An improved DV-hop positioning method optimized by a compact compression-improved false wild oat algorithm

Abstract

The present invention discloses a DV-Hop positioning method optimized by a compact compression improved false wild oat algorithm. It introduces the least squares method of the false wild oat algorithm for position estimation, and optimizes and compresses the false wild oat algorithm through a Gaussian probability model, thereby improving DV-Hop positioning. This not only improves the accuracy of positioning, but also helps to obtain the position estimation result faster, while reducing the demand for memory resources.

Description

Technical Field

[0001] The present invention relates to the technical field of wireless sensor network node positioning, and particularly to a DV-hop positioning method optimized by a compact compression improved false wild oat algorithm. Background Art

[0002] A wireless sensor network (WSN) consists of numerous inexpensive micro sensor nodes, which are deployed in a monitoring area and establish a multi-hop, self-organizing network system through wireless communication. The tasks of these nodes include sensing, collecting, and processing environmental data in the monitoring area, and then transmitting this information to an observer.

[0003] The application fields of WSN are very extensive, including but not limited to smart home, automated agriculture, forest fire monitoring, and environmental monitoring, etc. However, with limited land resources and the continuous maturity of technology, the focus of research has gradually shifted to the exploration of marine resources. These marine applications include underwater environmental monitoring, military activity support, marine biology tracking, underwater pipeline leak detection, and earthquake and tsunami early warning, etc. These applications also require node location information and data interaction, but due to the high cost and resource limitations of the Global Positioning System (GPS), it cannot be used on each sensor node. Therefore, it is crucial to use a small number of anchor nodes equipped with GPS modules for cooperative positioning.

[0004] The core idea of node positioning algorithms is for an unknown node to obtain the location information of its surrounding anchor nodes, and then indirectly estimate its own location based on the distance coordinate relationship. DV-Hop is a non-ranging algorithm, and its uniqueness lies in that it does not require additional hardware to measure distance, but relies on the number of hops between nodes to estimate distance. Compared with other non-ranging algorithms, DV-Hop shows a relatively high level of accuracy. However, the main defect of this method stems from using the least squares method for position estimation, resulting in a certain accumulation of positioning errors. Therefore, compared with ranging-based solutions, DV-Hop has larger errors. Currently, some intelligent optimization algorithms have emerged to replace the least squares method for position estimation. Although this reduces the positioning error, it will increase the requirements for hardware, especially the use of memory. Summary of the Invention

[0005] The purpose of the present invention is to provide a DV-hop positioning method optimized by a compact compression improved false wild oat algorithm, which can improve the positioning accuracy and reduce the demand for memory resources.

[0006] The purpose of the present invention is achieved through the following technical solutions:

[0007] A DV-hop positioning method optimized by a compact compression improved false wild oat algorithm, comprising:

[0008] Each anchor node transmits its own position information as broadcast information to the surrounding nodes. The nodes receive the broadcast information and record the minimum number of hops from each anchor node to itself. Each of the nodes includes a position node and an anchor node.

[0009] Calculate the average hop distance based on the distance between anchor nodes and the minimum number of hops, and then combine the minimum number of hops from each anchor node to each unknown node to calculate the distance between each anchor node and each unknown node.

[0010] Combine the distances between each anchor node and each unknown node, and establish an optimization model for node positioning in a three-dimensional wireless sensor network based on DV-Hop with the goal of the lowest positioning error.

[0011] Optimize the optimization model for node positioning in a three-dimensional wireless sensor network based on DV-Hop through the compact compressed wild oat algorithm improved by the Gaussian probability model. Among them, use the iteratively updated Gaussian model to initialize the wild oat individuals, substitute each wild oat individual into the optimization model for node positioning in a three-dimensional wireless sensor network based on DV-Hop for fitness evaluation, update the positions of the wild oat individuals through the wild oat algorithm, and iteratively update the Gaussian model. Finally, output the final positions of each wild oat individual, that is, the coordinates of the unknown nodes.

[0012] It can be seen from the technical solutions provided by the present invention described above that it introduces the least squares method of the wild oat algorithm for position estimation, and optimizes and compresses the wild oat algorithm through the Gaussian probability model, thereby improving the DV-Hop positioning. This not only improves the positioning accuracy, but also helps to obtain the position estimation result faster, and at the same time reduces the demand for memory resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0014] Figure 1 It is a flowchart of a DV-hop positioning method optimized by a compact compressed improved wild oat algorithm provided by an embodiment of the present invention.

[0015] Figure 2 It is a main flowchart of a DV-hop positioning method optimized by a compact compressed improved wild oat algorithm provided by an embodiment of the present invention.

[0016] Figure 3The distribution diagram of wireless sensor network nodes in a unimodal terrain provided by an embodiment of the present invention;

[0017] Figure 4 The distribution diagram of wireless sensor network nodes in a multi - peak terrain provided by an embodiment of the present invention;

[0018] Figure 5 The trend diagram of the positioning error of the present invention and a comparative algorithm varying with the anchor nodes in a unimodal terrain provided by an embodiment of the present invention;

[0019] Figure 6 The trend diagram of the positioning error of the present invention and a comparative algorithm varying with the anchor nodes in a multi - peak terrain provided by an embodiment of the present invention. Detailed implementation manners

[0020] Next, in combination with the accompanying drawings in the embodiments of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.

[0021] First, the following explanations are given for the terms that may be used in this article:

[0022] The description of terms such as "including", "comprising", "containing", "having" or other similar semantics should be interpreted as non - exclusive inclusion. For example, including a certain technical feature element (such as raw materials, components, ingredients, carriers, dosage forms, materials, dimensions, parts, components, mechanisms, devices, steps, processes, methods, reaction conditions, processing conditions, parameters, algorithms, signals, data, products or articles, etc.) should be interpreted as not only including the clearly listed certain technical feature element, but also including other well - known technical feature elements in the art that are not clearly listed.

[0023] The term "consisting of" means excluding any technical feature element that is not clearly listed. If this term is used in a claim, then this term will make the claim a closed - type claim, making it not contain technical feature elements other than the clearly listed technical feature elements, except for related conventional impurities. If this term only appears in a certain sub - clause of a claim, then it only limits the elements clearly listed in that sub - clause, and the elements recorded in other sub - clauses are not excluded from the overall claim.

[0024] The following provides a detailed description of a DV-hop positioning method optimized by a compact compression improved false wild oat algorithm provided by the present invention. The content not described in detail in the embodiments of the present invention belongs to the prior art well-known to those skilled in the art. In the embodiments of the present invention, those not specified in specific conditions are carried out according to the conventional conditions in the art or the conditions recommended by the manufacturer.

[0025] As Figure 1 shown, a DV-hop positioning method optimized by a compact compression improved false wild oat algorithm mainly includes the following steps:

[0026] Step 1: Each anchor node transmits its own position information as broadcast information to the surrounding nodes. The nodes receive the broadcast information and record the minimum hop count from each anchor node to itself. Each of the nodes includes a position node and an anchor node.

[0027] In this step, after receiving the broadcast information, the node determines whether it receives the broadcast information for the first time according to the position information of the anchor node in the broadcast information. If so, it records the hop count in the broadcast information, then adds 1 to the hop count and forwards it outward; if not, it judges the size of the hop count recorded by itself and the hop count in the broadcast information, records the smaller hop count, then adds 1 to the hop count and forwards it outward; finally, each node records the minimum hop count from each anchor node to itself.

[0028] Step 2: Calculate the average hop distance according to the distance between the anchor nodes and the minimum hop count, and then combine the minimum hop count from each anchor node to each unknown node to calculate the distance between each anchor node and each unknown node.

[0029] Step 3: Combine the distance between each anchor node and each unknown node, and establish an optimization model for wireless sensor network node positioning in a three-dimensional space based on DV-Hop with the goal of the lowest positioning error.

[0030] In this step, the objective function of the optimization model for wireless sensor network node positioning in a three-dimensional space based on DV-Hop is expressed as:

[0031]

[0032] The constraint conditions are:

[0033] x u ≤κ1,y u ≤κ2,z u ≤κ3

[0034] Among them, n represents the total number of anchor nodes in the network, is the objective function of the optimization model for wireless sensor network node positioning in a three-dimensional space based on DV-Hop, represents the three-dimensional coordinates of the optimized unknown node u, (xi , y i , z i ), which are the three-dimensional coordinates of the anchor node i, and d iu is the distance between the anchor node i and the unknown node u; κ1, κ2, and κ3 are the set maximum values of the x, y, and z axes, and the three maximum values together form the range of the wireless sensor network.

[0035] Step 4: Optimize the node localization optimization model of the wireless sensor network in three-dimensional space based on DV-Hop by using the compact compression type false wild oat algorithm improved by the Gaussian probability model. Among them, initialize the false wild oat individuals by using the iteratively updated Gaussian model, substitute each false wild oat individual into the node localization optimization model of the wireless sensor network in three-dimensional space based on DV-Hop, perform fitness evaluation, update the positions of the false wild oat individuals through the false wild oat algorithm, and iteratively update the Gaussian model. Finally, output the final positions of each false wild oat individual, that is, the coordinates of the unknown nodes.

[0036] The above solution provided by the embodiment of the present invention introduces the least squares method of the false wild oat algorithm for position estimation, and optimizes and compresses the false wild oat algorithm through the Gaussian probability model, thereby improving the DV-Hop positioning. This not only improves the positioning accuracy, but also helps to obtain the position estimation result faster, and at the same time reduces the demand for memory resources.

[0037] In order to more clearly show the technical solution provided by the present invention and the technical effects produced, the method provided by the embodiment of the present invention will be described in detail below with specific embodiments Figure 2 to present the main process of the method.

[0038] 1. Establish a wireless sensor network.

[0039] In the embodiment of the present invention, a three-dimensional space with specified length, width, and height and corresponding unimodal and multimodal terrains are generated, and a wireless sensor network with n nodes (for example, n = 200) is randomly generated in the two terrains.

[0040] In the embodiment of the present invention, the length, width, and height are κ1, κ2, and κ3 defined above. As an example, κ1 = κ2 = κ3 = 100 can be set.

[0041] As an example, the number of anchor nodes in the network can be set to 5, 10, 15, 20, 25, 30, and the communication radius of all nodes is 35m to test the performance of the present invention.

[0042] 2. Record the minimum number of hops from each node to each anchor node.

[0043] In this embodiment, the anchor nodes with known positions broadcast their position information to the surrounding neighboring nodes. The neighboring nodes receive the broadcast information and record the minimum number of hops from them to the anchor nodes until the broadcast of the entire network is completed. Specifically: The number of hops for each node to receive the broadcast information can be recorded by the Dijkstra method. For each node in the WSN network, if it receives the broadcast information for the first time, it directly records and increments the hop count by 1, and then continues to forward. If it has already received the broadcast information from this node, it makes a comparison, retains the information with the smaller hop value, increments the hop count by 1, and then continues to forward. In this way, the minimum number of hops from all nodes in the network to each anchor node can be calculated.

[0044] III. Calculate the distance between each unknown node and each anchor node.

[0045] In this embodiment, based on the corresponding distances between the known anchor nodes and the minimum number of hops between the known anchor nodes, the average hop distance is calculated. For different anchor nodes i and j, their position information is known information, and using three-dimensional coordinates, they are represented as (x i , y i , z i ), (x j , y j , z j ). The average hop distance hopSize between anchor node i and other anchor node j is calculated by the following formula i :

[0046]

[0047] where n represents the total number of anchor nodes in the network, and h ij is the minimum number of hops from anchor node i to other anchor node j.

[0048] Combined with the minimum number of hops from each anchor node to each unknown node, the distance between each anchor node and each unknown node is calculated; for anchor node i, its distance from unknown node u is calculated by the following formula:

[0049] d iu = HopSize i × h iu

[0050] where d iu is the distance between anchor node i and unknown node u, and h iu is the minimum number of hops from anchor node i to unknown node u.

[0051] IV. Establish an optimization model for node localization in a three-dimensional wireless sensor network based on DV-Hop with the goal of minimizing the positioning error.

[0052] In the embodiments of the present invention, the objective function of the optimized model for node localization in a three-dimensional space based on DV-Hop is expressed as:

[0053]

[0054] The constraint conditions are:

[0055] x u ≤κ1, y u ≤κ2, z u ≤κ3

[0056] Wherein, is the objective function of the optimized model for node localization in a three-dimensional space based on DV-Hop, represents the three-dimensional coordinates of the optimized unknown node u, (x i , y i , z i ) are the three-dimensional coordinates of the anchor node i, and d iu is the distance between the anchor node i and the unknown node u; κ1, κ2, and κ3 are the set maximum values of the x, y, and z axes, and the three maximum values together form the range of the wireless sensor network. For specific details, please refer to the description in the first part above.

[0057] V. Optimize the optimized model for node localization in a three-dimensional space based on DV-Hop through the compact compression type false oat algorithm improved by the Gaussian probability model, and solve the coordinates of the unknown nodes.

[0058] First, introduce the false oat algorithm; then, introduce the compact compression type false oat algorithm improved by the Gaussian probability model.

[0059] 1. False oat algorithm.

[0060] In the embodiments of the present invention, false oat (scientific name: Avena sterilis L.) is an annual herbaceous plant of the Gramineae family and the genus Avena. It is usually between 30 and 120 centimeters in height. It has slender leaves and slender spike-like inflorescences, and the spike-like inflorescences contain multiple spikelets. False oat is a weed with extremely strong vitality. Its reproduction speed is about 4-10 times that of wheat, and its seeds have the ability of long-term dormancy and can survive even in adverse environments, bringing great difficulties to prevention and control. This highly competitive weed competes with various crops for nutrients and water resources in the soil, ultimately resulting in a decrease in crop yields. When the seeds of false oat fall off from the mother plant, they will exhibit special movement patterns such as rolling and jumping. This unique movement mechanism is one of the reasons for its faster spread speed compared to other plants.

[0061] In the embodiments of the present invention, the false wild oat algorithm mainly targets the seed dissemination stage. The false wild oat seeds are approximately equivalent to the false wild oat individuals. The present invention involves some nouns related to the structures or characteristics of false wild oat seeds. For example, the main awn of the false wild oat is the main structure on which the rotational movement of the false wild oat seed depends. In the subsequent development stage, parameters such as the length of the main awn of the false wild oat will be used to calculate the position offset.

[0062] Randomly initialize several false wild oat individuals in the solution space to form an original population (not yet updated and iterated). The false wild oat individuals represent the solutions to the problem. In the embodiments of the present invention, a set of coordinates in the three-dimensional space corresponding to the problem, that is, the coordinates of the unknown node to be optimized. It is necessary to select the superior and eliminate the inferior according to the fitness value, update the population according to the algorithm steps, and obtain the optimal solution.

[0063] In the embodiments of the present invention, the false wild oat algorithm includes an initialization stage, an exploration stage, and a development stage; in each iteration, the exploration stage or the development stage is randomly selected with the same probability to update the positions of the false wild oat individuals; if the exploration stage is selected, the positions of the false wild oat individuals are directly updated through the exploration stage; if the development stage is selected, the relevant parameters are first calculated through the initialization stage, and then the positions of the false wild oat individuals are updated by using the development stage; the methods for each stage are introduced below.

[0064] (1) Initialization stage:

[0065]

[0066]

[0067] e′ = r × λ

[0068]

[0069] where index is the index of the false wild oat individual ranked according to the fitness, is the objective function of the DV-Hop-based three-dimensional wireless sensor network node positioning optimization model, rank is the ranking function; L is the length of the main awn of the false wild oat, dim is the problem dimension (it is the basis for the form of the solution. In the present invention, the coordinates of the unknown node are required, so it is a three-dimensional problem and its dimension is 3), e′ is the eccentric rotation coefficient of the false wild oat in the development stage, r is a random number in the interval [0, 1], m is the mass of the false wild oat individual, t is the current iteration number, iter_max is the maximum iteration number, and λ is a constant coefficient usually taken as 0.5.

[0070] (2) Exploration stage:

[0071] J = (2 × r - 1) × best x(g)- best x

[0072] X t+1 = best x + J

[0073] where J is the position offset caused by the false wild oat's energy storage and jumping, and best x(g) is the best position in a certain population, and best x is the global best position, and X t+1 is the position of the i-th false wild oat individual in the (t + 1)-th generation.

[0074] (3) In the development stage, calculate the sensitive parameters by combining the corresponding trigonometric functions, then calculate the corresponding position offset by combining the sensitive parameters, and then update the position of the false wild oat individual by combining the position offset.

[0075] In the embodiment of the present invention, in the development stage: it includes Scheme 1 and Scheme 2, and randomly selects one of them with equal probability.

[0076] Scheme 1:

[0077]

[0078]

[0079] X t+1 = best x + R

[0080] where a is the sensitive parameter, ub is the upper limit of the solution space, sin is the sine function, π is the symbol of pi, iter_max is the maximum number of iterations, R is the position offset caused by the rotational motion, unifrnd(-a, a) is a random number in the interval [-a, a], and X t+1 (i) is the position of the i-th false wild oat individual in the (t + 1)-th generation, and best x is the global best position.

[0081] Scheme 2:

[0082]

[0083]

[0084] X t+1 = best x + W

[0085] where b is a sensitive parameter, ub is the upper bound of the solution space, cos is the cosine function, π is the symbol of pi, iter_max is the maximum number of iterations, unifrnd(-b, b) is a random number in the interval [-b, b], W is the position offset caused by the natural environment, and X t+1 (i) is the position of the i-th false wild oat individual in the (t + 1)-th generation, and best x is the global best position.

[0086] 2. The compact compression type false wild oat algorithm improved by the Gaussian probability model.

[0087] In the initial stage, divide into multiple subpopulations and initialize the corresponding iteratively updated Gaussian probability model for each subpopulation; for each subpopulation: at the beginning of each round of iteration, use the corresponding iteratively updated Gaussian probability model to generate false wild oat individuals and conduct fitness evaluation, then update the positions of each false wild oat individual through the false wild oat algorithm; evaluate the fitness again, and determine the winner and loser respectively based on the fitness before and after the position update. Finally, update the Gaussian probability model according to the winner and loser and enter the next iteration. Keep repeating the iteration until the termination condition is met (i.e., the maximum number of iterations is reached). After the iteration is completed, output the false wild oat individual with the minimum fitness (optimal position) as the coordinates of an unknown node. The preferred implementation of the above process is as follows:

[0088] First, divide the original population into G subpopulations, where G is a positive integer. Each subpopulation performs iterative operations independently. After each subpopulation completes the iteration, the coordinates of an unknown node are obtained. By dividing the subpopulations and performing iterative operations as above, the accuracy of coordinate solution can be improved to a certain extent.

[0089] The iterative operation process of each subpopulation is the same. Here, the subpopulation g is taken as an example for introduction.

[0090] In the iterable Gaussian probability model, a perturbation vector is used to describe the distribution of false wild oat individuals in subpopulation g. The representation of the perturbation vector is as follows:

[0091]

[0092] where PV is the perturbation vector, and respectively represent the mean and standard deviation of the distribution of false wild oat individuals in subpopulation g at the t-th iteration (i.e., the current iteration), and have the corresponding probability density function (PDF). After normalizing the generated probability density function, false wild oat individuals are generated from it. The probability density function is constructed into a Chebyshev polynomial to obtain a cumulative distribution function (CDF) with values in the interval [0, 1]. The cumulative distribution function is as follows:

[0093]

[0094] Among them, CDF is the cumulative distribution function, PDF is the probability density function, the value range of x is from -1 to 1, d is the differential symbol, erf is the error function, and e is the natural constant.

[0095] It can be seen that the PDF belongs to a truncated Gaussian distribution and can be transformed into the following form:

[0096]

[0097] When the algorithm is executed, the following steps are involved: First, a random number r with a value range from 0 to 1 is generated through a uniform distribution; then, the generated random number is substituted into the inverse function of the cumulative distribution function corresponding to the perturbation vector to generate a solution y with a value range from -1 to 1; the inverse function of the cumulative distribution function is expressed as follows:

[0098]

[0099] Among them, erf -1 is the inverse function of erf, and the solution y from -1 to 1 is the false wild oat individual.

[0100] The false wild oat individuals generated by the above method are in the range from -1 to 1, and they are mapped to the actual solution space through the following formula:

[0101]

[0102] Among them, ub and lb are the upper and lower limits of the actual solution space, and y′ is the finally generated false wild oat individual.

[0103] According to the above process, multiple false wild oat individuals in the sub-population g in the t-th iteration are generated. For each false wild oat individual, fitness evaluation, position update, and fitness evaluation are performed again in sequence. The false wild oat individual with better fitness before and after position update becomes the winner, while the one with worse fitness becomes the loser. Taking the false wild oat individual y′ as an example, it is denoted as y″ after position update. If the fitness of the false wild oat individual y′ is better than the fitness y″ of the false wild oat individual y′, then the false wild oat individual y′ is the winner and y″ is the loser, and vice versa.

[0104] After that, the corresponding mean and standard deviation are updated according to the winner and the loser.

[0105] The update expression of the mean is as follows:

[0106]

[0107] Among them, represents the mean of the distribution of the false wild oat individuals in the sub-population g in the (t + 1)-th generation, N virtual is the size of the virtual population, which is a set parameter. For example, N virtual can take values within [20, 200]; winner g and loser g correspondingly represent the winner and loser in the sub-population g.

[0108] The update expression of the standard deviation is as follows:

[0109]

[0110] where represents the standard deviation of the distribution of the false wild oat individuals in the sub-population g in the (t + 1)-th generation.

[0111] After the above steps, the compact compression type false wild oat algorithm improved by the Gaussian probability model completes one iteration. The above process is repeated in a loop until the maximum number of iterations is reached, and the algorithm execution ends. During the algorithm execution, the false wild oat individual with the minimum fitness so far is always recorded. After the algorithm execution ends, the false wild oat individual (optimal position) with the minimum fitness is output, which is the coordinate of an unknown node.

[0112] In the embodiment of the present invention, the total number of executions of the algorithm and the input are determined according to the number of unknown nodes and the corresponding topological structure information. Each sub-population independently and parallelly executes the compact compression type false wild oat algorithm, and finally the coordinates of all unknown nodes are obtained.

[0113] During the execution of the compact compression type false wild oat algorithm above, the Gaussian probability models of each sub-population are independent. However, the initialized Gaussian probability models at the beginning stage are the same. For example, the initial values of the mean μ and the standard deviation σ can be 0 and 10.

[0114] In the above solution of the embodiment of the present invention, the compact compression type false wild oat algorithm improved by the Gaussian probability model replaces the distribution information of the actual population with the Gaussian probability model and reduces the number of false wild oat individuals in the actual population. At the same time, the population is divided into sub-populations, and a communication mechanism is set. When a certain number of iterations is reached, two sub-populations are randomly selected to communicate with each other, and the worse false wild oat individuals are replaced according to the principle of "survival of the fittest".

[0115] In this embodiment, the possible solutions represented by each false wild oat individual are divided into three groups according to the population diversity. The positions of the possible solutions within each group are close, and the differences between the possible solutions between groups are large. For example, it can be set that: the maximum number of iterations is 100, the problem dimension is 3D, and the problem boundary is [0, κ]. Here, the problem boundary refers to the value range of the solution, and here κ ∈ κ1, κ2, κ3.

[0116] To illustrate the performance of the above solution of the present invention, after outputting the position coordinates of unknown nodes, a relevant error analysis is carried out with the actual position coordinates. error u Quantifies the accumulation of the error between the true coordinates of each unknown node and the coordinates of the unknown nodes optimized by the algorithm, as follows:

[0117]

[0118] Where: (x u , y u , z u ) represents the actual three-dimensional coordinates of the unknown node u, represents the three-dimensional coordinates of the unknown node u optimized by the algorithm, and U is the number of unknown nodes.

[0119] Figures 3 to 4 Respectively give the node distribution diagrams of wireless sensor networks in unimodal terrain and multimodal terrain. Figures 4 to 5 Respectively give the trend diagrams of the positioning errors of the present invention and the comparative algorithms in unimodal terrain and multimodal terrain. The comparative algorithms include: Vector Hop positioning algorithm (3DDV-Hop), Particle Swarm Optimization algorithm based on Distance Vector Hop (PSO-3DDV-Hop), and Willow Catkin Optimization algorithm based on Distance Vector Hop (WCO-3DDV-Hop).

[0120] Through the description of the above embodiments, those skilled in the art can clearly understand that the above embodiments can be implemented by software or by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solutions of the above embodiments can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.), including several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in various embodiments of the present invention.

[0121] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by those skilled in the art within the technical scope disclosed by the present invention should be covered by the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims

1. A compact compression improved DV-hop positioning method optimized by the false wild oat algorithm, characterized in that: include: Each anchor node transmits its own location information as broadcast information to surrounding nodes. The nodes receive the broadcast information and record the minimum number of hops from each anchor node to itself. Each of the nodes includes a location node and an anchor node; The average hop distance is calculated based on the distance between anchor nodes and the minimum number of hops, and then the distance between each anchor node and each unknown node is calculated by combining the minimum number of hops from each anchor node to each unknown node; Combined with the distance between each anchor node and each unknown node, a DV-Hop-based wireless sensor network node positioning optimization model in three-dimensional space is established with the goal of minimizing positioning error. The compact compressed false oat algorithm improved by Gaussian probability model is used to optimize the node positioning optimization model of wireless sensor network in three-dimensional space based on DV-Hop. The false oat individuals are initialized by using the iteratively updated Gaussian model, and each false oat individual is substituted into the wireless sensor network node positioning optimization model in three-dimensional space based on DV-Hop for fitness evaluation. The position of the false oat individual is updated by the false oat algorithm, and the Gaussian model is iteratively updated to finally output the final position of each false oat individual, that is, the coordinates of the unknown node.

2. The DV-hop positioning method optimized by the compact compression improved false wild oat algorithm according to claim 1 is characterized in that: Each anchor node transmits its own location information as broadcast information to surrounding nodes. The nodes receive the broadcast information and record the minimum number of hops from each anchor node to itself, including: After receiving the broadcast information, the node determines whether it is the first time to receive the broadcast information based on the location information of the anchor node in the broadcast information. If so, it records the number of hops in the broadcast information, then increases the number of hops by 1 and forwards it outward; if not, it determines the size of the number of hops recorded by itself and the number of hops in the broadcast information, records the smaller number of hops, then increases the number of hops by 1 and forwards it outward; finally, each node will record the minimum number of hops from each anchor node to itself.

3. The DV-hop positioning method optimized by the compact compression improved false wild oat algorithm according to claim 1 is characterized in that: The method of calculating the average hop distance according to the distance between the anchor nodes and the minimum hop number, and then calculating the distance between each anchor node and each unknown node in combination with the minimum hop number from each anchor node to each unknown node includes: According to the corresponding distance between anchor nodes and the minimum number of hops between anchor nodes, the average hop distance is calculated; for different anchor nodes i, j, their location information is known information, and the corresponding three-dimensional coordinates are expressed as (x i ,y i , z i ), (x j ,y j , z j ), the average hop distance HopSize between anchor node i and anchor node j is calculated by the following formula: i : Where n represents the total number of anchor nodes in the network, h ij is the minimum number of hops from anchor node i to anchor node j; Combined with the minimum number of hops from each anchor node to each unknown node, the distance between each anchor node and each unknown node is calculated; for anchor node i, its distance to unknown node u is calculated by the following formula: d iu =HopSize i ×h iu Among them, d iu is the distance between anchor node i and unknown node u, h iu is the minimum number of hops from anchor node i to unknown node u.

4. A compact compression improved DV-hop positioning method optimized by the false wild oat algorithm according to claim 1, characterized in that: The objective function of the wireless sensor network node positioning optimization model in three-dimensional space based on DV-Hop is expressed as: The constraints are: x u ≤κ1,y u ≤κ2,z u ≤κ3 Where n represents the total number of anchor nodes in the network. is the objective function of the DV-Hop-based wireless sensor network node positioning optimization model in three-dimensional space. represents the three-dimensional coordinates of the optimized unknown node u, (x i ,y i , z i ) is the three-dimensional coordinate of anchor node i, d iu is the distance between anchor node i and unknown node u, κ1, κ2 and κ3 are the set maximum values ​​of x, y and z axes.

5. A compact compression improved false wild oat algorithm optimized DV-hop positioning method according to claim 1, characterized in that: The false wild oats algorithm includes an initialization phase, an exploration phase, and a development phase; each iteration randomly selects the exploration phase or the development phase with the same probability to update the individual positions of the false wild oats; if the exploration phase is selected, the individual positions of the false wild oats are directly updated through the exploration phase; If the development phase is selected, the relevant parameters are calculated through the initialization phase, and then the individual positions of the false wild oats are updated by utilizing the development phase; Initialization phase: e′=r×λ Among them, index is the index of the individual ranking of false wild oats according to fitness, is the objective function of the wireless sensor network node positioning optimization model in three-dimensional space based on DV-Hop, rank is the ranking function; L is the main awn length of false wild oats, dim is the problem dimension, e′ is the eccentric rotation coefficient of false wild oats in the development stage, r is a random number in the interval [0, 1], m is the individual mass of false wild oats, t is the current iteration number, iter_max is the maximum iteration number, and λ is a constant coefficient; Exploration phase: J=(2×r-1)×best x(g) -best x X t+1 =best x +J Where J is the position offset caused by the false wild oats energy jump, best x(g) For the best position in a group, best x is the global optimal position, X t+1 is the position of the i-th individual of false wild oats in the t+1th generation; During the development phase, sensitive parameters are calculated using trigonometric functions, and then the corresponding position offsets are calculated using the sensitive parameters. The positions of the false wild oats individuals are then updated using the position offsets.

6. A compact compression improved false wild oat algorithm optimized DV-hop positioning method according to claim 5, characterized in that: The method of calculating the sensitive parameters by combining the trigonometric function, calculating the corresponding position offset by combining the sensitive parameters, and updating the position of the false wild oat individual by combining the position offset comprises: X t+1 =best x +R Among them, a is a sensitive parameter, ub is the upper limit of the solution space, sin is the sine function, π is the symbol of pi, iter_max is the maximum number of iterations, R is the position offset of the rotational motion, unifrnd(-a, a) is a random number in the interval [-a, a], X t+1 is the position of the i-th false wild oat individual in the t+1th generation, best x is the global best position.

7. A compact compression improved false wild oat algorithm optimized DV-hop positioning method according to claim 5, characterized in that: The method of calculating the sensitive parameters by combining the trigonometric function, calculating the corresponding position offset by combining the sensitive parameters, and updating the position of the false wild oat individual by combining the position offset comprises: X t+1 =best x +W Where b is a sensitive parameter, ub is the upper limit of the solution space, cos is the cosine function, π is the symbol of pi, iter_max is the maximum number of iterations, unifrnd(-b, b) is a random number in the interval [-b, b], W is the position offset caused by the natural environment, and X t+1 is the position of the i-th false wild oat individual in the t+1th generation, best x is the global best position.

8. A compact compression improved false wild oat algorithm optimized DV-hop positioning method according to claim 5, characterized in that: The compact compression type false wild oats algorithm improved by the Gaussian probability model includes: In the initial stage, multiple sub-populations are divided, and the corresponding iteratively updated Gaussian probability model is initialized for each sub-population; for each sub-population: for each sub-population: at the beginning of each round of iteration, the corresponding iteratively updated Gaussian probability model is used to generate false wild oats individuals, and the fitness is evaluated, and then the position of each false wild oats individual is updated by the false wild oats algorithm; the fitness is evaluated again, and the winner and loser are determined by the fitness before and after the position update, and finally the Gaussian probability model is updated according to the winner and loser, and the next iteration is entered. The iteration is repeated until the termination condition is met. After the iteration is completed, the false wild oats individual with the minimum fitness is output as the coordinate of an unknown node.

9. A compact compression improved false wild oat algorithm optimized DV-hop positioning method according to claim 8, characterized in that: The original population is divided into G sub-populations, where G is a positive integer. Each sub-population performs iterative operations independently, and the iterative operations of different sub-populations have the same process; For subpopulation g: In the iteratively updated Gaussian probability model, a disturbance vector is used to describe the distribution of individuals of false wild oats in subpopulation g. The representation of the disturbance vector is: Where PV is the disturbance vector, and They represent the mean and standard deviation of the distribution of false wild oats individuals in the t-th iteration subpopulation g, and have corresponding probability density functions. After normalizing the generated probability density function, false wild oats individuals are generated from it, and the probability density function is constructed into a Chebyshev polynomial to obtain a cumulative distribution function with a value in the interval [0, 1]. The cumulative distribution function is as follows: Among them, CDF is the cumulative distribution function, PDF is the probability density function, the value range of x is -1 to 1, d is the differential sign, erf is the error function, and e is a natural constant; Generate a random number r ranging from 0 to 1 through uniform distribution, and substitute the generated random number into the inverse function of the cumulative distribution function corresponding to the perturbation vector to generate a solution y ranging from -1 to 1; the inverse function of the cumulative distribution function is expressed as follows: Among them, erf -1 It is the inverse function of erf, and the solution y from -1 to 1 is the individual of false wild oats; The false wild oats individuals generated using the above method are in the range of -1 to 1, and they are mapped to the actual solution space, which is achieved by the following formula: Among them, ub and lb are the upper and lower limits of the actual solution space, and y′ is the final generated false wild oat individual; According to the above process, multiple false wild oats individuals in the sub-population g in the tth iteration are generated, and the false wild oats individuals are evaluated for fitness, updated in position, and evaluated for fitness again in turn. The false wild oats individuals with better fitness before and after the position update become the winners, and the ones with poorer fitness become the losers.

10. A compact compression improved false wild oat algorithm optimized DV-hop positioning method according to claim 9, characterized in that: Updating the Gaussian probability model according to the winners and losers includes: updating the mean and standard deviation of the disturbance vector; The update expression of the mean is as follows: in, represents the mean value of the distribution of individuals of the subpopulation g in the t+1th generation, N virtual is the size of the virtual population; winner g and loser g Correspondingly, they represent the winners and losers in the subpopulation g; The updated expression of the standard deviation is as follows: in, represents the standard deviation of the distribution of individuals of the subpopulation g of wild oats in the t+1th generation.

Images