Cross-domain sensor resource scheduling method and system based on adaptive particle swarm

Abstract

The invention belongs to the field of sensor scheduling, and discloses a cross-domain sensor resource scheduling method and system based on an adaptive particle swarm, and the method comprises the steps: carrying out the initialization processing of a particle swarm; updating the initialized particle swarm in combination with a bee colony algorithm and a particle updating algorithm, updating the initialized particle swarm in a leading bee neighborhood search stage by adopting the particle updating algorithm to generate updated particles, and calculating to obtain target fitness corresponding to each updated particle; in the following bee following search stage, updating the updated particle swarm again by adopting a roulette algorithm based on the target fitness corresponding to each updated particle to obtain an updating result; the processes corresponding to the leading bee neighborhood search stage and the following bee following search stage are repeatedly executed until the maximum number of iterations is reached, and the current updating result serves as the optimal scheduling scheme of the sensor to be output; by adopting the method, the detection precision is improved, and the radiation exposure risk and the sensor task switching frequency are reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of sensor scheduling, and in particular relates to a cross-domain sensor resource scheduling method and system based on adaptive particle swarm. Background Art

[0002] In the confrontation of modern information warfare, occupying the "information high ground" and effectively exerting information advantages are the keys to victory. Sensors can be described as "clairvoyant eyes" and "superb ears" in the confrontation environment, and they play a vital role in obtaining information in the confrontation environment. However, in the confrontation environment, due to the presence of a large number of targets, the influence of the environment itself, and possible interference, the sensors will not work properly or will be destroyed and unable to continue to perform tasks. Single-domain sensors with very limited energy and performance cannot meet the monitoring and detection needs in the above complex confrontation environment, so multi-domain sensors need to be used to coordinate and cooperate to meet the above needs. However, the use of multi-sensor systems for detection in complex confrontation environments requires consideration of how to schedule resources for multi-domain sensor systems with limited resources.

[0003] Existing research on cross-domain sensor resource scheduling mainly includes linear programming methods, information theory-based methods, artificial intelligence-based methods, heuristic algorithm-based methods, etc. However, most of the existing research work focuses on the sensor scheduling problem in a single domain, or only considers the target detection problem, or only considers the target tracking problem, or considers these two problems separately, and almost none of them considers the multi-domain coordination problem. Although information theory methods can obtain good technical indicators, they cannot be well combined with actual combat needs, and often have problems such as high computational complexity and low timeliness. The heuristic algorithm itself may have problems such as slow convergence speed and low accuracy. In addition, there are few studies that use heuristic algorithms to solve the multi-domain coordination sensor resource scheduling problem.

[0004] It can be seen that most of the existing sensor scheduling studies focus on sensor scheduling in a single domain. For cross-domain sensor resource scheduling, the existing scheduling methods have problems such as insufficient global optimization capabilities, high switching costs, and high radiation exposure risks. Summary of the invention

[0005] The purpose of the present invention is to provide a cross-domain sensor resource scheduling method and system based on adaptive particle swarm, so as to solve the problems of insufficient global optimization capability, high switching cost and high radiation exposure risk in the existing scheduling methods.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: In a first aspect, the present invention provides a cross-domain sensor resource scheduling method based on adaptive particle swarm, comprising: Initializing the particle swarm, wherein the effective multi-domain sensor scheduling sequence is used as the initial particle of the particle swarm; The initialized particle swarm is updated by combining the bee swarm algorithm and the particle update algorithm. The specific steps are as follows: In the leader bee neighborhood search phase, the particle update algorithm is used to update the initialized particle swarm, generate updated particles, and calculate the target fitness corresponding to each updated particle; In the follower bee search phase, based on the target fitness corresponding to each updated particle, the roulette algorithm is used to update the updated particle group again, obtain the updated result and recalculate the target fitness corresponding to the updated particle; Repeat the corresponding processes of the leading bee neighborhood search phase and the following bee follow-up search phase until the maximum number of iterations is reached, and output the current update result as the optimal scheduling solution for the sensor.

[0007] Furthermore, the initialization process of the particle swarm specifically includes: Initialize the population size, target number, maximum number of iterations, and number of sensors corresponding to the particle swarm; For each target visual sensor set, a generation algorithm is used to generate an effective multi-domain sensor scheduling sequence as the initial particles of the particle swarm, where each dimension is different; For the initial particles of the particle swarm, a greedy strategy is used to calculate the initial particle fitness, initial individual optimal solution, initial global optimal solution, and the selection probability of each dimension of the particle for the current solution, individual optimal solution, global optimal solution and other optional solutions; Among them, the multi-domain sensors include space-based sensors, air-based sensors, sea-based sensors and land-based sensors.

[0008] Furthermore, in the leader bee neighborhood search phase, the specific steps of using the particle update algorithm to update the initialized particle swarm are as follows: Based on the selection probabilities corresponding to the current solution, individual optimal solution, global optimal solution and other optional value solutions, the roulette algorithm is used to update the initialized particle swarm and generate updated particles; Calculate the target fitness corresponding to the updated particle, as well as the selection probabilities corresponding to the current solution, individual optimal solution, global optimal solution and other optional value solutions of the updated particle.

[0009] Furthermore, the specific steps of updating the initialized particle swarm using the roulette algorithm include: The targets are grouped according to the distance between them, and the groups are sorted from large to small according to the average threat level within the group; the order of the two targets in the group is processed by enumeration method, and finally all possible target allocation sequences are obtained; For each target allocation sequence, sensors are selected in order according to the probability size, where: When any sensor is selected by the current dimension target, this sensor will be removed from the optional sensor sequence of the target in the following dimension; When two targets select the sensor with the same serial number, the probability sorting principle is adopted to select the particle corresponding component with a large probability and not repeated with the previously generated component for update; When updating to the kth dimension, first compare whether the sensor corresponding to the kth dimension of the current solution, the individual optimal solution, and the global optimal solution has been selected by the first k-1 dimensions of the new particle sequence; If the k-th dimension of the current solution already exists in the new particle sequence, the corresponding P1 marker is 0; if the k-th dimension of the individual optimal solution already exists in the new particle sequence, the corresponding P2 marker is 0; if the k-th dimension of the global optimal solution already exists in the new particle sequence, the corresponding P3 marker is 0; where P1, P2, P3, and P4 represent the selection probabilities corresponding to the current solution, the individual optimal solution, the global optimal solution, and other optional values, respectively; Compare the maximum values ​​of P1~P3 whose mark bit is 1, and record the maximum value; The roulette algorithm and target update sequence are used to update the particles in order. If the value of the newly generated dimension does not repeat the previously generated component, the update proceeds to the next dimension; otherwise, the solution component corresponding to the maximum value with a mark bit of 1 in P1~P3 is selected for update.

[0010] Furthermore, the specific calculation formulas of P1, P2, P3, and P4 are as follows:

[0011] In the formula, Cycle represents the current number of iterations, MaxCycle represents the set maximum number of iterations; p ranges from 0 to 1; represents the fitness value of the current solution, represents the fitness value of the individual optimal solution, Represents the fitness value of the global optimal solution.

[0012] Furthermore, in the following bee following search phase, based on the target fitness corresponding to each updated particle, the specific steps of using the roulette algorithm to update the updated particle swarm again are as follows: Normalize the fitness of each updated particle; Based on the normalized fitness, the roulette algorithm is then used to update the updated particle swarm again, and the fitness of the particles after the update is recalculated, as well as the selection probabilities corresponding to the current solution, individual optimal solution, global optimal solution and other optional value solutions of the particles after the update; Based on the current solution of the particle after the re-update, the individual optimal solution, the global optimal solution and the selection probabilities corresponding to the remaining optional values, the individual optimal solution and the global optimal solution are updated in combination with the greedy strategy to obtain the updated result.

[0013] Furthermore, it also includes a stage of eliminating low-quality particles. After executing the follow-up bee search stage, it is determined whether the number of particle updates has reached the set elimination upper limit. If so, a generation algorithm is used to regenerate a new particle to replace the old particle; otherwise, proceed to the next step.

[0014] In a second aspect, the present invention provides a cross-domain sensor resource scheduling system based on adaptive particle swarm, comprising: A particle swarm initialization module is used to initialize the particle swarm, wherein an effective multi-domain sensor scheduling sequence is used as the initial particle of the particle swarm; The particle swarm update module is used to update the initialized particle swarm by combining the bee swarm algorithm and the particle update algorithm, which specifically includes: The leading bee neighborhood search module is used to update the initialized particle swarm using a particle update algorithm during the leading bee neighborhood search phase, generate updated particles, and calculate the target fitness corresponding to each updated particle; The follower bee following search module is used to update the updated particle group again based on the target fitness corresponding to each updated particle using the roulette algorithm in the follower bee following search stage, obtain the update result and recalculate the target fitness corresponding to the updated particle; The scheduling scheme output module is used to repeatedly execute the corresponding processes of the leading bee neighborhood search phase and the following bee following search phase until the maximum number of iterations is reached, and output the current update result as the optimal scheduling scheme for the sensor.

[0015] In a third aspect, the present invention provides a device, comprising: Memory for storing computer programs; The processor is used to implement the steps of the above-mentioned cross-domain sensor resource scheduling method based on adaptive particle swarm when executing the computer program.

[0016] In a fourth aspect, the present invention provides a computer-readable storage medium storing a computer program, which, when executed by a processor, is used to implement the steps of the above-mentioned cross-domain sensor resource scheduling method based on adaptive particle swarm.

[0017] Compared with the prior art, the present invention has the following beneficial effects: The present invention provides a cross-domain sensor resource scheduling method based on adaptive particle swarms. The method combines the global search capability of a particle swarm with the local optimization capability of a swarm. The method can first quickly converge to a better solution through a particle swarm, and then use a swarm algorithm to perform local refinement optimization on the solution, thereby overcoming the problem of being easily trapped in a local optimum when the swarm algorithm is used alone. The method can effectively improve the quality of the solution in large-scale sensor resource scheduling problems. The method enhances the details of the local search through the information exchange and cooperation mechanism of the swarm. In particular, in sensor resource scheduling, the allocation of tasks and the configuration of resources usually have time-varying and spatial distribution characteristics. Information sharing and collaborative search are helpful to find the optimal or near-optimal resource scheduling scheme. The method can comprehensively consider the detection accuracy, radiation exposure risk and the number of sensor task switching under multi-domain sensors and multi-target constraints. The proposed method is superior to the swarm algorithm and the particle swarm optimization algorithm in terms of switching cost and global fitness, and can effectively solve suboptimal solutions or near-optimal solutions to meet the detection needs of multi-sensor systems in complex confrontation environments.

[0018] In addition, this scheduling method can give full play to the advantages of both in multi-objective optimization; the particle swarm algorithm is responsible for quickly finding a solution that meets the basic constraints by updating the group information, and the bee colony algorithm is responsible for fine-tuning the target on this basis to find a more balanced compromise solution. The combined algorithm can provide a better compromise solution in the multi-objective sensor resource scheduling problem to meet the complex requirements between various tasks and resources.

[0019] Preferably, in the present invention, a stage of eliminating low-quality particles is also set. When the number of particle updates reaches the set elimination upper limit, a generation algorithm is used to regenerate a new particle to replace the old particle, which plays a role in widely detecting particles that have jumped out of the local optimal solution, and avoids premature algorithm maturation to a certain extent. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 A flowchart of a cross-domain sensor resource scheduling method based on adaptive particle swarm provided in an embodiment of the present invention; Figure 2 The scheduling method provided in the embodiment of the present invention corresponds to the sensor scheduling sequence diagram at time 750-850; Figure 3 A sensor scheduling sequence diagram of the standard bee colony algorithm at time 750-850 provided by an embodiment of the present invention; Figure 4 A sensor scheduling sequence diagram at time 750-850 of the standard particle swarm algorithm provided by an embodiment of the present invention; Figure 5 A simulation diagram showing the fitness of the three algorithms at 100 moments as a function of the number of iterations provided by the embodiment of the present invention; Figure 6 A simulation diagram showing the fitness of the three algorithms at time 700 as a function of the number of iterations provided by an embodiment of the present invention; Figure 7 A simulation diagram showing the fitness of the three algorithms at time 850 as a function of the number of iterations provided by an embodiment of the present invention; Figure 8 A simulation diagram showing the fitness of the three algorithms at time 1000 as a function of the number of iterations provided by the embodiment of the present invention; Fig. 9 A flowchart of a cross-domain sensor resource scheduling method based on adaptive particle swarm provided by the present invention; Fig.10 A schematic structural diagram of a cross-domain sensor resource scheduling system based on adaptive particle swarm provided by the present invention. DETAILED DESCRIPTION

[0021] Example 1 As described in the background technology, most of the existing research focuses on the sensor scheduling problem in a single domain, or only considers the target detection problem, or only considers the target tracking problem, or considers these two problems separately, and almost no multi-domain coordination problem is considered. Although the information theory method can obtain good technical indicators, it cannot be well combined with the actual combat needs, and often has problems such as high computational complexity and low timeliness. The heuristic algorithm itself may have problems such as slow convergence speed and low accuracy. In addition, there are few studies that use heuristic algorithms to solve the multi-domain coordinated sensor resource scheduling problem.

[0022] In order to solve the above problems, the present invention provides a cross-domain sensor resource scheduling method based on adaptive particle swarm. The scheduling method can effectively solve the problems of single-domain sensors, slow convergence speed and easy falling into local optimality existing in the prior art.

[0023] The present embodiment is further explained and illustrated below in conjunction with the accompanying drawings: like Figure 1 As shown, this embodiment provides a cross-domain sensor resource scheduling method based on adaptive particle swarm, which specifically includes: S1. Initialize the particle swarm. In the parameter initialization stage, it is necessary to determine the size of the swarm, the number of targets, the maximum number of iterations, the upper limit of the number of leader bee eliminations, the number of sensors and other parameters.

[0024] S101. In this embodiment, the population size of the particle swarm is set to 60, the number of targets is 10, the maximum number of iterations is 100, the upper limit of the elimination times is 40, and the number of sensors is the total number of sensors in the four domains of space, air, sea and land. Among them, the number of space-based sensors is set to 6, the number of space-based sensors is 18, the number of sea-based sensors and land-based sensors are both 6, and the total number of cross-domain sensors is 36.

[0025] S2. In the particle initialization stage, according to the visible sensor set of each target, a generation algorithm is used to generate a sensor sequence that is different in each dimension, that is, the initial particle. The fitness of the initial particle is calculated according to the objective function, and the individual optimal solution, the global optimal solution, and P1 to P4 are initialized.

[0026] S201. In this embodiment, each dimension component of the particle is set to the sensor number visible to the target. Since the number of targets is 10, the dimension of each particle is 10. This particle coding method directly represents the sensor scheduling scheme, avoiding the calculation of invalid particles and facilitating the faster finding of better particles.

[0027] S202. The implementation process of the particle initialization algorithm is introduced below. First, each dimension component of the particle is randomly selected within the optional value range. Since the visible sensor numbers of two different targets may be the same, when a certain sensor is selected, the subsequent dimension optional sensor sequence will exclude this sensor, that is, this sensor will no longer be selected, which ensures that the selected sensor numbers in each dimension are not repeated. If the elements in the generated particle are not repeated, the particle can be considered legal.

[0028] S203. Specifically, the individual optimal solution is the optimal solution encountered by each particle during the iteration process, which is randomly generated according to the particle initialization algorithm at the beginning and continuously updated during the iteration process. For the global optimal solution, it refers to the optimal solution searched by all particles during the update process. At the beginning, the fitness of each initial particle is calculated, and the greedy strategy is used to obtain the current global optimal solution, which will also be continuously updated during the iteration process. P1 to P4 are calculated using the following formula, where Cycle represents the current iteration number, MaxCycle is the set maximum iteration number, P4 is a value that continuously decreases with the increase of the iteration number, and 0 < p < 1 is set; this is to ensure that the particles still have a certain ability to jump out of the local optimal solution in the later stage of the iteration. P1 to P3 are related to the fitness of the particle, where, represents the fitness value of the current solution, represents the fitness value of the individual optimal solution, represents the fitness value of the global optimal solution. Since in this embodiment, the smaller the fitness value, the better the corresponding sensor scheduling scheme, the reciprocal of the fitness value is taken during the calculation, that is, the smaller the fitness value of the sensor scheduling scheme, the larger the corresponding P value..

[0029]

[0030] S3. In the leading bee field search phase, it is necessary to use the particle update algorithm to generate updated particles based on the visual sensor set of each target.

[0031] S301. The implementation process of the particle update algorithm will be introduced in detail below. In the update process, there are four optional values ​​for the particle's i-th dimension, namely, the value of the particle's current solution's i-th dimension, the value of the individual optimal solution's i-th dimension, the value of the global optimal solution's i-th dimension, and the remaining optional values ​​of the i-th dimension. They correspond to P1~P4 respectively. For each dimension, the roulette algorithm is used. The larger the P value, the better the fitness of the solution and the greater the probability of being selected, which reflects the probability of the particle's selection after "thinking" about the relevant solution.

[0032] S302. In this embodiment, considering the influence of threat degree, the targets are grouped and sorted. First, they are grouped according to the distance between the targets. Targets with relatively close distances are divided into the same group, which means that the targets in the group are more likely to have sensor contention. Different groups are sorted according to the average threat degree within the group. Sensor resources are allocated preferentially to the targets with greater threat degree. The arrangement order of two targets in the group is processed by enumeration method, and finally all possible target allocation sequences are obtained.

[0033] S303, for each target allocation sequence, select sensors in order according to the probability. Since the sensor selected by the target in the previous dimension cannot be repeatedly selected by the target in the next dimension, when a sensor is selected by a target in a certain dimension, the optional sensor sequence of the target in the next dimension will exclude this sensor, that is, it is guaranteed not to be selected repeatedly. However, when the target selects a dimension of the individual optimal solution or the local optimal solution, it may also be repeated with the previously selected sensor. Therefore, when two targets select the sensor with the same sequence number, the probability sorting principle is adopted at this time, and the corresponding component of the particle with a large probability and not repeated with the previously generated component is selected for update. The implementation process of this method will be introduced below.

[0034] S304. Further, when updating to the kth dimension, it is first necessary to compare whether the sensors corresponding to the kth dimension of the current solution, the individual optimal solution, and the global optimal solution have been selected by the first k-1 dimensions of the new particle sequence. If the kth dimension of the current solution already exists in the new particle sequence, the corresponding P1 marker is 0; if the kth dimension of the individual optimal solution already exists in the new particle sequence, the corresponding P2 marker is 0; similarly, if the kth dimension of the global optimal solution already exists in the new particle sequence, the corresponding P3 marker is 0. Then compare the maximum values ​​of P1~P3 with a marker bit of 1, and record which probability is the maximum value. According to the roulette algorithm and target update sequence mentioned above, the particles are updated in sequence. If the value of the newly generated dimension does not repeat with the previously generated component, the update proceeds to the next dimension. If it repeats, select the solution component corresponding to the maximum value with a marker bit of 1 in P1~P3 for update, so as to ensure that the updated particles are legal, that is, each dimension is not repeated.

[0035] S4, calculate the fitness value of the updated particle, update P1~P4 according to the fitness, and use the greedy strategy to update the individual optimal solution and the global optimal solution.

[0036] S401, similar to the previous process, use the objective function to recalculate the fitness value corresponding to the updated particle. If the fitness is better, update the individual optimal solution and the global optimal solution. At the same time, update P1~P4 corresponding to each particle to prepare for the particle update operation in the following bee stage.

[0037] S5. In the follower bee search phase, the roulette algorithm is used, and particles with better fitness are more inclined to be followed, that is, a particle update operation is performed.

[0038] S501. First, the fitness of each particle is normalized, and then the roulette algorithm is used. The particle with better fitness may be closer to the optimal solution. Then, an update operation is performed on it to ensure that the particle continues to move toward the optimal solution.

[0039] S6, similar to S4, recalculate the fitness of the updated particles, update P1~P4 according to the fitness, and use the greedy strategy to update the individual optimal solution and the global optimal solution.

[0040] S7, the stage of eliminating low-quality particles, determines whether the number of particle updates has reached the set elimination upper limit of 40. If so, a generation algorithm is used to regenerate a new particle to replace the old particle. This plays a role in widely detecting particles that have jumped out of the local optimal solution, and avoids premature algorithm maturation to a certain extent.

[0041] S8. Determine whether the iteration reaches the preset maximum number of iterations. If so, output the optimal scheduling plan for the sensor and the program ends; if not, continue iterating according to the algorithm flow.

[0042] It can be seen that the sensor resource scheduling problem usually involves the reasonable allocation of limited sensor resources to meet multiple objectives under given time and space constraints. This problem has the characteristics of high nonlinearity, complex constraints, and a wide solution space. Therefore, traditional optimization methods are difficult to solve effectively. The particle swarm optimization algorithm is good at global search and can quickly find a good solution, but it may easily fall into local optimality. The swarm algorithm emphasizes local search and refined search, especially when it is close to the optimal solution, it can help find a more accurate solution. The cross-domain sensor resource scheduling method based on adaptive particle swarms proposed in this embodiment combines the advantages of these two algorithms. The particle swarm algorithm is responsible for global search in the solution space, and the swarm algorithm can help with local refined search, enhancing the algorithm's global search ability and local optimization ability.

[0043] Combining the bee colony and particle swarm algorithms can make the scheduling process more efficient when solving the sensor resource scheduling problem through the combination of global and local searches, and can quickly find a high-quality resource allocation solution. The sensor resource scheduling environment may be dynamic, such as the sudden disappearance of the target or the interference of the sensor. The method of the present invention has strong adaptability and can adjust the search strategy in real time according to environmental changes and respond quickly. Sensor resource scheduling problems are often multi-objective. For example, while meeting task requirements, the optimal allocation of sensor resources, power consumption, interference avoidance, etc. are taken into account. The scheduling method provided in this embodiment can flexibly handle these multi-objective constraints and provide a better compromise solution.

[0044] like Figure 2 , Figure 3 , Figure 4 As shown, the corresponding relationship between the target and the sensor of the three algorithms (the scheduling method provided in this embodiment, the bee swarm algorithm, and the particle swarm algorithm) at time 750-850 is respectively shown. Considering the overhead of sensor switching during the detection of the target, it can be seen that the scheduling method is the best in terms of the switching overhead, followed by the bee swarm algorithm and finally the particle swarm algorithm; this shows the advantage of the scheduling method in the sensor scheduling problem, and it can find a sensor scheduling solution with a relatively minimum switching cost.

[0045] Figure 5 , Figure 6 , Figure 7 , Figure 8The changes of fitness with the number of iterations at time 100, time 700, time 850, and time 1000 are respectively shown, where time 100 is the time when all targets are detected by satellites, time 700 is the time when some targets are detected by satellites and some targets are detected by sea, land, and air sensors, time 850 is the time when all targets are detected by sea, land, and air sensors, and time 1000 is the time when the number of detected targets decreases and the remaining targets are all detected by sea, land, and air sensors. It can be seen that the fitness of the three algorithms decreases with the increase of the number of iterations. The fitness value of the scheduling method at the end of the iteration at all the above moments is significantly better than that of the other two algorithms, which also shows the advantage of the scheduling method, which can find the sensor scheduling scheme with the relatively minimum fitness.

[0046] In summary, the cross-domain sensor resource scheduling method based on adaptive particle swarm proposed in this embodiment can flexibly handle these multi-objective constraints in the sensor scheduling problem and provide a relatively excellent scheduling solution.

[0047] Example 2 like Fig. 9 As shown, this embodiment provides a cross-domain sensor resource scheduling method based on adaptive particle swarm, including the following steps: Initializing the particle swarm, wherein the effective multi-domain sensor scheduling sequence is used as the initial particle of the particle swarm; The initialized particle swarm is updated by combining the bee swarm algorithm and the particle update algorithm. The specific steps are as follows: In the leader bee neighborhood search phase, the particle update algorithm is used to update the initialized particle swarm, generate updated particles, and calculate the target fitness corresponding to each updated particle; In the follower bee search phase, based on the target fitness corresponding to each updated particle, the roulette algorithm is used to update the updated particle group again, obtain the updated result and recalculate the target fitness corresponding to the updated particle; Repeat the corresponding processes of the leading bee neighborhood search phase and the following bee follow-up search phase until the maximum number of iterations is reached, and output the current update result as the optimal scheduling solution for the sensor.

[0048] Specifically, the initialization process of the particle swarm includes: Initialize the population size, target number, maximum number of iterations, and number of sensors corresponding to the particle swarm; For each target visual sensor set, a generation algorithm is used to generate an effective multi-domain sensor scheduling sequence as the initial particles of the particle swarm, in which each dimension is different; For the initial particles of the particle swarm, a greedy strategy is used to calculate the initial particle fitness, initial individual optimal solution, initial global optimal solution, and the selection probability of each dimension of the particle for the current solution, individual optimal solution, global optimal solution and other optional solutions; Among them, the multi-domain sensors include space-based sensors, air-based sensors, sea-based sensors and land-based sensors.

[0049] In this embodiment, the specific steps of using the particle update algorithm to update the initialized particle swarm in the leader bee neighborhood search phase are as follows: Based on the selection probabilities of the corresponding dimensions of the current solution, individual optimal solution, global optimal solution and other optional values, the roulette algorithm is used to update the initialized particle swarm and generate updated particles. Calculate the target fitness corresponding to the updated particle, as well as the selection probabilities corresponding to the current solution, individual optimal solution, global optimal solution and other optional value solutions of the updated particle.

[0050] In this embodiment, the roulette algorithm is used, and the specific steps of updating the initialized particle swarm include: The targets are grouped according to the distance between them, and the groups are sorted from large to small according to the average threat level within the group; the order of the two targets in the group is processed by enumeration method, and finally all possible target allocation sequences are obtained; For each target allocation sequence, sensors are selected in order according to the probability size, where: When any sensor is selected by the current dimension target, this sensor will be removed from the optional sensor sequence of the target in the following dimension; When two targets select the sensor with the same serial number, the probability sorting principle is adopted to select the particle corresponding component with a large probability and not repeated with the previously generated component for update; When updating to the kth dimension, first compare whether the sensor corresponding to the kth dimension of the current solution, the individual optimal solution, and the global optimal solution has been selected by the first k-1 dimensions of the new particle sequence; If the k-th dimension of the current solution already exists in the new particle sequence, the corresponding P1 marker is 0; if the k-th dimension of the individual optimal solution already exists in the new particle sequence, the corresponding P2 marker is 0; if the k-th dimension of the global optimal solution already exists in the new particle sequence, the corresponding P3 marker is 0; where P1, P2, P3, and P4 represent the selection probabilities corresponding to the current solution, the individual optimal solution, the global optimal solution, and other optional values, respectively; Compare the maximum values ​​of P1~P3 whose mark bit is 1, and record the maximum value; The roulette algorithm and target update sequence are used to update the particles in order. If the value of the newly generated dimension does not repeat the previously generated component, the update proceeds to the next dimension; otherwise, the solution component corresponding to the maximum value with a mark bit of 1 in P1~P3 is selected for update.

[0051] In this embodiment, the specific calculation formulas of P1, P2, P3, and P4 are as follows:

[0052] In the formula, Cycle represents the current number of iterations, MaxCycle represents the set maximum number of iterations; p ranges from 0 to 1; represents the fitness value of the current solution, represents the fitness value of the individual optimal solution, Represents the fitness value of the global optimal solution.

[0053] In this embodiment, in the follower bee following search phase, based on the target fitness corresponding to each updated particle, the specific steps of using the roulette algorithm to update the updated particle swarm again are as follows: Normalize the fitness of each updated particle; Based on the normalized fitness, the roulette algorithm is then used to update the updated particle swarm again, and the fitness of the particles after the update is recalculated, as well as the selection probabilities corresponding to the current solution, individual optimal solution, global optimal solution and other optional values ​​of the particles after the update; Based on the current solution of the particle after the re-update, the individual optimal solution, the global optimal solution and the selection probabilities corresponding to the remaining optional values, the individual optimal solution and the global optimal solution are updated in combination with the greedy strategy.

[0054] In this embodiment, it also includes a stage of eliminating low-quality particles. After executing the follower bee following search stage, it is determined whether the number of particle updates has reached the set elimination upper limit. If so, a generation algorithm is used to regenerate a new particle to replace the old particle; otherwise, proceed to the next step.

[0055] like Fig.10As shown, this embodiment also provides a cross-domain sensor resource scheduling system based on adaptive particle swarms, including: a particle swarm initialization module, used to initialize the particle swarm, wherein the effective multi-domain sensor scheduling sequence is used as the initial particle of the particle swarm; a particle swarm update module, used to update the initialized particle swarm in combination with the swarm algorithm and the particle update algorithm, specifically including: a leading bee neighborhood search module, used to update the initialized particle swarm using the particle update algorithm in the leading bee neighborhood search stage, generate updated particles, and calculate the target fitness corresponding to each updated particle; a following bee following search module, used to update the updated particle swarm again using the roulette algorithm based on the target fitness corresponding to each updated particle in the following bee following search stage, obtain the updated result and recalculate the target fitness corresponding to the updated particle; a scheduling scheme output module, used to repeatedly execute the processes corresponding to the leading bee neighborhood search stage and the following bee following search stage until the maximum number of iterations is reached, and output the current update result as the optimal scheduling scheme for the sensor.

[0056] The present invention also provides a device, comprising: a memory for storing a computer program; and a processor for implementing the steps of the cross-domain sensor resource scheduling method based on adaptive particle swarm when executing the computer program.

[0057] When the processor executes the computer program, the steps of cross-domain sensor resource scheduling based on adaptive particle swarm are implemented, for example: initializing the particle swarm, wherein the effective multi-domain sensor scheduling sequence is used as the initial particle of the particle swarm; updating the initialized particle swarm in combination with the swarm algorithm and the particle update algorithm, specifically including: in the leading bee neighborhood search stage, using the particle update algorithm to update the initialized particle swarm, generating updated particles, and calculating the target fitness corresponding to each updated particle; in the following bee following search stage, based on the target fitness corresponding to each updated particle, using the roulette algorithm to update the updated particle swarm again, obtaining the updated result and recalculating the target fitness corresponding to the updated particle; repeatedly executing the corresponding processes of the leading bee neighborhood search stage and the following bee following search stage until the maximum number of iterations is reached, and outputting the current updated result as the optimal scheduling solution for the sensor.

[0058] Alternatively, the processor implements the functions of each module in the above system when executing the computer program, for example: a particle swarm initialization module, which is used to initialize the particle swarm, wherein an effective multi-domain sensor scheduling sequence is used as the initial particle of the particle swarm; a particle swarm update module, which is used to update the initialized particle swarm in combination with the swarm algorithm and the particle update algorithm, specifically including: a leading bee neighborhood search module, which is used to update the initialized particle swarm using the particle update algorithm in the leading bee neighborhood search stage, generate updated particles, and calculate the target fitness corresponding to each updated particle; a following bee following search module, which is used to update the updated particle swarm again using the roulette algorithm based on the target fitness corresponding to each updated particle in the following bee following search stage, obtain the updated result and recalculate the target fitness corresponding to the updated particle; a scheduling plan output module, which is used to repeatedly execute the processes corresponding to the leading bee neighborhood search stage and the following bee following search stage until the maximum number of iterations is reached, and output the current update result as the optimal scheduling plan for the sensor.

[0059] Exemplarily, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of completing preset functions, and the instruction segments are used to describe the execution process of the computer program in the cross-domain sensor resource scheduling device based on adaptive particle swarm. For example, the computer program can be divided into a particle swarm initialization module, a particle swarm update module and a scheduling scheme output module; wherein the particle swarm update module includes a leading bee neighborhood search module and a following bee following search module; the specific functions of each module are as follows: a particle swarm initialization module, used to initialize the particle swarm, wherein an effective multi-domain sensor scheduling sequence is used as the initial particle of the particle swarm; a particle swarm update module, used to update the initialized particle swarm in combination with the swarm algorithm and the particle update algorithm, specifically including: a leading bee neighborhood search module, used to update the initialized particle swarm using the particle update algorithm in the leading bee neighborhood search phase, generate updated particles, and calculate the target fitness corresponding to each updated particle; a following bee following search module, used to update the updated particle swarm again using the roulette algorithm based on the target fitness corresponding to each updated particle in the following bee following search phase, obtain the updated result and recalculate the target fitness corresponding to the updated particle; a scheduling scheme output module, used to repeatedly execute the processes corresponding to the leading bee neighborhood search phase and the following bee following search phase until the maximum number of iterations is reached, and output the current updated result as the optimal scheduling scheme for the sensor.

[0060] The cross-domain sensor resource scheduling device based on adaptive particle swarm can be a computing device such as a desktop computer, a notebook, a PDA, and a cloud server. The cross-domain sensor resource scheduling device based on adaptive particle swarm can include, but is not limited to, a processor and a memory. Those skilled in the art can understand that the above is an example of a cross-domain sensor resource scheduling device based on adaptive particle swarm, and does not constitute a limitation on the cross-domain sensor resource scheduling device based on adaptive particle swarm, and can include more components than the above, or combine certain components, or different components. For example, the cross-domain sensor resource scheduling device based on adaptive particle swarm can also include input and output devices, network access devices, buses, etc.

[0061] The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may be any conventional processor, etc. The processor is the control center of the cross-domain sensor resource scheduling based on adaptive particle swarm, and uses various interfaces and lines to connect various parts of the cross-domain sensor resource scheduling device based on adaptive particle swarm.

[0062] The memory can be used to store the computer program and / or module, and the processor implements various functions of the cross-domain sensor resource scheduling device based on adaptive particle swarm by running or executing the computer program and / or module stored in the memory and calling the data stored in the memory.

[0063] The memory may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area may store data created according to the use of the mobile phone (such as audio data, a phone book, etc.), etc. In addition, the memory may include a high-speed random access memory, and may also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0064] The present invention also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the steps of the cross-domain sensor resource scheduling method based on adaptive particle swarm are implemented.

[0065] If the integrated modules / units of the cross-domain sensor resource scheduling system based on adaptive particle swarm are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium.

[0066] Based on this understanding, the present invention implements all or part of the processes in the above-mentioned cross-domain sensor resource scheduling method based on adaptive particle swarms, and can also be completed by instructing related hardware through a computer program, and the computer program can be stored in a computer-readable storage medium, and when the computer program is executed by a processor, the steps of the above-mentioned cross-domain sensor resource scheduling method based on adaptive particle swarms can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or preset intermediate form, etc.

[0067] The computer-readable storage medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal and software distribution medium, etc.

[0068] It should be noted that the content contained in the computer-readable storage medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable storage media do not include electrical carrier signals and telecommunication signals.

[0069] The above embodiment is only one of the implementation methods that can realize the technical solution of the present invention. The scope of protection claimed by the present invention is not limited only to this embodiment, but also includes changes, replacements and other implementation methods that can be easily thought of by any technician familiar with the technical field within the technical scope disclosed by the present invention.

Claims

1. A cross-domain sensor resource scheduling method based on adaptive particle swarm, characterized in that: include: Initializing the particle swarm, wherein the effective multi-domain sensor scheduling sequence is used as the initial particle of the particle swarm; The initialized particle swarm is updated by combining the bee swarm algorithm and the particle update algorithm. The specific steps are as follows: In the leader bee neighborhood search phase, the particle update algorithm is used to update the initialized particle swarm, generate updated particles, and calculate the target fitness corresponding to each updated particle; In the follower bee search phase, based on the target fitness corresponding to each updated particle, the roulette algorithm is used to update the updated particle group again, obtain the updated result and recalculate the target fitness corresponding to the updated particle; Repeat the corresponding processes of the leading bee neighborhood search phase and the following bee follow-up search phase until the maximum number of iterations is reached, and output the current update result as the optimal scheduling solution for the sensor.

2. The cross-domain sensor resource scheduling method based on adaptive particle swarm according to claim 1 is characterized in that: The initialization process of the particle swarm specifically includes: Initialize the population size, target number, maximum number of iterations, and number of sensors corresponding to the particle swarm; For each target visual sensor set, a generation algorithm is used to generate an effective multi-domain sensor scheduling sequence as the initial particles of the particle swarm, in which each dimension is different; For the initial particles of the particle swarm, a greedy strategy is used to calculate the initial particle fitness, initial individual optimal solution, initial global optimal solution, and the selection probability of each dimension of the particle for the current solution, individual optimal solution, global optimal solution and other optional solutions; Among them, the multi-domain sensors include space-based sensors, air-based sensors, sea-based sensors and land-based sensors.

3. The cross-domain sensor resource scheduling method based on adaptive particle swarm according to claim 2 is characterized in that: In the leader bee neighborhood search phase, the specific steps of using the particle update algorithm to update the initialized particle swarm are as follows: Based on the selection probabilities corresponding to the current solution, individual optimal solution, global optimal solution and other optional value solutions, the roulette algorithm is used to update the initialized particle swarm and generate updated particles; Calculate the target fitness corresponding to the updated particle, as well as the selection probabilities corresponding to the current solution, individual optimal solution, global optimal solution and other optional value solutions of the updated particle.

4. The cross-domain sensor resource scheduling method based on adaptive particle swarm according to claim 3 is characterized in that: in, Using the roulette algorithm, the specific steps of updating the initialized particle swarm include: The targets are grouped according to the distance between them, and the groups are sorted from large to small according to the average threat level within the group; the order of the two targets in the group is processed by enumeration method, and finally all possible target allocation sequences are obtained; For each target allocation sequence, sensors are selected in order according to the probability size, where: When any sensor is selected by the current dimension target, this sensor will be removed from the optional sensor sequence of the target in the following dimension; When two targets select the sensor with the same serial number, the probability sorting principle is adopted to select the particle corresponding component with a large probability and not repeated with the previously generated component for update; When updating to the kth dimension, first compare whether the sensor corresponding to the kth dimension of the current solution, the individual optimal solution, and the global optimal solution has been selected by the first k-1 dimensions of the new particle sequence; If the k-th dimension of the current solution already exists in the new particle sequence, the corresponding P1 marker is 0; if the k-th dimension of the individual optimal solution already exists in the new particle sequence, the corresponding P2 marker is 0; if the k-th dimension of the global optimal solution already exists in the new particle sequence, the corresponding P3 marker is 0; where P1, P2, P3, and P4 represent the selection probabilities corresponding to the current solution, the individual optimal solution, the global optimal solution, and other optional values, respectively; Compare the maximum values ​​of P1~P3 whose mark bit is 1, and record the maximum value; The roulette algorithm and target update sequence are used to update the particles in order. If the value of the newly generated dimension does not repeat the previously generated component, the update proceeds to the next dimension; otherwise, the solution component corresponding to the maximum value with a mark bit of 1 in P1~P3 is selected for update.

5. The cross-domain sensor resource scheduling method based on adaptive particle swarm according to claim 4 is characterized in that: The specific calculation formulas for P1, P2, P3, and P4 are as follows: In the formula, Cycle represents the current number of iterations, MaxCycle represents the set maximum number of iterations; p ranges from 0 to 1; represents the fitness value of the current solution, represents the fitness value of the individual optimal solution, Represents the fitness value of the global optimal solution.

6. The cross-domain sensor resource scheduling method based on adaptive particle swarm according to claim 5 is characterized in that: In the following bee following search phase, the specific steps of re-updating the updated particle swarm using the roulette algorithm based on the target fitness corresponding to each updated particle are as follows: Normalize the fitness of each updated particle; Based on the normalized fitness, the roulette algorithm is then used to update the updated particle swarm again, and the fitness of the particles after the update is recalculated, as well as the selection probabilities corresponding to the current solution, individual optimal solution, global optimal solution and other optional value solutions of the particles after the update; Based on the current solution of the particle after the re-update, the individual optimal solution, the global optimal solution and the selection probabilities corresponding to the remaining optional values, the individual optimal solution and the global optimal solution are updated in combination with the greedy strategy to obtain the updated result.

7. The cross-domain sensor resource scheduling method based on adaptive particle swarm according to claim 5 is characterized in that: in, It also includes the stage of eliminating low-quality particles. After executing the follow-up bee search stage, it is determined whether the number of particle updates has reached the set elimination upper limit. If so, a new particle is regenerated using the generation algorithm to replace the old particle; otherwise, proceed to the next step.

8. A cross-domain sensor resource scheduling system based on adaptive particle swarm, characterized in that: include: A particle swarm initialization module is used to initialize the particle swarm, wherein an effective multi-domain sensor scheduling sequence is used as the initial particle of the particle swarm; The particle swarm update module is used to update the initialized particle swarm by combining the bee swarm algorithm and the particle update algorithm, which specifically includes: The leading bee neighborhood search module is used to update the initialized particle swarm using a particle update algorithm during the leading bee neighborhood search phase, generate updated particles, and calculate the target fitness corresponding to each updated particle; The follower bee following search module is used to update the updated particle group again based on the target fitness corresponding to each updated particle using the roulette algorithm in the follower bee following search stage, obtain the update result and recalculate the target fitness corresponding to the updated particle; The scheduling scheme output module is used to repeatedly execute the corresponding processes of the leading bee neighborhood search phase and the following bee following search phase until the maximum number of iterations is reached, and output the current update result as the optimal scheduling scheme for the sensor.

9. A device, characterized in that: include: Memory for storing computer programs; A processor is used to implement the steps of the cross-domain sensor resource scheduling method based on adaptive particle swarms according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, it is used to implement the steps of the cross-domain sensor resource scheduling method based on adaptive particle swarm according to any one of claims 1 to 7.

Images