A distributed sensor target positioning method, device, and storage medium

Abstract

The present invention discloses a distributed sensor target positioning method, device, and storage medium. The method includes: obtaining node information of sensors; using a two-way learning particle swarm optimization framework with an adaptive communication interval adjustment mechanism and an external learning particle update strategy to optimize the node information: each node performs internal learning once or multiple times; neighbor nodes communicate with each other and transmit the current population solutions; an external learning is performed on the nodes; if the population converges, the optimization ends; otherwise, the communication interval is adaptively adjusted according to the change of the fitness value, and the next round of learning is continued. Through the distributed particle swarm optimization framework and the external learning strategy, the present invention applies the particle swarm optimization algorithm to distributed optimization problems, balances the exploration of the population and the consensus of the system, effectively improves the optimization performance of the algorithm, and ensures the consensus of the system. The present invention can be widely applied to the technical field of wireless sensor network target positioning.

Description

Technical Field

[0001] The present invention relates to the technical field of wireless sensor network target positioning, and particularly to a distributed sensor target positioning method, device and storage medium. Background Art

[0002] A wireless sensor network is a distributed system composed of a large number of spatially dispersed sensors. Thanks to the sensing, computing, communication and other capabilities of each sensor itself, a wireless sensor network can achieve many complex tasks, such as target positioning, environmental detection, emergency response, etc. Considering requirements such as network robustness, scalability, and low energy consumption of sensors, distributed optimization is a key technology for the deployment and optimization of wireless sensor networks.

[0003] The multi-target positioning problem in a wireless sensor network is a non-convex optimization problem, and an algorithm with good optimization performance and guaranteed convergence ability needs to be sought. Most of the existing distributed optimization technologies are gradient-based optimization algorithms. However, most of these methods can only optimize optimization problems with a convex objective function and require the objective function to satisfy certain mathematical properties. Summary of the Invention

[0004] To solve at least one of the technical problems existing in the prior art to a certain extent, the purpose of the present invention is to provide a distributed sensor target positioning method, device and storage medium.

[0005] The technical solution adopted by the present invention is as follows:

[0006] A distributed sensor target positioning method includes the following steps:

[0007] Obtain node information of sensors; wherein, the node information includes location information and routing information;

[0008] Adopt a two-way learning particle swarm optimization framework with an adaptive communication interval adjustment mechanism and an external learning particle update strategy to optimize the node information:

[0009] Each node performs one or more internal learning; neighbor nodes communicate with each other and transfer the current population solution;

[0010] Perform external learning on the nodes once; if the population converges, end the optimization; otherwise, adaptively adjust the communication interval according to the change of the fitness value and continue the next round of learning;

[0011] Wherein, the external learning particle update strategy is used to cooperate with different nodes to optimize the global objective.

[0012] Furthermore, the two-way learning particle swarm optimization framework with an adaptive communication interval adjustment mechanism is used to design the two processes of optimizing the local objective and the global objective of nodes in distributed optimization as two particle learning strategies, and the execution frequencies of the two learning strategies are different. By adaptively adjusting the execution frequency, that is, the communication interval, the swarm exploration and system consensus are balanced.

[0013] Furthermore, the adaptive adjustment of the execution frequency is achieved in the following way:

[0014] Before the start of the algorithm, initialize the fitness value archive D i ∈R n , and initialize the best fitness value bf = inf;

[0015] After internal learning and external learning, update the fitness value archive D according to the routing table i (i), send D i to neighbor nodes and receive the fitness value archive D sent by neighbor nodes j ; the node will update its own archive D according to the received information i , if the information of node a in the routing table comes from neighbor node b, then adopt the information of node b, that is, D i (a) = D b (a);

[0016] If then update the best fitness value bf; otherwise, if the best fitness value bf has not been updated for consecutive prediction times, then reduce the communication interval k = k - 1; where n represents the number of sensor nodes, and D i (j) represents the fitness value of node j recorded in node i.

[0017] Furthermore, the inertial velocity of the particle learning adopts the velocity of the previous external learning instead of the velocity of the particle's previous update to improve the convergence speed and stability of the algorithm.

[0018] Furthermore, the external learning particle update strategy is:

[0019]

[0020] x i，a (t + k + 1) = x i，a (t + k) + v i，a (t + k + 1)

[0021] where ω ij is the learning weight, satisfying r4 is a random number in the range [0, 1]; v i，a (t + k + 1) represents the particle velocity, vi，a (t) represents the inertial velocity of the particle, x j，a (t + k) represents the position of particle a in node j, x i，a (t + k) represents the position of particle a in node i.

[0022] Furthermore, the two-way learning particle swarm optimization framework with an adaptive communication interval adjustment mechanism and the external learning particle update strategy are adopted to optimize the node information, including:

[0023] A1. Randomly initialize the population p in each sensor node i ; Initialize the fitness value archive D i ∈R n , initialize the communication interval k = k0, initialize the best fitness value bf = inf; Initialize the routing table R through the flooding mechanism;

[0024] A2. Select learning objects for internal learning. Each individual selects two particles from its own current node as learning objects;

[0025] A3. Conduct internal learning to update the velocity and position of the particles;

[0026] A4. Update the fitness value of the particles;

[0027] A5. If the internal learning has been carried out for a preset number of times, go to step A6; otherwise, return to step A2;

[0028] A6. Send the solution {x i，1 , x i，n ,..., x i，m} of the current population to the neighbor nodes and receive the population solutions sent by the neighbor nodes

[0029] A7. Conduct external learning to update the velocity and position of the particles;

[0030] A8. Update the fitness value archive D i (i) according to the routing table, send D i to the neighbor nodes and receive the fitness value archive D j sent by the neighbor nodes;

[0031] A9. If then update the best fitness value bf and go to step A11; otherwise, go to step A10;

[0032] A10. If the best fitness value bf has not been updated for a continuous preset number of times, then decrease the communication interval k = k - 1;

[0033] A11. If the population has converged, end the program; otherwise, enter step A2.

[0034] Furthermore, internal learning is carried out through the following equations to update the velocity and position of the particles:

[0035] v i，a (t + 1) = r1v i，a (t) + r2(x i，b (t) - x i，a (t)) + r3(x i，c (t) - x i，a (t))

[0036] x i， a(t + 1) = x i，a (t) + v i，a (t + 1)

[0037] In the formula, x i，a is the currently updated particle, x i，b and x i，c are two selected learning objects, and r1, r2, and r3 are all random numbers within the range [0, 1]; v i，a represents the inertial velocity of the a-th particle in node i;

[0038] The fitness value of the particle is updated through the following equation:

[0039]

[0040] In the formula, N T represents the number of targets, is the distance between the sensor itself and the t-th target detected by the sensor, and x i，a，t is the position of the t-th target estimated by the current particle, and ω i is the position of the sensor itself.

[0041] Furthermore, the optimization goal is to locate the positions of multiple detected targets simultaneously. Therefore, the local goal of each sensor is set as the difference value between the sensor detection distance and the estimated position distance;

[0042] In the sensor target localization problem, the global optimization goal is the sum of the local goals:

[0043]

[0044] In the formula, f i (x) represents the predicted loss value of the i-th sensor, and n represents the number of sensor nodes.

[0045] Another technical solution adopted by the present invention is:

[0046] A distributed sensor target localization device, comprising:

[0047] At least one processor;

[0048] At least one memory for storing at least one program;

[0049] When the at least one program is executed by the at least one processor, the at least one processor implements the above-mentioned method.

[0050] Another technical solution adopted by the present invention is:

[0051] A computer-readable storage medium storing a program executable by a processor, the program executable by the processor being used to execute the method as described above when executed by the processor.

[0052] The beneficial effects of the present invention are: Through the distributed particle swarm optimization framework and the external learning strategy, the particle swarm optimization algorithm is applied to the distributed optimization problem, balancing the population exploration and the system consensus, effectively improving the optimization performance of the algorithm and ensuring the system consensus. Description of the Drawings

[0053] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following introduces the relevant technical solution drawings in the embodiments of the present invention or the prior art. It should be understood that the drawings below only conveniently and clearly illustrate some embodiments of the technical solutions in the present invention. For those skilled in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0054] Figure 1 is the flowchart of the bidirectional particle swarm optimization algorithm in the embodiment of the present invention;

[0055] Figure 2 is the flowchart of the method for generating a routing table based on broadcast flooding in the embodiment of the present invention. Detailed Embodiments

[0056] The following details the embodiments of the present invention. The examples of the embodiments are shown in the drawings, where the same or similar reference numerals represent the same or similar elements or elements with the same or similar functions from beginning to end. The embodiments described below with reference to the drawings are exemplary and are only used to explain the present invention and should not be construed as a limitation to the present invention. For the step numbers in the following embodiments, they are only set for the convenience of explanation and no limitation is made on the order between the steps. The execution order of each step in the embodiments can be adaptively adjusted according to the understanding of those skilled in the art.

[0057] In the description of the present invention, it should be understood that for the orientation description, such as the orientation or positional relationship indicated by up, down, front, back, left, right, etc., it is based on the orientation or positional relationship shown in the drawings. It is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation to the present invention.

[0058] In the description of the present invention, the meaning of several is one or more, the meaning of multiple is more than two, greater than, less than, exceeding, etc. are understood as not including the present number, above, below, within, etc. are understood as including the present number. If there is a description of first and second, it is only for the purpose of distinguishing technical features, and cannot be understood as indicating or implying relative importance or implicitly indicating the quantity of the indicated technical features or implicitly indicating the sequence relationship of the indicated technical features.

[0059] In the description of the present invention, unless otherwise clearly defined, words such as set, install, connect, etc. should be understood in a broad sense, and those skilled in the art can reasonably determine the specific meanings of the above words in the present invention in combination with the specific content of the technical solution.

[0060] Evolutionary computation does not depend on the mathematical characteristics of the problem to be solved and has excellent global search ability. Therefore, it has certain advantages in solving non-convex sensor target localization problems. Based on this, as Figure 1 shown, this embodiment provides a distributed sensor target localization method based on bidirectional learning particle swarm optimization, including the following steps:

[0061] S1. Randomly initialize the population p i in each sensor node. Each individual is encoded by a vector, representing the coordinates of all targets to be detected. In this embodiment, three-dimensional coordinates (x, y, z) are used. Therefore, the vector length is 3 times the number of targets to be measured. Initialize the fitness value archive D i ∈R n , initialize the best fitness value bf = inf. Initialize the communication interval k = k0. In this embodiment, k0 is 4. Initialize the routing table R through the flooding mechanism. Specifically, before the algorithm starts, the distributed sensor network will perform a flooding. The specific process is as shown in the flowchart in the appendix Figure 2 . First, each node sends a data packet to its neighbor nodes, identifies the data source, and sets the hop count to 0; after a node receives a data packet sent from a certain neighbor node, if the hop count of the data packet is greater than the hop count recorded in the routing table, it will be ignored; otherwise, the hop count is incremented by 1 and forwarded to other neighbor nodes. When the flooding ends, each node has successfully generated a routing table, recording the information sources of all other nodes in the network.

[0062] S2. Select learning objects for internal learning. Each individual selects two particles from its current node as learning objects. The learning object selection strategy in internal learning can apply existing particle swarm optimization methods, such as the level-based particle swarm algorithm (LLSO) and the competition-based particle swarm algorithm (CSO). It should be noted that the learning objects in the embodiments of the present invention must be selected from the current population, and the historical global optimal solution or the historical personal optimal solution cannot be selected as learning objects.

[0063] S3. Perform internal learning through the following formula to update the velocity and position of the particles:

[0064] v i，a (t + 1) = r1v i，a (t) + r2(x i,b (t) - x i，a (t)) + r3(x i，c (t) - x i，a (t)) # (6)

[0065] x i，a (t + 1) = x i，a (t) + v i，a (t + 1) # (7)

[0066] Among them, x i，a is the currently updated particle, x i，b and x i，c are the two selected learning objects, and r1, r2, and r3 are all random numbers within the range [0, 1].

[0067] S4. Update the particle fitness value through the following formula. The optimization goal of this embodiment is to simultaneously locate the positions of multiple detection targets. Therefore, the local goal of each sensor is set to the difference between the sensor detection distance and the estimated position distance. Among them, is the distance between the sensor itself and the t-th target detected, x i，a，t is the position of the t-th target estimated by the current particle, ω i is the position of the sensor itself.

[0068]

[0069] In the sensor target localization problem, the global optimization goal is the sum of the local goals:

[0070]

[0071] S5. If k times of internal learning have been performed, go to step S6; otherwise, return to step S2.

[0072] S6. The solution of the current population {xi，1 , x i，2 , ..., x i，m} is sent to neighbors and the population solutions sent by neighbors are received

[0073] S7. External learning is performed through the following formula to update the velocity and position of the particles:

[0074]

[0075] x i，a (t + k + 1) = x i，a (t + k) + v i，a (t + k + 1) # (11)

[0076] where ω ij is the learning weight, satisfying r4 is a random number within the range [0, 1]. Then, the fitness value of the particle is updated according to Equation (8). It should be noted that the inertial velocity used in external learning here is v i，a (t), that is, the velocity of the previous external learning.

[0077] S8. Update the fitness value file according to the routing table Send D i to neighbors and receive the fitness value file D sent by neighbors j . The node will update its own file D according to the received information i , if the information of node a in the routing table comes from neighbor node b, then the information of node b is adopted, that is, D i (a) = D b (a).

[0078] S9. If then update the best fitness value bf and perform step S11; otherwise, perform step S10.

[0079] S10. If the best fitness value bf has not been updated for 10 consecutive times, then reduce the communication interval k = k - 1.

[0080] S11. If the population has converged, end the program; otherwise, enter step S2.

[0081] In summary, the two-way particle swarm optimization algorithm proposed by the present invention can handle non-convex distributed optimization problems, ensure that the system can finally reach a consensus, and can effectively solve the multi-target localization problem in wireless sensor networks.

[0082] This embodiment also provides a distributed sensor target localization device, including:

[0083] At least one processor;

[0084] At least one memory for storing at least one program;

[0085] When the at least one program is executed by the at least one processor, such that the at least one processor implements Figure 1 and Figure 2 the method shown.

[0086] A distributed sensor target positioning device according to this embodiment can execute a distributed sensor target positioning method provided by an embodiment of the method of the present invention, can execute any combination of implementation steps of the method embodiment, and has corresponding functions and beneficial effects of the method.

[0087] This application embodiment also discloses a computer program product or a computer program. The computer program product or the computer program includes computer instructions, and the computer instructions are stored in a computer-readable storage medium. A processor of a computer device can read the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, such that the computer device executes Figure 1 and Figure 2 the method shown.

[0088] This embodiment also provides a storage medium storing instructions or a program that can execute a distributed sensor target positioning method provided by an embodiment of the method of the present invention. When the instructions or the program are run, any combination of implementation steps of the method embodiment can be executed, and corresponding functions and beneficial effects of the method are achieved.

[0089] In some alternative embodiments, the functions / operations mentioned in the block diagram may not occur in the order mentioned in the operation diagram. For example, depending on the functions / operations involved, two consecutive blocks shown can actually be executed substantially simultaneously or the blocks can sometimes be executed in the reverse order. In addition, the embodiments presented and described in the flowcharts of the present invention are provided by way of example for the purpose of providing a more comprehensive understanding of the technology. The disclosed method is not limited to the operations and logical flows presented herein. Alternative embodiments are anticipated, where the order of various operations is changed and sub-operations described as part of a larger operation are executed independently.

[0090] In addition, although the present invention has been described in the context of functional modules, it should be understood that, unless otherwise stated to the contrary, one or more of the described functions and / or features may be integrated in a single physical device and / or software module, or one or more functions and / or features may be implemented in separate physical devices or software modules. It should also be understood that a detailed discussion of the actual implementation of each module is not necessary for understanding the present invention. Rather, given the attributes, functions, and internal relationships of the various functional modules in the devices disclosed herein, the actual implementation of the modules will be understood within the ordinary skills of an engineer. Thus, those skilled in the art can implement the present invention as set forth in the claims without undue experimentation. It should also be understood that the specific concepts disclosed are merely illustrative and are not intended to limit the scope of the present invention, which is determined by the full scope of the appended claims and their equivalents.

[0091] If the described function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs that can store program codes.

[0092] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, apparatus, or device (such as a computer-based system, a system including a processor, or other systems that can fetch and execute instructions from the instruction execution system, apparatus, or device), or in conjunction with such instruction execution systems, apparatuses, or devices. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0093] More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection (electronic device) having one or more wirings, a portable computer diskette (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable media can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpretation, or otherwise processing as appropriate, and then storing it in a computer memory.

[0094] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware, or a combination thereof. In the above-described embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, any one or a combination of the following techniques well known in the art can be used: discrete logic circuits having logic gate circuits for implementing logical functions on data signals, application specific integrated circuits having appropriate combinational logic gate circuits, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), and the like.

[0095] In the above description of this specification, the descriptions referring to the terms "one embodiment / example", "another embodiment / example", or "certain embodiments / examples", etc. mean that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0096] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the claims and their equivalents.

[0097] The above has specifically described the preferred embodiments of the present invention, but the present invention is not limited to the above embodiments. Those skilled in the art can also make various equivalent deformations or substitutions without departing from the spirit of the present invention, and these equivalent deformations or substitutions are all included within the scope defined by the claims of this application.

Claims

1. A distributed sensor target positioning method, characterized in that, Including the following steps: Obtain the node information of the sensor; wherein, the node information includes location information and routing information; Adopt a two-way learning particle swarm optimization framework with an adaptive communication interval adjustment mechanism and an external learning particle update strategy to optimize the node information: Each node performs internal learning one or more times; neighboring nodes communicate with each other and transfer the current population solution; Perform external learning on the nodes; if the population converges, end the optimization; otherwise, adaptively adjust the communication interval according to the change of the fitness value and continue the next round of learning; Wherein, the external learning particle update strategy is used to cooperate with different nodes to optimize the global objective; The adopting of the two-way learning particle swarm optimization framework with an adaptive communication interval adjustment mechanism and the external learning particle update strategy to optimize the node information includes: A1. Randomly initialize the population p in each sensor node i ; Initialize the fitness value archive D i ∈R n , Initialize the communication interval k = k0, initialize the best fitness value bf = inf; Initialize the routing table R through the flooding mechanism; A2. Select learning objects for internal learning. Each individual selects two particles from its own current node as learning objects; A3. Perform internal learning and update the velocity and position of the particles; A4. Update the fitness value of the particles; A5. If the internal learning has been performed a preset number of times, perform step A6; otherwise, return to step A2; A6. Send the solutions {x i,1 , x i,2 , …, x i,m} of the current population to the neighbor nodes and receive the population solutions sent by the neighbor nodes A7. Perform external learning and update the velocity and position of the particles; A8. Update the fitness value file D according to the routing table i (i). Send D i to the neighbor nodes and receive the fitness value file D sent by the neighbor nodes j ; A9. If then update the best fitness value bf and proceed to step A11; otherwise, proceed to step A10； A10. If the best fitness value bf has not been updated for a preset number of consecutive times, reduce the communication interval k = k - 1; A11. If the population has converged, end the program; otherwise, enter step A2.

2. The distributed sensor target positioning method according to claim 1, wherein The two-way learning particle swarm optimization framework with an adaptive communication interval adjustment mechanism is used to design the two processes of optimizing the local objective and the global objective of nodes in distributed optimization as two particle learning strategies, and the execution frequencies of the two learning strategies are different. By adaptively adjusting the execution frequency, the group exploration and system consensus are balanced.

3. A distributed sensor target localization method according to claim 2, characterized in that, The adaptive adjustment of the execution frequency is achieved in the following way: Before the algorithm starts, initialize the fitness value archive D i ∈R n , initialize the best fitness value bf = inf; After internal learning and external learning, update the fitness profile D according to the routing table i (i), send D i to the neighbor nodes and receive the fitness profile D sent by the neighbor nodes j ; The node will update its own profile D according to the received information i , if the information of node a in the routing table comes from neighbor node b, then adopt the information of node b, that is, D i (a) = D b (a); If then update the best fitness value bf; otherwise, if the best fitness value bf has not been updated for consecutive prediction times, then reduce the communication interval k = k - 1; where n represents the number of network sensors, and D i (j) represents the fitness value of node j recorded in node i.

4. A distributed sensor target localization method according to claim 2, characterized in that, The inertial velocity of the particle learning adopts the velocity of the previous external learning instead of the velocity of the previous update of the particle, so as to improve the convergence speed and stability of the algorithm.

5. A distributed sensor target positioning method according to claim 4, characterized in that, The external learning particle update strategy is: x i,a (t + k + 1) = x i,a (t + k) + v i,a (t + k + 1) Among them, ω ij is the learning weight, satisfying r4 is a random number within the range [0, 1]; v i,a (t + k + 1) represents the particle velocity, v i,a (t) represents the particle inertial velocity, x j,a (t + k) represents the position of particle a in node j, x i,a (t + k) represents the position of particle a in node i.

6. A distributed sensor target positioning method according to claim 1, characterized in that Perform internal learning through the following formula to update the velocity and position of the particles: x i,a (t + 1)= x i,a (t)+ v i,a (t + 1) where x i,a is the currently updated particle, x i,b and x i,c are two selected learning objects, and r1, r2, r3 are all random numbers within the range [0, 1]; v i,a (t) represents the inertial velocity of the particle; Update the fitness value of the particles through the following formula: Where N T represents the number of targets, is the distance between the sensor itself and the t-th target detected by the sensor, and x i,a,t is the position of the t-th target estimated by the current particle, and ω i is the position of the sensor itself.

7. A distributed sensor target positioning method according to claim 6, characterized in that The optimization objective is to locate the positions of multiple detection targets simultaneously. Therefore, the local objective of each sensor is set as the difference value between the sensor detection distance and the estimated position distance; In the sensor target localization problem, the global optimization objective is the sum of the local objectives: where f i (x) represents the predicted loss value of the i-th sensor, and n represents the number of sensor nodes.

8. A distributed sensor target positioning device, characterized in that, Including: At least one processor; At least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements the method according to any one of claims 1 - 7.

9. A computer-readable storage medium storing a program executable by a processor, characterized in that, The program executable by the processor is used to execute the method according to any one of claims 1 - 7 when executed by the processor.

Images