WO2020177326A1 - 一种支持位置保护和查询隐私的可搜索加密方法 - Google Patents
一种支持位置保护和查询隐私的可搜索加密方法 Download PDFInfo
- Publication number
- WO2020177326A1 WO2020177326A1 PCT/CN2019/112048 CN2019112048W WO2020177326A1 WO 2020177326 A1 WO2020177326 A1 WO 2020177326A1 CN 2019112048 W CN2019112048 W CN 2019112048W WO 2020177326 A1 WO2020177326 A1 WO 2020177326A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- interest
- user
- point
- query
- service content
- 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.)
- Ceased
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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 OR CALCULATING; 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
Definitions
- the invention relates to the technical field of information security in the computer field, in particular to a searchable encryption method supporting location protection and query privacy.
- the purpose of the present invention is to overcome the shortcomings of the prior art and provide a searchable encryption method that supports location protection and query privacy.
- the present invention combines the searchable encryption method in cloud computing into the application scenario based on location services, and performs searchable encryption processing for query privacy based on location services.
- a searchable encryption method that supports location protection and query privacy.
- the specific steps include:
- the service provider preprocesses each point of interest; the point of interest includes the identification of the point of interest, the location of the point of interest, and the service content of the point of interest;
- the service provider uploads the preprocessed points of interest to the cloud server and shares the key with the user;
- the user submits a request for searching points of interest to the cloud server;
- the cloud server receives the point of interest search request sent by the user and searches according to the user's location;
- the cloud server generates verification objects, and sends the searched points of interest and verification objects to the user;
- the user receives the points of interest and verification objects searched by the cloud server;
- the user verifies the received points of interest and verification objects, and decrypts them to obtain search results.
- the preprocessing includes the generation of a key, the encryption of the service content of the point of interest, the construction of an index, the generation of a summary, and the fuzzification of the location information of the point of interest.
- the preprocessing method is: the service provider encrypts the service content of the point of interest held by the service provider to obtain the ciphertext point of interest service content set E, and the service provider obscures the location information of the point of interest to obtain a pseudo location L, and according to the obtained pseudo location
- the location L and point of interest service content collection E constructs a security index
- the service provider generates a message summary based on the points of interest, and each point of interest provided by the service provider can be represented as a node u.
- S is an n-bit random vector
- M 1 and M 2 are both n ⁇
- the invertible matrix of n, n represents the size of the generated dictionary, and k f is a symmetric encryption key.
- the service provider uses the hash table to construct the index.
- the Geohash algorithm is used for calculation, and the specific longitude or latitude is given for approximation coding;
- the formula for calculating the service content vector u.PV stored in node u is:
- the segmentation rule is applied to the key vector corresponding to the service content to obtain two random sub-vectors ⁇ P′,P′′ ⁇ , SK. S acts as a segmentation indicator, and the segmentation rules are:
- the service provider uploads the preprocessed points of interest to the cloud server and shares the key with the user.
- the service provider combines the encrypted ciphertext point of interest service content set E with the point of interest message Abstract, pseudo location information L and security index Upload to the cloud server; the key shared by the service provider and the user includes a symmetric key used to encrypt the service content and a secret key used to encrypt the trapdoor.
- the security key SK is only shared between the service provider and the user, and the cloud server does not know any information about the security key SK.
- the user needs to convert the query into a query trapdoor before searching for the interest point, the query trapdoor, the fuzzy result of the user's location information, and the number of the user's target interest points Is submitted to the cloud server.
- the encrypted form of Q is Then the service provider sends the query trapdoor T Q to the cloud server, and T Q includes And the number of target points of interest K to be acquired.
- the cloud server receives the search interest point request sent by the user and performs the retrieval step, the cloud server receives the query trapdoor and the target number sent from the user, and uses the fuzzy location information, ciphertext index and query trapdoor Perform a secure search to get the top-K most relevant encrypted points of interest.
- search method is:
- the cloud server obtains the short-distance result set R according to the user's fuzzy location (Geohash's calculation result) and the secure inner product of the unencrypted keyword set vector in the query content and the keyword vector of the service content, and constructs a verifiable object VO, the cloud server returns the result set R and VO to the user.
- the calculation formula of the correlation score between the encrypted form of u.PV and Q vector is:
- the calculation result indicates that the correlation score between the index and the query trapdoor is equal or proportional to the correlation score between the plaintext service content vector and the query trapdoor.
- the "key” in the hash table is set to the Geohash value of the fuzzy location information, and the "value” is the ciphertext service content, the ciphertext keyword vector of the service content, and the summary data of the points of interest; the target result set is represented by R, and the threshold is represented The minimum value of the correlation score between the node and the query in the current result set, K represents the number of target interest points to be obtained.
- the retrieval range of fuzzy positions can be appropriately expanded.
- the cloud server generates verification objects according to the relevance of the encrypted interest points and the query, and sends the top-K most relevant ciphertext interest points and verification objects to the user.
- the method for the user to verify the service content and the verification object is:
- the user uses the key k f to decrypt the search results and verify the correctness and completeness of the search results.
- Each element of the ciphertext location hash table contains the message digest of the current point of interest.
- the cloud server generates a verification object according to the obtained message digest of the top-K points of interest.
- the user decrypts the content of the points of interest, and generates the message digest of the points of interest in combination with the key k f .
- a new verification object VO' is generated.
- the present invention has the following beneficial effects:
- the present invention uses SecureKNN to realize symmetric multi-keyword ciphertext retrieval, and can realize sorting search results according to the degree of relevance to the query, and can verify the accuracy and completeness of the search results according to the verification object.
- the service provider constructs a ciphertext location hash table index in advance and utilizes it, which can effectively and quickly determine the relevant service content of the location range and improve the retrieval efficiency.
- the present invention uses a diagonal matrix to replace the full matrix, which reduces storage and calculation costs by an order of magnitude, and the time for matrix inversion is also greatly reduced, thereby greatly reducing the preprocessing costs of the service provider; At the same time, under the model of honest-but-curious (honest-but-curious), the adoption of the diagonal matrix does not reduce the security, so the present invention can increase the speed while ensuring the security.
- Figure 1 is a flowchart of a searchable encryption method that supports location protection and query privacy.
- Figure 2 is a system structure diagram of a searchable encryption method that supports location protection and query privacy.
- Fig. 3 is a cloud server and query data organization diagram based on a searchable encryption method that supports location protection and query privacy in the present invention.
- FIG. 1 and 2 respectively, a specific step diagram and corresponding system structure diagram of a searchable encryption method supporting location protection and query privacy.
- the specific steps include:
- the service provider preprocesses each point of interest; the point of interest includes the identification of the point of interest, the location of the point of interest, and the service content of the point of interest;
- the service provider uploads the preprocessed points of interest to the cloud server, and shares the key with the user;
- each service content is very small.
- the entire dictionary has only 6 keywords and 2 redundant keywords are introduced. So the entire generated dictionary size is 8.
- the service content of each point of interest in the service content set is as follows:
- the service content vector generated corresponding to the service content of each point of interest is as follows, where the weight value of the redundant keyword obeys the uniform distribution U(-0.01,0.01).
- the keyword vector corresponding to the service content of each point of interest is set as the service content vector.
- the cloud server receives the point of interest search request sent by the user and searches according to the user's location;
- the cloud server generates verification objects, and sends the searched points of interest and verification objects to the user;
- the user receives the points of interest and verification objects searched by the cloud server;
- the user verifies the received points of interest and verification objects, and decrypts them to obtain search results.
- the submitted query is "restaurant 5 star hotel charging pile"
- the number of target points of interest that need to be obtained is 2
- the user’s interest preference model established according to the user’s search history is assigned to different keywords Different weights are as follows: “Fruit Supermarket”: 2, "Convenience Store”: 5, “Parking Lot”: 1, “Five-star Hotel”: 8, "Restaurant”: 7, "Charging Pile”: 10.
- the query vector Q constructed according to the query submitted by the user and the encrypted forms Q′ and Q′′ of the sub-vectors generated by segmentation of the query vector Q are as follows:
- FIG. 3 shows a diagram of the cloud server and query data organization in this embodiment.
- the geohash algorithm When searching, the geohash algorithm is used to calculate the spatial position of the user's latitude and longitude on the ciphertext position hash table index, and then the Euclidean distance between the user position and the position of the point of interest is calculated according to the hash table index.
- 1000m is set as the default threshold by default. ; When there are not enough top-K service points within the default Euclidean distance, the default threshold is increased by a multiple of twice; within the default Euclidean distance, the encrypted keyword vector and user query trapdoor vector of each service point are calculated separately
- the safe inner product of, the mathematical meaning of the result is the similarity between the interest point and the user's query trapdoor vector.
- the first two ciphertext results are selected according to the similarity: (113.418115, 23.0788) and (113.386638, 23.064037).
- verifiable objects To achieve verifiability, you need to generate verifiable objects from the message digests of the nodes represented by (113.418115, 23.0788) and (113.386638, 23.064037). After sending them to the user, the user needs to decrypt the ciphertext service content and reconstruct each service content According to these message digests, the verifiable object is reconstructed. By determining whether the newly generated verifiable object and the verifiable object sent back by the server are equal to determine whether to accept the query result.
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Bioethics (AREA)
- General Health & Medical Sciences (AREA)
- Computer Security & Cryptography (AREA)
- Computer Hardware Design (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- Medical Informatics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
一种支持位置保护和查询隐私的可搜索加密方法,包括:服务提供者对每项兴趣点信息进行预处理;每项兴趣点均包括兴趣点标识、兴趣点位置、兴趣点服务内容;服务提供者将预处理后的数据上传到云服务器,并与用户共享密钥;用户向云服务器提交基于位置的搜索服务内容请求;云服务器接收用户发送的搜索服务内容请求并根据位置进行搜索;云服务器生成验证对象,并将搜索的服务内容及验证对象发送给用户;用户接收云服务器搜索的服务内容及验证对象;用户对接收的服务内容及验证对象进行验证,并解密得到搜索结果。该方法能够实现按照和查询的相关程度排序搜索结果,同时能够根据验证对象验证搜索结果的准确性和完整性。
Description
本发明涉及计算机领域中的信息安全技术领域,尤其涉及一种支持位置保护和查询隐私的可搜索加密方法。
随着云计算的日益普及,服务提供者为了减少数据管理、存储和计算的开销,降低日常基础设施运营的成本,会将其所拥有的大量的数据外包到云服务器进行存储或者处理。因此虽然云服务器给现如今的商业社会带来了便捷、低廉的数据处理成本,但是服务提供者失去了对数据的物理控制,使得服务提供者的数据被云服务器或者入侵者非法获取、访问。而在基于位置服务(Location-Based Services,LBS)领域,数据隐私问题也成为了LBS应用进一步发展的关键。因此,数据安全问题成为了制约云计算和基于位置服务应用发展的绊脚石,也成为了人们日益关注的话题。
常用的保护数据隐私的机制是将数据上传到云服务器之前,但是传统的通讯加密会极大地限制数据的可用性。而如果采用简单的下载、解密、处理机制的话就需要消耗大量的带宽和用户极大的计算开销,对于云计算这种理念是不适用的。虽然目前也有大量基于同态加密的方法或者基于公钥的可搜索加密方法被提了出来,但是这些方法的计算往往因其巨大的计算开销而使得实用性较低。所以人们仍然关注的是对称可搜索加密。在对称可搜索加密中,也有很多针对单关键字和多关键字的可搜索加密方法以及相应的改进方法被提了出来,但是这些方法的功能性相对较为单一,其中不少也存在较大的效率问题。目前对称可搜索加密的功能性仍然和明文的检索存在着非常大的差距,对称可搜索加密中的功能性如个性化检索、 逻辑检索、语义检索、模糊检索、动态更新等仍然有待进一步的研究。而在基于位置服务领域,针对用户身份隐私、位置隐私、查询隐私等关键隐私信息的隐私保护方法也被提出。
发明内容
本发明的目的在于克服现有技术的不足之处,提供一种支持位置保护和查询隐私的可搜索加密方法。本发明将云计算中的可搜索加密方法结合到基于位置服务的应用场景中,针对基于位置服务中的查询隐私进行可搜索加密处理。
本发明的目的能够通过以下技术方案实现:
一种支持位置保护和查询隐私的可搜索加密方法,具体步骤包括:
服务提供者对每项兴趣点进行预处理;兴趣点包括兴趣点标识、兴趣点位置、兴趣点服务内容;
服务提供者将预处理后的兴趣点上传到云服务器,并与用户共享密钥;
用户向云服务器提交搜索兴趣点请求;
云服务器接收用户发送的搜索兴趣点请求并根据用户位置进行搜索;
云服务器生成验证对象,并将搜索的兴趣点及验证对象发送给用户;
用户接收云服务器搜索的兴趣点及验证对象;
用户对接收的兴趣点及验证对象进行验证,并解密得到搜索结果。
具体地,所述服务提供者对每项兴趣点进行预处理步骤中,预处理包括密钥的生成、兴趣点服务内容的加密、索引的构造、摘要的生成以及兴趣点位置信息的模糊化。
预处理方法为:服务提供者将所持有的兴趣点服务内容加密得到密文兴趣点服务内容集合E,服务提供者对兴趣点的位置信息模糊化得到一个伪 位置L,并根据得到的伪位置L以及兴趣点服务内容集合E构造安全索引
同时服务提供者根据兴趣点生成消息摘要,服务提供者提供的每项兴趣点均可表示为一个节点u。
更进一步地,所述密钥生成的方法为:
初始化阶段,服务提供者生成一个四元组的安全密钥SK=(S,M
1,M
2,k
f),其中,S为一个n比特的随机向量,M
1和M
2均为n×n的可逆矩阵,n表示生成字典的大小,k
f为一个对称加密密钥。
更进一步地,所述构造索引的方法为:
服务提供者采用散列表来构造索引,在构造位置散列表时,采用Geohash算法进行计算,具体通过给定的经度或维度,进行逼近编码;
例如计算39.928167纬度,首先在区间[-90,90]进行二分为[-90,0),[0,90],称为左右区间,可以确定39.928167属于右区间,标记为1;接着将区间[0,90]进行二分为[0,45),[45,90]左右区间,可以确定39.928167属于左区间,标记为0;以此类推,取10位编码,得到10111 00011。
在构造兴趣点的服务内容向量的过程中,服务内容F
d中字典包含的关键字w
i对应的每一个维度的计算公式为:
根据Secure KNN算法,对节点u中存储的服务内容向量u.PV进行计算的公式为:
构造出明文位置散列表索引后,加密得到密文位置散列表索引,加密过程中,针对服务内容对应的关键字向量应用切分规则得到两个随机子向量{P′,P″},SK.S充当切分指示器,切分规则为:
具体地,所述服务提供者将预处理后的兴趣点上传到云服务器,并与用户共享密钥的步骤中,服务提供者将加密后的密文兴趣点服务内容集合E与兴趣点的消息摘要、伪位置信息L及安全索引
上传到云服务器;所述服务提供者与用户共享的密钥包括用于加密服务内容的对称密钥和加密陷门的秘密钥。安全密钥SK仅在服务提供者和用户之间共享,云服务器不知道安全密钥SK的任何信息。
具体地,所述用户向云服务器提交搜索兴趣点请求步骤中,用户在搜索兴趣点前,需要先将查询转换为查询陷门,查询陷门、用户位置信息模糊结果和用户的目标兴趣点数目被提交到云服务器。
更进一步地,生成查询陷门的方法为:
假定S
q={w
1,w
2,…,w
t}表示用户提交的查询内容中未加密的关键字集合,S
q的向量形式为Q,向量Q的每一个维的计算公式为:
然后进行归一化操作,Q被切分成两个随机子向量{Q′,Q"},SK.S充当切分指示器,切分规则如下:
具体地,所述云服务器接收用户发送的搜索兴趣点请求并进行检索步骤中,云服务器接收到来自用户发送过来的查询陷门和目标数目,利用模糊的位置信息、密文索引和查询陷门进行安全检索,得到top-K个最相关的加密兴趣点。
更进一步地,所述搜索方法为:
云服务器根据用户的模糊位置(Geohash的计算结果)和查询内容中未加密的关键字集合向量与服务内容的关键字向量的安全内积来获取近距离的结果集合R,构造一个可验证的对象VO,云服务器返回结果集合R和VO给用户,密文位置散列表索引的搜索算法执行过程中,u.PV和Q向量的加密形式之间相关性评分的计算公式为:
计算结果表示索引与查询陷门之间的相关性评分与明文服务内容向量和查询陷门之间的相关性评分相等或成正比关系。
利用位置散列表索引来加速查询的过程如下:
散列表中的“key”设置为模糊位置信息的Geohash值,“value”为密文服务内容、服务内容的密文关键字向量、兴趣点的摘要数据;目标结果集合用R表示,threshold则表示当前结果集合中节点和查询的相关性得分的最小值,K表示待获取的目标兴趣点数目。在检索阶段,在指定范围内的top-K个太低或者指定位置内不足top-K个兴趣点时,可以适当扩大模糊位置的检索范围。
具体地,云服务器按加密兴趣点与查询的相关性大小排序后生成验证对象,将top-K个最相关的密文兴趣点和验证对象发送给用户。
更进一步地,所述用户对服务内容及验证对象进行验证的方法为:
用户采用密钥k
f来解密搜索结果并验证搜索结果的正确性和完整性,密文位置散列表的每一个元素都包含当前兴趣点的消息摘要。云服务器根据获取的top-K个兴趣点的消息摘要生成验证对象,用户接收到top-K个密文兴趣点以及验证对象后,解密兴趣点内容,结合密钥k
f生成兴趣点的消息摘要。根据新生成的兴趣点的消息摘要生成新的验证对象VO′,通过判定云服务器返回的验证对象和用户新生成的验证对象是否相等,即VO′是否等于VO,用户决定是否接受此次查询的结果。
本发明相较于现有技术,具有以下的有益效果:
1、本发明采用了SecureKNN来实现了对称多关键字密文检索,并且能够实现按照和查询的相关程度排序搜索结果,同时能够根据验证对象验证搜索结果的准确性和完整性。
2、在本发明中,服务提供者通过预先构造了密文位置散列表索引并对其进行利用,能够有效地快速确定位置范围的相关服务内容,提升了检索的效率。
3、本发明采用对角矩阵来替代满矩阵,使得存储开销和计算开销降低了一个数量级,并且矩阵求逆的时间也极大减小,从而极大地降低了服务提供者的预处理的开销;同时在半可信(honest-but-curious即诚实但是好奇)的模型下,采用对角矩阵并未降低安全性,因此本发明能够在保证安全性的同时提升速度。
图1是一种支持位置保护和查询隐私的可搜索加密方法的流程图。
图2是一种支持位置保护和查询隐私的可搜索加密方法的系统结构图。
图3是本发明中基于一种支持位置保护和查询隐私的可搜索加密方法的云服务器以及查询数据组织形式图。
下面结合实施例及附图对本发明作进一步详细的描述,但本发明的实施方式不限于此。
实施例
如图1、2所述的分别为一种支持位置保护和查询隐私的可搜索加密方法的具体步骤图及对应系统结构示意图。
在本实施例中,具体步骤包括:
(1)服务提供者对每项兴趣点进行预处理;兴趣点包括兴趣点标识、兴趣点位置、兴趣点服务内容;
(2)服务提供者将预处理后的兴趣点上传到云服务器,并与用户共享密钥;
在本实施例中,为了方便说明程序的流程,各个服务内容都非常小。整个字典只有6个关键词,引入2个冗余关键词。所以整生成字典大小是 8。服务内容集合中各个兴趣点服务内容的内容如下:
(113.407695,23.059648):五星级酒店 餐厅
(113.386638,23.064037):餐厅 充电桩
(113.418115,23.0788):五星级酒店 充电桩
(113.321242,23.10114):便利店
(113.343807,23.069091):水果超市
(113.33619,23.092896):停车场
(113.319275,23.149813):五星级酒店 便利店 水果超市
(113.304641,23.070022):五星级酒店 充电桩 餐厅
生成字典后是:[水果超市,便利店,充电桩,餐厅,停车场,mugvnxze,五星级酒店,pzfv],其中“mugvnxzeh”和“pzfv”是引入的冗余关键词。
各个兴趣点服务内容对应生成的服务内容向量如下,其中冗余关键词的权重值服从均匀分布U(-0.01,0.01)。构造密文位置散列表索引的过程中,每个兴趣点服务内容对应的关键字向量设置为此服务内容向量。
(113.407695,23.059648)
五星级酒店:0.5493061443340549
餐厅:0.6496414920651304
[0.000000,0.000000,0.000000,0.649641,0.000000,-0.007514,0.549306,0.003004]
(113.386638,23.064037)
餐厅:0.6496414920651304
充电桩:0.6496414920651304
[0.000000,0.000000,0.649641,0.649641,0.000000,0.008282,0.000000,0.0 03478]
(113.418115,23.0788)
五星级酒店:0.5493061443340549
充电桩:0.6496414920651304
[0.000000,0.000000,0.649641,0.000000,0.000000,-0.008594,0.549306,-0.004946]
(113.321242,23.10114)
便利店:1.6094379124341003
[0.000000,1.609438,0.000000,0.000000,0.000000,-0.006176,0.000000,-0.008033]
(113.343807,23.069091)
水果超市:1.6094379124341003
[1.609438,0.000000,0.000000,0.000000,0.000000,0.003996,0.000000,0.007028]
(113.33619,23.092896)
停车场:2.1972245773362196
[0.000000,0.000000,0.000000,0.000000,2.197225,0.002741,0.000000,0.006191]
(113.319275,23.149813)
五星级酒店:0.3662040962227032
便利店:0.5364793041447
水果超市:0.5364793041447
[0.536479,0.536479,0.000000,0.000000,0.000000,-0.004668,0.366204,0. 000613]
(113.304641,23.070022)
五星级酒店:0.3662040962227032
餐厅:0.4330943280434203
充电桩:0.4330943280434203
[0.000000,0.000000,0.433094,0.433094,0.000000,-0.006085,0.366204,-0.003783]
加密明文位置散列表索引得到密文位置散列表索引,即根据切分规则切分每个节点中的修剪向量得到两个子向量P′、P″,然后使用可逆矩阵的转置矩阵加密两个子向量,并将相关的字段设置为NULL。
(3)用户向云服务器提交搜索兴趣点请求;
(4)云服务器接收用户发送的搜索兴趣点请求并根据用户位置进行搜索;
(5)云服务器生成验证对象,并将搜索的兴趣点及验证对象发送给用户;
(6)用户接收云服务器搜索的兴趣点及验证对象;
(7)用户对接收的兴趣点及验证对象进行验证,并解密得到搜索结果。
在本实施例中,提交的查询是"餐厅 五星级酒店 充电桩",需要获取的目标兴趣点个数是2,根据用户的搜索历史建立的用户的兴趣偏好模型,赋予不同的关键词以不同的权重,权重如下:"水果超市":2,"便利店":5,"停车场":1,"五星级酒店":8,"餐厅":7,"充电桩":10。那么根据用户提交的查询构造的查询向量Q以及查询向量Q经切分生成的子向量的加密形式Q′和Q″分别如下所示:
[0.000000,0.000000,115.059300,1.000000,0.000000,1.000000,19.450359,0.000000]
[174.797226,-190.718486,-16.424931,118.891982,-10.095257,58.659643,11.118955,-110.229204]
[2546.835577,-1077.082690,1838.242043,389.895225,-2904.909899,-1202.838724,1340.954562,-498.161811]
如图3所示为本实施例中云服务器以及查询数据组织形式图。
检索的时候,在密文位置散列表索引上采用geohash算法计算出用户经纬度的空间位置,然后根据散列表索引,计算用户位置与兴趣点的位置中的欧氏距离,这里默认设置1000m作为默认阈值;当默认欧式距离内没有足够的top-K个服务点,则以两倍的倍数增长默认阈值;在默认欧式距离内,分别计算每个服务点的加密的关键字向量和用户查询陷门向量的安全内积,其结果的数学意义就是兴趣点与用户查询陷门向量的相似程度,最后根据相似程度选出前2个密文结果:(113.418115,23.0788)和(113.386638,23.064037)。
要实现可验证性,需要将(113.418115,23.0788)和(113.386638,23.064037)所代表的节点的消息摘要生成可验证对象,发送到用户以后,用户需要解密密文服务内容,重新构造每篇服务内容的消息摘要,并根据这些消息摘要重新构造可验证对象,通过判定新生成的可验证对象和服务器发回的可验证对象是否相等以决定是否接受此次查询结果。
上述实施例为本发明较佳的实施方式,但本发明的实施方式并不受上述实施例的限制,其他的任何未背离本发明的精神实质与原理下所作的改变、修饰、替代、组合、简化,均应为等效的置换方式,都包含在本发明 的保护范围之内。
Claims (10)
- 一种支持位置保护和查询隐私的可搜索加密方法,其特征在于,具体步骤包括:服务提供者对每项兴趣点进行预处理,兴趣点包括兴趣点标识、兴趣点位置、兴趣点服务内容;服务提供者将预处理后的兴趣点上传到云服务器,并与用户共享密钥;用户向云服务器提交兴趣点搜索请求;云服务器接收用户发送的兴趣点搜索请求并根据用户位置进行搜索;云服务器生成验证对象,并将搜索的兴趣点及验证对象发送给用户;用户接收云服务器搜索的兴趣点及验证对象;用户对接收的兴趣点及验证对象进行验证,并解密得到搜索结果。
- 根据权利要求2所述的一种支持位置保护和查询隐私的可搜索加密方法,其特征在于,所述密钥的生成方法为:初始化阶段,服务提供者生成一个四元组的安全密钥SK=(S,M 1,M 2,k f),其中,S为一个n比特的随机向量,M 1和M 2均为n×n的可逆矩阵,n表示生成字典的大小,k f为一个对称加密密钥。
- 根据权利要求2所述的一种支持位置保护和查询隐私的可搜索加密方法,其特征在于,所述索引的构造方法为:服务提供者采用散列表来构造索引,在构造位置散列表时,采用 Geohash算法进行计算,具体通过给定的经度或维度,进行逼近编码;由兴趣点服务内容构造服务内容向量的过程中,服务内容向量中字典包含的关键字对应的每一个维度的计算公式为:其中, 表示关键词w i出现在兴趣点服务内容F d中的频率, 表示包含了关键词w i的兴趣点服务内容数量,N表示兴趣点服务内容集合中服务内容的个数,|F d|表示兴趣点服务内容F d的长度,即包含的关键词的数目;根据Secure KNN算法,对节点u中存储的每一个服务内容向量u.PV进行计算的公式为:构造出服务内容向量后,加密得到密文服务内容关键字向量,加密过程中,针对兴趣点服务内容对应的关键字向量应用切分规则得到两个随机子向量{P′,P″},SK.S充当切分指示器,切分规则为:
- 根据权利要求2所述的一种支持位置保护和查询隐私的可搜索加密方法,其特征在于,所述服务提供者与用户共享的密钥包括用于加密服务内容的对称密钥和加密陷门的秘密钥;安全密钥SK仅在服务提供者和用户之间共享,云服务器不知道安全密钥SK的任何信息。
- 根据权利要求1所述的一种支持位置保护和查询隐私的可搜索加密方法,其特征在于,所述用户向云服务器提交搜索兴趣点请求步骤中,用 户在搜索兴趣点前,需要先将查询转换为查询陷门,查询陷门、用户位置信息模糊结果和用户的目标兴趣点数目被提交到云服务器。
- 根据权利要求1所述的一种支持位置保护和查询隐私的可搜索加密方法,其特征在于,所述云服务器接收用户发送的搜索兴趣点请求并进行检索步骤中,云服务器接收到来自用户发送过来的查询陷门和目标兴趣点数目,利用模糊的位置信息、密文索引和查询陷门进行安全检索,得到top-K个最相关的加密兴趣点。
- 根据权利要求8所述的一种支持位置保护和查询隐私的可搜索加密方法,其特征在于,所述搜索方法为:云服务器根据用户的模糊位置和查询内容中未加密的关键字集合向量与服务内容的关键字向量的安全内积来获取近距离的结果集合R,构造一个可验证的对象VO,云服务器返回结果集合R和VO给用户,密文位置散列表索引的搜索算法执行过程中,u.PV和Q向量的加密形式之间相关性评分的计算公式为:利用位置散列表索引来加速查询的过程如下:散列表中的“key”设置为模糊位置信息的Geohash值,“value”为密文服务内容、服务内容的密文关键字向量、兴趣点的摘要数据;目标结果集合用R表示,threshold则表示当前结果集合中节点和查询的相关性得分的最小值,K表示待获取的目标兴趣点数目;在检索阶段,在指定范围内的top-K个太低或者指定位置内不足top-K个兴趣点时,适当扩大模糊位置的检索范围。
- 根据权利要求1所述的一种支持位置保护和查询隐私的可搜索加密方法,其特征在于,所述用户对接收的兴趣点及验证对象进行验证,并解密得到搜索结果步骤中,用户对兴趣点及验证对象进行验证的方法为:用户采用密钥k f来解密搜索结果并验证搜索结果的正确性和完整性,密文位置散列表的每一项元素都包含当前兴趣点的消息摘要;云服务器根据获取的top-K个兴趣点的消息摘要生成验证对象,用户接收到top-K个密文兴趣点以及验证对象后,解密每个密文兴趣点,结合密钥k f生成兴趣点的消息摘要;根据新生成的兴趣点的消息摘要生成新的验证对象VO′,通过判定云服务器返回的验证对象和用户新生成的验证对象是否相等,即VO′是否等于VO,用户决定是否接受此次查询的结果。
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910162593.4A CN109992995B (zh) | 2019-03-05 | 2019-03-05 | 一种支持位置保护和查询隐私的可搜索加密方法 |
| CN201910162593.4 | 2019-03-05 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2020177326A1 true WO2020177326A1 (zh) | 2020-09-10 |
Family
ID=67130382
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2019/112048 Ceased WO2020177326A1 (zh) | 2019-03-05 | 2019-10-18 | 一种支持位置保护和查询隐私的可搜索加密方法 |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN109992995B (zh) |
| WO (1) | WO2020177326A1 (zh) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116846604A (zh) * | 2023-06-13 | 2023-10-03 | 北京五八信息技术有限公司 | 一种信息保护方法、装置、设备及存储介质 |
Families Citing this family (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109992995B (zh) * | 2019-03-05 | 2021-05-14 | 华南理工大学 | 一种支持位置保护和查询隐私的可搜索加密方法 |
| CN110602099B (zh) * | 2019-09-16 | 2021-04-06 | 广西师范大学 | 基于可验证的对称可搜索加密的隐私保护方法 |
| CN112668015B (zh) * | 2019-12-12 | 2022-02-01 | 华控清交信息科技(北京)有限公司 | 一种数据处理方法、装置和用于数据处理的装置 |
| CN112257455B (zh) * | 2020-10-21 | 2024-04-30 | 西安电子科技大学 | 一种语义理解的密文空间关键字检索方法及系统 |
| CN112507362B (zh) * | 2020-12-14 | 2024-12-06 | 重庆邮电大学 | 一种数据外包隐私保护方法、系统及存储介质 |
| CN112488585B (zh) * | 2020-12-18 | 2024-01-12 | 湖南大学 | 众包环境下基于阈值相似性搜索的隐私保护任务匹配 |
| CN113068128B (zh) * | 2021-03-18 | 2021-11-23 | 西安电子科技大学 | 基于双云安全计算协议的用户地理位置近邻查询方法 |
| CN116156489B (zh) * | 2022-12-21 | 2026-03-31 | 广州大学 | 一种基于协同服务的位置隐私保护方法 |
| CN117272391B (zh) * | 2023-11-20 | 2024-02-27 | 哈尔滨工业大学(深圳)(哈尔滨工业大学深圳科技创新研究院) | 一种兴趣点查询方法及设备 |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120158845A1 (en) * | 2010-12-20 | 2012-06-21 | Harman International Industries, Incorporated | System and method for automatically posting user-location information to a social network |
| CN104092692A (zh) * | 2014-07-15 | 2014-10-08 | 福建师范大学 | 一种基于k-匿名与服务相似性相结合的位置隐私保护方法 |
| CN104507049A (zh) * | 2014-12-24 | 2015-04-08 | 福建师范大学 | 一种基于坐标变换的位置隐私保护方法 |
| CN105592085A (zh) * | 2015-12-23 | 2016-05-18 | 西安电子科技大学 | 面向位置感知推荐系统的隐私保护方法 |
| CN108388807A (zh) * | 2018-02-28 | 2018-08-10 | 华南理工大学 | 一种支持偏好搜索和逻辑搜索的高效可验证的多关键字排序可搜索加密方法 |
| CN109992995A (zh) * | 2019-03-05 | 2019-07-09 | 华南理工大学 | 一种支持位置保护和查询隐私的可搜索加密方法 |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| FR3040842B1 (fr) * | 2015-09-03 | 2018-12-07 | Commissariat A L'energie Atomique Et Aux Energies Alternatives | Methode d'interrogation confidentielle d'un service geodependant par cryptographie homomorphe |
| CN105791283B (zh) * | 2016-02-29 | 2018-09-21 | 电子科技大学 | 一种针对加密的空间数据的圆形范围搜索方法 |
| CN107220343B (zh) * | 2017-05-26 | 2020-09-01 | 福州大学 | 基于局部敏感哈希的中文多关键词模糊排序密文搜索方法 |
| CN108494768B (zh) * | 2018-03-22 | 2021-07-23 | 深圳大学 | 一种支持访问控制的密文搜索方法及系统 |
| CN109145079B (zh) * | 2018-07-24 | 2022-07-19 | 南京邮电大学 | 基于个人兴趣用户模型的云端可搜索加密方法 |
-
2019
- 2019-03-05 CN CN201910162593.4A patent/CN109992995B/zh not_active Expired - Fee Related
- 2019-10-18 WO PCT/CN2019/112048 patent/WO2020177326A1/zh not_active Ceased
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120158845A1 (en) * | 2010-12-20 | 2012-06-21 | Harman International Industries, Incorporated | System and method for automatically posting user-location information to a social network |
| CN104092692A (zh) * | 2014-07-15 | 2014-10-08 | 福建师范大学 | 一种基于k-匿名与服务相似性相结合的位置隐私保护方法 |
| CN104507049A (zh) * | 2014-12-24 | 2015-04-08 | 福建师范大学 | 一种基于坐标变换的位置隐私保护方法 |
| CN105592085A (zh) * | 2015-12-23 | 2016-05-18 | 西安电子科技大学 | 面向位置感知推荐系统的隐私保护方法 |
| CN108388807A (zh) * | 2018-02-28 | 2018-08-10 | 华南理工大学 | 一种支持偏好搜索和逻辑搜索的高效可验证的多关键字排序可搜索加密方法 |
| CN109992995A (zh) * | 2019-03-05 | 2019-07-09 | 华南理工大学 | 一种支持位置保护和查询隐私的可搜索加密方法 |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116846604A (zh) * | 2023-06-13 | 2023-10-03 | 北京五八信息技术有限公司 | 一种信息保护方法、装置、设备及存储介质 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109992995A (zh) | 2019-07-09 |
| CN109992995B (zh) | 2021-05-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109992995B (zh) | 一种支持位置保护和查询隐私的可搜索加密方法 | |
| Zhang et al. | PIC: Enable large-scale privacy preserving content-based image search on cloud | |
| CN111026788B (zh) | 一种混合云中基于同态加密的多关键词密文排序检索方法 | |
| CN107480163B (zh) | 一种云环境下支持隐私保护的高效密文图像检索方法 | |
| Su et al. | Privacy-preserving top-k spatial keyword queries in untrusted cloud environments | |
| WO2019165880A1 (zh) | 一种支持偏好搜索和逻辑搜索的高效可验证的多关键字排序可搜索加密方法 | |
| CN110138561A (zh) | 基于cp-abe自动校正的高效密文检索方法、云计算服务系统 | |
| CN106599311A (zh) | 一种基于云计算的互联网教育平台资源库的图像检索方法 | |
| CN107948146A (zh) | 一种混合云中基于属性加密的连接关键词检索方法 | |
| CN108959567A (zh) | 一种云环境下适用于大规模图像的安全检索方法 | |
| JP2024073565A (ja) | プライバシーを守る機械学習ラベリング | |
| CN111966778B (zh) | 一种基于关键词分组倒排索引的多关键词密文排序检索方法 | |
| Zou et al. | Efficient and secure encrypted image search in mobile cloud computing | |
| Shen et al. | Preferred keyword search over encrypted data in cloud computing | |
| CN111859421B (zh) | 一种基于词向量的多关键字密文存储、检索方法及系统 | |
| US11461551B1 (en) | Secure word search | |
| CN109885650A (zh) | 一种外包云环境隐私保护密文排序检索方法 | |
| CN111339539A (zh) | 一种多用户环境下的高效加密图像检索方法 | |
| CN118535759A (zh) | 一种基于树结构的隐私保护的跨模态语义检索方法 | |
| Tian et al. | A privacy-preserving image retrieval scheme with access control based on searchable encryption in media cloud | |
| CN118332570A (zh) | 一种隐私保护跨模态检索方法与系统 | |
| CN114386065A (zh) | 一种中文多关键字权重模糊搜索隐私保护方法 | |
| Li et al. | Segmented hash-based privacy-preserving image retrieval scheme in cloud-assisted iot | |
| Manasrah et al. | A privacy-preserving multi-keyword search approach in cloud computing: AM Manasrah et al. | |
| Guo et al. | Privacy preserving weighted similarity search scheme for encrypted data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 19917987 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 19917987 Country of ref document: EP Kind code of ref document: A1 |






