Method and system for suppressing clock synchronization network jitter when drone cluster formation changes

Abstract

The present invention discloses a method and system for suppressing jitter of a clock synchronization network when a drone cluster formation changes. The method of the present invention includes constructing a measurement network G for the drone cluster formation, randomly initializing the value of each edge weight in the measurement network G; then iterating: respectively updating the local clock of any node i in the measurement network G in the next iteration cycle; respectively calculating the probability that any node i in the measurement network G can establish a valid path in the measurement link, selecting the node with the largest probability of a valid path to establish the measurement link, and updating the edge weight of the measurement link; repeating the above iteration process until a preset end condition is reached. The present invention can quantitatively describe the topological characteristics of the measurement network and its changing rules, can reduce the jitter of the time measurement network caused by the change of the drone cluster formation, realize the smooth transition and rapid convergence of the distributed clock synchronization of the drone cluster, and effectively realize the clock synchronization control optimization for large-scale dynamic drone clusters.

Description

Technical Field

[0001] The present invention relates to unmanned aerial vehicle cluster technology, and in particular to a method and system for suppressing clock synchronization network jitter when a formation of an unmanned aerial vehicle cluster changes. Background Art

[0002] The drone cluster is a wireless sensor network (WSN) with dynamic changing properties in the air, and the solution to the WSN clock synchronization problem is mainly distributed clock synchronization. However, the drone cluster formation is constrained by the planned path for executing the task, and the network topology is also affected by the synchronous clock time-frequency transmission information mode, which will cause the drone cluster network topology to change frequently but not randomly like WSN; and the failure of some nodes and paths in the drone cluster formation and the addition of new drones will cause the drone cluster topology to change dynamically, thereby causing the jitter of the distributed clock synchronization network. The application background of this embodiment is to dynamically control and optimize the distributed clock network when the drone cluster formation changes, suppress the jitter of the clock synchronization network, and provide strong support for maintaining high-precision clock synchronization of the self-organizing network when the drone cluster formation changes. Existing distributed clock network jitter suppression methods such as average consistency time method, weighted maximum time method, time diffusion method, tree-based lightweight time synchronization method, etc., provide ideas for the rapid convergence method of distributed clock networks in the time domain, which is beneficial to the jitter suppression of drone cluster clock networks within a certain application range. However, jitter suppression based on time transmission is currently still used in point-to-point or fixed three-node drone cluster network topologies, and how to regulate and reduce the impact of dynamic changes in network topology on clock synchronization needs to be improved; moreover, the jitter suppression accuracy of existing distributed synchronous clocks used in sensor networks is generally in the microsecond range, and most jitter suppression algorithms cannot be directly applied to high-precision clock synchronization under dynamic changes in drone cluster formations. Summary of the invention

[0003] Technical problem to be solved by the present invention: In view of the above-mentioned problems in the prior art, a method and system for suppressing clock synchronization network jitter when a drone cluster formation changes are provided. The present invention can quantitatively describe the measurement network topology characteristics and their changing laws, reduce the time measurement network jitter caused by the drone cluster formation change, and achieve smooth transition and rapid convergence of distributed clock synchronization of the drone cluster, effectively realizing clock synchronization control optimization for large-scale dynamic drone clusters.

[0004] In order to solve the above technical problems, the technical solution adopted by the present invention is:

[0005] A method for suppressing clock synchronization network jitter when a drone cluster formation changes, comprising:

[0006] S1, build a measurement network G for the UAV cluster formation. The measurement network G includes a set of nodes, edges, and edge weights;

[0007] S2, randomly initialize the values ​​of the edge weights in the measurement network G;

[0008] S3, respectively update the local clock of any node i in the measurement network G in the next iteration cycle;

[0009] S4, respectively calculating the probability that any node i in the measurement network G can establish a valid path in the measurement link;

[0010] S5, respectively select the node with the largest probability of valid path for any node i in the measurement network G to establish a measurement link, and update the edge weight of the established measurement link;

[0011] S6, determine whether the preset end condition is met, if the preset end condition is met, end and exit; otherwise, jump to step S3 to continue iteration.

[0012] Optionally, the function expression of the measurement network G constructed in step S1 is:

[0013]

[0014] Where U is the node matrix of the measurement network, B is the edge between two nodes, i.e., the measurement link matrix, and W is the matrix of edge weights.

[0015] Optionally, in step S2, the values ​​of the edge weights in the measurement network G are randomly initialized, and the constraints are:

[0016]

[0017] In the above formula, U i represents the set of nodes connected to node i, w i,j Representation Node i and nodes j The edge weight between nodes j For and node i A collection of connected nodes U i The j nodes.

[0018] Optionally, in step S3, the function expression for updating the local clock of any node i in the measurement network G in the next iteration cycle is:

[0019]

[0020] In the above formula, For Nodei In the next iteration cycle t k +1 local clock, Representation Node i and nodes j The edge weight between the edges in this iteration cycle t k The weight within For Node j In this iteration cycle t k The local clock in U i Represents the set of nodes connected to node i, node j For and node i A collection of connected nodes U i The j nodes.

[0021] Optionally, in step S4, any node i may establish a calculation function expression for the probability of a valid path in the measurement link as follows:

[0022]

[0023] In the above formula, For Node i and nodes j In this iteration cycle t k The probability of a valid path in the measurement link can be established within For Node j In this iteration cycle t k The local clock in For Node i In this iteration cycle t k The local clock in For Node q In this iteration cycle t k The local clock in the node q For and node i A collection of connected nodes U i The j nodes.

[0024] Optionally, the function expression for updating the edge weight of the established measurement link in step S5 is:

[0025]

[0026] In the above formula, Representation Nodei and nodes j The edge weight between the edges in the next iteration cycle t k +1 weight, Representation Node i and nodes j The edge weight between the edges in this iteration cycle t k The weight within, s is the moving step length, For Node j In this iteration cycle t k The local clock in For Node i In this iteration cycle t k The local clock in the

[0027] Optionally, the end condition preset in step S6 refers to the number of iterations being equal to a set value, or satisfying the following dynamic clock synchronization network jitter suppression decision convergence determination adjustment:

[0028]

[0029] In the above formula, max means taking the maximum value, and min means taking the minimum value. For Node i In this iteration cycle t k The local clock in U i Represents the set of nodes connected to node i.

[0030] Optionally, before step S1, the step of determining the nodes and edges of the drone cluster formation is also included: taking a set time segment as a unit, detecting the drones in the drone cluster formation within the time segment as nodes, and the connections between the drones forming edges, thereby obtaining a sub-topology graph composed of nodes and edges; then combining the sub-topology graphs of multiple consecutive time segments constituting the overall test time T in pairs to form a complete topology graph containing the nodes and edges of the drone cluster formation.

[0031] In addition, the present invention also provides a clock synchronization network jitter suppression system when the drone cluster formation changes, including an interconnected microprocessor and a memory, and the microprocessor is programmed or configured to execute the steps of the clock synchronization network jitter suppression method when the drone cluster formation changes.

[0032] In addition, the present invention also provides a computer-readable storage medium, in which a computer program is stored. The computer program is used to be programmed or configured by a microprocessor to execute the steps of the clock synchronization network jitter suppression method when the drone cluster formation changes.

[0033] Compared with the prior art, the present invention mainly has the following advantages: the clock synchronization network jitter suppression method of the present invention when the formation of the drone cluster changes can quantitatively describe the measurement network topology characteristics and its changing rules by constructing a dynamic time measurement network topology map model of the drone cluster. In addition, based on the clock network control optimization algorithm proposed by the present invention and the adaptive controller with stable convergence for different formation types, the time measurement network jitter caused by the formation change of the drone cluster can be reduced, and the distributed clock synchronization of the drone cluster can be smoothly transitioned and quickly converged, effectively realizing high-precision clock synchronization for large-scale dynamic drone clusters. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 Schematic diagram of the basic flow of the method of the embodiment of the present invention.

[0035] Figure 2 4 is a schematic diagram of a bidirectional time measurement principle between adjacent node pairs in an embodiment of the present invention.

[0036] Figure 3 This is an example of measuring the clock time control of the current node in the network in an embodiment of the present invention.

[0037] Figure 4 It is a schematic diagram of the fusion of dynamic time-frequency measurement network topology graph in an embodiment of the present invention.

[0038] Figure 5 This is a comparison of network topology changes in clock synchronization measurement when the drone cluster changes in an embodiment of the present invention.

[0039] Figure 6 This is a schematic diagram of synchronous clock jitter before the control optimization method of this embodiment is added.

[0040] Figure 7 This is a schematic diagram of synchronous clock jitter after control optimization by the method of this embodiment. DETAILED DESCRIPTION

[0041] like Figure 1 As shown, the clock synchronization network jitter suppression method when the drone cluster formation changes in this embodiment includes:

[0042] S1, build a measurement network G for the UAV cluster formation. The measurement network G includes a set of nodes, edges, and edge weights;

[0043] S2, randomly initialize the values ​​of the edge weights in the measurement network G;

[0044] S3, respectively update the local clock of any node i in the measurement network G in the next iteration cycle;

[0045] S4, respectively calculating the probability that any node i in the measurement network G can establish a valid path in the measurement link;

[0046] S5, respectively select the node with the largest probability of valid path for any node i in the measurement network G to establish a measurement link, and update the edge weight of the established measurement link;

[0047] S6, determine whether the preset end condition is met, if the preset end condition is met, end and exit; otherwise, jump to step S3 to continue iteration.

[0048] In this embodiment, the weighted network topology model in complex network theory is used to construct a UAV cluster time measurement network. When a new UAV joins or leaves the cluster, all edge weights in the measurement network will change, and the degree of change depends on the actual situation of each edge. Therefore, the function expression of the measurement network G constructed in step S1 of this embodiment is:

[0049] , (1)

[0050] Where U is the node matrix of the measurement network, B is the edge between two nodes, i.e., the measurement link matrix, and W is the matrix of edge weights.

[0051] In this embodiment, the values ​​of the edge weights in the measurement network G are randomly initialized in step S2, and the constraints are:

[0052] , (2)

[0053] In the above formula, U i represents the set of nodes connected to node i (neighboring nodes of node i), w i,j Representation Node i and nodes j The edge weight between nodes j For and node i A collection of connected nodes U i The j Nodes. i and all nodes connected to it j The sum of the edge weights of the edges between them is 1.

[0054] In this embodiment, the function expression for updating the local clock of any node i in the measurement network G in the next iteration cycle in step S3 is:

[0055] , (3)

[0056] In the above formula, For Node i In the next iteration cycle t k +1 local clock, Representation Node i and nodes j The edge weight between the edges in this iteration cycle t k The weight within For Node j In this iteration cycle t k The local clock in U i Representation and Nodes i A collection of connected nodes, nodes j For and node i A collection of connected nodes U i The j nodes. According to the above formula, The corresponding node in this iteration cycle t k The sum of the products of the edge weights of the established measurement link nodes within and around the network and the clock time characteristics of the corresponding nodes.

[0057] In this embodiment, the calculation function expression of the probability that any node i can establish a valid path in the measurement link in step S4 is:

[0058] , (4)

[0059] In the above formula, For Node i and nodes j In this iteration cycle t k The probability of a valid path in the measurement link can be established within For Node j In this iteration cycle t k The local clock in For Node i In this iteration cycle t k The local clock in For Node q In this iteration cycle t k The local clock in the node q For and node i A collection of connected nodesU i The j Based on the above formula, the probability of establishing a valid path in the measurement link can be selected, that is, the greater the difference in clock time between two nodes, the greater the probability of establishing a measurement link.

[0060] In this embodiment, the function expression for updating the edge weight of the established measurement link in step S5 is:

[0061] , (5)

[0062] In the above formula, Representation Node i and nodes j The edge weight between the edges in the next iteration cycle t k +1 weight, Representation Node i and nodes j The edge weight between the edges in this iteration cycle t k The weight within, s is the moving step length, For Node j In this iteration cycle t k The local clock in For Node i In this iteration cycle t k The local clock in the

[0063] In this embodiment, the end condition preset in step S6 refers to the number of iterations being equal to a set value, or satisfying the following dynamic clock synchronization network jitter suppression decision convergence determination adjustment:

[0064] , (6)

[0065] In the above formula, max means taking the maximum value, and min means taking the minimum value. For Node i In this iteration cycle t k The local clock in U i Represents the set of nodes connected to node i.

[0066] See also Figure 2 The key to the method of this embodiment is the two-way time information interaction between adjacent node pairs. By using a two-way time measurement mechanism between two nodes ( U i → U j → Ui ), the relative clock deviation between any two nodes in the network can be obtained. U i Representation and Nodes i A collection of connected nodes, U j Representation and Nodes j A collection of connected nodes. Figure 2 Any adjacent node pair is shown i and j The process of two-way time information exchange between τ( i ) moment, node i Send time interaction group, and send time τ( i ) is recorded in the group; node j In τ( j,i ) receives the data from the node i The grouping, and in τ( i ) is replied at the same time, and τ( j,i ) and τ( j ) is recorded in the packet; the reply frame information is in τ( i,j ) arrives at the node i . Then the above time nodes have the following relationship:

[0067] , (7)

[0068] Among them, τ d0 Is a node i To Node j Path propagation time, τ d1 Is a node j To Node i The propagation time, Δ τ is the relative deviation between the clocks of the two nodes. Therefore, the formula for calculating the clock deviation Δτ is:

[0069] , (8)

[0070] If τ d0 Not equal to τ d1 , that is, there is a relative moving speed difference between the two nodes in the UAV formation change v , then the calculated error of the clock deviation between the two nodes is Δ ε It can be defined as:

[0071] , (9)

[0072] Since the node movement speed is much slower than the speed of radio waves, the node movement during the air transmission of the packet frame message between drones can be ignored. Therefore, the inconsistency of the round-trip delay recorded mainly depends on the time length from the i node receiving the time interaction packet sent by the j node to its reply. t pw , then the absolute value of the calculated error of the clock deviation is:

[0073] , (10)

[0074] Here, c is the speed of light in a vacuum.

[0075] After completing the two-way time information interaction, each node in the measurement network topology can obtain the clock time deviation with the node that can establish the measurement link, and then calculate the adjustment amount corresponding to the clock of the node through the edge weight recorded in the method of this embodiment, so as to complete the clock synchronization convergence of the global measurement network.

[0076] Assume that the measurement network G is Figure 3 As shown in the figure, the numbers in the figure are the node numbers. t k Time period (this iteration cycle t k ), node 3 can obtain the time deviation of the local node with which it establishes a link through two-way time information interaction and , node 3 can use these time deviations to calculate its own time adjustment:

[0077] , (11)

[0078] in, and is the edge weight generated by node 3 for each link node, satisfying:

[0079] , (12).

[0080] The process of node 3 adjusting its own clock time is:

[0081] , (13)

[0082] In the above formula, Indicates the time deviation of node 3. Similar processing is performed on other nodes in the time measurement network of the drone cluster. After several iterations, the entire network completes clock time synchronization.

[0083] It should be noted that the nodes and edges of the drone cluster formation can be manually specified based on the configuration of the drone cluster formation. In addition, considering that the drone cluster may change when performing tasks in a collaborative manner, the synchronous time measurement network topology of different time periods also changes dynamically, but based on the time information transmission mode and the task planning path constraints, the change of the synchronous time measurement network topology follows certain rules and mechanisms. Based on this, as an optional implementation method, the present embodiment adopts a real-time detection method, constructs a clock synchronization measurement topology map based on the overall test time T and fine granularity, and designs a synchronous time measurement network topology model from both macro and micro levels. Specifically, before step S1, it also includes the step of determining the nodes and edges of the drone cluster formation: taking the set time segment as a unit, detecting the drones in the drone cluster formation as nodes and the connections between drones in the time segment to form edges, thereby obtaining a sub-topology map composed of nodes and edges; then the sub-topology maps of multiple continuous time segments constituting the overall test time T are combined in pairs to form a complete topology map containing the nodes and edges of the drone cluster formation.

[0084] like Figure 4 As shown, multiple consecutive time segments are represented as t 1,1 ~t 1,4 , and the corresponding sub-topology graphs are represented as ~ , after pairwise combination, the sub-topology and Combine to get the sub-topology corresponding to time period t1 , sub-topology and Combine to get the sub-topology corresponding to time period t2 , and finally the sub-topology graph and The combined measurement network G can be expressed as:

[0085] , (14)

[0086] In the above formula, Indicates time period t i Corresponding sub-topology diagram, time period t i Total test time T Considering the convergence performance and stability requirements of the clock synchronization algorithm under static conditions, the overall test time T It is divided into multiple time segments, and the clock measurement network topology in one or more adjacent time segments can be merged into a sub-topology graph, whose model expression is:

[0087] , (15)

[0088] In the above formula, Indicates time period t i Next level time period t i , t j The corresponding sub-topology diagram. The clock measurement network topology in one or more adjacent time segments can be merged into a sub-topology diagram that can show the basic connectivity that the drone cluster time-frequency measurement network can achieve and the convergence consistency under regional static conditions; and the measurement network finally combined G It reflects the frequency and density of establishing measurement links between nodes in the drone cluster within the measurement time range, as well as the link building and time-frequency measurement capabilities of the corresponding edge weights of each node in the network. Figure 5 As shown, a schematic diagram comparing the topology change of the clock synchronization measurement network when the method of this embodiment and the traditional drone cluster change is shown, where (a) is a schematic diagram of the topology change of the clock synchronization measurement network when the traditional drone cluster changes, and (b) is a schematic diagram of the topology change of the clock synchronization measurement network when the drone cluster changes according to the method of this embodiment, wherein the topology controller is used to generate a measurement network G based on the sub-topology graphs of each time period in pairs, for comparison Figure 5 It can be seen that the traditional clock synchronization measurement network topology directly establishes links through time-frequency interaction between network nodes, which will cause drastic changes in the overall synchronization clock in the network, thereby causing jitter. The clock synchronization jitter suppression method designed in this embodiment is adopted to establish the synchronization network topology principle by introducing the minimum jitter principle in each iteration, and the purpose of suppressing jitter is achieved through network topology optimization and iterative regulation in time segments.

[0089] Finally, for the measurement nodes in the node matrix U, a multi-dimensional target search is performed based on the established edge matrix B, and each node has its own clock time characteristics. The clock deviation between each node will affect the clock time of the surrounding link building time measurement nodes. The clock time characteristics of each node and the generated edge weight value are related to the corresponding jitter suppression objective function. When the absolute deviation of the local maximum clock time and the minimum clock time in all adjusted nodes approaches 0 after the preset number of iterations, it means that it has a better target value, otherwise the target value is worse. Finally, the clock time characteristics of most nodes will converge to be consistent. Figure 6 and Figure 7 The figure shows the comparison effect of the clock synchronization measurement network jitter suppression when the method of this embodiment changes with the traditional drone cluster. Figure 6 The clock jitter of the clock synchronization measurement network is measured for the traditional drone cluster when the clock changes. Figure 7 This is the clock jitter after adding the control optimization operation of the method in this embodiment. Figure 6 and Figure 7 It can be seen that adding the control optimization operation of the method of this embodiment can significantly suppress the synchronous clock jitter.

[0090] In summary, most of the existing UAV cluster clock synchronization methods are used when the flight formation does not change, and there is a lack of effective clock synchronization control optimization methods when the dynamic UAV flight formation changes. The method of this embodiment can quantitatively describe the topological characteristics of the measurement network and its changing rules by constructing a dynamic time-frequency measurement network topology map model for the UAV cluster. In addition, the designed adaptive controller based on the firefly optimization algorithm and convergence stability for different formation types can reduce the jitter of the time-frequency measurement network caused by the change of the UAV cluster formation, realize the smooth transition and rapid convergence of the distributed clock synchronization of the UAV cluster, and provide an effective method for the optimization of clock synchronization control of large-scale dynamic UAV clusters.

[0091] In addition, this embodiment also provides a clock synchronization network jitter suppression system when the drone cluster formation changes, including an interconnected microprocessor and a memory, and the microprocessor is programmed or configured to execute the steps of the aforementioned clock synchronization network jitter suppression method when the drone cluster formation changes.

[0092] In addition, this embodiment also provides a computer-readable storage medium, which stores a computer program. The computer program is used to be programmed or configured by a microprocessor to execute the steps of the clock synchronization network jitter suppression method when the aforementioned drone cluster formation changes.

[0093] Those skilled in the art should understand that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application can take the form of a computer program product implemented on one or more computer-readable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the process Figure 1 A process or multiple processes and / or boxes Figure 1These computer program instructions can also be stored in a computer-readable memory that can guide a computer or other programmable data processing device to work in a specific way, so that the instructions stored in the computer-readable memory produce a product including an instruction device, which implements the functions specified in the process. Figure 1 A process or multiple processes and / or boxes Figure 1 These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable device provide for implementing the process in the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0094] The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. A method for suppressing clock synchronization network jitter when a drone cluster formation changes, characterized in that: include: S1, build a measurement network G for the UAV cluster formation. The measurement network G includes a set of nodes, edges, and edge weights; S2, randomly initializes the value of each edge weight in the measurement network G; S3, respectively update the local clock of any node i in the measurement network G in the next iteration cycle; S4, respectively calculating the probability that any node i in the measurement network G can establish a valid path in the measurement link; S5, respectively select the node with the largest probability of valid path for any node i in the measurement network G to establish a measurement link, and update the edge weight of the established measurement link; S6, judging whether the preset end condition is met, if the preset end condition is met, then ending and exiting; Otherwise, jump to step S3 to continue iteration.

2. The method for suppressing clock synchronization network jitter when the drone cluster formation changes according to claim 1 is characterized in that: The function expression of the measurement network G constructed in step S1 is: G={U,B,W} In the above formula, U is the node matrix of the measurement network, B is the edge between two nodes, that is, the measurement link matrix, and W is the matrix of edge weights.

3. The method for suppressing clock synchronization network jitter when the drone cluster formation changes according to claim 1 is characterized in that: In step S2, the values ​​of the edge weights in the measurement network G are randomly initialized, and the constraints are: In the above formula, U i represents the set of nodes connected to node i, w i,j Represents the edge weight of the edge between node i and node j, and node j is the node set U connected to node i i The jth node in .

4. The method for suppressing clock synchronization network jitter when the drone cluster formation changes according to claim 1 is characterized in that: In step S3, the function expression for updating the local clock of any node i in the measurement network G in the next iteration cycle is: In the above formula, is the node i in the next iteration cycle t k +1 within the local clock, Indicates the edge weight between node i and node j in this iteration cycle t k The weight within is the node j in this iteration cycle t k The local clock in U i represents the set of nodes connected to node i, and node j is the set of nodes connected to node i U i The jth node in .

5. The method for suppressing clock synchronization network jitter when the drone cluster formation changes according to claim 1 is characterized in that: In step S4, the calculation function expression of the probability that any node i can establish a valid path in the measurement link is: In the above formula, is the node i and node j in this iteration cycle t k The probability of a valid path in the measurement link can be established within is the node j in this iteration cycle t k The local clock in is the node i in this iteration cycle t k The local clock in is the node q in this iteration cycle t k The local clock in the node q is the node set U connected to the node i. i The jth node in .

6. The method for suppressing clock synchronization network jitter when the drone cluster formation changes according to claim 1 is characterized in that: The function expression for updating the edge weight of the established measurement link in step S5 is: In the above formula, Indicates the edge weight between node i and node j in the next iteration cycle t k +1 weight, Indicates the edge weight between node i and node j in this iteration cycle t k The weight within, s is the moving step length, is the node j in this iteration cycle t k The local clock in is the node i in this iteration cycle t k The local clock in the 7. The method for suppressing clock synchronization network jitter when the drone cluster formation changes according to claim 1 is characterized in that: The end condition preset in step S6 means that the number of iterations is equal to the set value, or the following dynamic clock synchronization network jitter suppression decision convergence determination adjustment is satisfied: In the above formula, max means taking the maximum value, and min means taking the minimum value. is the node i in this iteration cycle t k The local clock in U i Represents the set of nodes connected to node i.

8. The method for suppressing clock synchronization network jitter when the drone cluster formation changes according to claim 1 is characterized in that: Before step S1, the step of determining the nodes and edges of the drone cluster formation is also included: taking a set time segment as a unit, detecting the drones in the drone cluster formation within the time segment as nodes, and the connections between the drones forming edges, thereby obtaining a sub-topology graph composed of nodes and edges; then combining the sub-topology graphs of multiple consecutive time segments constituting the overall test time T in pairs to form a complete topology graph containing the nodes and edges of the drone cluster formation.

9. A clock synchronization network jitter suppression system for changes in drone cluster formation, comprising interconnected microprocessors and memories, characterized in that: The microprocessor is programmed or configured to execute the steps of the method for suppressing clock synchronization network jitter when the drone cluster formation changes as described in any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored therein, characterized in that: The computer program is used to be programmed or configured by a microprocessor to execute the steps of the method for suppressing clock synchronization network jitter when the drone cluster formation changes as described in any one of claims 1 to 8.

Images