 The mean square error related to the satellite is solved through a user ranging factor N;
the weighted least squares mathematical model of satellite pseudo-range single point positioning is:
WGΔX=Wb
where W is a weight matrix, w=diag (ω s1 ,ω s2 ,…,ω sf ) F is the selected memoryA number of satellites stored to the satellite dataset; g is a direction cosine matrix of the satellite, deltaX is a calculated three-dimensional position correction value and receiver clock error, and b consists of pseudo-range differences between each satellite pseudo-range observation value and a calculated receiver position inverse; solving equation Δx= (G) T W T WG) -1 G T W T Wb, repeatedly and iteratively calculating the coordinate position of the receiver to finish positioning.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310246306.4A CN116413758A (en) | 2023-03-14 | 2023-03-14 | Method for satellite positioning in urban complex environment with assistance of radio signals |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310246306.4A CN116413758A (en) | 2023-03-14 | 2023-03-14 | Method for satellite positioning in urban complex environment with assistance of radio signals |
Publications (1)
Publication Number | Publication Date |
---|---|
CN116413758A true CN116413758A (en) | 2023-07-11 |
Family
ID=87050696
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310246306.4A Pending CN116413758A (en) | 2023-03-14 | 2023-03-14 | Method for satellite positioning in urban complex environment with assistance of radio signals |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116413758A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117111126A (en) * | 2023-10-20 | 2023-11-24 | 成都格理特电子技术有限公司 | Beidou-based petrochemical worker joint positioning method and device |
-
2023
- 2023-03-14 CN CN202310246306.4A patent/CN116413758A/en active Pending
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117111126A (en) * | 2023-10-20 | 2023-11-24 | 成都格理特电子技术有限公司 | Beidou-based petrochemical worker joint positioning method and device |
CN117111126B (en) * | 2023-10-20 | 2023-12-22 | 成都格理特电子技术有限公司 | Beidou-based petrochemical worker joint positioning method and device |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105371870B (en) | A kind of in-orbit accuracy measurement method of star sensor based on star chart data | |
CN110275192B (en) | High-precision single-point positioning method and device based on smart phone | |
CN108919321B (en) | A kind of GNSS positioning Detection of Gross Errors method based on trial and error method | |
CN103363987B (en) | The method for recognising star map of a kind of many visual fields star sensor | |
CN112230247B (en) | GNSS integrity monitoring method used in urban complex environment | |
CN116413758A (en) | Method for satellite positioning in urban complex environment with assistance of radio signals | |
Sun et al. | Stacking ensemble learning for non-line-of-sight detection of global navigation satellite system | |
Sun et al. | Using dual‐polarization GPS antenna with optimized adaptive neuro‐fuzzy inference system to improve single point positioning accuracy in urban canyons | |
CN105388496B (en) | Traffic application vulnerability checking system and method based on GPS | |
CN108549062B (en) | System platform for sea surface search radar and multi-model target tracking method | |
CN111856513A (en) | Satellite observation value acquisition method and device, computer equipment and storage medium | |
CN105631191A (en) | Method and system for obtaining satellite performance | |
CN115079225A (en) | Navigation positioning method and device of marine receiver | |
Sun et al. | An integrated urban positioning algorithm using matching, particle swam optimized adaptive neuro fuzzy inference system and a spatial city model | |
CN101806905A (en) | Navigation positioning method and device for agricultural machines | |
CN110646822A (en) | Integer ambiguity Kalman filtering algorithm based on inertial navigation assistance | |
CN115616643A (en) | City area modeling auxiliary positioning method | |
CN113109843B (en) | Deception signal detection suppression method and device based on double-receiver pseudo-range double-difference | |
WO2024152673A1 (en) | Mobile terminal positioning method and apparatus, and device and storage medium | |
WO2023236643A1 (en) | Positioning method and apparatus, device and storage medium | |
CN111965674B (en) | Beidou positioning and resolving method and system based on self-adaptive cuckoo algorithm | |
CN113466909A (en) | GNSS multi-frequency system partial integer ambiguity subset selection method | |
CN115616637B (en) | Urban complex environment navigation positioning method based on three-dimensional grid multipath modeling | |
CN116859415A (en) | Quick, stable and high-precision multi-fault satellite identification and positioning method | |
CN111708055A (en) | Navigation deception signal detection method and device |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |