CN112153569A - Indoor fingerprint positioning-based optimization method - Google Patents
Indoor fingerprint positioning-based optimization method Download PDFInfo
- Publication number
- CN112153569A CN112153569A CN202010999359.XA CN202010999359A CN112153569A CN 112153569 A CN112153569 A CN 112153569A CN 202010999359 A CN202010999359 A CN 202010999359A CN 112153569 A CN112153569 A CN 112153569A
- Authority
- CN
- China
- Prior art keywords
- distance
- point
- positioning
- algorithm
- points
- 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
- 238000005457 optimization Methods 0.000 title claims abstract description 25
- 238000000034 method Methods 0.000 title claims abstract description 23
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 48
- 238000001914 filtration Methods 0.000 claims abstract description 18
- 239000002245 particle Substances 0.000 claims abstract description 13
- 238000007781 pre-processing Methods 0.000 claims abstract description 12
- 238000012360 testing method Methods 0.000 claims description 9
- 230000001133 acceleration Effects 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000001514 detection method Methods 0.000 claims description 3
- 239000011159 matrix material Substances 0.000 claims description 3
- 230000007704 transition Effects 0.000 claims description 3
- 238000007635 classification algorithm Methods 0.000 abstract description 3
- 238000005259 measurement Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 230000004807 localization Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- 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
- H04W4/021—Services related to particular areas, e.g. point of interest [POI] services, venue services or geofences
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/30—Services specially adapted for particular environments, situations or purposes
- H04W4/33—Services specially adapted for particular environments, situations or purposes for indoor environments, e.g. buildings
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W64/00—Locating users or terminals or network equipment for network management purposes, e.g. mobility management
Abstract
The invention relates to an optimization method based on indoor fingerprint positioning, which comprises the following steps: s1, discretely gridding the positioning area, deploying N AP points and determining a proper grid point distance; s2, selecting M reference points according to the distance, collecting RSSI between each reference point and an AP point to form a data set X, and preprocessing; s3, dividing reference point area IDs and constructing a fingerprint database; s4, selecting a real path, and acquiring RSSI between T undetermined positioning points and AP points on the real path; s5, obtaining the ID of the region to be positioned according to the data set X and the reference point region ID by adopting an SVM algorithm; s6, calculating the RSSI similarity between the to-be-positioned point after the region classification and the reference point in the region by utilizing the Euclidean distance, the Manhattan distance and the Chebyshev distance to obtain position estimation; and S7, taking the position estimation as an observation value, and combining a PDR algorithm to carry out particle filtering to obtain accurate positioning coordinates. The invention reduces the size of the search space by using the region classification algorithm to improve the efficiency, and simultaneously optimizes the matching algorithm to improve the positioning precision.
Description
Technical Field
The invention relates to the technical field of indoor positioning, in particular to an indoor fingerprint positioning-based optimization method.
Background
Providing real-time, accurate location information is the ultimate goal of Location Based Services (LBS). The complete satellite signals cannot be received in a closed indoor environment, and the deployment in the environment is complex and variable, so that the high-precision positioning is difficult to provide, and the research on the indoor positioning technology is essential.
In the prior art, the fingerprint positioning algorithm based on the RSSI and the pedestrian position deduction algorithm (PDR) are combined to realize positioning, but the RSSI is easily influenced by environmental factors, and the positioning accuracy is not high due to instability of the RSSI. In addition, a dynamic area division mechanism for dividing an area into sub-areas through dynamic linear boundaries between access point AP pairs is adopted, but the research cannot guarantee that boundary lines exist certainly, and the practicability of area division is seriously influenced.
Disclosure of Invention
The invention aims to provide an optimization method based on indoor fingerprint positioning, which reduces the size of a search space through a region classification algorithm, and simultaneously calculates the similarity by combining the Euclidean distance, the Manhattan distance and the Chebyshev distance to obtain position estimation, thereby improving the efficiency of the optimization algorithm.
In order to achieve the above purpose, the embodiments of the present invention provide the following technical solutions: an optimization method based on indoor fingerprint positioning comprises the following steps:
s1, discretely gridding the positioning area, deploying N AP points and determining a proper grid point distance;
s2, selecting M reference points according to the distance, collecting RSSI between each reference point and AP point to form a data set X with the dimension of (M.C) xN, and preprocessing, wherein C is the measuring frequency of each reference point;
s3, dividing reference point area IDs and constructing a fingerprint database;
s4, selecting a real path, and acquiring RSSI between T undetermined positioning points and AP points on the real path;
s5, obtaining the area ID of the point to be positioned according to the data set X and the reference point area ID by adopting an SVM algorithm, and traversing only the corresponding reference points in the area during matching to achieve the purposes of reducing the search space and improving the efficiency;
s6, calculating the RSSI similarity between the to-be-positioned point after the region classification and the reference point in the region by utilizing the Euclidean distance, the Manhattan distance and the Chebyshev distance to obtain position estimation;
and S7, taking the position estimation as an observation value, and combining a PDR algorithm to carry out particle filtering to obtain accurate positioning coordinates.
Further, in the step S2, the RSSI between each reference point and the AP point is preprocessed, and then the step S3 is performed, where the preprocessing is to remove the outlier DATA group in the original DATA set X, and then the basic fingerprint DATA set Y is obtained by performing the limiting average filtering and then the moving average filtering, and the basic fingerprint DATA set Y is combined with the reference point area ID to form the multi-feature fingerprint database DATA ═ Y ID.
Further, in the steps S4 and S5, the RSSI of T to-be-located points on the real path is collected in real time to form a data set X with a dimension of (T · C) × NtestPreprocessing the data to obtain a data set YtestSubstituting the classification decision function to obtain an area vector id with dimension T multiplied by 1, and realizing area classification:
wherein:
wi,wjfeatures representing reference points in a fingerprint database,
vi,vjan area ID indicating a reference point of the image,
αi,αjthe representation being a Lagrange multiplierThe solution of the method to the solution of the SVM fundamental,
K(wi,wj)＝wi Twjrepresenting a kernel function.
Further, in the step S6, K reference points with the highest similarity are selected, and the position estimate is calculated by taking the inverse of RSSI as a weight:
where p represents a variable parameter of the minkowski distance, which is the manhattan distance when p is 1; when p is 2, it is the euclidean distance; when p → ∞, this is the chebyshev distance; depending on the variation parameter, the Min distance may represent a class of distances:
wherein (x ', y ') represents a position estimate, (x 'k,y′k) Representing the coordinates of K reference points with the highest similarity, and respectively obtaining position estimation obtained by calculating the similarity through Euclidean distance, Manhattan distance and Chebyshev distance according to the value of p; the mean of the three sets of position estimates is taken as the position estimate for the improved algorithm:
where (x '", y'") indicates the position estimate of the improved algorithm, (x "")e,y″e),(x″m,y″m),(x″c,y″c) And respectively representing the position estimation of the calculation similarity of the Euclidean distance, the Manhattan distance and the Chebyshev distance.
Further, when the RSSI of the point to be positioned is collected, two Bluetooth gyroscope sensors are respectively bound to a left foot and a right foot, and the three-axis acceleration, the three-axis angular velocity and the three-axis angle when the user walks along a real path are recorded according to the frequency f; calculating the number of wave crests and wave troughs by a step frequency detection algorithm to obtain step number, thereby calculating the step length d of each step; and then, obtaining a heading angle theta of each step of walking by using a quaternion method, in the step S8, constructing a state transition matrix by using a PDR algorithm, improving the position estimation obtained by the algorithm to be used as an observed value, carrying out particle filtering together with the step length d and the heading angle theta, and outputting a final positioning coordinate:
wherein (x)0,y0) Coordinates representing the initial point, (x)i,yi) Coordinates representing the i-th step, diAnd thetaiThe step size and heading angle of the ith step are indicated.
Compared with the prior art, the invention has the beneficial effects that:
1. in the preprocessing stage, the RSSI redundant data caused by the mutant interference and the periodic interference is filtered by utilizing the advantages of amplitude limiting average filtering and moving average filtering, so that the influence caused by the instability of the RSSI data is reduced in an optimization algorithm based on indoor fingerprint positioning.
2. The search space size is reduced by using the region classification algorithm, and the test points are only traversed through the reference points corresponding to the region numbers when being matched, so that the efficiency of the optimization algorithm is improved.
3. The weight in the positioning process is optimized by utilizing the Euclidean distance, the Manhattan distance and the Chebyshev distance, the defect of Min's distance is overcome skillfully, the dimensions of all components of the constructed fingerprint database are consistent, namely the ' units ' of the features are the same, and the distribution of the features is also the same. And the PDR algorithm and the particle filter algorithm are combined, so that the positioning precision of the optimization algorithm is improved. Providing efficient location estimation for indoor positioning.
Drawings
Fig. 1 is a block diagram illustrating steps of an indoor fingerprint positioning-based optimization method according to an embodiment of the present invention;
fig. 2 is a deployment diagram of an AP point, a reference point, a point to be located, and a region partition in a selected location region of an optimization method based on indoor fingerprint location according to an embodiment of the present invention;
fig. 3 is a comparison between a predicted result and a real path of an indoor fingerprint positioning-based optimization method according to an embodiment of the present invention;
fig. 4 is a positioning error between each to-be-positioned point and a real coordinate, which is obtained by different algorithms of the optimization method based on indoor fingerprint positioning according to the embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, an embodiment of the present invention provides an indoor fingerprint positioning-based optimization method, including the following steps:
s1, discretely gridding the positioning area, deploying N AP points and determining a proper grid point distance;
s2, selecting M reference points according to the distance, collecting RSSI between each reference point and AP point to form a data set X with the dimension of (M.C) xN, and preprocessing, wherein C is the measuring frequency of each reference point;
s3, dividing reference point area IDs and constructing a fingerprint database;
s4, selecting a real path, and acquiring RSSI between T undetermined positioning points and AP points on the real path;
s5, obtaining the area ID of the point to be positioned according to the data set X and the reference point area ID by adopting an SVM algorithm, and traversing only the corresponding reference points in the area during matching to achieve the purposes of reducing the search space and improving the efficiency;
s6, calculating the RSSI similarity between the to-be-positioned point after the region classification and the reference point in the region by utilizing the Euclidean distance, the Manhattan distance and the Chebyshev distance to obtain position estimation;
and S7, taking the position estimation as an observation value, and combining a PDR algorithm to carry out particle filtering to obtain accurate positioning coordinates.
In the embodiment, the selected positioning area is discretely gridded according to the indoor environment, and the distance between grid points is determined to be 1 m; deploying 4 AP points, selecting 221 reference points according to the distance, collecting the received signal strength RSSI between each reference point and the AP point in real time, and forming a data set X (detailed in table 1) with the dimension of (221 & 10) multiplied by 4 which is 2210 multiplied by 4, wherein 10 is the measurement times of each reference point; meanwhile, the positioning area is divided into areas, and the area ID to which each reference point belongs is determined, as shown in fig. 2.
Table 1221 reference points real-time measurement RSSI number set X
In fig. 2, a circle ' o ' represents a reference point, an asterisk ' represents a point to be located, and a line connecting the points to be located represents a real path of walking. And dividing the regions, wherein each reference point and the to-be-positioned point have corresponding region numbers.
As an optimization scheme of the embodiment of the invention, real-time data is preprocessed, and outlier data groups in the original data set X are removed; after preprocessing of amplitude limiting average filtering and then moving average filtering, a basic fingerprint data set Y (detailed in Table 2) is obtained; and combining the basic fingerprint DATA set Y with the region vector ID to which the reference point belongs to form a multi-feature fingerprint database DATA ═ Y ID.
Where P denotes a threshold value of the clip average filter, and P ═ P4。
And
and (3) the RSSI of the c and c-1 real-time measurement between the m reference point and the n AP:
wherein L represents the sliding window size, and 0 < C ≦ C-L +1, and L ≦ 4.
Table 2 reference point RSSI data set Y after pre-processing filtering
Substituting a data set X and an area vector ID into an SVM algorithm to obtain a nonlinear classification decision function with a kernel function as an optimization scheme of the embodiment of the invention; selecting a real path, acquiring the RSSI of 48 to-be-positioned points on the path in real time, and forming a data set X with the dimension of (48.10) multiplied by 4-480 multiplied by 4test(see Table 3 for details), the data set Y is obtained after the preprocessing of the step 2test(see table 4 for details), substitute the classification decision function to obtain the area vector id with dimension of 48 × 1, implement the area classification:
wherein, wi，wjIndicating fingerThe characteristics of the reference points in the texture database,
vi,vjan area ID indicating a reference point of the image,
αi,αjrepresents the solution to the SVM fundamental form using the lagrange multiplier method,
K(wi,wj)＝wi Twjrepresenting a kernel function.
Table 348 points to be positioned real-time measuring RSSI number group Xtest
TABLE 4 preprocessed filtered RSSI data set Y with anchor pointstest
As the optimization scheme of the embodiment of the invention, according to the area ID of the point to be located, the area vector ID in the fingerprint database is matched, and only the reference point fingerprint database with the same area ID and the reference point fingerprint database with the same area ID are searched, so that the aims of reducing the search space and improving the efficiency are fulfilled; respectively utilizing Euclidean distance, Manhattan distance and Chebyshev distance to calculate RSSI similarity between the undetermined positioning point after region classification and a reference point in the region, selecting K reference points with highest similarity, and calculating position estimation by taking the reciprocal of RSSI as weight:
where p represents a variable parameter of the minkowski distance, which is the manhattan distance when p is 1; when p is 2, it is the euclidean distance; when p → ∞, this is the chebyshev distance; depending on the variation parameter, the Min distance may represent a class of distances:
wherein (x ', y ') represents a position estimate, (x 'k,y′k) Representing the coordinates of K reference points with the highest similarity, and respectively obtaining position estimation obtained by calculating the similarity through Euclidean distance, Manhattan distance and Chebyshev distance according to the value of p; the mean of the three sets of position estimates is taken as the position estimate for the improved algorithm:
where (x '", y'") indicates the position estimate of the improved algorithm, (x "")e,y″e),(x″m,y″m),(x″c,y″c) And respectively representing the position estimation of the calculation similarity of the Euclidean distance, the Manhattan distance and the Chebyshev distance.
TABLE 5 position estimation by similarity calculation using Euclidean distance, Manhattan distance, and Chebyshev distance
TABLE 6 mean of three sets of position estimates as position estimates for the improved algorithm
As an optimization scheme of the embodiment of the invention, when the RSSI of a point to be positioned is acquired in real time, two Bluetooth gyroscope sensors are respectively bound to a left foot and a right foot, and the three-axis acceleration, the three-axis angular velocity and the three-axis angle (see table 7 for details) when the mobile terminal walks along a real path are recorded according to the frequency of 20 hz; calculating the number of wave crests and wave troughs by a step frequency detection algorithm to obtain step number, thereby calculating the step length d of each step; and then, a quaternion method is utilized to obtain a heading angle theta of each step of walking (see table 8 for details).
TABLE 7 triaxial acceleration, triaxial angular velocity, and triaxial angle
TABLE 8 step length d and course Angle θ
| Ordinal number | Step size (m) | Course angle (°) |
| 1 | 1.005 | 0 |
| 2 | 0.991 | 0 |
| … | … | … |
| 47 | 1.001 | -90 |
| 48 | 1.004 | -90 |
As an optimization scheme of the embodiment of the present invention, a state transition matrix is constructed by using a PDR algorithm, a position estimation obtained by improving the algorithm is used as an observation value, particle filtering is performed together with a step length d and a course angle θ, and a final positioning coordinate is output (see table 9 and fig. 3 for details):
wherein (x)0,y0) Coordinates representing the initial point, (x)i,yi) Coordinates representing the i-th step, diAnd thetaiThe step size and heading angle of the ith step are indicated.
TABLE 9 particle filtered position estimation in conjunction with PDR algorithm
In fig. 3, the open square '□' represents the real coordinates of the point to be located, and the connecting line represents the real path; the hollow circle '. smallcircle' represents the position estimation of each point to be positioned after the improved algorithm, and is used as an observed value for carrying out particle filtering together with the PDR algorithm, and corresponding points are linked by a dotted line; asterisks' indicate final localization results.
Comparing RSSI data only fixes (WKNN); an improved positioning algorithm for calculating similarity by utilizing Euclidean distance, Manhattan distance and Chebyshev distance; taking the position estimation obtained by the improved algorithm as an observation value, and combining the observation value with the PDR algorithm to carry out particle filtering; positioning accuracy of three algorithms (see table 10 for details) and positioning error of each to-be-positioned point (see fig. 4 for details):
wherein (x)true,ytrue) Representing coordinates of points to be located on the true pathThe position coordinates.
TABLE 10 comparison of positioning accuracy
In fig. 4, the asterisk' indicates the positioning accuracy of each to-be-positioned point when positioning is performed only with RSSI (WKNN); the solid circle '●' represents the positioning accuracy of the improved algorithm when the similarity is calculated by using the Euclidean distance, the Manhattan distance and the Chebyshev distance to realize positioning; the filled squares ' ■ ' indicate the accuracy of particle filtered localization in conjunction with the PDR algorithm using the improved algorithm's position estimate as an observation. In combination with table 10, it can be seen that the improved algorithm improves the positioning accuracy from 2.73m to 2.35m, which is 13.92% higher. After being combined with the PRD algorithm and the particle filter algorithm, the precision can be improved to 0.75 m.
Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (5)
1. An optimization method based on indoor fingerprint positioning is characterized by comprising the following steps:
s1, discretely gridding the positioning area, deploying N AP points and determining a proper grid point distance;
s2, selecting M reference points according to the distance, collecting RSSI between each reference point and AP point to form a data set X with the dimension of (M.C) xN, and preprocessing, wherein C is the measuring frequency of each reference point;
s3, dividing reference point area IDs and constructing a fingerprint database;
s4, selecting a real path, and acquiring RSSI between T undetermined positioning points and AP points on the real path;
s5, obtaining the area ID of the point to be positioned according to the data set X and the reference point area ID by adopting an SVM algorithm, and traversing only the corresponding reference points in the area during matching to achieve the purposes of reducing the search space and improving the efficiency;
s6, calculating the RSSI similarity between the to-be-positioned point after the region classification and the reference point in the region by utilizing the Euclidean distance, the Manhattan distance and the Chebyshev distance to obtain position estimation;
and S7, taking the position estimation as an observation value, and combining a PDR algorithm to carry out particle filtering to obtain accurate positioning coordinates.
2. The indoor fingerprint based optimization method as claimed in claim 1, wherein: in the step S2, the RSSI between each reference point and the AP point is preprocessed, and then the step S3 is performed, where the preprocessing is to remove the outlier DATA group in the original DATA set X, and then the basic fingerprint DATA set Y is obtained by performing the limiting average filtering and then the sliding average filtering, and the basic fingerprint DATA set Y is combined with the reference point area ID to form the multi-feature fingerprint database DATA ═ Y ID.
3. The indoor fingerprint location-based optimization method of claim 2, wherein in the steps S4 and S5, the RSSI of T points to be located on the real path is collected in real time to form a data set X with dimension (T · C) xntestPreprocessing the data to obtain a data set YtestSubstituting the classification decision function to obtain an area vector id with dimension T multiplied by 1, and realizing area classification:
wherein:
wi,wjfeatures representing reference points in a fingerprint database,
vi,vjan area ID indicating a reference point of the image,
αi,αjrepresents the solution to the SVM fundamental form using the lagrange multiplier method,
K(wi,wj)＝wi Twjrepresenting a kernel function.
4. The indoor fingerprint location optimization method of claim 3, wherein the K reference points with the highest similarity are selected in the step S6, and the location estimate is calculated by taking the inverse RSSI as a weight:
where p represents a variable parameter of the minkowski distance, which is the manhattan distance when p is 1; when p is 2, it is the euclidean distance; when p → ∞, this is the chebyshev distance; depending on the variation parameter, the Min distance may represent a class of distances:
wherein, (x ", y") represents the position estimate, (x'k,y'k) Representing the coordinates of K reference points with the highest similarity, and respectively obtaining position estimation obtained by calculating the similarity through Euclidean distance, Manhattan distance and Chebyshev distance according to the value of p; the mean of the three sets of position estimates is taken as the position estimate for the improved algorithm:
where (x ' ", y '") represents the position estimate of the improved algorithm, (x ' ")e,y″e),(x″m,y″m),(x″c,y″c) And respectively representing the position estimation of the calculation similarity of the Euclidean distance, the Manhattan distance and the Chebyshev distance.
5. The indoor fingerprint positioning-based optimization method as claimed in claim 4, wherein when acquiring RSSI of the point to be positioned, two Bluetooth gyroscope sensors are respectively tied on the left foot and the right foot, and the three-axis acceleration, the three-axis angular velocity and the three-axis angle when walking along the real path are recorded according to the frequency f; calculating the number of wave crests and wave troughs by a step frequency detection algorithm to obtain step number, thereby calculating the step length d of each step; and then, obtaining a heading angle theta of each step of walking by using a quaternion method, in the step S8, constructing a state transition matrix by using a PDR algorithm, improving the position estimation obtained by the algorithm to be used as an observed value, carrying out particle filtering together with the step length d and the heading angle theta, and outputting a final positioning coordinate:
wherein (x)0,y0) Coordinates representing the initial point, (x)i,yi) Coordinates representing the i-th step, diAnd thetaiThe step size and heading angle of the ith step are indicated.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010999359.XA CN112153569A (en) | 2020-09-22 | 2020-09-22 | Indoor fingerprint positioning-based optimization method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010999359.XA CN112153569A (en) | 2020-09-22 | 2020-09-22 | Indoor fingerprint positioning-based optimization method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112153569A true CN112153569A (en) | 2020-12-29 |
Family
ID=73893549
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010999359.XA Pending CN112153569A (en) | 2020-09-22 | 2020-09-22 | Indoor fingerprint positioning-based optimization method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112153569A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113645561A (en) * | 2021-06-30 | 2021-11-12 | 南京邮电大学 | Self-adaptive switching positioning method based on indoor area division |
| CN114189809A (en) * | 2021-11-15 | 2022-03-15 | 华东师范大学 | Indoor positioning method based on convolutional neural network and high-dimensional 5G observation characteristics |
| CN114845388A (en) * | 2022-05-17 | 2022-08-02 | 电子科技大学 | Indoor positioning method for position fingerprint of sub-direction entropy weighting WKNN |
-
2020
- 2020-09-22 CN CN202010999359.XA patent/CN112153569A/en active Pending
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113645561A (en) * | 2021-06-30 | 2021-11-12 | 南京邮电大学 | Self-adaptive switching positioning method based on indoor area division |
| CN114189809A (en) * | 2021-11-15 | 2022-03-15 | 华东师范大学 | Indoor positioning method based on convolutional neural network and high-dimensional 5G observation characteristics |
| CN114189809B (en) * | 2021-11-15 | 2023-06-09 | 华东师范大学 | Indoor positioning method based on convolutional neural network and high-dimensional 5G observation feature |
| CN114845388A (en) * | 2022-05-17 | 2022-08-02 | 电子科技大学 | Indoor positioning method for position fingerprint of sub-direction entropy weighting WKNN |
| CN114845388B (en) * | 2022-05-17 | 2023-02-28 | 电子科技大学 | Indoor positioning method for position fingerprint of sub-direction entropy weighting WKNN |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112153569A (en) | Indoor fingerprint positioning-based optimization method | |
| CN103379619B (en) | A kind of localization method and system | |
| CN110118560B (en) | Indoor positioning method based on LSTM and multi-sensor fusion | |
| CN110081888B (en) | Indoor positioning algorithm based on credibility-based inertial navigation and geomagnetic fusion | |
| CN105806315B (en) | Noncooperative target relative measurement system and measuring method based on active coding information | |
| WO2021082571A1 (en) | Robot tracking method, device and equipment and computer readable storage medium | |
| Wang et al. | TOA-based NLOS error mitigation algorithm for 3D indoor localization | |
| CN108632761A (en) | A kind of indoor orientation method based on particle filter algorithm | |
| Seitz et al. | A hidden markov model for pedestrian navigation | |
| CN107179079A (en) | The indoor orientation method merged based on PDR with earth magnetism | |
| CN106597363A (en) | Pedestrian location method in indoor WLAN environment | |
| CN101907459A (en) | Monocular video based real-time posture estimation and distance measurement method for three-dimensional rigid body object | |
| CN102209382A (en) | Wireless sensor network node positioning method based on received signal strength indicator (RSSI) | |
| CN102802260A (en) | WLAN indoor positioning method based on matrix correlation | |
| CN109460539B (en) | Target positioning method based on simplified volume particle filtering | |
| CN113566820B (en) | Fused pedestrian positioning method based on position fingerprint and PDR algorithm | |
| Kwak et al. | Magnetic field based indoor localization system: A crowdsourcing approach | |
| CN106028450B (en) | Indoor positioning method and equipment | |
| Ding et al. | OGI-SLAM2: A hybrid map SLAM framework grounded in inertial-based SLAM | |
| Dong et al. | A wifi fingerprint augmentation method for 3-d crowdsourced indoor positioning systems | |
| CN111263295B (en) | WLAN indoor positioning method and device | |
| Wong et al. | Identifying centers of circulating and spiraling vector field patterns and its applications | |
| CN109600711B (en) | Indoor positioning method based on channel response frequency domain and spatial domain combined processing | |
| CN115406435A (en) | Indoor electronic map construction method and device based on WLAN and MEMS and storage medium | |
| CN113008226B (en) | Geomagnetic indoor positioning method based on gated cyclic neural network and particle filtering |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20201229 |