CN107437093B - Initial clustering center optimization selection method based on RFID data intensity-time distribution - Google Patents

Initial clustering center optimization selection method based on RFID data intensity-time distribution Download PDF

Info

Publication number
CN107437093B
CN107437093B CN201710545656.5A CN201710545656A CN107437093B CN 107437093 B CN107437093 B CN 107437093B CN 201710545656 A CN201710545656 A CN 201710545656A CN 107437093 B CN107437093 B CN 107437093B
Authority
CN
China
Prior art keywords
time
intensity
clustering
data
span
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.)
Active
Application number
CN201710545656.5A
Other languages
Chinese (zh)
Other versions
CN107437093A (en
Inventor
孙棣华
赵敏
郑林江
朱文霖
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.)
Liyang Smart City Research Institute Of Chongqing University
Original Assignee
Chongqing University
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 University filed Critical Chongqing University
Priority to CN201710545656.5A priority Critical patent/CN107437093B/en
Publication of CN107437093A publication Critical patent/CN107437093A/en
Application granted granted Critical
Publication of CN107437093B publication Critical patent/CN107437093B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • G06F18/232—Non-hierarchical techniques
    • G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
    • G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
    • G06K17/00—Methods or arrangements for effecting co-operative working between equipments covered by two or more of main groups G06K1/00 - G06K15/00, e.g. automatic card files incorporating conveying and reading operations
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
    • G06K7/00—Methods or arrangements for sensing record carriers, e.g. for reading patterns
    • G06K7/10—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation
    • G06K7/10009—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation sensing by radiation using wavelengths larger than 0.1 mm, e.g. radio-waves or microwaves
    • G06K7/10316—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation sensing by radiation using wavelengths larger than 0.1 mm, e.g. radio-waves or microwaves using at least one antenna particularly designed for interrogating the wireless record carriers
    • G06K7/10356—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation sensing by radiation using wavelengths larger than 0.1 mm, e.g. radio-waves or microwaves using at least one antenna particularly designed for interrogating the wireless record carriers using a plurality of antennas, e.g. configurations including means to resolve interference between the plurality of antennas
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
    • G06K7/00—Methods or arrangements for sensing record carriers, e.g. for reading patterns
    • G06K7/10—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation
    • G06K7/10009—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation sensing by radiation using wavelengths larger than 0.1 mm, e.g. radio-waves or microwaves
    • G06K7/10366—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation sensing by radiation using wavelengths larger than 0.1 mm, e.g. radio-waves or microwaves the interrogation device being adapted for miscellaneous applications
    • G06K7/10376—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation sensing by radiation using wavelengths larger than 0.1 mm, e.g. radio-waves or microwaves the interrogation device being adapted for miscellaneous applications the interrogation device being adapted for being moveable

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Toxicology (AREA)
  • General Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • Data Mining & Analysis (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Electromagnetism (AREA)
  • General Health & Medical Sciences (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Probability & Statistics with Applications (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Radar Systems Or Details Thereof (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

The invention discloses an initial clustering center optimization selection method based on RFID data intensity-time distribution, which comprises the following substeps: determining a time span according to the moving speed of the mobile scanning terminal and the specification of the goods package; screening the preprocessed RFID data set from the starting time on the data of which the intensity value in each time span is greater than a set threshold value; calculating the average intensity of the data in each time span, and storing the average intensity and the right end point of the time interval of the corresponding time span; dividing the stored average intensity into a clustering central area and a non-clustering central area by using a hill climbing method according to an intensity threshold; and connecting the continuous time intervals divided into the clustering center areas in intervals, taking the time middle points of the connected intervals as initial clustering centers, and solving all the initial clustering centers. The method selects the initial clustering center by using the intensity-time distribution rule of the RFID scanning data, improves the selection of the clustering center and improves the accuracy and stability of the clustering result.

Description

Initial clustering center optimization selection method based on RFID data intensity-time distribution
Technical Field
The invention relates to the technical field of initial clustering center selection, in particular to an initial clustering center optimal selection method based on RFID data intensity-time distribution.
Background
Rfid technology has many advantages not available with traditional wireless communication means: the method has the advantages of non-contact, non-line-of-sight and multi-target identification, high positioning precision, low cost, convenient arrangement, strong anti-interference capability and strong environment adaptability, thereby having wide attention in the warehouse goods package positioning technology. In the actual warehouse parcel positioning application, the intensity-time distribution of the RFID scanning data acquired by the mobile RFID reader-multi-antenna array has certain regularity, and the specific expression is as follows: when the mobile RFID reader-multi-antenna array is close to the RFID electronic tag, the RSSI value returned by the tag is larger; when the RFID electronic tag based on the multi-antenna array principle is moved, the RSSI value returned by the tag is small. The RFID scanning data can be divided according to the rule, and then the position of the goods package is judged.
The cluster analysis is a common analysis method in the field of data mining, can divide data objects into a plurality of clusters or groups, has higher similarity between the objects in the same group, has larger difference between the data objects in different groups, and can cluster the data objects according to the distribution characteristics of the RFID scanning data. Therefore, the problem of locating warehouse goods packages is converted into the problem of clustering of RFID scanning data. The current clustering analysis methods are numerous, the RFID scanning data can be accurately partitioned conveniently and efficiently according to the strength-time distribution characteristics of the RFID scanning data based on the partitioned K-Means clustering algorithm, the algorithm complexity is low, and a large amount of RFID scanning data can be processed in a short time. The traditional K-Means clustering algorithm has strong dependence on the initial clustering center, the clustering result is often unstable due to different selection of the initial clustering center, the iteration times can be increased, and the algorithm execution efficiency can be reduced.
Currently, many researchers are working on improving the initial cluster center selection method of the K-Means clustering algorithm. Korean is wave and the like, and an initial clustering center optimization selection algorithm is provided, and the algorithm selects k high-density points as initial clustering centers by calculating density parameters of each data object; zhanghealthy and others put forward a K-Means initial clustering center selection algorithm based on optimal division, the algorithm optimally divides the data object space by using a histogram, the initial clustering center is automatically determined according to the density distribution of the data object, the K value does not need to be preset, the dependence of the algorithm result on parameters is reduced, and the algorithm efficiency and the accuracy are improved; kaufman proposes a typical density estimation KR method, which estimates the density of data distribution by calculating the distance between two pairs, and then selects an initial clustering center from the data in the region with higher local density; the selection method based on the super-triangular fused grid density provided by the beautiful jade and the like can get rid of the interference of a preset k value; the algorithm avoids random selection of an initial clustering center in the traditional K-Means algorithm, can accelerate the convergence speed of the algorithm to a certain extent and improves the algorithm efficiency, but most of the algorithms are based on data density, and the applicability of the algorithm is greatly reduced aiming at the condition that the density distribution change of RFID scanning data in time is not obvious.
Disclosure of Invention
In view of this, the present invention provides an initial cluster center optimization selection method based on RFID data intensity-time distribution.
In order to achieve the above object, the present invention provides the following technical solution, a method for optimally selecting an initial clustering center based on RFID data intensity-time distribution, comprising the steps of: step 1: determining time span t according to moving speed and goods package specification of mobile scanning terminalspan(ii) a Step 2: starting the preprocessed RFID data set from the starting time, for each time span tspanScreening the data with the internal strength value larger than a set threshold value; and step 3: calculating each time span tspanMean intensity of inner data RargAnd the average intensity R is measuredargAnd corresponding time span tspanThe right end point of the time interval is stored; and 4, step 4: repeating the steps 2 and 3 until all data in the data set are completely processed and stored; and 5: using hill climbing method to store average intensity R in step 4argDividing a clustering central area and a non-clustering central area according to the intensity threshold MinR; step 6: connecting intervals of continuous time intervals divided into clustering center areas, taking time middle points of the connected intervals as initial clustering centers, and solving all the initial clustering centers; and 7: and 6, calculating and outputting all initial clustering center points according to the step 6, and then finishing the algorithm.
Further, the set threshold is-45 db.
Further, the method for dividing the clustering center area and the non-clustering center area comprises the following steps: if average intensity RargIf the time interval is greater than the threshold MinR, the time interval is considered to be a clustering central area, otherwise, the time interval is considered to be a non-clustering central area; the threshold MinR is according toAnd (4) setting empirically.
The invention has the beneficial effects that:
the method selects the initial clustering center by using the intensity-time distribution rule of the RFID scanning data, improves the selection of the clustering center and improves the accuracy and stability of the clustering result.
Drawings
In order to make the object, technical scheme and beneficial effect of the invention more clear, the invention provides the following drawings for explanation:
FIG. 1 is a diagram of an RFID data acquisition process;
FIG. 2 is a graph of RFID data intensity versus time;
FIG. 3 is a flow chart of the algorithm of the present invention.
Detailed Description
Preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings.
The method is characterized in that RFID electronic tags statically placed around the mobile scanning terminal are scanned according to the collection mode of RFID data in an actual application scene, namely the mobile scanning terminal moves along a specified direction, and the specific collection process is shown in figure 1. In the acquisition process, the intensity-time distribution of the RFID electronic tag at a certain position shows certain regularity, namely when a reader approaches the RFID electronic tag, the received RSSI value is larger; when the reader is far away from the RFID electronic tag, the received RSSI value is small, as shown in FIG. 2. According to the intensity-time distribution rule of the RFID scanning data, the time point corresponding to the maximum intensity value is selected as the initial clustering center point, so that the selection of the clustering center can be improved, and the accuracy and the stability of the clustering result are improved.
The specific invention content is as follows:
step 1: determining time span t according to moving speed of mobile scanning terminal and specification of goods packagespan。
Step 2: starting the preprocessed RFID data set from the starting time for each tspanThe data with internal intensity value larger than-45 db is selected because the intensity value of the electromagnetic wave propagating in the Y direction is gradually attenuated and seriously interfered, and the intensity value is too highSmall data can affect the accuracy of cluster center selection and must be culled.
In particular, t from the RFID data setmin(minimum time) and tmax(maximum time) and t calculated in step 1span(time span) time-segmenting the entire RFID data set; for each segment (time interval tmin，tmin+tspan]) And screening the data with the internal strength value larger than-45 db to reduce the influence of the data with serious interference and make the data strength change obviously.
And step 3: calculate each tspan(time interval [ t)min，tmin+tspan]) Mean intensity of inner data RargAnd R isargAnd corresponds to tspanRight end point of time interval (t)min+tspan) And (4) storing, wherein the average intensity calculation formula is as follows:
Figure BDA0001343021300000031
wherein n is tspanNumber of RFID data, x, after screening within a time intervaliRSSI represents tspanAnd the strength value of the ith RFID data in the time interval.
And 4, step 4: and repeating the steps 2 and 3 until all data in the data set are completely processed and stored.
And 5: average intensity R stored in the steps by using a hill climbing methodargAnd dividing the clustering center area and the non-clustering center area according to the intensity threshold MinR. Comparing the average intensity R of each time span along the time axisargWith an intensity threshold MinR, if RargAnd if the minimum ratio is greater than MinR, the cluster center area is considered, otherwise, the cluster center area is a non-cluster center area, and the non-cluster center area is used for distinguishing the cluster center areas.
Step 6: and (3) connecting the continuous time intervals divided into the clustering center areas in intervals, ensuring that the clustering centers are contained in the time intervals, taking the middle time of the time intervals as initial clustering centers, and solving all the initial clustering centers.
And 7: and 6, calculating and outputting all initial clustering center points according to the step 6, and then finishing the algorithm.
Finally, it is noted that the above-mentioned preferred embodiments illustrate rather than limit the invention, and that, although the invention has been described in detail with reference to the above-mentioned preferred embodiments, it will be understood by those skilled in the art that various changes in form and detail may be made therein without departing from the scope of the invention as defined by the appended claims.

Claims (3)

1. An initial clustering center optimization selection method based on RFID data intensity-time distribution is characterized in that: the method comprises the following steps:
step 1: determining time span t according to moving speed and goods package specification of mobile scanning terminalspan；
Step 2: starting the preprocessed RFID data set from the starting time, for each time span tspanScreening the data with the internal strength value larger than a set threshold value;
and step 3: calculating each time span tspanMean intensity of inner data RargAnd the average intensity R is measuredargAnd corresponding time span tspanThe right end point of the time interval is stored;
and 4, step 4: repeating the steps 2 and 3 until all data in the data set are completely processed and stored;
and 5: using hill climbing method to store average intensity R in step 4argDividing a clustering central area and a non-clustering central area according to the intensity threshold MinR;
step 6: connecting intervals of continuous time intervals divided into clustering center areas, taking time middle points of the connected intervals as initial clustering centers, and solving all the initial clustering centers;
and 7: and 6, calculating and outputting all initial clustering center points according to the step 6, and then finishing the algorithm.
2. The method for optimally selecting the initial cluster center based on the RFID data intensity-time distribution according to claim 1, wherein the method comprises the following steps: the set threshold is-45 db.
3. The method for optimally selecting the initial cluster center based on the RFID data intensity-time distribution according to claim 1, wherein the method comprises the following steps: the method for dividing the clustering central area and the non-clustering central area comprises the following steps: if average intensity RargIf the time interval is greater than the threshold MinR, the time interval is considered to be a clustering central area, otherwise, the time interval is considered to be a non-clustering central area; the threshold value MinR is set empirically.
CN201710545656.5A 2017-07-06 2017-07-06 Initial clustering center optimization selection method based on RFID data intensity-time distribution Active CN107437093B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710545656.5A CN107437093B (en) 2017-07-06 2017-07-06 Initial clustering center optimization selection method based on RFID data intensity-time distribution

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710545656.5A CN107437093B (en) 2017-07-06 2017-07-06 Initial clustering center optimization selection method based on RFID data intensity-time distribution

Publications (2)

Publication Number Publication Date
CN107437093A CN107437093A (en) 2017-12-05
CN107437093B true CN107437093B (en) 2020-09-25

Family

ID=60459767

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710545656.5A Active CN107437093B (en) 2017-07-06 2017-07-06 Initial clustering center optimization selection method based on RFID data intensity-time distribution

Country Status (1)

Country Link
CN (1) CN107437093B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112698268B (en) * 2020-12-10 2023-01-17 青岛海信网络科技股份有限公司 Target equipment positioning method and positioning terminal
CN114155717B (en) * 2022-02-10 2022-04-26 西南交通大学 Traffic flow data screening method, device, equipment and readable storage medium

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20080098227A (en) * 2007-05-04 2008-11-07 한국전기연구원 Wireless position recognition system and method thereof
CN101587182A (en) * 2009-06-25 2009-11-25 华南理工大学 Locating method for RFID indoor locating system
CN104602341A (en) * 2015-01-08 2015-05-06 重庆邮电大学 Indoor WLAN (Wireless Local Area Network) positioning method based on random user signal logic diagram mapping

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20080098227A (en) * 2007-05-04 2008-11-07 한국전기연구원 Wireless position recognition system and method thereof
CN101587182A (en) * 2009-06-25 2009-11-25 华南理工大学 Locating method for RFID indoor locating system
CN104602341A (en) * 2015-01-08 2015-05-06 重庆邮电大学 Indoor WLAN (Wireless Local Area Network) positioning method based on random user signal logic diagram mapping

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
基于最优划分的K-Means初始聚类中心选取算法;张健沛等;《系统仿真学报》;20090505;第21卷(第9期);第2587-2588页 *
时间强度模型的建立;王东方;《中国优秀硕士学位论文全文数据库 信息科技辑》;20170315(第03期);第31页、第47-48页、第51页、图3.7、图4.3 *

Also Published As

Publication number Publication date
CN107437093A (en) 2017-12-05

Similar Documents

Publication Publication Date Title
CN107437093B (en) Initial clustering center optimization selection method based on RFID data intensity-time distribution
CN107844058B (en) Motion curve discrete dynamic planning method
CN108765452A (en) A kind of detection of mobile target in complex background and tracking
CN100580694C (en) Rapid multi-threshold value dividing method for gray-scale image
CN111860340B (en) Efficient K-nearest neighbor search algorithm for unmanned three-dimensional laser radar point cloud
CN108491908B (en) Visual intelligent warehousing system and method based on radio frequency identification
CN105243348B (en) Reader optimal deployment method based on passive ultrahigh frequency RFID positioning system
Teng et al. A fast Q algorithm based on EPC generation-2 RFID protocol
CN108134661A (en) The pilot distribution method of low complex degree in a kind of extensive mimo system
CN110503354B (en) RFID (radio frequency identification) tag position estimation method based on deep learning
Wu et al. Capture-aware Bayesian RFID tag estimate for large-scale identification
CN115372995A (en) Laser radar target detection method and system based on European clustering
CN111736167B (en) Method and device for obtaining laser point cloud density
CN111832986A (en) Product storage method and system, storage medium and computer equipment
CN114973191A (en) Dynamic threshold determining method based on point cloud density and distance and Euclidean clustering method
CN110505293B (en) Cooperative caching method based on improved drosophila optimization algorithm in fog wireless access network
CN116486371A (en) Obstacle detection method and obstacle detection device based on laser point cloud
Kalache et al. Performances comparison of RFID anti-collision algorithms
CN110781700B (en) RFID multi-reader coordination method
CN114612512A (en) KCF-based target tracking algorithm
CN110568400B (en) Coarse positioning method for article labels in drawer on moving direction shaft of reader-writer
CN104866790B (en) A kind of RFID system collision-proof method for owing to determine the tree-like packet of self adaptation of blind separation
CN114051207A (en) Ultra-wideband accurate positioning method and device under signal interference and electronic equipment
CN117697768B (en) Target grabbing method, robot, electronic equipment and storage medium
Li et al. Dynamic frame slotted aloha algorithm based on improved tag estimation

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
TR01 Transfer of patent right

Effective date of registration: 20230208

Address after: 213399 room 5025, building B, 218 Hongkou Road, Kunlun Street, Liyang City, Changzhou City, Jiangsu Province

Patentee after: Liyang Smart City Research Institute of Chongqing University

Address before: 400044 No. 174, positive street, Shapingba District, Chongqing

Patentee before: Chongqing University

TR01 Transfer of patent right