Optimized deployment method based on three-dimensional space wireless sensor network nodes

Abstract

The invention discloses an optimal deployment method for wireless sensor network nodes based on a three-dimensional space, and the method comprises the steps: obtaining a to-be-deployed three-dimensional space model of the wireless sensor network nodes, and segmenting the to-be-deployed three-dimensional space through a KD tree; randomly throwing the movable wireless sensor network nodes to the to-be-deployed three-dimensional space, and obtaining the initial position of each wireless sensor network node; position deployment of wireless sensor network nodes is planned by adopting an ant lion optimization algorithm, a differential variation disturbance strategy is introduced, and the global search capability of the algorithm is improved; the RRT algorithm is used for performing path search, and meanwhile, a cost function is introduced for optimization, so that node movement collision is avoided. The problem of deployment of the wireless sensor network nodes in the three-dimensional space is solved by constructing the three-dimensional environment model, optimizing the particle swarm optimization algorithm, adopting path planning and the like, the search efficiency and convergence precision of the algorithm are improved, collision and interference between the nodes and obstacles and between the nodes are reduced, and the deployment efficiency of the wireless sensor network nodes is improved. The stability, the energy efficiency and the communication quality of the wireless sensor network can be effectively improved.

Description

Technical Field:

[0001] The present invention belongs to the technical field of wireless sensor networks, and particularly relates to an optimized deployment method for nodes of a three-dimensional space wireless sensor network. Background Art:

[0002] A wireless sensor network is formed by a large number of wireless sensor nodes through a self-organizing manner, and can perform real-time monitoring, collaborative sensing on a working area, and collect and process information on a monitored object and the working area. Node deployment is a prerequisite for a wireless sensor network to achieve a predetermined function and is the basis for ensuring the normal operation of the network. An excellent node deployment scheme is a prerequisite for ensuring the normal operation of a wireless sensor network in a working area and having a long life cycle.

[0003] Traditional wireless sensor network deployment methods mostly focus on node arrangement on a two-dimensional plane, ignoring the influence of complex environments and obstacles in three-dimensional space. However, in actual application scenarios, obstacles in three-dimensional space (such as buildings, trees, walls, etc.) will significantly affect the propagation of wireless signals, resulting in a decline in network communication quality and even loss of connectivity between nodes. Therefore, how to find an optimal node deployment location in three-dimensional space and reasonably avoid obstacles to ensure the coverage, connectivity, and stability of the network has become an important research topic in current wireless sensor network deployment.

[0004] Patent No. CN113242562B discloses an invention of a WSNs coverage enhancement method and system. In its solution, the maximum coverage rate of the wireless sensor network is used as a fitness function, and the positions of the ant population and the antlion population are updated through an improved antlion optimization algorithm after multiple iterations to obtain the position of the elite antlion, that is, to calculate the optimal deployment strategy of the corresponding sensor nodes, avoiding coverage holes and a large amount of redundancy of nodes. However, the global search ability of this method is not strong, the convergence speed is not fast enough, and it is limited to two-dimensional space.

[0005] Therefore, designing a deployment scheme for wireless sensor network nodes with a fast convergence speed, strong search ability, and high environmental fitness is a problem that needs to be solved.

[0006] The information disclosed in this background art section is only intended to enhance the overall understanding of the present invention and should not be regarded as an admission or any form of suggestion that this information constitutes prior art already known to those of ordinary skill in the art. Summary of the Invention:

[0007] The object of the present invention is to provide an optimized deployment method for nodes of a three-dimensional space wireless sensor network, using an improved ant lion optimization algorithm to find the optimal deployment scheme, and at the same time using an improved RRT algorithm to perform path planning for movable nodes according to the target points of the deployment scheme, so as to overcome the defects in the above-mentioned prior art.

[0008] To achieve the above object, the present invention provides an optimized deployment method for nodes of a three-dimensional space wireless sensor network, and its steps are as follows:

[0009] S01. Obtain a model of the three-dimensional space to be deployed by laser scanning. The model includes the positions, sizes of obstacles, and physical characteristics of the environment;

[0010] S02. Use a KD tree to divide the obtained three-dimensional space to be deployed;

[0011] S03. Randomly sprinkle movable wireless sensor network nodes into the three-dimensional space to be deployed, and obtain the initial positions of each wireless sensor network node;

[0012] S04. Use an ant lion optimization algorithm optimized by a differential mutation perturbation strategy to plan the position deployment of wireless sensor network nodes;

[0013] S05. Use an RRT algorithm optimized by introducing a cost function to construct an obstacle avoidance path planning from the initial node to the target point.

[0014] Preferably, in the technical solution, in step S02, the wireless sensor network nodes are surrounded by cubes with a slightly larger volume. The cube is regarded as a wireless sensor network node, and the volume size of the cube is set as the segmentation reference parameter, where the side length of the cube is equal to the diameter size of the wireless sensor network node; use a KD tree to divide the three-dimensional space to be deployed into cube spaces of the same size according to the reference parameter; check whether each cube space completely or partially contains an object. If there is no object part in a cube space, it can be removed.

[0015] Preferably, in the technical solution, in step S03, a three-dimensional space coordinate system is established, and the position of each cube in the three-dimensional space to be deployed is regarded as a coordinate point, and the initial positions of each wireless sensor network node are recorded according to the coordinate points.

[0016] Preferably, in the technical solution, in step S04, the position deployment process of the wireless sensor network nodes is as follows:

[0017] 4.1. Set the parameters of the ant lion optimization algorithm. The parameters of the ant lion optimization algorithm include the number of ant populations, the number of ant lion populations, the dimension and range of the search space, and the maximum number of iterations;

[0018] 4.2. Calculate the fitness values of each individual in the ant population and the antlion population, sort them in descending order, and assign the optimal individual to the elite antlion;

[0019] 4.3. Update the ant population using the differential mutation perturbation strategy, and generate new ant individuals using differential mutation;

[0020] 4.4. Randomly initialize the positions of ants and antlions, and simulate the random walk behavior of ants; perform normalization processing on the ant positions to ensure that the random walk of ants is always within the feasible region; antlions build traps, and the random walk behavior of ants located in the traps is affected:

[0021] 4.5. As the number of iterations increases, the activity range of ants gradually decreases, and then the elite antlion is selected;

[0022] 4.6. During the iteration process, ants perform random walks around the elite antlion and the antlion selected through the roulette wheel mechanism to determine their positions;

[0023] 4.7. Compare the fitness values of the antlion selected through the roulette wheel mechanism and the elite antlion. If the fitness value of the antlion selected through the roulette wheel mechanism is better than that of the elite antlion, the antlion preys on the ant and updates the position of the antlion population, and a new elite antlion is selected;

[0024] 4.8. Determine whether the number of iterations has reached the pre-set parameter; if not, increment the current iteration number t by 1, jump to step 4.3, and continue to execute; if so, end the algorithm and output the position of the elite antlion, which is the target position of node deployment.

[0025] Preferably, in the technical solution, in step 4.3, the differential mutation perturbation strategy formula is:

[0026] X(t) * = F[X AL - X(t)] - F[X A - X(t)] (1),

[0027] where t represents the current iteration number, F represents a random number between [0, 1], X AL is the current elite antlion individual, X A represents the antlion individual selected through the roulette wheel mechanism, X(t) is the current ant individual, and X(t) * represents the ant individual obtained by mutation update.

[0028] Preferably, in the technical solution, in step 4.4, if the random walk behavior of ants is represented as X(t), then:

[0029] X(t) = [0, cumsum(2r(t 1) - 1), cumsum(2r(t 2 ) - 1),..., cumsum(2r(t n ) - 1)](2),

[0030] Where cumsum() is the cumulative walking step of the ant, n represents the number of ants; r(t) is a random function that randomly selects 0 or 1 according to probability;

[0031] The normalization process of the ant position is expressed as:

[0032]

[0033] Where a i and b i respectively represent the lower and upper bounds of the random walking range of the i-th variable of the ant, and respectively represent the lower and upper bounds of the random walking range of the i-th variable of the ant at the t-th iteration;

[0034] The random walking behavior of the ant in the trap is expressed as:

[0035]

[0036] Where c t and d t respectively represent the lower and upper bounds of all variables of the ant at the t-th iteration, is the position of the antlion at the t-th iteration.

[0037] Preferably, in the technical solution, in step 4.5, the range of ant activities and decrease with the increase of the number of iterations, expressed as:

[0038]

[0039] Where w is a constant and iters represents the maximum number of iterations.

[0040] Preferably, in the technical solution, in step 4.6, the determined position of the ant is:

[0041]

[0042] Where represents the new position of the random walk of the antlion selected by the roulette mechanism at the t-th iteration, represents the new position of the random walk of the elite antlion at the t-th iteration, represents the position generated after weighted fusion of the i-th ant at the t-th iteration.

[0043] Preferably, in the technical solution, in step 4.7, the update of the elite antlion is expressed as:

[0044]

[0045] wherein, represents the fitness value of the elite antlion, represents the fitness value of the antlion selected by the roulette wheel mechanism.

[0046] Preferably, in the technical solution, in step S05, the obstacle avoidance path planning process is as follows:

[0047] 5.1 Randomly collect node X in the three-dimensional space to be deployed rand , and put the node X rand nearest to X near into the set T;

[0048] 5.2 Let X near grow along the direction pointing to X rand , and obtain a new node X new by specifying the step size;

[0049] 5.3 Introduce a cost function to impose constraints on the generation of random points;

[0050] 5.4 Perform a collision detection on the connection line between X near and X new . If a collision occurs, perform the next round of loop. If no collision occurs, put the new node X new into the node set T, and set X near as the parent node of X new ;

[0051] 5.5 If X new is equal to the target end point X goal , or the distance from X new to X goal is less than the set threshold, the path is successfully found, that is, exit the loop.

[0052] Preferably, in the technical solution, in step 5.3, the introduced cost function is:

[0053] f(n) = g(n) + h(n) (10),

[0054] where f(n) is the evaluation function from the initial point through node n to the target node, g(n) is the actual cost from the initial node to node n in the state space, and h(n) is the estimated cost of the best path from node n to the target node.

[0055] Compared with the prior art, the present invention has the following beneficial effects:

[0056] The KD tree is used to divide the three-dimensional space to be deployed into cube spaces of the same size as the wireless sensor network nodes, clarifying the coordinate positions of the nodes and obstacles in the three-dimensional space, which helps to quickly find the target position and perform path planning and deployment. The ant lion optimization algorithm using the differential mutation perturbation strategy enriches the diversity of the population and improves the global search ability of the algorithm. The RRT algorithm optimized by introducing a cost function is used to perform path planning for the movable nodes, effectively avoiding collisions between the nodes and obstacles. The integrated optimization deployment and path planning of nodes in a three-dimensional complex environment are realized. Brief Description of the Drawings:

[0057] Figure 1 It is a flowchart of the optimization deployment method for wireless sensor network nodes based on a three-dimensional space according to the present invention;

[0058] Figure 2 It is a flowchart of the ant lion optimization algorithm improved by the differential mutation strategy according to the present invention. Detailed Embodiment:

[0059] The following is a detailed description of the specific embodiments of the present invention, but it should be understood that the protection scope of the present invention is not limited by the specific embodiments.

[0060] Unless otherwise clearly stated, throughout the specification and claims, the term "comprising" or its variations such as "including" or "having" etc. will be understood to include the stated elements or components, without excluding other elements or other components.

[0061] As Figure 1 shown, for the optimization deployment method of wireless sensor network nodes based on a three-dimensional space, the steps are as follows:

[0062] S01. Laser scanning is used to obtain the model of the three-dimensional space to be deployed, and the model includes the positions, sizes of the obstacles and the physical characteristics of the environment;

[0063] S02. The obtained three-dimensional space to be deployed is divided using the KD tree; the wireless sensor network nodes are surrounded by cubes with slightly larger volumes, and this cube is regarded as the wireless sensor network node, and the volume size of this cube is set as the cutting control parameter, where the side length of the cube is equal to the diameter size of the wireless sensor network node; the three-dimensional space to be deployed is divided into cube spaces of the same size according to the control parameter using the KD tree; check whether each cube space completely or partially contains an object, and if there is no object part in a cube space, it can be removed;

[0064] Each cuboid space is regarded as a coordinate point, which helps to quickly find the final deployment position after iterative update by the ant lion optimization algorithm and set its coordinate point as the final target point of the RRT algorithm path planning;

[0065] S03. Randomly sprinkle the nodes of the mobile wireless sensor network into the three-dimensional space to be deployed, and obtain the initial positions of each wireless sensor network node; establish a three-dimensional space coordinate system, regard the position of each cube in the three-dimensional space to be deployed as a coordinate point, and record the initial positions of each wireless sensor network node according to the coordinate points;

[0066] S04. Use the ant lion optimization algorithm optimized by the differential mutation perturbation strategy to plan the position deployment of the wireless sensor network nodes;

[0067] As Figure 2 shown, the process of position deployment of the wireless sensor network nodes is as follows:

[0068] 4.1. Set the parameters of the ant lion optimization algorithm, and the parameters of the ant lion optimization algorithm include the number of ant populations, the number of ant lion populations, the dimension and range of the search space, and the maximum number of iterations;

[0069] 4.2. Calculate the fitness values of each individual in the ant population and the ant lion population, sort them in descending order, and assign the optimal individual to the elite ant lion;

[0070] 4.3. Use the differential mutation perturbation strategy to update the ant population and generate new ant individuals by differential mutation;

[0071] The formula of the differential mutation perturbation strategy is:

[0072] X(t) * = F[X AL - X(t)] - F[X A - X(t)] (1),

[0073] where t represents the current iteration number, F represents a random number between [0, 1], X AL is the current elite ant lion individual, X A represents the ant lion individual selected by the roulette wheel mechanism, X(t) is the current ant individual, and X(t) * represents the ant individual obtained by mutation update;

[0074] 4.4. Randomly initialize the positions of ants and ant lions, and simulate the random walk behavior X(t) of ants, then:

[0075] X(t) = [0, cumsum(2r(t 1 ) - 1), cumsum(2r(t 2) - 1),..., cumsum(2r(t n ) - 1)](2),

[0076] where cumsum() is the cumulative walking steps of the ants, n represents the number of ants; r(t) is a random function that randomly selects 0 or 1 according to probability;

[0077] Normalize the ant positions to ensure that the ants' random walks are always within the feasible region,

[0078]

[0079] where a i and b i represent the lower and upper bounds of the random walk range of the i-th variable of the ant respectively, and represent the lower and upper bounds of the random walk range of the i-th variable of the ant at the t-th iteration respectively;

[0080] The antlion constructs a trap, and the random walk behavior of the ants in the trap is expressed as:

[0081]

[0082] where c t and d t represent the lower and upper bounds of all variables of the ant at the t-th iteration respectively, is the position of the antlion at the t-th iteration;

[0083] 4.5. The range of ant activities and decrease as the number of iterations increases, and then the elite antlions are selected:

[0084]

[0085] where w is a constant and iters represents the maximum number of iterations;

[0086] 4.6. During the iteration process, the ants perform random walks around the elite antlions and the antlions selected through the roulette wheel mechanism to determine their positions:

[0087]

[0088] where represents the new position of the random walk of the antlion selected by the roulette wheel mechanism at the t-th iteration, represents the new position of the random walk of the elite antlion at the t-th iteration, represents the position generated after weighted fusion of the i-th ant at the t-th iteration;

[0089] 4.7. Compare the fitness values of the ant lions selected through the roulette wheel mechanism and the elite ant lions. If the fitness value of the ant lion selected through the roulette wheel mechanism is better than that of the elite ant lion, the ant lion preys on the ant and updates the position of the ant lion population, and selects a new elite ant lion;

[0090]

[0091] Among them, represents the fitness value of the elite ant lion, represents the fitness value of the ant lion selected by the roulette wheel mechanism;

[0092] 4.8. Determine whether the number of iterations reaches the preset parameter; if not, increment the current iteration number t by 1, jump to step 4.3, and continue to execute; if so, end the algorithm and output the position of the elite ant lion, that is, the target position of node deployment

[0093] S05. Use the RRT algorithm optimized by introducing a cost function to construct an obstacle avoidance path planning from the initial node to the target point; the process of obstacle avoidance path planning is as follows:

[0094] 5.1. Randomly collect a node X rand in the three-dimensional space to be deployed, and put the node X rand nearest to X near into the set T;

[0095] 5.2. Let X near grow along the direction pointing to X rand , and obtain a new node X new by specifying a step size;

[0096] 5.3. Introduce a cost function to impose a constraint condition on the generation of random points;

[0097] f(n) = g(n) + h(n) (10),

[0098] where f(n) is the evaluation function from the initial point through node n to the target node, g(n) is the actual cost from the initial node to node n in the state space, and h(n) is the estimated cost of the best path from node n to the target node;

[0099] Take f(n) as one of the decision conditions for generating random points, perform cost detection on the newly generated random points, and select the point with f(n) less than the value of its parent node plus a step size distance as the expansion point; here, g(n) is the product of the step size and the number of random points generated on the current path; for the calculation of the estimated cost h(n), in order to ensure the randomness of node generation in the RRT algorithm, the Euclidean distance that can grow in any direction is selected;

[0100] 5.4. For X near and Xnew Perform collision detection on the connection lines. If a collision occurs, perform the next round of loop. If no collision occurs, add the new node X new to the node set T and set X near as the parent node of X new ;

[0101] 5.5. If X new is equal to the target end point X goal , or the distance from X new to X goal is less than the set threshold, the path is successfully found, i.e., exit the loop.

[0102] The foregoing description of the specific exemplary embodiments of the present invention is for the purposes of illustration and exemplification. These descriptions are not intended to limit the invention to the precise forms disclosed, and it is apparent that many changes and variations are possible in light of the above teaching. The purpose of selecting and describing the exemplary embodiments is to explain the specific principles of the invention and its practical applications, so that those skilled in the art can implement and utilize the various different exemplary embodiments of the invention, as well as various different selections and changes. The scope of the invention is intended to be defined by the claims and their equivalents.

Claims

1. The optimized deployment method of wireless sensor network nodes based on three-dimensional space includes the following steps: S01. Laser scanning obtains a model of the three-dimensional space to be deployed, the model including the location and size of obstacles and the physical characteristics of the environment; S02. Use a KD tree to segment the acquired three-dimensional space to be deployed; S03, randomly scattering the movable wireless sensor network nodes to the three-dimensional space to be deployed, and obtaining the initial position of each wireless sensor network node; S04, using the ant lion optimization algorithm optimized by differential mutation perturbation strategy to plan the location deployment of wireless sensor network nodes; S05. Use the RRT algorithm with cost function optimization to construct an obstacle avoidance path planning from the initial node to the target point.

2. The method for optimizing deployment of three-dimensional wireless sensor network nodes according to claim 1, characterized in that: In step S02, the wireless sensor network node is surrounded by a slightly larger cube, which is regarded as the wireless sensor network node, and the size of the cube is set as the cut comparison parameter, where the side length of the cube is equal to the diameter of the wireless sensor network node; the KD tree is used to divide the three-dimensional space to be deployed into cube spaces of the same size according to the comparison parameter; and each cube space is checked to see whether it completely or partially contains an object. If a cube space does not contain part of the object, it can be removed.

3. The method for optimizing deployment of three-dimensional wireless sensor network nodes according to claim 1, characterized in that: In step S03, a three-dimensional space coordinate system is established, and the location of each cube in the three-dimensional space to be deployed is regarded as a coordinate point, and the initial position of each wireless sensor network node is recorded according to the coordinate point.

4. The method for optimizing deployment of nodes in a three-dimensional wireless sensor network according to claim 1, characterized in that: In step S04, the location deployment process of the wireless sensor network nodes is as follows: 4.

1. Set the parameters of the Ant Lion Optimization Algorithm, which include the number of ant populations, the number of ant lion populations, the dimension and range of the search space, and the maximum number of iterations; 4.

2. Calculate the fitness value of each individual in the ant population and antlion population, sort them in descending order, and assign the best individual to the elite antlion; 4.

3. Use differential mutation disturbance strategy to update the ant population and generate new ant individuals by differential mutation; 4.

4. Randomly initialize the positions of ants and antlions to simulate the random walk of ants; normalize the positions of ants to ensure that the random walk of ants is always within the feasible domain; antlions build traps, and the random walk of ants in the traps is affected: 4.

5. As the number of iterations increases, the range of ant activities gradually decreases, and elite ant lions are selected; 4.

6. During the iteration process, the ants randomly walk around the elite antlion and the antlion selected by the roulette mechanism to determine their positions; 4.

7. Compare the fitness values ​​of the antlion selected by the roulette mechanism and the elite antlion. If the antlion selected by the roulette mechanism has a better fitness value than the elite antlion, the antlion preys on ants and updates the position of the antlion population to select a new elite antlion. 4.

8. Determine whether the number of iterations reaches the preset parameter; if not, add 1 to the current number of iterations t, jump to step 4.3 and continue execution; if so, end the algorithm and output the elite ant lion position, that is, the node deployment target position.

5. The method for optimizing deployment of three-dimensional wireless sensor network nodes according to claim 4, characterized in that: In step 4.3, the differential mutation perturbation strategy formula is: X(t) * =F[X AL -X(t)]-F[X A -X(t)](1), Where t represents the current iteration number, F represents a random number between [0,1], and X AL Is the current elite antlion individual, X A represents the antlion individual selected by the roulette mechanism, X(t) is the current ant individual, and X(t) * Represents the ant individual obtained by mutation update.

6. The method for optimizing deployment of three-dimensional wireless sensor network nodes according to claim 5, characterized in that: In step 4.4, the random walk behavior of the ant is represented by X(t), then: X(t)=[0,cumsum(2r(t1)-1),cumsum(2r(t2)-1),...,cumsum(2r(t n )-1)](2), Where cumsum() is the cumulative walking step length of the ants, n is the number of ants; r(t) is a random function that randomly selects 0 or 1 according to probability; The normalized processing of ant positions is expressed as: where a i and b i They represent the lower and upper bounds of the random walk range of the ant’s i-th variable, and They represent the lower and upper bounds of the random walk range of the ant’s i-th variable at the t-th iteration respectively; The random walk behavior of an ant in a trap is expressed as: where c t and d t They represent the lower and upper bounds of all ant variables at the tth iteration, is the position of the antlion at the tth iteration.

7. The method for optimizing deployment of three-dimensional wireless sensor network nodes according to claim 6, characterized in that: In step 4.5, the range of ant activities and It decreases as the number of iterations increases, expressed as: Where w is a constant and iters represents the maximum number of iterations.

8. The method for optimizing deployment of three-dimensional wireless sensor network nodes according to claim 7, characterized in that: In step 4.6, the ant determines the position as: in represents the new position of the antlion random walk selected by the roulette mechanism at the tth iteration, represents the new position of the elite ant lion's random walk at the tth iteration, It represents the position of the i-th ant after weighted fusion at the t-th iteration.

9. The method for optimizing deployment of nodes in a three-dimensional wireless sensor network according to claim 8, characterized in that: In step 4.7, the update of the elite antlion is expressed as: in, Represents the fitness value of the elite antlion, Represents the fitness value of the antlion selected by the roulette mechanism.

10. The method for optimizing deployment of three-dimensional wireless sensor network nodes according to claim 1, characterized in that: In step S05, the obstacle avoidance path planning process is: 5.

1. Randomly collect nodes X in the three-dimensional space to be deployed rand , and the distance X rand The nearest node X near Put it into the set T; 5.

2. Let X near Along the direction X rand The direction of growth is specified, and the new node X is obtained by specifying the step size. new ; 5.

3. Introduce the cost function to impose constraints on the generation of random points; 5.

4. X near With X new If a collision occurs, the next cycle will be performed. If no collision occurs, the new node X new Put it into the node set T and set X near For X new The parent node of 5.5 If X new Equal to the target terminal node X goal , or X new To X goal If the distance is less than the set threshold, the path is successfully found and the loop is exited.

11. The method for optimizing deployment of three-dimensional wireless sensor network nodes according to claim 10, characterized in that: In step 5.3, the introduced cost function is: f(n)=g(n)+h(n)(10), Where f(n) is the evaluation function from the initial point via node n to the target node, g(n) is the actual cost from the initial node to node n in the state space, and h(n) is the estimated cost of the best path from node n to the target node.

Images