CN110891241A - Fingerprint positioning method based on long-time memory network model and access point selection strategy - Google Patents

Fingerprint positioning method based on long-time memory network model and access point selection strategy Download PDF

Info

Publication number
CN110891241A
CN110891241A CN201911152184.2A CN201911152184A CN110891241A CN 110891241 A CN110891241 A CN 110891241A CN 201911152184 A CN201911152184 A CN 201911152184A CN 110891241 A CN110891241 A CN 110891241A
Authority
CN
China
Prior art keywords
access point
rss
fingerprint
data
network model
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
Application number
CN201911152184.2A
Other languages
Chinese (zh)
Inventor
费泽松
史新宇
郭婧
尹睿锐
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Chongqing Innovation Center of Beijing University of Technology
Original Assignee
Chongqing Innovation Center of Beijing University of Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Chongqing Innovation Center of Beijing University of Technology filed Critical Chongqing Innovation Center of Beijing University of Technology
Priority to CN201911152184.2A priority Critical patent/CN110891241A/en
Publication of CN110891241A publication Critical patent/CN110891241A/en
Pending legal-status Critical Current

Links

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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/29—Geographical information databases
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/049—Temporal neural networks, e.g. delay elements, oscillating neurons or pulsed inputs
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • 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
    • H04W4/023—Services making use of location information using mutual or relative location information between multiple location based services [LBS] targets or of distance thresholds
    • 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

Abstract

The invention discloses a fingerprint positioning method based on a long-time memory network model and an access point selection strategy, which at least comprises the following steps: s1: indoor RSS signal data acquisition, S3: finishing screening of access point data based on a preset judgment strategy; s4: using a sliding window to locally extract the characteristics of the fingerprint database; s5: and training the updated fingerprint database by adopting a long-time memory network model to obtain a trained network model, and realizing indoor positioning. The method adopts the access point selection strategy to reconstruct the database so as to reduce the scale of the database, utilizes the characteristic point extraction method to process the data selectively output by the AP so as to extract the characteristics with larger information quantity, inputs the extracted characteristics into the LSTM network model, achieves the effect of reducing the calculated quantity and simultaneously enhancing the noise robustness, solves the positioning problem in the indoor complex environment, and reduces the influence on the positioning precision in the non-line-of-sight transmission background.

Description

Fingerprint positioning method based on long-time memory network model and access point selection strategy
Technical Field
The invention belongs to the technical field of fingerprint positioning, and particularly relates to a fingerprint positioning method based on a long-time memory network model and an access point selection strategy.
Background
With the wide popularization and rapid development of wireless communication network technology, the demand of people for location services is sharply increased. Positioning services can be divided into outdoor positioning and indoor positioning according to different scenes of demands. The outdoor positioning can utilize a satellite high-precision wireless navigation positioning system, but is influenced by indoor complex environment and non-line-of-sight transmission, and the method is not suitable for indoor positioning, so that the method has wide interest in scientific research personnel on how to realize precise positioning in the indoor environment. Indoor positioning services and location-based applications (such as tracking, patient monitoring, navigation, and the like) are closely related to the current society and smart cities, and have become popular research fields in the fields of modern communication and information technology, however, the existing indoor positioning technology generally has the disadvantages of large positioning error and high calculation complexity, and cannot meet the requirement of people on indoor high-precision positioning.
To support indoor positioning services and overcome the limitations of satellite positioning systems, scientists have studied a number of approaches. The existing indoor positioning system mainly comprises three positioning methods, namely triangulation, angle measurement and signal strength measurement, wherein the signal strength measurement is also called a fingerprint positioning method. Triangulation methods utilize TOA/TDOA techniques; the angle measurement method uses the AOA technology; the signal strength measurement method utilizes the wireless received signal strength RSS to construct a fingerprint database. The specific implementation can be divided into two stages: collecting RSS signals transmitted by different APs at different reference points in an off-line stage, recording position coordinates of the reference points, and forming a fingerprint database by all the signals and the corresponding position coordinates; and training the database obtained in the off-line stage by a machine learning method in the on-line stage to construct a network model. When a group of unknown RSS signal intensity data is received, the original database can be matched in the trained network model, so that the reference point position coordinates corresponding to the unknown RSS data are obtained, and the indoor positioning effect can be realized. The fingerprint positioning has the advantages of high positioning precision, small influence of non-line-of-sight transmission, capability of fully utilizing the existing facilities, easiness in realization, no need of or only a few additional devices in the system, and small influence of upgrading and maintenance on users. The disadvantages are the high up-front workload and the unsuitability for areas where the environment changes too fast.
Existing fingerprint positioning techniques can be broadly divided into two aspects: on one hand, fingerprint databases are reconstructed by using a clustering algorithm, an access point selection algorithm and the like, and on the other hand, indoor positioning algorithms are improved, and traditional algorithms comprise a k-nearest neighbor algorithm (KNN), a convolutional neural network and other algorithms for positioning. However, the existing fingerprint positioning technology does not take both aspects into consideration. The invention aims to solve the technical defects, considers a novel long-short term memory model with the advantage of processing data according to time sequence, and provides a fingerprint positioning method based on an access point selection scheme and the long-short term memory model.
Disclosure of Invention
The invention aims to improve positioning accuracy by utilizing a machine learning algorithm and an access point selection algorithm, reduce calculated amount, enhance noise robustness, solve the positioning problem in an indoor complex environment, and reduce the influence on the positioning accuracy under a non-line-of-sight transmission background, thereby providing a fingerprint positioning method based on a long-time memory network model and an access point selection strategy.
The purpose of the invention is realized by the following technical scheme:
a fingerprint positioning method based on a long-time memory network model and an access point selection strategy at least comprises the following steps: s1: indoor RSS signal data acquisition; s3: finishing screening of access point data based on a preset judgment strategy; s4: using a sliding window to locally extract the characteristics of the fingerprint database; s5: and training the updated fingerprint database by adopting a long-time memory network model to obtain a trained network model, and realizing indoor positioning.
The invention adopts the access point selection strategy to reconstruct the database so as to reduce the scale of the database, utilizes the characteristic point extraction method to process the data selectively output by the AP so as to extract the characteristics with larger information quantity, inputs the extracted characteristics into the LSTM network model and realizes the indoor positioning with high precision and low complexity.
According to a preferred embodiment, the step S1, the acquiring indoor RSS data includes: q groups of sample data collected by each reference point in the N reference points, and M RSS signals from the access point included in each group of sample data; where the number of APs is M, the APs are denoted as AP1, AP 2.
According to a preferred embodiment, the screening of the access point data in step S3 specifically includes the following steps:
s3.1 traversing all reference points in indoor scene, and calculating RSS signal average value psi from different access points on each reference pointj：
Figure BDA0002283855670000031
Wherein, M represents the number of access points,
Figure BDA0002283855670000032
is the RSS signal, Ψ, for the ith access point, jth reference pointjRepresents the average of the RSS signals from the M access points at the jth reference point;
s3.2, replacing the original signal intensity with the RSS signal average value, namely updating the original database with the new fingerprint database Ψ j;
s3.3 calculating the Standard deviation of each Access Point
Figure BDA0002283855670000033
Figure BDA0002283855670000034
Where Q represents the number of RSS signal samples per access point received at each reference point, and the RSSjRepresenting the jth signal received at the lth reference point,
Figure BDA0002283855670000035
represents the average of Q signal samples.
S3.4 calculating the frequency of occurrence of different access points
Figure BDA0002283855670000041
Figure BDA0002283855670000042
Wherein N isiFor the number of times the ith access point appears in the fingerprint database,
Figure BDA0002283855670000043
is the number of all access points.
S3.5 calculating the stability of each Access Point
Figure BDA0002283855670000044
Figure BDA0002283855670000045
Wherein the content of the first and second substances,
Figure BDA0002283855670000046
indicating the frequency of occurrence of the ith access point,
Figure BDA0002283855670000047
represents the standard deviation of the ith access point, epsilon is a positive number approaching zero;
s3.6 are arranged from big to small
Figure BDA0002283855670000048
Selecting an access point corresponding to the first K values and RSS data corresponding to the access point to reconstruct the two-dimensional fingerprint database psiKDiscarding the remaining access points and their associated RSS data; therein ΨKIs N × Q, N being the number of reference points, Q being the number of RSS signal samples received at each reference point for each access point.
Step S3 optimizes the RSS value using a stable access point selection algorithm, reduces network traffic and data calculation generated during the node position estimation process, and improves the accuracy of positioning.
According to a preferred embodiment, the fingerprint location method further comprises a step S2 of preprocessing the collected RSS signal data.
According to a preferred embodiment, in step S2, the preprocessing of the collected RSS signal data includes: duplicate entries in the collected dataset are deleted.
According to a preferred embodiment, the step S2 further includes: the RSS signal strength values in the data set remaining after deleting duplicate entries in the data set that are not in the range of-40 dBm to-100 dBm are filled to 100 dBm.
According to a preferred embodiment, in step S4, the fingerprint database Ψ is accessedKThe localized extraction of features of (1) comprises: s3.1 fingerprint database Ψ with dimension size of NxQKDividing the access points into numgroup small groups, wherein the dimension size of each small group is M multiplied by q, M is the number of access points, and q is the number of samples contained in each small group; s3.2, calculating the maximum value and the minimum value of the RSS in each group and the quantiles of 25%, 50% and 75%, and using the data to complete the updating of the fingerprint database.
In step S3, the invention adopts the sliding window to extract local features from the original data, effectively processes the received value of the intensity of the complex RSS signal, and realizes the purposes of reducing the time consumption of positioning and improving the positioning precision.
The main scheme and the further selection schemes can be freely combined to form a plurality of schemes which are all adopted and claimed by the invention; in the invention, the selection (each non-conflict selection) and other selections can be freely combined. The skilled person in the art can understand that there are many combinations, which are all the technical solutions to be protected by the present invention, according to the prior art and the common general knowledge after understanding the scheme of the present invention, and the technical solutions are not exhaustive herein.
The invention has the beneficial effects that: the method adopts the access point selection strategy to reconstruct the database so as to reduce the scale of the database, utilizes the characteristic point extraction method to process the data selectively output by the AP so as to extract the characteristics with larger information quantity, inputs the extracted characteristics into the LSTM network model, achieves the effect of reducing the calculated quantity and simultaneously enhancing the noise robustness, solves the positioning problem in the indoor complex environment, and reduces the influence on the positioning precision in the non-line-of-sight transmission background.
Drawings
FIG. 1 is a flowchart of a fingerprint positioning method based on a long-and-short-term memory network model and an access point selection strategy according to the present invention;
FIG. 2 is a flowchart of an off-line phase and an on-line phase of a fingerprint positioning method based on a long-and-short-term memory network model and an access point selection strategy according to the present invention;
fig. 3 is a flowchart illustrating the operation of performing localized extraction on the features of the fingerprint database by using a sliding window in step S4 in the fingerprint positioning method based on the long-term memory network model and the access point selection policy according to the present invention.
Detailed Description
The embodiments of the present invention are described below with reference to specific embodiments, and other advantages and effects of the present invention will be easily understood by those skilled in the art from the disclosure of the present specification. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention. It is to be noted that the features in the following embodiments and examples may be combined with each other without conflict.
It should be noted that, in order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, 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 some embodiments of the present invention, but not all embodiments.
Thus, the following detailed description of the embodiments of the present invention is not intended to limit the scope of the invention as claimed, but is merely representative of selected embodiments of the invention. 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.
Example 1:
fig. 1 is a flowchart of a fingerprint positioning method based on a long-and-short-term memory network model and an access point selection policy according to the present invention, fig. 2 is a flowchart of an offline stage and an online stage of a fingerprint positioning method based on a long-and-short-term memory network model and an access point selection policy according to the present invention, and fig. 3 is a flowchart of a work flow of performing localized extraction on features of a fingerprint database using a sliding window in step S4 in a fingerprint positioning method based on a long-and-short-term memory network model and an access point selection policy according to the present invention.
The method is described in specific application scenarios based on fig. 1 to fig. 3, and the embodiment describes the method of the present invention applied to positioning in a shopping mall environment.
With the rapid development of social economy and the wide application of the internet of things, the indoor positioning technology is continuously promoted by the heat tide of science and technology, and the requirement of many public places such as hospitals, markets, warehouses and the like on indoor positioning urges the positioning technology to become more mature. In recent years, the market scale is gradually enlarged, the efficiency of emergency treatment on the security scheduling of the market by market managers is very low, the police resources of the market cannot be reasonably distributed, great inconvenience is brought to consumers, and the manpower resources of the market are wasted. Considering that the WIFI fingerprint positioning technology has high efficiency, universality, low consumption and higher accuracy for positioning services of large public places such as shopping malls, the invention provides the fingerprint positioning method based on the long-time memory network model and the access point selection strategy, which can improve the positioning accuracy in the shopping malls and realize the reasonable distribution of shopping mall resources.
The practice of the invention is generally divided into two stages: the first phase is an off-line phase, i.e., data is collected and the model is trained. The main work of the stage is to divide the market into 5 square grids with equal area, collect RSS signal intensity data of the central position of the divided square in the market, further form a position fingerprint database, each fingerprint information corresponds to a specific square grid, and then train a positioning model by using the method provided by the invention; the second phase is an on-line phase (user position determining phase), which determines the actual position of the user by determining the real-time data information (RSS signal strength data) of the user to be positioned through the receiver and then determining which data matches with the fingerprint database by adopting a corresponding configuration algorithm.
The implementation of the fingerprint positioning method based on the long-time memory network model and the access point selection strategy specifically comprises the following steps:
step S1: and (4) indoor RSS signal data acquisition. The collected data includes 96 reference points, 620 access points, which may be denoted as AP1，AP2，...，APMAnd acquiring 800 groups of sample data at each reference point, wherein each group of sample data contains 620 RSS signals from the access point.
Step S2: the collected RSS signal data is pre-processed. The specific operation is as follows: duplicate entries in the collected dataset are deleted. The remaining data set was examined and RSS signal strength values that were not in the range of-40 dBm to-100 dBm were filled to 100 dBm.
Step S3: and finishing the screening of the access point data based on a preset judgment strategy. Judging and retaining important APs, and discarding irrelevant APs at the same time, specifically:
step S3.1, traversing all reference points in the indoor scene, and calculating the RSS signal average value psi from 620 access points on each reference pointj：
Figure BDA0002283855670000081
Wherein the content of the first and second substances,
Figure BDA0002283855670000082
is the RSS signal of the ith access point, the jth reference point, ΨjRepresenting the average of the RSS signals from the 620 access points at the jth reference point.
Step S3.2 replaces the original signal strength with the RSS signal mean, i.e. with the new fingerprint database ΨjAnd updating the original database.
Step 3.3 calculate the standard deviation of each access point
Figure BDA0002283855670000083
Figure BDA0002283855670000084
Wherein the RSSjRepresenting the jth signal received at the lth reference point,
Figure BDA0002283855670000091
represents the average of 800 signal samples.
Step 3.4 calculating the frequency of occurrence of different access points
Figure BDA0002283855670000092
Figure BDA0002283855670000093
Wherein N isiIs the number of times the ith access point appears in the fingerprint database,
Figure BDA0002283855670000094
is the number of all access points.
Step 3.5 calculate the stability of each access point
Figure BDA0002283855670000095
Figure BDA0002283855670000096
Wherein the content of the first and second substances,
Figure BDA0002283855670000097
indicating the frequency of occurrence of the ith access point,
Figure BDA0002283855670000098
the standard deviation of the ith access point is represented, and epsilon is set to be 0.0001, so that the invalid condition that the denominator is 0 occurs in the stability calculation of the access point is prevented.
Step 3.6 is arranged from big to small
Figure BDA0002283855670000099
Selecting the access point corresponding to the first 50 values and the RSS data corresponding to the access point to reconstruct the two-dimensional fingerprint database ΨKThe remaining access points and their associated RSS data are discarded. Therein ΨKIs 96 × 800 × 620, 96 is the number of reference points, and 800 is the number of RSS signal samples received at each reference point for each access point.
Step S4, using sliding window to fingerprint database ΨKThe features of (2) are locally extracted.
Step 4.1 fingerprint database Ψ with dimension size of 96 × 800 × 620KThe size of each group is 40 × 620, wherein 600 is the number of access points, and 40 is the number of samples contained in each group.
Step 4.2 calculates the RSS maximum, minimum and 25%, 50%, 75% quantiles in each group and updates the fingerprint database with these data.
And step four, training the updated fingerprint database by adopting the long-time memory network model to obtain a mature network model, and when acquiring the user real-time data information (RSS signal intensity data) of the position to be measured, determining which data of the fingerprint database is matched with the user real-time data information by using the trained model so as to obtain the actual position of the user.
In summary, the fingerprint positioning method based on the long-and-short-term memory network model and the access point selection strategy provided by the invention aims to optimize an indoor positioning method, improve positioning accuracy and reduce positioning power consumption and cost. The access point selection scheme adopted by the invention reduces the scale of the original fingerprint database, reduces the influence of noise on the positioning precision, improves the positioning precision and simultaneously reduces the calculation complexity by utilizing the local feature extraction method, and finally realizes the indoor positioning service with high precision and low power consumption by utilizing the long-time memory model to match data.
At the same time, the validity of the proposed method is verified by using the open source database. The result shows that compared with other methods such as LSTM and KNN without AP selection strategy, the algorithm has the highest positioning precision and the shortest running time. That is, the present technology has better performance than other existing solutions.
The foregoing basic embodiments of the invention and their various further alternatives can be freely combined to form multiple embodiments, all of which are contemplated and claimed herein. In the scheme of the invention, each selection example can be combined with any other basic example and selection example at will. Numerous combinations will be known to those skilled in the art.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents and improvements made within the spirit and principle of the present invention are intended to be included within the scope of the present invention.

Claims (7)

1. A fingerprint positioning method based on a long-time memory network model and an access point selection strategy is characterized in that,
the fingerprint positioning method at least comprises the following steps:
s1: the indoor RSS signal data acquisition is performed,
s3: finishing screening of access point data based on a preset judgment strategy;
s4: using a sliding window to locally extract the characteristics of the fingerprint database;
s5: and training the updated fingerprint database by adopting a long-time memory network model to obtain a trained network model, and realizing indoor positioning.
2. The fingerprint positioning method according to claim 1, wherein the step S1 of collecting indoor RSS data includes: q groups of sample data collected by each reference point in the N reference points, and M RSS signals from the access point included in each group of sample data; where the number of access points is M, the access points are denoted as { AP1，AP2，...，APM}。
3. The method as claimed in claim 2, wherein the step S3 of screening the ap data includes the following steps:
s3.1 traversing all reference points in indoor scene, and calculating RSS signal average value psi from different access points on each reference pointj：
Figure FDA0002283855660000011
Wherein, M represents the number of access points,
Figure FDA0002283855660000012
is the RSS signal, Ψ, for the ith access point, jth reference pointjRepresents the average of the RSS signals from the M access points at the jth reference point;
s3.2 replace the original signal strength by the RSS signal mean, i.e. use the new fingerprint database ΨjUpdating the original database;
s3.3 calculating the Standard deviation of each Access Point
Figure FDA0002283855660000021
Figure FDA0002283855660000022
Where Q represents the number of RSS signal samples per access point received at each reference point, and the RSSjRepresenting the jth signal received at the lth reference point,
Figure FDA0002283855660000023
represents the average of Q signal samples.
S3.4 calculating the frequency of occurrence of different access points
Figure FDA0002283855660000024
Figure FDA0002283855660000025
Wherein N isiFor the number of times the ith access point appears in the fingerprint database,
Figure FDA0002283855660000026
is the number of all access points.
S3.5 calculating the stability of each Access Point
Figure FDA0002283855660000027
Figure FDA0002283855660000028
Wherein the content of the first and second substances,
Figure FDA0002283855660000029
indicating the frequency of occurrence of the ith access point,
Figure FDA00022838556600000210
denotes the standard deviation of the ith access point, ε is the approximateA positive number of zero;
s3.6 are arranged from big to small
Figure FDA00022838556600000211
Selecting an access point corresponding to the first K values and RSS data corresponding to the access point to reconstruct the two-dimensional fingerprint database psiKDiscarding the remaining access points and their associated RSS data; therein ΨKIs N × Q, N being the number of reference points, Q being the number of RSS signal samples received at each reference point for each access point.
4. The fingerprint positioning method according to claim 1, wherein the fingerprint positioning method further comprises a step S2 of preprocessing the collected RSS signal data.
5. The fingerprint positioning method according to claim 4, wherein the preprocessing of the collected RSS signal data in step S2 includes: duplicate entries in the collected dataset are deleted.
6. The fingerprint positioning method according to claim 5, wherein the step S2 further comprises: the RSS signal strength values in the data set remaining after deleting duplicate entries in the data set that are not in the range of-40 dBm to-100 dBm are filled to 100 dBm.
7. The method for fingerprint location based on long-and-short-term memory network model and access point selection strategy as claimed in claim 1, wherein in step S3, the fingerprint database Ψ is processedKThe localized extraction of features of (1) comprises:
s3.1 fingerprint database Ψ with dimension size of NxQKDividing the data into numcroup groups, wherein the dimension size of each group is Mxq, and M is access pointsNumber, q is the number of samples each subgroup contains;
s3.2, calculating the maximum value and the minimum value of the RSS in each group and the quantiles of 25%, 50% and 75%, and using the data to complete the updating of the fingerprint database.
CN201911152184.2A 2020-01-08 2020-01-08 Fingerprint positioning method based on long-time memory network model and access point selection strategy Pending CN110891241A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201911152184.2A CN110891241A (en) 2020-01-08 2020-01-08 Fingerprint positioning method based on long-time memory network model and access point selection strategy

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201911152184.2A CN110891241A (en) 2020-01-08 2020-01-08 Fingerprint positioning method based on long-time memory network model and access point selection strategy

Publications (1)

Publication Number Publication Date
CN110891241A true CN110891241A (en) 2020-03-17

Family

ID=69748374

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201911152184.2A Pending CN110891241A (en) 2020-01-08 2020-01-08 Fingerprint positioning method based on long-time memory network model and access point selection strategy

Country Status (1)

Country Link
CN (1) CN110891241A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113030853A (en) * 2021-03-07 2021-06-25 中国人民解放军陆军工程大学 RSS and AOA combined measurement-based multi-radiation source passive positioning method

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108717175A (en) * 2018-04-18 2018-10-30 同济大学 Indoor fingerprint positioning method based on region division and sparse support vector regression
US20190137621A1 (en) * 2017-06-02 2019-05-09 Apple Inc. Compressing Radio Maps Using Different Compression Models
CN110118560A (en) * 2019-05-28 2019-08-13 东北大学 A kind of indoor orientation method based on LSTM and Multi-sensor Fusion
CN110166930A (en) * 2019-04-03 2019-08-23 华中科技大学 A kind of indoor orientation method and system based on WiFi signal
CN110519693A (en) * 2019-09-29 2019-11-29 东北大学 A kind of fusion and positioning method towards intelligent mobile terminal
CN110602651A (en) * 2019-09-20 2019-12-20 北京智芯微电子科技有限公司 Positioning method based on WIFI position fingerprint and positioning system of robot

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20190137621A1 (en) * 2017-06-02 2019-05-09 Apple Inc. Compressing Radio Maps Using Different Compression Models
CN108717175A (en) * 2018-04-18 2018-10-30 同济大学 Indoor fingerprint positioning method based on region division and sparse support vector regression
CN110166930A (en) * 2019-04-03 2019-08-23 华中科技大学 A kind of indoor orientation method and system based on WiFi signal
CN110118560A (en) * 2019-05-28 2019-08-13 东北大学 A kind of indoor orientation method based on LSTM and Multi-sensor Fusion
CN110602651A (en) * 2019-09-20 2019-12-20 北京智芯微电子科技有限公司 Positioning method based on WIFI position fingerprint and positioning system of robot
CN110519693A (en) * 2019-09-29 2019-11-29 东北大学 A kind of fusion and positioning method towards intelligent mobile terminal

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
AYESHA SAHAR ET.AL: "An LSTM-based Indoor Positioning Method Using Wi-Fi Signals", 《ICVISP 2018: PROCEEDINGS OF THE 2ND INTERNATIONAL CONFERENCE ON VISION， IMAGE AND SIGNAL PROCESSING》 *
ZHENGHUA CHEN ET.AL: "WiFi Fingerprinting Indoor Localization Using Local Feature-Based Deep LSTM", 《IEEE SYSTEMS JOURNAL》 *
沈阳: "基于指纹的无线室内定位中接入点选择算法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113030853A (en) * 2021-03-07 2021-06-25 中国人民解放军陆军工程大学 RSS and AOA combined measurement-based multi-radiation source passive positioning method

Similar Documents

Publication Publication Date Title
Song et al. A novel convolutional neural network based indoor localization framework with WiFi fingerprinting
CN108989976B (en) Fingerprint positioning method and system in intelligent classroom
Song et al. Cnnloc: Deep-learning based indoor localization with wifi fingerprinting
CN108536851B (en) User identity recognition method based on moving track similarity comparison
CN105044662B (en) A kind of fingerprint cluster multi-point joint indoor orientation method based on WIFI signal intensity
CN108919177B (en) Positioning map construction method based on virtual information source estimation and track correction
CN105120479B (en) The signal intensity difference modification method of terminal room Wi-Fi signal
CN106102163B (en) WLAN fingerprint positioning method based on RSS linear correlation Yu secondary weighted centroid algorithm
CN109195110B (en) Indoor positioning method based on hierarchical clustering technology and online extreme learning machine
CN103763769B (en) Based on the indoor fingerprint positioning method that access point reselection procedure and self-adapting cluster divide
CN105120433A (en) WLAN indoor positioning method based on continuous sampling and fuzzy clustering
CN108566620B (en) Indoor positioning method based on WIFI
CN107677989B (en) A kind of indoor location localization method carrying out RSSI removal noise based on RSSI maximum value
CN108627798A (en) WLAN indoor positioning algorithms based on linear discriminant analysis and gradient boosted tree
Yang et al. Research on Wi-Fi indoor positioning in a smart exhibition hall based on received signal strength indication
Huang et al. MAPS: Indoor localization algorithm based on multiple AP selection
CN113543026B (en) Multi-floor indoor positioning method based on radial basis function network
WO2022242018A1 (en) Indoor target positioning method based on improved cnn model
CN109541537B (en) Universal indoor positioning method based on ranging
CN110891241A (en) Fingerprint positioning method based on long-time memory network model and access point selection strategy
CN105657653A (en) Indoor positioning method based on fingerprint data compression
CN114095853B (en) Method and device for generating indoor map
Jia et al. A fingerprint-based localization algorithm based on LSTM and data expansion method for sparse samples
Xiang et al. Crowdsourcing-based indoor localization with knowledge-aided fingerprint transfer
Zhang et al. Research and optimization of BLE fingerprint indoor positioning algorithm based on fusion clustering

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: 20200317