CN105682224B - A kind of distributed wireless fingerprint positioning method for exempting from off-line training - Google Patents
A kind of distributed wireless fingerprint positioning method for exempting from off-line training Download PDFInfo
- Publication number
- CN105682224B CN105682224B CN201610163617.4A CN201610163617A CN105682224B CN 105682224 B CN105682224 B CN 105682224B CN 201610163617 A CN201610163617 A CN 201610163617A CN 105682224 B CN105682224 B CN 105682224B
- Authority
- CN
- China
- Prior art keywords
- anchor node
- fingerprint
- node
- anchor
- information
- 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.)
- Expired - Fee Related
Links
Classifications
-
- 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
-
- 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
-
- H04W4/04—
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Position Fixing By Use Of Radio Waves (AREA)
- Mobile Radio Communication Systems (AREA)
- Collating Specific Patterns (AREA)
Abstract
A kind of distributed wireless fingerprint positioning method for exempting from off-line training, including fingerprint base establishment stage, fingerprint matching stage, positioning stage, it is characterized by: without carrying out offline information collection, but each anchor node sends request message to neighbours' anchor node, the neighbor node for each receiving request calculates signal strength, then the anchor node for sending request is returned result to, the anchor node for sending request generates fingerprint base information fi, fingerprint base information fiIt is stored directly in the node and neighbours' anchor node；In positioning system, anchor node quantity is larger and when being evenly distributed, it is desirable that the off-line training workload and fingerprint database maintenance workload of positioning system are small as far as possible, and when individual anchor nodes power-off, damage occur or increasing, system can be normally carried out work.
Description
Technical field
The present invention relates to indoor wireless positioning fields, especially in certain large-scale indoor scenes, using wireless fingerprint positioning,
When it is a large amount of, be uniformly distributed anchor node when, it is desirable that it is adoptable a kind of fixed when positioning system has preferable stability and scalability
Position method.
Background technique
Wireless fingerprint positioning is one of the basic skills of current indoor positioning, generally there is 5 key links: Radio Map
Foundation, clustering block, AP(Access Point, be also possible to anchor node, such as the ZigBee node of known coordinate here)
Selection, location feature extract and location estimation, wherein the foundation of Radio Map and clustering block belong to off-line phase, behind three
A step belongs to on-line stage.The first step, RSSI and physical location dependence of the system to all AP in target localization region
It is described, with the realization of offline fingerprint collecting method.Entire localization region is divided into several locators in off-line phase by second step
Region constructs location model in positioning subregion；User terminal is categorized into positioning subregion, then by on-line stage in advance
It realizes and is accurately positioned in positioning subregion again.Location model is limited in positioning subregion using clustering block algorithm, letter
Change the mapping relations of RSSI and physical location, to lay the foundation to improve positioning accuracy, reducing computation complexity.Third step,
To in stator region, choosing optimal AP subclass for positioning, removal differentiates that stationkeeping ability is weak, the biggish AP of noise.4th
Step is extracted on the basis of giving optimal AP subclass and differentiates location feature.5th step, in the location feature database constructed
On the basis of, user location is estimated by matching type or learning-oriented location algorithm.
But using fingerprint location method, there are such several difficulties: 1) RSSI signal is uncertain, so that different physical bits
The RSS signal mode set is difficult to distinguish one another, and seriously affects positioning accuracy.The probabilistic principal element of RSS is caused to have: within doors
The influence of construction pattern, the walking about of personnel, the multipath effect of wireless signal and other 2.4G open frequency range signals coexisted.2)
The power consumption as caused by computation complexity is larger, and limitation finger print matching method uses in certain terminals, and 3) fingerprint technique is in offline rank
The foundation of the Radio Map of section requires a great deal of time and energy, and the later period is due to environmental change or certain AP nodes
Damage and increase, and need to re-establish Radio Map, the maintenance workload of entire fingerprint database is huge.Above-mentioned reason harm
The popularization and application of fingerprint location method are hindered.
There are a kind of application scenarios, such as underground large parking lot, area is bigger, transmits and senses using wireless sensor network
Signal (such as automobile parking signal), and using the position of wireless sensor network positioning car owner or automobile (by positioning car owner
Parking card on hand realizes that practical parking card is also a radio node), in this case, due to wireless sensor network
Node is numerous and distribution uniform, and has known these node coordinate positions in advance, can using in the present invention exempt from from
The Distributed localization method of line training.
Summary of the invention
The technical problems to be solved by the present invention are: designing a kind of indoor orientation method based on wireless fingerprint, positioning
In system, anchor node quantity is larger and when being evenly distributed, it is desirable that the off-line training workload and fingerprint database of positioning system are tieed up
It is small as far as possible to protect workload, when individual anchor node power-off, damage occur or increasing, system can be normally carried out work.
In order to solve the above-mentioned technical problem, the present invention uses following scheme:
A kind of distributed wireless fingerprint positioning method for exempting from off-line training, including fingerprint base establishment stage, fingerprint matching rank
Section, positioning stage, it is characterised in that: without carrying out offline information collection, but each anchor node is sent to neighbours' anchor node
Request message, the neighbor node for each receiving request calculate signal strength, then return result to the anchor node for sending request,
The anchor node for sending request generates fingerprint base information fi, fingerprint base information fiIt is stored directly in the node and neighbours' anchor node.
The fingerprint matching stage uses distributed fingerprint matching algorithm: node ID to be positionediAnchor node peripherad first
Location Request is sent, anchor node receives the signal strength that request then calculates the node, then signal strength information, itself position
It sets coordinate and mark number is sent to IDi, node generation finger print information S to be positioned, then node to be positioned broadcasts finger print information S
It is sent to neighbouring anchor node, by the fingerprint base information f in anchor node on anchor nodeiIt is matched with finger print information S.
Positioning stage use K- nearest neighbor algorithm, that is, find out k anchor node for meeting threshold values d, then utilize centroid algorithm or
The centroid algorithm of person's weighting, the mass center or weighted mass center coordinate for calculating k anchor node are as location estimation.
Fingerprint base is distributed storage, and according to different indoor environment and application requirement, decision is stored in how many neighbours
It occupies in anchor node.
Often dynamic updates the fingerprint base in anchor node, and the characteristic of autonomous production fingerprint base information is capable of using anchor node,
According to different environmental quality and handling characteristics, the signal strength of anchor node periodically acquisition surrounding anchor node generates fingerprint base
Information.
The fingerprint matching in fingerprint matching stage further includes steps of the finger print information S acquired in real time and finger first
Line library information fiIt is poor to make, and is standard particularly as being according to the anchor node ID in real-time acquisition finger print information S, searches fingerprint base information
fiIn whether have corresponding anchor node, if so, signal strength RSSI value both then make it is poor；If it is not, signal is strong
Degree RSSI value is considered as 0, and it is poor then to make, and finally compares the similitude of difference, if difference keeps approximately constant value, shows section undetermined
Point fingerprint and anchor node fingerprint are matched.
The present invention has the advantage that be with effect: 1) this method does not need complicated offline fingerprint base collection work, by
The RSSI of surrounding anchor node can be collected automatically every some cycles in anchor node, the maintenance workload of fingerprint database is also very
It is small；2) the positioning result precision of this method is comparable in positioning accuracy with traditional centralized location method.According to test result,
It is uniformly distributed in anchor node, and when medium density, two kinds of localization methods and survey of the real node position position error within 2 meters
Example on probation is all 85% or more；3) complicated lower algorithm is calculated since fingerprint matching and location algorithm all use, and anchor node
Between the data volume transmitted it is small, so anchor node small power consumption, and system is stablized, this wireless biography stringenter for power consumption requirements
Sensor network (such as wireless network of ZigBee node composition) is of practical significance；4) finger print information distributed storage is in anchor node
In, it ensure that certain redundancy, the robustness of algorithm can be improved.It, still can benefit even if currently there is certain anchor nodes to fail
Nodes of locations is positioned with the finger print information, to be held in position the positioning accuracy of algorithm.
Detailed description of the invention
Fig. 1 is the wireless fingerprint positioning basic flow chart in the present invention；
Fig. 2 a, Fig. 2 b are the fingerprint letter structures pair in conventional fingerprint location fingerprint library message structure and the present invention in anchor node
Than figure；
Fig. 3 is the communication process figure of positioning node and surrounding anchor node in the present invention；
Fig. 4 is that the positioning robustness in different fingerprint positioning methods, under anchor node failure conditions compares figure.
Specific embodiment
Fig. 1 is overview flow chart of the present invention, and basic step is as follows:
1. fingerprint base establishment stage:
Fingerprint base collection step is still used in the present invention, but the Radio Map of this method no longer uses reference point, and
It is to collect the signal strength RSSI value of neighbouring anchor node in each anchor node, and result is redundantly distributed to k anchor of surrounding
On node.Without carrying out offline information collection, but each anchor node sends request message to neighbours' anchor node, each receives
The neighbor node of request calculates signal strength, then returns result to the anchor node for sending request, sends the anchor node of request
Generate fingerprint base information fi, fingerprint base information fiIt is stored directly in the node and neighbours' anchor node.
Often dynamic updates the fingerprint base in anchor node, and the characteristic of autonomous production fingerprint base information is capable of using anchor node,
According to different environmental quality and handling characteristics, the signal strength of anchor node periodically acquisition surrounding anchor node generates fingerprint base
Information, a series of problems that can thus overcome environmental change, joint behavior variation caused.
Fingerprint base is distributed storage, and according to different indoor environment and application requirement, decision is stored in how many neighbours
It occupies in anchor node.If the redundancy of fingerprint base is bigger, the robustness of location algorithm is better, but will increase wireless sense network
In the traffic and occupy a large amount of node storage space.
2. the fingerprint matching stage:
A) node ID to be positionediAnchor node peripherad first sends Location Request, and anchor node receives request and then calculates
Then self-position coordinate, signal strength information and mark number are sent to ID by the signal strength of the nodei, node to be positioned
IDiGenerate real time fingerprint information S.
B) node ID to be positionediFinger print information S is broadcast to neighbouring anchor node, by anchor node on anchor node
In fingerprint base information fiIt is matched with finger print information S.
C) first by the finger print information S acquired in real time and fingerprint base information fiMake it is poor, particularly as be according in real time acquire fingerprint
Anchor node ID in information S is standard, searches finger print information fiIn whether have corresponding anchor node, if so, letter both then
It is poor that number intensity RSSI value is made；If it is not, signal strength RSSI value is considered as 0, it is poor then to make.
D) finally compare the similitude of difference, if difference keeps approximately constant value, show node fingerprint undetermined and anchor section
Point fingerprint is matched.The basic skills for evaluating this difference consistency is to calculate the variance of above-mentioned RSSI difference, or calculate
The absolute value of these differences and these difference average values and.
The present invention uses distributed fingerprint matching algorithm, and to reduce computation complexity, simple fingerprint matching can be used
Algorithm.
3. positioning stage:
Fairly simple location algorithm is selected, is positioned using nearest neighbor method or K- nearest neighbour method.Nearest neighbor method, directly
Select most matched anchor node, location estimation of the position as positioning node.K- nearest neighbor algorithm finds out k for meeting threshold values d
Anchor node, then using centroid algorithm or the centroid algorithm of weighting, the mass center or weighted mass center coordinate for calculating k anchor node are made
For location estimation.
Fig. 2 a is that conventional fingerprint positioning mode is collecting fingerprint stage, the data format of fingerprint base.In conventional method, each
At reference point, information is sent using the peripherad anchor node of a mobile sensor node, collects the radio of surrounding anchor node
Signal strength forms finger print information, stores the RID into central databaseiIndicate the identification number of each reference point.IDiIndicate attached
The identification number of close anchor node.RSSIiIndicate the identification number ID searched from reference point locationsiAnchor node signal strength.
Fig. 2 b is the finger print information structure for the method for the present invention, wherein IDjIndicate the identification number of anchor node.IDiNear indicating
Anchor node identification number.RSSIiIndicate neighbours' anchor node (identification number IDi) identification number is arrived as IDjAnchor node signal it is strong
Degree.<xi,yi> expression identification number is IDiAnchor node coordinate, as unit of grid.
Fig. 3 is the peripherad anchor section of the communication process between anchor node and positioning node, first positioning node in the present invention
Point sends Location Request, and anchor node receives request and then calculates the signal strength of the node, then signal strength information, itself
Position coordinates and mark number are sent to IDi.Unknown node generates finger print information S, and structure is as shown in Figure 2 b, and then positioning node will
Finger print information S broadcasts to neighbouring anchor node, by the fingerprint base information f in anchor nodeiIt is matched with finger print information S.
After the completion of matching, greater than the anchor node of threshold value d, anchor node ID itselfiWith the difference variance being calculated in matching step or
It absolute difference and passes positioning node back, carries out location algorithm operation.
Fig. 4 is the method in traditional centralized fingerprint method and the present invention, and there are the positioning accuracies in the case of failure node
Compare.Fingerprint information-distribution type is stored in anchor node in the present invention, has certain redundancy, the robust of algorithm can be improved
Property.In figure, k=1 indicates that fingerprint is merely stored in current anchor node, and redundancy is not present, just can not benefit after anchor node failure
With the finger print information in the anchor node；K=3 indicates that fingerprint is not merely stored in current anchor node, and is stored in two outer neighbours
In anchor node；K=6 indicates that fingerprint will be also stored in other 5 neighbours' anchor nodes.In this way, even if current anchor node failure, according to
It so can use the finger print information to position nodes of locations, to be held in position the positioning accuracy of algorithm.It can from figure
Out, in same percentage of failures, K value is bigger, and positioning accuracy is higher, for example, tradition is fixed under 50% failure conditions
Position method, error is up to 10 meters, and this method, and at 7 meters hereinafter, in K=6, positioning accuracy error only has 5 meters, it is seen that this
Method robustness improves a lot.
Claims (3)
1. a kind of distributed wireless fingerprint positioning method for exempting from off-line training, including fingerprint base establishment stage, fingerprint matching stage
And positioning stage, it is characterised in that: without carrying out offline information collection, but each anchor node is asked to the transmission of neighbours' anchor node
Message is sought, the neighbor node for each receiving request calculates signal strength, then returns result to the anchor node for sending request, hair
The anchor node of request is sent to generate fingerprint base information fi, fingerprint base information fiIt is stored directly in the node and neighbours' anchor node, passes through
Often dynamic updates the fingerprint base in anchor node, the characteristic of autonomous production finger print information is capable of using anchor node, according to different rings
Border feature and handling characteristics, the signal strength of anchor node periodically acquisition surrounding anchor node, generate fingerprint base information；
The fingerprint matching stage uses distributed fingerprint matching algorithm: node ID to be positionediIt is fixed that anchor node peripherad first is sent
Position request, anchor node receives the signal strength that request then calculates the node, then signal strength information, self-position coordinate
ID is sent to mark numberi, node generation finger print information S to be positioned, then node to be positioned broadcasts to finger print information S
Neighbouring anchor node, by the fingerprint base information f in anchor node on anchor nodeiIt is matched with finger print information S；
The fingerprint matching in fingerprint matching stage further includes steps of the finger print information S and fingerprint base that will first acquire in real time
Information fiIt is poor to make, and is standard particularly as being according to the anchor node ID in real-time acquisition finger print information S, searches fingerprint base information fiIn
Whether corresponding anchor node is had, if so, then the signal strength RSSI value work of the two is poor；If it is not, by signal strength
RSSI value is considered as 0, and it is poor then to make, and finally compares the similitude of difference, if difference keeps approximately constant value, shows node undetermined
Fingerprint is matched with anchor node fingerprint；
The fingerprint location stage uses arest neighbors or K- nearest neighbor algorithm.
2. a kind of distributed wireless fingerprint positioning method for exempting from off-line training as described in claim 1, it is characterised in that described
K- nearest neighbor algorithm finds out k anchor node for meeting threshold value d, then using centroid algorithm or the centroid algorithm of weighting, calculate
The mass center or weighted mass center coordinate of k anchor node are as location estimation.
3. a kind of distributed wireless fingerprint positioning method for exempting from off-line training as described in claim 1, it is characterised in that fingerprint
Library is distributed storage, and according to different indoor environment and application requirement, decision is stored in how many a neighbours' anchor nodes.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610163617.4A CN105682224B (en) | 2016-03-22 | 2016-03-22 | A kind of distributed wireless fingerprint positioning method for exempting from off-line training |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610163617.4A CN105682224B (en) | 2016-03-22 | 2016-03-22 | A kind of distributed wireless fingerprint positioning method for exempting from off-line training |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105682224A CN105682224A (en) | 2016-06-15 |
| CN105682224B true CN105682224B (en) | 2019-07-05 |
Family
ID=56311339
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610163617.4A Expired - Fee Related CN105682224B (en) | 2016-03-22 | 2016-03-22 | A kind of distributed wireless fingerprint positioning method for exempting from off-line training |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105682224B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106162529B (en) * | 2016-07-08 | 2020-02-14 | 北京邮电大学 | Indoor positioning method and device |
| CN106535097B (en) * | 2016-11-14 | 2019-11-08 | 安徽华米信息科技有限公司 | Indoor orientation method, device and server |
| CN108377546A (en) * | 2016-11-21 | 2018-08-07 | 中兴通讯股份有限公司 | A kind of indoor orientation method and device, server, user equipment |
| WO2018204200A1 (en) | 2017-05-05 | 2018-11-08 | Pcms Holdings, Inc. | Privacy-preserving location based services |
| CN108710104A (en) * | 2018-03-15 | 2018-10-26 | 内蒙古科技大学 | The method and system of object for coal mine down-hole tunnel positioned in real time |
| CN108834045B (en) * | 2018-05-31 | 2020-06-23 | 北京邮电大学 | Positioning method and device based on positioning model |
| CN109195123B (en) * | 2018-08-22 | 2020-10-30 | 普联技术有限公司 | Fingerprint information updating method, device, storage medium and system in indoor positioning |
| CN109905840B (en) * | 2019-03-18 | 2021-04-20 | 成都西加云杉科技有限公司 | Positioning method, device and system and wireless access equipment |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103905992A (en) * | 2014-03-04 | 2014-07-02 | 华南理工大学 | Indoor positioning method based on wireless sensor networks of fingerprint data |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102395198A (en) * | 2011-11-04 | 2012-03-28 | 上海电机学院 | Signal intensity-based node positioning method and device for wireless sensing network |
| CN103384358A (en) * | 2013-06-25 | 2013-11-06 | 云南大学 | Indoor positioning method based on virtual reality and WIFI space field strength |
| CN104540219B (en) * | 2014-12-29 | 2018-05-01 | 北京工业大学 | A kind of Wi-Fi fingerprint indoor orientation methods of low complex degree |
-
2016
- 2016-03-22 CN CN201610163617.4A patent/CN105682224B/en not_active Expired - Fee Related
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103905992A (en) * | 2014-03-04 | 2014-07-02 | 华南理工大学 | Indoor positioning method based on wireless sensor networks of fingerprint data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105682224A (en) | 2016-06-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105682224B (en) | A kind of distributed wireless fingerprint positioning method for exempting from off-line training | |
| Dong et al. | A calibration-free localization solution for handling signal strength variance | |
| Yang et al. | Locating in fingerprint space: Wireless indoor localization with little human intervention | |
| CN103476118B (en) | A kind of WLAN indoor location fingerprint positioning method for monitoring in real time | |
| CN102546059B (en) | Non-supervision clustering-based distributed cooperative spectrum sensing method for cognitive self-organizing network | |
| CN103747419B (en) | A kind of indoor orientation method based on signal strength difference and dynamic linear interpolation | |
| CN105636201B (en) | Indoor orientation method based on sparse signal fingerprint database | |
| KR100948412B1 (en) | Method and system for estimating the location using a Received Signal Strength Indication | |
| KR102129400B1 (en) | Radio map construction method | |
| CN105243844A (en) | Road state identification method based on mobile phone signal | |
| CN103533647A (en) | Radio frequency map self-adaption positioning method based on clustering mechanism and robust regression | |
| CN103052128A (en) | Wireless sensor network-based energy-efficient collaborative scheduling method | |
| CN103648164A (en) | Time difference of arrival and Gossip algorithm based wireless sensor network distributed positioning method | |
| CN106793085A (en) | Fingerprint positioning method based on normality assumption inspection | |
| Du et al. | KF-KNN: Low-cost and high-accurate FM-based indoor localization model via fingerprint technology | |
| CN113207089A (en) | Position fingerprint positioning method based on CSI and crowdsourcing migration self-calibration updating | |
| CN109541537B (en) | Universal indoor positioning method based on ranging | |
| Le et al. | Received signal strength based indoor positioning algorithm using advanced clustering and kernel ridge regression | |
| CN111405461A (en) | Wireless indoor positioning method for optimizing equal-interval fingerprint sampling number | |
| Sitanayah et al. | Heuristic algorithm for finding boundary cycles in location-free low density wireless sensor networks | |
| Xiang et al. | Crowdsourcing-based indoor localization with knowledge-aided fingerprint transfer | |
| Zhou et al. | IMLours: Indoor mapping and localization using time-stamped WLAN received signal strength | |
| Pourahmadi et al. | Indoor positioning and distance-aware graph-based semi-supervised learning method | |
| CN103442431B (en) | Fingerprint positioning method and server | |
| CN108521631A (en) | A kind of mobile AP recognition methods towards indoor positioning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20190705 Termination date: 20200322 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |