Unmanned aerial vehicle air computing communication integrated transmission and flight path planning method and system

Abstract

The invention discloses an unmanned aerial vehicle air computing communication integrated transmission and flight path planning method and system, and aims to realize that an unmanned aerial vehicle assists ground user communication and air computing at the same time, improve communication efficiency and effectively utilize spectrum resources. In consideration of the problem of interference between communication and air computing, an unmanned aerial vehicle air computing and communication integrated transmission framework is provided, and an unmanned aerial vehicle is used as an air aggregation center to execute an air computing data aggregation task of a sensor and is also used as an air base station to perform wireless communication with a ground user. The transmitting power of the user, the transmitting coefficient of the sensor, the normalization factor and the track of the unmanned aerial vehicle are comprehensively considered, an algorithm framework based on deep reinforcement learning is proposed to ensure the precision requirement of air calculation, interference is reduced, and the transmission quality of the user is improved to the maximum extent. In addition, a reward function is proposed to combine with unmanned aerial vehicle destination requirements and user scheduling fairness. Simulation shows that compared with a comparison scheme, the proposal has advantages in performance.

Description

Background Art

[0002] Sensor networks are enabling technologies for ubiquitous sensing, computing, communication, and everything-intelligent connection in future wireless networks, providing services such as information sensing, information transmission, and information processing, and have been widely applied in fields such as public services, smart farms, smart logistics, and the Internet of Vehicles. However, this development trend has also brought unprecedented challenges in large-scale device connection and massive data processing. On the one hand, a large number of devices deployed in various applications need to achieve high-speed and low-latency data transmission through wireless networks to ensure the timeliness of tasks and the stable operation of the system; on the other hand, massive and complex device data needs to be transmitted to servers (base stations, cloud platforms, or edge computing nodes) for computing, processing, and analysis to achieve device decision-making and management, such as model parameter aggregation in large-scale distributed learning, state estimation and control optimization in smart grids, etc. This large-scale computing and transmission load significantly increases the communication and computing delays of the network and is difficult to meet the requirements of many applications with extremely high real-time requirements.

[0003] To address this issue, Over-the-Air Computation (AirComp), as an innovative data processing paradigm, has received extensive attention. Different from the traditional "transmit first, then compute" mode, AirComp utilizes the signal superposition characteristics of wireless multiple access channels to directly complete data fusion computing during the simultaneous data transmission of multiple devices, such as calculating arithmetic means or weighted sums. Compared with traditional methods, AirComp significantly reduces data transmission overhead and computing latency, providing an efficient data aggregation mechanism for large-scale wireless sensor networks, and is particularly suitable for real-time computing applications such as federated learning, intelligent sensing, and cooperative control. However, the computing performance of AirComp highly depends on the quality of the wireless channel. In complex environments, such as terrain occlusion, interference signals, and poor wireless propagation conditions, channel fading will cause an increase in data fusion errors, affecting computing accuracy and thus reducing the reliability of the system. Introducing drones as flexible aerial communication and computing platforms in this regard can be deployed in areas with poor channel conditions and establish high-quality line-of-sight channels, thereby significantly improving channel quality, enhancing AirComp data fusion accuracy, and computing efficiency.

[0004] Due to the advantages of unmanned aerial vehicle (UAV)-assisted communication and aerial computing, the application scenarios that integrate UAV-assisted communication and UAV-assisted aerial computing have attracted much attention. While this integration of communication and computing improves communication efficiency and the utilization efficiency of spectrum resources, it also brings inevitable interference challenges between communication and aerial computing. However, a large number of literature reviews show that existing research on UAV-assisted aerial computing mainly focuses on optimizing the computing accuracy strategy in single or multiple aerial computing network scenarios, neglecting the analysis of communication and computing requirements and interference mitigation in complex scenarios where UAVs assist both communication and aerial computing simultaneously. Therefore, there is an urgent need for an efficient computing and transmission scheme specifically designed for the integrated transmission scenario of UAV-assisted aerial computing and communication to maximize the quality of service of users while ensuring the accuracy of aerial computing. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a method and system for integrated transmission and trajectory planning of UAV aerial computing and communication in view of the deficiencies in the above-mentioned prior art, so as to solve the technical problems of insufficient interference management and resource collaborative optimization in the scenario where UAVs assist both communication and aerial computing simultaneously, which affect computing accuracy and quality of service.

[0006] The present invention adopts the following technical solutions: A method for integrated transmission and trajectory planning of UAV aerial computing and communication includes the following steps: Construct an integrated model of UAV aerial computing and communication including UAVs, M users, and J sensor nodes; Based on the obtained integrated model of UAV aerial computing and communication, respectively construct a sensor-UAV aerial computing model and a user transmission model based on time division multiple access (TDMA); according to the sensor-UAV aerial computing model and the user transmission model based on TDMA, introduce the maximum error constraint of aerial computing, and construct an optimization model that maximizes the sum of user transmission rates; Based on the current user access strategy and UAV trajectory coordinates, design a transmission strategy based on an alternating iteration algorithm for the optimization model; Design a user access strategy and a UAV trajectory planning strategy based on a deep reinforcement learning framework. During the deep reinforcement learning training process, obtain the optimal transmission strategy according to the transmission strategy optimization algorithm in each time slot, and learn the user access strategy and the UAV trajectory planning strategy through the training policy network until convergence to generate a dynamically optimized smooth trajectory.

[0007] Preferably, in the integrated model of UAV aerial computing and communication, in the n th time slot, the channel model between user and the UAV is:

[0008]

[0009] Among them, is the carrier frequency, c is the speed of light, and are the additional losses of the line-of-sight and non-line-of-sight links respectively, is the user and the probability of the line-of-sight link between the user and the UAV; In time slot the signal received by the UAV is:

[0010] Among them, is the transmit precoding coefficient of the sensor ; is the preprocessed signal of the sensor ; is the user scheduling strategy of the user ; is the transmit power of the user ; is the transmit signal of the user ; is the additive white Gaussian noise.

[0011] Preferably, the sensor-UAV air computing model is:

[0012] Among them, represents taking the expected value of the randomness of the original signal and , the receiver noise ; is the user interference received by the UAV during air computing in time slot ; is the background noise power; In the user transmission model based on time division multiple access, in time slot the transmission rate of user communication is:

[0013] Among them, is the interference power of air computing; Considering that the UAV can only access one user at most in each time slot, the user scheduling strategy is defined as:

[0014]

[0015] Constrain the total number of time slots for user access to the drone:

[0016] Among them, Denotes floor function.

[0017] Preferably, in the time slot The calculation objective function of the sensor data is:

[0018] Among them, Is the number of sensors, Is the sensor set, Is the sensor Preprocessed signal transmitted; Expected estimated average value:

[0019] Among them, Is the normalization factor received by the drone, Is the signal received by the drone in the n Time slot.

[0020] Preferably, the optimization model for maximizing the sum of user transmission rates is:

[0021] Among them, constraint C1 is the constraint on the mean squared error (MSE) of air computing, Γ is the maximum error threshold of air computing, constraint C2 and C3 are the maximum power constraints of sensors and users respectively, And Are the maximum transmission powers of sensors and users respectively, constraint C8 and C9 limit the upper limits of the transmission powers of sensors and users respectively, Is the maximum flight speed of the drone, Is the n Time slot in which the sensor Transmission precoding coefficient, Is the n Time slot in which the user m Transmission power, Is the n Time slot in which the user m Access strategy, Is the n Time slot normalization factor received by the drone, Is the n Time slot position coordinates of the drone, Is the n Time slot transmission rate of the user, Is the nTime-slot user m Channel gain between the user and the UAV Standard deviation of Gaussian white noise Power of Gaussian white noise For the n Time-slot sensor j Channel gain between the sensor and the UAV Set of time slots Set of sensors Number of users Set of users Maximum flight speed of the UAV Duration of each time slot Take-off position coordinates of the UAV Destination coordinates of the UAV

[0022] Preferably, the optimization problem of the transmission strategy is as follows

[0023] Wherein For the n Time slot of the user m Access strategy For the n Received normalization factor of the UAV in the time slot For the n Time slot of the user m Channel gain between the user and the UAV For the n Time slot of the user m Transmission power Introduced slack variable, as the Upper bound of the term Introduced slack variable to approximate the objective function For the n Time slot of the sensor Transmission precoding coefficient For the n Time slot of the sensor j Channel gain between the sensor and the UAV Maximum allowable calculation error for aerial computing Number of sensors Maximum transmission power of the sensors Standard deviation of Gaussian white noise Power of Gaussian white noise

[0024] Preferably, the solution of the UAV normalization factor is .

[0026] Preferably, during the deep reinforcement learning training process, the user access and UAV trajectory planning problem is described as a Markov decision process, including: State space : The state value of the th time slot is defined as the set of the access state of the current user, the sum of the current user transmission rates, the current air computing error, and the horizontal coordinate of the current UAV position; Action space : The action value of the th time slot is defined as the change in the UAV position in the two-dimensional coordinates and the access policy of the current time slot; Reward function : The th time slot, for the state taking the action the obtained reward function is defined as the sum of the user access reward and the UAV trajectory reward .

[0027] Preferably, the state value of the th time slot , the action value of the th time slot , the user access reward and the UAV trajectory reward are respectively:

[0028]

[0029]

[0030] Among them, is the access policy of the user n in the m th time slot, is the transmission rate of the user in the n th time slot, is the computing error of air computing in the n th time slot, is the coordinate of the UAV position on the x-axis in the n th time slot, is the coordinate of the UAV position on the y-axis in the n th time slot, is the reward for the UAV to reach the destination in the last time slot, is the n th time slot, is the nThe position coordinates of the time-slot UAV are the destination coordinates of the UAV and are the duration of each time slot.

[0031] In a second aspect, an embodiment of the present invention provides a UAV integrated air computing and communication transmission and trajectory planning system, including: A construction module that constructs a UAV integrated air computing and communication model including UAVs, M users, and J sensor nodes. Based on the UAV integrated air computing and communication model, a sensor-UAV air computing model and a user transmission model based on time division multiple access are respectively constructed; A constraint module that, according to the sensor-UAV air computing model and the user transmission model based on time division multiple access, introduces an air computing maximum error constraint to construct an optimization model for maximizing the sum of user transmission rates; A strategy module that designs a transmission strategy based on an alternating iteration algorithm for the optimization model based on the current user access strategy and the UAV trajectory coordinates; An output module that designs a user access strategy and a UAV trajectory planning strategy based on a deep reinforcement learning framework. During the deep reinforcement learning training process, the optimal transmission strategy is obtained according to the transmission strategy optimization algorithm in each time slot, and the user access strategy and the UAV trajectory planning strategy are learned through training the strategy network until convergence, generating a dynamically optimized smooth trajectory.

[0032] In a third aspect, a computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the above UAV integrated air computing and communication transmission and trajectory planning method are implemented.

[0033] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium including a computer program, and when the computer program is executed by a processor, the steps of the above UAV integrated air computing and communication transmission and trajectory planning method are implemented.

[0034] In a fifth aspect, a chip includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the above UAV integrated air computing and communication transmission and trajectory planning method are implemented.

[0035] In a sixth aspect, an embodiment of the present invention provides an electronic device including a computer program, and when the computer program is executed by the electronic device, the steps of the above UAV integrated air computing and communication transmission and trajectory planning method are implemented.

[0036] Compared with the prior art, the present invention has at least the following beneficial effects: A method for integrated transmission and trajectory planning of unmanned aerial vehicle (UAV) aerial computing and communication, which enables the UAV to assist ground users in communication and aerial computing simultaneously, can improve communication efficiency and effectively utilize spectrum resources. In particular, considering the interference problem between communication and aerial computing, since the UAV has limited energy and computing power, it is particularly sensitive to interference. To address this problem, a framework for integrated transmission of UAV aerial computing and communication is proposed. The UAV serves as an aerial aggregation center to perform the task of aggregating aerial computing data of sensors, and also serves as an aerial base station to communicate wirelessly with ground users. Considering the transmit power of users, the transmit coefficient of sensors, the normalization factor, and the UAV trajectory comprehensively, an algorithm framework based on deep reinforcement learning is proposed to ensure the accuracy requirements of aerial computing, reduce interference, and maximize the transmission quality of users. In addition, the proposed reward function combines the requirements of the UAV destination and the fairness of user scheduling. Simulations show that the proposed scheme has advantages in performance compared with the comparison scheme.

[0037] Furthermore, by integrating UAVs, users, and sensor nodes, a multi-dimensional collaborative network architecture is established to solve the problem of communication and computing coupling in dynamic scenarios; this model uses the UAV as a mobile base station and an aerial computing platform to optimize the channel quality through flexible deployment. Especially in complex environments with terrain occlusion or severe interference, the signal superposition accuracy (such as arithmetic mean, weighted sum, etc. computing tasks) can be significantly improved. At the same time, the model supports the heterogeneous communication requirements of multiple devices. For example, M users transmit data through time division multiple access (TDMA), and J sensors directly complete data fusion through aerial computing, reducing the communication overhead and delay in the traditional "transmission first and then computing" mode. This architecture provides a basic framework for the physical layer and network layer for subsequent optimization problems, ensuring the global coordination of resource allocation.

[0038] Furthermore, the sensor-UAV model utilizes the wireless channel superposition characteristic to directly complete the calculation (such as aggregating the model parameters of federated learning) during the signal transmission process, greatly reducing the transmission delay and energy consumption, and is applicable to scenarios with high real-time requirements (such as smart grid state estimation). The user transmission model based on TDMA avoids multi-user interference through orthogonal time slot allocation, maximizing the utilization rate of channel resources; combined with a dynamic time slot allocation algorithm (such as AWPSO-SAA), it can predict user needs and optimize time slot allocation, reducing transmission jitter (the delay jitter is as low as 0.11). The separate design of the two ensures both the calculation accuracy and meets the high-throughput communication requirements.

[0039] Furthermore, an aerial computing maximum error constraint (such as mean square error threshold) is introduced to ensure that the data fusion accuracy is not affected by channel fading, and at the same time, with the goal of maximizing the sum of user rates, the power and time slot allocation are optimized.

[0040] Furthermore, alternating iteration methods (such as block coordinate descent and successive convex approximation) split the joint optimization problem into independent sub-problems. For example, when the trajectory is fixed, optimize the sensor transmission power, or when the power is fixed, optimize the UAV position, reducing the computational complexity and accelerating convergence. By iteratively approaching the global optimal solution, it avoids the difficulties of non-convex optimization caused by multi-variable coupling. In addition, it supports real-time adjustment in a dynamic environment, such as updating the denoising factor according to the channel state, ensuring that the aerial computing error is always below the threshold.

[0041] Furthermore, deep reinforcement learning (DRL) combines the Actor-Critic framework, uses deep convolutional networks to perceive the environmental state in real time (such as user distribution, channel quality), and generates optimal trajectories and access strategies. During the training process, experience replay and target network techniques stabilize the learning process and avoid local optima. After convergence, the policy network outputs smooth trajectories that satisfy the UAV dynamics constraints (maximum speed, turning radius). In actual deployment, the DRL algorithm can dynamically adjust the trajectory to avoid sudden interferences (such as newly added obstacles), and correct the trajectory deviation through Kalman filtering to ensure communication and computing performance (experiments show an average success rate of 90%). Compared with traditional path planning algorithms, DRL significantly improves the response speed and adaptability in complex scenarios.

[0042] It can be understood that the beneficial effects of the second to sixth aspects above can be referred to the relevant descriptions in the first aspect above, and will not be elaborated here.

[0043] In summary, through phased modeling and optimization, the present invention solves the problems of interference management, resource allocation, and dynamic planning in UAV-assisted aerial computing. By constructing a basic model to separate computing and communication requirements; achieving efficient resource allocation through multi-objective optimization and alternating iteration; introducing DRL to enhance the intelligent decision-making ability in a dynamic environment; and demonstrating the comprehensive advantages of high-precision computing and low-latency communication in scenarios such as federated learning and vehicle-to-everything networks.

[0044] Next, through the drawings and embodiments, the technical solutions of the present invention will be further described in detail. Description of the Drawings

[0045] To more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required to be used in the embodiments of the present application will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application, and those of ordinary skill in the art can obtain other drawings based on these drawings without creative efforts.

[0046] Figure 1 It is a flowchart of a method and system for integrated transmission and flight path planning of UAV aerial computing and communication proposed by the present invention; Figure 2 This is a comparison chart of UAV trajectories under different air computing error constraints in the separated distributed topology (where the distribution areas of users and sensors are far apart) provided by an embodiment of the present invention; Figure 3 This is a performance comparison chart provided by an embodiment of the present invention with fixed user transmission strategies and fixed sensor transmission strategies under different task durations; Figure 4 This is a performance comparison chart provided by an embodiment of the present invention with the straight - line flight trajectory scheme in the separated distributed topology and the hybrid distributed topology (where the distribution areas of users and sensors are relatively close); Figure 5 This is a schematic diagram of a computer device provided by an embodiment of the present invention; Figure 6 This is a block diagram of an electronic device provided by an embodiment of the present invention.

[0047] Among them, 60. Computer device; 61. Processor; 62. Memory; 63. Computer program; 600. Electronic device; 610. Processing unit; 620. Storage unit; 6201. Random access storage unit; 6202. Cache storage unit; 6203. Read - only storage unit; 6204. Program / utilities; 6205. Program module; 630. Bus; 640. Display unit; 650. Input / output interface; 660. Network adapter; 700. External device. Detailed implementation manners

[0048] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0049] In the description of the present invention, it should be understood that the terms "include" and "comprise" indicate the presence of the described features, wholes, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their combinations.

[0050] It should also be understood that the terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the specification of the present invention and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an", and "the" are intended to include the plural forms.

[0051] It should be further understood that the term "and / or" as used in the specification and appended claims of the present invention refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations. For example, A and / or B can represent three situations: A exists alone, A and B exist simultaneously, and B exists alone. Additionally, in the present invention, the character " / " generally indicates an "or" relationship between the associated objects before and after.

[0052] It should be understood that although terms such as first, second, third, etc. may be used in the embodiments of the present invention to describe preset ranges, etc., these preset ranges should not be limited to these terms. These terms are only used to distinguish the preset ranges from each other. For example, without departing from the scope of the embodiments of the present invention, the first preset range may also be referred to as the second preset range, and similarly, the second preset range may also be referred to as the first preset range.

[0053] Depending on the context, the word "if" as used herein may be interpreted as "when" or "while" or "in response to determining" or "in response to detecting". Similarly, depending on the context, the phrase "if determined" or "if detected (stated condition or event)" may be interpreted as "when determined" or "in response to determining" or "when detecting (stated condition or event)" or "in response to detecting (stated condition or event)".

[0054] Various structural schematic diagrams according to the disclosed embodiments of the present invention are shown in the drawings. These figures are not drawn to scale, where for the purpose of clear expression, some details are enlarged and some details may be omitted. The shapes of various regions and layers shown in the figures and their relative sizes and positional relationships are merely exemplary, and in practice, there may be deviations due to manufacturing tolerances or technical limitations, and those skilled in the art can design regions / layers with different shapes, sizes, and relative positions according to actual needs.

[0055] The present invention provides a method for integrated transmission and trajectory planning of unmanned aerial vehicle (UAV) air computing and communication, adopting.

[0056] Embodiment 1 A method for integrated transmission and trajectory planning of UAV air computing and communication according to the present invention includes the following steps: S1. Construct a UAV air computing and communication integration model composed of one UAV, M users, and J sensor nodes; Consider a UAV air computing and communication integrated transmission model, mainly including one UAV, communication users, , air computing sensors, The position coordinates of the user and the sensor are represented by and respectively.

[0057] The flight time of the UAV is discretized into time slots, which are respectively denoted as . The duration of each time slot is . At the n th time slot, the UAV flight path point is denoted as , and are respectively the coordinates and coordinates of the UAV, H being the flight altitude of the UAV. The starting point and the ending point of the UAV are and respectively. In each time slot, the UAV receives sensor and user data simultaneously, where the user communicates with the UAV using the time division multiple access protocol. At the n th time slot, the channel model between the user and the UAV is:[[]]

[0058]

[0059] where is the carrier frequency, c is the speed of light, and are the additional losses of the line-of-sight and non-line-of-sight links respectively. is the probability of the line-of-sight link between the user and the UAV, ,

[0060] where and depend on the propagation environment. Similarly, the channel between the channel sensor j and the UAV is defined as , and its form is the same as .

[0061] To avoid interference between users, the user scheduling is defined as a set of binary variables . In the time slot , the signal received by the UAV is:[[]]

[0062] where is the transmit precoding coefficient of the sensor , is the preprocessed signal of the sensor , is the user 's user scheduling strategy is the transmit power of the user , is the transmit signal of the user , is additive white Gaussian noise, i.e., .

[0063] S2. Based on the UAV air computing and communication integrated model in step S1, construct a sensor-UAV air computing model; In the UAV air computing and communication integrated model, sensor data is aggregated by the UAV through air computing technology. In time slot , the preprocessed signal transmitted by the sensor is expressed as:

[0064] where is the preprocessing function of the sensor in time slot , is the original data of the sensor in time slot . In the data aggregation process of the UAV, the sum of the distributed signals from all sensors corresponds to the natural superposition of signals in the channel, and then the UAV post-processes the superimposed signals to obtain the target calculated value. In time slot , the calculation objective function of sensor data is:

[0065] Here, the mean operation of sensor data is used as the post-processing function of the UAV. In practice, the UAV post-processes and scales the signal to obtain the expected estimated average value:

[0066] where is the UAV receive normalization factor for signal power compensation and noise suppression.

[0067] Due to the influence of channel fading, interference, and noise, the mean square error (MSE) between the estimated value and the true value is needed to quantify the air computing error. In time slot , the MSE of air computing is expressed as:

[0068] Among them, represents taking the expected value of the randomness of the original signal and , the receiver noise . The sensor transmission symbols are independent and normalized with zero mean and unit variance, that is , , .

[0069] The MSE is further expressed as:

[0070] Among them, is the user interference received by the UAV during air computing in time slot , is the background noise power.

[0071] Meanwhile, based on the UAV air computing and communication integrated model in step S1, a user transmission model based on time division multiple access is constructed; In the UAV air computing and communication integrated model, for the uplink data transmission from the user to the UAV, in time slot , the transmission rate of user communication is:

[0072] Among them, is the interference power of air computing.

[0073] Considering that the UAV can access at most one user in each time slot, the user scheduling strategy is defined as:

[0074]

[0075] In addition, to ensure the fairness of the user scheduling strategy, the total number of time slots for users to access the UAV is constrained:

[0076] Among them, represents floor function.

[0077] S3. According to the sensor-UAV air computing model and the user transmission model based on time division multiple access, considering the mutual interference problem between air computing and communication, introducing the maximum error constraint of air computing, and constructing an optimization model for maximizing the sum of user transmission rates; Considering the transmission interference of aerial computing and communication, by jointly optimizing the user transmission power, sensor transmission precoding coefficients, UAV normalization factor, user scheduling strategy, and UAV trajectory, the sum of user transmission rates is maximized under the constraint of aerial computing error. Mathematically, the optimization problem is formulated as:

[0078] Among them, constraint C1 is the constraint of the mean squared error (MSE) of aerial computing, Γ is the maximum error threshold of aerial computing, constraints C2 and C3 are the maximum power constraints of the sensor and the user respectively, and are the maximum transmission powers of the sensor and the user respectively. Constraints C8 and C9 limit the upper bounds of the transmission powers of the sensor and the user respectively, is the maximum flight speed of the UAV.

[0079] S4. Based on the current user access strategy and UAV flight path coordinates, design a transmission strategy based on the alternating iteration algorithm for the optimization model in step S3; Based on the current time slot fixed user access strategy and UAV flight path coordinates , design a transmission strategy by optimizing the following optimization problem.

[0080]

[0081] This problem optimizes the transmission strategies of the user and the sensor and the UAV normalization factor alternately. The transmission strategy optimization problem is expressed as follows:

[0082] Since the objective function in this problem is non-jointly concave with respect to the transmission power and precoding coefficients, it is necessary to introduce a slack variable as the upper bound to relax the problem, and transform this objective function into:

[0083] Since the objective function in the above form is still non-jointly concave with respect to the user transmission power and the slack variable , it is necessary to further introduce a slack variable to process the objective function as follows:

[0084] Among them, is the optimal solution.

[0085] So far, the transmission strategy optimization problem can be re-expressed in the following form:

[0086] This optimization problem can be solved by alternately optimizing and until the result converges.

[0087] For the UAV normalization factor optimization problem, this problem is expressed as the following feasibility check problem:

[0088] This problem is mainly affected by the constraint condition C1. Since the constraint C1 is a quadratic function of , the most satisfactory solution to this feasibility check problem can be expressed as a quadratic optimization problem with respect to the constraint C1, which is expressed as follows:

[0089] It is easy to obtain that the solution of the UAV normalization factor is:

[0090] S5. Design a user access strategy and a UAV trajectory planning strategy based on the deep reinforcement learning framework. During the deep reinforcement learning training process, the optimal transmission strategy is obtained according to the transmission strategy optimization algorithm in step S4 in each time slot, and the user access strategy and the UAV trajectory planning strategy are learned by training the policy network until the algorithm converges.

[0091] Design an algorithm framework based on deep reinforcement learning to optimize the user access strategy and the UAV trajectory planning strategy, and adopt the transmission strategy optimization algorithm in step S4 to obtain the optimal transmission strategy in each time slot.

[0092] The deep reinforcement learning algorithm needs to describe the user access and UAV trajectory planning problems as a Markov decision process, which mainly consists of the following three parts: ① State space : The state value at the -th time slot is defined as the set of the access state of the current user, the sum of the current user transmission rates, the current air computing error, and the horizontal coordinate of the current UAV position, that is

[0093] ② Action space : The action value at the -th time slot is defined as the change amount of the UAV position in the two-dimensional coordinate and the access strategy of the current time slot, that is:

[0094] ③ Reward function : The time slot, for the state take an action The obtained reward function is defined as the user access reward and the UAV trajectory reward The sum is defined as follows:

[0095] where is the reward for the UAV to reach the destination at the last time slot.

[0096] Those skilled in the art can understand that various aspects of the present invention can be implemented as a system, method or program product. Therefore, various aspects of the present invention can be specifically implemented in the following forms, namely: a complete hardware implementation, a complete software implementation (including firmware, microcode, etc.), or an implementation combining hardware and software aspects, which can be collectively referred to as "circuit", "module" or "platform" here.

[0097] Embodiment 2 The present invention provides a UAV integrated air computing and communication transmission and trajectory planning system, which can be used to implement the above-mentioned UAV integrated air computing and communication transmission and trajectory planning method. Specifically, the UAV integrated air computing and communication transmission and trajectory planning system includes a construction module, a constraint module, a strategy module and an output module.

[0098] Among them, the construction module constructs a UAV integrated air computing and communication model including UAVs, M users and J sensor nodes. Based on the UAV integrated air computing and communication model, a sensor-UAV air computing model and a user transmission model based on time division multiple access are respectively constructed; The constraint module introduces the maximum error constraint of air computing according to the sensor-UAV air computing model and the user transmission model based on time division multiple access, and constructs an optimization model for maximizing the sum of user transmission rates; The strategy module designs a transmission strategy based on the alternating iteration algorithm for the optimization model based on the current user access strategy and the UAV trajectory coordinates; The output module designs a user access strategy and a UAV trajectory planning strategy based on the deep reinforcement learning framework. During the deep reinforcement learning training process, the optimal transmission strategy is obtained according to the transmission strategy optimization algorithm in each time slot, and the user access strategy and the UAV trajectory planning strategy are learned by training the strategy network until convergence, and a dynamically optimized smooth trajectory is generated.

[0099] Embodiment 3 The present invention provides a terminal device, which includes a processor and a memory. The memory is used to store a computer program, and the computer program includes program instructions. The processor is used to execute the program instructions stored in the computer storage medium. The processor may be a Central Processing Unit (CPU), or may also be other general-purpose processors, Graphics Processing Unit (GPU), Tensor Processing Unit (TPU), Digital Signal Processor (DSP), Application Specific Integrated Circuit (ASIC), Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, and is suitable for implementing one or more instructions. Specifically, it is suitable for loading and executing one or more instructions to implement the corresponding method flow or corresponding function. The processor described in the embodiments of the present invention can be used for the operations of the integrated transmission and trajectory planning method of unmanned aerial vehicle (UAV) air computing and communication, including: Constructing an integrated UAV air computing and communication model including UAVs, M users, and J sensor nodes; based on the obtained integrated UAV air computing and communication model, respectively constructing a sensor-UAV air computing model and a user transmission model based on time division multiple access; according to the sensor-UAV air computing model and the user transmission model based on time division multiple access, introducing an air computing maximum error constraint, and constructing an optimization model for maximizing the sum of user transmission rates; based on the current user access strategy and UAV trajectory coordinates, designing a transmission strategy based on an alternating iteration algorithm for the optimization model; designing a user access strategy and a UAV trajectory planning strategy based on a deep reinforcement learning framework. During the deep reinforcement learning training process, the optimal transmission strategy is obtained according to the transmission strategy optimization algorithm in each time slot, and the user access strategy and the UAV trajectory planning strategy are learned through training the policy network until convergence, generating a dynamically optimized smooth trajectory.

[0100] Please refer to Figure 5, the terminal device is a computer device. The computer device 60 in this embodiment includes: a processor 61, a memory 62, and a computer program 63 stored in the memory 62 and executable on the processor 61. When the computer program 63 is executed by the processor 61, it implements the integrated transmission and trajectory planning method of the unmanned aerial vehicle for aerial computing and communication in the embodiment. To avoid repetition, it will not be elaborated here one by one. Alternatively, when the computer program 63 is executed by the processor 61, it implements the functions of each model / unit in the integrated transmission and trajectory planning system of the unmanned aerial vehicle for aerial computing and communication in the embodiment. To avoid repetition, it will not be elaborated here one by one.

[0101] The computer device 60 can be a computing device such as a desktop computer, a notebook, a palm computer, and a cloud server. The computer device 60 may include, but is not limited to, a processor 61 and a memory 62. Those skilled in the art can understand that Figure 5 merely examples of the computer device 60, which do not constitute a limitation on the computer device 60, may include more or fewer components than shown in the figure, or combine certain components, or different components. For example, the computer device may also include input / output devices, network access devices, buses, etc.

[0102] The so-called processor 61 may be a central processing unit (CPU), or may also be other general-purpose processors, a graphics processing unit (GPU), a tensor processing unit (TPU), a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0103] The memory 62 may be an internal storage unit of the computer device 60, such as the hard disk or memory of the computer device 60. The memory 62 may also be an external storage device of the computer device 60, such as a plug-in hard disk equipped on the computer device 60, a smart media card (SMC), a secure digital (SD) card, a flash card, etc.

[0104] Further, the memory 62 may also include both the internal storage unit of the computer device 60 and external storage devices. The memory 62 is used to store computer programs as well as other programs and data required by the computer device. The memory 62 may also be used to temporarily store data that has been output or is to be output.

[0105] Please refer to Figure 6 , the terminal device is the electronic device 600, and the electronic device 600 is presented in the form of a general computing device. The components of the electronic device may include but are not limited to: at least one processing unit 610, at least one storage unit 620, a bus 630 connecting different platform components (including the storage unit 620 and the processing unit 610), a display unit 640, etc.

[0106] Among them, the storage unit stores program codes, and the program codes can be executed by the processing unit 610, so that the processing unit 610 executes the steps according to various exemplary embodiments of the present invention described in the above method part of this specification. For example, the processing unit 610 may execute steps as shown in Figure 1 .

[0107] The storage unit 620 may include a readable medium in the form of a volatile storage unit, such as a random access storage unit (RAM) 6201 and / or a cache storage unit 6202, and may further include a read-only storage unit (ROM) 6203.

[0108] The storage unit 620 may also include a program / utilities 6204 having a set (at least one) of program modules 6205. Such program modules 6205 include but are not limited to: an operating system, one or more application programs, other program modules, and program data. Each or some combination of these examples may include the implementation of a network environment.

[0109] The bus 630 may represent one or more of several types of bus structures, including a memory bus or a memory controller, a peripheral bus, a graphics acceleration port, a processing unit, or a local bus using any one of the multiple bus structures.

[0110] The electronic device 600 can also communicate with one or more external devices 700 (such as a keyboard, a pointing device, a Bluetooth device, etc.), and can also communicate with one or more devices that enable a user to interact with the electronic device 600, and / or communicate with any device (such as a router, a modem) that enables the electronic device 600 to communicate with one or more other computing devices. Such communication can be carried out through the input / output interface 650. Moreover, the electronic device 600 can also communicate with one or more networks (such as a local area network, a wide area network, and / or a public network, such as the Internet) through the network adapter 660. The network adapter 660 can communicate with other modules of the electronic device 600 through the bus 630. It should be understood that, although not shown in the figure, other hardware and / or software modules can be used in combination with the electronic device 600, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage platforms, etc.

[0111] Embodiment 4 The present invention also provides a storage medium, specifically a computer-readable storage medium, which is a memory device in a terminal device for storing programs and data. It can be understood that the computer-readable storage medium here can include both the built-in storage medium in the terminal device and, of course, the extended storage medium supported by the terminal device. It can be any tangible medium that contains or stores a program, and this program can be used by or in combination with an instruction execution system, apparatus, or device. The computer-readable storage medium provides a storage space that stores the operating system of the terminal. And, in this storage space, one or more instructions suitable for being loaded and executed by the processor are also stored. These instructions can be one or more computer programs (including program codes). It should be noted that more specific examples of the computer-readable storage medium here include: an electrical connection having one or more wires, a portable disk, a hard disk, a random access memory, a read-only memory, an erasable programmable read-only memory, an optical fiber, a portable compact disk read-only memory, an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0112] The computer-readable storage medium also includes a data signal propagated in a baseband or as part of a carrier wave, which carries the readable program code. Such a propagated data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The readable storage medium can also be any readable medium other than the readable storage medium, and this readable medium can send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on the readable storage medium can be transmitted using any appropriate medium, including but not limited to wireless, wired, optical fiber, radio frequency, etc., or any suitable combination of the above.

[0113] The program code for performing the operations of the present invention can be written in any combination of one or more programming languages. The programming languages include object-oriented programming languages such as Java, C++, etc., and also include conventional procedural programming languages such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computing device, partially on the user's device, executed as an independent software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device can be connected to the user's computing device through any type of network, including a local area network or a wide area network, or can be connected to an external computing device (for example, by using an Internet service provider to connect through the Internet).

[0114] One or more instructions stored in the computer-readable storage medium can be loaded and executed by a processor to implement the corresponding steps of the method for integrated transmission and trajectory planning of unmanned aerial vehicle air computing communication in the above embodiments; one or more instructions in the computer-readable storage medium are loaded and executed by the processor to perform the following steps: Construct a model for integrated unmanned aerial vehicle air computing communication including an unmanned aerial vehicle, M users, and J sensor nodes; based on the obtained model for integrated unmanned aerial vehicle air computing communication, respectively construct a sensor-unmanned aerial vehicle air computing model and a user transmission model based on time division multiple access; according to the sensor-unmanned aerial vehicle air computing model and the user transmission model based on time division multiple access, introduce an air computing maximum error constraint, and construct an optimization model for maximizing the sum of user transmission rates; based on the current user access strategy and the unmanned aerial vehicle trajectory coordinates, design a transmission strategy based on an alternating iteration algorithm for the optimization model; design a user access strategy and an unmanned aerial vehicle trajectory planning strategy based on a deep reinforcement learning framework. During the deep reinforcement learning training process, obtain the optimal transmission strategy according to the transmission strategy optimization algorithm in each time slot, and learn the user access strategy and the unmanned aerial vehicle trajectory planning strategy through training the policy network until convergence, and generate a dynamically optimized smooth trajectory.

[0115] In each of the embodiments provided in this application, the databases involved may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., and is not limited thereto. In each of the embodiments provided in this application, the processors involved may be general-purpose processors, central processors, graphics processors, digital signal processors, programmable logic devices, data processing logics based on quantum computing, etc., and are not limited thereto.

[0116] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Usually, the components of the present invention described and shown in the accompanying drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed present invention, but merely represents selected embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0117] The technical effects of the present invention will be described in detail below in combination with simulations.

[0118] The present invention simulates the integrated transmission and flight planning method and system of unmanned aerial vehicle (UAV) air computing and communication to verify the superiority of the solution of the present invention. The specific steps are as follows: The designed basic parameters are that in a network area of 1000m×1000m, there are 15 users, 36 sensors, and one UAV distributed randomly in this area. The height of the UAV is 100m, the maximum speed is 30m / s, the total mission duration is 60s, and the duration of each time slot is 1s.

[0119] The channel noise power is -95dBm, the carrier frequency is 2GHz, and the channel parameters adopt the air-ground channel parameters in a suburban environment.

[0120] The maximum transmission powers of the sensors and users are 0.05W and 0.2W respectively. The simulation results of the relevant performance of the present invention are as Figure 2 shown.

[0121] In summary, for the integrated transmission and trajectory planning method and system of UAV aerial computing and communication of the present invention, it can realize that the UAV simultaneously assists ground users in communication and aerial computing, which can improve communication efficiency and effectively utilize spectrum resources. In particular, considering the interference problem between communication and aerial computing, since the UAV has limited energy and computing power, it is particularly sensitive to interference. To address this issue, an integrated transmission framework for UAV aerial computing and communication is proposed. The UAV serves as an aerial aggregation center to execute the aerial computing data aggregation task of sensors, and at the same time serves as an aerial base station to communicate wirelessly with ground users. Considering the transmit power of users, the transmit coefficient of sensors, the normalization factor, and the UAV trajectory comprehensively, an algorithm framework based on deep reinforcement learning is proposed to ensure the accuracy requirements of aerial computing, reduce interference, and maximize the transmission quality of users. In addition, the proposed reward function combines the UAV destination requirements and the fairness of user scheduling. Simulations show that the proposed solution has advantages in performance compared with the comparison scheme.

[0122] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the above division of each functional unit and module is used as an example. In actual applications, the above functions can be assigned to different functional units and modules according to needs, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. Each functional unit and module in the embodiment can be integrated into a processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above integrated unit can be implemented in the form of hardware or in the form of a software functional unit. In addition, the specific names of each functional unit and module are only for the convenience of mutual distinction and do not limit the protection scope of the present application. The specific working processes of the units and modules in the above system can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0123] In the above embodiments, the descriptions of the respective embodiments have their own emphases. For parts not detailed or recorded in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0124] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed in the present invention can be implemented by electronic hardware, or by a combination of computer software and electronic hardware. Whether these functions are executed in hardware or software depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

[0125] In the embodiments provided by the present invention, it should be understood that the disclosed device / terminal and method can be implemented in other ways. For example, the device / terminal embodiments described above are merely illustrative. For example, the division of the modules or units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection between each other can be through some interfaces. The indirect coupling or communication connection of the device or unit can be in electrical, mechanical or other forms.

[0126] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0127] In addition, in each embodiment of the present invention, the functional units can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above integrated units can be implemented in the form of hardware or in the form of software functional units.

[0128] If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, to implement all or part of the processes in the above method embodiments of the present invention, it can also be completed by a computer program instructing relevant hardware. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file or some intermediate form. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc. It should be noted that the content included in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not include electrical carrier signals and telecommunication signals.

[0129] This application is described with reference to the flowcharts and / or block diagrams of methods, apparatus, and computer program products according to embodiments of the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing device generate means for implementing the functions specified in one or more flows and / or blocks Figure 1 one or more flows and / or blocks Figure 1 in one or more blocks.

[0130] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including instruction means that implement the functions specified in one or more flows and / or blocks Figure 1 one or more flows and / or blocks Figure 1 in one or more blocks.

[0131] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operational steps are performed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one or more flows and / or blocks Figure 1 one or more flows and / or blocks Figure 1 in one or more blocks.

[0132] The above content is only for explaining the technical idea of the present invention and cannot be used to limit the protection scope of the present invention. Any modification made on the basis of the technical solution according to the technical idea proposed by the present invention falls within the protection scope of the claims of the present invention.

Claims

1. An integrated transmission and trajectory planning method for airborne computing and communication of drones, characterized in that, It includes the following steps: Construct a unified model of aerial computing and communication for drones that includes drones, M users, and J sensor nodes; Based on the obtained integrated aerial computing and communication model of the unmanned aerial vehicle (UAV), respectively construct a sensor-UAV aerial computing model and a user transmission model based on time division multiple access (TDMA); according to the sensor-UAV aerial computing model and the user transmission model based on TDMA, introduce the maximum error constraint of aerial computing, and construct an optimization model for maximizing the sum of user transmission rates; Based on the current user access strategy and UAV trajectory coordinates, design a transmission strategy based on the alternating iteration algorithm for the optimization model; Design a user access strategy and a UAV trajectory planning strategy based on the deep reinforcement learning framework. During the deep reinforcement learning training process, obtain the optimal transmission strategy according to the transmission strategy optimization algorithm in each time slot, and learn the user access strategy and the UAV trajectory planning strategy through training the policy network until convergence, generating a dynamically optimized smooth trajectory.

2. The integrated transmission and flight path planning method for aerial computing and communication of the drone according to claim 1, characterized in that In the integrated air computing and communication model of the UAV, at the n time slot, the channel model between the user and the UAV is as follows: Wherein, is the carrier frequency, c is the speed of light, and are the additional losses of the LOS and NLOS links respectively, is the user and the probability of the LOS link between the user and the UAV; In a time slot the signal received by the drone is: wherein, is the transmit precoding coefficient of the sensor ; is the preprocessed signal of the sensor ; is the user scheduling strategy of the user ; is the transmit power of the user ; is the transmit signal of the user ; is the additive white Gaussian noise.

3. The integrated transmission and trajectory planning method for aerial computing and communication of an unmanned aerial vehicle according to claim 1, characterized in that, The sensor-UAV aerial computing model is: Among them, represents the original signal and , the receiver noise takes the expected value of randomness, is the user interference received by the UAV during air computing in time slot , and is the background noise power; In the user transmission model based on time division multiple access, in a time slot the transmission rate of user communication is as follows: Among them, is the interference power of aerial computing; Considering that a drone can only connect to one user at most in each time slot, a user scheduling strategy is defined as follows: Constrain the total number of time slots for users to access the UAV: Among them, represents rounding down.

4. The integrated transmission and trajectory planning method for aerial computing and communication of an unmanned aerial vehicle according to claim 3, characterized in that, In a time slot the objective function for calculating sensor data is: Among them, is the number of sensors, is the sensor set, is the sensor preprocessed signal transmitted; Expected estimated average value: Among them, is the UAV reception normalization factor, is the signal received by the UAV in the n time slot.

5. The integrated transmission and trajectory planning method for aerial computing and communication of an unmanned aerial vehicle according to claim 1, characterized in that The optimization model for maximizing the sum of user transmission rates is: Among them, the constraint C1 is the constraint of the air-computation error MSE, Γ is the maximum error threshold of air-computation, and the constraints C2 and C3 are the maximum power constraints of the sensor and the user respectively. and are the maximum transmission powers of the sensor and the user respectively. The constraints C8 and C9 respectively limit the upper limits of the transmission powers of the sensor and the user. is the maximum flight speed of the UAV. is the n transmission precoding coefficient of the sensor in the th time slot. is the n transmission power of the user in the m th time slot. is the n access strategy of the user in the m th time slot. is the n received normalization factor of the UAV in the th time slot. n is the position coordinate of the UAV in the th time slot. n is the transmission rate of the user in the th time slot. n is the m channel gain between the user and the UAV in the th time slot. is the standard deviation of Gaussian white noise. is the n channel gain between the sensor and the UAV in the j th time slot. is the set of time slots. is the set of sensors. is the number of users. is the user set. is the maximum flight speed of the UAV. is the duration of each time slot. is the take-off position coordinate of the UAV. is the destination coordinate of the UAV.

6. The integrated transmission and trajectory planning method of UAV air computing and communication according to claim 1, characterized in that The optimization problem of the transmission strategy is as follows: wherein, is the access strategy of the user n in the m time slot, is the received normalization factor of the UAV in the n time slot, is the n channel gain between the user m and the UAV in the time slot, n is the transmit power of the user m in the time slot, where is the introduced slack variable and serves as the upper bound of the term, is the introduced slack variable for approximating the objective function, is the transmit precoding coefficient of the sensor n in the time slot, is the n channel gain between the sensor j and the UAV in the time slot, is the maximum allowable computational error for aerial computing, is the number of sensors, is the maximum transmit power of the sensors, is the standard deviation of the Gaussian white noise, is the power of the Gaussian white noise.

7. The integrated transmission and trajectory planning method for aerial computing and communication of an unmanned aerial vehicle according to claim 6, wherein The solution of the UAV normalization factor is: 。 8. The integrated transmission and trajectory planning method for airborne computing and communication of an unmanned aerial vehicle according to claim 1, wherein During the deep reinforcement learning training process, describe the user access and UAV trajectory planning problems as a Markov decision process, including: State space : The state value of the nth time slot is defined as the set of the access state of the current user, the sum of the current user transmission rates, the current air computing error, and the horizontal coordinate of the current UAV position; Action space : The action value of the nth time slot is defined as the change in the UAV position on the two-dimensional coordinates and the access strategy for the current time slot; Reward function : At the th time slot, for the state , when taking the action , the obtained reward function is defined as the sum of the user access reward and the UAV trajectory reward .

9. The integrated transmission and trajectory planning method for aerial computing and communication of an unmanned aerial vehicle according to claim 8, characterized in that, The status value of the time slot , the action value of the time slot , user access reward and UAV flight path reward are respectively: Among them, is the access strategy of the user n in the m th time slot, is the transmission rate of the user in the n th time slot, is the calculation error of air computing in the n th time slot, is the coordinate of the UAV position on the x-axis in the n th time slot, is the coordinate of the UAV position on the y-axis in the n th time slot, is the reward for the UAV to reach the destination in the last time slot, is the n th time slot, is the position coordinate of the UAV in the n th time slot, is the destination coordinate of the UAV, is the duration of each time slot.

10. An integrated transmission and trajectory planning system for airborne computing and communication of drones, characterized in that, It includes: Building module, building includes drones, M users and J a unified model of airborne computing and communication for drones with sensor nodes. Based on the unified model of airborne computing and communication for drones, a sensor-drone airborne computing model and a user transmission model based on time division multiple access are respectively constructed; A constraint module, according to the sensor-UAV aerial computing model and the user transmission model based on TDMA, introduce the maximum error constraint of aerial computing, and construct an optimization model for maximizing the sum of user transmission rates; A policy module, based on the current user access strategy and UAV trajectory coordinates, design a transmission strategy based on the alternating iteration algorithm for the optimization model; An output module, design a user access strategy and a UAV trajectory planning strategy based on the deep reinforcement learning framework. During the deep reinforcement learning training process, obtain the optimal transmission strategy according to the transmission strategy optimization algorithm in each time slot, and learn the user access strategy and the UAV trajectory planning strategy through training the policy network until convergence, generating a dynamically optimized smooth trajectory.

Images