CN112100646A - Spatial data privacy protection matching method based on two-stage grid conversion - Google Patents

Spatial data privacy protection matching method based on two-stage grid conversion Download PDF

Info

Publication number
CN112100646A
CN112100646A CN202010274446.9A CN202010274446A CN112100646A CN 112100646 A CN112100646 A CN 112100646A CN 202010274446 A CN202010274446 A CN 202010274446A CN 112100646 A CN112100646 A CN 112100646A
Authority
CN
China
Prior art keywords
matching
data
space
point
point set
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
CN202010274446.9A
Other languages
Chinese (zh)
Inventor
张海涛
李济平
朱少楠
焦东来
王新光
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.)
Nanjing University of Posts and Telecommunications
Original Assignee
Nanjing University of Posts and Telecommunications
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 Nanjing University of Posts and Telecommunications filed Critical Nanjing University of Posts and Telecommunications
Priority to CN202010274446.9A priority Critical patent/CN112100646A/en
Publication of CN112100646A publication Critical patent/CN112100646A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
    • G06F21/60—Protecting data
    • G06F21/62—Protecting access to data via a platform, e.g. using keys or access control rules
    • G06F21/6218—Protecting access to data via a platform, e.g. using keys or access control rules to a system of files or objects, e.g. local or distributed file system or database
    • G06F21/6227—Protecting access to data via a platform, e.g. using keys or access control rules to a system of files or objects, e.g. local or distributed file system or database where protection concerns the structure of data, e.g. records, types, queries
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
    • G06F21/60—Protecting data
    • G06F21/62—Protecting access to data via a platform, e.g. using keys or access control rules
    • G06F21/6218—Protecting access to data via a platform, e.g. using keys or access control rules to a system of files or objects, e.g. local or distributed file system or database
    • G06F21/6245—Protecting personal data, e.g. for financial or medical purposes
    • G06F21/6254—Protecting personal data, e.g. for financial or medical purposes by anonymising data, e.g. decorrelating personal data from the owner's identification

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Databases & Information Systems (AREA)
  • Health & Medical Sciences (AREA)
  • Bioethics (AREA)
  • General Health & Medical Sciences (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Computer Hardware Design (AREA)
  • Computational Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Computer Security & Cryptography (AREA)
  • Pure & Applied Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Computing Systems (AREA)
  • Algebra (AREA)
  • Medical Informatics (AREA)
  • Collating Specific Patterns (AREA)

Abstract

The invention provides a space data privacy protection matching method based on two-stage grid conversion, which comprises two parts of space conversion processing of original point set data and privacy protection matching of coordinates of two-stage grid expression point sets. And a two-stage grid conversion method is adopted, so that the attack based on background knowledge is effectively avoided. The 1-level grid discretizes a data space, and the accurate coordinates of data points are hidden; the 2-level grid adopts local coordinates, so that background attack based on data outlier analysis is avoided. And the mapping values of the unmatched points are sequentially filtered by adopting a step-by-step screening method, so that the matching calculation efficiency is improved. By adopting a multi-time matching calculation method, only the corresponding original point set of the point pairs matched in all the rotated two-stage grid spaces is judged as final matching, so that the precision of the matching result is ensured.

Description

Spatial data privacy protection matching method based on two-stage grid conversion
Technical Field
The invention relates to the technical field of spatial data privacy protection, in particular to a spatial data privacy protection matching method based on two-stage grid conversion.
Background
In recent years, with the widespread of global positioning systems, sensor networks, and mobile devices, a large amount of emerging spatial data has been generated. The emerging spatial data have the advantages that the number of users is large, the space-time scale is large, and the like, which are not replaceable by the traditional spatial data. Through analysis of emerging spatial data, rules with rich semantic metaphors are found, and certain auxiliary decisions can be provided for scientific management of departments such as city planning and traffic.
At present, a very serious common problem exists for a plurality of analysis applications aiming at emerging spatial data: the bias of the data. That is, using emerging spatial data from a single source for analysis, it is difficult to achieve a complete activity description for users in an area. For example, subscriber location data generated in a mobile communication network is typically collected and stored by different communication carriers (e.g., three carriers in china: china mobile, china unicom, china telecom, dual card subscriber account for a small percentage), and analysis based on subscriber location data typically cannot cover all subscribers in the area. Therefore, to ensure unbiased performance for emerging spatial data analysis applications, integrated analysis of emerging spatial data from different sources is required.
The main means for realizing the spatial data integration analysis are as follows: and various spatial data mining technologies (such as decision trees, association rules, clustering and the like) which take the matched spatial data as objects and take the implicit knowledge and the spatial relationship as discovery targets. Spatial data matching is the basis of spatial data mining, and privacy protection matching is the core for ensuring data security analysis. The early privacy protection matching mainly adopts SMC technology, and has the problems of high computing and communication cost and low practicability. The existing mainstream method mainly adopts a data mapping technology based on geometric or algebraic conversion. However, the existing spatial data matching based on the data mapping technology generally has the problem that the attack based on the third-party background knowledge is difficult to effectively deal with.
Disclosure of Invention
The invention aims to provide a space data privacy protection matching method based on two-stage grid conversion, which can effectively avoid attack based on background knowledge and has the advantages of high calculation speed and high matching precision.
The invention provides a space data privacy protection matching method based on two-stage grid conversion, which comprises two parts of space conversion processing of original point set data and privacy protection matching of coordinates of two-stage grid expression point sets, wherein the space conversion processing of the original point set data comprises the following steps:
the method comprises the following steps: expressing the point sets of the original data in a unified coordinate system by both sides with the original data, and making the conversion parameters consistent;
step two: expressing the point set coordinates in respective original spaces by both sides in a rotating space coordinate system based on the conversion parameters, and further converting the point set coordinates into point set coordinates expressed based on a two-stage grid;
step three: the two parties respectively send point set coordinates based on two-stage grid expression and a matching threshold value based on Euclidean distance to a third party;
the privacy protection matching of the two-level grid expression point set coordinates comprises the following steps:
step four: the third party sequentially carries out space matching and threshold comparison based on Euclidean distance on the point set coordinates expressed based on the two-stage grids, and a step-by-step screening method is adopted to obtain a final matching point pair;
step five: the third party respectively sends the numbers of the final matching point pairs back to the two parties with the original point set data;
step six: the two parties possessing the original point set data exchange the original point set data with each other according to the numbers of the matching point pairs.
The further improvement lies in that: the specific steps of the spatial conversion processing of the original point set data are as follows:
the method comprises the following steps: the original data owner A, B expresses its point set in a unified coordinate system and agrees with the conversion parameters;
step two: the data owner A, B creates rotation transformation matrices respectively, and maps the point set coordinates in the original space to the point set coordinates in the rotation space based on the rotation transformation matrices;
step three: the data owner A, B converts the point set coordinates in the rotation space to point set coordinates in the level 1 grid space, respectively;
step four: the data owner A, B converts the point set coordinates in level 1 grid space to point set coordinates in level 2 grid space, respectively;
step five: the data owner A, B sends the point set data in level 2 grid space and the euclidean distance based matching threshold to the third party, respectively.
The further improvement lies in that: the privacy protection matching method of the two-level grid expression point set coordinates comprises the following specific steps:
step six: calculating the Euclidean distance between pairs of data set points in the 2-level grid space after the 1 st rotation by a third party;
step seven: the third party obtains a matching point pair in the 2-level grid space after the 1 st rotation by screening based on the comparison between the Euclidean distance and the matching threshold;
step eight: calculating Euclidean distances of the matching point pairs in the 2-level grid space after the 2 nd rotation, and screening the matching point pairs in the 2 nd-level grid space after the 2 nd rotation through threshold value comparison;
step nine: adopting a recursion mode to screen step by step to obtain all the matched point pairs in the rotated 2-level grid space;
step ten: the third party sends the numbers of the final pairs of matching points to the data owner A, B, respectively, which exchanges the original point set data with each other according to the numbers of the pairs of matching points.
The further improvement lies in that: five definitions are involved in the method, which are respectively: the method comprises the steps of rotating a point set in a space, a point set in a level 1 grid space, a point set in a level 2 grid space, and Euclidean distances and matching point pairs among points in the level 2 grid space.
The invention has the beneficial effects that: the attack based on background knowledge can be effectively avoided, and the method has the advantages of high calculation speed and high matching precision. And a two-stage grid conversion method is adopted, so that the attack based on background knowledge is effectively avoided. The 1-level grid discretizes a data space, and the accurate coordinates of data points are hidden; the 2-level grid adopts local coordinates, so that background attack based on data outlier analysis is avoided. And the mapping values of the unmatched points are sequentially filtered by adopting a step-by-step screening method, so that the matching calculation efficiency is improved. By adopting a multi-time matching calculation method, only the corresponding original point set of the point pairs matched in all the rotated two-stage grid spaces is judged as final matching, so that the precision of the matching result is ensured.
Drawings
FIG. 1 is a set of points OPs in the original space of data owner A of the present inventionASchematic representation of the expression of (a).
FIG. 2 is a set of points OPs in the original space of data owner B of the present inventionBSchematic representation of the expression of (a).
FIG. 3 is a set of points in rotational space of data owner A of the present invention
Figure BDA0002444268510000051
Schematic representation of the expression.
FIG. 4 is a set of points in rotational space of data owner B of the present invention
Figure BDA0002444268510000052
Schematic representation of the expression.
FIG. 5 is a set of points in level 1 grid space of data owner A of the present invention
Figure BDA0002444268510000053
Schematic representation of the expression.
FIG. 6 is a set of points in level 1 grid space of data owner B of the present invention
Figure BDA0002444268510000054
Schematic representation of the expression.
FIG. 7 is a set of points in a 2-level grid space of data owner A of the present invention
Figure BDA0002444268510000055
Schematic representation of the expression.
FIG. 8 is a set of points in 2-level grid space of data owner B of the present invention
Figure BDA0002444268510000056
Schematic representation of the expression.
Fig. 9 is a schematic diagram showing 9 distribution expressions of point pairs when the euclidean distance between the point pairs is calculated according to the present invention.
FIG. 10 is a diagram of a data set in two levels of grid space after the 1 st rotation of the present invention
Figure BDA0002444268510000057
Is shown.
FIG. 11 is a diagram of the present invention in a two-level grid space after 2 nd rotation corresponding to the matching point pairs in FIG. 10
Figure BDA0002444268510000058
And
Figure BDA0002444268510000059
point pair diagram in (1).
Fig. 12 is a schematic diagram of the matched point pairs obtained by screening the point pairs in fig. 11 according to the present invention.
FIG. 13 is a diagram of a data set in a two-level grid space after a 3 rd rotation corresponding to the matching point pairs of FIG. 12 according to the present invention
Figure BDA00024442685100000510
And
Figure BDA00024442685100000511
point pair diagram in (1).
Fig. 14 is a schematic diagram of the matched point pairs obtained by screening the point pairs in fig. 13 according to the present invention.
FIG. 15 is a diagram of a data set in a two-level grid space after 4 th rotation corresponding to the matching point pairs of FIG. 14 according to the present invention
Figure BDA0002444268510000061
And
Figure BDA0002444268510000062
point pair diagram in (1).
Fig. 16 is a schematic diagram of the matched point pairs obtained by screening the point pairs in fig. 15 according to the present invention.
FIG. 17 is a diagram of a data set in two levels of grid space after the 5 th rotation corresponding to the matching point pairs of FIG. 16 according to the present invention
Figure BDA0002444268510000063
And
Figure BDA0002444268510000064
point pair diagram in (1).
Fig. 18 is a schematic diagram of the matched point pairs obtained by screening the point pairs in fig. 17 according to the present invention.
Detailed Description
For the purpose of enhancing understanding of the present invention, the present invention will be further described in detail with reference to the following examples, which are provided for illustration only and are not to be construed as limiting the scope of the present invention.
First, several basic definitions are given:
definition 1: given a set of points in the original space, OPs ═ p1,p2,p3,., after the ith rotation, the corresponding points in the rotation space are collected as
Figure BDA0002444268510000065
Wherein p is1,p2,p3,.. are points in the original set of spatial points OPs,
Figure BDA0002444268510000066
is a set of points RPs in rotation spaceiFor point p in OPs, it is associated with RPsiPoint q in (1)iThe mapping conversion rule of (1) is as follows:
[qi·x qi·y 1]T＝Mi×[p·x p·y 1]T，
wherein q isi·x，qiY represents the point qiP x, p y respectively represent the abscissa and ordinate of the point p,
Figure BDA0002444268510000067
is around point uiRotation alphaiThe rotation of the angle (counterclockwise direction of rotation is the positive direction) transforms the matrix.
Figure BDA0002444268510000068
Is rotated by alpha about the originiA rotation matrix of the angle is formed,
Figure BDA0002444268510000069
is translated to point uiTranslation matrix ui·x，uiY is the point uiThe abscissa and the ordinate of (a).
Definition 2: set of points in level 1 grid space: given a set of points in the rotation space after the ith rotation
Figure BDA0002444268510000071
Its corresponding set of points in level 1 grid space is
Figure BDA0002444268510000072
Wherein the content of the first and second substances,
Figure BDA0002444268510000073
is a set of points RPs in rotation spaceiThe point (b) in (c) is,
Figure BDA0002444268510000074
is a set of points FPs in a level 1 grid spaceiPoint of (1) for RPsiPoint q in (1)iOf it with FPsiPoint d iniThe mapping conversion rule of (1) is as follows:
Figure BDA0002444268510000075
wherein d isi·x，diY represents the point diAbscissa and ordinate of (a), qi·x，qiY represents the point qiThe abscissa and the ordinate of (a), w represents the upper bound of the level 1 grid space,
Figure BDA0002444268510000076
indicating a rounding down.
Definition 3: set of points in 2-level grid space: given the set of points in level 1 mesh space after the ith rotation
Figure BDA0002444268510000077
Its corresponding set of points in 2-level grid space is
Figure BDA0002444268510000078
Wherein the content of the first and second substances,
Figure BDA0002444268510000079
is a set of points FPs in a level 1 grid spaceiThe point (b) in (c) is,
Figure BDA00024442685100000710
is a set of points SPs in a 2-level grid spaceiPoints in (1) for FPsiPoint d iniWith SPsiPoint c iniThe mapping conversion rule of (1) is as follows:
Figure BDA00024442685100000711
wherein, ci·x，ciY represents the point ciAbscissa and ordinate of (d)i·x，diY represents the point diW represents the upper bound of the 1-level grid space, k represents the upper bound of the 2-level grid space,
Figure BDA00024442685100000712
indicating a rounding down.
Definition 4: euclidean distances between points in level 2 grid space: given two 2-level grid space midpoints cAAnd cBAnd the euclidean distance therebetween is defined as:
Figure BDA00024442685100000713
wherein the content of the first and second substances,
Figure BDA00024442685100000714
Figure BDA0002444268510000081
Δxand ΔyRespectively represent cAAnd c' a difference between the horizontal and vertical coordinates, cAX and c'. x represent cAAnd the abscissa of c', cAY and c'. y denote cAAnd the ordinate of c'. c ═ cB·x+i,cB·y+j),i∈{-k,0,k},j∈{-k,0,k}，cB·x，cBY represents cBK is the upper bound of the 2-level grid space.
Definition 5: matching point pairs: given two 2-level grid space midpoints cAAnd cBAnd a set euclidean distance matching threshold', if the condition is satisfied: cdist (c)A,cB) If <' > then is called cAAnd cBTo match a point pair, wherein cdist (c)A,cB) Is a point cAAnd cBHas an Euclidean distance of ═ k/w, and is cAAnd cBCorresponding to the Euclidean distance matching threshold in the original space, w is cAAnd cBUpper bound of the corresponding level 1 grid space, k being cAAnd cBThe upper bound of the corresponding level 2 grid space.
The embodiment comprises two stages of space conversion processing of original point set data and privacy protection matching of two-stage grid expression point set coordinates, namely a first stage: spatial transformation processing of raw point set data
Step 1) original data owner A, B expresses its point set in a uniform coordinate system and agrees with the conversion parameter;
in this example, the sets of points in the original space of the data owner A, B under a certain unified coordinate system are: OPsA＝{p1A,p2A,p3A,p4A}＝{(0.05,0.1),(0.2,0.2),(0.7,0.3),(0.6,0.65)}，OPsB＝{p1B,p2B,p3B,p4B(0.6 ), (0.4,0.1), (0.8,0.4), (0.2,0.9) }, and the corresponding graphic representations are shown in fig. 1 and 2, in which (a) is a set of points OPs in the original space of the data owner aAExpression, (B) is the set of points OPs in the original space of data owner BBAnd (4) expressing. The conversion parameters agreed by both parties are: w is 0.5, k is 5, k is 0.1, the number of rotations i is 5, and a is a corresponding rotation angle1＝30°,a2＝37°,a3＝45°,a4＝53°,a560 °, rotation point u1＝u2＝u3＝u4＝u5＝(0,1)。
Step 2) the data owner A, B creates rotation transformation matrices respectively, and maps the point set coordinates in the original space to the point set coordinates in the rotation space based on the rotation transformation matrices; in this example, the data owner A, B depends on the rotation angle a1＝30°,a2＝37°,a3＝45°,a4＝53°,a560 °, and a rotation point u1＝u2＝u3＝u4＝u5Create 5 rotation transformation matrices M as defined by 1 ═ 0,11～M5And obtaining point sets OPs respectivelyA、OPsBSet of points in 5 rotation spaces
Figure BDA0002444268510000091
We use OPsAPoint p in (1)1A(0.05,0.1), winding point u1Rotation α ═ 0,11After 30 °, a point in the rotation space is obtained
Figure BDA00024442685100000910
For example, a specific calculation process is given. Wherein the content of the first and second substances,
Figure BDA0002444268510000092
Figure BDA0002444268510000093
Figure BDA0002444268510000094
Figure BDA0002444268510000095
and finally calculating to obtain:
Figure BDA0002444268510000096
similarly, for point sets OPsAOther points in (1): p is a radical of2A(0.2,0.2)、p3A(0.7,0.3)、p4A(0.6,0.65), calculated to yield:
Figure BDA0002444268510000097
Figure BDA0002444268510000098
namely:
Figure BDA0002444268510000099
then, point set OPsAAround point u1Rotation α ═ 0,12＝37°、α3＝45°、α4＝53°、α5After 60 degrees, the corresponding point sets in the rotation space are respectively:
Figure BDA0002444268510000101
Figure BDA0002444268510000102
Figure BDA0002444268510000103
Figure BDA0002444268510000104
further, the point sets OPsBAround point u1Rotation α ═ 0,11＝30°，α2＝37°、α3＝45°、α4＝53°、α5After 60 degrees, the corresponding point sets in the rotation space are respectively:
Figure BDA0002444268510000105
Figure BDA0002444268510000106
Figure BDA0002444268510000107
Figure BDA0002444268510000108
Figure BDA0002444268510000109
Figure BDA00024442685100001010
and
Figure BDA00024442685100001011
the graphic representation is shown in FIGS. 3 and 4, wherein a 1-a 5 are point sets in the rotation space of the data owner A
Figure BDA00024442685100001012
Expression, b1 ℃B5 is a set of points in the rotation space of data owner B
Figure BDA00024442685100001013
And (4) expressing.
Step 3) the data owner A, B converts the point set coordinates in the rotation space into point set coordinates in the 1-level grid space, respectively;
in this example, the data owner A, B sets the points in the rotation space according to definition 2, with the set parameter w equal to 0.5
Figure BDA00024442685100001014
Mapping conversion is carried out to obtain a corresponding point set in the 1-level grid space
Figure BDA00024442685100001015
We collect with points
Figure BDA00024442685100001016
Point of (5)
Figure BDA00024442685100001017
Obtaining points in level 1 grid space
Figure BDA0002444268510000111
For example, a specific calculation process is given:
Figure BDA0002444268510000112
Figure BDA0002444268510000113
that is, obtain
Figure BDA0002444268510000114
Similarly, for point sets
Figure BDA0002444268510000115
Other points in (1):
Figure BDA0002444268510000116
Figure BDA0002444268510000117
and calculating to obtain:
Figure BDA0002444268510000118
Figure BDA0002444268510000119
namely:
Figure BDA00024442685100001110
then, it is calculated to obtain
Figure BDA00024442685100001111
The corresponding point sets in the level 1 grid space are respectively:
Figure BDA00024442685100001112
Figure BDA00024442685100001113
Figure BDA00024442685100001114
Figure BDA00024442685100001115
further, the calculation results in
Figure BDA00024442685100001116
The corresponding point sets in the level 1 grid space are respectively:
Figure BDA00024442685100001117
Figure BDA00024442685100001118
Figure BDA00024442685100001119
Figure BDA00024442685100001120
Figure BDA00024442685100001121
Figure BDA00024442685100001122
and
Figure BDA00024442685100001123
is shown in FIGS. 5 and 6, wherein a 1-a 5 are point sets in the level 1 grid space of data owner A
Figure BDA00024442685100001124
Expression, B1-B5 are sets of points in level 1 grid space of data owner B
Figure BDA00024442685100001125
And (4) expressing.
Step 4) the data owner A, B converts the point set coordinates in the 1-level grid space into point set coordinates in the 2-level grid space, respectively;
in this example, the data owner A, B sets the points in the 1-level mesh space according to definition 3 based on the set parameters w-0.5 and k-5
Figure BDA0002444268510000121
Mapping conversion is carried out to obtain a corresponding point set in the 2-level grid space
Figure BDA0002444268510000122
We collect with points
Figure BDA0002444268510000123
Point of (5)
Figure BDA0002444268510000124
Obtaining points in 2-level grid space
Figure BDA0002444268510000125
For example, a specific calculation process is given:
Figure BDA0002444268510000126
Figure BDA0002444268510000127
that is, obtain
Figure BDA0002444268510000128
Similarly, for point sets
Figure BDA0002444268510000129
Other points in (1):
Figure BDA00024442685100001210
Figure BDA00024442685100001211
and calculating to obtain:
Figure BDA00024442685100001212
Figure BDA00024442685100001213
namely:
Figure BDA00024442685100001214
then, it is calculated to obtain
Figure BDA00024442685100001215
The corresponding point sets in the 2-level grid space are respectively:
Figure BDA00024442685100001216
Figure BDA00024442685100001217
further, the calculation results in
Figure BDA00024442685100001218
The corresponding point sets in the 2-level grid space are respectively:
Figure BDA00024442685100001219
Figure BDA00024442685100001220
Figure BDA00024442685100001221
Figure BDA00024442685100001222
and
Figure BDA00024442685100001223
is shown in FIGS. 7 and 8, wherein a 1-a 5 are point sets in 2-level grid space of data owner A
Figure BDA00024442685100001224
Expression, B1-B5 are sets of points in the 2-level grid space of data owner B
Figure BDA0002444268510000131
And (4) expressing.
Step 5) the data owner A, B sends the point set data in level 2 grid space and the matching threshold based on euclidean distance to the third party, respectively.
In this example, the following is calculated according to the parameters 0.1, k 5 and w 0.5: the matching threshold value' based on the euclidean distance is 0.1 × 5/0.5 is 1. Data owner A, B separately gathers points in 2-level grid space
Figure BDA0002444268510000132
Figure BDA0002444268510000133
And sending the matching threshold value' 1 based on the Euclidean distance to a third party.
And a second stage: privacy preserving matching of two-level grid expression point set coordinates
Step 6), the third party calculates the Euclidean distance between data set point pairs in the 2-level grid space after the 1 st rotation;
in this example, the data owner A, B sends SPs's according to definition 4 based on the set parameter k being 5A、SPsBThe 1 st rotated point set contained in
Figure BDA0002444268510000134
The euclidean distance between them.
We compute a set of points
Figure BDA0002444268510000135
Point of (5)
Figure BDA0002444268510000136
Sum point set
Figure BDA0002444268510000137
Point of (5)
Figure BDA0002444268510000138
The euclidean distance between them is taken as an example to illustrate a specific implementation process.
Figure BDA0002444268510000139
And
Figure BDA00024442685100001310
the euclidean distance therebetween is defined as:
Figure BDA00024442685100001311
wherein the content of the first and second substances,
Figure BDA00024442685100001312
by
Figure BDA00024442685100001313
k is 5, calculated as:
Figure BDA0002444268510000141
by
Figure BDA0002444268510000142
Taking c' ═ (-3, -1) as an example, further calculation yields:
Figure BDA0002444268510000143
c'·x＝-3，
Figure BDA0002444268510000144
Figure BDA0002444268510000145
c'·y＝4，
Figure BDA0002444268510000146
therefore, the temperature of the molten metal is controlled,
Figure BDA0002444268510000147
in the same way, all possible conditions are obtained
Figure BDA0002444268510000148
And
Figure BDA0002444268510000149
the euclidean distance therebetween is:
Figure BDA00024442685100001410
in summary,
Figure BDA00024442685100001411
point pair
Figure BDA00024442685100001412
And
Figure BDA00024442685100001413
is shown in fig. 9, where ● represents a data set
Figure BDA00024442685100001414
Point of (5)
Figure BDA00024442685100001415
A means a data set
Figure BDA00024442685100001416
Point of (5)
Figure BDA00024442685100001417
In the same way, calculate
Figure BDA00024442685100001418
And
Figure BDA00024442685100001419
the euclidean distance between the other 15 points in (a) results are as follows:
Figure BDA00024442685100001420
Figure BDA00024442685100001421
Figure BDA0002444268510000151
Figure BDA0002444268510000152
step 7), the third party obtains a matching point pair in the 2-level grid space after the 1 st rotation by screening based on the comparison between the Euclidean distance and the matching threshold;
in this example, according to definition 5, the euclidean distance between the point pairs calculated in step 6 is compared with a matching threshold' ═ 1 to obtain matching point pairs. The specific process is as follows:
Figure BDA0002444268510000153
mismatch is not achieved;
Figure BDA0002444268510000154
mismatch is not achieved;
Figure BDA0002444268510000155
matching;
Figure BDA0002444268510000156
mismatch is not achieved;
Figure BDA0002444268510000157
mismatch is not achieved;
Figure BDA0002444268510000158
mismatch is not achieved;
Figure BDA0002444268510000159
matching;
Figure BDA00024442685100001510
mismatch is not achieved;
Figure BDA00024442685100001511
mismatch is not achieved;
Figure BDA00024442685100001512
mismatch is not achieved;
Figure BDA00024442685100001513
matching;
Figure BDA00024442685100001514
mismatch is not achieved;
Figure BDA00024442685100001515
matching;
Figure BDA00024442685100001516
matching;
Figure BDA00024442685100001517
mismatch is not achieved;
Figure BDA00024442685100001518
and (6) matching.
The obtained matching point pairs are:
Figure BDA00024442685100001519
and
Figure BDA00024442685100001520
Figure BDA00024442685100001521
and
Figure BDA00024442685100001522
Figure BDA00024442685100001523
and
Figure BDA00024442685100001524
Figure BDA00024442685100001525
and
Figure BDA00024442685100001526
Figure BDA00024442685100001527
and
Figure BDA00024442685100001528
Figure BDA00024442685100001529
and
Figure BDA00024442685100001530
the graphical representation is shown in FIG. 10, where ● represents a data set
Figure BDA00024442685100001531
The points in (A) represent the data set
Figure BDA00024442685100001532
Point (2).
And 8) calculating the Euclidean distance of the matching point pairs in the 2-level grid space after the 2 nd rotation, and screening to obtain the matching point pairs in the 2-level grid space after the 2 nd rotation through threshold value comparison.
In this example, SPs are obtained according to the matching point pairs obtained in step 7A、SPsBPoint set after 2 nd rotation
Figure BDA00024442685100001533
And
Figure BDA00024442685100001534
point pair (5):
Figure BDA00024442685100001535
and
Figure BDA00024442685100001536
Figure BDA00024442685100001537
and
Figure BDA0002444268510000161
Figure BDA0002444268510000162
and
Figure BDA0002444268510000163
Figure BDA0002444268510000164
and
Figure BDA0002444268510000165
Figure BDA0002444268510000166
and
Figure BDA0002444268510000167
Figure BDA0002444268510000168
and
Figure BDA0002444268510000169
the graphical representation is shown in FIG. 11, where ● represents a data set
Figure BDA00024442685100001610
The points in (A) represent the data set
Figure BDA00024442685100001611
Point (2).
According to definition 4 and definition 5, the euclidean distance between the point pairs is calculated and compared with a matching threshold' 1, resulting in the following:
Figure BDA00024442685100001612
mismatch is not achieved;
Figure BDA00024442685100001613
matching;
Figure BDA00024442685100001614
mismatch is not achieved;
Figure BDA00024442685100001615
matching;
Figure BDA00024442685100001616
mismatch is not achieved;
Figure BDA00024442685100001617
and (6) matching.
The obtained matching point pairs are:
Figure BDA00024442685100001618
and
Figure BDA00024442685100001619
Figure BDA00024442685100001620
and
Figure BDA00024442685100001621
Figure BDA00024442685100001622
and
Figure BDA00024442685100001623
the graphical representation is shown in FIG. 12, where ● represents a data set
Figure BDA00024442685100001624
The points in (A) represent the data set
Figure BDA00024442685100001625
Point (2).
And 9) screening step by adopting a recursion mode to obtain all the rotated matching point pairs in the 2-level grid space.
In this example, SPs are obtained according to the matching point pairs obtained in step 8A、SPsBPoint set after 3 rd rotation
Figure BDA00024442685100001626
And
Figure BDA00024442685100001627
point pair (5):
Figure BDA00024442685100001628
and
Figure BDA00024442685100001629
Figure BDA00024442685100001630
and
Figure BDA00024442685100001631
Figure BDA00024442685100001632
and
Figure BDA00024442685100001633
the graphical representation is shown in FIG. 13, where ● represents a data set
Figure BDA00024442685100001634
The points in (A) represent the data set
Figure BDA00024442685100001635
Point (2).
In accordance with definitions 4 and 5, the euclidean distances between the pairs of points are calculated and compared with a matching threshold' of 1, respectively, with the following results:
Figure BDA00024442685100001636
mismatch is not achieved;
Figure BDA00024442685100001637
matching;
Figure BDA00024442685100001638
and (6) matching.
The obtained matching point pairs are:
Figure BDA00024442685100001639
and
Figure BDA00024442685100001640
Figure BDA00024442685100001641
and
Figure BDA00024442685100001642
the graphical representation is shown in FIG. 14, where ● represents a data set
Figure BDA0002444268510000171
The points in (a) and the triangular points represent data sets
Figure BDA0002444268510000172
Point (2).
Further, from the matching point pairs
Figure BDA0002444268510000173
And
Figure BDA0002444268510000174
Figure BDA0002444268510000175
and
Figure BDA0002444268510000176
corresponding to obtain SPsA、SPsBSet of points after the 4 th rotation
Figure BDA0002444268510000177
And
Figure BDA0002444268510000178
point pair (5):
Figure BDA0002444268510000179
and
Figure BDA00024442685100001710
Figure BDA00024442685100001711
and
Figure BDA00024442685100001712
the graphical representation is shown in FIG. 15, where ● represents a data set
Figure BDA00024442685100001713
The points in (A) represent the data set
Figure BDA00024442685100001714
Point (2).
In accordance with definitions 4 and 5, the euclidean distances between the pairs of points are calculated and compared with a matching threshold' of 1, respectively, with the following results:
Figure BDA00024442685100001715
matching;
Figure BDA00024442685100001716
and (6) matching.
The obtained matching point pairs are:
Figure BDA00024442685100001717
and
Figure BDA00024442685100001718
Figure BDA00024442685100001719
and
Figure BDA00024442685100001720
the graphical representation is shown in FIG. 16, where ● represents a data set
Figure BDA00024442685100001721
The points in (A) represent the data set
Figure BDA00024442685100001722
Point (2).
Further, from the matching point pairs
Figure BDA00024442685100001723
And
Figure BDA00024442685100001724
Figure BDA00024442685100001725
and
Figure BDA00024442685100001726
corresponding to obtain SPsA、SPsBPoint set after the 5 th rotation
Figure BDA00024442685100001727
And
Figure BDA00024442685100001728
point pair (5):
Figure BDA00024442685100001729
and
Figure BDA00024442685100001730
Figure BDA00024442685100001731
and
Figure BDA00024442685100001732
as shown in fig. 17, wherein ● represents a data set
Figure BDA00024442685100001733
The points in (A) represent the data set
Figure BDA00024442685100001734
Point (2).
In accordance with definitions 4 and 5, the euclidean distances between the pairs of points are calculated and compared with a matching threshold' of 1, respectively, with the following results:
Figure BDA00024442685100001735
matching;
Figure BDA00024442685100001736
and not matched.
The obtained matching point pairs are:
Figure BDA00024442685100001737
and
Figure BDA00024442685100001738
the graphical representation is shown in FIG. 18, where ● represents a data set
Figure BDA00024442685100001739
The points in (A) represent the data set
Figure BDA00024442685100001740
Point (2).
At this point, the matching calculation is finished, and finally the matching point pairs meeting the requirements are obtained as follows:
Figure BDA00024442685100001741
and
Figure BDA00024442685100001742
step 10) the third party sends the numbers of the final pairs of matching points to the data owner A, B, respectively, which exchanges the original point set data with each other according to the numbers of the pairs of matching points.
In this example, the third party will
Figure BDA0002444268510000181
Sent to the data owner A and will
Figure BDA0002444268510000182
To the data owner B. Data owner A by number
Figure BDA0002444268510000183
Sets of points OPs from the original spaceAIn (1), selecting p4A(0.6,0.65) to the data owner B. Accordingly, data owner B is numbered accordingly
Figure BDA0002444268510000184
Sets of points OPs from the original spaceBIn (1), selecting p1B(0.6 ) to the data owner a. And finally, realizing the exchange of the data of the matched original point set.

Claims (4)

1. A space data privacy protection matching method based on two-stage grid conversion is characterized in that: the matching method comprises two parts of space conversion processing of original point set data and privacy protection matching of two-stage grid expression point set coordinates, wherein the space conversion processing of the original point set data comprises the following steps:
the method comprises the following steps: expressing the point sets of the original data in a unified coordinate system by both sides with the original data, and making the conversion parameters consistent;
step two: expressing the point set coordinates in respective original spaces by both sides in a rotating space coordinate system based on the conversion parameters, and further converting the point set coordinates into point set coordinates expressed based on a two-stage grid;
step three: the two parties respectively send point set coordinates based on two-stage grid expression and a matching threshold value based on Euclidean distance to a third party;
the privacy protection matching of the two-level grid expression point set coordinates comprises the following steps:
step four: the third party sequentially carries out space matching and threshold comparison based on Euclidean distance on the point set coordinates expressed based on the two-stage grids, and a step-by-step screening method is adopted to obtain a final matching point pair;
step five: the third party respectively sends the numbers of the final matching point pairs back to the two parties with the original point set data;
step six: the two parties possessing the original point set data exchange the original point set data with each other according to the numbers of the matching point pairs.
2. The spatial data privacy protection matching method based on two-stage grid transformation as claimed in claim 1, characterized in that: the specific steps of the spatial conversion processing of the original point set data are as follows:
the method comprises the following steps: the original data owner A, B expresses its point set in a unified coordinate system and agrees with the conversion parameters;
step two: the data owner A, B creates rotation transformation matrices respectively, and maps the point set coordinates in the original space to the point set coordinates in the rotation space based on the rotation transformation matrices;
step three: the data owner A, B converts the point set coordinates in the rotation space to point set coordinates in the level 1 grid space, respectively;
step four: the data owner A, B converts the point set coordinates in level 1 grid space to point set coordinates in level 2 grid space, respectively;
step five: the data owner A, B sends the point set data in level 2 grid space and the euclidean distance based matching threshold to the third party, respectively.
3. The spatial data privacy protection matching method based on two-stage grid transformation as claimed in claim 1, characterized in that: the privacy protection matching method of the two-level grid expression point set coordinates comprises the following specific steps:
step six: calculating the Euclidean distance between pairs of data set points in the 2-level grid space after the 1 st rotation by a third party;
step seven: the third party obtains a matching point pair in the 2-level grid space after the 1 st rotation by screening based on the comparison between the Euclidean distance and the matching threshold;
step eight: calculating Euclidean distances of the matching point pairs in the 2-level grid space after the 2 nd rotation, and screening the matching point pairs in the 2 nd-level grid space after the 2 nd rotation through threshold value comparison;
step nine: adopting a recursion mode to screen step by step to obtain all the matched point pairs in the rotated 2-level grid space;
step ten: the third party sends the numbers of the final pairs of matching points to the data owner A, B, respectively, which exchanges the original point set data with each other according to the numbers of the pairs of matching points.
4. The spatial data privacy protection matching method based on two-stage grid transformation as claimed in claim 1, characterized in that: five definitions are involved in the method, which are respectively: the method comprises the steps of rotating a point set in a space, a point set in a level 1 grid space, a point set in a level 2 grid space, and Euclidean distances and matching point pairs among points in the level 2 grid space.
CN202010274446.9A 2020-04-09 2020-04-09 Spatial data privacy protection matching method based on two-stage grid conversion Pending CN112100646A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010274446.9A CN112100646A (en) 2020-04-09 2020-04-09 Spatial data privacy protection matching method based on two-stage grid conversion

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010274446.9A CN112100646A (en) 2020-04-09 2020-04-09 Spatial data privacy protection matching method based on two-stage grid conversion

Publications (1)

Publication Number Publication Date
CN112100646A true CN112100646A (en) 2020-12-18

Family

ID=73750307

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010274446.9A Pending CN112100646A (en) 2020-04-09 2020-04-09 Spatial data privacy protection matching method based on two-stage grid conversion

Country Status (1)

Country Link
CN (1) CN112100646A (en)

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110102435A1 (en) * 2009-11-04 2011-05-05 Tomtec Imaging Systems Gmbh Method and device for visualizing surface-like structures in volumetric data sets
CN106649262A (en) * 2016-10-31 2017-05-10 复旦大学 Protection method for enterprise hardware facility sensitive information in social media

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110102435A1 (en) * 2009-11-04 2011-05-05 Tomtec Imaging Systems Gmbh Method and device for visualizing surface-like structures in volumetric data sets
CN106649262A (en) * 2016-10-31 2017-05-10 复旦大学 Protection method for enterprise hardware facility sensitive information in social media

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
G.GHINITA,C.R.VICENTE,N.SHANG,AND E.BERTINO: "Privacy-preserving matching of spatial datasets with protection against background knowledge", 《PRIVACY-PRESERVING MATCHING OF SPATIAL DATASETS WITH PROTECTION AGAINST BACKGROUND KNOWLEDGE》 *

Similar Documents

Publication Publication Date Title
CN106504326B (en) Take the landform altitude sampled point encryption method of form precision into account
WO2022151654A1 (en) Random greedy algorithm-based horizontal federated gradient boosted tree optimization method
CN102915423B (en) A kind of power business data filtering system based on rough set and gene expression and method
Castiglia et al. Compressed-vfl: Communication-efficient learning with vertically partitioned data
Li et al. Retracted: Design of multimedia blockchain privacy protection system based on distributed trusted communication
CN113033652B (en) Image recognition system and method based on block chain and federal learning
CN107222851A (en) A kind of method of utilization difference secret protection Wifi Fingerprint indoor locating system privacies
CN111639356A (en) Smart city data sharing system and method
CN114897736A (en) Cloud computing-based hidden image recovery method for big data processing
CN114444940B (en) Enterprise data acquisition and analysis system based on big data
CN105069710A (en) Data processing method and system for intelligent power grid
CN112100646A (en) Spatial data privacy protection matching method based on two-stage grid conversion
Zhang et al. Geo-indistinguishable Mechanisms for Spatial Crowdsourcing via Multi-Objective Evolutionary Optimization
Gu et al. A Spatial-Temporal Transformer Network for City-Level Cellular Traffic Analysis and Prediction
CN204859226U (en) Big data process system of E -Government
CN111460513B (en) Similarity-binning-based space point set data privacy protection matching method
CN111915327A (en) Intelligent distribution method and device based on event characteristics
CN111767567A (en) Social information security management method
CN113255884B (en) Network abnormal traffic identification and classification method based on collaborative learning
CN107943807A (en) A kind of data processing method and storage device
CN112686559A (en) Achievement transformation online supply and demand matching method based on big data and artificial intelligence
Fan et al. A Malicious Traffic Detection Method Based on Attention Mechanism
CN112085815A (en) Polar coordinate image expression method and transformation algorithm thereof
CN112235787A (en) Position privacy protection method based on generation countermeasure network
CN117077161B (en) Privacy protection depth model construction method and system based on dynamic programming solution

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: 20201218