Reliable service coverage enhancement method for unmanned aerial vehicle

Abstract

The invention provides an unmanned aerial vehicle reliable service coverage enhancement method. The method comprises the following steps: deriving a successful edge calculation service probability to measure service reliability; taking a successful edge calculation service probability as a reliability constraint, and constructing a service coverage maximization problem by optimizing the height of the unmanned aerial vehicle and a user calculation task unloading probability; the constructed optimization problem relates to a complex coupling relationship among the flight height of the unmanned aerial vehicle, ground service coverage, service reliability and a user access strategy; the coupling action relations are researched, and an improved golden section search method is adopted to solve the optimization problem; and the overall service coverage range of the system is further expanded.

Description

Technical Field

[0001] This invention relates to the fields of unmanned aerial vehicles (UAVs) and wireless communication technology, and specifically to a method for enhancing reliable service coverage for UAVs. Background Technology

[0002] Mobile edge computing enables rapid, localized service response by pushing lightweight micro-cloud servers to the network edge. It allows users to offload some or all of their computing tasks to nearby edge servers. Upon receiving the offloaded computing tasks, the edge server executes the corresponding computation and returns the results to the user. By rapidly processing task offloading within the wireless access network, mobile edge computing significantly reduces service response time, effectively supporting emerging applications such as online gaming, virtual reality, and autonomous driving.

[0003] Mobile edge computing is typically integrated into existing terrestrial mobile communication systems, such as 4G / 5G systems. While this integration is a low-cost method for realizing mobile edge computing, its performance is highly dependent on the terrestrial infrastructure deployment of the mobile system. Specifically, limited by fixed infrastructure deployment, traditional terrestrial-based mobile edge computing networks struggle to handle high-dynamic workload fluctuations while maintaining stringent quality of service, such as sudden surges in traffic in hotspot areas. Furthermore, in critical scenarios (such as emergency rescue, coverage blind spots, and densely populated urban areas), terrestrial-based mobile edge computing networks may fail to provide reliable low-latency service performance due to factors such as infrastructure damage, service overload, or deep channel fading. Therefore, finding alternative service resources to improve service reliability has become a critical issue that mobile edge computing urgently needs to address.

[0004] Unmanned aerial vehicle (UAV)-based airborne edge computing systems have become an important technical means to enhance the reliability of terrestrial edge computing network services, and integrated air-ground mobile edge computing networks have become an important component of current mobile edge computing research. UAVs, leveraging their flexibility and mobility, can provide service support capabilities and enhance local area performance on demand. Simultaneously, the strong line-of-sight transmission component of the ground-to-air transmission link will further improve the reliability of offloading computing tasks. However, existing research mostly considers the horizontal trajectory optimization design of UAVs at fixed flight altitudes, insufficiently utilizing the three-dimensional spatial degrees of freedom of UAVs and failing to fully leverage the performance potential of UAV mobile edge computing platforms. While some studies have considered UAV three-dimensional trajectory optimization, they have not fully considered the coupling between UAV three-dimensional trajectory design and service coverage and quality of service, limiting the applicability of the research solutions.

[0005] In summary, the existing technologies have the following problems: how to construct a theoretical research framework in an integrated air-ground network, comprehensively consider the coupled impact of UAV spatial location on service coverage, air-ground transmission link quality, and service reliability, and jointly optimize UAV flight trajectory and ground user access strategies to achieve enhanced reliable service coverage. Summary of the Invention

[0006] The purpose of this invention is to address how to construct a theoretical research framework in an integrated air-ground network, comprehensively consider the coupled impact of UAV spatial location on service coverage, air-ground transmission link quality, and service reliability, and jointly optimize UAV flight trajectory and ground user access strategies to achieve enhanced reliable service coverage.

[0007] Therefore, the present invention provides a method for enhancing reliable service coverage of unmanned aerial vehicles (UAVs), the method comprising the following steps:

[0008] The probability of successfully introducing edge computing services is used to measure the reliability of system services.

[0009] Construct a model for maximizing drone service coverage based on the probability of successful edge computing services;

[0010] Analyze the coupling variables in the model of the drone service coverage maximization problem;

[0011] Utilize the obtained analysis results to optimize the coverage of drone services.

[0012] Specifically, the probability of a successful edge computing service is:

[0013]

[0014] Where β is the probability of a user unloading the drone. This represents the probability of successful communication for the drone. Calculate the probability of success for the drone. This represents the probability of successful communication at the ground base station. The probability of success is calculated at the ground base station.

[0015] Specifically, the model for maximizing drone service coverage is as follows:

[0016]

[0017] stp s (β,H)≥1-ε

[0018] 0≤β≤1

[0019] H>0

[0020] Where 1-ε is the minimum lower limit requirement for the probability of successful edge computing service within the drone service coverage area, β is the probability of user computing task unloading, and the value of β is between 0 and 1; H is the drone altitude, and the value of H is non-negative.

[0021] Specifically, the analysis of the coupling variables in the model for maximizing drone service coverage includes:

[0022] This study investigates the implicit features in the model for maximizing unmanned aerial vehicle (UAV) service coverage and analyzes the influence between coupling variables.

[0023] The reliable service coverage optimization problem is solved by utilizing the influence between the coupled variables.

[0024] Specifically, the coupling variables include: the UAV's spatial location and the probability of the user's computational task being unloaded.

[0025] The above technical solution has the following beneficial effects:

[0026] The technical solution of this invention demonstrates significant technical effects in many aspects, such as improving service reliability, expanding service coverage, and optimizing resource utilization, as detailed below:

[0027] Ensuring Service Reliability: A service reliability evaluation system under an integrated air-ground network architecture was constructed. Successful communication and successful computation probabilities were derived sequentially, quantifying the reliability of offloading transmission and computation processing available to users accessing UAVs or ground base stations. Subsequently, considering both communication and computation dimensions, the probability of successful edge computing services was defined to assess network service reliability. Then, given the minimum edge computing service probability requirement within the service coverage area, reliability within the coverage area was ensured by comprehensively utilizing the flexibility of UAVs and the basic service coverage of ground infrastructure.

[0028] Expanding service coverage: With the optimization goal of maximizing reliable service coverage, this approach effectively utilizes the limited service resources of the integrated air-to-ground mobile edge computing network by optimizing drone altitude and user task offloading probability, thereby increasing its service coverage and ultimately enhancing the number of users that limited edge computing resources can support. On one hand, adjusting drone altitude can alter the service coverage radius; on the other hand, adjusting the user computing task offloading probability β coordinates the resource utilization of drones and ground base stations, further expanding the overall service coverage of the system.

[0029] Efficiently Solving Optimization Problems: Addressing the difficulty in solving the service coverage maximization problem due to the coupling between drone altitude and user computation task offloading probability, this paper delves into the structural characteristics of the problem, analyzing the combined impact of drone altitude and user computation task offloading probability on communication and computing service reliability. Based on this analysis, an improved golden section search algorithm is proposed. This algorithm effectively addresses the limitation of standard golden section search due to the unknown range of search variable values. By setting benchmark values ​​and dynamically updating the search range, it efficiently solves for the optimal drone altitude and task offloading probability while ensuring feasibility and optimality. This makes the entire technical solution highly operable and practical in real-world applications.

[0030] Adapting to Real-World Network Environments: Considering that the number and location of users are difficult to obtain in advance and change dynamically over time, random geometry is used to handle the randomness of user locations. A homogeneous Poisson point process is used to model the random spatial distribution of users, which better reflects the characteristics of real-world network scenarios. Simultaneously, a more accurate ground-to-air transmission link model is employed, comprehensively considering the small-scale fading and large-scale path loss exponent of the UAV altitude-to-ground-to-air transmission link. This allows the system model to more accurately reflect the actual link characteristics, enabling the entire technical solution to operate more stably and efficiently in real-world network environments. Attached Figure Description

[0031] Figure 1 This is a flowchart of a method for enhancing reliable service coverage for unmanned aerial vehicles (UAVs) according to an embodiment of the present invention;

[0032] Figure 2 This is a performance comparison chart of a method for enhancing reliable service coverage for unmanned aerial vehicles (UAVs) and a non-cooperative solution provided in an embodiment of the present invention.

[0033] Figure 3 This is a performance comparison chart of a reliable service coverage enhancement method for unmanned aerial vehicles (UAVs) and a fixed-altitude solution provided in an embodiment of the present invention. Detailed Implementation

[0034] The technical solutions of 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. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0035] In embodiments of the present invention, such as Figure 1 This invention provides a method for enhancing reliable service coverage for unmanned aerial vehicles (UAVs), the method comprising the following steps:

[0036] S101: Introduce the probability of successful edge computing services to measure system service reliability; use the probability of successful edge computing services as a service reliability constraint.

[0037] S102: Construct a model for maximizing drone service coverage based on the probability of successful edge computing services; construct the service coverage maximization problem by optimizing drone altitude and task offloading probability;

[0038] S103: Analyze the coupling variables in the model of the UAV service coverage maximization problem; study the implicit structural characteristics of the constructed problem, and analyze the influence between the coupling variables;

[0039] S104: Optimize the service coverage of UAVs using the obtained analysis results; develop a low-complexity optimization method based on the golden section search to solve the constructed reliable service coverage optimization problem by utilizing the influence between coupled variables; verify the accuracy of the theoretical analysis and the superiority of the proposed service coverage enhancement method using numerical simulation.

[0040] As one implementation method, using the obtained success probability of edge computing services as a performance constraint, the reliable service coverage maximization problem is first constructed as follows:

[0041]

[0042] stp s (β,H)≥1-ε

[0043] 0≤β≤1

[0044] H>0

[0045] Specifically, regarding the probability of successful communication, since its expression is independent of the user's calculated task unloading probability, it is only necessary to analyze the impact of the drone's altitude on the probability of successful communication. This is achieved by solving for the probability of successful communication. Regarding the first derivative information of the drone's altitude H, we can obtain the following: as the drone's altitude increases, the probability of successful communication at the drone decreases monotonically, while the probability of successful communication at the ground base station first increases and then decreases.

[0046] To calculate the probability of success, analyze the probability of successful communication. Regarding the first derivatives of the drone altitude H and the user offloading probability β, we can conclude that: as the drone altitude increases, the probability of successful computation at both the drone and the ground base station monotonically decreases. Conversely, as the probability of user computation task migration increases, the probability of successful computation at the drone monotonically decreases, while the probability of successful computation at the ground base station monotonically increases.

[0047] Considering the impact of drone altitude on the probabilities of successful communication and successful computation, proof by contradiction shows that when the constructed reliable service coverage optimization problem reaches its optimal value, the constraint on the probability of successful edge computation necessarily holds in equation form. Furthermore, for a given value of the probability of successful edge computation, as the user unloading probability monotonically decreases from 1 to 0, the corresponding maximum drone flight altitude first increases and then decreases. Utilizing this unimodal characteristic, a globally optimal algorithm based on the golden section algorithm can be constructed to obtain the optimal drone altitude and the probability of user computation migration and unloading. Theoretical analysis proves that the proposed scheme converges to the globally optimal solution with logarithmic complexity.

[0048] Furthermore, this invention constructs an integrated air-ground network architecture that accurately characterizes the impact of UAV spatial location on reliable service coverage capabilities. The architecture comprises the following main components:

[0049] The study depicted the impact of the UAV's spatial location on large-scale path loss and small-scale random fading in the ground-to-air transmission link.

[0050] The impact of large-scale path loss and small-scale random fading on the ground-to-ground transmission link between users and ground base stations within the coverage area of ​​the UAV service was analyzed.

[0051] The impact of drone spatial location on service coverage and computing load is described;

[0052] The probability of successful computation is used as a reliability index, and a theoretical expression for the probability of successful computation is derived.

[0053] Specifically, consider an integrated air-ground mobile edge computing network consisting of a drone and a ground base station. The spatial distribution of users is modeled as a uniform Poisson point process with density λ. Each user's computational task is represented by a triple [I,C,T], where I is the task size (bits), C is the number of CPU cycles required for the computational task, and T is the maximum tolerable latency of the computational task. Each computational task is indivisible; therefore, a probabilistic full offload model is adopted, i.e., the user randomly offloads the entire task to the drone with an offload probability β, or to the ground base station with a probability 1-β. The offload probability β quantifies the user's preference for accessing the drone to obtain service support. By optimizing the offload probability β, the resource utilization between the drone and the ground base station can be coordinated.

[0054] Specifically, to avoid co-channel interference caused by offloading computing tasks from different users, all users adopt an orthogonal multiple access strategy to access the drone or ground base station. (The rest of the text appears to be incomplete and requires further context.) u ,y u The user of [0] decides to send the data to the location V = [x] v ,y v For the drone unloading task of [H], the signal-to-noise ratio received by the drone is:

[0055]

[0056] Where Ω represents the small-scale fading power gain of the ground-to-air transmission link, and P t This represents the user's transmit power, ||·|| represents the Euclidean distance, and α is the transmit power. v σ represents the large-scale path transmission loss factor of the ground-to-air transmission link. 2 It is the additive white Gaussian noise power at the drone receiver.

[0057] Considering the impact of UAV altitude on the ground-to-air transmission link, the small-scale fading power gain Ω follows a Ricean distribution related to UAV altitude, with the probability distribution function being:

[0058]

[0059] Where I0(·) is the zeroth-order modified Bessel function, K > 0 represents the Rice factor, defined as the ratio of the power of the direct transmission component to the power of the scattered component in the transmission link. To describe the gradual increase in the proportion of the direct path component as the drone's altitude increases due to the decrease in surrounding scattering objects, the Rice factor K can be further defined as...

[0060]

[0061] Here, a1>0 and b1>0 represent environmental factors, the specific values ​​of which are related to factors such as the electromagnetic spectrum used and the characteristics of the surrounding environment. Let θ be the elevation angle between the ground user and the drone. Given a certain ground user's location, as the drone's altitude increases, the elevation angle θ between the user and the drone monotonically increases.

[0062] Furthermore, as the altitude of drones increases, the ground-to-air transmission link gradually approaches the free-space link transmission model, and the large-scale path transmission loss factor α... v Gradually decrease, that is

[0063]

[0064] Among them, a2<0, a3>0, b2>0, and b3>0 are environmental factors, and their specific values ​​are related to factors such as the electromagnetic spectrum used and the characteristics of the surrounding environment.

[0065] If user U = [x u ,y u ,0] Select access ground base station B = [x b ,y b The signal-to-noise ratio received by the ground base station is: [0]

[0066]

[0067] Where g represents the small-scale fading power gain of the ground-to-ground transmission link, and its value follows a Rayleigh distribution, α b This represents the large-scale path transmission loss factor of a ground-to-ground transmission link.

[0068] Specifically, through virtualization technology, edge servers can generate multiple logically isolated virtual machines, each handling one computing task, thus simultaneously processing multiple computing tasks offloaded by different users. However, because multiple virtual machines generated by the same edge server share the same underlying physical resources, I / O interference will occur between the virtual machines, and the actual usable computing frequency of each virtual machine will decrease as the number of virtual machines generated by the edge server increases. For running M... s The computation frequency expression for an edge server with 1 virtual machine is:

[0069]

[0070] Where s∈{v,b} is used to identify the edge server located at the UAV (denoted as v) or at the ground base station (denoted as b), f s (M s ) to run M simultaneously on edge servers s The computing frequency (CPU cycles / second) of each virtual machine when there are multiple virtual machines, f s,0 d represents the computing frequency available to a virtual machine when the server is running only a single virtual machine. s >0 represents a performance degradation factor that depends on factors such as the operating system and processor hardware configuration. Given a user computation task requiring C CPUs, and a computation latency of...

[0071]

[0072] Among them, T s,0 =C / f s,0 This indicates the computational latency when the edge server is running only one virtual machine.

[0073] To describe the reliability of edge computing services, we first define the probability of successful communication as follows, specifically for the user computing task offloading and transmission phase:

[0074]

[0075] Where s∈{v,b} is used to identify the received signal-to-noise ratio at the UAV (denoted as v) or the ground base station (denoted as b). This is the signal-to-noise ratio threshold for successful reception of user computing tasks by drones or ground base stations. R represents the area covered by the drone service, and R is the radius of the drone service coverage area. ψ represents the minimum elevation angle limit when a ground user accesses the drone.

[0076] Taking into account both the spatial distribution of users and the distribution function of transmission links, we can derive the probability of successful communication when a user chooses to access drone service support.

[0077]

[0078] Here, Q(·,·) is the Marcum Q function.

[0079] When a user connects to a terrestrial base station to obtain service support, the probability of successful communication is:

[0080]

[0081] Wherein, the lower limit of integration D l Defined as

[0082]

[0083] D is the distance D from the user to the ground base station within the drone service coverage area. b The probability density distribution function. Based on geometric principles, let l be the distance from the UAV's ground projection point to the ground base station, and R be the radius of the UAV's ground service coverage area. The expression falls into the following two categories:

[0084] (1) When l≥R, D b The cumulative probability distribution function is

[0085]

[0086] (2) When l < R, D b The cumulative probability distribution function is

[0087]

[0088] in,

[0089]

[0090]

[0091] Subsequently, for the computation processing phase of the user's computation task, the probability of successful computation is defined as:

[0092]

[0093] Considering that the spatial distribution of users follows a uniform Poisson process, and the number of users within a given area follows a Poisson distribution, the probability of successful computation when a user accesses an edge computing service via a drone can be obtained as follows:

[0094]

[0095] in, It is determined by the computational delay requirement The maximum number of virtual machines that a given edge server on a drone can support. It is the floor operator, (·). + =max{0,·}.

[0096] When a user chooses to access a terrestrial base station to obtain services, the probability of success is calculated as follows:

[0097]

[0098] in, It is determined by the computational delay requirement The maximum number of virtual machines that an edge server mounted on a given ground base station can support.

[0099] Considering both the reliability of task offloading and transmission, and the reliability of computation processing, the probability of a successful edge computing service is defined as follows:

[0100]

[0101] Based on the law of total probability, the expression for calculating the probability of success margin can be obtained as follows:

[0102]

[0103] Where β is the probability of a user unloading the drone. This represents the probability of successful communication for the drone. Calculate the probability of success for the drone. This represents the probability of successful communication at the ground base station. The probability of success is calculated at the ground base station.

[0104] As one implementation method, a method for enhancing reliable service coverage for unmanned aerial vehicles (UAVs) specifically includes the following steps:

[0105] I. System Model

[0106] (I) System Model

[0107] Consider an integrated air-ground mobile edge computing network consisting of a fixed ground base station, a mobile drone, and multiple randomly distributed users. Each ground base station and drone is equipped with an edge server. All three are equipped with a single antenna. Drones move along arbitrary horizontal trajectories, and user mobility is modeled using a random walk model. The system follows a time-slot structure, assuming that the period length of each time slot is sufficiently small such that the changes in drone and user positions within a single time slot are negligible, changing only between adjacent time slots. Within each time slot, the user distribution is modeled by a homogeneous Poisson point process with density λ. Users within the drone's coverage area can choose to offload computational tasks to either the base station or the drone.

[0108] (II) Link Model

[0109] 1. Terrestrial Transmission Link: A terrestrial transmission link model from the user to the base station, considering both large-scale path loss and small-scale random fading, expressed as follows: Where g represents the small-scale Rayleigh fading channel power gain, i.e., g ~ exp(1), d b α represents the transmission distance from the user to the base station. b >2 represents the large-scale path transmission loss factor of the terrestrial transmission link.

[0110] 2. Ground-to-Air Transmission Link: To describe the highly correlated link characteristics, a ground-to-air link model is adopted, with a channel gain of... Where Ω represents independent and identically distributed small-scale fading, d is the transmission distance from the user to the drone, and α v Let be the large-scale path transmission loss factor of the ground-to-air transmission link. The small-scale fading of the ground-to-air transmission link follows a Ricean distribution, with the probability density function being:

[0111] Its probability distribution function is

[0112]

[0113] Where I0(·) is the zeroth-order modified Bessel function, K > 0 represents the Rice factor, defined as the ratio of the power of the direct transmission component to the power of the scattered component in the transmission link. To describe the gradual increase in the proportion of the direct path component as the drone's altitude increases due to the decrease in surrounding scattering objects, the Rice factor K can be further defined as:

[0114]

[0115] Here, a1>0 and b1>0 represent environmental factors, the specific values ​​of which are related to factors such as the electromagnetic spectrum used and the characteristics of the surrounding environment. Let θ be the elevation angle between the ground user and the drone. Given a certain ground user's location, as the drone's altitude increases, the elevation angle θ between the user and the drone monotonically increases.

[0116] Furthermore, as the altitude of drones increases, the ground-to-air transmission link gradually approaches the free-space link transmission model, and the large-scale path transmission loss factor α... v Gradually decrease, that is

[0117]

[0118] Among them, a2<0, a3>0, b2>0, and b3>0 are environmental factors, and their specific values ​​are related to factors such as the electromagnetic spectrum used and the characteristics of the surrounding environment.

[0119] (III) Task and Unloading Model

[0120] A user's computational task is represented by a triple [I,C,T], where I is the number of bits required for the task, C is the number of CPU cycles required, and T is the maximum tolerable latency. Since each computational task is indivisible, a probabilistic full offload model is adopted. The user randomly offloads the entire task to a drone with an offload probability β, or to a ground base station with a probability of 1-β. The offload probability β quantifies the user's preference for accessing edge computing services via a drone. By optimizing the user's offload probability β, the resource utilization efficiency of both the drone and the ground base station can be coordinated.

[0121] (4) Communication model

[0122] To avoid co-channel interference caused by offloading computing tasks from different users, all users adopt an orthogonal multiple access strategy to access the drone or ground base station. (The rest of the text appears to be incomplete and requires further context.) u ,y u The user of [0] decides to send the data to the location V = [x] v ,y v For the drone unloading task of [H], the signal-to-noise ratio received by the drone is:

[0123]

[0124] Where Ω represents the small-scale fading power gain of the ground-to-air transmission link, and P t This represents the user's transmit power, ||·|| represents the Euclidean distance, and α is the transmit power. v σ represents the large-scale path transmission loss factor of the ground-to-air transmission link. 2 It is the additive white Gaussian noise power at the drone receiver.

[0125] If user U = [x u ,y u ,0] Select access ground base station B = [x b ,y b The signal-to-noise ratio received by the ground base station is: [0]

[0126]

[0127] Where g represents the small-scale fading power gain of the ground-to-ground transmission link, α b This represents the large-scale path transmission loss factor of a ground-to-ground transmission link.

[0128] (V) Computational Model

[0129] Edge servers typically employ parallel computing for rapid response. Specifically, through virtualization technology, an edge server can generate multiple logically isolated virtual machines (VMs), each handling a single computational task, thus simultaneously processing multiple computational tasks offloaded by different users. However, because multiple VMs generated by the same edge server share the same underlying physical resources, I / O interference occurs between them, and the actual usable computational frequency of each VM decreases as the number of VMs generated by the edge server increases. For running M... s The computation frequency expression for an edge server with 1 virtual machine is:

[0130]

[0131] Where s∈{v,b} is used to identify the edge server located at the UAV (denoted as v) or at the ground base station (denoted as b), f s (M s ) to run M simultaneously on edge servers s The computing frequency (CPU cycles / second) of each virtual machine when there are multiple virtual machines, f s,0 d represents the computing frequency available to a virtual machine when the server is running only a single virtual machine. s >0 represents a performance degradation factor that depends on factors such as the operating system and processor hardware configuration. Given a user computation task requiring C CPUs, and a computation latency of...

[0132]

[0133] Among them, T s,0 =C / f s,0 This indicates the computational latency when the edge server is running only one virtual machine.

[0134] After the edge server completes the task computation and processing, it immediately sends the result back to the user. This application considers mobile edge computing applications with a small number of bits in the computation result, such as video surveillance and other IoT applications. In this case, the result distribution latency is much smaller than the task offloading and transmission latency, so this application ignores the result distribution latency.

[0135] II. Service Reliability Analysis

[0136] Since user spatial locations follow a uniform Poisson process, their distribution is stationary and isotropic. Therefore, any user within the service coverage area can be selected as a typical user for performance analysis, and the analysis results obtained at the typical user are the same as the average service performance within the service coverage area. This application first analyzes the communication reliability and computational reliability of the typical user, and then derives the probability of successful edge computing service to evaluate the overall service reliability of the system.

[0137] (I) Communication Reliability Analysis

[0138] Users at a fixed transmission rate Unload the computing task, where W is the transmission bandwidth. Transmit the signal-to-noise ratio (SNR) value to the target. Successful transmission requires a received SNR of not less than [value missing]. Otherwise, the task unloading will result in a transmission interruption. Therefore, the probability of successful communication is defined as:

[0139]

[0140] Where s∈{v,b} is used to identify the received signal-to-noise ratio at the UAV (denoted as v) or the ground base station (denoted as b). This is the signal-to-noise ratio threshold for successful reception of user computing tasks by drones or ground base stations. R represents the area covered by the drone service, and R is the radius of the drone service coverage area. Indicates the service coverage area Take the average expectation.

[0141] To meet communication transmission delay requirements, the signal-to-noise ratio threshold is... The value of must satisfy the following constraints:

[0142]

[0143] Where τ is the duration of the user's computation task offloading transmission, and $I$ is the number of task bits that the user offloads.

[0144] 1. Derivation of the successful communication probability of offloading to the drone

[0145] Drones are typically equipped with directional antennas for enhancement, so they only connect at elevation angles above a threshold. Users can access the drone service. Therefore, the drone's coverage radius satisfies the spatial constraints:

[0146]

[0147] D v Let D be the distance between the user and the ground projection of the drone. Since the user's position follows a Poisson point process, D v The probability density function is:

[0148]

[0149] Therefore, when a user decides to offload a task to a drone, according to D v Given the probability distribution of the ground-to-air transmission link, the probability of successful communication at the UAV is:

[0150]

[0151] Where Q(·,·) is the Marcum Q function. In the formula, the Rice factor K and the path loss exponent α... v Both are functions of user location and drone altitude.

[0152] 2. Derivation of the successful communication probability when offloading to a ground base station

[0153] Let l represent the horizontal distance between the ground base station and the UAV's ground projection point within each time slot, and let D be the distance between them. b Let D be the distance from the user to the fixed ground base station. To analyze the probability of successful communication at the ground base station, we need to first derive D. b The distribution of D can be obtained. b The cumulative probability distribution function is

[0154] (1) When l≥R, D b The cumulative probability distribution function is

[0155]

[0156] (2) When l < R, D b The cumulative probability distribution function is

[0157]

[0158] in,

[0159]

[0160] Through calculation D can be obtained directly b The probability density function. At this point, the probability of successful communication when a user accesses a ground base station can be obtained as:

[0161]

[0162] Wherein, the lower limit of integration D l Defined as

[0163]

[0164] (II) Computational Reliability Analysis

[0165] To evaluate system computational performance, the probability of successful computation is used to describe the system's ability to respond to the computational demands of the currently carried task within a given latency requirement T. Its performance metric is defined as follows:

[0166]

[0167] Where s∈{v,b} is used to identify the computation latency at the UAV (denoted as v) or the ground base station (denoted as b), and T is the maximum tolerable latency requirement for the user's computation task.

[0168] As defined, the probability of successful computation depends on the number of virtual machines simultaneously generated by the mobile edge computing server. Specifically, as the number of virtual machines increases, the parallel computing gain improves, and more users can obtain immediate responses on the same server. However, because multi-user computing tasks share the same underlying physical resources on the same edge server, severe I / O interference occurs, leading to a deterioration in the actual achievable computational latency performance of a single computing task. To address computation timeouts caused by I / O interference, each edge server, based on its computational latency requirements... The maximum number of users that can be supported is determined. Newly arrived user computing tasks are discarded only when the actual number of users received exceeds the maximum user threshold, ensuring that the currently executing user computing tasks can be successfully completed.

[0169] After adopting the above-mentioned mobile edge computing server task control strategy, when a user accesses a drone to obtain edge computing services, the probability of successful computation is:

[0170]

[0171] in, It is determined by the computational delay requirement The maximum number of virtual machines that a given edge server on a drone can support. It is the floor operator, (·). + =max{0,·}.

[0172] When a user chooses to access a terrestrial base station to obtain services, the probability of success is calculated as follows:

[0173]

[0174] in, It is determined by the computational delay requirement The maximum number of virtual machines that an edge server mounted on a given ground base station can support.

[0175] (III) Probability of Successful Edge Computing Services

[0176] Based on the combined analysis of communication reliability and computational reliability, a successful edge computing service probability is introduced to measure the overall service performance of the considered air-to-ground integrated mobile edge computing network. The successful edge computing service probability estimates the average performance of users within the service coverage area and can approximate the proportion of successful service areas to the entire UAV coverage area, defined as:

[0177]

[0178] Based on the law of total probability, the expression for calculating the probability of success margin can be obtained as follows:

[0179]

[0180] Where β is the probability of a user unloading the drone. This represents the probability of successful communication for the drone. Calculate the probability of success for the drone. This represents the probability of successful communication at the ground base station. The probability of success is calculated at the ground base station.

[0181] III. Optimal UAV Altitude and Mission Unloading Probability Design

[0182] Using the obtained success probability of edge computing services as a performance constraint, this application first constructs a reliable service coverage maximization problem. Then, by exploring the properties of the constructed problem, a method is proposed to find the optimal UAV altitude H and mission offloading probability β. The convergence, optimality and complexity of the proposed method are analyzed.

[0183] (I) Problem Formulation for Maximizing Service Coverage

[0184] Given the user spatial distribution density λ, and using the derived successful edge computing service probability to specify the target service quality that users need to achieve, a service coverage maximization problem model is constructed for each time slot as follows:

[0185]

[0186] stp s (β,H)≥1-ε

[0187] 0≤β≤1

[0188] H>0

[0189] The objective function maximizes the service coverage area of ​​a single time slot UAV, 1-ε is the minimum lower bound requirement for the probability of successful edge computing service within the UAV service coverage area, the user computing task unloading probability β is between 0 and 1, and the UAV altitude H is non-negative.

[0190] (II) Algorithm Design

[0191] To solve the service reliability constraint expression p s The complex coupling problem between UAV altitude H and user computation task offloading probability β in (β,H) is first analyzed. The impact of UAV altitude H and user computation task offloading probability β on service reliability is analyzed.

[0192] For the probability of successful communication, since its expression is independent of the user's calculated task unloading probability, it is only necessary to analyze the impact of the drone's altitude on the probability of successful communication. This is achieved by solving for the probability of successful communication. Regarding the first derivative information of the drone's altitude H, we can obtain the following: as the drone's altitude increases, the probability of successful communication at the drone decreases monotonically, while the probability of successful communication at the ground base station first increases and then decreases.

[0193] To calculate the probability of success, analyze the probability of successful communication. Based on the first derivative information of the UAV altitude H and the user computation task offloading probability β, we can conclude that: as the UAV altitude increases, the probability of successful computation at both the UAV and the ground base station monotonically decreases. Conversely, as the user computation task migration probability increases, the probability of successful computation at the UAV monotonically decreases, while the probability of successful computation at the ground base station monotonically increases.

[0194] Considering the impact of drone altitude on the probabilities of successful communication and successful computation, proof by contradiction shows that when the constructed reliable service coverage optimization problem reaches its optimal value, the constraint on the probability of successful edge computation necessarily holds in equation form. Furthermore, for a given value of the probability of successful edge computation, as the user unloading probability monotonically decreases from 1 to 0, the corresponding maximum drone flight altitude first increases and then decreases. Utilizing this unimodal characteristic, a globally optimal algorithm based on the golden section algorithm can be constructed to obtain the optimal drone altitude and the probability of user computation migration and unloading. Theoretical analysis proves that the proposed scheme converges to the globally optimal solution with logarithmic complexity.

[0195] remember When calculating the task unloading probability β for a given user, the constraint condition p must be satisfied. s A relatively large UAV flight altitude value of (β,H)=1-ε. Due to the difficulty in obtaining... Because the expression is closed, the golden section algorithm cannot be directly used to solve the problem. Therefore, this project proposes a low-complexity optimization method based on golden section search, the specific steps of which are as follows:

[0196] First, using β=1 and H v Initialize the algorithm and compare it with the benchmark H. th Initialize to H v H v Satisfy constraint p s(1,H v )=1-ε.

[0197] In each subsequent iterative search, the golden ratio principle is used to determine the access probabilities of the two users to be tested, and the corresponding values ​​are calculated for each of these two access probabilities.

[0198] Let the access probabilities of the two users being tested be β1 and β2, where β1 < β2. In each round of the search, there are a total of four cases and corresponding update strategies:

[0199] Scenario 1: When and When both β1 and β2 have real solutions, they are both feasible solutions to the original reliable service coverage maximization problem. In this case, if or Exceeding the benchmark H th Then use Update the comparison benchmark H th Meanwhile, when the constraints are met If the search interval [0, β1] is reached, then the search interval [β2, 1] is deleted directly; otherwise, the search interval [β2, 1] is deleted. as well as All are lower than the benchmark H th H th The algorithm remains unchanged, and the search interval [0, β2] is directly deleted. Note that, due to the unimodal property, the deleted interval cannot contain the optimal solution, thus not compromising the algorithm's optimality.

[0200] Scenario 2: When It does not exist. If they exist, then β1 is an infeasible solution and β2 is a feasible solution. If Then use Update the comparison benchmark H th And delete the infeasible interval [0, β1]. Otherwise, if The search interval [0, β2] is directly deleted, and the benchmark H is not updated. th .

[0201] Scenario 3: When and When neither exists, then β1 and β2 are both infeasible solutions. Therefore, the benchmark H cannot be updated. th The search interval [0, β2] is directly deleted.

[0202] Scenario 4: When exist, When β1 does not exist, β2 is a feasible solution and β1 is an infeasible solution. This situation is impossible in the constructed problem due to the unimodal characteristic.

[0203] Repeat the above steps until the size of the remaining search interval is lower than the convergence accuracy threshold Δ.

[0204] (III) Algorithm Analysis

[0205] Convergence analysis: The proposed algorithm continuously narrows the search range of the user computation task unloading probability β during the search process. Since the value of the user computation task unloading probability β is restricted to a finite range β∈[0,1], the proposed algorithm can eventually converge to a sufficiently small β search range, which is lower than the accuracy threshold Δ.

[0206] Optimality Analysis: Since the optimal user computation task unloading probability β exhibits a unimodal relationship with the UAV flight altitude, and the interval deletion process ensures that the deleted range does not contain the optimal solution, combined with convergence analysis, the proposed method can ultimately converge to a unique optimal solution.

[0207] Complexity Analysis: The computational complexity of the proposed reliable service coverage enhancement mainly comes from the deletion operation. Specifically, the search range is continuously narrowed by a factor of 1-κ or κ to reduce the range of values ​​for the user's computational task offloading probability β, where... Therefore, given the target accuracy requirement Δ, the maximum number of iterations is... and Between these, the overall computational complexity required by the algorithm is at least O(n log n). At most

[0208] like Figure 2 As shown, by comparing the service coverage performance of the proposed air-to-ground cooperation scheme with that of a ground base station computing platform only (denoted as "BS-only") and an airborne computing platform only (denoted as "UAV-only"), it can be seen that the proposed air-to-ground cooperation scheme can bring significant performance gains. Specifically, the airborne computing platform only scheme, since it does not interact with the ground base station, has service coverage performance independent of the distance between the UAV's ground projection and the ground base station. However, with a ground base station computing platform only scheme, the reliable service coverage range gradually decreases as the service coverage center moves further away from the base station. In contrast, in the proposed scheme, although the overall service coverage decreases due to the reduced availability of ground computing resources as the coverage center moves further away from the base station, the reduction in service coverage can be mitigated by efficiently utilizing the UAV's airborne computing platform.

[0209] like Figure 3As shown, the service coverage performance of the proposed scheme is compared with that of the traditional altitude-free optimization scheme. In the scheme, the service coverage area at a fixed altitude is determined by an exhaustive search method to find the optimal user access probability, thus determining the maximum coverage area that satisfies service reliability constraints at the current flight altitude. It can be seen that the proposed scheme can flexibly adjust the UAV's flight altitude based on the availability of ground resources, effectively controlling co-channel interference during the task offloading and transmission phase and I / O interference during the computation and processing phase, thereby improving service coverage. Simulation results further verify that utilizing the three-dimensional spatial flexibility of the UAV platform can bring more significant service performance gains.

[0210] It should be understood that the specific order or hierarchy of steps in the disclosed process is an example of an exemplary method. Based on design preferences, it should be understood that the specific order or hierarchy of steps in the process may be rearranged without departing from the scope of this disclosure. The appended method claims provide elements of various steps in an exemplary order and are not intended to limit the scope to the specific order or hierarchy described.

[0211] In the above detailed description, various features are combined together in a single embodiment to simplify this disclosure. This approach to disclosure should not be construed as reflecting an intention that embodiments of the claimed subject matter require more features than are explicitly stated in each claim. Rather, as reflected in the appended claims, the invention is presented with fewer features than all of the features of the single disclosed embodiment. Therefore, the appended claims are hereby explicitly incorporated into the detailed description, wherein each claim stands alone as a preferred embodiment of the invention.

[0212] The disclosed embodiments have been described above to enable any person skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be applied to other embodiments without departing from the spirit and scope of this disclosure. Therefore, this disclosure is not limited to the embodiments given herein, but is consistent with the broadest scope of the principles and novel features disclosed herein.

[0213] The foregoing description includes examples of one or more embodiments. It is certainly impossible to describe all possible combinations of components or methods in order to describe the above embodiments, but those skilled in the art will recognize that further combinations and arrangements of the various embodiments are possible. Therefore, the embodiments described in this application are intended to cover all such changes, modifications, and variations that fall within the scope of the appended claims. Furthermore, the term "comprising" as used in the specification or claims is interpreted in a manner similar to the term "including," as interpreted when used as a conjunction in the claims. Additionally, the use of any term "or" in the specification of the claims is intended to mean "non-exclusive or."

[0214] Those skilled in the art will also understand that the various illustrative logical blocks, units, and steps listed in the embodiments of the present invention can be implemented by electronic hardware, computer software, or a combination of both. To clearly demonstrate the interchangeability of hardware and software, the functions of the various illustrative components, units, and steps described above have been generally described. Whether such functionality is implemented through hardware or software depends on the specific application and the overall system design requirements. Those skilled in the art can implement the described functions using various methods for each specific application, but such implementation should not be construed as exceeding the scope of protection of the embodiments of the present invention.

[0215] The various illustrative logic blocks or units described in the embodiments of this invention can be implemented or operate the described functions using a general-purpose processor, digital signal processor, application-specific integrated circuit (ASIC), field-programmable gate array or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof. The general-purpose processor can be a microprocessor; alternatively, it can be any conventional processor, controller, microcontroller, or state machine. The processor can also be implemented using a combination of computing devices, such as a digital signal processor and a microprocessor, multiple microprocessors, one or more microprocessors combined with a digital signal processor core, or any other similar configuration.

[0216] The steps of the methods or algorithms described in the embodiments of this invention can be directly embedded in hardware, a software module executed by a processor, or a combination of both. The software module can be stored in RAM, flash memory, ROM, EPROM, EEPROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium in the art. Exemplarily, the storage medium can be connected to the processor so that the processor can read information from and write information to the storage medium. Optionally, the storage medium can also be integrated into the processor. The processor and storage medium can be housed in an ASIC, which can be housed in a user terminal. Optionally, the processor and storage medium can also be housed in different components of the user terminal.

[0217] In one or more exemplary designs, the functions described in the embodiments of the present invention can be implemented in hardware, software, firmware, or any combination of these three. If implemented in software, these functions can be stored on a computer-readable medium or transmitted on a computer-readable medium in the form of one or more instructions or code. Computer-readable media include computer storage media and communication media that facilitate the transfer of computer programs from one place to another. Storage media can be any available media that can be accessed by a general-purpose or special-purpose computer. For example, such computer-readable media can include, but is not limited to, RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to carry or store program code in the form of instructions or data structures and other forms that can be read by a general-purpose or special-purpose computer, or a general-purpose or special-purpose processor. Furthermore, any connection can be suitably defined as a computer-readable medium, for example, if the software is transmitted from a website, server or other remote resource via a coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL) or wirelessly, such as infrared, wireless and microwave, it is also included in the defined computer-readable medium. The disks and discs mentioned include compressed disks, laser discs, optical discs, DVDs, floppy disks, and Blu-ray discs. Disks typically copy data magnetically, while disks typically copy data optically using lasers. Combinations of the above can also be contained in computer-readable media.

[0218] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for enhancing reliable service coverage for unmanned aerial vehicles (UAVs), characterized in that, The method includes the following steps: The probability of successfully introducing edge computing services is used to measure the reliability of system services. Construct a model for maximizing drone service coverage based on the probability of successful edge computing services; Analyze the coupling variables in the model of the drone service coverage maximization problem; Utilize the obtained analysis results to optimize the coverage of drone services.

2. The method for enhancing reliable service coverage for unmanned aerial vehicles (UAVs) according to claim 1, characterized in that, The probability of a successful edge computing service is: Where β is the probability of a user unloading the drone. This represents the probability of successful communication for the drone. Calculate the probability of success for the drone. This represents the probability of successful communication at the ground base station. The probability of success is calculated at the ground base station.

3. The method for enhancing reliable service coverage for unmanned aerial vehicles (UAVs) according to claim 1, characterized in that, The model for maximizing drone service coverage is as follows: stp s (β,H)≥1-ε 0≤β≤1 H>0 Where 1-ε is the minimum lower limit requirement for the probability of successful edge computing service within the drone service coverage area, β is the probability of user computing task unloading, and the value of β is between 0 and 1; H is the drone altitude, and the value of H is non-negative.

4. The method for enhancing reliable service coverage for unmanned aerial vehicles (UAVs) according to claim 1, characterized in that, The analysis of the coupling variables in the model for maximizing drone service coverage includes: This study investigates the implicit features in the model for maximizing unmanned aerial vehicle (UAV) service coverage and analyzes the influence between coupling variables. The reliable service coverage optimization problem is solved by utilizing the influence between the coupled variables.

5. The method for enhancing reliable service coverage for unmanned aerial vehicles (UAVs) according to claim 4, characterized in that, The coupling variables include: the UAV's spatial location and the probability of the user's computational task being unloaded.

Images