CN115103351A - Wireless sensor network distributed positioning method under privacy protection - Google Patents

Abstract

The invention discloses a distributed positioning method for a wireless sensor network under privacy protection. Assuming that all sensor nodes in the scene are located in the convex hull of the anchor node, each sensor node can find three adjacent nodes as a triangulation set , so that it is located in the convex hull formed by these adjacent nodes; sensor nodes and adjacent nodes can communicate in pairs, while considering the privacy of node locations; when sensors communicate and interact, add node location information based on privacy protection policies. The time-varying incremental noise makes it impossible for attackers to steal the private location of the node, so that the wireless sensor network can be precisely located while keeping the location of the node private.

Description

A privacy-preserving distributed localization method for wireless sensor networks

technical field

The invention relates to the technical field of wireless sensor network positioning, and more particularly, to a distributed positioning method for wireless sensor networks under privacy protection.

Background technique

Localization, a fundamental problem in wireless sensor networks, has attracted extensive attention of researchers in recent years, inspiring many potential applications, such as location support services, smart home systems, and object tracking. Sensor location information is key to accurately processing sensor measurements, but in applications where sensors are deployed randomly, the exact location of the sensor is unknown. Therefore, how to design a suitable localization algorithm is the key to realize the efficient and accurate work of the sensor.

At present, the distributed iterative positioning algorithm based on barycentric coordinates has become a widely used high-precision distributed positioning algorithm. By measuring the relative distance between sensor nodes, the algorithm expresses the positioning process as an iterative process in the form of a matrix vector, which can Globally converge to the precise location of the sensor. It is worth noting that although sensor localization algorithms based on distributed frameworks have been proposed, most of them do not consider that the information interaction between sensors will cause serious privacy issues.

The information exchange in the form of unencrypted plaintext is easily attacked by external attackers. The attacker can steal information through the communication link between intrusion nodes, thereby infringing the privacy information of sensor nodes and destroying the entire positioning system. Additionally, attackers can also infer the location of other non-anchor nodes by correlating distance measurements with the monitored area. Therefore, formulating corresponding strategies to protect the private locations of sensor nodes is a topic worthy of attention.

SUMMARY OF THE INVENTION

The purpose of the present invention is to overcome the deficiencies of the prior art and provide a distributed positioning method for wireless sensor networks under privacy protection. The private location of the sensor node.

In order to achieve the above purpose of the invention, a distributed positioning method for a wireless sensor network under privacy protection of the present invention is characterized in that, it includes the following steps:

(1), set the wireless sensor network

Suppose the wireless sensor network to be located consists of n nodes in a two-dimensional space, each node represents a wireless sensor;

和非锚节点集

其中，记锚节点集

由s个锚节点X_i组成，记为

记非锚节点集

由n-s个非锚节点Y_j组成，记为

In the wireless sensor network to be located, the node whose coordinate position is known is recorded as the anchor node, and the node whose own coordinate position is unknown is recorded as the non-anchor node, so that the nodes in the wireless sensor network are divided into anchor node sets

and non-anchor node sets

Among them, the anchor node set is recorded

It consists of s anchor nodes X _i , denoted as

set of non-anchor nodes

It consists of ns non-anchor nodes Y _j , denoted as

(2), record the coordinates of anchor nodes and non-anchor nodes;

中各锚节点的坐标，记为P_a＝[p₁,...,p_i,...,p_s]^T，其中，p_i为第i个锚节点X_i的坐标；record anchor node set

The coordinates of each anchor node in , denoted as P _a =[p ₁ ,..., _pi ,...,p _s ] ^T , where p _i is the coordinate of the _ith anchor node Xi;

中的非锚节点的坐标表示为P_b＝[p_s+1,...,p_s+j,...,p_n]^T，p_s+j表示第j个非锚节点Y_j的坐标；set non-anchor nodes

The coordinates of the non-anchor nodes in are expressed as P _b =[ps ₊₁ ,...,ps _+j ,...,p _n ] ^T , where ps _+j represents the jth non-anchor node Y _j coordinate;

(3), set the adjacent node set of the non-anchor node;

其中，K_j是由距离非锚节点Y_j最近的3个节点d_j1,d_j2,d_j3构成；Define the adjacent node set of non-anchor node Y _j

Among them, K _j is composed of three nodes d _j1 , d _j2 , and d _j3 that are closest to the non-anchor node Y _j ;

(4) Establish a distributed positioning model for the location of non-anchor nodes;

中所有非锚节点都位于锚节点集

的锚节点的凸包内，遍历非锚节点集

在每个迭代时刻t，更新非锚节点Y_j的位置；Assuming a non-anchor node set

All non-anchor nodes in the anchor node set

Within the convex hull of the anchor node, traverse the set of non-anchor nodes

At each iteration time t, update the position of the non-anchor node Y _j ;

Among them, p _j (t) represents the coordinate of the jth non-anchor node Y _j at time t, γ is the gain coefficient, p _k is the position coordinate of the adjacent node d _jk of the non-anchor node Y _j at time t, and a _jk is The barycentric coordinates of the non-anchor node Y _j relative to its adjacent node d _jk ;

(5) Synchronously iteratively locate the location of non-anchor nodes under privacy protection;

(5.1), the set maximum iteration time t ^* ; the initial positions of the anchor node X _i and the non-anchor node Y _j are randomly set, wherein, the initial position of the anchor node X _i is set to p _i (0)=p _i , set randomly The initial position of the non-anchor node Y _j is p _j (0);

其中锚节点X_i的噪声参数δ_i＝1；非锚节点集Y_j的噪声参数δ_j满足约束

Setting time-varying incremental noise of each node in wireless sensor network

where the noise parameter δ _i =1 of the anchor node X _i ; the noise parameter δ _{j of the non-anchor node set Y j satisfies} the constraint

(5.2), at iterative time t, add time-varying incremental noise σ _w (t) to the position estimation of each node, thereby enhancing privacy protection;

Among them, 0<ζ<1 is a constant value parameter;

(5.3) At iterative time t, calculate the position estimates of anchor node X _i and non-anchor node set Y _j after introducing time-varying incremental noise as:

in,

(5.4) After passing the privacy protection operation of formula (3), each node sends its current position estimate to adjacent nodes, and further obtains the position estimate of each node at the next moment;

(5.5), write the barycentric coordinates of all non-anchor nodes relative to their adjacent node set K _j in matrix-vector form as:

Among them, F and H are sub-random matrices, [FH] is a row random matrix, when the adjacent node d _jk belongs to the triangulation set of the non-anchor node Y _j , [FH] _jk =a _jk , when the adjacent node d When _jk does not belong to the triangulation set of the non-anchor node Y _j , [FH] _jk =0, I _s and I _ns are identity matrices;

中基于锚节点集

表示的位置为：(5.6), when the iteration time t reaches the set maximum iteration time t ^* , calculate the non-anchor node set

anchor node set

The indicated location is:

的位置定位完成。So far, the non-anchor node set based on the privacy protection policy

The location positioning is completed.

The purpose of the invention of the present invention is achieved in this way:

The present invention is a distributed positioning method for wireless sensor network under privacy protection. It is assumed that all sensor nodes in the scene are located in the convex hull of the anchor node, and each sensor node can find three adjacent nodes as a triangulation set, so that the It is located in the convex hull formed by these adjacent nodes; sensor nodes and adjacent nodes can communicate in pairs, while considering the privacy of the node location; when the sensor is communicating and interacting, the time-varying node location information is added based on the privacy protection strategy. Incremental noise makes it impossible for attackers to steal the private location of the node, so that the wireless sensor network can be precisely located while keeping the location of the node private.

At the same time, the distributed positioning method of the wireless sensor network under the privacy protection of the present invention also has the following beneficial effects:

(1) In the distributed positioning iteration process of the present invention, the noise terms that satisfy the constraints are added to the position estimation of the sensor nodes. This method not only has no effect on the positioning accuracy of the sensor network, but also prevents the attacker from stealing the position information of the neighbor nodes. In this case, the privacy of the precise location of the node can be effectively guaranteed.

(2) The present invention considers that in some real sensor network environments, attackers with remote observation capabilities may steal part of the inbound information of their neighbors, while the classical privacy protection method is limited to the fact that attackers can only steal their neighbors A node's outbound information, but cannot steal the inbound information of its neighbors. Therefore, the present invention proposes a new privacy protection strategy that introduces time-varying incremental noise, which can still protect the privacy of the node well when the attacker steals the inbound information of its neighbor nodes.

Description of drawings

1 is a flowchart of a distributed positioning method for a wireless sensor network under a privacy protection of the present invention;

Figure 2 is a schematic diagram of the placement of a group of Raspberry Pis set up in the experiment;

Figure 3 is a schematic diagram of the convex hull of a node in the wireless sensor network located at its adjacent node;

Fig. 4 is the interaction topology diagram between 10 sensor nodes in the wireless sensor network;

Figure 5 is the initial position randomly set by 7 non-anchor nodes in the wireless sensor network;

Figure 6 is the precise location to which the 7 non-anchor nodes finally converge after 40 iterations.

Detailed ways

The specific embodiments of the present invention are described below with reference to the accompanying drawings, so that those skilled in the art can better understand the present invention. It should be noted that, in the following description, when the detailed description of known functions and designs may dilute the main content of the present invention, these descriptions will be omitted here.

Example

FIG. 1 is a flowchart of a distributed positioning method for a wireless sensor network under privacy protection according to the present invention.

In this embodiment, as shown in FIG. 1 , a distributed positioning method for a wireless sensor network under privacy protection of the present invention includes the following steps:

S1. Set up the wireless sensor network;

和非锚节点集

锚节点的位置是静态已知的。在本实施例中，锚节点集

由3个锚节点X_i组成，记为

非锚节点集

由7个非锚节点Y_j组成，记为

In this embodiment, as shown in Fig. 2, the wireless sensor network is located in a distributed manner through a group of Raspberry Pi, which consists of a total of n=10 nodes in the two-dimensional space, and each node represents a wireless sensor , the nodes in the wireless sensor network are further divided into anchor node sets

and non-anchor node sets

The location of the anchor node is known statically. In this embodiment, the anchor node set

It consists of 3 anchor nodes X _i , denoted as

non-anchor node set

It consists of 7 non-anchor nodes Y _j , denoted as

S2, record the coordinates of anchor nodes and non-anchor nodes;

中各锚节点的坐标，记为P_a＝[p₁,p₂,p₃]^T；record anchor node set

The coordinates of each anchor node in , denoted as P _a =[p ₁ ,p ₂ ,p ₃ ] ^T ;

中这些未知坐标的非锚节点的坐标表示为P_b＝[p₄,...,p_3+j,...,p₁₀]^T，p_3+j表示第j个非锚节点Y_j的坐标；set non-anchor nodes

The coordinates of these non-anchor nodes with unknown coordinates in are expressed as P _b =[p ₄ ,...,p _3+j ,...,p ₁₀ ] ^T , where p _3+j represents the jth non-anchor node Y _j coordinate of;

S3. Set the adjacent node set of the non-anchor node;

其中，K_j由距离非锚节点Y_j最近的3个节点d_j1,d_j2,d_j3构成；Define the adjacent node set of non-anchor node Y _j

Among them, K _j consists of three nodes d _j1 , d _j2 , and d _j3 closest to the non-anchor node Y _j ;

S4. Establish a distributed positioning model for the location of non-anchor nodes;

中所有非锚节点都位于

的锚节点的凸包内，遍历非锚节点集

在每个迭代时刻t，更新非锚节点Y_j的位置；In this embodiment, FIG. 4 is an interaction topology diagram of each sensor in the wireless sensor network. There are 3 anchor nodes in the wireless sensor network, which are represented by Anchori(i=1,2,3); there are 7 non-anchor nodes in the network, which are represented by nodej(j=4,5,...,10). Assuming a non-anchor node set

All non-anchor nodes in

Within the convex hull of the anchor node, traverse the set of non-anchor nodes

At each iteration time t, update the position of the non-anchor node Y _j ;

Among them, p _j (t) represents the coordinate of the jth non-anchor node Y _j at time t, γ is the gain coefficient, p _k is the position coordinate of the adjacent node d _jk of the non-anchor node Y _j at time t, and a _jk is The barycentric coordinates of the non-anchor node Y _j relative to its adjacent node d _jk ; .

In this embodiment, as shown in FIG. 3 , in a triangulation set of non-anchor node Y _j in the wireless sensor network, the barycentric coordinate of non-anchor node Y _j relative to three adjacent nodes r, s, and t is a _jr , a _js , a _jt , the position coordinate p _{j of the non-anchor node Y j satisfies} :

p _j =a _jr p _r +a _js p _s +a _jt p _t (2)

Among them, a _jr +a _js +a _jt =1;

The barycentric coordinates of the non-anchor nodes Y _j can be calculated by specifying the ratio of the signed areas between the triangles:

Among them, S _Δjst , S _Δjrt , S _Δjrs , S _Δrst can be obtained by measuring the relative distance between nodes, and calculated by the Cayley-Menger determinant, taking S _Δjst as an example:

S5. Synchronously iteratively locate the position of the non-anchor node under privacy protection;

S5.1, the maximum iteration time t ^* set; randomly set the initial positions of the anchor node X _i and the non-anchor node Y _j , wherein, the initial position of the anchor node X _i is set to p _i (0)=p _i , set randomly The initial position of the non-anchor node Y _j is p _j (0);

As shown in Figure 4, there are frequent communication interactions among the nodes of the wireless sensor network. In the process of information exchange between nodes, in order to prevent attackers from stealing information, a privacy protection strategy of time-varying incremental noise is introduced.

其中锚节点X_i的噪声参数δ_i＝1；非锚节点集Y_j的噪声参数δ_j满足约束

Setting time-varying incremental noise of each node in wireless sensor network

where the noise parameter δ _i =1 of the anchor node X _i ; the noise parameter δ _{j of the non-anchor node set Y j satisfies} the constraint

那么在迭代时刻t，对每个节点的位置估计增加时变增量噪声σ_w(t)为：S5.2. Add time-varying incremental noise to the position estimation of each node

Then at the iteration time t, the time-varying incremental noise σ _w (t) is added to the position estimate of each node as:

Among them, 0<ζ<1 is a constant value parameter set randomly, and it can be found that the added noise increases with the passage of the iteration time.

S5.3. At the iteration time t, calculate the position estimates of the anchor node X _i and the non-anchor node set Y _j respectively after introducing the time-varying incremental noise as:

in,

S5.4. After passing the privacy protection operation of formula (6), each node sends its current position estimate to adjacent nodes, and further obtains the position estimate of each node at the next moment;

S5.5. In this embodiment, the barycentric coordinates of 7 non-anchor nodes relative to their adjacent node set K _j are written in matrix-vector form as:

Among them, F and H are sub-random matrices, [FH] is a row random matrix, when the adjacent node d _jk belongs to the triangulation set of the non-anchor node Y _j , [FH] _jk =a _jk , when the adjacent node d When _jk does not belong to the triangulation set of the non-anchor node Y _j , [FH] _jk =0, I _s and I _ns are identity matrices;

中基于锚节点集

表示的位置为：S5.6. When the iteration time t reaches the set maximum iteration time t ^* , calculate the non-anchor node set

anchor node set

The indicated location is:

的位置定位完成。So far, based on the non-anchor node set under the privacy protection policy

The location positioning is completed.

Figure 5 is the initial position randomly set by 7 non-anchor nodes in the wireless sensor network, where Node j (j=4, 5,..., 10) is the given initial position estimate of the non-anchor node, it can be seen that the initial position There is a large error between the estimate and the actual position. Using the iterative positioning algorithm under the privacy protection of the present invention, the estimated position of the non-anchor node gradually approaches the actual position. Figure 6 is the exact location of the 7 non-anchor nodes in the sensor network after 40 iterations. It can be seen that the non-anchor nodes Node j (j=4, 5,..., 10) can all converge to the sensor nodes actual location.

Although the illustrative specific embodiments of the present invention have been described above to facilitate the understanding of the present invention by those skilled in the art, it should be clear that the present invention is not limited to the scope of the specific embodiments. For those skilled in the art, As long as various changes are within the spirit and scope of the present invention as defined and determined by the appended claims, these changes are obvious, and all inventions and creations utilizing the inventive concept are included in the protection list.

Claims (1)

1. A wireless sensor network distributed positioning method under privacy protection is characterized by comprising the following steps:

(1) setting a wireless sensor network;

the method comprises the steps that a wireless sensor network to be positioned is arranged and composed of n nodes in a two-dimensional space, and each node represents a wireless sensor;

in the wireless sensor network to be positioned, nodes with known self coordinate positions are marked as anchor nodes, nodes with unknown self coordinate positions are marked as non-anchor nodes, and therefore the nodes in the wireless sensor network are divided into anchor node sets

And a set of non-anchor nodes

Wherein the set of anchor nodes

By s anchor nodes X _i Composition of, is

Anchorage node set

From n-s non-anchor nodes Y _j Composition, as

(2) Recording the coordinates of the anchor nodes and the non-anchor nodes;

recording anchor node set

Coordinates of each anchor node in P _a ＝[p ₁ ,…,p _i ,…,p _s ] ^T Wherein p is _i For the ith anchor node X _i The coordinates of (a);

aggregating non-anchor nodes

Coordinate of the non-anchor node in (1) is expressed as P _b ＝[p _s+1 ,…,p _s+j ,…,p _n ] ^T ，p _s+j Denotes the jth non-anchor node Y _j The coordinates of (a);

(3) setting an adjacent node set of non-anchor nodes;

defining non-anchor nodes Y _j Adjacent node set of

Wherein, K _j Is formed by a distance non-anchor node Y _j Nearest 3 nodes d _j1 ,d _j2 ,d _j3 Forming;

(4) establishing a distributed positioning model of the non-anchor node position;

assuming a set of non-anchor nodes

Wherein all non-anchor nodes are located in the anchor node set

Within the convex hull of the anchor node, traverse through the set of non-anchor nodes

At each iteration time t, updating the non-anchor node Y _j The position of (a);

wherein p is _j (t) denotes the jth non-anchor node Y _j Coordinate at time t, gamma is the gain factor, p _k Is a non-anchor node Y at time t _j Adjacent node d _jk Position coordinates of (a) _jk Is a non-anchor node Y _j Relative to its neighboring node d _jk The barycentric coordinates of (a);

(5) synchronously and iteratively positioning the position of the non-anchor node under privacy protection;

(5.1) set maximum iteration time t ^* (ii) a Randomly setting anchor node X _i And non-anchor node Y _j In which the anchor node X _i Is set to p _i (0)＝p _i Randomly setting non-anchor node Y _j Is p as the initial position _j (0)；

Setting time-varying incremental noise of each node in wireless sensor network

Wherein anchor node X _i Noise parameter delta of _i 1 is ═ 1; set of non-anchor nodes Y _j Noise parameter delta of _j Satisfy constraints

(5.2) adding a time-varying delta noise sigma to the position estimate of each node at iteration time t _w (t), further enhancing privacy protection;

wherein, zeta 0 & lt 1 is a constant parameter;

(5.3) respectively calculating anchor nodes X at iteration time t _i And a set of non-anchor nodes Y _j The position estimate after introducing time-varying delta noise is:

wherein,

(5.4) after the privacy protection operation of the formula (3), each node sends the position estimation of the current time of the node to the adjacent node, and further obtains the position estimation of each node at the next time;

(5.5) all non-anchor nodes relative to the adjacent node set K _j The barycentric coordinates of (a) are written in a matrix-vector form:

wherein F, H are all sub-random matrices, [ F H ]]Is a row random matrix when the adjacent node d _jk Belonging to non-anchor nodes Y _j When the triangle is split into two sets, [ F H ]] _jk ＝a _jk When the adjacent node d _jk Not belonging to non-anchor nodes Y _j When the triangle is split into two sets, [ F H ]] _jk ＝0，I _s And I _n-s Is an identity matrix;

(5.6) when the iteration time t reaches the set maximum iteration time t ^* Then, a set of non-anchor nodes is calculated

Based on anchor node set

The positions indicated are:

to this end, a set of non-anchor nodes based on a privacy preserving policy

The position location of (2) is completed.

Images