Unmanned aerial vehicle high-definition image transmission scheduling system for 5g millimeter wave communication

Abstract

The application discloses a 5G millimeter wave communication unmanned aerial vehicle high-definition image transmission scheduling system and belongs to the technical field of unmanned aerial vehicle communication. The system comprises an unmanned aerial vehicle end, a ground station end and a 5G millimeter wave communication network; the unmanned aerial vehicle end comprises an image acquisition module, a data processing module, a 5G millimeter wave communication module, a flight control module and an edge computing module; and the ground station end comprises a 5G millimeter wave communication module, a data receiving and calculating module, an image transmission scheduling module, a display module and an edge server. Through the fusion of 5G-A millimeter wave communication and radar sensing, the closed loop of "communication-sensing-control" is realized for the first time, and a breakthrough is made in the field of environment sensing and communication cooperation compared with the prior art. Through the edge node, local real-time processing is realized, the cloud end is used for model training and global optimization, and the delay is reduced by 60% compared with the traditional centralized architecture. In combination with the blockchain authentication and frequency hopping technology, an end-to-end secure transmission channel is constructed, and the anti-interference capability is improved by 50% compared with the use of AES-256 alone.

Description

Technical Field

[0001] This invention relates to the field of drone communication technology, and more specifically, to a high-definition image transmission and dispatching system for drones using 5G millimeter-wave communication. Background Technology

[0002] With the widespread application of drones in surveying, security, logistics, and many other fields, the demand for real-time transmission of high-definition image data from drones to ground stations is becoming increasingly urgent. Currently, traditional drone image transmission often uses the 2.4GHz or 5.8GHz frequency bands. These bands have limited spectrum resources and are highly susceptible to interference, leading to problems such as stuttering and frame drops in image transmission, making it difficult to meet the requirements for real-time transmission of high-definition images. Furthermore, as scenarios involving multiple drones working collaboratively increase, how to efficiently allocate drone image transmission resources has become a pressing problem to be solved.

[0003] Despite the new opportunities brought by 5G technology, research and application of 5G millimeter-wave communication in high-definition image transmission and scheduling for drones are still in their early stages. Existing related technologies, such as the 3D UPF beam tracking technology developed by the Opto-Intelligence team, involve the application of millimeter-wave communication in drones, but lack deep integration with edge computing and intelligent resource allocation algorithms. This leads to problems such as resource allocation imbalance and excessive transmission latency when multiple drones are working collaboratively. Beijing Telecom's 3D collaborative perception algorithm has achieved centimeter-level positioning and path optimization, but it has not been effectively combined with communication resource allocation, resulting in a mismatch between communication link quality and positioning accuracy, and an image transmission packet loss rate greater than 5%. Task scheduling algorithms based on multi-objective optimization, such as KnCMPSO, lack intelligent decision-making capabilities when facing the dynamically changing communication environment of drones. Therefore, mature and innovative systems and solutions are still relatively lacking.

[0004] In view of this, the present invention is proposed to solve the above-mentioned technical problems. Summary of the Invention

[0005] The purpose of this invention is to provide a 5G millimeter-wave communication-based high-definition image transmission and dispatch system for unmanned aerial vehicles (UAVs) to solve the proposed technical problems.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] A 5G millimeter-wave communication-based high-definition image transmission and dispatch system for drones, including the drone terminal, the ground station terminal, and the 5G millimeter-wave communication network;

[0008] The drone terminal includes:

[0009] The image acquisition module is used to acquire high-definition image data during the flight of the drone;

[0010] The data processing module preprocesses the high-definition image data acquired by the image acquisition module;

[0011] The 5G millimeter-wave communication module transmits the image data, which has been preprocessed by the data processing module, to the ground station via the 5G millimeter-wave communication network, while simultaneously collecting communication link quality parameters in real time.

[0012] The flight control module controls the UAV's flight attitude and flight path according to the instructions sent from the ground station;

[0013] The edge computing module integrates the TensorRT framework to perform local computations such as image preprocessing and target recognition, and synchronizes model parameters with the ground station through edge nodes;

[0014] Ground station terminals include:

[0015] The 5G millimeter-wave communication module receives image data sent from the drone.

[0016] The data receiving and decoding module decompresses and decodes the data received by the 5G millimeter-wave communication module to restore the original high-definition image data.

[0017] The image transmission scheduling module dynamically allocates image transmission resources and optimizes image transmission paths based on factors such as the mission requirements, flight status, and communication link quality of multiple drones.

[0018] The display module displays the high-definition image data recovered by the data receiving and processing module;

[0019] Edge servers support real-time data processing and AI model training, and use TensorFlow Serving to update communication quality prediction models in real time.

[0020] 5G millimeter-wave communication networks provide high-speed and stable communication links to ensure fast and reliable transmission of image data between UAVs and ground stations. This includes 5G base stations and related signal processing equipment, which can efficiently transmit, receive and process millimeter-wave signals. Using the 5G-A NR protocol, the base station integrates millimeter-wave radar to achieve real-time perception of the UAV's position and speed, and fuses this with communication link quality data.

[0021] The image transmission scheduling module includes:

[0022] Real-time status monitoring: The drone terminal collects its own flight parameters and communication link quality parameters in real time and sends them to the ground station terminal through the 5G millimeter wave communication network. The ground station terminal simultaneously monitors the mission execution status of each drone.

[0023] Resource allocation strategy: The image transmission scheduling module adopts a priority-based and communication quality-based resource allocation algorithm based on the received UAV status information and task information. This algorithm introduces a reinforcement learning (DRL) model, takes task priority, communication quality prediction value, and UAV remaining power parameters as input, and outputs the optimal bandwidth allocation scheme.

[0024] Communication quality prediction: Using an LSTM neural network, input historical signal strength, bit error rate, and environmental parameters, predict the communication quality score for the next 500ms.

[0025] 5G base stations in 5G millimeter-wave communication networks use phased array antennas and beamforming technology. Based on the real-time location information of the UAV, the antenna array phase is adjusted to form a directional beam with a beam pointing angle tracking accuracy of ≤±2°, which improves the millimeter-wave signal gain by 8-15dB. At the same time, radar echo data is used to construct a three-dimensional environment map.

[0026] The edge computing module on the drone uses the NVIDIA Jetson AGX Orin platform to perform target detection on the image before transmission, and only uploads the region of interest (ROI) data, reducing the amount of transmission by 30%-50%.

[0027] Furthermore, the image acquisition module on the drone uses a high-resolution camera with a resolution of ≥4K to acquire high-definition image data at a frame rate of 15-30 frames per second, depending on the drone's flight mission and scenario requirements.

[0028] Furthermore, the data processing module on the drone uses the H.265 algorithm to compress and encode the acquired image data, with a compression ratio controlled between 10:1 and 20:1, and uses the AES-256 algorithm to encrypt the compressed data.

[0029] Furthermore, the data receiving and processing module at the ground station decrypts the received data using the AES-256 algorithm and decompresses it using the H.265 decoding algorithm. During the processing, it uses the CRC32 check algorithm and LDPC error correction code for data verification and error correction.

[0030] Furthermore, TensorFlow Serving is deployed on the edge servers at the ground stations to update the communication quality prediction model in real time, with a model update cycle of ≤1 minute.

[0031] Furthermore, the drone and ground station use blockchain technology to achieve two-way identity authentication, and the drone integrates a Beidou short message module, which automatically switches to satellite communication when the 5G network is interrupted.

[0032] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0033] (1) Multimodal collaborative perception: It integrates 5G-A millimeter wave communication and radar perception, and realizes the "communication-perception-control" closed loop for the first time, which is a significant breakthrough in environmental perception and communication collaboration compared with existing technologies.

[0034] (2) Edge-cloud collaborative intelligence: Real-time local processing is achieved through edge nodes, and model training and global optimization are performed in the cloud, reducing latency by 60% compared to the traditional centralized architecture.

[0035] (3) Anti-interference security system: Combining blockchain authentication and frequency hopping technology, an end-to-end secure transmission channel is constructed, and the anti-interference capability is improved by 50% compared with AES-256 alone. Attached Figure Description

[0036] The accompanying drawings, which form part of this application, are used to provide a further understanding of the present invention. The illustrative embodiments and descriptions of the present invention are used to explain the present invention, but do not constitute an undue limitation of the present invention. Obviously, the drawings described below are merely some embodiments; those skilled in the art can obtain other drawings based on these drawings without creative effort. In the drawings:

[0037] Figure 1 This is a flowchart of the image transmission scheduling system provided in this embodiment of the application. Detailed Implementation

[0038] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0039] See Figure 1 As shown, this application provides a 5G millimeter-wave communication-based high-definition image transmission and dispatching system for drones, including a drone terminal, a ground station terminal, and a 5G millimeter-wave communication network;

[0040] The drone terminal includes:

[0041] The image acquisition module is used to acquire high-definition image data during the flight of the drone;

[0042] The data processing module preprocesses the high-definition image data acquired by the image acquisition module, including compression and encoding.

[0043] The 5G millimeter-wave communication module transmits the image data, which has been preprocessed by the data processing module, to the ground station via the 5G millimeter-wave communication network. At the same time, it collects communication link quality parameters in real time, including signal strength, bit error rate, and signal-to-noise ratio.

[0044] The flight control module controls the UAV's flight attitude and flight path according to the instructions sent from the ground station;

[0045] The edge computing module integrates the TensorRT framework to perform local computations such as image preprocessing and target recognition, and synchronizes model parameters with the ground station through edge nodes;

[0046] The ground station includes:

[0047] The 5G millimeter-wave communication module receives image data sent from the drone.

[0048] The data receiving and decoding module decompresses and decodes the data received by the 5G millimeter-wave communication module to restore the original high-definition image data.

[0049] The image transmission scheduling module dynamically allocates image transmission resources, including transmission bandwidth, communication frequency, and transmission power, and optimizes image transmission paths based on factors such as the mission requirements, flight status, and communication link quality of multiple drones.

[0050] The display module displays the high-definition image data recovered by the data receiving and processing module;

[0051] Edge servers support real-time data processing and AI model training, and use TensorFlow Serving to update communication quality prediction models in real time.

[0052] 5G millimeter-wave communication networks provide high-speed and stable communication links to ensure fast and reliable transmission of image data between UAVs and ground stations. This includes 5G base stations and related signal processing equipment, which can efficiently transmit, receive, and process millimeter-wave signals. Using the 5G-A NR protocol, the base station integrates millimeter-wave radar to achieve real-time perception of the UAV's position (accuracy ≤ 0.5 meters) and speed (accuracy ≤ 0.1 m / s), and fuses this with communication link quality data.

[0053] The image transmission scheduling module includes:

[0054] Real-time status monitoring: The UAV collects its own flight parameters (position, speed, altitude) and communication link quality parameters in real time (signal strength ≥ -85dBm is preferred, bit error rate ≤ 10⁻). 6 (For priority), and transmit to the ground station via 5G millimeter wave communication network. The ground station simultaneously monitors the mission execution status of each drone (mission priority is divided into four levels: emergency / high / medium / low).

[0055] Resource allocation strategy: The image transmission scheduling module uses a priority-based and communication quality-based resource allocation algorithm based on the received UAV status and task information. This algorithm incorporates a reinforcement learning (DRL) model, taking task priority, predicted communication quality, and remaining UAV battery parameters as inputs, and outputs the optimal bandwidth allocation scheme. The convergence speed is 40% faster than traditional algorithms, including:

[0056] Priority quantification: Urgent tasks have a weight of 0.8, high priority 0.6, medium priority 0.4, and low priority 0.2;

[0057] Communication quality score: A comprehensive score is calculated based on signal strength, bit error rate, and signal-to-noise ratio (out of 100).

[0058] Resource allocation coefficient = priority weight × 0.6 + communication quality score × 0.4, and bandwidth resources are allocated according to the coefficient ratio;

[0059] For high-priority tasks, more image transmission resources are allocated first, and for drones with good communication link quality, their transmission bandwidth is appropriately increased. At the same time, a distributed alliance cooperation algorithm is adopted to dynamically divide the task area and allocate spectrum resources according to parameters such as drone type, task priority, and remaining battery power. Adjacent drones use frequency hopping technology (frequency hopping rate ≥ 1000 times / second) combined with power control (formula: P=P0・e^(-d / λ), where λ is the attenuation coefficient) to avoid mutual interference of millimeter wave signals between adjacent drones.

[0060] Dynamic adjustment: When the communication quality score of a drone is below 60 points for three consecutive sampling periods (sampling period is 1 second), the image transmission scheduling module dynamically adjusts the allocation of image transmission resources in real time, reducing its transmission bandwidth to 50%-70% of the original allocation, and allocating the released resources to drones with a communication quality score ≥80 points. The flight control module then sends a path adjustment command to the drone, causing it to fly to an area with a signal strength ≥-80dBm.

[0061] Communication quality prediction: Using an LSTM neural network, input historical signal strength, bit error rate, and environmental parameters, predict the communication quality score for the next 500ms (accuracy ≥ 90%).

[0062] The image acquisition module on the drone uses a high-resolution camera with a resolution of ≥4K to acquire high-definition image data at a frame rate of 15-30 frames per second, depending on the drone's flight mission and scenario requirements.

[0063] The data processing module on the drone uses the H.265 algorithm to compress and encode the acquired image data, with a compression ratio controlled between 10:1 and 20:1, and uses the AES-256 algorithm to encrypt the compressed data.

[0064] 5G base stations in 5G millimeter-wave communication networks use phased array antennas (32T32R) and beamforming technology. Based on the real-time location information of the UAV (positioning accuracy ≤1 meter), the antenna array phase is adjusted to form a directional beam with a beam pointing angle tracking accuracy ≤±2°, which improves the millimeter-wave signal gain by 8-15dB. At the same time, radar echo data is used to construct a three-dimensional environment map.

[0065] The data receiving and processing module at the ground station decrypts the received data using the AES-256 algorithm and decompresses it using the H.265 decoding algorithm. During the processing, it uses the CRC32 check algorithm and LDPC error correction code (coding rate ≥ 0.8) for data verification and error correction.

[0066] The edge computing module on the drone uses the NVIDIA Jetson AGX Orin platform to perform target detection on the image before transmission, and only uploads the region of interest (ROI) data, reducing the amount of transmission by 30%-50%.

[0067] TensorFlow Serving is deployed on the edge servers at the ground station to update the communication quality prediction model in real time, with a model update cycle of ≤1 minute.

[0068] The drone and ground station use blockchain technology (PBFT consensus algorithm) to achieve two-way identity authentication with an authentication latency of ≤50ms and a 90% improvement in resistance to man-in-the-middle attacks. The drone also integrates a Beidou short message module, which automatically switches to satellite communication when the 5G network is interrupted.

[0069] Drone-side workflow

[0070] Image Acquisition: The image acquisition module uses a high-resolution camera with a resolution of ≥4K to acquire high-definition image data at a frame rate of 15-30 frames per second, depending on the UAV flight mission and scenario requirements. For example, in surveying and mapping missions, the camera can be set to acquire 4K image data at a frame rate of 30 frames per second.

[0071] Data preprocessing: The data processing module uses the H.265 algorithm to compress and encode the acquired image data, with a compression ratio controlled between 10:1 and 20:1. While ensuring image quality, the data size is compressed to a suitable range for wireless transmission. Simultaneously, the compressed data is encrypted using the AES-256 algorithm to ensure data transmission security.

[0072] Edge computing processing: The edge computing module uses the NVIDIA Jetson AGX Orin platform to perform object detection on the image before transmission and only uploads the region of interest (ROI) data.

[0073] Data transmission: The 5G millimeter-wave communication module modulates and encodes the pre-processed image data according to the 5G-A NR protocol, and then transmits it through the 5G millimeter-wave band. During transmission, the transmit power and transmission rate are dynamically adjusted according to the communication link quality to ensure reliable data transmission. At the same time, communication link quality parameters are collected in real time and sent to the ground station.

[0074] Flight Control: The flight control module receives flight commands from the ground station, adjusts the UAV's flight attitude, speed, altitude, and other parameters accordingly, and feeds back the UAV's real-time flight status information to the ground station. It integrates a BeiDou short message module, automatically switching to satellite communication when the 5G network is interrupted.

[0075] Ground station workflow

[0076] Data reception: The 5G millimeter-wave communication module receives image data from the drone, performs demodulation, decoding and other processing, and restores the compressed and encrypted image data.

[0077] Data processing: The data receiving and processing module uses the AES-256 algorithm to decrypt the recovered image data and the H.265 decoding algorithm to decompress it, restoring the original high-definition image data. During processing, the CRC32 checksum algorithm and LDPC error correction code (coding rate ≥ 0.8) are used for data verification and error correction to ensure the integrity and accuracy of the image data.

[0078] Edge server processing: The edge server deploys TensorFlow Serving, updates the communication quality prediction model in real time, receives model parameters from the edge computing module on the drone and synchronizes them.

[0079] Image transmission scheduling: The image transmission scheduling module acquires real-time flight status, communication link quality, and mission information for each UAV. Based on a priority- and communication quality-based resource allocation algorithm (including a reinforcement learning model), it calculates the image transmission resources to be allocated to each UAV, such as transmission bandwidth and transmit power. For UAVs performing emergency rescue missions, high transmission bandwidth is prioritized to ensure timely image transmission. Simultaneously, frequency planning and power control coordinate the millimeter-wave signal transmission of multiple UAVs to avoid interference. The communication quality prediction model is trained using historical communication data (containing over 200,000 samples), employing the Adam optimizer with a learning rate of 0.001, a training cycle of 50 epochs, and a test set MAE ≤ 5 points (out of 100). The reinforcement learning model runs a deep Q-network (DQN) at the ground station. The state space includes 12-dimensional parameters such as UAV position, mission priority, and predicted communication quality, while the action space includes bandwidth allocation levels (1-10 levels). An experience replay mechanism is used to improve training efficiency.

[0080] Identity authentication: Two-way identity authentication between the drone and the device is achieved using blockchain technology (PBFT consensus algorithm).

[0081] Image display: The display module displays the processed high-definition image data, and the operator can view the image information in real time through the monitor, so as to make decisions and adjustments for the UAV flight mission.

[0082] Workflow of 5G millimeter wave communication networks

[0083] Signal Transmission and Reception: The 5G base station transmits downlink signals to the drone via a millimeter-wave antenna (32T32R) and simultaneously receives uplink signals from the drone. The base station employs beamforming technology, adjusting the antenna array phase to form a directional beam based on the drone's real-time location information (positioning accuracy ≤1 meter). The beam pointing angle tracking accuracy is ≤±2°, increasing the millimeter-wave signal gain by 8-15dB, improving signal directivity and gain, and enhancing signal coverage and transmission quality. The base station integrates millimeter-wave radar to achieve real-time perception of the drone's position and speed.

[0084] Signal processing: The base station demodulates, decodes, and corrects errors in the received signals, extracting image data and UAV status information. Simultaneously, it encodes and modulates the commands and control information to be sent to the UAV before transmitting them via a millimeter-wave antenna. A 3D environment map is constructed using radar echo data.

[0085] Network Management: The 5G network management system monitors and manages the entire 5G millimeter-wave communication network, including base station operating status, signal strength, and interference. Based on network conditions and drone communication needs, it dynamically adjusts network parameters such as bandwidth allocation and power control to ensure stable network operation and efficient communication.

[0086] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely preferred examples and are not intended to limit the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

A 1.5G millimeter-wave communication-based UAV high-definition image transmission and dispatch system is characterized by: This includes drone terminals, ground station terminals, and 5G millimeter-wave communication networks; The drone terminal includes: The image acquisition module is used to acquire high-definition image data during the flight of the drone; The data processing module preprocesses the high-definition image data acquired by the image acquisition module; The 5G millimeter-wave communication module transmits the image data preprocessed by the data processing module to the ground station via the 5G millimeter-wave communication network, while simultaneously collecting communication link quality parameters in real time. The flight control module controls the UAV's flight attitude and flight path according to the instructions sent from the ground station; The edge computing module integrates the TensorRT framework to realize local computation of image preprocessing and target recognition, and synchronizes model parameters with the ground station through edge nodes; The ground station includes: The 5G millimeter-wave communication module receives image data sent from the drone. The data receiving and decoding module decompresses and decodes the data received by the 5G millimeter-wave communication module to restore the original high-definition image data. The image transmission scheduling module dynamically allocates image transmission resources and optimizes image transmission paths based on the mission requirements, flight status, and communication link quality factors of multiple drones. The display module displays the high-definition image data recovered by the data receiving and processing module; Edge servers support real-time data processing and AI model training, and use TensorFlow Serving to update communication quality prediction models in real time. The 5G millimeter-wave communication network provides a high-speed and stable communication link to ensure fast and reliable transmission of image data between the UAV and the ground station. It includes 5G base stations and related signal processing equipment, which can efficiently transmit, receive and process millimeter-wave signals. It adopts the 5G-A NR protocol, and the base station integrates millimeter-wave radar to realize real-time perception of the UAV's position and speed, and fuses it with the communication link quality data. The image transmission scheduling module includes: Real-time status monitoring: The drone terminal collects its own flight parameters and communication link quality parameters in real time and sends them to the ground station terminal through the 5G millimeter wave communication network. The ground station terminal simultaneously monitors the mission execution status of each drone. Resource allocation strategy: The image transmission scheduling module adopts a priority-based and communication quality-based resource allocation algorithm based on the received UAV status information and task information. This algorithm introduces a reinforcement learning (DRL) model, takes task priority, communication quality prediction value, and UAV remaining power parameters as input, and outputs the optimal bandwidth allocation scheme. Communication quality prediction: Using an LSTM neural network, input historical signal strength, bit error rate, and environmental parameters, predict the communication quality score for the next 500ms. The 5G base station in the 5G millimeter-wave communication network adopts phased array antennas and beamforming technology. Based on the real-time location information of the UAV, the antenna array phase is adjusted to form a directional beam with a beam pointing angle tracking accuracy of ≤±2°, which improves the millimeter-wave signal gain by 8-15dB. At the same time, a three-dimensional environment map is constructed using radar echo data. The edge computing module on the drone uses the NVIDIA Jetson AGX Orin platform to perform target detection on the image before transmission, and only uploads the region of interest (ROI) data, reducing the transmission volume by 30%-50%.

2. The 5G millimeter-wave communication UAV high-definition image transmission and dispatch system according to claim 1, characterized in that, The image acquisition module on the drone uses a high-resolution camera with a resolution of ≥4K to acquire high-definition image data at a frame rate of 15-30 frames per second, depending on the drone's flight mission and scenario requirements.

3. The 5G millimeter-wave communication UAV high-definition image transmission and dispatch system according to claim 2, characterized in that, The data processing module on the drone uses the H.265 algorithm to compress and encode the acquired image data, with a compression ratio controlled between 10:1 and 20:1, and uses the AES-256 algorithm to encrypt the compressed data.

4. The 5G millimeter-wave communication UAV high-definition image transmission and dispatch system according to claim 3, characterized in that, The data receiving and processing module at the ground station decrypts the received data using the AES-256 algorithm and decompresses it using the H.265 decoding algorithm. During the processing, it uses the CRC32 check algorithm and LDPC error correction code for data verification and error correction.

5. The 5G millimeter-wave communication UAV high-definition image transmission and dispatch system according to claim 4, characterized in that, The edge server at the ground station deploys TensorFlow Serving to update the communication quality prediction model in real time, with a model update cycle of ≤1 minute.

6. The 5G millimeter-wave communication UAV high-definition image transmission and dispatch system according to claim 5, characterized in that, The drone terminal and the ground station terminal use blockchain technology to achieve two-way identity authentication, and the drone terminal integrates a Beidou short message module, which automatically switches to satellite communication when the 5G network is interrupted.

Images