CN109444813A - A kind of RFID indoor orientation method based on BP and DNN amphineura network - Google Patents
A kind of RFID indoor orientation method based on BP and DNN amphineura network Download PDFInfo
- Publication number
- CN109444813A CN109444813A CN201811255127.2A CN201811255127A CN109444813A CN 109444813 A CN109444813 A CN 109444813A CN 201811255127 A CN201811255127 A CN 201811255127A CN 109444813 A CN109444813 A CN 109444813A
- Authority
- CN
- China
- Prior art keywords
- rssi
- network
- path loss
- dnn
- loss coefficient
- 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
-
- 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
- G01S5/0278—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 involving statistical or probabilistic considerations
-
- 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
- G06N3/045—Combinations of networks
-
- 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
- G06N3/084—Backpropagation, e.g. using gradient descent
Abstract
A kind of RFID indoor orientation method based on BP and DNN amphineura network, compensate for it is traditional based on path loss coefficient n is set as constant under same environment in RSSI indoor positioning technologies the shortcomings that, by combining nerual network technique, establish the transformation model of signal strength Yu path loss coefficient, Accurate Prediction goes out the path loss coefficient n of different location, the error that tradition is fixed based on RSSI localization method path loss coefficient n and generated is reduced, system accuracy is improved;In conjunction with BP network and deep neural network, the path loss coefficient n for the different location that BP network is exported, and input of the tag signal strength to be measured received as DNN, path loss coefficient n can be answered according to different environment output phasies, to more accurately predict the coordinate of label to be measured, and improve system robustness;Tag coordinate to be measured is exported in conjunction with deep neural network, mark label can be treated in real time and positioned, the disadvantage of conventional mapping methods real-time difference is overcome.
Description
Technical field
The invention belongs to wirelessly communicate field of locating technology, specifically propose a kind of for passing through combination mind in indoor environment
Through network technology, to improve the RFID indoor orientation method based on BP and DNN amphineura network of positioning accuracy.
Background technique
Research combines the wireless technology of the characteristics of indoor environments and existing maturation both at home and abroad, proposes a variety of solution party
Case.According to the difference of perception environmental parameter and data acquisition modes, indoor positioning can be divided into seven classes, be based on WiFi skill respectively
Art, based on Bluetooth technology, based on ZigBee technology, based on ultra-wide band technology, based on ultrasonic technology, based on infrared technology with
And the indoor positioning based on RFID technique.Radio frequency identification (Radio Frequency Identification, RFID) is a kind of
Using radiofrequency signal automatic identification echo signal object and obtain the technology of relevant information.It is non-as possessed by RFID technique to connect
Touching, non line of sight, identification is without manual intervention and can identify multiple labels simultaneously, while having positioning accuracy swift and convenient to operate
The advantages that high, becomes preferred indoor positioning technologies.
There are two types of indoor orientation methods based on RFID, is distance measuring method and non-ranging method respectively.Determined based on ranging
Position method mainly includes arrival time (TOA), reaching time-difference (TDOA), angle of arrival (AOA) and received signal strength (RSSI)
Four kinds of methods include scene analysis positioning and degree of approximation positioning based on non-ranging localization method.Due to the positioning based on RSSI
Method ranging is simple, it has also become most common indoor orientation method, but that there are positioning accuracies is not high, it is affected by environment larger etc. to ask
Topic.
Summary of the invention
The present invention proposes a kind of RFID localization method of combination nerual network technique, solves the existing interior based on RFID
Influence of the complicated indoor environment to positioning in location technology, and improve positioning accuracy, overcome traditional location algorithm real-time
The problem of difference.
A kind of RFID indoor orientation method based on BP and DNN amphineura network, it is characterised in that: comprise the steps of:
Step 1: by reader and pre-prepd RFID reference label according to the actual conditions in indoor positioning region according to
Certain regular arrangement, the distance of measurement reference label to reader, the signal for recording each reader reception RFID reference label are strong
The coordinate of angle value RSSI and label position obtain original training data collection;
Step 2: noise suppression preprocessing is carried out to collected signal strength indication RSSI;The company that the same reader receives
Continue multiple RSSI signals and meet Gaussian Profile, in order to reduce interference, excludes error caused by small probability event, pass through Gauss model
The RSSI value in maximum probability section is chosen as valid data, then asks its arithmetic mean of instantaneous value as the output of filtering;
Step 3: the RSSI value received according to the distance of the reference label measured to reader and reader, by it
It substitutes into ranging formula, calculates path loss coefficient n;
Step 4: filtered received signal strength indication RSSI and path loss coefficient n are concentrated in together, constructed new
The training set of BP neural network model concentrates in together RSSI and path loss coefficient n and tag coordinate to be measured, and building is new
DNN model training set;
Step 5: training BP neural network;The BP network is three-decker, using filtered RSSI value as BP network
Input, path loss coefficient niAs output, BP network model;
Step 6: training deep neural network;The environment loss factor n that filtered RSSI value and BP network are exportedi
As the input of deep neural network, the actual coordinate of reference label is as output, training deep neural network model;
Step 7: predicted path loss factor in BP network model is input to after the RSSI that reader receives is filtered
ni, then it is input in DNN model with RSSI, predict tag coordinate to be measured in real time online.
Further, in the step 2, specific denoising step are as follows:
Firstly, RSSI value obeys (0, σ2) Gaussian Profile, probability density function is shown below:
Wherein,
Then section (μ-σ < RSSIk< μ+σ) new probability formula it is as follows:
P (μ-σ < RSSIk< μ+σ)=F (μ+σ)-F (μ-σ)=φ (1)-φ (- 1)=0.6826
Calculate the arithmetic mean of instantaneous value of these RSSI values:
Wherein, N is the number for meeting the signal strength indication in maximum probability section in continuous measurement k times.
Further, in the step 3, the ranging formula are as follows:
RSSI (d)=RSSI (d0)-10nlgd
Further, very difficult due to acquiring mass data under actual conditions in the step 5, it is rolled over using k-
Cross-validation method is trained, and collected data set D is divided into k size similar exclusive subsets, a subset is protected
It gives over to as test set, other k-1 subset repeats k time as training set, cross validation, the average value of k test result of return.
Compared with the conventional method, the invention has the following advantages:
(1) method proposed by the present invention compensate for it is traditional based in RSSI indoor positioning technologies on the road same environment Xia Jiang
The shortcomings that diameter loss factor n is set as constant establishes turn of signal strength and path loss coefficient by combining nerual network technique
Change model, Accurate Prediction goes out the path loss coefficient n of different location, reduces tradition based on RSSI localization method path loss
Coefficient n is fixed and the error that generates, improves the positioning accuracy of system.
(2) method combination BP network proposed by the present invention and deep neural network, the different location that BP network is exported
Path loss coefficient n, and the input of the tag signal strength to be measured that receives as DNN, can be defeated according to different environment
Corresponding path loss coefficient n out to more accurately predict the coordinate of label to be measured, and improves the robustness of system.
(3) combination deep neural network proposed by the present invention exports tag coordinate to be measured, can treat mark label in real time
It is positioned, overcomes the disadvantage of conventional mapping methods real-time difference.
Detailed description of the invention
Fig. 1 is the flow diagram of the method for the present invention.
Fig. 2 is the indoor locating system room top view based on the method for the present invention.
Fig. 3 is the structural schematic diagram of BP neural network.
Fig. 4 is the structural schematic diagram of deep neural network.
Specific embodiment
Technical solution of the present invention is described in further detail with reference to the accompanying drawings of the specification.
A kind of RFID indoor orientation method based on BP and DNN amphineura network proposed by the present invention, key point are this
Method combines tradition based on the indoor orientation method of RSSI with neural network, can be fitted using neural network any one
The characteristics of a continuous functional relation, the functional relation being fitted between RSSI and path loss coefficient n and tag coordinate to be measured.
It specifically includes that off-line phase arranges reader and label in the room, carries out data acquisition and data de-noising pretreatment, obtain
Training dataset constructs simultaneously BP network model and deep neural network model;Online collect is read after training model
Signal strength indication RSSI and n the input DNN model that device receives are read, tag coordinate is finally obtained.
Description of specific embodiments of the present invention by taking indoor environment shown in Fig. 2 as an example.
Step 1: progress environment arrangement first acquires data.Positioning system designed by the present invention includes reader,
RFID label tag and terminal.Four readers are placed at four angles respectively in the room, are prepared reference label and are placed on room
At 50 reference points being inside uniformly arranged in advance.Four readers obtain to each reference label continuous sampling 30 times respectively
The signal strength indication RSSI of i labelk I, j, it is denoted as Ri, wherein i=1,2,3 ..., 50, j=1,2,3,4, k=1,2,3 ...,
30, and record the coordinate P of corresponding i-th of labeli(xi, yi).(signal strength indication of each label is united with coordinate,
Obtain original training data collection the D={ (R of Noise1,, P1),(R2, P2) ... (Ri, Pi)}。)
Step 2: noise suppression preprocessing is carried out to collected signal strength indication RSSI.The company that the same reader receives
Continue multiple RSSI signals and meet Gaussian Profile, in order to reduce interference, excludes error caused by small probability event, pass through Gauss model
The RSSI value in maximum probability section is chosen as valid data, the relationship of signal strength indication and Gaussian function is as follows:
Wherein,
RSSIk i,jFor k-th of signal strength indication of i-th of label that j-th of reader is read, i=1,2,3 ...,
50, j=1,2,3,4, k=1,2,3 ..., 30.
It takes the received signal strength indication in maximum probability section to retain as valid data, then has to all in k measurement
Effect data are averaged, and process is as follows:
Wherein m is the number for meeting the signal strength indication in maximum probability section in continuous measurement k times, D 'i,jIt indicates i-th
The average value for the filtered signal strength indication RSSI that label is read by j-th of reader remembers R 'i={ D 'i,1, D 'i,2,
D’i,3, D 'I, 4}。
Step 3: ten are chosen after filtering apart from reader d0The signal strength indication RSSI of reference label at=1m, i.e.,
D’d, d=1,2 ..., 10 calculates its average valuePath loss coefficient n can be acquired by formulai, it is as follows:
Step 4: Offline training data collection is established.The data set that high quality can be obtained according to step 2 will obtain after filtering
Signal strength indication RSSI and path loss coefficient n put together, obtain the training dataset D ' of BP neural network1={ (R '1,
n1),(R’2, n2) ..., (R 'i, ni), by filtered signal strength indication RiWith path loss coefficient niWith tag coordinate PiConnection
It is combined to obtain the training dataset D ' of new deep neural network2={ (R '1, n1, P1), (R '2, n2,, P2) ... (R 'i,
ni, Pi), i=1,2,3 ..., 50.
Step 5: BP network model.After obtaining new training dataset, cross-validation method is rolled over using k- to carry out
Training.By data set D '1It is divided into 10 parts, a copy of it regards test set, other 9 parts are used as training set, repeats to divide and train
Model 10 times, the smallest model of extensive error can be obtained, to obtain the different path loss coefficients under different environment
N avoids influence of the environment to positioning accuracy.
Step 6: training DNN network model.It is similar with training BP neural network model, obtain the data set D ' of DNN2It
Afterwards, it is trained using k- folding cross-validation method, Optimized model parameter obtains optimal DNN model, which ensure that positioning
Real-time, effectively avoid the position error generated by indoor environment complicated and changeable.
Step 7: tag coordinate to be measured is predicted in real time online.After thering is RFID label tag to enter reader receiving area, warp
The data prediction for crossing step 3 obtains filtered signal strength indication RSSI, is entered into trained BP neural network mould
It is input in DNN, i.e., exportable label to be measured by type, outgoing route loss factor n with filtered signal strength indication RSSI
Accurate coordinates.
The foregoing is merely better embodiment of the invention, protection scope of the present invention is not with above embodiment
Limit, as long as those of ordinary skill in the art's equivalent modification or variation made by disclosure according to the present invention, should all be included in power
The protection scope recorded in sharp claim.
Claims (4)
1. a kind of RFID indoor orientation method based on BP and DNN amphineura network, it is characterised in that: comprise the steps of:
Step 1: by reader and pre-prepd RFID reference label according to the actual conditions in indoor positioning region according to certain
Regular arrangement, the distance of measurement reference label to reader record the signal strength indication that each reader receives RFID reference label
The coordinate of RSSI and label position obtain original training data collection;
Step 2: noise suppression preprocessing is carried out to collected signal strength indication RSSI;The same reader receives continuous more
A RSSI signal meets Gaussian Profile, in order to reduce interference, excludes error caused by small probability event, is chosen by Gauss model
RSSI value in maximum probability section is as valid data, then asks its arithmetic mean of instantaneous value as the output of filtering;
Step 3: the RSSI value received according to the distance of the reference label measured to reader and reader is substituted into
In ranging formula, path loss coefficient n is calculated;
Step 4: filtered received signal strength indication RSSI and path loss coefficient n are concentrated in together, and construct new BP mind
RSSI and path loss coefficient n and tag coordinate to be measured are concentrated in together, are constructed new by the training set through network model
The training set of DNN model;
Step 5: training BP neural network;The BP network is three-decker, using filtered RSSI value as the defeated of BP network
Enter, path loss coefficient niAs output, BP network model;
Step 6: training deep neural network;The environment loss factor n that filtered RSSI value and BP network are exportediAs depth
The input of neural network is spent, the actual coordinate of reference label is as output, training deep neural network model;
Step 7: predicted path loss factor n in BP network model is input to after the RSSI that reader receives is filteredi, then will
It is input in DNN model with RSSI, predicts tag coordinate to be measured in real time online.
2. a kind of RFID indoor orientation method based on BP and DNN amphineura network according to claim 1, feature exist
In: in the step 2, specific denoising step are as follows:
Firstly, RSSI value obeys (0, σ2) Gaussian Profile, probability density function is shown below:
Wherein,
Then section (μ-σ < RSSIk< μ+σ) new probability formula it is as follows:
P (μ-σ < RSSIk< μ+σ)=F (μ+σ)-F (μ-σ)=φ (1)-φ (- 1)=0.6826
Calculate the arithmetic mean of instantaneous value of these RSSI values:
Wherein, N is the number for meeting the signal strength indication in maximum probability section in continuous measurement k times.
3. a kind of RFID indoor orientation method based on BP and DNN amphineura network according to claim 1, feature exist
In: in the step 3, the ranging formula are as follows:
RSSI (d)=RSSI (d0)-10n lg d
4. a kind of RFID indoor orientation method based on BP and DNN amphineura network according to claim 1, feature exist
In: in the step 5, due under actual conditions acquire mass data it is very difficult, using k- folding cross-validation method come into
Row training, is divided into the similar exclusive subsets of k size for collected data set D, and a subset is retained as test set,
His k-1 subset repeats k time as training set, cross validation, the average value of k test result of return.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811255127.2A CN109444813B (en) | 2018-10-26 | 2018-10-26 | RFID indoor positioning method based on BP and DNN double neural networks |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811255127.2A CN109444813B (en) | 2018-10-26 | 2018-10-26 | RFID indoor positioning method based on BP and DNN double neural networks |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109444813A true CN109444813A (en) | 2019-03-08 |
CN109444813B CN109444813B (en) | 2022-10-18 |
Family
ID=65548691
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811255127.2A Active CN109444813B (en) | 2018-10-26 | 2018-10-26 | RFID indoor positioning method based on BP and DNN double neural networks |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109444813B (en) |
Cited By (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109996172A (en) * | 2019-03-14 | 2019-07-09 | 北京工业大学 | One kind being based on BP neural network precision indoor positioning system and localization method |
CN110320513A (en) * | 2019-07-05 | 2019-10-11 | 南京简睿捷软件开发有限公司 | A kind of production factors positioning system and method for large area workshop based on RFID |
CN110381580A (en) * | 2019-07-25 | 2019-10-25 | 上海开域信息科技有限公司 | A kind of WiFi localization method based on ratio optimization |
CN110888105A (en) * | 2019-10-09 | 2020-03-17 | 哈尔滨工程大学 | DOA estimation method based on convolutional neural network and received signal strength |
CN110909873A (en) * | 2019-10-08 | 2020-03-24 | 北京建筑大学 | Method and device for denoising passive RFID (radio frequency identification) mobile ranging data and computer equipment |
CN110990500A (en) * | 2019-03-29 | 2020-04-10 | 天维讯达(湖南)科技有限公司 | Propagation path model map establishing method and path loss determining method |
CN111031477A (en) * | 2019-12-25 | 2020-04-17 | 中山大学 | Millimeter wave indoor positioning method based on deep learning |
CN111523667A (en) * | 2020-04-30 | 2020-08-11 | 天津大学 | Neural network-based RFID (radio frequency identification) positioning method |
CN111724495A (en) * | 2020-04-30 | 2020-09-29 | 杭州任联科技有限公司 | Non-inductive attendance system and method based on dual-orientation gateway, RFID and BP neural network |
CN111741524A (en) * | 2019-06-04 | 2020-10-02 | 腾讯科技(深圳)有限公司 | Positioning method, positioning device, computer readable storage medium and computer equipment |
CN113203419A (en) * | 2021-04-25 | 2021-08-03 | 重庆大学 | Indoor inspection robot correction positioning method based on neural network |
CN113640740A (en) * | 2021-08-04 | 2021-11-12 | 成都诚骏科技有限公司 | Indoor high-precision positioning method for intelligent warehousing management system |
WO2022100952A1 (en) * | 2020-11-13 | 2022-05-19 | Telefonaktiebolaget Lm Ericsson (Publ) | Ue driven antenna tilt |
CN115113168A (en) * | 2022-08-25 | 2022-09-27 | 南京宇安防务科技有限公司 | Radar clutter suppression method based on neural network |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103945533A (en) * | 2014-05-15 | 2014-07-23 | 济南嘉科电子技术有限公司 | Big data based wireless real-time position positioning method |
US9569589B1 (en) * | 2015-02-06 | 2017-02-14 | David Laborde | System, medical item including RFID chip, data collection engine, server and method for capturing medical data |
CN106959444A (en) * | 2017-03-07 | 2017-07-18 | 上海工程技术大学 | A kind of RFID indoor locating systems and method based on artificial neural network |
CN107247260A (en) * | 2017-07-06 | 2017-10-13 | 合肥工业大学 | A kind of RFID localization methods based on adaptive depth confidence network |
-
2018
- 2018-10-26 CN CN201811255127.2A patent/CN109444813B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103945533A (en) * | 2014-05-15 | 2014-07-23 | 济南嘉科电子技术有限公司 | Big data based wireless real-time position positioning method |
US9569589B1 (en) * | 2015-02-06 | 2017-02-14 | David Laborde | System, medical item including RFID chip, data collection engine, server and method for capturing medical data |
CN106959444A (en) * | 2017-03-07 | 2017-07-18 | 上海工程技术大学 | A kind of RFID indoor locating systems and method based on artificial neural network |
CN107247260A (en) * | 2017-07-06 | 2017-10-13 | 合肥工业大学 | A kind of RFID localization methods based on adaptive depth confidence network |
Non-Patent Citations (3)
Title |
---|
吴超等: "基于BP神经网络的RFID室内定位算法研究", 《计算机仿真》 * |
费扬等: "基于BP神经网络模型的RSSI测距方法研究", 《电波科学学报》 * |
陈增强等: "基于模糊神经网络建模的RFID室内定位算法", 《系统科学与数学》 * |
Cited By (20)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109996172A (en) * | 2019-03-14 | 2019-07-09 | 北京工业大学 | One kind being based on BP neural network precision indoor positioning system and localization method |
CN110990500A (en) * | 2019-03-29 | 2020-04-10 | 天维讯达(湖南)科技有限公司 | Propagation path model map establishing method and path loss determining method |
CN110990500B (en) * | 2019-03-29 | 2023-12-15 | 天维讯达(湖南)科技有限公司 | Propagation path model map establishment method and path loss determination method |
CN111741524A (en) * | 2019-06-04 | 2020-10-02 | 腾讯科技(深圳)有限公司 | Positioning method, positioning device, computer readable storage medium and computer equipment |
CN110320513A (en) * | 2019-07-05 | 2019-10-11 | 南京简睿捷软件开发有限公司 | A kind of production factors positioning system and method for large area workshop based on RFID |
CN110381580B (en) * | 2019-07-25 | 2021-03-26 | 上海开域信息科技有限公司 | WiFi positioning method based on proportion optimization |
CN110381580A (en) * | 2019-07-25 | 2019-10-25 | 上海开域信息科技有限公司 | A kind of WiFi localization method based on ratio optimization |
CN110909873A (en) * | 2019-10-08 | 2020-03-24 | 北京建筑大学 | Method and device for denoising passive RFID (radio frequency identification) mobile ranging data and computer equipment |
CN110909873B (en) * | 2019-10-08 | 2022-07-22 | 北京建筑大学 | Method and device for denoising passive RFID (radio frequency identification) mobile ranging data and computer equipment |
CN110888105A (en) * | 2019-10-09 | 2020-03-17 | 哈尔滨工程大学 | DOA estimation method based on convolutional neural network and received signal strength |
CN111031477A (en) * | 2019-12-25 | 2020-04-17 | 中山大学 | Millimeter wave indoor positioning method based on deep learning |
CN111523667A (en) * | 2020-04-30 | 2020-08-11 | 天津大学 | Neural network-based RFID (radio frequency identification) positioning method |
CN111724495A (en) * | 2020-04-30 | 2020-09-29 | 杭州任联科技有限公司 | Non-inductive attendance system and method based on dual-orientation gateway, RFID and BP neural network |
CN111523667B (en) * | 2020-04-30 | 2023-06-27 | 天津大学 | RFID positioning method based on neural network |
WO2022100952A1 (en) * | 2020-11-13 | 2022-05-19 | Telefonaktiebolaget Lm Ericsson (Publ) | Ue driven antenna tilt |
CN113203419B (en) * | 2021-04-25 | 2023-11-10 | 重庆大学 | Indoor inspection robot correction positioning method based on neural network |
CN113203419A (en) * | 2021-04-25 | 2021-08-03 | 重庆大学 | Indoor inspection robot correction positioning method based on neural network |
CN113640740A (en) * | 2021-08-04 | 2021-11-12 | 成都诚骏科技有限公司 | Indoor high-precision positioning method for intelligent warehousing management system |
CN113640740B (en) * | 2021-08-04 | 2023-09-29 | 成都诚骏科技有限公司 | Indoor high-precision positioning method for intelligent warehouse management system |
CN115113168A (en) * | 2022-08-25 | 2022-09-27 | 南京宇安防务科技有限公司 | Radar clutter suppression method based on neural network |
Also Published As
Publication number | Publication date |
---|---|
CN109444813B (en) | 2022-10-18 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109444813A (en) | A kind of RFID indoor orientation method based on BP and DNN amphineura network | |
CN108989976B (en) | Fingerprint positioning method and system in intelligent classroom | |
CN105636197B (en) | Method for estimating distance and device and node positioning method and equipment | |
CN103945332B (en) | A kind of received signal strength and multi-path information united NNs indoor orientation method | |
CN105704652B (en) | Fingerprint base acquisition and optimization method in a kind of positioning of WLAN/ bluetooth | |
CN104330771B (en) | A kind of indoor RFID precise positionings method and apparatus | |
CN112533149B (en) | Moving target positioning algorithm based on UWB mobile node | |
CN105636201B (en) | Indoor orientation method based on sparse signal fingerprint database | |
CN104602342A (en) | IBeacon device based efficient indoor positioning method | |
CN108307767B (en) | Detection of obstacles obstacle avoidance system and method suitable for full-automatic weeder | |
CN105717483B (en) | A kind of location determining method and device based on multi-source positioning method | |
CN105120433A (en) | WLAN indoor positioning method based on continuous sampling and fuzzy clustering | |
CN106970379B (en) | Based on Taylor series expansion to the distance-measuring and positioning method of indoor objects | |
CN106028290A (en) | WSN multidimensional vector fingerprint positioning method based on Kriging | |
CN108668245A (en) | A kind of indoor orientation method based on WiFi Yu earth's magnetic field database | |
CN104053129A (en) | Wireless sensor network indoor positioning method and device based on sparse RF fingerprint interpolations | |
CN109640269A (en) | Fingerprint positioning method based on CSI Yu Time Domain Fusion algorithm | |
CN109672973A (en) | A kind of indoor positioning blending algorithm based on most strong AP method | |
CN103561380A (en) | Location fingerprint positioning method and device | |
CN104038901A (en) | Indoor positioning method for reducing fingerprint data acquisition workload | |
CN104507097A (en) | Semi-supervised training method based on WiFi (wireless fidelity) position fingerprints | |
CN108983204A (en) | A kind of keyless entry and without key activation system localization method | |
CN109375168A (en) | A kind of low speed move vehicle localization method based on RSSI | |
Siyang et al. | WKNN indoor Wi-Fi localization method using k-means clustering based radio mapping | |
CN112333642A (en) | Indoor positioning method based on channel state information |
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 |