A location privacy protection method in an edge computing location-based service scenario

Abstract

This invention relates to a location privacy protection method in an edge computing location service scenario, which generates a fake location for a user within the coverage area of ​​all edge nodes that directly provide services to the user A. The method includes: S1: Let E = {e1, e2, ..., e...}, where E is the set of edge nodes that directly provide services to user A. n}. Using user A's real location P as the center, obtain a circle C that is inscribed within the coverage area of ​​each edge node in set E; S2: Add a noise vector to user A's real location P to obtain a false location P', ensuring that the false location P' is within the coverage area of ​​circle C; S3: Upload the false location P' to each edge node e in set E. i This invention provides location services. The advantage of this invention is that, in edge computing scenarios, it generates a fake user location within the coverage area of ​​all edge nodes directly providing services to that user, making it impossible for malicious edge nodes to identify the authenticity of the fake location.

Description

Technical Field

[0001] This invention relates to the field of location privacy protection, and more specifically to a method for location privacy protection in edge computing location-based service scenarios. Background Technology

[0002] Location-based services (LBS) have become widely integrated into people's lives, bringing them great convenience. LBS primarily has the following application scenarios: POI search services, navigation services, motion detection services, advertising push services, and nearest neighbor detection services. However, while providing convenience to users, LBS also brings the risk of privacy breaches. In the field of edge computing location privacy protection, both domestically and internationally, the main approach is currently based on traditional differential privacy mechanisms.

[0003] Differential privacy, proposed by Dwork in 2006, is a novel privacy protection mechanism to address the problem of privacy leakage. Differential privacy is applicable to statistical datasets; when publishing results of a statistical dataset, noise can be added before publication, thereby protecting the privacy of the original dataset. Differential privacy protects privacy by adding noise, and the privacy budget ε is related to the privacy protection capability; the smaller ε, the higher the privacy protection capability.

[0004] In edge computing location-based service scenarios, if an end user directly uploads location data to an edge node, the user's real location must be within the coverage area of ​​that edge node. Therefore, the fake location data uploaded by the user must also be within the coverage area of ​​that edge node. Otherwise, the edge node will identify the user's fake location data and refuse to provide location services. Traditional differential privacy mechanisms may generate fake locations outside the coverage area of ​​the edge node, making it easy for service providers to identify fake locations and causing service unavailability.

[0005] Therefore, in edge computing location-based service scenarios, protecting user location privacy while preventing edge nodes from identifying false locations is an urgent problem to be solved. Summary of the Invention

[0006] To address the aforementioned issues, this invention provides a method for protecting location privacy in edge computing location-based service scenarios.

[0007] The technical solution of this invention is: a method for protecting location privacy in edge computing location-based service scenarios, comprising:

[0008] Step S1: Assume the set of edge nodes that directly provide services to user A is E = {e1, e2, ..., e...} n}, with the real location P of user A as the center, we obtain a circular region C in which user A is inscribed by the coverage area of ​​each edge node in the edge node set E;

[0009] Step S2: Add a noise vector to the real location of user A to obtain a false location P', so that the false location P' is within the coverage area of ​​the circular region C;

[0010] Step S3: Upload the dummy location P' to each edge node in the edge node set E to obtain location services.

[0011] Compared with the prior art, the present invention has the following advantages:

[0012] In the field of location-based edge computing, this invention adds reasonable noise to a user's real location, generating a fake location within the coverage area of ​​the edge nodes directly serving that user. This method can generate a fake location for a user within the coverage area of ​​all edge nodes simultaneously, making it impossible for any single edge node to identify the authenticity of the fake location. Attached Figure Description

[0013] Figure 1 This is a flowchart illustrating a location privacy protection method in a location-based service scenario using edge computing, as described in an embodiment of the present invention.

[0014] Figure 2 This is a schematic diagram of the effective circular area in an embodiment of the present invention;

[0015] Figure 3 This is a polar coordinate diagram of the user's real and fake positions in an embodiment of the present invention. Detailed Implementation

[0016] This invention provides a location privacy protection method for edge computing location-based service scenarios. It utilizes differential privacy and geographic indistinguishability to protect user location privacy and can prevent edge nodes from identifying the authenticity of fake locations.

[0017] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below through specific implementations and in conjunction with the accompanying drawings.

[0018] like Figure 1 As shown in the figure, an embodiment of the present invention provides a location privacy protection method in an edge computing location-based service scenario, comprising the following steps:

[0019] Step S1: Assume the set of edge nodes that directly provide services to user A is E = {e1, e2, ..., e...} n}, with the real location P of user A as the center, we obtain a circular region C in which user A is inscribed by the coverage area of ​​each edge node in the edge node set E;

[0020] Step S2: Add a noise vector to the real location P of user A to obtain a false location P', so that the false location P' is within the coverage area of ​​the circular region C;

[0021] Step S3: Upload the dummy location P' to each edge node in the edge node set E to obtain location services.

[0022] In one embodiment, step S1 above: Assume the set of edge nodes that directly provide services to user A is E = {e1, e2, ..., e}. n Taking user A's actual location P as the center, we obtain a circular region C in which user A is internally tangent to the coverage area of ​​each edge node in the edge node set E, specifically including:

[0023] Step S11: Obtain the Cartesian coordinates of the real location P of user A, and obtain the set E = {e1, e2, ..., e...} of all edge nodes that directly provide services to user A. n};

[0024] Step S12: Traverse each edge node e in set E i Calculate the distance from user A to edge node e i The minimum value d of the coverage boundary i Save it to set D;

[0025] Step S13: Obtain a circular region C with user A's actual location P as the center and a radius equal to the minimum value R in set D. This region is the valid region for user A, such as... Figure 2 The circular valid areas generated for terminal user 1 and terminal user 2 are shown respectively.

[0026] In one embodiment, step S2 above, which involves adding a noise vector to the real location P of user A to obtain a false location P', such that the false location is within the coverage area of ​​the circular region C, specifically includes:

[0027] The method proposed in this invention is based on a geographic indistinguishability mechanism. The core idea is to add noise to user A's real location and then generate a false location within a circular region C to prevent edge nodes from recognizing the false location data. The method uses the location privacy protection probability density function formula (1) to add a noise vector to user A's real location to generate a false location:

[0028]

[0029] like Figure 3As shown, R represents the radius of the effective region centered on the real location, ε is the privacy budget, and the privacy protection capability decreases as the privacy budget increases, r represents the distance between the false location and the real location, and θ is the angle between the ray from the real location to the false location and the polar axis in polar coordinates.

[0030] Integrating over r, we obtain the probability density of r, as shown in equation (2):

[0031]

[0032] Integrating over θ, we obtain the probability density of θ, as shown in equation (3):

[0033]

[0034] Step S21: Convert the Cartesian coordinates of user A's actual location data to polar coordinates, such as... Figure 3 As shown, the pole of the polar coordinates is the real position of user A, and r represents the distance between the false position and the real position;

[0035] Step S22: User A sets a privacy budget ε, which, together with the radius R of User A's effective area, is used as a parameter in formula (1);

[0036] Step S23: Use formula (1) to randomly generate r and θ;

[0037] Step S24: Use the following formulas (4) and (5) to generate the Cartesian coordinates of the pseudo-position:

[0038] P′.x=P.x+r·cos(θ) (4)

[0039] P′.y=P.y+r·sin(θ) (5).

[0040] Prove that the location privacy protection probability density function formula (1) used in step S2 has a total probability of 1 within the effective area;

[0041] For the location privacy protection probability density function formula (1) used in step S2, integrate it over its effective area to calculate the total probability within the effective range:

[0042]

[0043] As shown in equation (6) above, it is proven that step S2 will not generate false positions outside the effective region, demonstrating the rationality of the probability density function formula (1) proposed and used in this invention;

[0044] It is proven that the fake location generation method described in step S2 satisfies ε-geographic indistinguishability. For ease of description, this method is denoted as GIMIA.

[0045] For a user's true location x and its neighboring locations x′, it is proven that after applying the proposed location privacy protection mechanism, the probability of being confused from the true location x to z, Pr(GIMIA(x)) = z, and the probability of being confused from x′ to z, Pr(GIMIA(x′)), satisfy the definition of ε-geographic indistinguishability:

[0046] P r (G(x)=z)≤e εd(x，x′) P r (G(x′)=z) (7)

[0047] The proof is shown in equation (8) below:

[0048]

[0049] The above evidence demonstrates that the location privacy protection method proposed in this invention satisfies the ε-geographic indistinguishability definition, and this method can provide a rigorous theoretical proof of privacy protection for end users of location-based services in edge computing scenarios.

[0050] In one embodiment, step S3 above, which involves uploading the dummy location P' to each edge node in E to obtain location services, specifically includes:

[0051] User A uploads the false location P' calculated in step S2 to each edge node that directly provides services to it. The edge nodes then provide location services to user A based on the false location P'.

[0052] This invention discloses a location privacy protection method for edge computing location-based service scenarios. Theoretically, it is proven that the method satisfies the ε-geographic indistinguishability definition. By adding reasonable noise to the user's real location, a fake location is generated within the coverage area of ​​the edge node that directly provides services to the user. When multiple edge nodes directly serve a user at the same time, the method can generate the user's fake location within the coverage area of ​​all edge nodes, so that no edge node can identify the authenticity of the fake location.

[0053] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for protecting location privacy in edge computing location-based service scenarios, characterized in that, The method includes the following steps: Step S1: The set of edge nodes that directly provide services to user A is E={e1,e2,…,e...} n }, where e i As an edge node, based on user A's real location Using the circle as the center, we obtain a circular region C in which the coverage area of ​​each edge node in the edge node set E is inscribed; Step S2: Determine the actual location of user A Adding noise vectors to obtain false positions This makes the false position Within the coverage area of ​​circle C; Step S3: Place the false position Upload the information to each edge node in the edge node set E to obtain location services; In step S2, the actual location of user A is determined. Add noise vectors to generate fake positions In polar coordinates with user A's actual location as the pole, the probability density function of this noise vector is described as follows: In formula (1), R represents the actual position. The radius of the effective region centered on the circle. For privacy budget, This represents the distance between the false location and the true location. The angle between the ray from the real position to the pseudo position and the polar axis in polar coordinates; Integrating over r, we obtain the probability density of r, as shown in equation (2): right Integral, to obtain The probability density is as follows (3): False position The specific calculation method is as follows: Input: The user's actual Cartesian coordinates P(x,y); the set of edge nodes E={e1,e2,…,e...} that directly provide services to the user. n }; and the user-defined privacy budget. ; Calculate the minimum Euclidean distance from the Cartesian coordinate P(x,y) to the boundary of the coverage area of ​​each edge node in the edge node set E, and obtain the result set D, where R is the minimum value in set D. Convert the Cartesian coordinate P(x,y) to polar coordinates, where the Cartesian coordinate P(x,y) is the pole of the polar coordinates. R and As parameters of formula (1), a noise vector is added to the Cartesian coordinates P(x,y) using formula (1), and r and are randomly generated. The false position is obtained using the following formulas (4) and (5). : 。 2. The method as described in claim 1, characterized in that, step In S1, we obtain a value based on user A's real location. Let A be a circular region C centered at a point and tangent to the coverage areas of all edge nodes that provide services to A. The radius of this circular region C is the minimum Euclidean distance between user A and the boundaries of the coverage areas of all edge nodes that provide services to A.

3. The method as described in claim 1, characterized in that, In step S2, users can set a privacy budget. Determine the ability to protect location privacy.

4. The method as described in claim 1, characterized in that, In step S3, when the user uses the location service, they use formula (1) to determine their real location. Add noise vectors to generate fake positions , to the false position Send the location service to all edge nodes in set E that directly provide the service.

Images