Unmanned cluster collaborative ad hoc network time slot resource allocation method and device

Abstract

The embodiment of the invention provides an unmanned cluster collaborative ad hoc network time slot resource allocation method and device, and the method and device achieve the effective mapping of demands through the innovative design of a task configuration system, a state machine and a resource matrix. A time slot allocation mechanism is constructed, and a reliable scheduling scheme is established in combination with connectivity analysis and resource matching. Dynamic optimization is introduced, and the adaptability of distribution is ensured through state monitoring and strategy adjustment. According to the method, the defects of the traditional technology in the aspects of task configuration, time slot allocation, dynamic adjustment and the like are effectively overcome, and technical support is provided for unmanned cluster collaboration.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of unmanned aerial vehicles, in particular to a method and device for allocating time slot resources in cooperative ad hoc networks of unmanned clusters. BACKGROUND

[0002] The existing time slot allocation method for unmanned clusters has obvious deficiencies. Traditional systems perform poorly in task configuration and resource allocation, and fail to effectively support search and rescue task requirements, affecting communication efficiency.

[0003] In addition, the existing technology has bottlenecks in time slot division and network connectivity. Most systems lack perfect time frame structure and state analysis strategies, resulting in unreasonable resource scheduling.

[0004] The existing system has technical shortcomings in dynamic adjustment. Lack of in-depth analysis of node state, it is difficult to achieve flexible resource allocation through strategy optimization, affecting system performance. The solution to these problems is of great significance to improve the cooperative ability of unmanned clusters. SUMMARY

[0005] To solve the problems in the prior art, the present application provides a method and device for allocating time slot resources in cooperative ad hoc networks of unmanned clusters, which can effectively solve the deficiencies of traditional technology in task configuration, time slot allocation and dynamic adjustment, and provide technical support for cooperative unmanned clusters.

[0006] To solve at least one of the above problems, the present application provides the following technical solutions: In a first aspect, the present application provides a method for allocating time slot resources in cooperative ad hoc networks of unmanned clusters, comprising: Setting cross-domain cluster nodes as gateway control nodes, setting job group nodes as water surface sensing nodes, setting unmanned platform nodes as execution search and rescue nodes, generating a task configuration matrix, constructing a task state machine according to the task configuration matrix, performing task deduction and calculation based on the task state machine for search and rescue area coverage, target search tracking and wounded transport salvage, converting task requirements into a time slot resource matrix, and writing the time slot resource matrix into a resource allocation unit; Dividing the time frame into a time synchronization area and a data interaction area, writing the time frame structure into each water surface unmanned node control unit, reading the cache queue length of adjacent search and rescue nodes, generating a time slot occupation table reflecting the communication state of the unmanned cluster, calculating a network connectivity matrix between water surface nodes according to the time slot occupation table, performing matching operation on the network connectivity matrix and the time slot resource matrix, generating an initial time slot allocation scheme suitable for search and rescue scenarios, and writing the initial time slot allocation scheme into a resource scheduling unit; The control information is written in the data frame header in a piggyback structure, a framing configuration table is generated according to the electromagnetic propagation characteristics on the sea, the framing configuration table is used for data framing processing, the time slot occupation table is read for distributed time slot selection of the search and rescue node, a unmanned node queue state table is established based on the network connection matrix, a time slot occupation dynamic adjustment coefficient is calculated according to the unmanned node queue state table, the distribution scheme is corrected according to the dynamic adjustment coefficient, the corrected distribution scheme is written into a time slot resource scheduling unit, and a time slot resource scheduling strategy supporting the sea maneuver of the unmanned cluster is generated.

[0007] Further, the method further comprises: constructing a cluster organization table comprising a control level, a command relationship and an operation area; reading a cross-domain cluster node identifier and mapping it to a gateway control node; reading an operation group node identifier and mapping it to a water surface sensing node; reading a unmanned platform node identifier and mapping it to an execution search and rescue node; and writing the cluster organization table into a node management unit. The method further comprises: constructing a task planning table comprising a task attribute, a message type and a communication period; dividing a node communication level according to the cluster organization table; generating a task configuration matrix comprising a node identifier, a level attribute and a communication demand; and writing the task configuration matrix into a task management unit.

[0008] Further, the method further comprises: constructing a state machine configuration table comprising a state transition condition, a task trigger condition and a deduction parameter; generating a task state machine according to the task configuration matrix; performing Monte Carlo deduction on search and rescue area coverage, target search tracking and wounded personnel transfer salvage based on the state machine; generating a task demand prediction table; and writing the task demand prediction table into a deduction engine. The method further comprises: constructing a resource mapping table comprising a communication performance, a data capacity and a quality index; calculating a communication resource demand according to the task demand prediction table; generating a time slot resource matrix; and writing the time slot resource matrix into a resource allocation unit according to a priority.

[0009] Further, the method further comprises: constructing a time frame structure table comprising a synchronization frame length, a data frame length and a guard interval; dividing a time frame into a time synchronization area and a data interaction area according to a fixed interval; generating a framing parameter table comprising a frame format, a time slot number and a time slot length; and writing the framing parameter table into each water surface unmanned node control unit. The method further comprises: constructing a neighbor state table comprising a node address, a queue length and a survival time; reading a cache queue length information of a neighboring search and rescue node; generating a time slot occupation table comprising a time slot state, reservation information and hop number information; and writing the time slot occupation table into a state management unit.

[0010] Further, it also comprises: constructing a network topology table containing node identification, link state, time slot state, calculating the connectivity between water surface nodes according to the time slot occupation table, generating a network connectivity matrix, and writing the network connectivity matrix into the topology management unit; Constructing a matching strategy table containing resource demand, connectivity state, and allocation weight, performing weighted matching operation on the network connectivity matrix and the time slot resource matrix, generating an initial time slot allocation scheme, and writing the initial time slot allocation scheme into the resource scheduling unit.

[0011] Further, it also comprises: constructing a dedicated frame header table containing frame type, reservation field, and queue length, writing control information into the data frame header in a piggyback structure, generating a framing configuration table according to the characteristics of electromagnetic propagation on the sea, and writing the framing configuration table into the data processing unit; Constructing a time slot state table containing idle state, occupation state, and busy state, reading the time slot occupation table for state judgment, generating a distributed selection matrix containing transmission time slot, reception time slot, and reservation time slot, and writing the distributed selection matrix into the time slot scheduling unit.

[0012] Further, it also comprises: constructing a state monitoring table containing node identification, queue length, and life cycle, calculating the queue state distribution between nodes according to the network connectivity matrix, generating an unmanned node queue state table, calculating a time slot occupation dynamic adjustment coefficient according to the unmanned node queue state table, and writing the dynamic adjustment coefficient into the adjustment control unit; Constructing a resource reallocation table containing initial scheme, adjustment coefficient, and priority, modifying the initial time slot allocation scheme according to the dynamic adjustment coefficient, generating a dynamic time slot allocation scheme, and writing the dynamic time slot allocation scheme into the time slot resource scheduling unit.

[0013] In a second aspect, the application provides a time slot resource allocation device for unmanned cluster cooperative ad hoc network, comprising: A task configuration module is configured to set a cross-domain cluster node as a gateway control node, set a job group node as a water surface sensing node, set an unmanned platform node as an execution search and rescue node, generate a task configuration matrix, construct a task state machine according to the task configuration matrix, perform task deduction and calculation based on the task state machine for search and rescue area coverage, target search tracking, and wounded personnel transfer salvage, convert task demand into a time slot resource matrix, and write the time slot resource matrix into the resource allocation unit. The resource scheduling module is configured to divide a time frame into a time synchronization area and a data interaction area, write the time frame structure into each water surface unmanned node control unit, read the buffer queue length of adjacent search and rescue nodes, generate a time slot occupation table reflecting the unmanned cluster communication state, calculate a network connectivity matrix between water surface nodes according to the time slot occupation table, perform matching operation on the network connectivity matrix and the time slot resource matrix, and generate an initial time slot allocation scheme suitable for a search and rescue scene and write the initial time slot allocation scheme into a resource scheduling unit. The time slot adjustment module is configured to write control information into a data frame header in a piggyback structure, generate a frame configuration table according to the sea electromagnetic propagation characteristics, use the frame configuration table for data framing processing, read the time slot occupation table to perform distributed time slot selection on the search and rescue nodes, establish an unmanned node queue state table based on the network connectivity matrix, calculate a time slot occupation dynamic adjustment coefficient according to the unmanned node queue state table, correct the allocation scheme according to the dynamic adjustment coefficient, write the corrected allocation scheme into a time slot resource scheduling unit, and generate a time slot resource scheduling strategy supporting unmanned cluster sea mobility.

[0014] In a third aspect, the present application provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the unmanned cluster cooperative ad hoc network time slot resource allocation method when executing the program.

[0015] In a fourth aspect, the present application provides a computer readable storage medium having a computer program stored thereon, wherein the computer program is executable on a processor to implement the steps of the unmanned cluster cooperative ad hoc network time slot resource allocation method.

[0016] In a fifth aspect, the present application provides a computer program product comprising a computer program / instruction, wherein the computer program / instruction is executable on a processor to implement the steps of the unmanned cluster cooperative ad hoc network time slot resource allocation method.

[0017] From the above technical solutions, the present application provides an unmanned cluster cooperative ad hoc network time slot resource allocation method and device, which effectively maps the requirements by innovatively designing a task configuration system, through a state machine and a resource matrix. A time slot allocation mechanism is constructed, combined with connectivity analysis and resource matching, to establish a reliable scheduling scheme. Dynamic optimization is introduced, through state monitoring and strategy adjustment, to ensure the adaptability of the allocation. The method effectively solves the deficiencies of traditional technologies in task configuration, time slot allocation, and dynamic adjustment, and provides technical support for unmanned cluster cooperation. BRIEF DESCRIPTION OF DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0019] Figure 1 The flowchart of the time slot resource allocation method for the unmanned cluster cooperative ad hoc network in the embodiments of the present application; Figure 2 The structural diagram of the time slot resource allocation device for the unmanned cluster cooperative ad hoc network in the embodiments of the present application; Figure 3 The structural diagram of the electronic device in the embodiments of the present application.

[0020] Reference signs: Electronic device 9600, central processor 9100, memory 9140, communication module 9110, input unit 9120, audio processor 9130, display 9160, power supply 9170, buffer memory 9141, application / function storage part 9142, data storage part 9143, driver program storage part 9144, antenna 9111, speaker 9131, microphone 9132. DETAILED DESCRIPTION

[0021] In order to make the purpose, technical solutions and advantages of the embodiments of the present application more clear, the following will combine the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are some embodiments of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0022] The acquisition, storage, use, processing and the like of data in the technical solutions of the present application all conform to the relevant provisions of laws and regulations.

[0023] Considering the problems in the prior art, the present application provides a time slot resource allocation method and device for an unmanned cluster cooperative ad hoc network. By innovatively designing a task configuration system, the effective mapping of requirements is realized through a state machine and a resource matrix. A time slot allocation mechanism is constructed, combined with connectivity analysis and resource matching, to establish a reliable scheduling scheme. Dynamic optimization is introduced, and through state monitoring and strategy adjustment, the adaptability of the allocation is ensured. This method effectively solves the deficiencies of traditional technologies in task configuration, time slot allocation and dynamic adjustment, and provides technical support for unmanned cluster cooperation.

[0024] In order to effectively solve the deficiencies of the traditional technology in task configuration, time slot allocation and dynamic adjustment, and provide technical support for unmanned cluster cooperation, an embodiment of an unmanned cluster cooperation ad hoc network time slot resource allocation method is provided, as shown in Figure 1 The unmanned cluster cooperation ad hoc network time slot resource allocation method specifically includes the following contents: Step S101: Set the cross-domain cluster node as a gateway control node, set the job group node as a water surface sensing node, set the unmanned platform node as an execution search and rescue node, generate a task configuration matrix, construct a task state machine according to the task configuration matrix, perform task deduction and calculation on search and rescue area coverage, target search tracking and wounded personnel transfer salvage based on the task state machine, convert task requirements into a time slot resource matrix, and write the time slot resource matrix into a resource allocation unit; This embodiment is developed for the typical mixed grouping scene of maritime disaster rescue. The precondition is that the node identity registration and link connection test have been completed, and the node list contains cross-domain cluster nodes that can be transmitted back by satellite / 5G, job group nodes that undertake air reconnaissance and water surface patrol, and a large number of unmanned platform nodes that perform net search, throw rescue and tow salvage. The first action of S101 is to map the node identifiers of different sources into three types of operational functions: the cross-domain cluster node is mapped to the gateway control node, which is used for satellite or 5G uplink and cross-domain command; the job group node is mapped to the water surface sensing node, which is responsible for reconnaissance, situation awareness and backbone relay; the unmanned platform node is mapped to the execution search and rescue node, which undertakes side-scan search, close-range tracking and salvage. Mapping is not simply renaming, but constructing an organization table containing control levels, command relationships and operation areas in the node management unit, and solidifying the level attributes according to the link capacity, energy surplus and load sensor type reported by the node, as the basis for subsequent message cycles and resource priorities.

[0025] After the mapping is completed, the embodiment generates a task configuration matrix around the bidirectional constraints of the task-communication. The specific process is as follows: first, a task planning table is constructed, task attributes (for example, “area coverage”, “target tracking”, “wounded evacuation and salvage”) are defined, message types (cross-domain instruction broadcast, group situation issuance, group member state reporting, heartbeat, synchronization), and communication cycle intervals are defined. Then, the matrix is expanded in a hierarchical dimension, the rows of the matrix correspond to node identifiers, the columns correspond to message types and cycle requirements, and the cells record the communication cost profile (expected message rate, average load, timing jitter tolerance) of the node in the current task phase. For example, the cross-domain broadcast of the gateway control node is set to the 1-5 Hz interval, the group internal situation issuance of the water surface sensing node is 0.5 Hz and is linearly associated with the group internal size, the state reporting of the search and rescue node is maintained at 1 Hz, and when a suspected fallen person is captured, a high-priority event frame is temporarily superimposed. The task configuration matrix obtained in this way not only depicts the message cycle, but also embeds priority labels and transmission reliability requirements, which facilitates subsequent mapping of task requirements to resources.

[0026] To map the task dynamics to the time-varying constraints of the communication resources, the embodiment constructs a task state machine according to the task configuration matrix. The state machine configuration table defines a state set (area coverage S_cov, target tracking S_trk, wounded evacuation S_med, etc.), state transition conditions (for example, sensing node target confidence exceeds threshold, rescue boat approaches target to safe radius, cross-domain node issues convergence instruction), and inference parameters (sea state level, group member density, target occurrence rate). The state machine is carried in the inference engine for Monte Carlo inference: taking time as the axis, under the given initial partition and node distribution of the task, repeatedly sampling target occurrence and movement, link fading, and node joining / leaving to drive state switching. On each sample path, the message cycle and priority are taken from the task configuration matrix according to the current state, and are converted into transmission requirements per unit time frame. The influence of sampled index parameters such as target density, node spacing, and link SNR on the results follows natural laws: increasing target density will increase sensing broadcast and tracking reporting frequency, increasing node spacing will increase routing hop count, and unit message will occupy more time slots to offset packet loss, and decreasing SNR will trigger retransmission budget increase, all of which are reflected through the in-frame acknowledgement and retransmission upper limit model.

[0027] After the task demand prediction table is derived, the embodiment converts the "business side dimension" into a "time slot side dimension". Here, a resource mapping table is constructed, and the entries include communication performance (delay tolerance, reliability level), data capacity (message length statistics, peak / average ratio), and quality indicators (allowed retransmission times, maximum jitter). The mapping process uses time frame parameters (frame length, time slot length, guard interval) and message length fields to calculate the nominal occupation of each type of message in a frame. When there is an acknowledgement / retransmission mechanism, the expected retransmission period is estimated based on the link quality parameters and added to the nominal occupation. To facilitate resource arrangement, the embodiment sums up the time slot requirements of each type of node and each type of message in each state, and arranges them in priority layers to obtain a time slot resource matrix. The rows correspond to nodes / hierarchies, the columns correspond to classified time slot pools in a time frame (such as control priority pool, general data pool, and recyclable synchronization pool), and the matrix elements are demand quotas and tolerance fluctuation amplitudes.

[0028] It should be emphasized that the derivation is not a one-time offline conclusion. In a maritime search and rescue scenario, the state changes from regional coverage to target tracking, which is often triggered by a perception node, which will immediately change the resource structure: the proportion of high-priority control frames and tracking reports increases, and periodic broadcasts are moderately compressed; during the process of transferring the wounded, the confirmation link between the search and rescue node and the gateway control node must be more stable, and the proportion of time slot pools reserved for confirmation frames is increased. The state machine counts these periodic structural changes during the derivation, and the time slot resource matrix forms two sets of quotas accordingly: "steady state + transition state". The resource allocation unit reduces the shock when switching strategies during the running period based on this.

[0029] In order to enable the resource matrix to be directly consumed by subsequent connectivity and occupation table matching operations, the embodiment embeds two types of keys in the matrix: hierarchy keys (gateway / perception / search and rescue) and message keys (instruction / situation / report / heartbeat / synchronization). After receiving the matrix, the resource allocation unit saves it as a versioned entry and marks the task state stamp. When the network connectivity matrix is read in subsequent S201 and subsequent steps, weighted matching can be performed on the same time slice, avoiding "resource demand but no connectivity bearing" fragmented idling. Considering that synchronization frames can be transferred to data time slots when there is no synchronization demand, the embodiment configures a transferable proportion for the synchronization area in the matrix, which can be temporarily supplemented to high-priority event streams during state switching.

[0030] On the level of effect, this embodiment solves three specific problems. First, the traditional static time slot planning is difficult to adjust with the evolution of tasks. This embodiment quantifies the message structure changes driven by task events in advance through state machine + deduction, and the time slot resource matrix has priority and retransmission budget, avoiding on-site passive traffic jam. Second, the reliability fluctuation caused by the drift of the electromagnetic environment at sea. By introducing link quality and retransmission upper limit in the mapping table, the natural SNR-packet loss-time slot overhead relationship is reflected in the quota, and resource scheduling is no longer dependent on the threshold of hitting the head. Third, the coexistence of layered communication across domains and within groups, the matrix distinguishes control and job flow with hierarchical keys, avoiding mutual compression of high-level broadcast and close-range tracking, and facilitating subsequent matching operations by layer weighting.

[0031] To help readers correspond the abstract process with the actual operation, two small segments are given. Segment one: in the morning, the visibility is poor, and the sensing node triggers two suspected targets in the southeast sector. The state machine transitions from S_cov to S_trk, the quota for tracking reporting and control confirmation in the matrix is adjusted upward, the gateway broadcast is reduced to the lower limit, and the resource allocation unit records the version and waits for the connection matrix matching. Segment two: the salvage boat successfully approaches, and enters S_med, the confirmation ratio between the search and rescue node and the gateway is further increased, the intra-group situation broadcast is maintained, but the cycle is synchronized to support the short burst flow of the salvage state reporting. In the two processes, the resource matrix is written into the resource allocation unit in chronological order, ready for joint decision with the time slot occupation table-connection matrix after S102, to complete the closed loop from "task-demand-time slot".

[0032] Step S102: divide the time frame into time synchronization area and data interaction area, write the time frame structure into each surface unmanned node control unit, read the buffer queue length of the adjacent search and rescue node, generate a time slot occupation table reflecting the communication state of the unmanned cluster, calculate the network connectivity matrix between the surface nodes according to the time slot occupation table, perform matching operation on the network connectivity matrix and the time slot resource matrix, and generate an initial time slot allocation scheme suitable for the search and rescue scene, and write the initial time slot allocation scheme into the resource scheduling unit; This embodiment develops around the link of "time frame structure-state acquisition-occupation table-connectivity matrix-matching with resource matrix-initial allocation scheme" in S102, and the scene is limited to maritime search and rescue formation: the gateway control node is responsible for cross-domain backhaul, the surface sensing node undertakes partition search and situation broadcast, and the search and rescue node frequently reports and triggers events in close-range maneuver. The last step S101 has generated a time slot resource matrix with priority and retransmission budget, and S102 will land this matrix to the actual time frame and executable allocation scheme.

[0033] The embodiment first constructs a time frame structure table, which includes the length of a synchronization frame, the length of a data frame, and a guard interval. Time is divided into time frames of fixed periods, each frame being composed of a time synchronization area (which can be given up when synchronization is not required) and a data interaction area. Each data time slot is subdivided into four segments: data transmission, data propagation guard interval, acknowledgement transmission, and acknowledgement propagation guard interval. To unify the behavior of multiple nodes, the time frame structure table and a frame division parameter table (frame format, number of time slots, time slot length) are distributed to each surface unmanned node control unit, enabling the nodes to make distributed selection and avoidance under a unified time base. Considering the time-varying nature of electromagnetic propagation at sea, the transferable proportion of the synchronization area is marked, and resource scheduling can be temporarily incorporated into the data interaction area in emergency situations.

[0034] After having a unified time base, the embodiment begins to collect neighborhood load. Each node listens to data packets from neighboring search and rescue nodes according to a piggyback frame header, reads the buffer_size field and the occupy_next bit, and maintains a neighbor state table (node address, queue length, time to live TTL) in conjunction. The update process follows the natural propagation and activity rules: if a neighbor has a message in the last frame, its TTL is reset to the initial value; if it is not seen for a number of frames, its TTL is decremented, and when it decreases to 0, it is considered to be out of range or temporarily inactive and is removed from the table. The resulting buffer queue length distribution is accumulated within a two-hop range, providing a basis for subsequent fairness and on-demand.

[0035] Based on the above listening results, the embodiment generates a time slot occupation table. The occupation table records the state according to the time slot index: idle, occupied, and busy, and additionally includes two key supplementary fields: OCPNXT and HOPCNT. OCPNXT identifies whether the next frame will continue to be unavailable, and HOPCNT records the number of hops from the source of the state to the local. In conflict determination, the "closer first" principle is adopted: if a two-hop node B announces the reservation of a certain time slot in the last frame, and the current frame receives actual data of another-hop node in the same time slot, the time slot is marked as occupied by C and HOPCNT is reset to 1, effectively resolving the cross-hop reservation conflict. For detecting that the same time slot is reserved by two or more nodes at the same time, the state is marked as busy to prevent further selection and cause a collision. Since the acknowledgement short frame and the data frame are in a closed loop in the same time slot, the embodiment considers the acknowledgement segment as an implicit occupation of the time slot, and does not list the time slot, but marks the retransmission probability in the state machine, facilitating consideration of the additional cost when matching the quota later.

[0036] The process of computing the network connectivity matrix is not just looking at physical adjacency. This embodiment extracts the activity and conflict of visible links from the slot occupancy table: if a pair of nodes has a stable, conflict-free receive-acknowledge sequence in the last W frames, then the edge is given a "strong connectivity" in the connectivity matrix; if it is visible but has been busy for a long time, it is marked as "weak connectivity"; if the TTL has been zeroed or only has a two-hop conflicting reservation trace, it is temporarily set to zero. The metrics of the connectivity matrix embody three natural constraints: neighbor priori (neighbors are better than distant neighbors), conflict penalty (the more busy, the lower the effective connectivity), and confirmation integrity (lack of confirmation is considered unreliable). To avoid falling into an overly pessimistic graph when the sea conditions deteriorate, this embodiment allows weak connectivity to accept a small amount of control class quota in the boundary scenario, which is used as an exploration path for expanding the route.

[0037] The matching of the connectivity matrix and the slot resource matrix is the core of S102. This embodiment constructs a matching strategy table containing resource requirements, connectivity status, and allocation weights. The weight reflects the business priority (from S101), the link strength (given by the connectivity matrix), and the conflict cost (statistical from the occupancy table). The matching operation follows the idea of hierarchical weighting: control class and upper layer protocol control frame are preferentially satisfied on strong connectivity edges; general data flow can overflow to weak connectivity when strong connectivity is insufficient, but the length of continuous time slots is reduced to reduce the risk of collision; recyclable synchronization area is injected into high priority flow in proportion in the event state. To define the operation relationship, this embodiment gives a weighted expression of the total weight: W(i,j,c)=α·P(c)+β·C(i,j)-γ·K(i,j), where P(c) is the priority of business category c, representing the urgency of the task side to this category of message; C(i,j) is the connectivity strength between nodes i and j, reflecting the reliable interaction degree in the past window; K(i,j) is the conflict cost, which comes from the busy count of i-j related time slots; α, β, γ are weight coefficients set in the system, used to weigh the business urgency, link reliability and collision risk in different sea conditions and task stages. The dimensions and physical meanings in this formula come from the objective properties of tasks and links, without introducing assumptions that violate the natural propagation law.

[0038] According to the above weight, this embodiment generates an initial time slot allocation scheme suitable for search and rescue scenarios in each time frame. The generation process is divided into two steps: first, for high-priority classes (upper-layer control, instructions, tracking reporting) on strong connectivity pairs, do an "integer packing" to preferentially select time slots that are currently idle and OCPNXT=0; then, for general data and heartbeat, load the remaining resources, and skip the time slots marked as busy. If the resource is tight, this embodiment scales the number of time slots for each node according to the proportion of the neighbor queue length, to avoid the saturation of a single node blocking the neighborhood. After scaling, the necessary confirmation budget will be converted into the reservation of adjacent available time slots to avoid the overlap of confirmation segments and other reservations.

[0039] The initial scheme is written into the resource scheduling unit, with a version number and a time stamp, for subsequent dynamic adjustment. To illustrate its actual behavior, two short examples are given: one, the southeast subzone has a sudden multi-target burst, the queue length of the sensing node rises rapidly, the occupancy table has a large number of busy increases in a number of low-numbered time slots, the connectivity matrix gives a high C value to the sensing-gateway edge, and the matching layer will track the report migration to the medium-high numbered free time slots and borrow part of the synchronization area; the second, the search and rescue node approaches the wounded, and needs more intensive confirmation in a short time. The initial scheme continuously arranges a number of short time slot pairs for each pair, which contains a data and acknowledgment closed loop, to ensure reliable feedback under near-distance water surface fast maneuvering.

[0040] From the technical problems and effects, the embodiment solves three points: first, the abstract task resource demand is projected onto the time frame and the connectivity graph, eliminating the gap between "quota allocation and no bearing"; second, the multi-source marker of the occupancy table and the hop number are used to resolve reservation conflicts, reducing collisions in a distributed environment; third, priority, connectivity strength and conflict cost are considered in parallel in matching, so that the initial scheme not only conforms to the urgency of search and rescue business but also does not violate the objective constraints of marine wireless propagation, leaving a margin for subsequent dynamic adjustment based on queue state.

[0041] Step S103: write the control information in the back-mounted structure into the data frame header, generate a framing configuration table according to the marine electromagnetic propagation characteristics, use the framing configuration table for data framing processing, read the time slot occupancy table for distributed time slot selection of the search and rescue node, establish an unmanned node queue state table based on the network connectivity matrix, calculate a time slot occupancy dynamic adjustment coefficient according to the unmanned node queue state table, modify the allocation scheme according to the dynamic adjustment coefficient, write the modified allocation scheme into the time slot resource scheduling unit, and generate a time slot resource scheduling strategy that supports unmanned cluster maritime maneuvering.

[0042] This embodiment focuses on S103, and "backs up" the network side control information into the frame header, so that control and business are co-loaded and propagated. In the water surface search and rescue formation with variable sea conditions and fluctuating links, framing processing, distributed time slot selection and dynamic modification according to queue state are completed. The premise is that S102 has obtained the initial time slot allocation scheme and the connectivity matrix, and each node locally maintains the time slot occupancy table and the neighbor state table. The goal is to convert the initial scheme into a scheduling strategy that can be executed every frame, and to maintain the stability and on-demand of resource allocation under disturbances such as maneuvering, network entry, and link fading.

[0043] The embodiment first defines a special frame header table in the data processing unit, Type is used for packet type distinction, occupy_next is used for cross-frame reservation release, buffer_size carries the total length of the node to-be-sent queue, and the reserved bit is aligned according to the protocol. The piggybacked writing of control information follows two constraints, one is that the upper-layer service fragmentation boundary cannot be changed, and the other is that the frame header length is constant so that the adjacent node can be parsed with a fixed offset. Considering the sea multipath and non-line-of-sight reflection, the embodiment generates a framing configuration table according to the wireless link statistics under the combination of ship height, sea state level and carrier frequency, gives the matching range of the synchronous frame, data frame and guard interval in a frame, and confirms that the propagation protection required by the segment is reserved according to the upper bound of link delay jitter; when there is no synchronization demand, the configuration table marks the proportion of the transferable synchronization area as an elastic resource pool for burst events. The framing configuration table is issued to each node, and the subsequent framing processing frames the upper-layer message and arranges the confirmation timing according to the table, so that the confirmation is returned in a closed loop in the same time slot, reducing the cross-slot interference.

[0044] In the distributed time slot selection stage, the embodiment reads the free / occupied / busy markers of the local time slot occupation table and the OCPNXT and HOPCNT side marks, and performs secondary decision on the initial allocation scheme issued in S102. The logic path is: first, filter out the busy and OCPNXT=1 time slots to avoid conflicts and cross-frame reservations; then, for the occupied time slots from two hops, give way when there is one-hop alternative free selection in the current frame, and follow the closer priority; finally, fill the target number of sending time slots according to the priority sequence of the service category. For event class reporting, if the initial scheme allocation is insufficient and the configuration table allows the use of the synchronization area, then additional short time slots are added within the allowed proportion, and the length depends on the total overhead of the event frame and the confirmation. The distributed selection does not depend on central coordination, but achieves "visible fairness" through local listening and protocol control bits, and the nodes converge due to the consensus of the same occupation table and the initial scheme.

[0045] In order to make the dynamic correction have objective basis, the embodiment establishes an unmanned node queue state table based on the network connection matrix, and counts the buffer_size, TTL and the sending success / failure count of the near T frames of the one-hop neighbor. The state table aggregates the information of three directions according to the node identifier: communication demand intensity (queue length and its growth rate), bearing possibility (connection strength and confirmation integrity), and local congestion risk (busy ratio of the time slot related to the node). The internal relationship between these indicators is natural: the longer and faster the queue grows, the higher the short-time sending demand is; the stronger the connection is, the easier it is to convert the allocated time slot into effective throughput under the current topology and sea conditions; and the higher the busy ratio is, the greater the conflict risk in the same neighborhood is, which needs to be inhibited. In order to map the above relationship to the resource layer, the embodiment calculates the time slot occupation dynamic adjustment coefficient δ of each node, adopts a set of weighted and normalized construction: δ = f(ρd, ρc, ρk), wherein ρd represents the demand intensity dimension, such as the normalized quantity of the buffer_size of the node and its time difference; ρc represents the connection dimension, such as the weighted mean of the edge weight of the connection matrix; and ρk represents the conflict cost dimension, such as the busy ratio of the occupation table statistics. The function f is monotonically increasing with respect to ρd and ρc and decreasing with respect to ρk, which ensures that the nodes with “demand and accessibility” are moderately expanded, and the “high conflict” neighborhood is moderately contracted. The physical meaning of the parameters is clear: ρd describes the business pressure, ρc describes the realizability, and ρk describes the risk of shared medium, without involving the head-in-the-clouds factor that is out of the scene.

[0046] Based on δ, the embodiment corrects the initial allocation scheme. The correction is developed in two layers: in the node, multiply the time slot share of the node itself according to the priority sequence of the service category, re-allocate the time slot share in the category, and ensure the basic disc of the upper layer protocol control frame and the key tracking report; and in the neighborhood, relatively stretch the total time slot budget, and the principle is not to break through the maximum load of the frame configuration table and the lower bound of confirmation protection. If there are multiple high-δ nodes in the neighborhood, adopt proportional division and reserve a small amount of balanced time slots to prevent starvation; if the δ of a node maintains a low value for a plurality of frames and the TTL tends to zero, release its cross-frame reservation, shorten the reservation time of OCPNXT, and make the resource quickly flow back. The correction output is still expressed in the form of a specific time slot index, and is written into the time slot resource scheduling unit with a version stamp for direct execution in the next frame.

[0047] To test the mechanism under the behavior of maneuvering and bursting, two fragments close to rescue are listed in this embodiment. Fragment one, the perception node detects a new target in the southeast sector, and the buffer_size jumps up in a short time. The connectivity matrix shows that it has a strong edge with the gateway, and the ρd and ρc are rising synchronously, and the δ is lifted, and after correction, it adds time slots to the middle number of free slots and moves part of the synchronization area; since the same domain busy ratio has not risen, ρk remains, and expansion can be converted into effective reporting. Fragment two, the salvage boat approaches the swimmer at high speed, and the body shielding and the wave surface cause the confirmation failure in the time slot, the connectivity decreases for a short time, ρc drops, and δ is suppressed, and the scheduling splits the long time slot into multiple short slots and aligns with the confirmation, reduces the collision duration, and waits for the connectivity to be repaired before refilling.

[0048] The key problems solved by this embodiment are reflected in three aspects. First, the queue length and reservation intention are propagated in the actual service packet through the piggyback frame header, avoiding additional control overhead and ensuring the consistency of perception-scheduling. Second, the frame configuration table combines the delay and jitter boundary of electromagnetic propagation at sea to confirm the timing and protection interval, which are no longer set empirically, reducing cross-gap reply failure. Third, the three-dimensional feedback is formed based on the queue state table and dynamic coefficient δ, which is oriented to "demand-availability-conflict". The modified scheme suppresses local collision while ensuring the completion of key flow, maintaining the communication resilience of the rescue formation during maneuvering. The strategy written into the time slot resource scheduling unit has a timestamp and a state label, which facilitates subsequent steps to continue adjustment based on version rollback and window statistics without destroying the established distributed consensus.

[0049] As can be seen from the above description, the time slot resource allocation method for unmanned cluster cooperative self-organizing network provided by the embodiments of the present application can effectively map the demand by innovatively designing the task configuration system, through the state machine and the resource matrix. The time slot allocation mechanism is constructed, combined with connectivity analysis and resource matching, to establish a reliable scheduling scheme. Dynamic optimization is introduced to ensure the adaptability of the allocation through state monitoring and strategy adjustment. This method effectively solves the deficiencies of traditional technologies in task configuration, time slot allocation and dynamic adjustment, and provides technical support for unmanned cluster cooperation.

[0050] In an embodiment of the time slot resource allocation method for unmanned cluster cooperative self-organizing network of the present application, the following contents can also be specifically included: Step S201: constructing a cluster organization table containing control levels, command relationships and operation areas, mapping the cross-domain cluster node identifier to the gateway control node, mapping the operation group node identifier to the water surface perception node, mapping the unmanned platform node identifier to the execution search and rescue node, and writing the cluster organization table into the node management unit; Step S202: Construct a task planning table containing task attributes, message types, and communication cycles, divide node communication levels according to the cluster organization table, generate a task configuration matrix containing node identifiers, level attributes, and communication requirements, and write the task configuration matrix into the task management unit.

[0051] This embodiment is developed around S201-S202, and the goal is to convert the loose node list into two types of core data structures that can be consumed by the task-driven allocator: the cluster organization table and the task configuration matrix. The scenario continues the sea rescue organization: the cross-domain cluster nodes undertake satellite / 5G backhaul and cross-domain command; the operation group nodes are oriented to sea and air sensing and partitioned relay; and the unmanned platform nodes are responsible for close-range search, approach, and salvage. The upper resource allocation unit will subsequently read the level, command relationship, and message load from these structures to form an initial scheme matching the time slot and connectivity graph.

[0052] This embodiment first collects node basic metadata at S201, which comes from the report at the time of network entry and link testing, and the fields include hardware platform, wireless standard, carrier frequency and bandwidth, antenna height, energy reserve, sensor / load type, location, and operation sector. A cluster organization table containing control levels, command relationships, and operation areas is constructed, and the table header is divided into three groups of keys: level keys (group leader / group leader / group member), relationship keys (uplink command object, downlink jurisdiction range, and redundant takeover person), and spatial keys (geographical polygon or grid index). The mapping rule is not statically specified by name, but a capability threshold and responsibility matching is introduced: those with cross-domain links and stability passing the threshold are mapped to gateway control nodes; those with multi-source sensing load and relay power margin are mapped to water surface sensing nodes; and the remaining operation function-based nodes are entered into the search and rescue nodes. To avoid single-point failure, the command relationship allows one-to-many and redundant connection, and the organization table configures a backup group leader field for each group leader, which records the version and timestamp when written into the node management unit, facilitating subsequent reconstruction and backtracking.

[0053] There is an easily overlooked detail in the mapping: the operation area and the command relationship need to be coupled. Sea rescue is usually divided by sector, and if the effective coverage of a sensing node is inconsistent with the responsibility sector of its superior group leader, it is easy to cause cross-border situation broadcast. This embodiment performs consistency checking on the spatial key when generating the organization table, and if cross-border is found, it preferentially adjusts the intra-group boundary or triggers command relationship reconnection. For dynamically joined nodes, the read mapping process executed by S201 adds a "cold start listening" stage: the node first only listens to a frame, collects intra-group heartbeat and synchronization, and confirms reachability before writing into the organization table, avoiding introducing dithering up-down level binding in the topology critical state.

[0054] After the solidification of the organizational structure is completed, S202 turns to the generation of the task planning table and the task configuration matrix. The task planning table is oriented to the business layer, and the fields include task attributes (coverage / tracking / transit / synchronization, etc.), message types (cross-domain broadcast, group member situation, state reporting, heartbeat, synchronization), communication periods and their upper and lower limits, message length statistics, reliability level, and time delay tolerance. For example, the group leader broadcasts the group member situation at 0.5 Hz with the cost linearly superimposed according to the number of group members, the state reporting is at 1 Hz, and the heartbeat and synchronization are short frame period broadcasts. After the planning table is established, the nodes are divided into communication levels according to the hierarchical relationship in the cluster organization table in this embodiment, and further consideration is given to the actual bearing influence of the differences in group size, routing hop number, and link quality on the period and reliability: for branches with large group size and high hop number, even if the nominal period of the task side is consistent, time slot redundancy corresponding to the retransmission probability needs to be reserved in the matrix.

[0055] The construction of the task configuration matrix is the key of S202, which compresses the three-dimensional elements of “node x message type x communication period / reliability” into a calculable communication demand image. The matrix row is indexed by the node identifier, the column is expanded according to the message type, and the cell records four items: expected message rate, single frame average byte, reliability level, and priority label; the auxiliary column records the routing hop number estimate and the confirmation budget. In order to make the matrix be used by the subsequent matching, this embodiment externalizes the “hierarchical attributes” to the metadata of the matrix, so that the resource scheduling can allocate the quota according to the level. Moreover, the matrix is connected with the task state label, which indicates the switching rules of the cell in the coverage, tracking, transit, and other states, facilitating the subsequent state machine to switch the readings according to the stage.

[0056] For the determination of the matrix values, this embodiment adopts task-link joint reasoning. First, the nominal period is given by the task planning, and then the average hop number and link reliability of each type of message are estimated according to the spatial key in the organization table and the historical routing, and the expected occupation of confirmation and retransmission is converted. In order to give clear dimension conversion, a calculation relationship is listed: S_req = r_msg·L_msg / τ_slot + p_rt·Δ_ack, where S_req represents the nominal time slot demand of the message in the unit time frame, r_msg is the message rate of the message type, L_msg is the average byte number, τ_slot is the number of bytes that can be carried by a single time slot, p_rt is the expected number of retransmission triggers, and Δ_ack is the additional overhead of confirmation and protection to the time slot. r_msg, L_msg come from the task planning table, τ_slot comes from the frame parameter, p_rt is given by the combination of the connectivity quality and the hop number estimate, and Δ_ack depends on the confirmation embedded timing and the protection interval. Each parameter of the relationship has physical meaning and follows the constraints of maritime wireless propagation and MAC confirmation mechanism.

[0057] To avoid the matrix from deviating from the actual topology, the embodiment projects the redundant command relationship in the organization table into a resource constraint: once the primary group leader fails, after the backup group leader switches, the "in-group situation" column of the group members is automatically redirected to the new superior, the period remains unchanged but the hop count and the confirmation budget are recalculated according to the new path. In the high-variability scenario of disaster rescue, this way of reflecting command switching in the communication budget can reduce the jitter in the subsequent time slot allocation phase. Another detail is the handling of synchronization and heartbeat. The embodiment sets a transferable flag and an upper limit proportion for the "synchronization" column in the matrix, so as to facilitate borrowing by the resource scheduling in emergency.

[0058] Two small segments are given to illustrate how the matrix reflects task differences. First, in the daytime, the visibility is good, coverage is the main task, and the "in-group situation" column of the water surface sensing node grows linearly according to the in-group size. The search and rescue node only keeps reporting the state at 1 Hz, and the priority label of the matrix is more biased towards upper-layer control and situation broadcast. Second, in the night, a falling signal suddenly occurs, the state switches to tracking, the search and rescue node activates high priority in the "event report" column, the message rate is increased, and the matrix increases the Δ_ack of the confirmation budget to resist the night link fluctuation. The matrix keeps the minimum overhead of heartbeat and synchronization in both states, but marks the transferable proportion of synchronization in the event state.

[0059] From the technical problems and effects, the embodiment integrates "who commands who", "where to work", and "what message to send and at what rhythm" into one, avoiding the drawbacks of disconnection between organizational relationships and communication profiles in traditional methods; by coupling the message period and the link reliability into quantifiable time slot demand, the subsequent matching and scheduling stages no longer rely on empirical coefficients; by making the redundant command and transferable synchronization explicit at the matrix level, when a node fails or a state suddenly occurs, the mapping from task to resource does not need to be rebuilt, but only needs to be switched to the adjacent version, reducing the interruption time. Finally, the cluster organization table and the task configuration matrix are written into the node management unit and the task management unit, respectively, to be solidified with version number, time stamp, and state label, waiting for the resource allocation and time slot scheduling unit to complete the landing from task to resource in S102-S103.

[0060] In an embodiment of the time slot resource allocation method of the unmanned cluster cooperative ad hoc network in the application, the following contents can be specifically included: Step S301: constructing a state machine configuration table containing state transition conditions, task trigger conditions, and deduction parameters, generating a task state machine according to the task configuration matrix, performing Monte Carlo deduction on search area coverage, target search tracking, and casualty evacuation and salvage based on the state machine, and generating a task demand prediction table, and writing the task demand prediction table into a deduction engine; Step S302: Constructing a resource mapping table containing communication performance, data capacity, and quality indicators, calculating communication resource requirements based on the task requirement prediction table, generating a time slot resource matrix, and writing the time slot resource matrix into the resource allocation unit according to priority.

[0061] This embodiment is based on S301-S302, and the purpose is to project the state evolution of the task side into a calculable way to the communication resource side, forming an executable input at the time slot level. It is known that S201-S202 gives the task configuration matrix (node identification, hierarchical attribute, message type, and periodic demand), and S102-S103 will match and schedule the time slot resource matrix output by this step. The environment of maritime search and rescue is unstable and has obvious event-driven characteristics, and the state switching is not a linear process, so it is necessary to use state machines and Monte Carlo reasoning to describe multiple possible scenario paths to avoid resource mismatch caused by single path assumptions.

[0062] This embodiment first constructs a state machine configuration table at S301, with three core fields in the table: state transition conditions, task trigger conditions, and reasoning parameters. The state set includes search area coverage S_cov, target search and tracking S_trk, medical evacuation and salvage S_med, and auxiliary states such as idle and collection. The state transition conditions reflect natural causality, for example, "if the confidence of the perception node on the target in a certain sector exceeds the threshold and is stable within a continuous preset number of windows, then S_cov→S_trk", and "if the distance between the search and rescue node and the target is less than the safety radius and the upper level authorizes, then S_trk→S_med". The task trigger conditions are driven by upper-level commands or external events, such as collection, re-partition search, and temporary no-fly zone updates by cross-domain gateways. The reasoning parameters include sea state level (which affects channel jitter and motion speed), node density, target occurrence rate, link SNR statistics, and node entry / exit network probability, all of which are consistent with the physical processes in the scene and do not introduce assumptions that contradict the laws of propagation or motion.

[0063] Based on the configuration table, the embodiment generates a task state machine according to the task configuration matrix. Each state of the state machine is bound to a group of communication image pointers, which point to the message period, length, priority and reliability level in the corresponding state in the matrix. Then, Monte Carlo deduction is entered: the time axis is discretized into frames, a large number of sample path simulations are repeatedly performed, and the target appearance and movement, link quality fluctuation, neighbor topology change and command triggering time are randomly sampled in each path; the state switching is driven according to the state transition condition, and the bound communication image is read on each frame slice to be converted into the “demand slice” of the current frame. In order to make the deduction closer to the actual situation, the embodiment embeds the linkage between the routing hop number and the confirmation overhead in the sample: the sparse topology or the deterioration of the sea conditions will increase the average hop number and the packet loss, and then increase the confirmation and retransmission budget; on the contrary, when the dense grouping and the sea conditions are stable, the demand slice tends to the nominal value. After the accumulation of multiple sample paths, the task demand prediction table is obtained, which includes the time slice, the state label, the message rate distribution of each message type, the average length, the expected retransmission number and the priority interval. The table is written into the deduction engine as the direct input of subsequent resource mapping.

[0064] Turning to S302, the embodiment constructs a resource mapping table to establish correspondence between communication performance, data capacity, quality index and time frame parameters. The communication performance includes delay tolerance, reliability level, confirmation timing model; the data capacity includes length statistics (mean / peak) and arrival burst degree of each message; the quality index includes the maximum number of allowed retransmissions and acceptable jitter. The time frame parameters are given by the system (frame length, time slot length, guard interval, confirmation segment proportion). The core of the mapping process is to convert “the business side demand of each type of message in each state” into “the time slot demand of the unit time frame”. The embodiment adopts component combination: the nominal load is determined by the message rate and the length, the confirmation and retransmission overhead is determined by the link quality and the hop number estimation, and the guard interval depends on the upper bound of the sea propagation delay jitter. In order to facilitate review, the embodiment describes the conversion with a relationship: S_req(u, c, t) = r(u, c, t) · L(u, c) / τ_slot + E_rt(u, c, t) · Δ_ack, where S_req represents the unit time frame time slot demand of node u for class c message at time t; r(u, c, t) is the message rate obtained by deduction; L(u, c) is the average byte of the message of this type; τ_slot is the number of bytes that can be carried by a single time slot (determined by the time slot length and the modulation and coding); E_rt(u, c, t) is the expected number of retransmissions at time t (obtained by connectivity quality, hop number and SNR statistics); Δ_ack is the equivalent time slot overhead of single confirmation and protection. The physical meaning of each parameter is consistent before and after, which meets the natural relationship of “the greater the load, the worse the link, the more stringent the confirmation, and the more occupied”.

[0065] In order to let the time slot resource matrix reflect the demand strength and express the priority and the deliverability, the embodiment is divided into layers and categories in the aggregation of S_req. The first layer is the control class and the upper layer protocol control frame, which supports multi-hop routing and command achievement, and is marked as non-degradable quota; the second layer is the mission critical flow (tracking reporting, salvage process confirmation), which can borrow the synchronization area in emergency state; the third layer is the general data and situation broadcast, which has a scalable label. The resource mapping table gives the priority sequence and the degradation strategy of each layer, for example, when the S_req sum exceeds the time slots available in the current frame, the continuous occupation of the third layer is reduced first to maintain the integrity of the confirmation closed loop. After the aggregation is completed, the time slot resource matrix is obtained, the matrix row is indexed by the node or the layer, the column is expanded according to the classified time slot pool (control pool, critical pool, general pool, deliverable synchronization pool), the cell records the quota and the fluctuation upper limit, and is attached with a state label for S102 matching.

[0066] It should be pointed out that the coupling of deduction and mapping avoids the distortion problem of "static cycle → rigid quota". Two segments are given to illustrate the internal logic. In the morning, the wind is small and the target is sparse, the S_cov residence time is long in the deduction, the message rate is low and the E_rt is close to zero, the matrix allocation is mainly for situation broadcast and basic heartbeat, and the critical pool occupies a small proportion; in the evening, the wind increases and the thermal noise rises, S_trk is frequently triggered, E_rt moves up, the matrix automatically increases the confirmation budget in the critical pool, part of the synchronization deliverable label is activated, and the control pool remains unchanged. Both segments follow the natural chain of wireless link quality → retransmission budget → time slot occupation, and do not produce allocation that contradicts the physical conditions.

[0067] From the engineering effect, the embodiment explicitly shows the task trigger and state transition through the state machine configuration table, covers multiple feasible paths by using Monte Carlo deduction, and the output task demand prediction table has time and uncertainty semantics; by means of the resource mapping table, the performance, capacity and quality index are unified to the time slot dimension, forming a time slot resource matrix with priority and degradation strategy. In this way, the subsequent connectivity matching and distributed selection are no longer trimmed by experience, but are landed in the order of "demand → reachability → risk", and can switch to the matrix version of the adjacent state when the node enters the network or the link deteriorates, reducing the cost of re-planning. Finally, the time slot resource matrix is written into the resource allocation unit according to the priority order, and records the version and state stamp, providing a solid upstream basis for the matching of S102 and the dynamic correction of S103.

[0068] In an embodiment of the time slot resource allocation method of the unmanned cluster cooperative ad hoc network in the application, the following contents can also be specifically included: Step S401: constructing a time frame structure table containing the length of a synchronization frame, the length of a data frame, and a guard interval, dividing the time frame into a time synchronization area and a data interaction area according to a fixed interval, generating a frame division parameter table containing a frame format, a time slot number, and a time slot length, and writing the frame division parameter table into each water surface unmanned node control unit; Step S402: constructing a neighbor state table containing a node address, a queue length, and a survival time, reading the length information of a cache queue of a neighboring search and rescue node, generating a time slot occupation table containing a time slot state, reservation information, and hop number information, and writing the time slot occupation table into a state management unit.

[0069] The embodiment S401-S402 is directed to frame division and neighbor state collection of a maritime search and rescue formation under a medium wave / microwave link, and the goal is to bring a unified physical time base and load observation to each water surface unmanned node, so that subsequent matching and scheduling have a consistent reference surface under distributed conditions. The prelude S201-S202 has specified the hierarchy and message cycle, and S301-S302 has given the time slot dimension of resource mapping, and this step encapsulates the time frame structure and neighbor information into two types of basic tables and writes them into the node side control / state unit.

[0070] The embodiment first constructs a time frame structure table in S401, and the fields are the length of a synchronization frame Ts, the length of a data frame Td, and a guard interval Tg. The time axis is divided into continuous time frames by a fixed interval, and each frame is composed of a time synchronization area and a data interaction area; each data time slot in the data interaction area is subdivided into four segments: data transmission, data propagation protection, acknowledgement transmission, and acknowledgement propagation protection. The guard interval Tg is set with reference to the maximum round-trip delay and the upper bound of multipath spread of sea surface electromagnetic propagation, to ensure that the same gap acknowledgement can be reliably received after the physical layer falls back, and not to leak across the gap. According to different sea state levels (smooth, swell, and wind wave), the embodiment gives a recommended combination interval of Ts:Td:Tg, and marks the transferable proportion of the synchronization area, which is converted into a recoverable data gap to undertake a burst task when the synchronization load is low. Then a frame division parameter table is generated, which unifies the frame format, the time slot number Nslot, and the time slot length τslot, to ensure that the nodes perform analysis and reservation release of the control information on the same time base. To avoid the beat drift caused by the deployment of heterogeneous bandwidth, the frame division parameter table contains a modulation and coding indication and a payload threshold, to map τslot and the number of bytes that can be carried, and after being written into each water surface unmanned node control unit, the frame alignment and gap counting are completed by the MAC timer thereof.

[0071] The frame parameter distribution is not equal to readiness. In this embodiment, a local self-check is implemented at each node side: the radio frequency front-end time deviation is read, the last frame received synchronization broadcast timestamp is compared, and if the deviation exceeds the threshold, the synchronization area is temporarily expanded by several sub-slots to complete the re-alignment. The reason is that the change of sea conditions or platform maneuvering will cause the local oscillator frequency deviation to accumulate, and the self-healing window is left out in advance to reduce the subsequent collision. For gateway control nodes carrying satellites / 5G backhaul, a cross-domain bridging flag is added to the frame parameter table, indicating that the node needs to reserve part of the fixed number of time slots to interface the uplink backhaul, avoiding the impact of intra-group occupation on cross-domain commands.

[0072] In S402, this embodiment constructs a neighbor state table at the node side, including neighbor node address (MAC), queue length buffer_size, time to live TTL, and optional last received time and source level. Each node receives data frames or upper layer protocol control frames, parses buffer_size and occupy_next (cross-frame reservation intention) from the dedicated frame header, refreshes the queue length of the neighbor to the local table, and resets TTL to the preset initial value. If a neighbor does not receive any packet for a number of consecutive frames, this embodiment decays its TTL by frame, and deletes the item when it falls to zero, avoiding the impact of obsolete information on local time slot selection probability. The reason for using TTL instead of a simple timestamp is that intermittent links in wide-area ad hoc networks can be caused by wave blocking or attitude changes. TTL provides a tolerance window for "may still be present but not received", so that the scheduling is not excessively contracted due to short shadow areas.

[0073] On the basis of stable update of neighbor state, this embodiment generates a time slot occupation table. Each node maintains a state table indexed by slot number for the current frame, including three types of states: idle, occupied, and busy, with supporting records of reservation information OCPNXT and hop count information HOPCNT. The definition of occupation follows the observation priority principle: if a local or one-hop neighbor successfully transmits data in a slot and receives an acknowledgment, the slot is marked as occupied and HOPCNT=1; if two or more neighbors declare reservation in the same slot, the slot is marked as busy and new reservations are prohibited; if only a two-hop neighbor declared OCPNXT=1 in the last frame and a one-hop neighbor appears actual transmission in the current frame, the two-hop reservation is overwritten with the more recent priority principle, and HOPCNT is reset to 1. OCPNXT records the cross-frame reservation intention, which is used to pre-occupy the unavailable set at the start of the next frame, reducing the collision probability of newly connected nodes. The rolling update of the occupation table occurs at the frame boundary: first, the state is preset according to OCPNXT of the last frame, then it is corrected according to the transmission and reception events in the frame, and finally it is written to the state management unit at the end of the frame for S102 / S103 to read.

[0074] To let the occupancy table reflect the natural constraints of physical propagation and MAC acknowledgement, this embodiment takes the acknowledgement short frame as "endogenous occupancy" sharing the same slot with the data frame, without setting up a separate acknowledgement slot; but attaches an acknowledgement result flag in the status field, for subsequent estimation of the reliability of the slot. If a peer appears in multiple frames in the pattern of "data success - acknowledgement missing", it will also generate a "soft collision" count locally, even if it is not marked as busy, for the matching operator to reduce the priority of the slot. The basis for this approach is that the sea reflection path may make the effective SNR of the acknowledgement segment lower than that of the data segment, and in the short term it can still be compensated by retransmission, but it is not appropriate to continue to concentrate the loading of key flows.

[0075] This embodiment further illustrates the logical association of the occupancy table and the neighbor state. The buffer_size of the neighbor reflects its sending pressure. If the buffer_size of multiple neighbors in an area is synchronously lifted, and the busy count of the same numbered slot on the occupancy table increases, it can be judged that the area is facing the effect of contention amplification, and the node should tend to skip these high contention slots when making reservations, and disperse the reservations to the medium and high numbered interval, to avoid the vicious cycle of "crowded - collision - retransmission - more crowded". Conversely, when the TTL is generally reduced and the busy count is rare, it indicates that the neighborhood is sparse or the link is good, and this embodiment allows the node to be given a more continuous time slot segment without breaking the protection interval and the acknowledgement closed loop, to improve the consistency of throughput.

[0076] To show the details in the landing process, two segments are given. Segment one, the sea conditions are stable in the morning, and the group leader node broadcasts the situation to the group members at a rhythm of 0.5 Hz. The node reads from the frame header that the occupy_next of the group leader is 0, indicating that it does not retain cross-frame occupancy; the occupancy table therefore presents intermittent occupancy stripes in low numbered slots, and the surrounding nodes prefer to fill in the blanks when making reservations, to avoid squeezing continuous acknowledgement chains. Segment two, the wind and waves rise in the evening, and the search and rescue node continuously reports suspected targets, buffer_size jumps, and to ensure acknowledgement of the return, it sets OCPNXT=1 in several consecutive slots, and the neighborhood nodes accordingly pre-empt these slots as unusable, and the contention pressure is transferred to the unreserved interval, and the collision probability decreases according to the natural law.

[0077] From the perspective of technical problems and effects, the embodiment parameterizes the length, structure and protection boundary of the time frame at S401, solves the inconsistency problem of reservations caused by the lack of a unified time base in multiple nodes; at S402, through the neighbor state table driven by the piggyback frame header and the occupation table containing OCPNXT / HOPCNT, the factual problems of how to reliably perceive "who is using", "who will use" and "how far away" in a distributed environment are solved. The combination of the occupation table and the neighbor state provides a verifiable underlying evidence chain for subsequent connectivity matrix calculation and matching, so that resource scheduling can converge without relying on central broadcast, and reasonable avoidance and bearing are maintained when the sea is mobile, nodes join or exit and sea conditions fluctuate. The above two types of tables are written into the node control unit and the state management unit with version stamps and time markers, facilitating cross-frame association and backtracking in subsequent steps.

[0078] In an embodiment of the time slot resource allocation method of the unmanned cluster cooperative ad hoc network of the present application, the following contents can also be specifically included: Step S501: constructing a network topology table containing node identification, link state and time slot state, calculating the connectivity relationship between water surface nodes according to the time slot occupation table, generating a network connectivity matrix, and writing the network connectivity matrix into a topology management unit; Step S502: constructing a matching strategy table containing resource demand, connectivity state and allocation weight, performing weighted matching operation on the network connectivity matrix and the time slot resource matrix to generate an initial time slot allocation scheme, and writing the initial time slot allocation scheme into a resource scheduling unit.

[0079] The embodiment expands around S501-S502, and weaves the occupation facts maintained by the node side and the quota demand of the task side into a set of executable initial time slot allocation output. The prelude S401-S402 has issued the framing parameters and formed the time slot occupation table, S301-S302 has given the time slot resource matrix layered by state and service category, S102 has a rudiment of matching logic, and this step further explicitizes the topology reachability and completes the calculation closed loop of weighted matching based thereon.

[0080] The embodiment first constructs a network topology table at S501, the core fields being node identification, link state and time slot state. The node identification comes from the node management unit, the link state is derived from the transceiving record in the near W frame, the confirmation success rate and RSSI / SNR statistics, and the time slot state is directly taken from the local time slot occupation table, including free / occupied / busy and its cross-frame reservation OCPNXT, hop count HOPCNT. Two constraints are introduced during construction: one is to weaken the "visibility" obtained only by energy detection by taking the complete successful transceiving pair as strong evidence; the other is to set a soft penalty for the confirmation loss caused by multipath, and not to directly reduce a single loss to incommunicability. Then the connectivity between the water surface nodes is calculated, the method being to take the "available gap intersection" of each pair of nodes in the window as the basic quantity of bearable degree, and then to reduce it by the conflict factors of busy rate and OCPNXT coverage rate; for pairs that only have two-hop reservation without one-hop actual flow, mark them as weakly connected to prevent mistaking "will use in the future" as "available now". Finally, the target network connectivity matrix is obtained, the element Cij reflecting the reachable strength between nodes i and j under the current frame structure and occupation fact, and the matrix is timestamped and window parameter written into the topology management unit for calling by the matching stage and subsequent dynamic adjustment.

[0081] The generation of the connectivity matrix is not simply topology graph extraction, but also contains timing judgment consistent with the MAC behavior. In the embodiment, the continuous "busy" on the same numbered time slot is regarded as a neighborhood preemption hotspot, and local retreat is applied to the Cij of adjacent edges; for the pair with long-term OCPNXT=1 on the fixed numbered gap, it is regarded as a cross-frame reserved channel, and Cij is weighted in the gap neighborhood, but will not spread to the full gap domain, so as to avoid false amplification. For example, the group leader and three group members form a stable "broadcast-confirmation" closed loop in the middle numbered gap, so the Cij of these pairs of edges on these gaps is high; if another search and rescue node repeatedly reserves the same numbered gap but collides, then the Cij between it and its neighbors decreases in that gap, but does not affect its potential reachability in high-numbered free gaps. Such processing makes C both "number sensitive" and retains the overall bearing view of the node pair.

[0082] At S502, the embodiment constructs a matching policy table, with fields for resource requirements (quotas and priorities from the time-slot resource matrix), connectivity status (edge strengths and hotspot / reserve flags from C), and allocation weights (for multi-factor tradeoff). The matching goal is to place high-priority flows into high-reachable low-collision edges and slot numbers, ensuring control and critical flow confirmation loop completion, with scalable space reserved for general flows. To avoid abstract description, the embodiment defines a total weight value W(i,j,c,s) for each candidate allocation, representing the attractiveness of carrying traffic of class c on node pair (i,j) in time slot s. The weight is composed of three things: task-side urgency P(c), topology reachability strength Cij(s), and collision cost Kij(s).

[0083] The matching operation is performed in the order of "hard constraints first, then weights". The first round is a hard constraint filter: eliminate candidates that are marked as busy or OCPNXT=1 and reserved by others on slot s; for classes that require confirmation loop, only consider numbers that can carry confirmation in the same slot. The second round is a weight-driven "bin packing": traverse the candidates in Cij(s) descending order by P(c) from high to low, split the quota into the smallest available pieces and fill them in, tracking the remaining capacity and collision accumulation of each edge and each slot. To avoid local congestion, a "dispersion threshold" is set in the matching policy table: when the Kij(s) of a numbered slot grows too fast, subsequent bin packing is forced to migrate to an idle slot far away from the number, even if C is slightly lower, in exchange for lower collision risk. The third round handles the deliverable synchronization pool: when the event state and critical flow overallocation, allow the reserved proportion in the synchronization area to be converted into the critical pool, the rule is to maintain the minimum demand for synchronization in the next frame, and not to cross the protection interval boundary.

[0084] After generating the initial time-slot allocation scheme, the embodiment performs two consistency checks on it. First, confirmation loop check: each allocated control / critical flow occupation point can form a "data-confirmation" path in the same slot time slice, and this confirmation path does not conflict with others. Second, neighborhood fairness quick check: count the effective slot proportion obtained by each node within a window, if it is far higher than the median in the neighborhood and its neighbor queue is generally high, trigger a slight rollback, move a small number of general flows to a higher numbered backup slot, leave space for S103 secondary modification based on queue coefficient. These two checks directly correspond to the engineering needs of maritime rescue: stable and reliable key link, and no excessive "squeeze" in the neighborhood.

[0085] To illustrate the adaptability of the matching to different sea conditions and task stages, two segments are given in this embodiment. Segment one, stable sea conditions during the day, Cij(s) is high in multiple numbered intervals, the resource matrix is mainly situation broadcast and heartbeat, and the packing tends to form a periodic rhythm in low-numbered gaps, and part of the number is fixed to the cross-domain gateway backhaul; Segment two, strong wind and waves at night and suspected fallen person, the key flow P(c) is up-regulated, E_rt is improved during the resource mapping period, and W is increased on the medium and high-numbered idle gaps, the matching result moves the key flow to these gaps, and starts the synchronization pool delivery, sacrificing the continuity of general flow in a small range. This migration follows the natural logic of "worse link → high retransmission budget → need more dispersed and low collision bearing".

[0086] In summary, the embodiment converges the dispersed occupation facts and confirmation evidence into the connectivity matrix C through the network topology table, and expresses the link reachability in a way with numbered sensitivity; through the matching strategy table, the task priority, connectivity strength and conflict risk are combined into a calculable weight to complete the construction of the initial time slot allocation scheme; through the closed loop and fairness check, the reliable reachability of the key message and the sustainable bearing of the neighborhood are guaranteed. The final scheme is written into the resource scheduling unit in the form of a three-tuple of time slot index, node pair and message category, with version stamp and state label, for the next step of fine-grained dynamic correction in S103 combined with the queue state coefficient δ.

[0087] In an embodiment of the time slot resource allocation method of the unmanned cluster cooperative ad hoc network of the present application, the following contents can also be specifically included: Step S601: constructing a special frame header table containing frame type, reservation field and queue length, writing control information into the data frame header in a piggyback structure, generating a framing configuration table according to the electromagnetic propagation characteristics on the sea, and writing the framing configuration table into the data processing unit; Step S602: constructing a time slot state table containing idle state, occupied state and busy state, reading the time slot occupation table for state judgment, generating a distributed selection matrix containing sending time slot, receiving time slot and reservation time slot, and writing the distributed selection matrix into the time slot scheduling unit.

[0088] The embodiment faces the link fluctuation under the conditions of maneuvering, shielding and multipath of the maritime search and rescue formation, encapsulates the control information and business data into a unified data frame around S601-S602, and selects and reserves the available time slots of each frame in a distributed manner on the node side. The framing parameters and time slot occupation table have been established in the preceding S401-S402, and the initial allocation scheme and connectivity matrix have been output in S501-S502, and the current step connects "frame header carried control information - framing configuration - occupation sensing - selection matrix" into an executable closed loop.

[0089] The embodiment first constructs a special frame header table at S601. The fields include frame type Type (distinguishing data frame, upper layer protocol control frame, and acknowledgement frame), reservation field occupy_next (declaring whether to continue using the current time slot number in the next frame), and queue length buffer_size (current total length of the sending end MAC queue), and the remaining reserved bits are aligned by bytes. The control information is written into the data frame header in a piggyback structure, and the frame header length is fixed, the upper layer message fragmentation boundary is not changed, the neighbor node is parsed with a fixed offset, and no additional control channel is required. To avoid interference between control bits and service load, the frame header field is checked by a CRC unit at the link layer, the data segment maintains the upper layer check independence, the error processing paths of the two are separated, and the false triggering of retransmission is reduced. For the delay jitter and multipath expansion of electromagnetic propagation at sea, the embodiment generates a frame configuration table according to historical SNR, platform height, and sea state level, which gives the upper and lower bounds of the lengths of the synchronization frame, the data frame, the acknowledgement segment, and the guard interval, and clearly indicates which sub-slot can be donated as data when there is no synchronization requirement. The configuration table is written into the data processing unit, so that the MAC can allocate the timing according to the table when encapsulating: data transmission-guard-acknowledgement-guard, to ensure that the acknowledgement can be returned in a closed loop within the same slot and does not cross the protection boundary.

[0090] The frame configuration table is not a static constant, but has a scene key value. The embodiment maintains two sets of configuration files in the data processing unit: smooth sea state file and severe sea state file; when the acknowledgement missing rate or round-trip delay variance in the recent window exceeds the threshold, switch to the file of "expanding protection and shrinking data", and vice versa. The switching conforms to the natural propagation law: sea state deterioration leads to multipath expansion, and the protection segment needs to be lengthened, while the number of bytes carried by a single slot decreases, which is equivalent to the upper layer that more time slots are needed for unit load; when it is smooth, it is reversed to shrink and return to the nominal carrying capacity. The special frame header table and the frame configuration table are issued together, and the node writes the current configuration file number used into the reserved bit when sending a packet, and the neighbor interprets the acknowledgement timing accordingly, reducing the beat misjudgment across nodes.

[0091] In S602, the embodiment constructs a time slot state table on the node side, the state set contains three types of idle, occupied, and busy, and the index is aligned with the system unified time slot number. The state judgment reads the local time slot occupation table and superimposes the frame observation: if more than two neighbors announce reservation or collision on a slot, it is marked as busy; if it is successfully transmitted and received with confirmation by the local or one-hop neighbor, it is marked as occupied; the rest is idle. The occupied entry is accompanied by the source hop count HOPCNT and the reserved information whether it is declared by the opposite end OCPNXT=1, which is used for preset masking in the next frame. Then a distributed selection matrix is generated, the process is not simply filled with idle, but combined with the initial allocation scheme of S502 and the local service queue: first, determine the target sending / receiving / reservation share of this frame according to the initial scheme, then filter the busy and other OCPNXT=1 numbers in the time slot state table, and preferentially select the number set corresponding to the one-hop strong connectivity edge; for upper-layer control and key reporting that need to be confirmed in a closed loop, the number that can carry confirmation in the same slot is forced to be selected. If there is not enough idle, the embodiment scales the request amount of the node according to the neighbor queue length ratio, releases part of the reservation to the more "hungry" neighbors, and avoids triggering a chain of retransmissions in the high contention area.

[0092] In order to express the selection logic as a reusable calculation, the embodiment calculates the score of each candidate number s locally, and fills it in according to the score ranking, which depends on the natural indicators of task urgency, connectivity and conflict, which can be expressed by a linear combination: Score(s)=α·P_local+β·C_local(s)-γ·K_local(s). Where P_local is the urgency of the current highest priority queue of the node (mapped by the queue length and expiration time), C_local(s) is the reachable strength of number s on the one-hop peer (derived from the local slice of the connectivity matrix), K_local(s) is the busy / collision history strength of number s in the neighborhood; α, β, γ are the scheduling bias coefficients of the node, written in the node configuration. The higher P_local indicates that the key queue needs to be emptied faster, the higher C_local(s) indicates that it is more likely to succeed on this number, and the higher K_local(s) means that uploading will trigger conflict risk, the linear relationship is consistent with the physical common sense of wireless shared medium. Score is only used for local sorting and does not participate in cross-node coordination to avoid introducing centralized dependence.

[0093] The selection matrix contains three blocks: the sending time slot set S_tx, the receiving time slot set S_rx and the reservation time slot set S_res. S_tx is selected locally according to Score and quota; S_rx is obtained by back inference on the initial scheme and neighbor OCPNXT, ensuring that a receiving window is left on the number declared by the opposite end; S_res is used to declare the number to be used in the next frame, and the rule is that control and key reporting are short-term reserved on the number confirmed to be stable, and ordinary data is not reserved across frames to reduce unnecessary shielding. After the matrix is generated, it is written into the time slot scheduling unit, with the configuration file number and timestamp, for the MAC timer to trigger reception and transmission according to the plan in the frame.

[0094] Two scene fragments are given. One is in a stable sea state, the perception node broadcasts the situation at 0.5 Hz, the confirmation is stable, Score is higher in the low number interval, S_tx presents periodic occupation, and S_res is sparse; the neighbor leaves S_rx for it according to the reserved bit, and the collision is less. The second is that the wind and waves increase and a suspected swimmer appears, the key queue of the search and rescue node expands, P_local rises, the frame configuration switches to the expansion protection file, C_local drops in part of the number, K_local rises in the low number, and the sorting result migrates the key report to the idle gap in the middle and high number, and sets OCPNXT=1 for a number of numbers. After seeing the declaration, the surrounding nodes avoid in the next frame, and the confirmation closed loop during the event period is maintained stable.

[0095] From the technical problem and effect point of view, the embodiment propagates the queue length and reservation intention in the service stream through the special frame header, avoids extra control load and ensures information freshness; maps the delay jitter of electromagnetic propagation at sea into the time budget of confirmation and protection through the frame configuration table, reduces the cross-gap reply failure; through the time slot state table and the distributed selection matrix, the "visible occupation fact" and the "task side priority" are fused locally at the node to form a centralized synchronous decision, reducing the collision and idling caused by rapid changes in topology. Finally, the selection matrix is written into the time slot scheduling unit, cooperates with the dynamic adjustment coefficient δ in S103, and supports continuous communication of the unmanned cluster in the sea mobile process.

[0096] In an embodiment of the time slot resource allocation method of the unmanned cluster cooperative self-organizing network in the application, the following contents can be specifically included: Step S701: constructing a state monitoring table containing node identification, queue length and survival period, calculating the queue state distribution between nodes according to the network connectivity matrix, generating an unmanned node queue state table, calculating a time slot occupation dynamic adjustment coefficient according to the unmanned node queue state table, and writing the dynamic adjustment coefficient into an adjustment control unit; Step S702: constructing a resource reallocation table containing the initial scheme, adjustment coefficient, and priority, modifying the initial time slot allocation scheme according to the dynamic adjustment coefficient, generating a dynamic time slot allocation scheme, and writing the dynamic time slot allocation scheme into the time slot resource scheduling unit.

[0097] The embodiment proposes the implementation path of S701-S702 for resource self-adaptation of the maritime search and rescue formation in the operation period, and connects the network connectivity matrix C and the initial time slot allocation scheme generated in S501-S502, and combines the neighbor queue length and time slot occupation facts maintained by each node in S401-S402, to form a three-dimensional feedback of “on-demand-achievable-conflict” to dynamically modify the time slot occupation and ensure that the key services can still be reliably carried under the maneuvering, network entry and exit, and sea state fluctuation.

[0098] The embodiment first constructs a state monitoring table in S701, the fields of which include node identification (u), queue length (buffer_size, including the mean value and increment in the time window), survival period TTL (used to determine the activity), and two types of operation evidence, one being the edge strength snapshot from the connectivity matrix C, and the other being the busy ratio and cross-frame reservation ratio of the gap associated with the node in the time slot occupation table. The state monitoring table is updated frame by frame: when the node receives the neighbor back-mounted frame header, the buffer_size is refreshed and the TTL is reset; if no packet is seen for a plurality of frames, the TTL is decayed to zero and temporarily removed from the table. After the monitoring table is constructed, the queue state distribution between nodes is calculated according to the connectivity matrix, and the embodiment adopts a one-hop aggregation and two-hop dilution strategy, the former reflecting the direct competition relationship, and the latter used to prompt the risk of regional congestion spread. The summary obtains the queue state table of the unmanned node, the core of which is to generate three types of indexes for each node: demand intensity ρd (normalized by the queue length and its growth rate), achievable capacity ρc (composed of the edge weight weighted mean value of C and the confirmation integrity), and conflict pressure ρk (given by the busy ratio and the overlap probability of OCPNXT). The internal relationship of the three indexes conforms to the natural law: the more the business accumulates, the more the link is accessible, and the more meaningful the resource expansion is; the higher the conflict is, the more blind expansion will lead to increased retransmission and will not be worth the loss.

[0099] Based on the above three indicators, the time slot occupation dynamic adjustment coefficient δ(u) is calculated in this embodiment. Considering the non-stationarity of sea conditions and topology, δ needs to be resistant to short-term impact and responsive to sustained pressure. This embodiment introduces time smoothing and threshold constraints in the calculation, which takes a monotonically increasing relationship with ρd, ρc, and a monotonically decreasing relationship with ρk. In order to facilitate the interface with different load scales, δ is normalized to the interval [0, 1], and a value greater than 1 indicates that the node's share of the next frame should be increased. To clarify the dimensional relationship, a calculation formula is given for explanation: δ(u)=σ{α·ρd(u)+β·ρc(u)-γ·ρk(u)}, where σ is a truncated-normalization function, α, β, γ are system weighting coefficients; ρd(u) represents the sending demand intensity of the node in the monitoring window; ρc(u) represents the possibility of converting quota to effective throughput under the current topology; ρk(u) represents the conflict risk of the node's neighborhood. Each term in the formula can be directly calculated from the monitoring table and the connection matrix, independent of unobservable quantities, and consistent with the physical intuition of "load- reachable-conflict". The calculated δ is written into the adjustment control unit together with the node identifier, and the time stamp and state label are recorded for subsequent reallocation.

[0100] In S702, this embodiment constructs a resource reallocation table around the three-element information of "initial scheme + adjustment coefficient + priority". The initial scheme provides the packing result of specific slot number, node pair and message category; the adjustment coefficient comes from S701, reflecting the current expansion and contraction demand of each node; the priority inherits from the time slot resource matrix, marking the rigid and transferable boundaries of different categories. The first step of reallocation is the re-division of node quota: for each node, scale its obtained quota of each category by δ(u), high-δ nodes maintain or slightly expand key categories (control, tracking reporting, transfer confirmation), and low-δ nodes compress general categories (situation broadcast, non-urgent data), but do not break the minimum guarantee line of key categories. The second step is to eliminate neighborhood conflicts: if a certain numbered slot has K rapid accumulation (busy count rises rapidly) in the neighborhood, migrate the general category allocation that falls on this slot, and preferentially move to idle slots with a longer number distance and a lower C value, and keep the key category in place to form a confirmation closed loop. The third step is cross-node balancing: when multiple adjacent nodes have high δ and insufficient available slots, this embodiment proportionally divides the transferable synchronization pool and general pool to ensure that there is no single long-term starvation; when the TTL of a node tends to zero or ρc drops rapidly, cancel its cross-frame reservation and release OCPNXT occupation to return resources to active and reachable subjects.

[0101] In order to let the redistribution be in place within a single frame, the embodiment keeps the "number sensitive" packing granularity, does not change the timing of the acknowledgement-data same-gap closed loop, and avoids introducing the hidden danger of cross-gap reply failure. The resource redistribution table is checked for consistency after generation: one is the key flow guarantee check, which checks item by item that the control and tracking reports on the corresponding edge still have the same gap acknowledgement path; the other is the time base and protection boundary check, which prevents the expansion operation from breaking through the protection interval or covering the reserved gap of the cross-domain gateway. The dynamic time slot allocation scheme is output through the four-tuple sequence table of "node pair-gap number-category-occupancy period", with a version stamp written into the time slot resource scheduling unit for each node to execute in the next frame according to the piggyback reservation bit.

[0102] The embodiment tests the mechanism in two segments close to maritime rescue. Segment one, the southeast sector wind and wave rise, the confirmation success rate of several search and rescue nodes drops for a short time, ρc decreases, K of the neighborhood increases, and δ is suppressed; the redistribution breaks up the original long string of general category occupation, and maintains the confirmation budget on the number more dispersed free gap, waiting for the sea conditions to recover after the stability is restored. Segment two, a new suspected target appears at night, the buffer_size of a certain water surface perception node increases sharply while it maintains a high position with the gateway C, δ jumps, the system allocates the deliverable proportion of the synchronization pool to the high priority report of the node, and the general category part of other nodes in the group is moderately reduced, while the lower limit quota of heartbeat and synchronization is still reserved to maintain the basic time base.

[0103] From the perspective of technical problems and effects, the embodiment binds node load, activity and local conflict facts together through the state monitoring table, and the combination of the queue state table and the connection matrix provides an objective basis for "who to expand and who to shrink"; through the lightweight calculation of δ coefficient and the number sensitive secondary packing, the initial scheme is modified on demand without destroying the confirmation closed loop and the protection interval; through the explicit indication of priority and deliverable boundary, the key task is stably guaranteed in resource shortage rather than being diluted by averaging. The output dynamic time slot allocation scheme has a time stamp and a state label, which can work together with the piggyback reservation in S103 and local monitoring in subsequent work, and maintain consistent resource allocation order in the maritime formation with frequent maneuvers and in-out network.

[0104] In order to effectively solve the deficiencies of the traditional technology in task configuration, time slot allocation and dynamic adjustment, and provide technical support for unmanned cluster cooperation, an embodiment of an unmanned cluster cooperative ad hoc network time slot resource allocation device for implementing all or part of the contents of the unmanned cluster cooperative ad hoc network time slot resource allocation method is provided, as shown in Figure 2 The unmanned cluster cooperative ad hoc network time slot resource allocation device specifically includes the following contents: The task configuration module 10 is configured to set the cross-domain cluster nodes as gateway control nodes, set the job group nodes as water surface sensing nodes, set the unmanned platform nodes as search and rescue execution nodes, generate a task configuration matrix, construct a task state machine according to the task configuration matrix, perform task deduction and calculation based on the task state machine for search and rescue area coverage, target search tracking, and wounded personnel transfer and salvage, convert task requirements into a time slot resource matrix, and write the time slot resource matrix into a resource allocation unit. The resource scheduling module 20 is configured to divide a time frame into a time synchronization area and a data interaction area, write the time frame structure into each water surface unmanned node control unit, read the buffer queue length of adjacent search and rescue nodes, generate a time slot occupation table reflecting the communication state of the unmanned cluster, calculate a network connectivity matrix between the water surface nodes according to the time slot occupation table, perform matching operation on the network connectivity matrix and the time slot resource matrix, generate an initial time slot allocation scheme suitable for a search and rescue scene, and write the initial time slot allocation scheme into a resource scheduling unit. The time slot adjustment module 30 is configured to write control information into a data frame header in a piggyback structure, generate a framing configuration table according to the sea electromagnetic propagation characteristics, use the framing configuration table for data framing processing, read the time slot occupation table to perform distributed time slot selection for the search and rescue nodes, establish an unmanned node queue state table based on the network connectivity matrix, calculate a time slot occupation dynamic adjustment coefficient according to the unmanned node queue state table, correct the allocation scheme according to the dynamic adjustment coefficient, write the corrected allocation scheme into a time slot resource scheduling unit, and generate a time slot resource scheduling strategy supporting sea maneuvering of the unmanned cluster.

[0105] As can be seen from the above description, the unmanned cluster cooperative self-organizing network time slot resource allocation device provided by the embodiments of the present application can effectively map requirements through the innovative design of the task configuration system, the state machine, and the resource matrix. The time slot allocation mechanism is constructed, the reliable scheduling scheme is established by combining connectivity analysis and resource matching, the adaptability of the allocation is ensured through state monitoring and strategy adjustment by introducing dynamic optimization. The method effectively solves the deficiencies of the traditional technology in task configuration, time slot allocation, and dynamic adjustment, and provides technical support for unmanned cluster cooperation.

[0106] In order to further illustrate the present application, the present application further provides a specific application example of an unmanned cluster cooperative self-organizing network time slot resource allocation method using the above unmanned cluster cooperative self-organizing network time slot resource allocation device, which specifically includes the following contents: The time slot resource allocation method based on task driving is applied to the constructed sea unmanned cluster cooperative operation scene relying on simulation technology, and the network reconstruction adaptive ability of the method under task driving is verified. In the scene, a UAV cluster sea scene is constructed, and three different tasks of bearing tasks, new node network access tasks, and formation change tasks are set to verify.

[0107] The load task refers to a task of changing different types of traffic according to combat requirements; the new node network access task refers to a task of unmanned aerial vehicle joining the network in an actual battlefield; and the formation change task refers to a task of unmanned cluster changing formation according to actual environment changes under a combat order.

[0108] (1) Network reconstruction verification driven by a load task The network reconstruction verification driven by a load task mainly tests whether the technology can adapt to changing time slot allocation strategies when different types of traffic change. See Table 1 below: Table 1 Parameter settings of dynamic scalable access protocol

[0109] When the simulation is performed to 30s, the frequency of formation control messages changes from 2Hz to 5Hz, and the frequency of state reporting messages changes from 0.5Hz to 0.25Hz.

[0110] As can be seen from the key parameter settings, the frame length is 100ms, the 300th frame corresponds to the simulation time of 30s, the periodic traffic is increased at 30s in the scene, and the periodic time slot allocation is increased after the 300th frame, which indicates that the network can reconstruct time slot resources in the load task.

[0111] (2) Network reconstruction verification driven by a new node network access task The network reconstruction verification driven by a new node network access task mainly tests whether the technology can reconstruct the intra-cluster network when a new node accesses the network. See Table 2 below: Table 2 C-OLSR parameter settings

[0112] When the simulation is performed to 20s, ten silent nodes access the network in the scene.

[0113] In the cluster structure, the intra-cluster members report information to the cluster head, and the cluster head communicates with other nodes through the backbone network. When a new member joins the cluster, it can communicate with other nodes in the network through the cluster head after discovering the cluster head. Therefore, the network access time is defined as the time from the start of the node to the discovery of the cluster head. Through simulation verification, the network access time of the intra-cluster nodes is shown in the table. See Table 3: Table 3 Node network access time

[0114] (3) Network reconstruction verification driven by a formation change task The network reconstruction verification driven by a formation change task tests whether the technology can maintain the stability of the network topology in motion. See Table 4: Table 4 C-OLSR parameter settings

[0115] After the simulation starts, the cluster of 100 nodes is gradually divided into four platoons, and it can be found from the data transmission of the collected communication link that when the node platoon changes and the distance between the node and the cluster head is far away, the link communication can still be maintained normally.

[0116] From the hardware level, in order to effectively solve the deficiencies of traditional technology in task configuration, time slot allocation and dynamic adjustment, and provide technical support for unmanned cluster cooperation, the present application provides an embodiment of an electronic device for implementing all or part of the contents of the time slot resource allocation method for unmanned cluster cooperative ad hoc network, which specifically includes the following contents: A processor, a memory, a communication interface and a bus; wherein the processor, the memory and the communication interface complete the communication among each other through the bus; the communication interface is used to realize the information transmission between the unmanned cluster cooperative ad hoc network time slot resource allocation device and the core business system, the user terminal and the related database and other related devices; the logic controller can be a desktop computer, a tablet computer and a mobile terminal, etc., and the present embodiment is not limited thereto. In the present embodiment, the logic controller can be implemented with reference to the embodiments of the unmanned cluster cooperative ad hoc network time slot resource allocation method and the embodiments of the unmanned cluster cooperative ad hoc network time slot resource allocation device, the contents of which are incorporated herein, and the repeated parts will not be described herein.

[0117] It can be understood that the user terminal can include a smart phone, a tablet electronic device, a network set-top box, a portable computer, a desktop computer, a personal digital assistant (PDA), a vehicle-mounted device, a smart wearable device, etc. The smart wearable device can include smart glasses, a smart watch, a smart bracelet, etc.

[0118] In practical applications, part of the unmanned cluster cooperative ad hoc network time slot resource allocation method can be executed on the electronic device as described above, or all operations can be completed in the client device. Specifically, the selection can be made according to the processing capacity of the client device and the limitation of the user's use scenario, etc. The present application is not limited thereto. If all operations are completed in the client device, the client device can also include a processor.

[0119] The client device described above can have a communication module (i.e., a communication unit) that can be in communication connection with a remote server to realize data transmission with the server. The server can include a server of a task scheduling center side, and in other implementation scenarios, can also include a server of an intermediate platform, such as a server of a third-party server platform that is in communication link with the server of the task scheduling center. The server can include a single computer device, or can include a server cluster composed of multiple servers, or a server structure of a distributed device.

[0120] Figure 3 A schematic block diagram of a system configuration of an electronic device 9600 according to an embodiment of the present application is shown in FIG. 9. As shown in FIG. 9, the electronic device 9600 can include a central processor 9100 and a memory 9140; the memory 9140 is coupled to the central processor 9100. It is noted that the structure shown in FIG. 9 is exemplary; other types of structures can also be used to supplement or replace the structure to realize telecommunication functions or other functions. Figure 3 Figure 3 The structure shown in FIG. 9 is exemplary; other types of structures can also be used to supplement or replace the structure to realize telecommunication functions or other functions.

[0121] In an embodiment, the time slot resource allocation method for unmanned cluster cooperative ad hoc network can be integrated into the central processor 9100. The central processor 9100 can be configured to perform the following control: Step S101: setting a cross-domain cluster node as a gateway control node, setting a job group node as a water surface sensing node, setting an unmanned platform node as an execution search and rescue node, generating a task configuration matrix, constructing a task state machine according to the task configuration matrix, performing task deduction and calculation based on the task state machine for search and rescue area coverage, target search tracking, and wounded personnel transfer salvage, converting task requirements into a time slot resource matrix, and writing the time slot resource matrix into a resource allocation unit; Step S102: dividing a time frame into a time synchronization area and a data interaction area, writing the time frame structure into each water surface unmanned node control unit, reading the length of a buffer queue of a neighboring search and rescue node, generating a time slot occupation table reflecting the communication state of the unmanned cluster, calculating a network connectivity matrix between water surface nodes according to the time slot occupation table, performing matching operation on the network connectivity matrix and the time slot resource matrix, generating an initial time slot allocation scheme suitable for a search and rescue scenario, and writing the initial time slot allocation scheme into a resource scheduling unit; ​Step S103: write the control information in the data frame header in a piggyback structure, generate a framing configuration table according to the electromagnetic propagation characteristics on the sea, use the framing configuration table for data framing processing, read the time slot occupation table to perform distributed time slot selection on the search and rescue node, establish an unmanned node queue state table based on the network connectivity matrix, calculate a time slot occupation dynamic adjustment coefficient according to the unmanned node queue state table, correct the allocation scheme according to the dynamic adjustment coefficient, write the corrected allocation scheme into a time slot resource scheduling unit, and generate a time slot resource scheduling strategy supporting the sea maneuver of the unmanned cluster.

[0122] From the above description, the electronic device provided by the embodiments of the present application can effectively map the requirements by innovatively designing the task configuration system, through the state machine and the resource matrix. The time slot allocation mechanism is constructed, the reliable scheduling scheme is established by combining the connectivity analysis and the resource matching. The adaptability of the allocation is ensured by introducing the dynamic optimization through the state monitoring and the strategy adjustment. The method effectively solves the deficiencies of the traditional technology in the task configuration, the time slot allocation and the dynamic adjustment, and provides a technical support for the unmanned cluster collaboration.

[0123] In another embodiment, the unmanned cluster collaboration ad hoc network time slot resource allocation device can be configured separately from the central processor 9100, for example, the unmanned cluster collaboration ad hoc network time slot resource allocation device can be configured as a chip connected with the central processor 9100, and the function of the unmanned cluster collaboration ad hoc network time slot resource allocation method is realized through the control of the central processor.

[0124] As shown in Figure 3 , the electronic device 9600 can also include a communication module 9110, an input unit 9120, an audio processor 9130, a display 9160, and a power supply 9170. It should be noted that the electronic device 9600 does not necessarily include all the components shown in Figure 3 ; in addition, the electronic device 9600 can also include components not shown in Figure 3 , which can be referred to the prior art.

[0125] As shown in Figure 3 , the central processor 9100 is also sometimes referred to as a controller or an operation control, which can include a microprocessor or other processor device and / or a logic device, the central processor 9100 receives input and controls the operation of each component of the electronic device 9600.

[0126] The memory 9140, for example, can be one or more of a buffer, a flash memory, a hard drive, a removable media, a volatile memory, a non-volatile memory, or other suitable device. The above-mentioned information related to failure can be stored, and in addition, a program for executing the information related to failure can be stored. The central processing unit 9100 can execute the program stored in the memory 9140 to achieve information storage or processing, and the like.

[0127] The input unit 9120 provides input to the central processing unit 9100. The input unit 9120 is, for example, a key or a touch input device. The power supply 9170 is used to supply power to the electronic device 9600. The display 9160 is used to display display objects such as images and characters. The display can be, for example, an LCD display, but is not limited thereto.

[0128] The memory 9140 can be a solid state memory such as a read only memory (ROM), a random access memory (RAM), a SIM card, and the like. It can also be a memory that retains information even when power is off, can be selectively erased, and is provided with more data, and examples of such a memory are sometimes referred to as an EPROM, and the like. The memory 9140 can also be some other type of device. The memory 9140 includes a buffer memory 9141 (sometimes referred to as a buffer). The memory 9140 can include an application / function storage section 9142 for storing application programs and function programs or for storing a flow for executing operations of the electronic device 9600 by the central processing unit 9100.

[0129] The memory 9140 can also include a data storage section 9143 for storing data such as contacts, digital data, pictures, sounds, and / or any other data used by the electronic device. A driver storage section 9144 of the memory 9140 can include various drivers of the electronic device for a communication function and / or for executing other functions of the electronic device such as a messaging application, an address book application, and the like.

[0130] The communication module 9110 is a transmitter / receiver that transmits and receives signals via an antenna 9111. The communication module 9110 (transmitter / receiver) is coupled to the central processing unit 9100 to provide input signals and receive output signals, which can be the same as in the case of a conventional mobile communication terminal.

[0131] Based on different communication technologies, multiple communication modules 9110, such as a cellular network module, a Bluetooth module, and / or a wireless local area network module, etc., can be provided in the same electronic device. The communication module 9110 (transmitter / receiver) is also coupled to a speaker 9131 and a microphone 9132 via an audio processor 9130 to provide audio output via the speaker 9131 and to receive audio input from the microphone 9132, thereby enabling typical telecommunication functions. The audio processor 9130 can include any suitable buffers, decoders, amplifiers, etc. In addition, the audio processor 9130 is also coupled to the central processor 9100, thereby enabling the recording of audio on the local device via the microphone 9132 and enabling the playing of stored audio on the local device via the speaker 9131.

[0132] The embodiments of the present application further provide a computer readable storage medium capable of implementing all steps of the unmanned cluster cooperative ad hoc network time slot resource allocation method in which the execution subject is a server or a client in the above-mentioned embodiments, and the computer program is stored on the computer readable storage medium, and when the processor executes the computer program, all steps of the unmanned cluster cooperative ad hoc network time slot resource allocation method in which the execution subject is a server or a client in the above-mentioned embodiments are implemented, for example, the following steps are implemented when the processor executes the computer program: Step S101: setting a cross-domain cluster node as a gateway control node, setting a job group node as a water surface sensing node, setting an unmanned platform node as an execution search and rescue node, generating a task configuration matrix, constructing a task state machine according to the task configuration matrix, performing task deduction and calculation based on the task state machine on search and rescue area coverage, target search tracking, and wounded personnel transfer salvage, converting task requirements into a time slot resource matrix, and writing the time slot resource matrix into a resource allocation unit; Step S102: dividing a time frame into a time synchronization area and a data interaction area, writing the time frame structure into each water surface unmanned node control unit, reading the length of a buffer queue of a neighboring search and rescue node, generating a time slot occupation table reflecting the communication state of the unmanned cluster, calculating a network connectivity matrix between water surface nodes according to the time slot occupation table, performing matching operation on the network connectivity matrix and the time slot resource matrix, generating an initial time slot allocation scheme suitable for a search and rescue scene, and writing the initial time slot allocation scheme into a resource scheduling unit; Step S103: write the control information in the data frame header in a piggyback structure, generate a framing configuration table according to the electromagnetic propagation characteristics in the sea, use the framing configuration table for data framing processing, read the time slot occupation table for distributed time slot selection of the search and rescue node, establish an unmanned node queue state table based on the network connectivity matrix, calculate a time slot occupation dynamic adjustment coefficient according to the unmanned node queue state table, correct the allocation scheme according to the dynamic adjustment coefficient, write the corrected allocation scheme into a time slot resource scheduling unit, and generate a time slot resource scheduling strategy supporting the sea maneuver of the unmanned cluster.

[0133] From the above description, the computer readable storage medium provided by the embodiments of the application can realize effective mapping of requirements by innovatively designing a task configuration system, through a state machine and a resource matrix. A time slot allocation mechanism is constructed, combined with connectivity analysis and resource matching, to establish a reliable scheduling scheme. Dynamic optimization is introduced, through state monitoring and strategy adjustment, to ensure the adaptability of the allocation. The method effectively solves the deficiencies of traditional technologies in task configuration, time slot allocation, and dynamic adjustment, and provides technical support for unmanned cluster collaboration.

[0134] The embodiments of the application also provide a computer program product capable of realizing all steps of the unmanned cluster collaboration ad hoc network time slot resource allocation method in the above-mentioned embodiments, wherein the computer program / instruction is executed by a processor to realize the steps of the unmanned cluster collaboration ad hoc network time slot resource allocation method, for example, the computer program / instruction realizes the following steps: Step S101: set a cross-domain cluster node as a gateway control node, set a job group node as a water surface sensing node, set an unmanned platform node as an execution search and rescue node, generate a task configuration matrix, construct a task state machine according to the task configuration matrix, perform task deduction and calculation on search and rescue area coverage, target search tracking, and wounded personnel transfer salvage based on the task state machine, and convert task requirements into a time slot resource matrix and write the time slot resource matrix into a resource allocation unit; Step S102: divide a time frame into a time synchronization area and a data interaction area, write the time frame structure into each water surface unmanned node control unit, read the buffer queue length of adjacent search and rescue nodes, generate a time slot occupation table reflecting the communication state of the unmanned cluster, calculate a network connectivity matrix between water surface nodes according to the time slot occupation table, perform matching operation on the network connectivity matrix and the time slot resource matrix, generate an initial time slot allocation scheme suitable for a search and rescue scene, and write the initial time slot allocation scheme into a resource scheduling unit; Step S103: write the control information in the data frame header in a piggyback structure, generate a framing configuration table according to the electromagnetic propagation characteristics on the sea, use the framing configuration table for data framing processing, read the time slot occupation table to perform distributed time slot selection on the search and rescue node, establish an unmanned node queue state table based on the network connectivity matrix, calculate a time slot occupation dynamic adjustment coefficient according to the unmanned node queue state table, correct the distribution scheme according to the dynamic adjustment coefficient, write the corrected distribution scheme into a time slot resource scheduling unit, and generate a time slot resource scheduling strategy supporting the sea maneuver of the unmanned cluster.

[0135] From the above description, the computer program product provided by the embodiments of the present application can realize effective mapping of requirements by innovatively designing a task configuration system, through a state machine and a resource matrix. A time slot allocation mechanism is constructed, combined with connectivity analysis and resource matching, to establish a reliable scheduling scheme. Dynamic optimization is introduced, through state monitoring and strategy adjustment, to ensure the adaptability of the allocation. The method effectively solves the deficiencies of traditional technologies in task configuration, time slot allocation, and dynamic adjustment, and provides technical support for unmanned cluster collaboration.

[0136] Those skilled in the art will understand that the embodiments of the present application can be provided as a method, device, or computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer usable program code.

[0137] The present application is described with reference to flowcharts and / or block diagrams of the method, device (apparatus), and computer program product according to the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of flows and / or blocks in the flowcharts and / or block diagrams 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 apparatus to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing apparatus produce a device that implements the functions specified in the flowcharts and / or block diagrams. Figure 1 The device that implements the functions specified in one or more flows and / or blocks. Figure 1 The device that implements the functions specified in one or more flows and / or blocks.

[0138] These computer program instructions can also be stored in a computer readable storage medium that can direct the computer or other programmable data processing apparatus to work in a specific manner, so that the instructions stored in the computer readable storage medium produce a manufactured product including instruction devices that implement the functions specified in the flowcharts and / or block diagrams. Figure 1 The device that implements the functions specified in one or more flows and / or blocks.Figure 1 the function specified in the one or more blocks.

[0139] These computer program instructions can also be loaded into computer or other programmable data processing devices, so that a series of operation steps are performed on the computer or other programmable data processing devices to generate computer-implemented processes, so that the instructions executed on the computer or other programmable devices provide processes for implementing the flow Figure 1 the flow or flows and / or blocks Figure 1 the steps of the function specified in the one or more blocks.

[0140] The principles and implementation manners of the present application are described in the specific embodiments in the present application, and the above embodiment descriptions are only used to help understand the method of the present application and its core idea; meanwhile, for those skilled in the art, according to the idea of the present application, the specific implementation manners and application ranges will have changes, and the above descriptions should not be understood as limitations on the present application.

Claims

1. A method for allocating time slot resources in an unmanned cluster cooperative self-organizing network, characterized in that, The method includes: Set the cross-domain cluster node as the gateway control node, the work group node as the surface sensing node, and the unmanned platform node as the search and rescue execution node. Generate a task configuration matrix, construct a task state machine based on the task configuration matrix, perform task simulation and calculation on the search and rescue area coverage, target search and tracking, and casualty transfer and retrieval based on the task state machine, convert the task requirements into a time slot resource matrix, and write the time slot resource matrix into the resource allocation unit. The time frame is divided into a time synchronization area and a data interaction area. The time frame structure is written into the control unit of each surface unmanned node. The buffer queue length of the adjacent search and rescue node is read to generate a time slot occupancy table that reflects the communication status of the unmanned cluster. The network connectivity matrix between the surface nodes is calculated based on the time slot occupancy table. The network connectivity matrix is ​​matched with the time slot resource matrix to generate an initial time slot allocation scheme suitable for the search and rescue scenario. The initial time slot allocation scheme is written into the resource scheduling unit. Control information is written into the header of the data frame in a backpack structure. A frame configuration table is generated based on the electromagnetic propagation characteristics at sea. The frame configuration table is used for data frame processing. The time slot occupancy table is read to perform distributed time slot selection for search and rescue nodes. An unmanned node queue status table is established based on the network connectivity matrix. The time slot occupancy dynamic adjustment coefficient is calculated based on the unmanned node queue status table. The allocation scheme is corrected according to the dynamic adjustment coefficient. The corrected allocation scheme is written into the time slot resource scheduling unit to generate a time slot resource scheduling strategy that supports the maritime maneuvering of unmanned swarms.

2. The unmanned cluster cooperative self-organizing network time slot resource allocation method according to claim 1, characterized in that, The step of setting cross-domain cluster nodes as gateway control nodes, setting work group nodes as surface sensing nodes, and setting unmanned platform nodes as search and rescue execution nodes, and generating a task configuration matrix, includes: Construct a cluster organization table that includes control hierarchy, command relationship and operation area, read the cross-domain cluster node identifier and map it as gateway control node, read the operation group node identifier and map it as surface sensing node, read the unmanned platform node identifier and map it as search and rescue execution node, and write the cluster organization table into the node management unit. Construct a task planning table that includes task attributes, message types, and communication cycles. Divide the node communication levels according to the cluster organization table, generate a task configuration matrix that includes node identifiers, level attributes, and communication requirements, and write the task configuration matrix into the task management unit.

3. The unmanned cluster cooperative self-organizing network time slot resource allocation method according to claim 1, characterized in that, The process of constructing a task state machine based on the task configuration matrix, performing task simulation calculations on search and rescue area coverage, target search and tracking, and casualty transfer and retrieval based on the task state machine, converting task requirements into a time-slot resource matrix, and writing the time-slot resource matrix into a resource allocation unit includes: Construct a state machine configuration table that includes state transition conditions, task triggering conditions, and simulation parameters. Generate a task state machine based on the task configuration matrix. Perform Monte Carlo simulations on search and rescue area coverage, target search and tracking, and casualty transfer and retrieval based on the state machine. Generate a task requirement prediction table and write the task requirement prediction table into the simulation engine. A resource mapping table containing communication performance, data capacity, and quality indicators is constructed. Communication resource requirements are calculated based on the task requirement prediction table, a time slot resource matrix is ​​generated, and the time slot resource matrix is ​​written into the resource allocation unit according to priority.

4. The unmanned cluster cooperative self-organizing network time slot resource allocation method according to claim 1, characterized in that, The process of dividing the time frame into a time synchronization area and a data interaction area, writing the time frame structure into the control unit of each surface unmanned node, reading the buffer queue length of adjacent search and rescue nodes, and generating a time slot occupancy table reflecting the communication status of the unmanned cluster includes: Construct a time frame structure table that includes synchronization frame length, data frame length, and protection interval. Divide the time frame into a time synchronization area and a data interaction area according to a fixed interval. Generate a sub-frame parameter table that includes frame format, number of time slots, and time slot length. Write the sub-frame parameter table into the control unit of each unmanned node on the water surface. A neighbor status table containing node address, queue length, and lifetime is constructed. The cache queue length information of neighboring search and rescue nodes is read, and a time slot occupancy table containing time slot status, reservation information, and hop count information is generated. The time slot occupancy table is then written into the status management unit.

5. The unmanned cluster cooperative self-organizing network time slot resource allocation method according to claim 1, characterized in that, The step of calculating the network connectivity matrix between surface nodes based on the time slot occupancy table, matching the network connectivity matrix with the time slot resource matrix to generate an initial time slot allocation scheme suitable for the search and rescue scenario, and writing the initial time slot allocation scheme into the resource scheduling unit includes: A network topology table containing node identifiers, link status, and time slot status is constructed. The connectivity between water surface nodes is calculated based on the time slot occupancy table, a network connectivity matrix is ​​generated, and the network connectivity matrix is ​​written into the topology management unit. A matching strategy table containing resource requirements, connectivity status, and allocation weights is constructed. The network connectivity matrix and the time slot resource matrix are subjected to weighted matching operations to generate an initial time slot allocation scheme. The initial time slot allocation scheme is then written into the resource scheduling unit.

6. The unmanned cluster cooperative self-organizing network time slot resource allocation method according to claim 1, characterized in that, The steps include writing control information into the data frame header in a backpack structure, generating a frame configuration table based on the electromagnetic propagation characteristics at sea, using the frame configuration table for data frame processing, and reading the time slot occupancy table to perform distributed time slot selection for search and rescue nodes, including: A dedicated frame header table containing frame type, reservation field, and queue length is constructed. Control information is written into the data frame header in a back-mounted structure. A sub-frame configuration table is generated based on the electromagnetic propagation characteristics at sea. The sub-frame configuration table is then written into the data processing unit. A time slot status table containing idle, occupied, and busy states is constructed. The time slot occupancy table is read to determine the status, and a distributed selection matrix containing sending time slots, receiving time slots, and reserved time slots is generated. The distributed selection matrix is ​​then written into the time slot scheduling unit.

7. The unmanned cluster cooperative self-organizing network time slot resource allocation method according to claim 1, characterized in that, The process involves establishing an unmanned node queue status table based on the network connectivity matrix, calculating a dynamic adjustment coefficient for time slot occupancy based on the unmanned node queue status table, correcting the allocation scheme according to the dynamic adjustment coefficient, writing the corrected allocation scheme into the time slot resource scheduling unit, and generating a time slot resource scheduling strategy that supports unmanned swarm maritime maneuvering, including: Construct a status monitoring table that includes node identifier, queue length, and lifespan; calculate the queue status distribution among nodes based on the network connectivity matrix; generate an unmanned node queue status table; calculate the time slot occupancy dynamic adjustment coefficient based on the unmanned node queue status table; and write the dynamic adjustment coefficient into the adjustment control unit. A resource reallocation table containing an initial scheme, adjustment coefficients, and priorities is constructed. The initial time slot allocation scheme is modified according to the dynamic adjustment coefficients to generate a dynamic time slot allocation scheme. The dynamic time slot allocation scheme is then written into the time slot resource scheduling unit.

8. A time slot resource allocation device for an unmanned cluster cooperative self-organizing network, characterized in that, The device includes: The task configuration module is used to set cross-domain cluster nodes as gateway control nodes, work group nodes as surface sensing nodes, and unmanned platform nodes as search and rescue execution nodes, generate a task configuration matrix, construct a task state machine based on the task configuration matrix, perform task simulation and calculation on search and rescue area coverage, target search and tracking, and casualty transfer and retrieval based on the task state machine, convert task requirements into a time slot resource matrix, and write the time slot resource matrix into the resource allocation unit. The resource scheduling module is used to divide the time frame into a time synchronization area and a data interaction area, write the time frame structure into the control unit of each surface unmanned node, read the buffer queue length of the adjacent search and rescue node, generate a time slot occupancy table reflecting the communication status of the unmanned cluster, calculate the network connectivity matrix between the surface nodes according to the time slot occupancy table, perform matching operation between the network connectivity matrix and the time slot resource matrix to generate an initial time slot allocation scheme suitable for the search and rescue scenario, and write the initial time slot allocation scheme into the resource scheduling unit. The time slot adjustment module is used to write control information into the data frame header in a backpack structure, generate a frame configuration table based on the electromagnetic propagation characteristics at sea, use the frame configuration table for data frame processing, read the time slot occupancy table to perform distributed time slot selection for search and rescue nodes, establish an unmanned node queue status table based on the network connectivity matrix, calculate the time slot occupancy dynamic adjustment coefficient based on the unmanned node queue status table, correct the allocation scheme according to the dynamic adjustment coefficient, write the corrected allocation scheme into the time slot resource scheduling unit, and generate a time slot resource scheduling strategy that supports unmanned swarm maritime maneuvering.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the unmanned cluster cooperative self-organizing network time slot resource allocation method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the unmanned cluster cooperative self-organizing network time slot resource allocation method according to any one of claims 1 to 7.

Images