Dual-frequency master-slave unmanned aerial vehicle array communication networking method based on TDMA

Abstract

The invention belongs to the technical field of unmanned aerial vehicle measurement and control communication, and discloses a dual-frequency master-slave type unmanned aerial vehicle array communication networking method based on TDMA. The method comprises the following steps: a ground control station broadcasts an initialization instruction through a first frequency band and receives self-inspection feedback information of each unmanned aerial vehicle; based on the self-inspection feedback information, dynamically allocating a communication time slot to the unmanned aerial vehicle which works normally; the ground control station performs instruction interaction with the target unmanned aerial vehicle through the first frequency band according to the time slot allocation; when the interactive instruction triggers image return, the target unmanned aerial vehicle returns image data through a second high-frequency band; and the ground control station dynamically determines a recovery priority according to the state information returned by the plurality of unmanned aerial vehicles in real time, and controls the unmanned aerial vehicles to be recovered according to the priority in sequence through a first frequency band broadcast instruction. According to the invention, a dual-frequency separation design is adopted, so that the safety and stability of cluster operation are improved; through precise allocation of TDMA time slots, the utilization rate of network resources is improved, and stability, controllability and operability of the system are guaranteed.

Description

Technical Field

[0001] This invention belongs to the field of UAV telemetry, control and communication technology, and particularly relates to a UAV array (cluster) communication networking method suitable for wide-area visual perception scenarios. It is applicable to UAV cluster communication, image transmission, supervision and management in scenarios such as environmental monitoring and on-site monitoring, including network structure, command frame structure and time slot allocation. Background Technology

[0002] With the booming development of the low-altitude economy, the application depth and breadth of drone technology in environmental monitoring, on-site surveillance, and emergency disaster relief are continuously expanding, and large-scale drone swarm collaborative operation scenarios are increasing. In these scenarios, multiple drones need to achieve efficient and reliable command interaction with the ground control station, while simultaneously completing high-speed transmission of massive amounts of data such as high-resolution images and real-time video. The networking method, as the core support for communication and data transmission, directly determines the stability and efficiency of swarm operations, becoming a key pain point in current technological breakthroughs.

[0003] In the networking design of existing UAV flight control systems, Time Division Multiple Access (TDMA) has become the mainstream network access method for multi-UAV collaborative scenarios because it can effectively avoid command conflicts and reduce network congestion risks. This approach allocates a dedicated time slot to each UAV, limiting the terminal to complete command and data transmission only within the corresponding time slot. It further improves network throughput and channel utilization by optimizing time slot allocation strategies. Several research achievements have been made related to this technology: Chongqing University of Posts and Telecommunications, in its patent "Multi-hop TDMA Access Method for UAV Ad Hoc Networks Based on Time Slot Assisted Allocation and Use," proposed two mechanisms: "one-hop node-assisted allocation of data time slots" and "idle data time slot-assisted neighbor node transmission," significantly improving time slot allocation efficiency and channel utilization. Shi Boya et al., in their patent "Time Slot Allocation Method Based on TDMA Networks," balanced bandwidth utilization and network node expansion capabilities through time slot function division and dynamic frame length adjustment. The 54th Research Institute of China Electronics Technology Group Corporation, in its patent "A Dynamic TDMA Time Slot Allocation Method for Variable Time Slot Cluster UAV Frequency Hopping Ad Hoc Networks," divided time frames into narrowband and wideband time slots, improving the transmission capacity of the cluster network through differentiated time slot length design and dynamic adjustment mechanisms. However, all the above-mentioned TDMA-related networking and time slot allocation schemes are designed for single-frequency transmission systems. Due to the half-duplex communication mode, command transmission and data transmission need to share the same channel resources, which leads to mutual encroachment and limited processing efficiency, making it difficult to meet the dual requirements of real-time command transmission and high-speed data transmission for cluster collaboration.

[0004] Regarding the completeness and universality of networking methods, existing technologies still have significant shortcomings: The patent "A Networking Method Applicable to Large-Scale UAV Ground-to-Air Networks" by Beijing Institute of Technology and the 54th Research Institute of China Electronics Technology Group Corporation (CETC) details the workflow management by dividing core stages such as network search, network entry, and link establishment, but does not provide specific time slot allocation logic and command framing details; The patent "UAV Cluster Self-Organizing Network Communication System and Method Based on Broadband and Narrowband Integration" published by Xi'an University of Electronic Science and Technology solves the cluster management problem under dynamic topology by integrating CSMA / CA and MC-TDMA hybrid access methods, but does not involve the specific design of command frame structure and command transmission control process; The "UAV Self-Organizing Network TDMA Access Method for Emergency Communication Networks" proposed by An Guochen et al. designs a networking protocol with 5 stages and a detailed time slot allocation mechanism for emergency scenarios, but the working state machine and time slot allocation logic of this scheme are highly adapted to emergency scenarios and lack compatibility with general scenarios such as environmental monitoring and on-site monitoring, thus limiting its applicability.

[0005] Furthermore, single-band transmission systems have inherent drawbacks: the structural design and time slot allocation of command and data transmission frames must consider both types of service requirements, leading to logical complexity. This reduces the flexibility of UAV swarm management and scheduling, and further limits the real-time performance and efficiency of command transmission. Although some scholars have proposed a dual-band collaborative UAV array communication architecture, using the 840.5-845MHz band for command transmission and the 2.4GHz band for data transmission, this system lacks clear supporting network procedures, time slot allocation strategies, and command control details, failing to form a complete technical solution. Therefore, how to fully leverage the resource advantages of dual-band transmission and the anti-collision characteristics of TDMA networking to design a complete communication network method that balances highly reliable command control and high-speed data transmission has become a pressing technical challenge in the field of UAV swarm communication.

[0006] Based on the above analysis, the problems and shortcomings of the existing technology are as follows:

[0007] (1) Existing TDMA-related networking and time slot allocation schemes are all designed for single-frequency transmission systems. Due to the half-duplex communication mode, command transmission and data transmission share the same channel resources, which leads to mutual encroachment and limited processing efficiency, making it difficult to meet the dual requirements of real-time command and high-speed data for cluster collaboration.

[0008] (2) In terms of the completeness and universality of the networking method, the existing technology does not provide specific time slot allocation logic and instruction framing details, or does not involve the specific design of instruction frame structure and the control process of instruction transmission, and lacks compatibility with common scenarios such as environmental monitoring and on-site monitoring, thus limiting its scope of application.

[0009] (3) The logic of the existing single-band transmission system is complex, which reduces the flexibility of UAV cluster management and scheduling, and further limits the real-time performance and efficiency of command transmission. At the same time, the supporting networking process, time slot allocation strategy and command control details are not clearly defined, and a complete technical solution has not been formed. Summary of the Invention

[0010] To address the compatibility issues of command and image transmission data and the TDMA self-organizing network problem in existing single-band UAV array communication systems, this invention discloses a TDMA-based dual-band master-slave UAV array (cluster) communication networking method. This method aims to provide an efficient and reliable communication networking solution for UAV arrays to meet the communication and data transmission requirements of dual-band master-slave UAV arrays. The technical solution is as follows:

[0011] This invention is implemented as follows: a TDMA-based dual-frequency master-slave UAV array communication networking method is executed collaboratively by a ground control station acting as the master station and multiple UAVs acting as slave stations, including the following stages:

[0012] S1. Power-on and networking phase: The ground control station broadcasts a power-on self-test command containing time slot allocation information in a loop through the first frequency band; each UAV performs a self-test and feeds back its status through the first frequency band; the ground control station determines the set of UAVs that are working normally based on the feedback and completes system initialization and initial time slot allocation;

[0013] S2. Mission Execution Phase: The ground control station sends uplink control commands to the target UAV through the first frequency band in the designated unicast time slot according to the time slot allocation; the target UAV replies with downlink status commands through the first frequency band in the corresponding feedback time slot; when the uplink control command includes an image acquisition command, the target UAV transmits image data back to the ground control station through the second frequency band with a frequency higher than the first frequency band.

[0014] S3. Shutdown and Recovery Phase: The ground control station broadcasts recovery preparation instructions and shutdown and recovery instructions sequentially through the first frequency band. The priority of the recovery group is dynamically adjusted according to the real-time status of the UAV, and the UAV cluster is controlled to perform shutdown and recovery operations in sequence according to the priority order.

[0015] Furthermore, the first frequency band is the 433MHz low-frequency band, and the second frequency band is the 5GHz high-frequency band;

[0016] Instructions transmitted via the first frequency band employ forward error correction coding, while image data transmitted via the second frequency band uses a hybrid automatic repeat request (HARQ) mechanism for error control.

[0017] When the receiving end detects an error in the received image data, it uses FEC to correct the error; if the data cannot be recovered after error correction, it sends a retransmission request to the sending end.

[0018] Furthermore, the time slot allocation is based on the TDMA mechanism, including broadcast time slots, uplink unicast command time slots, downlink feedback time slots, and image data return time slots; the length of each time slot is dynamically calculated and determined according to the length of the corresponding command frame or data frame and the nominal transmission rate of the channel.

[0019] Furthermore, during the instruction and data transmission process, a time-slot-based transmission scheduling mechanism is adopted. The ground control station allocates a fixed time slot to each UAV, and the UAV array receives instructions and feeds back the execution results of the control instructions from the ground control station within the corresponding time slot.

[0020] The ground control station sends control commands to the UAV, which then initiates the data acquisition process according to the commands. The acquired image data is compressed and encoded; the encoded data stream is divided into three data segments, which are modulated sequentially onto three carrier frequencies of 5.15GHz, 5.35GHz and 5.75GHz and transmitted in parallel.

[0021] Furthermore, the instruction frame structure includes five fields: frame header, target ID, instruction type, instruction content, and checksum. The frame header is used to identify the start and synchronization of the instruction frame, the target ID field is used to specify the target of the response instruction, the instruction type is used to clearly indicate the instruction and operation type, and the checksum field is used to verify the integrity and accuracy of the instruction frame to ensure that the instruction has not been tampered with or damaged during transmission.

[0022] The command types include broadcast commands and unicast commands. Broadcast commands include power-on self-test broadcast commands, system initialization broadcast commands, recovery preparation broadcast commands, and power-off recovery broadcast commands, which are used for time slot allocation, position preset, and UAV status self-test at the beginning of system operation.

[0023] Unicast commands include uplink control commands and downlink status commands, used to control individual UAVs in the UAV array during system operation. Uplink control commands are commands from the ground control station to the UAV, including position adjustment commands, emergency avoidance commands, image acquisition commands, handshake commands, and retransmission commands. Downlink status commands are commands from the UAV to the ground control station, including status feedback commands, execution result commands, priority adjustment request commands, self-test feedback commands, and data feedback commands.

[0024] Furthermore, the frame structure of the power-on self-test broadcast command includes a time slot allocation priority list, which enables the ground control station to allocate downlink transmission time slots to different UAVs according to their priority levels;

[0025] The frame structure of the system initialization broadcast command contains a list of preset position information, which is used to assign three-dimensional target coordinates, start time and path to each UAV;

[0026] Both the recycling preparation broadcast command and the power-off recycling broadcast command contain queue or queue priority information in their frame structures for dynamically adjusting the recycling order. In addition, the power-off recycling broadcast command includes the recycling start time.

[0027] Furthermore, the position adjustment command is used to instruct the UAV to adjust the three-dimensional coordinates of the target position, and the adjustment path node includes the position information of two transit nodes, which are used to instruct the UAV to adjust the coordinate path.

[0028] Emergency avoidance commands are configured as high-priority commands to indicate the three-dimensional coordinates of the target location for the UAV to avoid, interrupting or preempting ongoing normal command transmission time slots;

[0029] Image acquisition commands are used to instruct the image acquisition parameters of the UAV. Image acquisition parameters include acquisition parameters used to control the image acquisition equipment and image data processing operations.

[0030] The priority adjustment request instruction includes power status information that indicates the urgency of drone recovery, which is used to trigger the ground control station to dynamically adjust the recovery priority during the shutdown recovery phase.

[0031] The execution result command is sent within the command time slot assigned to the UAV and is used by the UAV slave station to feed back the command execution result to the ground master station during the mission execution phase.

[0032] Status feedback commands are sent within the command time slot assigned to the UAV and are used to instruct the UAV slave station on real-time status monitoring during the mission execution phase.

[0033] The self-test feedback command is used during the power-on self-test phase to provide feedback on the self-test results of the drone slave station and to count the number of valid drone devices.

[0034] The data return command is a command sent in parallel after the acquired data is segmented and framed during the task execution phase. It is used for high-definition image data return.

[0035] Furthermore, before transmitting image data back via the second frequency band, a handshake mechanism is performed, including:

[0036] The ground control station sends a handshake command to the target UAV. The target UAV responds to the handshake command and reports its own status information. After confirming that it can receive the command, it sends an image acquisition command.

[0037] Furthermore, the image data transmitted back via the second frequency band is as follows: the target UAV divides the image data stream into multiple data segments and modulates them onto different subcarrier frequencies within the second frequency band for parallel transmission.

[0038] Another objective of this invention is to provide a ground control station for implementing the aforementioned TDMA-based dual-frequency master-slave UAV array communication networking method, the ground control station comprising:

[0039] The control module is used to generate various commands, execute TDMA-based time slot scheduling, process UAV feedback information, and dynamically adjust recovery priorities.

[0040] The first communication module is configured to operate in the 433MHz frequency band and is used to send all broadcast commands and uplink control commands, and to receive all downlink status commands.

[0041] The second communication module is configured to operate in the 5GHz band and is used to receive image data transmitted back by the drone.

[0042] Another object of the present invention is to provide a drone for implementing the aforementioned TDMA-based dual-frequency master-slave drone array communication networking method, the drone comprising:

[0043] The first airborne communication module is configured to operate in the 433MHz frequency band to receive broadcast commands and uplink control commands from the ground control station, and to send downlink status commands within the allocated feedback time slots.

[0044] The image acquisition and processing module is used to respond to image acquisition commands, acquire and compress image data;

[0045] The second airborne communication module is configured to operate in the 5GHz band and is used to transmit compressed image data back to the ground control station.

[0046] Combining all the above technical solutions, the beneficial effects of this invention are as follows:

[0047] First, based on the core design of this invention, which combines "dual-frequency separation + TDMA time slot allocation + master-slave architecture + complete networking process", and considering the pain points of existing technologies, the following beneficial effects are extracted from the dimensions of communication reliability, transmission efficiency, universality, and operability:

[0048] 1. Dual-frequency separation design breaks through the performance bottleneck of single-frequency transmission.

[0049] Achieving parallel and conflict-free command and data transmission: A dedicated 433MHz band carries various control commands (including broadcast, uplink, and downlink commands), while the 5G band is dedicated to high-capacity data backhaul, completely resolving the issue of command and data sharing and competing for channels in single-frequency systems. The 433MHz band boasts strong anti-interference capabilities and long propagation distances, ensuring the real-time performance and reliability of command transmission. The 5G band offers ample bandwidth to meet the high-speed transmission needs of massive amounts of data such as high-resolution images and real-time video. Employing three-way parallel data transmission, the data transmission rate is more than three times higher than that of a single-frequency system (taking the 2.4G band as an example, with a maximum transmission bandwidth of 40MHz). If the 5G band uses an 80MHz transmission bandwidth, the rate increase can reach five times.

[0050] Reduce frame structure design complexity: Command frames and data transmission frames are designed independently to adapt to the characteristics of their respective frequency bands, eliminating the need to consider the compatibility of the two types of services. This simplifies the frame structure logic and parsing process, reduces the command processing latency of the UAV terminal to the millisecond level, and improves the response speed of cluster scheduling.

[0051] 2. TDMA time slots are precisely allocated, improving network resource utilization.

[0052] Differentiated time slots adapt to diverse service needs: For 14 types of command frame structures and data transmission characteristics, a dedicated time slot allocation strategy is designed. Broadcast time slots enable the master station to achieve unified management and control of the cluster (such as power-on self-test and time slot allocation), while unicast time slots ensure point-to-point command interaction (such as position adjustment and status feedback), avoiding command conflicts and network congestion. Channel utilization can be improved by more than 100% compared to the traditional fixed time slot allocation method.

[0053] Among the beneficial effects of this invention, the conclusion that "channel utilization is improved by more than 100% compared with the traditional fixed time slot allocation method" is derived based on the channel utilization calculation formula in the TDMA networking scenario and the time slot allocation design of this invention. The specific calculation method and comparison logic are as follows:

[0054] I. Core Calculation Model for Channel Utilization (Generally applicable to TDMA scenarios)

[0055] Channel utilization (η) refers to the proportion of actual effective transmission time to the total channel resource time, and is calculated using the following formula:

[0056]

[0057] The key parameters are defined as follows:

[0058] 1. The total duration of a TDMA frame is the sum of the duration of all time slots allocated to UAV nodes plus the time slot protection interval (idle time to avoid interference between time slots).

[0059] 2. The actual time for all nodes to transmit useful data / instructions within their respective time slots must be deducted from the synchronization overhead, frame header overhead, and invalid waiting time (such as idle time caused by the mismatch between time slot allocation and business needs in traditional solutions).

[0060] Assuming a total timeslot period of 100ms, no guard interval, 10 UAV nodes, a fixed timeslot length of 10ms per UAV, and an effective transmission time of 5ms for each node (1ms command + 4ms data), with 10 nodes operating and 5 nodes transmitting data, the total effective transmission time is... Corresponding channel utilization Fixed time slot allocation mechanism,

[0061] For the dual-frequency cooperative dynamic time slot allocation mechanism proposed in this invention, under the same conditions (1ms for instruction transmission and 4ms for data transmission per node, 10 nodes working, and 5 nodes transmitting data), the instruction time slot (433MHz) and data time slot (5G) operate independently and in parallel, without the need for time-division transmission, and the effective transmission time is superimposed:

[0062] Assume the total time slot period of the command channel (Includes instruction time slots for 10 nodes, with an average of 5ms per time slot and no idle time); Total time slot period for the data channel (Includes 10 data time slots, each time slot is 5ms, with no idle time); Effective transmission time of the command channel Effective transmission time of data channel Under dual-frequency parallel operation, the channel utilization rate of this invention is... Furthermore, this invention allocates time slots according to service type (broadcast / unicast), and the instruction time slots are dynamically adjusted according to the actual instruction length; the master station adjusts the number of time slots according to the number of nodes and service load, which can avoid the idleness of dedicated time slots and enhance the flexibility of time slot allocation.

[0063] Supports dynamic adaptation of cluster size: Based on master-slave architecture, the ground master station can flexibly adjust the time slot length and allocation scheme according to the number of drones joining the network and changes in business load. It can not only adapt to the efficient communication of small-scale drone formations (3-5 drones) but also meet the collaborative operation needs of large-scale clusters (20 or more drones), solving the problem of insufficient universality of existing solutions.

[0064] 3. The master-slave architecture, combined with the complete networking process, ensures the stability, controllability, and operability of the system.

[0065] Simplify network management and reduce collaboration complexity: Adopting a master-slave architecture of "centralized control by ground master station and collaborative execution by UAV slaves", the master station is responsible for network scheduling, time slot allocation, command issuance and data aggregation. Slave devices do not need to participate in routing calculations or complex negotiations, but only need to respond to the master station's commands, which significantly reduces the hardware power consumption and software complexity of UAV terminals and greatly improves the success rate of cluster networking.

[0066] The entire process ensures network reliability: It covers three complete stages: power-on networking, task execution, and power-off recovery. It clarifies the instruction interaction logic and time slot allocation rules for each stage, solving the problem that existing technologies only provide network stage divisions and lack practical details.

[0067] 4. High versatility and strong adaptability expand the boundaries of application scenarios.

[0068] Adaptable to various scenarios and general needs: Breaking away from the scenario limitations of existing emergency communication solutions, through a flexible time slot adjustment mechanism and standardized command frame structure, it can seamlessly adapt to various drone swarm collaboration scenarios such as environmental monitoring, on-site monitoring, agricultural plant protection, and low-altitude inspection, without the need for major modifications for specific scenarios, thus reducing the cost of technology implementation.

[0069] Compatible with dual-band hardware ecosystem: Both the command band (433MHz) and the data band (5G) are mature and universal frequency bands, making hardware selection convenient and highly compatible. It can be directly adapted to the communication modules of existing mainstream drones without the need for additional dedicated hardware development, significantly enhancing the promotion and application value of the solution.

[0070] 5. The instruction set is improved, enhancing the precision of cluster control.

[0071] The multi-dimensional command system covers the entire business process: It is designed with 14 types of commands in three categories: broadcast, uplink, and downlink, covering the entire process requirements such as power-on self-test, time slot allocation, position adjustment, emergency avoidance, data transmission, and status feedback. It solves the problem of lack of command control details in the existing dual-frequency solution and realizes precise management and control of the entire life cycle of the UAV swarm "from networking to recovery".

[0072] Priority adaptation ensures critical business operations: Through instruction classification design, critical instructions such as emergency avoidance and handshake instructions can occupy time slot resources first, ensuring that core control needs are responded to first in complex environments (such as sudden interference and terrain obstruction), thereby improving the safety and stability of cluster operations.

[0073] Secondly, this invention continuously optimizes and adjusts the protocol parameters, dynamically adjusting the time slot length and interval of command transmission according to the number of UAVs and the flight environment in actual tasks; and adjusting the priority and retransmission strategy in the data transmission protocol according to the real-time nature and importance of image data. Through these implementation steps and optimization measures, this invention ensures that the UAV array command communication and data transmission protocol system can operate efficiently and stably in practical applications, meeting the communication needs of UAV arrays in various complex tasks.

[0074] Third, this invention utilizes the mature and universal 433MHz (command) and 5G (data) frequency bands, eliminating the need for dedicated hardware modules and allowing direct compatibility with existing mainstream drone communication terminals. The master-slave architecture simplifies the software logic and computing power requirements of drone terminals, significantly reducing equipment procurement and maintenance costs for swarm operations, thus possessing a strong market promotion advantage. This communication method can be widely applied in environmental monitoring, low-altitude inspection, agricultural plant protection, emergency rescue and disaster relief, aerial filming, and many other fields, providing drone companies with a "high-reliability + high-speed + low-cost" networking solution. Simultaneously, it can generate value-added services such as time-slot scheduling algorithm licensing and customized networking scheme design, forming a diversified profit model of software licensing and joint technical services, with enormous commercial monetization potential.

[0075] This invention fills the gap in a complete technical solution for TDMA networking under dual-frequency cooperation: In existing technologies, dual-frequency UAV communication systems only propose a frequency band separation architecture concept, without specifying the networking process, time slot allocation strategy, and command frame structure details; while TDMA networking solutions are all limited to single-frequency transmission, lacking adaptation design for dual-frequency architecture. This invention, for the first time, constructs a complete technical system of "dual-frequency separation transmission + precise TDMA time slot allocation + master-slave full-process networking," clarifying 14 command frame structures, a three-stage networking process, and differentiated time slot allocation rules, filling the technical gap of no complete practical solution for TDMA networking in dual-frequency scenarios.

[0076] This invention successfully solves the multiple requirements of achieving highly reliable command control, high-speed data transmission, low-complexity network management, and wide-scenario adaptability in large-scale UAV collaborative scenarios. For a long time, the industry has faced a persistent technical dilemma: while single-frequency TDMA solutions can avoid command conflicts, they are limited by half-duplex mode and cannot simultaneously guarantee command real-time performance and high-speed data transmission; dual-frequency architectures can separate command and data transmission, but lack supporting network construction processes and time slot allocation mechanisms, making them difficult to implement; general-purpose network solutions generally suffer from the problem of "adaptability and performance being mutually exclusive." These challenges have hindered the upgrade of UAV swarms from small-scale formations to large-scale collaboration, representing a long-standing technical bottleneck that the industry has urgently needed to overcome.

[0077] This invention successfully overcomes the significant technical bias in the field of UAV swarm communication that "general-purpose UAV networking solutions must sacrifice performance adaptability," providing a new direction for industry technological development. The industry generally believes that UAV swarm communication scenarios vary greatly (e.g., emergency rescue requires rapid network construction, environmental monitoring requires stable transmission, and low-altitude inspection requires high-speed data transmission). General-purpose networking solutions must simplify technical design and reduce performance indicators to cover multiple scenarios, leading to the misconception that "generality and high performance are mutually exclusive." This invention, through an innovative combination of "dynamic time slot allocation + standardized command system + dual-frequency performance adaptation," ensures universal adaptability to multiple scenarios through standardized networking processes and command frame structures. It also meets the core performance requirements of reliable commands and high-speed data transmission in different scenarios through dual-frequency separation design. Furthermore, it adapts to the service load characteristics of different scenarios through differentiated time slot allocation strategies, achieving a synergistic unity of generality and high performance. This breaks the industry's perception of the performance limitations of general-purpose networking solutions and provides a new path for the development of large-scale general-purpose UAV swarm communication technology. Attached Figure Description

[0078] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure;

[0079] Figure 1 This is a flowchart of a TDMA-based dual-frequency master-slave UAV array communication networking method provided in an embodiment of the present invention;

[0080] Figure 2 This is a schematic diagram of broadcast instructions provided in an embodiment of the present invention;

[0081] Figure 3 This is a schematic diagram of a unicast instruction provided in an embodiment of the present invention;

[0082] Figure 4 This is a diagram of the power-on self-test broadcast command frame structure provided in an embodiment of the present invention;

[0083] Figure 5 This is a system initialization broadcast instruction frame structure diagram provided in an embodiment of the present invention;

[0084] Figure 6 This is a diagram of the structure of the recycling preparation broadcast instruction frame provided in an embodiment of the present invention (where N is at most 13; the dashed part is an optional step).

[0085] Figure 7 This is a diagram of the power-off recycling broadcast instruction frame structure provided in an embodiment of the present invention (the dashed part is an optional component).

[0086] Figure 8 This is a structural diagram of the position adjustment instruction frame provided in an embodiment of the present invention;

[0087] Figure 9 This is a structural diagram of the emergency avoidance command frame provided in an embodiment of the present invention;

[0088] Figure 10 This is a structural diagram of the image acquisition instruction frame provided in an embodiment of the present invention;

[0089] Figure 11 This is a diagram of the handshake instruction frame structure provided in an embodiment of the present invention;

[0090] Figure 12 This is a diagram of the retransmission instruction frame structure provided in an embodiment of the present invention;

[0091] Figure 13 This is a diagram of the state rotation instruction frame structure provided in an embodiment of the present invention;

[0092] Figure 14 This is a diagram of the execution result instruction frame structure provided in an embodiment of the present invention;

[0093] Figure 15 This is a structure diagram of the priority adjustment request instruction frame provided in an embodiment of the present invention;

[0094] Figure 16 This is a diagram of the self-test feedback instruction frame structure provided in an embodiment of the present invention;

[0095] Figure 17 This is a data rollback instruction frame structure diagram provided in an embodiment of the present invention;

[0096] Figure 18 This is a power-on networking time slot allocation diagram provided in an embodiment of the present invention (the dashed lines represent optional steps);

[0097] Figure 19 This is a power-on initialization flowchart provided in an embodiment of the present invention (M represents a maximum of 100 drones, and X represents the number of drones that normally provide feedback).

[0098] Figure 20 This is a service stage time slot allocation diagram provided in an embodiment of the present invention (the dashed part is an optional stage);

[0099] Figure 21 This is a time slot allocation diagram for the shutdown and recycling phase provided in an embodiment of the present invention (N is a maximum of 13 in the diagram). Detailed Implementation

[0100] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Many specific details are set forth in the following description to provide a thorough understanding of the present invention. However, the present invention can be practiced in many other ways different from those described herein, and those skilled in the art can make similar modifications without departing from the spirit of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0101] The innovation of the TDMA-based dual-frequency master-slave UAV array communication networking method provided in this invention is as follows:

[0102] First, this invention proposes for the first time a dual-frequency separation and centralized master-slave management network architecture, addressing the pain points of single-frequency and distributed networking. Specifically, it explicitly defines the 433MHz band as dedicated to carrying all types of control commands (broadcast, uplink, and downlink commands), while the 5G band is dedicated to massive data backhaul. This achieves physical channel separation between command and data transmission, completely resolving the core contradiction of shared channels and mutual contention in traditional single-frequency systems. Simultaneously, it leverages the strong anti-interference capabilities of 433MHz to ensure command reliability and utilizes the high bandwidth of 5G to meet high-speed data transmission requirements, achieving a balance between reliability and efficiency. A strictly centralized management model is constructed for the ground master station and UAV slaves. The ground master station is uniformly responsible for network scheduling, time slot allocation, command issuance, and data aggregation. UAV slaves do not need to participate in complex routing negotiations or time slot contention; they only need to respond to the master station's commands to complete collaborative operations. This simplifies the hardware complexity and software logic of UAV terminals, while avoiding the path optimization challenges of distributed networking, significantly improving the stability and controllability of cluster networking.

[0103] Secondly, a standardized command system adapted to the three-stage work cycle was designed. For the first time, a complete networking process encompassing three stages—power-on networking, task execution, and power-off recovery—was proposed. This clearly defines the core objectives, command interaction logic, and time slot allocation rules for each stage, supporting scenarios such as rapid synchronization when UAVs enter the network late and reconnection after a connection failure. It addresses the problem of existing technologies only dividing the networking stage and lacking detailed operational procedures for the entire process, achieving standardized management and control of the cluster throughout its entire lifecycle from startup to recovery. A standardized command system of fourteen types in three categories was constructed. Innovatively, commands are categorized by transmission method into broadcast commands, uplink unicast commands, and downlink unicast commands. Functionally, it covers all business scenarios such as power-on self-test, time slot allocation, position adjustment, emergency avoidance, data backhaul, and status feedback. Each command corresponds to a dedicated frame structure design, filling the technical gap in existing dual-frequency communication systems that lack specific command control details, achieving precise and standardized cluster management and control.

[0104] Third, this invention proposes a differentiated dynamic time slot allocation strategy for TDMA that adapts to dual-frequency and diverse service requirements. A precise matching mechanism for service type, frequency band characteristics, and time slot resources is designed: Dedicated TDMA time slots are configured for the global synchronization requirements of broadcast commands and the point-to-point interaction requirements of unicast commands with dual-frequency communication. Broadcast time slots are issued at fixed intervals, while unicast time slots are allocated according to node services, enabling dual-frequency parallel operation. This significantly improves channel utilization compared to traditional fixed time slot allocation methods. Dynamic adaptation to cluster size and service load is achieved: the ground master station can flexibly adjust the number, length, and allocation scheme of time slots based on the number of UAVs joining the network. This adapts to both efficient communication in small formations of 3-5 UAVs and the collaborative operation needs of large-scale clusters of 20 or more UAVs, overcoming the bottleneck of limited node expansion capabilities in traditional TDMA solutions.

[0105] Fourth, a technical solution that balances generalization and loose coupling is proposed to improve scenario adaptability and feasibility. A design strategy combining standardized design and dynamic adjustment is adopted to achieve universal adaptability across multiple scenarios: through standardized command frame structures, time slot allocation rules, and networking processes, it can seamlessly adapt to various drone swarm collaboration scenarios without significant modifications for specific scenarios (environmental monitoring, emergency rescue, agricultural plant protection, etc.), solving the problem that existing solutions are mostly designed for single scenarios and lack versatility. Based on existing hardware: using the mature and common 433MHz and 5G frequency bands, it can directly adapt to the communication modules of existing mainstream drones without the need to develop dedicated hardware. At the same time, the master-slave architecture reduces the computing power and power consumption requirements of the drone terminal, significantly lowering the hardware modification costs and promotion barriers for technology implementation, and possessing strong engineering practice value.

[0106] Example 1: The instruction communication protocol for dual-frequency communication systems provided in this embodiment of the invention adopts a TDMA-based communication networking method, which specifies the format, process, timing and control mechanism of instruction and data transmission to ensure that the various parts of the dual-frequency communication system can work together efficiently and reliably.

[0107] Example 1: Small-to-medium scale UAV array networking application in environmental monitoring scenarios (3-5 UAVs);

[0108] I. Application Scenarios and Technology Adaptation; This embodiment targets an urban periphery ecological environment monitoring scenario, requiring four UAVs to form a collaborative array to complete vegetation coverage surveys and water quality sampling point image acquisition tasks within a 10km² area. Core requirements of the scenario: stable command transmission (avoiding flight path deviations due to building obstruction or electromagnetic interference), real-time data transmission (high-definition images need to be promptly synchronized to the ground station for on-site analysis), and a simple networking process (rapid deployment and task initiation). Technical solutions used:

[0109] 1. It adopts a combination of "433MHz command band + 5.15GHz / 5.35GHz data band", with 433MHz ensuring penetration and anti-interference, and dual data bands carrying high-definition video transmission in parallel;

[0110] 2. Following the three-stage process of power-on networking, mission execution, and power-off recovery, the ground master station centrally allocates TDMA time slots, with each UAV allocated one command time slot and one data time slot;

[0111] 3. The command system focuses on the specific needs of environmental monitoring, adding commands for sampling point positioning, image resolution adjustment, and flight path correction. Data transmission commands prioritize the transmission of images of areas with abnormal water quality.

[0112] II. Specific Application Process

[0113] 1. Power-on and network setup phase (≤20 seconds):

[0114] After powering on, the three drones automatically switch to the 433MHz frequency band and listen for the power-on self-test command broadcast by the ground master station, which includes a time slot allocation command (2.32ms per command time slot, no data time slot allocation). The drones that complete the self-test send self-test feedback commands in sequence according to the allocated time slots. The master station completes identity authentication and channel quality assessment through the protection time slots and allocates command time slot 1 (15~75ms), data time slot 1 (259.3ms), preset position and path to drone 1. The remaining drones are allocated subsequent time slots in sequence.

[0115] 2. Task execution phase:

[0116] Three drones flew to the designated location according to a preset path and time. Drone 1 collected vegetation images, Drone 2 collected water quality images, and Drone 3 was responsible for supplementing images in the edge areas. The main station sent data acquisition commands to Drone 2, which adjusted its shooting parameters and collected high-definition images. The data was transmitted back in parallel via the 5.15GHz and 5.35GHz dual-band frequencies, with a data transmission rate of 160Mbps. The ground station received the data in real time and analyzed the regional water quality. When encountering tall buildings that obstructed the signal, the 433MHz command band maintained stable communication. The main station adjusted the drone's flight altitude using "position adjustment commands" to avoid signal interruption.

[0117] 3. Shutdown and recycling phase:

[0118] After the mission is completed, the main station sends a broadcast recovery preparation command. The drone performs a status self-check and sends a priority adjustment request based on the actual status. The main station adjusts the shutdown recovery order and issues a shutdown recovery command. The drones shut down and are recovered in sequence according to the issued recovery order information and location information, completing the network closed loop.

[0119] III. Technical Problems Solved and Positive Effects

[0120] 1. Technical problems to be solved: Addressing the pain points of command interference and untimely data transmission in small-to-medium-scale array environmental monitoring; traditional single-frequency solutions suffer from channel congestion caused by control commands and data transmission commands, leading to untimely command response and data transmission, and command loss. For example, untimely response to emergency avoidance commands could directly cause damage or even crashes to drones.

[0121] 2. Positive effects: High command transmission success rate, with no command loss or delay issues caused by interference; high data return rate, using dual-band parallel transmission, and visible high-definition image transmission delay compression of 50%.

[0122] Example 2: Large-scale UAV array networking application in emergency rescue scenarios (20-40 UAVs)

[0123] I. Application Scenarios and Technology Adaptation

[0124] This embodiment targets a mountain earthquake relief scenario, deploying a search and rescue array of 30 drones to locate trapped individuals within a 30km² area, transmit disaster information, and guide the delivery of emergency supplies. Key requirements for this scenario include: high fault tolerance (communication remains possible even when some drones are obstructed by terrain), high-speed, high-volume data transmission (real-time transmission of multiple search and rescue video feeds), priority response to emergency commands (e.g., dispatching rescue efforts after the discovery of trapped individuals), and support for delayed network entry (allowing subsequent reinforcement drones to quickly join the network). Technical adaptation solution:

[0125] 1. It adopts a combination of 433MHz command frequency band and three data frequency bands of 5.15GHz / 5.35GHz / 5.75GHz. The 433MHz frequency band strengthens the relay mechanism, and the data frequency band is classified and carried according to search and rescue video (5.15GHz), disaster data (5.35GHz) and delivery guidance data (5.75GHz);

[0126] 2. TDMA time slots are dynamically adjusted, adding 5 emergency time slots and 3 late network access time slots in addition to the original instruction time slots and data time slots;

[0127] 3. The command system has added instructions for locating and reporting trapped personnel, and instructions for dispatching emergency reinforcements and coordinates for delivering supplies. The instructions for dispatching emergency reinforcements have the highest priority and can take over ordinary time slots.

[0128] II. Specific Application Process

[0129] 1. Power-on and Networking Phase (≤60 seconds): After the first batch of 27 drones are powered on, they enter a 433MHz listening state. The ground master station sends emergency networking instructions via broadcast time slots, including a time slot allocation table, preset locations and paths, and relay node election rules (the 3 drones with the strongest signals become relay nodes). After the drones complete handshake authentication, the master station allocates instruction time slots and corresponding data frequency band time slots to each drone. Relay nodes are additionally allocated relay time slots (for forwarding data from drones that are being blocked). 30 minutes later, 3 reinforcement drones are powered on and send network access requests via late network access time slots. The master station dynamically adjusts the time slot allocation table, adding 3 sets of instruction-data time slots. The reinforcement drones complete the network setup within 20 seconds without interrupting the existing search and rescue mission.

[0130] 2. Mission Execution Phase: 27 drones were deployed to conduct search and rescue operations in grid-based areas. Drone 7 located trapped personnel in a valley and sent a personnel location command via an emergency command time slot. The main station immediately dispatched three surrounding drones to coordinate image capture. Image data was transmitted back in real time via the 5.15GHz band at a transmission rate of 80Mbps, and the ground rescue team simultaneously received the location and on-site images. Drone 12 lost direct communication with the main station due to mountain obstruction and automatically switched to relay mode. It forwarded commands via the 433MHz time slot of Drone 5 (relay node), and the main station guided Drone 12 to deliver emergency supplies to the trapped personnel by sending coordinate commands. Massive amounts of disaster data (terrain damage, road blockages) were transmitted via dual-band 5.35GHz and 5.75GHz without data loss or delay, and the ground command center generated a disaster distribution map in real time.

[0131] 3. Shutdown and Recovery Phase: After the search and rescue mission is completed, the main station sends a shutdown preparation command. The drones are then recovered in formation according to the principle of shutting down drones in unobstructed areas first, followed by drones in relay nodes and valley areas, to avoid data residue caused by simultaneous shutdown. After the drones confirm the shutdown, the main station broadcasts the shutdown and recovery command. The drones are then shut down and recovered in formation according to the recovery order and location information, completing the network loop.

[0132] III. Technical problems solved and positive effects:

[0133] 1. Technical problems to be solved: pain points such as insufficient network resilience, slow response to emergency commands, and slow transmission of massive amounts of data in large-scale emergency disaster relief; serious problems such as communication interruption caused by obstruction and channel congestion caused by emergency commands and data in traditional single-frequency solutions.

[0134] 2. Positive Effects: The relay mechanism significantly improves the communication success rate in complex terrain compared to traditional solutions; the emergency command response latency is less than 3ms, buying critical time for rescue; the three-data-band classification and carrying ensures uncongested transmission of massive data, with disaster data and search and rescue videos being transmitted back synchronously, improving rescue decision-making efficiency by more than 60%; the late-entry network design supports dynamic personnel addition without restarting the network, adapting to the dynamic needs of emergency disaster relief, and its networking flexibility is significantly better than existing fixed-time-slot solutions.

[0135] like Figure 1 As shown, the dual-frequency communication system command transmission process provided in this embodiment of the invention is divided into three stages. The first stage involves the master station continuously broadcasting commands for 10 seconds after device initialization, used for system status self-checking and initialization settings. The second stage is the task execution process, where the master station randomly sends unicast uplink commands according to task requirements, instructing the UAV to adjust its attitude, perform emergency avoidance, and transmit data back. The third stage is the task completion and UAV recovery / shutdown stage, where the master station broadcasts recovery preparation and recovery / shutdown commands to complete the recovery and shutdown of the UAV array.

[0136] Typical exception handling process:

[0137] 1. The drone has lost contact;

[0138] During the mission execution phase, if the drone is unable to communicate with the main station due to mountain obstruction, urban high-rise buildings blocking the way, or electromagnetic interference, and the duration exceeds 3 TDMA frame cycles (e.g., 1 second), the main station will trigger the disconnection anomaly handling mechanism.

[0139] If the master station fails to receive status feedback from the target drone within three consecutive command slots after sending a handshake command, and no corresponding data is received in the data slots, the drone is deemed to be out of contact. The drone's ID and allocated time slot resources are then marked. The master station broadcasts a relay search command via a time slot, notifying 3-5 normally communicating drones (pre-defined as candidate relay nodes) in the vicinity of the drone's location to activate their 433MHz band listening mode and search for the missing drone's signal. Once a relay node detects the missing drone's signal, the master station forwards a reconnection command, including the latest time slot allocation, through the relay node. After the missing drone successfully reconnects, normal operational procedures resume.

[0140] Upon detecting a communication interruption with the master station, the UAV automatically switches to disconnection mode, shortening the listening cycle of the 433MHz command band and sending a distress signal (including its GPS location and remaining battery power) every 100ms. Upon receiving a relay instruction from a nearby relay node, the UAV completes a two-way handshake with the master station through the relay node, synchronizing the latest time slot parameters and re-establishing the communication link (reconnection time ≤ 2 seconds). After successful reconnection, the UAV immediately reports the mission execution status during the disconnection period. The master station adjusts subsequent mission instructions based on the feedback to avoid duplicate or missed tasks. If reconnection fails for 10 consecutive minutes, the UAV initiates an autonomous return-to-home process, flying along a preset return route while continuously sending distress signals. Upon returning to a signal coverage area, it automatically reconnects to the master station. If the return-to-home fails, an emergency landing command is triggered to ensure equipment safety.

[0141] Based on the above solutions, this invention can achieve communication recovery in complex environments through relay scheduling, and the time slot resource reservation mechanism avoids the time consumption of reconnection and re-networking. The latency of task recovery after reconnection is no more than 5 seconds, effectively controlling the loss of task efficiency. The design of UAV autonomous return and emergency landing can effectively reduce the risk of equipment loss and can cope with the problems of task interruption and equipment damage after UAV loses connection in traditional solutions.

[0142] 2. Command verification failed;

[0143] After receiving instructions from the master station, the UAV detects errors in the instruction frame data or that the instruction parameters are outside the reasonable range (such as abnormal position coordinates) through CRC check, triggering an instruction verification failure exception. At the end of the current instruction time slot, the UAV sends an instruction verification failure instruction (containing error codes: 01 → CRC error, 02 → parameter abnormality) via the 433MHz frequency band, while discarding the erroneous instruction and performing no operation.

[0144] When a command verification fails on the master station side, if the command frame verification fails, a retransmission command is immediately sent at the end of the current command time slot; if the data frame verification fails, a retransmission command is immediately sent in the next command cycle corresponding to the UAV's command time slot. If three consecutive command verifications fail, the master station determines that the channel is severely interfered with and initiates relay node forwarding of commands.

[0145] Therefore, this invention can effectively avoid operational accidents caused by UAVs executing incorrect commands (such as deviation from the flight path and collision with obstacles), and significantly improve the accuracy of command transmission; by using emergency retransmission and channel switching, the impact of interference on command interaction is reduced, and the probability of mission interruption is reduced.

[0146] The command frame structure consists of five parts: frame header, target ID, command type, command content, and checksum. The frame header identifies the start and synchronization of the command frame; the target ID field specifies the target of the response command; the command type field clearly indicates the command and operation type (e.g., broadcast command, takeoff command, status detection command, handshake command, interrupt retransmission command, position indication command, image acquisition command, etc.); different commands have their own parameter content. Taking the takeoff command as an example, its command parameter field specifies the detailed parameters of the operation (e.g., target coordinates, flight speed, formation parameters, etc.); and the checksum field verifies the integrity and accuracy of the command frame, ensuring that the command has not been tampered with or damaged during transmission.

[0147] Regarding the command communication protocol, the communication system involved in this invention adopts a master-slave communication architecture, that is, the ground control station acts as the master station, responsible for initiating, managing, and terminating the entire command transmission process. The UAV terminal performs corresponding operations according to the instructions of the ground control station.

[0148] Control commands include two types: broadcast commands and unicast commands. Broadcast commands are used for time slot allocation, position preset, and UAV status self-checks at the beginning of system operation. Unicast commands are used for controlling individual UAVs in the UAV array during system operation. Broadcast commands include power-on self-check broadcast commands, system initialization broadcast commands, recovery preparation broadcast commands, and power-off recovery broadcast commands. See [link to command classification] for details. Figure 2 .

[0149] Unicast commands include two types: uplink commands and downlink commands. Uplink commands are those from the ground control station (master station) to the UAV (slave station), including position adjustment commands, emergency avoidance commands, image acquisition commands, handshake commands, and retransmission commands. Downlink commands are those from the UAV (slave station) to the ground control station (master station), including status feedback commands, execution result commands, priority adjustment request commands, self-test feedback commands, and data feedback commands. See [link to command classification] for details. Figure 3 .

[0150] Before each data transmission command, the master station first sends a handshake signal to the UAV, informing it of the type and length of the command to be sent. Upon receiving the handshake signal, the UAV immediately responds, providing its own status information (such as battery level, flight attitude, and communication link quality) and whether it is ready to receive commands. This handshake mechanism effectively avoids misreception and conflicts during command transmission, improving the reliability of command transmission.

[0151] During command transmission, the system employs a time-slot-based transmission scheduling mechanism. The master station allocates a dedicated time slot to each UAV based on the size of the UAV array and mission requirements. Slave stations can then report the execution results of the master station's control commands within their corresponding time slots. The master station sends control commands to the UAV, and the UAV receives the commands and reports the execution results within its own time slot. For specific time slot allocation details, see [link to details]. Figure 18 , 20 21. This time slot allocation mechanism not only ensures the orderly transmission of commands, but also avoids interference between command signals from different UAVs, thus improving system control and information transmission performance.

[0152] In addition, the command communication protocol is designed with idle time slots and preemption mechanisms to deal with emergencies and urgent tasks. For example, during a normal flight mission, if an obstacle or other emergency suddenly appears, the ground control station can immediately send a high-priority emergency avoidance command. This command can preempt the communication channel and be quickly transmitted to the UAV, enabling the UAV to take timely obstacle avoidance or emergency measures to ensure flight safety.

[0153] Regarding the data transmission protocol, a time-slot-based transmission scheduling mechanism is still used. The slave station only transmits image data within its allocated time slot, effectively solving the spectrum congestion problem. The UAV initiates the data acquisition process according to control commands. First, the acquired image data is compressed and encoded to reduce the amount of data transmitted. The encoded data stream is then divided into three data segments and framed (including adding frame headers, redundant bits, and checksums), and modulated onto three carrier frequencies of 5.15GHz, 5.35GHz, and 5.75GHz respectively for parallel transmission.

[0154] The uplink data acquisition command frame includes a frame header, target ID, image acquisition parameters, and checksum. The image acquisition parameters include parameters such as camera focal length, aperture, angle, and frame rate, as well as operations such as annotation, storage, and playback of image data, so that ground operators can remotely control and manage the image acquisition and transmission of the UAV.

[0155] The command frame structure consists of five parts: frame header, target ID, command type, command content, and checksum. The frame header identifies the start and synchronization of the command frame; the target ID field specifies the target of the response command; the command type field clearly indicates the command and operation type (e.g., broadcast command, takeoff command, status detection command, handshake command, interrupt retransmission command, position indication command, image acquisition command, etc.); different commands have their own parameter content. Taking the takeoff command as an example, its command parameter field specifies the detailed parameters of the operation (e.g., target coordinates, flight speed, formation parameters, etc.); and the checksum field verifies the integrity and accuracy of the command frame, ensuring that the command has not been tampered with or damaged during transmission.

[0156] In addition, to improve the reliability and efficiency of command transmission, Hybrid Automatic Repeat Request (HARQ) technology is adopted. HARQ technology combines the advantages of forward error correction coding and automatic repeat request. When the receiver detects an error in the received image data, it first attempts to correct the error using FEC. If the data cannot be recovered after error correction, it sends a retransmission request to the sender.

[0157] Example 2: The dual-frequency master-slave UAV communication system provided in this embodiment of the invention uses two independent frequency bands for command communication and data transmission respectively. Due to the significant difference between the amount of command information and image transmission data, and the different requirements for data transmission bandwidth and rate, a lower frequency is typically used to transmit control commands, while a higher frequency band is used to transmit image data. The low frequency is at the MHz level; for example, the 433MHz wireless communication band can achieve a single-frequency communication bandwidth of up to 100 Mbps. The high frequency uses the 5G communication band, with a maximum single-frequency bandwidth of up to 80 Mbps. The communication system adopts a master-slave communication architecture. The ground control station acts as the sole master station, responsible for the unified management and scheduling of the entire UAV array's flight missions and operational commands. The UAV array acts as a slave station, receiving allocation and management from the master station and executing corresponding tasks based on the received command information. Hardware requirements: Both the ground control master station and the UAV slave stations must be equipped with wireless transmission modules and image transmission modules. The system integrates adaptive modulation and demodulation technology, which can adjust the modulation method according to channel conditions and transmission requirements to maximize system transmission efficiency. To improve the reliability and anti-interference capability of command transmission, the system also employs forward error correction coding (FEC) technology. At the sending end, the command data is encoded and redundant information is added, enabling the receiving end to correct errors and recover the original command data even if some data is corrupted. Furthermore, the system incorporates an Automatic Repeat Request (ARQ) mechanism. When the receiving end detects an error in a received command frame, it sends a retransmission request to the sending end. Upon receiving the request, the sending end quickly retransmits the command frame, ensuring the accuracy of the returned command information.

[0158] Example 3: This embodiment of the invention provides a ground control station for implementing the TDMA-based dual-frequency master-slave UAV array communication networking method. The ground control station includes:

[0159] The control module is used to generate various commands, execute TDMA-based time slot scheduling, process UAV feedback information, and dynamically adjust recovery priorities.

[0160] The first communication module is configured to operate in the 433MHz frequency band and is used to send all broadcast commands and uplink control commands, and to receive all downlink status commands.

[0161] The second communication module is configured to operate in the 5GHz band and is used to receive image data transmitted back by the drone.

[0162] Example 4: This embodiment of the invention provides a drone for implementing the TDMA-based dual-frequency master-slave drone array communication networking method. The drone includes:

[0163] The first airborne communication module is configured to operate in the 433MHz frequency band to receive broadcast commands and uplink control commands from the ground control station, and to send downlink status commands within the allocated feedback time slots.

[0164] The image acquisition and processing module is used to respond to image acquisition commands, acquire and compress image data;

[0165] The second airborne communication module is configured to operate in the 5GHz band and is used to transmit compressed image data back to the ground control station.

[0166] To further demonstrate the positive effects of the above embodiments, the present invention conducts the following experiments based on the above technical solutions.

[0167] I. Implementation of communication protocols;

[0168] The key to implementing this communication protocol lies in the dynamic time slot allocation mechanism of the TDMA access method and the instruction frame structure design for dual-frequency communication. The dynamic time slot allocation mechanism design includes the time slot interval, the number of time slots, and the dynamic adjustment mechanism, which needs to be based on the instruction length and data transmission rate. The instruction frame structure design includes the content of each field and bit allocation, which needs to be based on the instruction type, instruction content, and carrier frequency.

[0169] The length and content of the commands vary depending on the command type. The dual-frequency communication data transmission system for UAV arrays described in this invention involves a total of 12 command types. Among them, there are 4 broadcast commands, which need to cover all UAV terminals. There are 10 unicast commands, broadcast to a single UAV. The specific frame structure is as follows:

[0170] II. Broadcast instructions;

[0171] 1. The power-on self-test broadcast command includes a frame header, command type, time slot allocation priority list, idle codeword, and checksum. The command length is a maximum of 1024 bits. bits The frame structure consists of an 8-bit header, a 4-bit instruction type, a 1000-bit time slot priority allocation list (each 10 bits corresponds to one drone, with the first 3 bits specifying the drone's terminal priority and the last 7 bits indicating the downlink transmission time slot for each drone), an 8-bit checksum for frame verification, and 4 bits of free space. See the detailed frame structure below. Figure 4 .

[0172] 2. The system initialization broadcast command includes a frame header, command type, time slot allocation list, preset location information list, and checksum. The maximum command length is 10424 bits. bits The frame header is 8 bits long, the instruction type is 4 bits long, and the table length of the preset location information list is [missing information]. Each 96 bits corresponds to one UAV, used to indicate the UAV's 3D target position information; each UAV's time slot is allocated 8 bits, the check code is 8 bits long, used for frame verification; 4 bits are free codewords. See the detailed frame structure below. Figure 5 .

[0173] 3. The broadcast command for preparing to reclaim resources includes a frame header, command type, preset formation information, preset reclamation coordinates, reclamation time interval, idle bits, and checksum. The command length is [length missing]. bits (or The number of drone formations is 8 bits, where the frame header is 8 bits long and the instruction type is 4 bits long. Let the total number of drones in a formation be X, with 8 drones forming one formation. Drones less than one formation are formed into another formation. The total number of formations should be X. ( (Indicates an integer division upwards) A single drone formation uses 4 bits, and information on an X-drone formation uses... Used to indicate the final drone recovery team and priority; recovery is carried out in formation, with 8 drones recovered each time, requiring 8 coordinate information. Each preset recovery coordinate occupies 96 bits, therefore the preset recovery coordinate information occupies... The frame checksum is 8 bits long and is used for frame verification. If X is odd, the free codeword is 4 bits. If X is even, the free codeword is 0 bits. See the detailed frame structure below. Figure 6 .

[0174] 4. For the power-off recycling broadcast command, the following are included: frame header, command type, adjustment recycling queue, recycling start time, recycling time interval, idle codeword, and checksum. The command length is: ( bits (or The maximum instruction length is 448 bits. This includes an 8-bit frame header, a 4-bit instruction type, and information on adjusting and recycling queues. This is used to indicate the final drone recovery team and priority; the recovery start time occupies 20 bits, the recovery time interval is 8 bits, and the checksum is 8 bits long, used for frame verification. If X is odd, the idle codeword is 4 bits. If X is even, the idle codeword is 0 bits. See [link to frame structure] for details. Figure 7 .

[0175] Unicast commands are divided into uplink commands and downlink commands. Uplink commands include five types of control commands: position adjustment command, emergency avoidance command, image acquisition command, handshake command, and retransmission command. Downlink commands include five types of control commands: status feedback command, execution result command, priority adjustment request command, self-test feedback command, and data feedback command. The specific command structure design is as follows:

[0176] 1. For position adjustment commands, the command length is 320 bits, including an 8-bit frame header, a 4-bit command type, an 8-bit target ID, and 96 bits of target coordinates (indicating the UAV's adjustment of the target's 3D coordinates). The adjustment path nodes include two relay node position information, totaling 192 bits, used to indicate the UAV's coordinate adjustment path. The checksum is 8 bits long for frame verification, and there are 4 bits of free codewords. See the detailed frame structure below. Figure 8 .

[0177] 2. For emergency avoidance commands, the command length is 224 bits, including an 8-bit frame header, a 4-bit command type, an 8-bit target ID, and 96-bit avoidance coordinates indicating the 3D coordinates of the target location for the UAV. The avoidance path node includes a transit node location information, totaling 96 bits, used to indicate the UAV's coordinate adjustment path. The checksum is 8 bits long for frame verification, and there are 4 bits of free codewords. See the detailed frame structure below. Figure 9 .

[0178] 3. For image acquisition commands, the command length is 80 bits, including an 8-bit frame header, a 4-bit command type, an 8-bit target ID, a 16-bit acquisition frame rate, a 16-bit focal length level, an 8-bit gimbal rotation angle, and an 8-bit IMU data compensation, used to indicate UAV image acquisition parameters; an 8-bit checksum for frame verification; and 4 bits of free codewords. See the detailed frame structure below. Figure 10 .

[0179] 4. For handshake commands, the command length is 32 bits, including an 8-bit frame header, a 4-bit command type, an 8-bit target ID, a 1-bit information indicating connection success or failure, an 8-bit checksum for frame verification, and 3 bits of free codewords. See the detailed frame structure below. Figure 11 .

[0180] 5. For retransmission commands, the command length is 32 bits, including an 8-bit frame header, a 4-bit command type, an 8-bit destination ID, an 8-bit checksum for frame verification, and 4 bits of free space codewords. See the detailed frame structure below. Figure 12 .

[0181] 6. For status feedback commands, the command length is 40 bits, including an 8-bit frame header, a 4-bit command type, an 8-bit target ID, and 8 bits for battery status, motor status, sensor status, and (normal / abnormal) status (total 8 bits) used to indicate the UAV equipment status; an 8-bit checksum for frame verification; and 4 bits of free code. See the detailed frame structure below. Figure 13 .

[0182] 7. For execution result instructions, the instruction length is 32 bits, including an 8-bit frame header, a 4-bit instruction type, an 8-bit target ID, a 1-bit information indicating success or failure, an 8-bit checksum for frame verification, and 3 bits of free codewords. See the detailed frame structure below. Figure 14 .

[0183] 8. For priority adjustment request instructions, the instruction length is 32 bits, including an 8-bit frame header, a 4-bit instruction type, an 8-bit target ID, and a 4-bit battery status information. The remaining battery status indicates the urgency of recovery, prioritizing drones with longer return times or lower battery levels by setting their group to higher priority and arranging their recovery order accordingly to ensure safety and efficiency. The checksum is 8 bits long and used for frame verification. See the detailed frame structure below. Figure 15 .

[0184] 9. For self-test feedback commands, the command length is 32 bits. The frame header is 8 bits long, the command type is 4 bits long, the target ID is 8 bits long (used to identify the target UAV), the status feedback is 4 bits, and the checksum is 8 bits long (used for frame verification). See the detailed frame structure below. Figure 16 .

[0185] 10. For data return commands: The command length is 20736.032 kbits. This includes an 8-bit frame header, a 4-bit command type, an 8-bit target ID, a 20736 kbit image data information, an 8-bit checksum for frame verification, and a 4-bit idle code. See the detailed frame structure below. Figure 17 .

[0186] III. Time slot allocation method;

[0187] Based on the required transmission time for different command contents, and according to the command interaction process of the dual-frequency UAV communication data transmission system, time slot allocation design can be performed. 433MHz is used as the carrier frequency for control command transmission, employing FSK modulation. With a transmission distance of 1km, the maximum data transmission rate can reach 100kbps. The command transmission time can be calculated based on the command length.

[0188]

[0189] Where v represents the data transmission rate and d represents the instruction frame length.

[0190] The following durations are calculated based on the normal response of all drones, namely: For the power-on self-test broadcast command, the command frame length is 10.24ms and the required time slot length is 104.24ms; for the system initialization broadcast command, the frame length is 104.24ms; for the recovery preparation broadcast, the required time slot length is 11.96ms; for the power-off recovery broadcast, the required time slot length is 4.48ms; for the position adjustment command, the frame length is 3.2ms; for the emergency avoidance command, the frame length is 2.24ms; and for the image acquisition command, the frame length is 0. 8ms; 0.32ms is required for handshake instruction frame length; 0.32ms is required for retransmission instruction frame length; 0.4ms is required for status transition instruction frame length; 0.32ms is required for execution result instruction frame length; 0.32ms is required for priority adjustment request instruction frame length; 0.32ms is required for self-test feedback instruction frame length; 259.3ms is required for data return instruction frame length.

[0191] The data frames of this invention are transmitted via three frequency points of the 5G communication band, using an adaptive modulation method. The maximum data transmission rate can reach 240Mbps. Assuming the image acquisition frame rate is f=30fps and the image resolution is... This allows us to calculate the actual total data transmission per second. Required data transmission time .

[0192] IV. The working process of the communication system;

[0193] 1. Power-on and networking phase; After the system powers on, it first performs initialization settings. The ground control terminal sends a power-on self-test broadcast command for 128ms. X drones perform self-test operations and report the status of X drones. The feedback interval is 320ms. Therefore, the system boot-up and networking phase takes 448ms. Then all drones responded normally and entered the operational phase. If show If the drone is malfunctioning, the ground control unit continues to send 128ms self-test commands and waits for feedback from the drone's self-test. If all feedback is normal, the next time slot allocation will proceed; otherwise... Then record bad pixels. Only the drones (X units) in their normal state at this time will be retained for subsequent operations. The specific time slot allocation and power-on initialization process are as follows: Figure 18 and Figure 19 As shown.

[0194] 2. Mission Execution Phase; After the UAVs complete initialization, multiple normally functioning UAVs (a total of X) will enter the mission execution phase. This phase mainly involves four tasks: position adjustment, emergency avoidance, data acquisition and transmission, and data retransmission. Figure 20 (The dashed box indicates an optional step).

[0195] Position adjustment task: If there are problems with the shooting angle or altitude in the returned data, the ground will send a position adjustment command, and the drone will adjust its position or aerial attitude according to the command.

[0196] Emergency obstacle avoidance mission: When the ground control terminal detects that a drone may be in conflict with other flying objects, it will send an emergency command to trigger the drone to perform obstacle avoidance actions.

[0197] Data acquisition and transmission tasks: After receiving the data acquisition command from the master station within the allocated time slot, the UAV acquires data, compresses and frames the image data, and transmits the data transmission command to the ground control master station on the designated frequency. This step is the core of the service transmission phase.

[0198] Data retransmission task: The ground terminal parses the received data. If data errors or transmission abnormalities are found, a retransmission command is sent to the corresponding UAV, requesting it to retransmit the data.

[0199] During the mission execution phase, a handshake signal test between the ground control station and the UAV is first conducted within the allocated time slot. The ground control station sends the handshake signal according to the format specified in the protocol, including the UAV's identification code, command type pre-indication, and other information. Upon receiving the handshake signal, the UAV immediately sends a response signal containing its current status information to the ground control station through the feedback channel of its own 433MHz command receiver. The ground control station determines whether the UAV is in a command-receiving state based on the response signal, such as whether the battery is sufficient and the communication link is stable.

[0200] After confirming that the UAV can receive commands, the ground control station sends control commands within the designated time slot according to the time slot allocation scheme. The command frame is encapsulated in a preset format and transmitted via a 433MHz command transmitter. The UAV receives the command frame within its own time slot and verifies it, checking the frame header, checksum, and other information for correctness. If the verification passes, the flight control system parses the command type and parameters and executes the corresponding operation; if the verification fails, the ground control station sends a retransmission command. Upon receiving the retransmission request, the UAV retransmits the command frame.

[0201] The entire task execution process operates in TDMA mode across multiple frequency bands, with a guard interval set for each band (the guard interval is 10~15ms, calculated using the following formula:). This ensures reliable command transmission and system resilience against faults.

[0202] 3. Shutdown and Recovery Phase; When the UAV array enters the shutdown and recovery phase, the ground control first sends a recovery preparation broadcast command. This broadcast is sent 10 times and is used to initially set up the recovery formation for the X currently operating UAVs. Subsequently, the ground control... After waiting, a shutdown and recovery broadcast is issued. This broadcast carries the adjusted priority recovery group order and shutdown and recovery start time. The ground station, based on the remaining battery power and available time transmitted back by each drone in real time, calculates the urgency of recovery and sets the groups of drones with longer return times or lower battery power as high priority, prioritizing their recovery order to ensure safety and recovery efficiency. Finally, the system executes the recovery operation according to the adjusted group order. Each group performs shutdown and recovery at 60-second intervals, proceeding sequentially according to the group numbering order, until all drones are safely recovered. The entire process, through status awareness and dynamic scheduling, achieves efficient and reliable recovery in multi-drone collaborative scenarios. See the recovery time slot allocation section. Figure 21 .

[0203] Regarding the implementation of the data transmission protocol, parameters are configured for the UAV image acquisition equipment, such as setting the frame rate to 30fps and the resolution to 1920×1080, to meet the format requirements for image data transmission. During image data transmission, the data link layer adjusts the modulation and demodulation methods and coding rates in real time based on channel quality. For example, when a high bit error rate is detected on the 5.15GHz channel, the modulation method for this band is automatically switched from 16QAM (Quadrature Amplitude Modulation) to QPSK (Quadrature Phase Shift Keying), and the coding rate is reduced to improve the reliability of data transmission. Simultaneously, a hybrid automatic repeat request mechanism performs preliminary error correction on the received image data at the receiving end. If uncorrectable errors still exist, a retransmission request is sent to the sending end. Upon receiving the request, the sending end combines the previously transmitted data, recodes and modulates the data using a soft combination method, and retransmits it through the corresponding frequency band.

[0204] Throughout the implementation of the communication protocol, the protocol parameters are continuously optimized and adjusted. For example, the time slot length and interval of command transmission are dynamically adjusted according to the number of UAVs and the flight environment in the actual mission; the priority and retransmission strategy in the data transmission protocol are adjusted according to the real-time nature and importance of the image data. Through these implementation steps and optimization measures, the UAV array command communication and data transmission protocol system of this invention can operate efficiently and stably in practical applications, meeting the communication needs of UAV arrays in various complex tasks.

[0205] The above description is merely a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications, equivalent substitutions, and improvements made by those skilled in the art within the scope of the technology disclosed in the present invention, and within the spirit and principles of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A TDMA-based dual-frequency master-slave UAV array communication networking method, characterized in that, This method is executed collaboratively by a ground control station acting as the master station and multiple UAVs acting as slave stations, and specifically includes the following steps: S1. Power-on and networking phase: The ground control station broadcasts a power-on self-test command containing time slot allocation information in a loop through the first frequency band; each UAV performs a self-test and feeds back its status through the first frequency band; the ground control station determines the set of UAVs that are working normally based on the feedback and completes system initialization and initial time slot allocation; S2. Mission Execution Phase: The ground control station sends uplink control commands to the target UAV through the first frequency band in the designated unicast time slot according to the time slot allocation; the target UAV replies with downlink status commands through the first frequency band in the corresponding feedback time slot; when the uplink control command includes an image acquisition command, the target UAV transmits image data back to the ground control station through the second frequency band with a frequency higher than the first frequency band. S3. Shutdown and Recovery Phase: The ground control station broadcasts recovery preparation instructions and shutdown and recovery instructions sequentially through the first frequency band. It dynamically adjusts the priority of the recovery group according to the real-time status of the UAVs and controls the UAV cluster to perform shutdown and recovery operations in sequence according to the priority order.

2. The TDMA-based dual-frequency master-slave UAV array communication networking method according to claim 1, characterized in that, The first frequency band is the 433MHz low-frequency band, and the second frequency band is the 5GHz high-frequency band; Instructions transmitted via the first frequency band employ forward error correction coding, while image data transmitted via the second frequency band uses a hybrid automatic repeat request (HARQ) mechanism for error control. When the receiving end detects an error in the received image data, it uses FEC for error correction. If the data cannot be recovered after error correction, a retransmission request is sent to the sender.

3. The TDMA-based dual-frequency master-slave UAV array communication networking method according to claim 1, characterized in that, The time slot allocation is based on the TDMA mechanism and includes broadcast time slots, uplink unicast command time slots, downlink feedback time slots, and image data return time slots. The length of each time slot is dynamically calculated and determined according to the length of the corresponding command frame or data frame and the nominal transmission rate of the channel.

4. The TDMA-based dual-frequency master-slave UAV array communication networking method according to claim 1, characterized in that, During command and data transmission, a time-slot-based transmission scheduling mechanism is adopted. The ground control station allocates a fixed time slot to each UAV, and the UAV array receives commands and feeds back the execution results of the control commands from the ground control station within the corresponding time slot. The ground control station sends control commands to the UAV, and the UAV initiates the data acquisition process according to the control commands, compressing and encoding the acquired image data. The encoded data stream is divided into three data segments, which are modulated sequentially onto three carrier frequencies of 5.15 GHz, 5.35 GHz, and 5.75 GHz and transmitted in parallel.

5. The TDMA-based dual-frequency master-slave UAV array communication networking method according to claim 1, characterized in that, The instruction frame structure includes five fields: frame header, target ID, instruction type, instruction content, and checksum. The frame header is used to identify the start and synchronization of the instruction frame, the target ID field is used to specify the target of the response instruction, the instruction type is used to clearly indicate the instruction and operation type, and the checksum field is used to verify the integrity and accuracy of the instruction frame to ensure that the instruction has not been tampered with or damaged during transmission. The command types include broadcast commands and unicast commands. Broadcast commands include power-on self-test broadcast commands, system initialization broadcast commands, recovery preparation broadcast commands, and power-off recovery broadcast commands, which are used for time slot allocation, position preset, and UAV status self-test at the beginning of system operation. Unicast commands include uplink control commands and downlink status commands, used to control individual UAVs in the UAV array during system operation. Uplink control commands are commands from the ground control station to the UAV, including position adjustment commands, emergency avoidance commands, image acquisition commands, handshake commands, and retransmission commands. Downlink status commands are commands from the UAV to the ground control station, including status feedback commands, execution result commands, priority adjustment request commands, self-test feedback commands, and data feedback commands.

6. The TDMA-based dual-frequency master-slave UAV array communication networking method according to claim 5, characterized in that, The frame structure of the power-on self-test broadcast command contains a time slot allocation priority list, which is used by the ground control station to allocate downlink transmission time slots to different UAVs according to priority level; The frame structure of the system initialization broadcast command contains a list of preset position information, which is used to assign three-dimensional target coordinate information to each UAV; Both the recycling preparation broadcast command and the shutdown recycling broadcast command contain queue or queue priority information in their frame structures for dynamically adjusting the recycling order.

7. The TDMA-based dual-frequency master-slave UAV array communication networking method according to claim 5, characterized in that, The position adjustment command is used to instruct the UAV to adjust the three-dimensional coordinates of the target position. The adjustment path node includes the position information of two transit nodes, which are used to instruct the UAV to adjust the coordinate path. Emergency avoidance commands are configured as high-priority commands to indicate the three-dimensional coordinates of the target location for the UAV to avoid, interrupting or preempting ongoing normal command transmission time slots; Image acquisition commands are used to instruct the image acquisition parameters of the UAV. Image acquisition parameters include acquisition parameters used to control the image acquisition equipment and image data processing operations. The priority adjustment request instruction includes power status information that indicates the urgency of drone recovery, which is used to trigger the ground control station to dynamically adjust the recovery priority during the shutdown recovery phase.

8. The TDMA-based dual-frequency master-slave UAV array communication networking method according to claim 1, characterized in that, Before transmitting image data back via the second frequency band, a handshake mechanism is performed, including: The ground control station sends a handshake command to the target UAV. The target UAV responds to the handshake command and reports its own status information. After confirming that it can receive the command, it sends an image acquisition command. The image data transmitted via the second frequency band is as follows: the target UAV divides the image data stream into multiple data segments and modulates them onto different subcarrier frequencies within the second frequency band for parallel transmission.

9. A ground control station for implementing the TDMA-based dual-frequency master-slave UAV array communication networking method as described in any one of claims 1-8, characterized in that, The ground control station includes: The control module is used to generate various commands, execute TDMA-based time slot scheduling, process UAV feedback information, and dynamically adjust recovery priorities. The first communication module is configured to operate in the 433MHz frequency band and is used to send all broadcast commands and uplink control commands, and to receive all downlink status commands. The second communication module is configured to operate in the 5GHz band and is used to receive image data transmitted back by the drone.

10. A drone for implementing the TDMA-based dual-frequency master-slave drone array communication networking method as described in any one of claims 1-8, characterized in that, The drone includes: The first airborne communication module is configured to operate in the 433MHz frequency band to receive broadcast commands and uplink control commands from the ground control station, and to send downlink status commands within the allocated feedback time slots. The image acquisition and processing module is used to respond to image acquisition commands, acquire and compress image data; The second airborne communication module is configured to operate in the 5GHz band and is used to transmit compressed image data back to the ground control station.

Images