Security and Communication Networks

Security and Communication Networks / 2018 / Article
Special Issue

Cyberspace Security for Future Internet

View this Special Issue

Research Article | Open Access

Volume 2018 |Article ID 4724815 |

Lu Ou, Hui Yin, Zheng Qin, Sheng Xiao, Guangyi Yang, Yupeng Hu, "An Efficient and Privacy-Preserving Multiuser Cloud-Based LBS Query Scheme", Security and Communication Networks, vol. 2018, Article ID 4724815, 11 pages, 2018.

An Efficient and Privacy-Preserving Multiuser Cloud-Based LBS Query Scheme

Academic Editor: Zhenyu Li
Received24 Sep 2017
Accepted31 Dec 2017
Published08 Mar 2018


Location-based services (LBSs) are increasingly popular in today’s society. People reveal their location information to LBS providers to obtain personalized services such as map directions, restaurant recommendations, and taxi reservations. Usually, LBS providers offer user privacy protection statement to assure users that their private location information would not be given away. However, many LBSs run on third-party cloud infrastructures. It is challenging to guarantee user location privacy against curious cloud operators while still permitting users to query their own location information data. In this paper, we propose an efficient privacy-preserving cloud-based LBS query scheme for the multiuser setting. We encrypt LBS data and LBS queries with a hybrid encryption mechanism, which can efficiently implement privacy-preserving search over encrypted LBS data and is very suitable for the multiuser setting with secure and effective user enrollment and user revocation. This paper contains security analysis and performance experiments to demonstrate the privacy-preserving properties and efficiency of our proposed scheme.

1. Introduction

Location-based services (LBSs) are increasingly popular in today’s society. It is reported that up to 150 million people have enjoyed LBSs as early as 2014 [1]. People reveal their location information to LBS providers to obtain personalized services such as map directions, restaurant recommendations, and taxi reservations.

The most common and most important service in an LBS system is a location query service. In LBS applications, a user is able to use his powerful smartphone equipped with GPS modules to obtain accurate location information anytime and anywhere by submitting a query keyword of interest (e.g., hotel) to the LBS system. Upon receiving an location query request from the user, an LBS provider rapidly returns back a target location list, in which all locations are ranked in an ascending order based on distances between the query user and these target locations.

Every coin has two sides: although the LBS greatly facilitates people’s life nowadays, the user privacy disclosure problems for LBS applications are more and more serious. The LBS providers can mine LBS users’ privacy by analyzing LBS queries or recovering the spatial correlated data [2]. For example, LBS providers can easily obtain user’s mobility trace or even infer user’s real identity, healthy status, hobbies, and so on [3, 4]. To address the privacy challenge in the LBS system, many solutions have been proposed such as the pseudoidentity technique [5], location fuzzy [69], and private information retrieval in the trusted third party (TTP) [5, 7]. These schemes significantly promote the further development of LBS applications.

With the rapid development of the cloud computing, more and more LBS providers are beginning to consider to outsource their location data and services to the cloud server for enjoying the numerous advantages brought by the cloud computing such as economic savings, great flexibility, quick deployment, excellent computation performance, and abundant bandwidth resources. However, the cloud server is not fully trusted usually by the LBS providers due to being operated by the remote commercial organizations. Once the location data is outsourced to the cloud server in the plaintext form, the data security would not be guaranteed. For example, a corrupted administrator of the cloud server may sell the location data outsourced by the LBS provider to other one for obtaining illegal profit. Presently, the most effective way to protect the confidentiality of outsourced location data is to encrypt data before outsourcing it to the cloud server [10]. On the other hand, the bare user query requests also provide more opportunities for the cloud server to mine user’s privacy just like a traditional LBS system. Therefore, the user requests should also be encrypted before being submitted to the cloud server. However, data encryption makes the available location query service a challenging task, since the ciphertext no longer bears the natures of numerical computation and character match in the plaintext field. Therefore, there are two essential problems that need to be solved in a cloud-based LBS application over the encrypted outsourced location data: how to find out all target locations over the encrypted location data according to the encrypted user request; how to compute or compare distances between these target locations and user current location over the encrypted outsourced location data.

A recent work in [11] sets out to explore the challenging issue that how to implement the cloud-based LBS system over encrypted location data and proposes a privacy-preserving cloud-based LBS query scheme, called “EPQ.” The scheme enables the cloud server to perform LBS query over the encrypted LBS data without divulging users’ location information. However, the scheme only can enforce a user location coordinate query according to a user’s current location. In a practical LBS application, a goal-oriented keyword query is necessary for the user to accurately locate locations of interest (e.g., the user may need to accurately search hotels near to him/her). Compared with the existing work, in this paper, we propose an efficient and secure keyword-based query scheme that allows the user to be able to first accurately locate desirable locations according to the encrypted query request and then rank distances between these target locations and the user’s current location, which greatly improves the user’s location server experiences. Moreover, our scheme is very suitable for a multiuser cloud environment by equipping with flexible users enrollment and users revocation mechanisms.

In this paper, we make the following three key contributions:(i)First, we propose an efficient and privacy-preserving cloud-based LBS query scheme. For protecting the security of location data and user requests against the curious cloud server, we adopt a hybrid encryption to encrypt the outsourced location data and user requests while the cloud server can still provide accurately LBS query services for users by performing privacy-preserving and efficient search over the encrypted locations data. In addition, our scheme is very suitable for the multiuser setting by equipping with flexible user enrollment and user revocation mechanisms.(ii)Second, we provide detailed correction analysis and security analysis. The analyses show that our scheme is correct and can achieve user privacy preservation and confidentiality of LBS data, simultaneously.(iii)Lastly, we implement our scheme in Java and evaluate the performance on a real data set. Experimental results demonstrate that our proposed scheme is efficient and practical.

The rest of our paper is organized as follows. In Section 2, we review some related literatures. In Section 3, we recall a bilinear pairing map, secure kNN, and a difficulty assumption of discrete logarithm problem as the preliminaries. Then, we formalize a system model and a threat model and depict problem statements in Section 4. We present our approach in Section 5. What is more, some analyses and performance evaluations are conducted in Sections 6 and 7, respectively. Finally, we draw our conclusions in Section 8.

In this section, we review some related works about privacy protection in traditional LBSs and cloud-based LBSs, respectively.

2.1. Traditional LBS Privacy Protection

Privacy leakage problem in traditional LBSs has drawn much attention of researchers, and we review mainly related literatures.

Firstly, a location -anonymity model is introduced, which guarantees that an individual cannot be identified from other individuals [12]. What is more, in a distributed environment, an anonymous approach based on homomorphic encryption [13] is proposed to protect location privacy. However, when the anonymous region is sensitive or individuals are the same place, the sensitive location will still be leaked. Thus the third party (TTP) is proposed to manage the location information centrally [1416]. To achieve an accurate query, a method is proposed to convert original locations of LBS data and query, maintaining a spatial relationship between the LBS data and query [16]. However, because of many users’ sensitive information in the TTP, attackers would aim at attacking it easily. Then a scheme without the TTP is proposed, which protects the locations through private information retrieval [7]. Recently, considering mobile nodes, a distributed anonymizing protocol based on peer-to-peer architecture is proposed [17]. A specific zone is responded by each mobile node. Besides, an information-theoretic notion was introduced to protect privacy in LBS systems [18]. An approach is proposed to protect both client’s location privacy and the server’s database security by improving the oblivious transfer protocol [19]. For providing privacy-preserving map services, a new multiple mix zones location privacy protection is proposed. By using this method, users are able to query a route between two endpoints on the map, without revealing any confidential location and query information [20].

2.2. Cloud-Based LBS Privacy Protection

Considering the low computation and communication cost, the LBS providers outsource the LBS data to the cloud server to compute accurate LBS queries, whereas the cloud server is semitrusted. Hence the privacy problem is still a challenge in the cloud-based LBS. There are some literatures about this problem. Firstly, a spatiotemporal predicate-based encryption is proposed for fine-grained access control [21]. Then an improved homomorphic encryption [11] is proposed to protect users’ privacy and LBS data privacy. A privacy extension in crowdsourced LBS [22] is proposed. To handle the long-term privacy protection and fake path generation for LBS, a dummy-based long-term location privacy protection [23] is proposed. Recently, two-tier lightweight network coding based on pseudonym scheme in a group LBS [24] is proposed to protect privacy. What is more, a query scheme by using Bloom filter and bilinear pairing is proposed [25]. However, the literatures above did not consider the multiusers condition (i.e., joining of registered users and revocation of expired users). But unregistered users and expired users access for cloud-based LBSs is a typical scenario. Therefore, providing an efficient and privacy-preserving cloud-based LBS in multiuser environments is an unnegligible issue.

3. Preliminaries

In this section, we introduce several necessary tools used in our scheme, including a bilinear pairing map, secure kNN computation techniques, and the difficulty assumption of discrete logarithm problem.

3.1. Bilinear Pairing Map

Let and be two multiplication cyclic groups with large prime order . A bilinear pairing map [26, 27] satisfies the following properties:(i)Bilinear: for all and , .(ii)Nondegenerate: if , then .(iii)Computable: for any element , there exists a polynomial time algorithm to compute .

3.2. Secure kNN

Secure kNN [28] enables an efficient kNN computation over encrypted data points. It adopts an asymmetric scalar-product-preserving encryption (ASPE) to achieve a distance comparison between two encrypted data vectors. We synoptically introduce the principle of this technique as follows.

Definition 1 (asymmetric scalar-product-preserving encryption). Let be an encryption function and be an encryption of a point under a key . is an ASPE if and only if just preserves the scalar product between encrypted data points and an encrypted query points; that is, (1), where is one encrypted data point and is one encrypted query point;(2), where and are two encrypted data points.For ease of understanding, we describe the APSE scheme in Algorithm 1. As shown in Algorithm 1, this scheme includes five parts, that is, a key, a tuple encryption function, a query encryption function, a distance comparison operator, and a decryption function.

APSE Scheme
Key: a invertible matrix .
Tuple Encryption Function : Consider an LBS data which will be stored in a cloud server.
(1) Create a -dimensional point .
(2) The encrypted data .
Query Encryption Function  : Consider an LBS query .
(1) Generate a random number .
(2) Create a -dimensional point .
(3) The encrypted query .
Distance Comparison Operator  :
Let and be the encrypted data of and respectively. To determine whether is nearer to a
query than is, the system checks whether , where is the encrypted point of .
Decryption Function  : Consider an encrypted data .
The original data , where is a matrix which projects on the first dimensions
and where is the identity matrix.
3.3. Difficulty Assumption of Discrete Logarithm Problem

Given a multiplication group with the prime order , is its generator. An element is selected from randomly, computing . The definition of the difficulty assumption of discrete logarithm problem (DLP) is as follows.

Definition 2 (difficulty assumption of discrete logarithm problem). Given and , it is difficult to compute the correct value of . In other words, given a tuple , there is not an efficient polynomial time algorithm to output .

4. Background

In this section, we formally introduce our system model and threat model and then state our proposed problem.

4.1. System Model

In our system model, there are three entities: an LBS provider, a cloud server, and a group of LBS users, as shown in Figure 1. Next, we introduce each entity of our model as follows.

(i) LBS Provider. An LBS provider is a location data owner. It outsources large-scale location data to the cloud server for enjoying the low-cost storage services and powerful computation services. To ensure the confidentiality of location data, all location data is uploaded to the cloud server after being encrypted by the LBS provider. In addition, when an LBS user wants to join the system, the LBS provider provides authentication and registration service for the LBS user. Once the LBS user passes authentication, the LBS provider sends some important security parameters to the user via secure communication channels. Correspondingly, the LBS provider is also able to revoke any expired LBS user, who no longer has the query capabilities for the outsourced location data when being revoked by the LBS provider.

(ii) A Group of LBS Users. A group of LBS users are the location data users, who enjoy convenient LBSs by submitting LBS query requests to the LBS provider anywhere and anytime. To hide query requests of LBS users for protecting privacy, LBS users first encrypt their query requests and then submit the encrypted query requests to the cloud server. Note that the LBS users are usually referred to the legal registered users and unregistered users and revoked users from the provider cannot enjoy LBSs.

(iii) Cloud Server. Upon receiving the encrypted LBS query request submitted by a legal LBS user, the cloud server is responsible for performing the query over encrypted outsourced location data on behalf of the LBS user and returning the satisfied query results to the LBS user. In the whole query processes, the cloud server does not know any contents about outsourced location data, the user’s query request, and the current location of the LBS user.

4.2. Problem Statements

In a conventional LBS system, the LBS data construction usually is organized as the category set and the location data set, as shown in Table 1(a). A denotes the general name of location data, which contains multiple concrete location data. Each concrete location data is a four-tuple , which describes the detailed information of a certain location. When a registered user searches an interested location, he/she submits the specified CATEGORY and his current location coordinates to the LBS system. The LBS system first searches over the category set according to the submitted CATEGORY to obtain all target locations and then sorts target locations in an ascending order based on the distances between the user’s current location and these target locations, which are easily computed according to the user’s coordinate and each target location’s coordinate, and finally returns the first nearest locations to the query user, if the query user sends an optional parameter to the LBS system. It means that the LBS system can analyze what are the LBS user interested in and his/her real-time location, when receiving an LBS query.


Location Data Set

  , Hunan University, , ,
  , Center South University, , .
  , the Second Xiangya Hospital, , III-A ,
  , the Union Hospital, , III-A .
  , Huatian Hotel, , ,
  , Westin Hotel, , .


Location Data Set

   , (Hunan University), , ,
   , (Center South University), , .
   , (the Xiangya Hospital), , ( III-A ,
   , (the Union Hospital), , ( III-A
   , (Huatian Hotel), , (  ,
   , (Westin Hotel), , (  .

To ensure the confidentiality of LBS data and enable registered users to enforce efficient location query in the privacy-preserving manner when the LBS provider outsources LBS data and query services to the cloud server, in this paper, we adopt a hybrid encryption mechanism to encrypt the LBS data; the encryption version of LBS data is shown in Table 1(b), where , , and denote different encryption scheme, respectively, whose constructions will be formally proposed in the next section. By encrypting different fields of LBS data using the different encryptions , , and , our scheme allows the cloud server to provide totally the same query service over encrypted location data as the plaintext environment aforementioned while information about the location data and user’s query request is exposed to the cloud server.

From the point of view of LBS users, compared with the LBS system in the plaintext environment, the only difference in our scheme is that an LBS user needs to encrypt the interested and his/her location coordinates to generate a query trapdoor. The cloud server performs the LBS query over encrypted outsourced location data according to the query trapdoor. Of course, the necessary decryption operations need to be involved for the LBS user once the encrypted LBS query results are received; however, this is not our concerned problem in this paper.

In addition, the LBS system is a typical multiuser application, our scheme designs efficient and flexible user registration and user revocation mechanisms to guarantee that only registered users are able to use the LBS system and unregistered users or revoked users have not access to it.

5. A Privacy-Preserving Multiuser LBS Query Scheme Based on Hybrid Encryption

In this section, we describe the implementation details of our privacy-preserving multiuser LBS query scheme. From the system-level point of view, our scheme includes six key modules: system initialization, new user grant, location data encryption, query trapdoor generation, search over encrypted location data, and user revocation. Each module is operated by one entity independently or multiple entities interactively and all modules integrally constitute our privacy-preserving multiuser LBS query system.

5.1. System Initialization

The system initialization operation is executed by the LBS provider to set up the system running environment. The LBS provider takes a large security parameter as input and first generates two multiplication cyclic groups and with the large prime order equipping the bilinear pairing map . Let be a generator of . Then, the algorithm defines a cryptographical hash function , which maps a message of arbitrary length to an element in . Lastly, the algorithm chooses a random value and generates a invertible matrix that are kept secretly by the LBS provider and opens the public parameter .

5.2. New User Grant

When a new LBS user wants to join the system, the LBS provider registers the new user in this phase. At first, the LBS provider selects a random value for and computes and the inverse matrix of . Then, , , and are sent to the user by secure communication channels. When receives , , and , he/she randomly selects and keeps , secretly and then further computes .

According to the received value , computes his/her register secret key :

At last, is sent to the cloud server and the cloud server stores this tuple into a user list .

5.3. Location Data Encryption

To guarantee the security of location data, the LBS provider needs to encrypt all location information before outsourcing them to the cloud server. In this paper, to enable an efficient and privacy-preserving LBS query, we use different encryption mechanisms to encrypt the different attributes of the location data. Without loss of generality, we use to denote any location data belongs to    in category set. The LBS provider takes the following three steps to encrypt the location data .

First, the LBS provider uses its secret value to code as and further employs the bilinear pairing map to calculate . We use to denote the code operation of attribute of the location data such that

Second, the LBS provider uses the secretly preserved invertible matrix to encrypt ’s coordinate as Here, correspondingly, we use to denote this encryption operation.

Third, for the remaining other attributes TITLE and DESCRIPTION, the LBS provider adopts any semantically secure symmetric encryption such as AES under a given key to encrypt them, denoted as in our paper such that

5.4. Query Trapdoor Generation

To preserve user’s query privacy and enable correct search over encrypted location data, a query user with the current location coordinate needs to encrypt his/her query request before it is submitted to the cloud server. In this paper, a query trapdoor generation is conducted in two steps.

First, the user chooses a desired query objective denoted as (e.g., ) and uses the secret value granted by the LBS provider and the secret value randomly chosen by himself/herself in the user grant phase to encrypt as and .

Second, the user generates a random number and uses the matrix granted by the LBS provider to encrypt the current location coordinate as . Ultimately, the query trapdoor of can be denoted as follows:

5.5. Search over Encrypted Location Data

After the query user generates a query trapdoor , he/she submits and a parameter to the cloud server. Upon receiving the query trapdoor and , the powerful cloud server is responsible for searching over encrypted outsourced location data on behalf of the query use, without knowing any plaintext information of the outsourced location data and the user query request. If the user is a legal user, the cloud server returns back the first encrypted target locations that satisfy the query and are nearest to the query user. Therefore, in the whole query process, the cloud server must perform two key operations under the encrypted environment: searching the encrypted category set according to the query trapdoor to obtain all target locations; sorting the distances between target locations and user’s current location in an ascending order. To achieve the above goal, the cloud sever processes the search in two steps.

First, the cloud server looks up the query user ’s registration information from the user list . If the user information does not exist, the cloud server rejects the query; otherwise it linearly scans the encrypted category set and obtains all encrypted target location data if it finds out an encrypted    in the encrypted category set that satisfies the following equation:

Second, upon obtaining all target locations, the cloud server sorts the distances between target locations and user’s query location by evaluatingwhere and are any two locations satisfying the query with the encrypted coordinate and , respectively.

If the above inequality holds, then this indicates that the target location is closer to the query user than the target location . Hence, is sorted in the front of . Finally, the cloud server returns the first encrypted locations to the query user .

5.6. User Revocation

User revocation is an essential yet challenging task in a practical multiuser application such as an LBS system. In some related literatures supporting user revocation, to prevent revoked users from continuing to access outsourced cloud data, the data provider usually has to rebuild data index or reencrypt large amounts of data and reuploads them to the cloud server and issues new keys to the remaining users. It unavoidably brings heavy computation and communication cost for the data provider because of the high of dynamic of users in the cloud environment. In this paper, we propose an efficient user revocation mechanism without any data reencryption and keys update operations while being able to effectively prevent the revoked user from searching outsourced location data. More concretely, for a user who will be revoked by the LBS provider, the LBS provider first sends the user information about to the cloud server. Then, the cloud server scans user information in the user list to find out the information of and deletes . Once is deleted from , no longer has the capability to search location data stored at the cloud server. Since without , the cloud server cannot complete matching between the trapdoor and encrypted according to the query scheme proposed in Section 5.5, can still generate a legal query trapdoor.

6. Analysis

In this section, we analyze the search correctness and security to prove that our proposed scheme is correct and secure.

6.1. Search Correctness Analysis

When an authorized query user submits his/her query trapdoor to the cloud server, the cloud server firstly obtains the all encrypted locations satisfying the query by performing a matching operation between an encrypted and . Specifically, the cloud server judges whether holds or not for an encryption   . If the equation holds, then this indicates that the query correctly matches and the cloud server obtains all target locations belong to   . The correctness can be easily verified as follows:Then, for any two target locations and , the cloud server is able to determine whether is closer to the query current location than by evaluatingThis is because that where (, resp.) denotes the Euclidean distance between the location (, resp.) and the user’s current location. So,

Thus, the cloud server is able to sort all target locations according to the above distance comparisons in an ascending order and returns back the first nearest locations to the query user.

6.2. Security Analysis

In our proposed scheme, three encryption schemes , , and are employed to protect the confidentiality of LBS data. In this section, we will analyze the security of our scheme against the “honest-but-curious” cloud server in the multiuser environment.

is a semantically secure symmetric encryption such as AES that encrypts the TITLE and DESCRIPTION fields of LBS data. The semantic security of AES guarantees the security of TITLE and DESCRIPTION fields of LBS data. We use secure kNN encryption technique denoted as to encrypt the attribute of LBS data and query user’s coordinate to enable a secure and effective distance comparison. The security of the attribute of LBS data and the query user’s coordinate mainly relies on the security of secure kNN scheme. For the attribute of LBS data, we use to encrypt it to enable secure and flexible search over encrypted location data. Specifically, given a location data with attribute , the ciphertext can be denoted as . Since is a group element in with a large prime order, the secret key is acknowledgedly intractable from in the large number field due to the well-known DLP assumption. Without the secret key kept secretly by the LBS provider, the cloud server cannot recover the message from encryption .

In addition, in the multiuser environment, the system should prevent an illegal user from requesting for query LBS data stored in the cloud server. In our scheme, when a registered user wants to query the LBS location data using , uses secret query keys and to generate the query trapdoor of , . Under the assumption of DLP, attackers cannot compute out and according to . Without the correct and , an illegal user cannot generate the correct query trapdoor such that cannot perform the correct query over encrypted location data. For a revoked user , although can generate the trapdoor for , still cannot let the cloud server perform a correct query on behalf of him/her due to lacking of the necessary query parameter that has been deleted from the list by the cloud server in the phase of user revocation.

7. Evaluation

In this section, we evaluate the performance of our proposed scheme from the perspective of the LBS provider, the LBS user, and the cloud server, respectively. The software and hardware configurations of the LBS provider and LBS user side are Windows 7 desktop systems with Intel Core 2 Duo CPU 2.26 GHZ, 3 GB memory, and 320 GHZ hard driver and the cloud sever side is a virtual machine with Core 2 Duo CPU 4 × 2.394 GHZ, 8 GB memory on the Dell blade sever M610, and the Linux Centos 5.8 OS.

All programs are developed using Java language and the JPBC library [29] is employed to implement group operations. We execute all experiments in a real data set collected from the open street map [30] with 50 categories and the number of concrete location data being 1000 by extracting the location data belonging to Yuelu District, Changsha, China.

7.1. LBS Data Encryption

Figure 2(a) shows that the time cost of encrypting LBS data for the LBS provider increases linearly with the increasing size of category set when the total number of location data remains unchanged . Figure 2(b) shows the number of concrete location data has little influence on the time cost of encrypting LBS data when fixing the size of category set . Recall that, in our scheme, is used to encrypt categories in category set and and are used to encrypt location data. Experimental results from Figure 2 illustrate that the time cost of encrypting LBS data is closely related to the encryption while not being almost affected by and . It is reasonable that the consists of the time-consuming pair operation and exponent operation over the ellipse curve group while and almost do not consume time when an extremely small message and 3-dimensional vector are encrypted by them, respectively.

7.2. Query Request Encryption

According to the query trapdoor generation proposed in the Section 5.4, in the whole processes of the query request encryption, three key operations are involved to encrypt an interested query keyword and current location coordinate for a registered LBS user (i.e., the hash operation, the exponentiation operation on group, and the matrix multiplication operation between a matrix and a 3-dimensional column vector). The time cost of each operation based on our software/hardware setting is shown in Table 2. Therefore, the total time cost of generating a query request for an LBS user can be denoted as  ms; it is extremely efficient in practice.

Notations Descriptions Time (ms)

Hash function ≈93
Exponentiation operation on group ≈34
Matrix multiplication operation<1

7.3. Query over Encrypted LBS Data

Figures 3(a) and 3(b) show the time cost of search over encrypted location data for the cloud server. We can observe that the number of categories and encrypted location data have little influence on the overhead of search for the cloud server. This is because that the main time cost for the search is to only enforce two relatively time-consuming pairing operations while linear search over 50 categories according to the query trapdoor for target location data and 3-dimensional vector computation for distance comparison almost does not consume time.

Figure 3(c) shows the average response time of our query scheme for different sizes of query users. We can see that the response time grows linearly with the increasing number of query users. When the number of query users achieves 100, the response time is about 6.82 s, and this is extremely efficient in practical application.

8. Conclusion

In this paper, we propose a privacy-preserving multiuser LBS query scheme based on the hybrid encryption in the cloud environment. Adopting different encryptions on different attributes of LBS data, our proposed scheme can achieve users’ location privacy protection and the confidentiality of LBS data. In particular, the LBS query is performed in the cipher environment, thus the LBS users can get the accurate LBS query results without disclosing their private information. Besides, we consider LBS user accountability and LBS user dynamics, for preventing the unregistered users and expired users accessing. And extensive experiments show that our proposed scheme is highly efficient. In the future, we will consider collusion attacks in the cloud-based LBSs.

Conflicts of Interest

The authors declare that there are no conflicts of interest regarding the publication of this article.

Authors’ Contributions

Hui Yin and Zheng Qin equally contributed to this work.


This work is partially supported by the National Science Foundation of China under Grants nos. 61472131 and 61772191; the Science and Technology Key Projects of Hunan Province under Grants nos. 2015TP1004, 2015SK2087, 2015JC1001, and 2016JC2012; the Natural Science Foundation of Hunan Province under Grant no. 2017JJ2292; Outstanding Youth Research Project of Provincial Education Department of Hunan under Grant no. 17B030; the Science and Technology Planning Project of Changsha under Grant no. k1705018.


  1. Statista, “Number of location-based service users in the United States from 2013 to 2018 (in millions),” Statista; 2017. View at: Google Scholar
  2. K. Xie, X. Ning, X. Wang et al., “Recover corrupted data in sensor networks: a matrix completion solution,” IEEE Transactions on Mobile Computing, vol. 16, no. 5, pp. 1434–1448, 2017. View at: Publisher Site | Google Scholar
  3. M. Li, H. Zhu, Z. Gao et al., “All your Location are belong to us: breaking mobile social networks for automated user location tracking,” in Proceedings of the 15th ACM International Symposium on Mobile Ad Hoc Networking and Computing, MobiHoc 2014, pp. 43–52, USA, August 2014. View at: Publisher Site | Google Scholar
  4. J. Shao, R. Lu, and X. Lin, “FINE: a fine-grained privacy-preserving location-based service framework for mobile devices,” in Proceedings of the IEEE INFOCOM, pp. 244–252, IEEE, Ontario, Canada, May 2014. View at: Publisher Site | Google Scholar
  5. G. Ghinita, P. Kalnis, A. Khoshgozaran, C. Shahabi, and K.-L. Tan, “Private queries in location based services: anonymizers are not necessary,” in Proceedings of the ACM SIGMOD International Conference on Management of Data (SIGMOD '08), pp. 121–132, ACM, 2008. View at: Publisher Site | Google Scholar
  6. B. Gedik and L. Liu, “Protecting location privacy with personalized k-anonymity: architecture and algorithms,” IEEE Transactions on Mobile Computing, vol. 7, no. 1, pp. 1–18, 2008. View at: Publisher Site | Google Scholar
  7. R. Paulet, M. G. Kaosar, X. Yi, and E. Bertino, “Privacy-preserving and content-protecting location based queries,” IEEE Transactions on Knowledge and Data Engineering, vol. 26, no. 5, pp. 1200–1210, 2014. View at: Publisher Site | Google Scholar
  8. L. Sweeney, “Achieving k-anonymity privacy protection using generalization and suppression,” International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems, vol. 10, no. 5, pp. 571–588, 2002. View at: Publisher Site | Google Scholar
  9. Q. Wang, C. Xu, and M. Sun, “Multi-dimensional K-anonymity based on mapping for protecting privacy,” Journal of Software, vol. 6, no. 10, pp. 1937–1944, 2011. View at: Publisher Site | Google Scholar
  10. K. Xie, X. Ning, X. Wang et al., “An efficient privacy-preserving compressive data gathering scheme in WSNs,” Information Sciences, vol. 390, pp. 82–94, 2017. View at: Publisher Site | Google Scholar
  11. H. Zhu, R. Lu, C. Huang, L. Chen, and H. Li, “An efficient privacy-preserving location-based services query scheme in outsourced cloud,” IEEE Transactions on Vehicular Technology, vol. 65, no. 9, pp. 7729–7739, 2016. View at: Publisher Site | Google Scholar
  12. M. Gruteser and D. Grunwald, “Anonymous usage of location-based services through spatial and temporal cloaking,” in Proceedings of the 1st International Conference on Mobile Systems, Applications and Services, pp. 31–42, ACM, San Francisco, Calif, USA, May 2003. View at: Publisher Site | Google Scholar
  13. G. Zhong and U. Hengartner, “Toward a distributed k-anonymity protocol for location privacy,” in Proceedings of the 7th ACM Workshop on Privacy in the Electronic Society (WPES '08), pp. 33–37, ACM, Alexandria, VA, USA, October 2008. View at: Publisher Site | Google Scholar
  14. C.-Y. Chow, M. F. Mokbel, and W. G. Aref, “: query processing for location services without compromising privacy,” ACM Transactions on Database Systems (TODS), vol. 34, no. 4, Article ID 24, 2009. View at: Publisher Site | Google Scholar
  15. B. Gedik and L. Liu, “Location privacy in mobile systems: a personalized anonymization model,” in Proceedings of the 25th IEEE International Conference on Distributed Computing Systems (ICDCS '05), pp. 620–629, IEEE, June 2005. View at: Google Scholar
  16. A. Khoshgozaran and C. Shahabi, “Blind evaluation of nearest neighbor queries using space transformation to preserve location privacy,” in Proceedings of the Advances in Spatial and Temporal Databases, pp. 239–257, Springer, 2007. View at: Google Scholar
  17. M. Ghaffari, N. Ghadiri, M. H. Manshaei, and M. S. Lahijani, “: a peer-to-peer privacy preserving query service for location-based mobile applications,” IEEE Transactions on Vehicular Technology, vol. 66, no. 10, pp. 9458–9469, 2017. View at: Publisher Site | Google Scholar
  18. Z. Montazeri, A. Houmansadr, and H. Pishro-Nik, “Achieving perfect location privacy in wireless devices using anonymization,” IEEE Transactions on Information Forensics and Security, vol. 12, no. 11, pp. 2683–2698, 2017. View at: Publisher Site | Google Scholar
  19. H. Jannati and B. Bahrak, “An oblivious transfer protocol based on elgamal encryption for preserving location privacy,” Wireless Personal Communications, vol. 97, no. 2, pp. 1–11, 2017. View at: Publisher Site | Google Scholar
  20. I. Memon, Q. A. Arain, M. H. Memon, F. A. Mangi, and R. Akhtar, “Search me if you can: multiple mix zones with location privacy protection for mapping services,” International Journal of Communication Systems, vol. 30, no. 16, Article ID e3312, 2017. View at: Publisher Site | Google Scholar
  21. Y. Zhu, D. Ma, D. Huang, and C. Hu, “Enabling secure location-based services in mobile cloud computing,” in Proceedings of the 2013 2nd ACM SIGCOMM Workshop on Mobile Cloud Computing, MCC 2013, pp. 27–32, China, August 2013. View at: Publisher Site | Google Scholar
  22. J. B. Abdo, T. Bourgeau, J. Demerjian, and H. Chaouchi, “Extended privacy in crowdsourced location-based services using mobile cloud computing,” Mobile Information Systems, vol. 2016, Article ID 7867206, 13 pages, 2016. View at: Publisher Site | Google Scholar
  23. F. Tang, J. Li, I. You, and M. Guo, “Long-term location privacy protection for location-based services in mobile cloud computing,” Soft Computing, vol. 20, no. 5, pp. 1735–1747, 2016. View at: Publisher Site | Google Scholar
  24. J. Y. Chen and C. L. Wang, “Privacy protection for mobile cloud data: a network coding approach,” View at: Google Scholar
  25. H. Yin, Z. Qin, L. Ou, and K. Li, “A query privacy-enhanced and secure search scheme over encrypted data in cloud computing,” Journal of Computer and System Sciences, vol. 90, pp. 14–27, 2017. View at: Publisher Site | Google Scholar | MathSciNet
  26. D. Boneh and M. Franklin, “Identity-based encryption from the Weil pairing,” SIAM Journal on Computing, vol. 32, no. 3, pp. 586–615, 2003. View at: Publisher Site | Google Scholar | MathSciNet
  27. H. Yin, Z. Qin, J. Zhang, and K. Li, “Achieving secure, universal, and fine-grained query results verification for secure search scheme over encrypted cloud data,” IEEE Transactions on Cloud Computing, no. 99, 2017. View at: Publisher Site | Google Scholar
  28. W. K. Wong, D. W. Cheung, B. Kao, and N. Mamoulis, “Secure kNN computation on encrypted databases,” in Proceedings of the International Conference on Management of Data and 28th Symposium on Principles of Database Systems (SIGMOD-PODS '09), pp. 139–152, Providence, RI, USA, July 2009. View at: Publisher Site | Google Scholar
  29. A. de Caro and V. Iovino, “jPBC: java pairing based cryptography,” in Proceedings of the 16th IEEE Symposium on Computers and Communications (ISCC '11), pp. 850–855, July 2011. View at: Publisher Site | Google Scholar
  30. Openstreetmap Foundation UK West Midlands, Openstreetmap. Openstreetmap Foundation West Midlands, U, K; 2017, View at: Publisher Site

Copyright © 2018 Lu Ou et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

More related articles

1088 Views | 565 Downloads | 6 Citations
 PDF Download Citation Citation
 Download other formatsMore
 Order printed copiesOrder

Related articles

We are committed to sharing findings related to COVID-19 as quickly as possible. We will be providing unlimited waivers of publication charges for accepted research articles as well as case reports and case series related to COVID-19. Review articles are excluded from this waiver policy. Sign up here as a reviewer to help fast-track new submissions.