CN109640262A - A kind of localization method and system, equipment, storage medium based on mixed-fingerprint - Google Patents
A kind of localization method and system, equipment, storage medium based on mixed-fingerprint Download PDFInfo
- Publication number
- CN109640262A CN109640262A CN201811452581.7A CN201811452581A CN109640262A CN 109640262 A CN109640262 A CN 109640262A CN 201811452581 A CN201811452581 A CN 201811452581A CN 109640262 A CN109640262 A CN 109640262A
- Authority
- CN
- China
- Prior art keywords
- fingerprint
- reference point
- cluster
- mixed
- point
- 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.)
- Granted
Links
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
- H04W4/029—Location-based management or tracking services
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W16/00—Network planning, e.g. coverage or traffic planning tools; Network deployment, e.g. resource partitioning or cells structures
- H04W16/22—Traffic simulation tools or models
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/80—Services using short range communication, e.g. near-field communication [NFC], radio-frequency identification [RFID] or low energy communication
-
- 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
- H04W64/006—Locating users or terminals or network equipment for network management purposes, e.g. mobility management with additional information processing, e.g. for direction or speed determination
Abstract
The invention discloses a kind of localization method and system based on mixed-fingerprint, equipment, storage medium, position and mixed-fingerprint of the present invention by acquisition reference point, and cluster is carried out to reference point and realizes that localization region divides, the reference point of one cluster should train a reference point disaggregated model, point to be determined reference point disaggregated model corresponding with its is recycled to carry out reference point classification to obtain the reference point and its confidence level that confidence level is greater than default confidence level, according to the mixed-fingerprint of acquired reference point, confidence level, the position of location information acquisition point to be determined, overcome to exist in the prior art and realizes the low technical problem of the positioning accuracy of positioning using single fingerprint, effectively increase positioning accuracy.
Description
Technical field
The present invention relates to positioning field, especially a kind of localization method and system based on mixed-fingerprint, equipment, storage are situated between
Matter.
Background technique
RSSI, Received Signal Strength Indicator, signal strength instruction.
CSI, Channel State Information, channel state information.
WiFi location technology is one of academic circles at present and the research emphasis of industry.Currently taken in the positioning of outdoor scene
Business is mainly provided by GPS, network based positioning system.But since urban canyons phenomenon is increasingly severe, it is fixed to have seriously affected
Position precision.Due to popularizing for Wi-Fi hotspot, WiFi location technology is increasingly becoming research hotspot.WiFi signal strength information is due to tool
There is the feature simply easily acquired, has been widely used in current WiFi location technology.RSSI information, which can be used easily, appoints
It anticipates one and the wireless receiving network interface card of 802.11a/g/n agreement is supported to obtain.But it also have the shortcomings that it is one very big, that is, it is fixed
Position precision is unstable, especially in complicated architectural environment, since wireless signal will receive the barrier of barrier in the air,
The communication process that will lead to signal becomes extremely complex.Especially many of environment mobile object when, the influence of multipath effect,
Meeting causes positioning accuracy to generate biggish fluctuation so that the increase of RSSI signal fluctuation amplitude.
For the precision fluctuation for solving the problems, such as RSSI location technology, Halperin D proposes CSI location technology at first.WiFi
Channel state information can portray the state of each OFDM subchannel, be a more stable signal index.CSI location technology is
The characteristics of using OFDM technology frequency diversity, realizes the precise measurement of sub-channel information, obtains finer subchannel
Information.Since the CSI of only some subchannel can generate fluctuation, CSI location technology can obtain more stable wireless channel
Fingerprint, to improve positioning accuracy and stability.Due to the isomerism of space environment, the feature for being included to CSI fingerprint is mentioned
Taking is a significant challenge.To sum up, RSSI fingerprint or CSI fingerprint (such as FILA algorithm, ConFi is used alone in existing locating scheme
Algorithm), a part of finger print information can be all lost, position inaccurate is caused, position error is larger.
Summary of the invention
The present invention is directed to solve at least some of the technical problems in related technologies.For this purpose, of the invention
One purpose is to provide a kind of localization method based on mixed-fingerprint and system, equipment, storage medium, effectively improves positioning accurate
Degree.
The technical scheme adopted by the invention is that:
In a first aspect, the present invention provides a kind of localization method based on mixed-fingerprint, comprising the following steps:
Information collection step acquires the location information of multiple reference points in localization region and comes from wireless signal transmitting terminal
Mixed-fingerprint, the mixed-fingerprint includes RSSI fingerprint and CSI fingerprint；
Sorting procedure clusters the reference point to obtain multiple clusters；
Model training step, according to the mixed-fingerprint of the reference point in cluster training reference point disaggregated model, a cluster is corresponding
One reference point disaggregated model；
Cluster obtaining step to be positioned obtains the point to be determined pair according to the mixed-fingerprint of point to be determined and the multiple cluster
The cluster to be positioned answered；
Reference point obtaining step, according to reference point corresponding to the mixed-fingerprint of the point to be determined and the cluster to be positioned
For disaggregated model to obtain multiple reference points and its confidence level, the confidence level of the reference point is greater than default confidence level；
Point to be determined position acquisition step is believed according to the position of the multiple reference point and its confidence level, the reference point
Breath obtains the location information of the point to be determined.
Further, the sorting procedure specifically includes:
The number of cluster is determined by elbow method；
Reference point is clustered according to the number of the cluster, the RSSI fingerprint of the reference point and k-means algorithm
To obtain multiple clusters.
Further, using the Euclidean distance of RSSI vector as distance metric when cluster.
Further, the cluster obtaining step to be positioned specifically includes:
Obtain the RSSI fingerprint of the point to be determined to each cluster mean vector distance；
Obtain to be positioned cluster of the cluster as the point to be determined corresponding to the smallest mean vector.
Further, the reference point disaggregated model is neural network reference point disaggregated model.
Further, the point to be determined position acquisition step specifically includes:
Believed according to the position of three side location algorithm of square weighting, the multiple reference point and its confidence level, the reference point
Breath obtains the location information of the point to be determined.
Second aspect, the present invention provide a kind of positioning system based on mixed-fingerprint, comprising:
Information acquisition unit, for acquiring the location information of multiple reference points in localization region and being sent out from wireless signal
The mixed-fingerprint at end is penetrated, the mixed-fingerprint includes RSSI fingerprint and CSI fingerprint；
Cluster cell, for being clustered to the reference point to obtain multiple clusters；
Model training unit, for according to the mixed-fingerprint of the reference point in cluster training reference point disaggregated model, a cluster
A corresponding reference point disaggregated model；
Cluster acquiring unit to be positioned, it is described to be positioned for being obtained according to the mixed-fingerprint of point to be determined and the multiple cluster
The corresponding cluster to be positioned of point；
Reference point acquiring unit, for the ginseng according to corresponding to the mixed-fingerprint of the point to be determined and the cluster to be positioned
For examination point disaggregated model to obtain multiple reference points and its confidence level, the confidence level of the reference point is greater than default confidence level；
Point to be determined position acquisition unit, for the position according to the multiple reference point and its confidence level, the reference point
Set the location information of point to be determined described in acquisition of information.
The third aspect, the present invention provide a kind of positioning system based on mixed-fingerprint, comprising:
Wireless signal transmitting terminal, for emitting wireless signal；
Mobile terminal, for acquiring the location information of multiple reference points in localization region and from wireless signal transmitting terminal
Mixed-fingerprint, the mixed-fingerprint includes RSSI fingerprint and CSI fingerprint；
Cluster cell, for being clustered to the reference point to obtain multiple clusters；
Model training unit, for according to the mixed-fingerprint of the reference point in cluster training reference point disaggregated model, a cluster
A corresponding reference point disaggregated model；
Cluster acquiring unit to be positioned, it is described to be positioned for being obtained according to the mixed-fingerprint of point to be determined and the multiple cluster
The corresponding cluster to be positioned of point；
Reference point acquiring unit, for the ginseng according to corresponding to the mixed-fingerprint of the point to be determined and the cluster to be positioned
For examination point disaggregated model to obtain multiple reference points and its confidence level, the confidence level of the reference point is greater than default confidence level；
Point to be determined position acquisition unit, for the position according to the multiple reference point and its confidence level, the reference point
Set the location information of point to be determined described in acquisition of information.
Fourth aspect, the present invention provide a kind of positioning device based on mixed-fingerprint, comprising:
At least one processor；And the memory being connect at least one described processor communication；Wherein,
The memory is stored with the instruction that can be executed by least one described processor, and described instruction is by described at least one
A processor executes, so that at least one described processor is able to carry out the localization method based on mixed-fingerprint.
5th aspect, the present invention provide a kind of computer readable storage medium, the computer-readable recording medium storage
There are computer executable instructions, the computer executable instructions are used to make computer execution is described to determine based on mixed-fingerprint
Position method.
The beneficial effects of the present invention are:
The present invention carries out cluster to reference point and realizes that localization region is drawn by obtaining position and the mixed-fingerprint of reference point
Point, the reference point of a cluster should train a reference point disaggregated model, recycle point to be determined reference point corresponding with its
Disaggregated model carries out reference point classification to obtain the reference point and its confidence level that confidence level is greater than default confidence level, according to acquired
The mixed-fingerprint of reference point, confidence level, location information obtain the position of point to be determined, overcome and exist in the prior art using single
One fingerprint realizes the low technical problem of the positioning accuracy of positioning, effectively increases positioning accuracy.
In addition, the present invention is also by according to three side location algorithm of square weighting, multiple reference points and its confidence level, reference point
Location information obtain point to be determined location information, can further improve positioning accuracy, obtain the position of accurate point to be determined
Confidence breath.
Detailed description of the invention
Fig. 1 is an a kind of specific embodiment method flow diagram of the localization method based on mixed-fingerprint in the present invention；
Fig. 2 is an a kind of specific embodiment specific flow chart of the localization method based on mixed-fingerprint in the present invention；
Fig. 3 is a kind of specific embodiment signal of sorting procedure in the localization method based on mixed-fingerprint in the present invention
Figure；
Fig. 4 is the one of three side location algorithm of square weighting specific in a kind of localization method based on mixed-fingerprint in the present invention
Embodiment schematic diagram.
Specific embodiment
It should be noted that in the absence of conflict, the features in the embodiments and the embodiments of the present application can phase
Mutually combination.
Embodiment 1
In fact, being the feature more stable with the variation of distance than RSSI value the present invention is based on CSI value, and combine empty
Between closer two points of positional distance, Euclidean distance between the fingerprint of two points also can this smaller rule propose one kind
Localization method based on mixed-fingerprint, the localization method are the region segmentation location models based on mixed-fingerprint and propose, base
It in the region segmentation location model of mixed-fingerprint is clustered based on the reference point first to localization region, again in each cluster
Orientation problem classification problem as a reference point is belonged to probabilistic fingerprint location model by the thought classified.Specifically
Ground is an a kind of specific embodiment method flow diagram of the localization method based on mixed-fingerprint in the present invention with reference to Fig. 1, Fig. 1, one
Localization method of the kind based on mixed-fingerprint, comprising the following steps:
Information collection step acquires the location information of multiple reference points in localization region and comes from wireless signal transmitting terminal
Mixed-fingerprint, mixed-fingerprint includes RSSI fingerprint and CSI fingerprint；
Sorting procedure clusters reference point to obtain multiple clusters；
Model training step, according to the mixed-fingerprint of the reference point in cluster training reference point disaggregated model, a cluster is corresponding
One reference point disaggregated model；
It is corresponding undetermined to obtain point to be determined according to the mixed-fingerprint of point to be determined and multiple clusters for cluster obtaining step to be positioned
Position cluster；
Reference point obtaining step, according to reference point disaggregated model corresponding to the mixed-fingerprint of point to be determined and cluster to be positioned
To obtain multiple reference points and its confidence level, the confidence level of the reference point is greater than default confidence level；
Point to be determined position acquisition step, according to the location information of multiple reference points and its confidence level, reference point obtain to
The location information of anchor point.
Precise positioning is realized using above-mentioned steps, is overcome in conjunction with mixed-fingerprint, reference point cluster and reference point classification existing
Have in technology to exist and realize the low technical problem of the positioning accuracy of positioning using single fingerprint, effectively increases positioning accuracy.
It is an a kind of specific embodiment detailed process of the localization method based on mixed-fingerprint in the present invention with reference to Fig. 2, Fig. 2
Figure, is below specifically described localization method:
First stage is off-line phase:
Information collection step: the wireless signal transmitting terminal in localization region launches wireless signal, wireless signal transmitting terminal
Can have multiple, in the present embodiment, wireless signal transmitting terminal is Wi-Fi hotspot, and Wi-Fi hotspot launches WiFi signal.It can benefit
The mixed-fingerprint that reference point in localization region is acquired with mobile terminal (such as laptop), when mobile terminal is in some reference point
When can connect with some Wi-Fi hotspot, then communication connection is established to complete the acquisition of mixed-fingerprint, the energy in a reference point
Collect the mixed-fingerprint from least one wireless signal transmitting terminal；Specifically, for the reliability being held in position, at this
The hot spot that the signal strength mean value that reference point can receive is greater than some threshold value just carries out mixed-fingerprint acquisition, this threshold value one
As take a biggish value；The hot spot for being greater than some threshold value to RSSI value just carries out CSI value acquisition.And for the position of reference point
Information can obtain the opposite geographical position coordinates (x, y) of each reference point, by spatial position modeling so as to be converted into
Latitude and longitude coordinates.In addition, each reference point has a reference point number, and there is each Wi-Fi hotspot a Wi-Fi hotspot to compile
Number.
Sorting procedure: referring to Fig. 3, and Fig. 3 is sorting procedure in a kind of localization method based on mixed-fingerprint in the present invention
One specific embodiment schematic diagram determines the number K of cluster by elbow method first；Number K, reference further according to cluster
The RSSI finger print data collection and k-means algorithm of point cluster reference point to obtain K cluster, with RSSI vector when cluster
Euclidean distance gathers the lesser reference point of RSSI vector Euclidean distance in one class as distance metric.Specific cluster
For process as shown in figure 3, can determine the cluster of each reference point after cluster, the reference point of the same cluster forms a ginseng
Examination point collection is realized and carries out region division to localization region according to the cluster result of RSSI fingerprint, cluster to obtain k to reference point
Cluster, as shown in Fig. 2, N in Fig. 2 is K, then available multiple mixed-fingerprint collection, mixed-fingerprint collection, region 2 such as region 1
Mixed-fingerprint collection etc.；Correspondingly, the central point of the fingerprint of each cluster can be obtained, central point is the mixed of all reference points in cluster
Close the mean value of fingerprint vector, i.e. mean vector.
Model training step: reference point disaggregated model is neural network reference point disaggregated model, neural network reference point point
The activation primitive of the neural network that class model is one 3 layers, the hidden layer of neural network uses line rectification unit (ReLU), mind
The activation primitive of output layer through network is Softmax, and the neuron number of output layer is equal to the number of reference point in cluster with defeated
The similarity degree of the mixed-fingerprint of the mixed-fingerprint and each reference point of the point inputted out, i.e. matching probability namely confidence level.
Cluster belonging to each reference point is obtained through the above steps, while having obtained the reference point set that each cluster includes.With every
A cluster is (the neural network reference point classification in such as Fig. 2 of the corresponding neural network reference point disaggregated model of module training
Model 1, neural network reference point disaggregated model 2 etc.).Using in same cluster each reference point acquisition mixed-fingerprint as
Training sample, using reference point number as class label, respectively each cluster training reference point disaggregated model.
Second stage is on-line stage:
Cluster obtaining step to be positioned: first obtain point to be determined mixed-fingerprint in RSSI fingerprint to each cluster mean value to
The distance of amount；Obtain to be positioned cluster of the cluster as point to be determined corresponding to the smallest mean vector.
Reference point obtaining step: corresponding reference point disaggregated model is obtained according to cluster to be positioned and (has trained the mould finished
Type), according to reference point disaggregated model corresponding to the mixed-fingerprint of point to be determined and cluster to be positioned with obtain multiple reference points and
Its confidence level, wherein the confidence level that multiple reference points meet reference point is greater than default confidence level, and default confidence level is adjustable,
In the present embodiment, maximum 3 reference points of confidence level are taken.
Point to be determined position acquisition step: in the present embodiment, according to three side location algorithm of square weighting, three reference points and
Its confidence level, three reference points location information obtain the location information of point to be determined.It is one kind in the present invention with reference to Fig. 4, Fig. 4
A specific embodiment schematic diagram of three side location algorithm of square weighting in localization method based on mixed-fingerprint, wherein WiFi AP
For Wi-Fi hotspot, cross star phenotypic marker indicates the position of reference point, and dot indicates the actual position of point to be determined, RP1, RP2, RP3
For with maximum three reference points of the confidence level of point to be determined, reference point RP1, RP2, RP3Position coordinates be respectively (x1, y1)、
(x2, y2) and (x3, y3), according to three corresponding confidence levels of reference point, reference point RP1, RP2, RP3Confidence level be respectively P1, P2
And P3, using the position coordinates of reference point, (reference point higher for confidence level, distributes higher power with confidence level square weighting
Weight) position coordinates (x of point to be determined can be calculatede, ye):
After obtaining the maximum multiple reference points of confidence level with point to be determined, it also can use maximum likelihood positioning and calculate
Method calculates the position coordinates of point to be determined, but position error is larger.According to the position coordinates (x of point to be determined obtainede,
ye) with the true geographical location (x of point to be determinedr, yr) calculate Euclidean distance E between themdisNamely position error, it utilizes
The size of position error can assess the positioning performance quality of localization method.
Localization method of the invention uses RSSI and CSI fingerprint characteristic simultaneously, and core was made of 3 stages.First
A stage is reference point fingerprint collecting, off-line training step.In the region for needing to establish WiFi location model, mobile terminal is allowed to adopt
Collect RSSI the and CSI mixed-fingerprint characteristics of several reference points.Second stage is divided into two steps.The first step is according to reference point
RSSI fingerprint carries out region division, while calculating and saving the RSSI vector at the center of cluster belonging to each region；Second step is adopted
It is each regional training reference point disaggregated model with neural network algorithm.Three phases are the tuning on-line stages.For one
Location Request first acquires mixed-fingerprint, then determines the Location Request affiliated area according to RSSI fingerprint.Finally by mixed-fingerprint
The corresponding neural network classification model of affiliated area is inputted, is acquired and the highest multiple references of fingerprint matching probability to be positioned
Point, and calculate according to the square weighting value of matching probability the estimated coordinates of fingerprint to be positioned.
Embodiment 2
A kind of positioning system based on mixed-fingerprint, comprising:
Information acquisition unit, for acquiring the location information of multiple reference points in localization region and being sent out from wireless signal
The mixed-fingerprint at end is penetrated, mixed-fingerprint includes RSSI fingerprint and CSI fingerprint；
Cluster cell, for being clustered to reference point to obtain multiple clusters；
Model training unit, for according to the mixed-fingerprint of the reference point in cluster training reference point disaggregated model, a cluster
A corresponding reference point disaggregated model；
Cluster acquiring unit to be positioned, it is corresponding for obtaining point to be determined according to the mixed-fingerprint of point to be determined and multiple clusters
Cluster to be positioned；
Reference point acquiring unit is classified for the reference point according to corresponding to the mixed-fingerprint of point to be determined and cluster to be positioned
For model to obtain multiple reference points and its confidence level, the confidence level of the reference point is greater than default confidence level；
Point to be determined position acquisition unit, for being obtained according to the location information of multiple reference points and its confidence level, reference point
Take the location information of point to be determined.
The specific work process of positioning system based on mixed-fingerprint can refer to the description of embodiment 1, repeat no more.
Embodiment 3
A kind of positioning system based on mixed-fingerprint, comprising:
Wireless signal transmitting terminal, for emitting wireless signal, such as Wi-Fi hotspot；
Mobile terminal, for acquiring the location information of multiple reference points in localization region and from wireless signal transmitting terminal
Mixed-fingerprint, mixed-fingerprint includes RSSI fingerprint and CSI fingerprint, wherein mobile terminal can be laptop；
Cluster cell, for being clustered to reference point to obtain multiple clusters；
Model training unit, for according to the mixed-fingerprint of the reference point in cluster training reference point disaggregated model, a cluster
A corresponding reference point disaggregated model；
Cluster acquiring unit to be positioned, it is corresponding for obtaining point to be determined according to the mixed-fingerprint of point to be determined and multiple clusters
Cluster to be positioned；
Reference point acquiring unit is classified for the reference point according to corresponding to the mixed-fingerprint of point to be determined and cluster to be positioned
For model to obtain multiple reference points and its confidence level, the confidence level of the reference point is greater than default confidence level；
Point to be determined position acquisition unit, for being obtained according to the location information of multiple reference points and its confidence level, reference point
Take the location information of point to be determined.
The specific work process of positioning system based on mixed-fingerprint can refer to the description of embodiment 1, repeat no more.
Embodiment 4
A kind of positioning device based on mixed-fingerprint, comprising:
At least one processor；And the memory being connect at least one described processor communication；Wherein,
The memory is stored with the instruction that can be executed by least one described processor, and described instruction is by described at least one
A processor executes, so that at least one described processor is able to carry out the localization method based on mixed-fingerprint.About
The specific descriptions of localization method based on mixed-fingerprint can refer to the description of embodiment 1, repeat no more.
Embodiment 5
A kind of computer readable storage medium, the computer-readable recording medium storage have computer executable instructions,
The computer executable instructions are used to that computer to be made to execute the localization method based on mixed-fingerprint.About based on mixing
The specific descriptions of the localization method of fingerprint can refer to the description of embodiment 1, repeat no more.
It is to be illustrated to preferable implementation of the invention, but the invention is not limited to the implementation above
Example, those skilled in the art can also make various equivalent variations on the premise of without prejudice to spirit of the invention or replace
It changes, these equivalent deformations or replacement are all included in the scope defined by the claims of the present application.
Claims (10)
1. a kind of localization method based on mixed-fingerprint, which comprises the following steps:
Information collection step acquires the location information of multiple reference points in localization region and mixing from wireless signal transmitting terminal
Fingerprint is closed, the mixed-fingerprint includes RSSI fingerprint and CSI fingerprint；
Sorting procedure clusters the reference point to obtain multiple clusters；
Model training step, according to the mixed-fingerprint of the reference point in cluster training reference point disaggregated model, a cluster is one corresponding
Reference point disaggregated model；
It is corresponding to obtain the point to be determined according to the mixed-fingerprint of point to be determined and the multiple cluster for cluster obtaining step to be positioned
Cluster to be positioned；
Reference point obtaining step is classified according to reference point corresponding to the mixed-fingerprint of the point to be determined and the cluster to be positioned
For model to obtain multiple reference points and its confidence level, the confidence level of the reference point is greater than default confidence level；
Point to be determined position acquisition step, is obtained according to the location information of the multiple reference point and its confidence level, the reference point
Take the location information of the point to be determined.
2. the localization method according to claim 1 based on mixed-fingerprint, which is characterized in that the sorting procedure specifically wraps
It includes:
The number of cluster is determined by elbow method；
Reference point is clustered to obtain according to the number of the cluster, the RSSI fingerprint of the reference point and k-means algorithm
Take multiple clusters.
3. the localization method according to claim 2 based on mixed-fingerprint, which is characterized in that with RSSI vector when cluster
Euclidean distance is as distance metric.
4. the localization method according to claim 3 based on mixed-fingerprint, which is characterized in that the cluster to be positioned obtains step
Suddenly it specifically includes:
Obtain the RSSI fingerprint of the point to be determined to each cluster mean vector distance；
Obtain to be positioned cluster of the cluster as the point to be determined corresponding to the smallest mean vector.
5. the localization method according to any one of claims 1 to 4 based on mixed-fingerprint, which is characterized in that the reference
Point disaggregated model is neural network reference point disaggregated model.
6. the localization method according to any one of claims 1 to 4 based on mixed-fingerprint, which is characterized in that described undetermined
Site location obtaining step specifically includes:
It is obtained according to the location information of three side location algorithm of square weighting, the multiple reference point and its confidence level, the reference point
Take the location information of the point to be determined.
7. a kind of positioning system based on mixed-fingerprint characterized by comprising
Information acquisition unit, for acquiring the location information of multiple reference points in localization region and from wireless signal transmitting terminal
Mixed-fingerprint, the mixed-fingerprint includes RSSI fingerprint and CSI fingerprint；
Cluster cell, for being clustered to the reference point to obtain multiple clusters；
Model training unit, for according to the mixed-fingerprint of the reference point in cluster training reference point disaggregated model, a cluster to be corresponding
One reference point disaggregated model；
Cluster acquiring unit to be positioned, for obtaining the point to be determined pair according to the mixed-fingerprint of point to be determined and the multiple cluster
The cluster to be positioned answered；
Reference point acquiring unit, for the reference point according to corresponding to the mixed-fingerprint of the point to be determined and the cluster to be positioned
For disaggregated model to obtain multiple reference points and its confidence level, the confidence level of the reference point is greater than default confidence level；
Point to be determined position acquisition unit, for being believed according to the position of the multiple reference point and its confidence level, the reference point
Breath obtains the location information of the point to be determined.
8. a kind of positioning system based on mixed-fingerprint characterized by comprising
Wireless signal transmitting terminal, for emitting wireless signal；
Mobile terminal, for acquiring the location information of multiple reference points in localization region and mixing from wireless signal transmitting terminal
Fingerprint is closed, the mixed-fingerprint includes RSSI fingerprint and CSI fingerprint；
Cluster cell, for being clustered to the reference point to obtain multiple clusters；
Model training unit, for according to the mixed-fingerprint of the reference point in cluster training reference point disaggregated model, a cluster to be corresponding
One reference point disaggregated model；
Cluster acquiring unit to be positioned, for obtaining the point to be determined pair according to the mixed-fingerprint of point to be determined and the multiple cluster
The cluster to be positioned answered；
Reference point acquiring unit, for the reference point according to corresponding to the mixed-fingerprint of the point to be determined and the cluster to be positioned
For disaggregated model to obtain multiple reference points and its confidence level, the confidence level of the reference point is greater than default confidence level；
Point to be determined position acquisition unit, for being believed according to the position of the multiple reference point and its confidence level, the reference point
Breath obtains the location information of the point to be determined.
9. a kind of positioning device based on mixed-fingerprint characterized by comprising
At least one processor；And the memory being connect at least one described processor communication；Wherein,
The memory is stored with the instruction that can be executed by least one described processor, and described instruction is by described at least one
It manages device to execute, so that at least one described processor is able to carry out as claimed in any one of claims 1 to 6 based on mixed-fingerprint
Localization method.
10. a kind of computer readable storage medium, which is characterized in that the computer-readable recording medium storage has computer can
It executes instruction, the computer executable instructions are for executing computer as claimed in any one of claims 1 to 6 based on mixed
Close the localization method of fingerprint.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811452581.7A CN109640262B (en) | 2018-11-30 | 2018-11-30 | Positioning method, system, equipment and storage medium based on mixed fingerprints |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811452581.7A CN109640262B (en) | 2018-11-30 | 2018-11-30 | Positioning method, system, equipment and storage medium based on mixed fingerprints |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109640262A true CN109640262A (en) | 2019-04-16 |
| CN109640262B CN109640262B (en) | 2021-01-05 |
Family
ID=66070407
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811452581.7A Active CN109640262B (en) | 2018-11-30 | 2018-11-30 | Positioning method, system, equipment and storage medium based on mixed fingerprints |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109640262B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110381440A (en) * | 2019-06-16 | 2019-10-25 | 西安电子科技大学 | The fingerprint indoor orientation method of joint RSS and CSI based on deep learning |
| CN111417066A (en) * | 2020-03-05 | 2020-07-14 | 滴图(北京)科技有限公司 | Positioning method, positioning device, electronic equipment and computer readable storage medium |
| CN113204715A (en) * | 2021-04-13 | 2021-08-03 | Oppo广东移动通信有限公司 | User recommendation method and device, electronic equipment and storage medium |
| CN113784280A (en) * | 2021-07-28 | 2021-12-10 | 中国矿业大学 | Fireman indoor positioning method based on WiFi fingerprint |
| CN114495243A (en) * | 2022-04-06 | 2022-05-13 | 第六镜科技(成都)有限公司 | Image recognition model training method, image recognition model training device, image recognition method, image recognition device and electronic equipment |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102932738A (en) * | 2012-10-31 | 2013-02-13 | 北京交通大学 | Improved positioning method of indoor fingerprint based on clustering neural network |
| CN104185275A (en) * | 2014-09-10 | 2014-12-03 | 北京航空航天大学 | Indoor positioning method based on WLAN |
| CN105163282A (en) * | 2015-09-22 | 2015-12-16 | 济南东朔微电子有限公司 | Indoor positioning system and positioning method based on Bluetooth location fingerprint |
| US20160286519A1 (en) * | 2015-03-24 | 2016-09-29 | Ramot At Tel-Aviv University Ltd. | Determining a location of an ofdm transmitter |
| EP3372024A1 (en) * | 2015-11-06 | 2018-09-12 | Telefonaktiebolaget LM Ericsson (publ) | Positioning in wlan systems |
| CN108540931A (en) * | 2018-04-08 | 2018-09-14 | 中国矿业大学 | A kind of underground section segmented sighting distance node cooperation location algorithm |
| CN108650628A (en) * | 2018-03-24 | 2018-10-12 | 西安电子科技大学 | The indoor orientation method that ranging and fingerprint based on Wi-Fi network are combined |
-
2018
- 2018-11-30 CN CN201811452581.7A patent/CN109640262B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102932738A (en) * | 2012-10-31 | 2013-02-13 | 北京交通大学 | Improved positioning method of indoor fingerprint based on clustering neural network |
| CN104185275A (en) * | 2014-09-10 | 2014-12-03 | 北京航空航天大学 | Indoor positioning method based on WLAN |
| US20160286519A1 (en) * | 2015-03-24 | 2016-09-29 | Ramot At Tel-Aviv University Ltd. | Determining a location of an ofdm transmitter |
| CN105163282A (en) * | 2015-09-22 | 2015-12-16 | 济南东朔微电子有限公司 | Indoor positioning system and positioning method based on Bluetooth location fingerprint |
| EP3372024A1 (en) * | 2015-11-06 | 2018-09-12 | Telefonaktiebolaget LM Ericsson (publ) | Positioning in wlan systems |
| CN108650628A (en) * | 2018-03-24 | 2018-10-12 | 西安电子科技大学 | The indoor orientation method that ranging and fingerprint based on Wi-Fi network are combined |
| CN108540931A (en) * | 2018-04-08 | 2018-09-14 | 中国矿业大学 | A kind of underground section segmented sighting distance node cooperation location algorithm |
Non-Patent Citations (1)
| Title |
|---|
| 党小超: "一种RSS和CSI融合的二阶段室内定位方法", 《传感技术学报》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110381440A (en) * | 2019-06-16 | 2019-10-25 | 西安电子科技大学 | The fingerprint indoor orientation method of joint RSS and CSI based on deep learning |
| CN111417066A (en) * | 2020-03-05 | 2020-07-14 | 滴图(北京)科技有限公司 | Positioning method, positioning device, electronic equipment and computer readable storage medium |
| CN113204715A (en) * | 2021-04-13 | 2021-08-03 | Oppo广东移动通信有限公司 | User recommendation method and device, electronic equipment and storage medium |
| CN113204715B (en) * | 2021-04-13 | 2024-03-29 | Oppo广东移动通信有限公司 | User recommendation method and device, electronic equipment and storage medium |
| CN113784280A (en) * | 2021-07-28 | 2021-12-10 | 中国矿业大学 | Fireman indoor positioning method based on WiFi fingerprint |
| CN114495243A (en) * | 2022-04-06 | 2022-05-13 | 第六镜科技(成都)有限公司 | Image recognition model training method, image recognition model training device, image recognition method, image recognition device and electronic equipment |
| CN114495243B (en) * | 2022-04-06 | 2022-07-05 | 第六镜科技(成都)有限公司 | Image recognition model training and image recognition method and device, and electronic equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109640262B (en) | 2021-01-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109640262A (en) | A kind of localization method and system, equipment, storage medium based on mixed-fingerprint | |
| Li et al. | AF-DCGAN: Amplitude feature deep convolutional GAN for fingerprint construction in indoor localization systems | |
| CN102480677B (en) | A kind of determination method and apparatus of fingerprint positioning error | |
| CN104185275B (en) | A kind of indoor orientation method based on WLAN | |
| CN102802260B (en) | WLAN indoor positioning method based on matrix correlation | |
| CN102480678B (en) | Fingerprint positioning method and system | |
| WO2015154438A1 (en) | Positioning method and device | |
| CN106125045B (en) | A kind of ADAPTIVE MIXED indoor orientation method based on Wi-Fi | |
| CN105898713A (en) | WiFi fingerprint indoor positioning method based on weighted cosine similarity | |
| CN106912105A (en) | 3-D positioning method based on PSO_BP neutral nets | |
| Ding et al. | Overview of received signal strength based fingerprinting localization in indoor wireless LAN environments | |
| CN103901398B (en) | A kind of location fingerprint localization method based on combination collating sort | |
| CN104038901B (en) | Indoor positioning method for reducing fingerprint data acquisition workload | |
| CN111479231A (en) | Indoor fingerprint positioning method for millimeter wave large-scale MIMO system | |
| CN109951798A (en) | Merge the enhancing location fingerprint indoor orientation method of Wi-Fi and bluetooth | |
| CN109640269A (en) | Fingerprint positioning method based on CSI Yu Time Domain Fusion algorithm | |
| CN109084774A (en) | A kind of unmanned plane cluster localization method based on channel state information | |
| CN107995682B (en) | Wireless access independent positioning method, device, electronic equipment and storage medium | |
| CN103634902A (en) | Novel indoor positioning method based on fingerprint cluster | |
| CN110222776A (en) | A kind of indoor Passive Location based on CSI finger print information | |
| CN113207089A (en) | Position fingerprint positioning method based on CSI and crowdsourcing migration self-calibration updating | |
| CN110049549A (en) | More fusion indoor orientation methods and its system based on WiFi fingerprint | |
| CN109633530B (en) | Positioning method, positioning system, positioning equipment and storage medium | |
| Wang et al. | A novel fingerprint localization algorithm based on modified channel state information using Kalman filter | |
| CN112867021B (en) | Improved TrAdaBoost-based indoor positioning method for transfer learning |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |