Cloud edge resource scheduling scheme generation method and device, equipment and storage medium

Abstract

The application provides a cloud-edge-end resource scheduling scheme generation method and device, equipment and a storage medium, and belongs to the technical field of edge computing. The method comprises the following steps: based on a Rayleigh fading model, a first transmission rate of a terminal device unloading a task to an edge server and a second transmission rate of the terminal device unloading the task to a cloud server are calculated; based on the computing power of the terminal device, the computing density of the computing task, the first transmission rate and the second transmission rate, a first processing time delay required by the edge server to process the task, a second processing time delay required by the cloud server to process the task, and a third processing time delay required by the terminal device to process the task are determined, and then an optimization problem is generated; a differential evolution algorithm is used to solve the optimization problem, and a cloud-edge-end resource scheduling scheme is obtained. The cloud-edge-end resource scheduling scheme generated by the application can reduce the task processing time delay of cloud-edge-end collaborative computing in a high-speed moving scene and improve the task processing efficiency.

Description

Technical Field

[0001] This invention relates to the field of edge computing technology, and in particular to a method, apparatus, device, and storage medium for generating cloud-edge-device resource scheduling schemes. Background Technology

[0002] With the development of intelligent services on high-speed trains, more data needs to be analyzed and processed. However, the computing power of the onboard terminals is insufficient to complete the processing tasks within the specified time limit. Therefore, it is necessary to offload some computing tasks to edge servers and cloud servers to achieve lower processing latency. The transmission rate between the computing terminals and edge servers / cloud servers is limited; if all computing tasks are offloaded, there will be high transmission latency. Therefore, it is necessary to rationally design the task offloading ratio and communication and computing resource management schemes to achieve the minimum processing latency.

[0003] Most current edge computing systems are geared towards low-speed mobile or stationary devices, failing to fully consider the impact of high-speed device movement on task offloading rates. Inaccurate modeling of the transmission rate of the high-speed train's vehicle-to-ground channel results in the inability to obtain precise task offloading and resource allocation strategies, leading to poor adaptability to high-speed mobile scenarios.

[0004] Therefore, there is an urgent need for a method to generate cloud-edge-device resource scheduling schemes for high-speed moving trains. Summary of the Invention

[0005] This invention provides a method, apparatus, device, and storage medium for generating cloud-edge-device resource scheduling schemes, in order to solve the problem that existing edge computing systems do not fully consider the impact of high-speed device movement on task unloading rate, resulting in the inability to obtain accurate task unloading and resource allocation strategies.

[0006] This invention provides a method for generating cloud-edge-device resource scheduling schemes, comprising the following steps.

[0007] Based on the Rayleigh fading model, a first transmission rate and a second transmission rate are calculated. The first transmission rate is the transmission rate at which the terminal device offloads the task to the edge server, and the second transmission rate is the transmission rate at which the terminal device offloads the task to the cloud server. Based on the computing power of the terminal device, the computing density of the computing task, the first transmission rate and the second transmission rate, a first processing delay, a second processing delay and a third processing delay are determined. The first processing delay is the time required for the edge server to process the task, the second processing delay is the time required for the cloud server to process the task, and the third processing delay is the time required for the terminal device to process the task. An optimization problem is generated based on the first processing delay, the second processing delay, and the third processing delay; The differential evolution algorithm is used to solve the optimization problem, resulting in a cloud-edge-device resource scheduling scheme.

[0008] According to the cloud-edge-device resource scheduling scheme generation method provided by the present invention, the first transmission rate is calculated based on the following formula: ; in, Indicates the first transmission rate. This indicates the transmission bandwidth for the terminal device to offload tasks to the edge server. This represents the fading coefficient of the fast fading in the vehicle-to-ground channel between the terminal device and the edge server. This represents the fading coefficient of the slow fading in the vehicle-to-ground channel between the terminal device and the edge server. This indicates the transmit power of the edge server. Represents the noise power spectral density. express The conditional probability density function.

[0009] According to the cloud-edge-device resource scheduling scheme generation method provided by the present invention, the second transmission rate is calculated based on the following formula: ; in, Indicates the second transmission rate. This indicates the transmission bandwidth for the terminal device to offload tasks to the cloud server. This represents the fading coefficient of the fast fading in the vehicle-to-ground channel between the terminal device and the edge server. This represents the fading coefficient of the slow fading in the vehicle-to-ground channel between the terminal device and the edge server. This indicates the transmit power of the edge server. Represents the noise power spectral density. express The conditional probability density function.

[0010] According to the cloud-edge-device resource scheduling scheme generation method provided by the present invention, the Rayleigh fading model in... The formula for calculating the conditional probability density function is as follows: ; in, , ; Indicates the delay of channel estimation. This indicates expired channel state information. Represents the time-domain correlation coefficient. Denotes the zeroth-order Bessel function of the first kind. Indicates the channel estimation error. This represents the first kind of modified zeroth-order Bessel function. This indicates the maximum Doppler frequency shift of the vehicle-to-ground channel.

[0011] According to the cloud-edge-device resource scheduling scheme generation method provided by the present invention, the objective function of the differential evolution algorithm is as follows: ; in, , ; Describe the objective function. This represents the total processing time of the computational task. N Indicates the number of computational tasks. J This indicates the number of constraints in the optimization problem. Represents the penalty function. This represents the penalty factor for each constraint. Indicates the first j One constraint condition.

[0012] The present invention also provides a cloud-edge-device resource scheduling scheme generation device, comprising the following modules: The rate calculation module is used to: calculate a first transmission rate and a second transmission rate based on the Rayleigh fading model, wherein the first transmission rate is the transmission rate at which the terminal device offloads the task to the edge server, and the second transmission rate is the transmission rate at which the terminal device offloads the task to the cloud server. The latency calculation module is used to: determine a first processing latency, a second processing latency, and a third processing latency based on the computing power of the terminal device, the computing density of the computing task, the first transmission rate, and the second transmission rate, wherein the first processing latency is the time required for the edge server to process the task, the second processing latency is the time required for the cloud server to process the task, and the third processing latency is the time required for the terminal device to process the task. The problem generation module is used to generate optimization problems based on the first processing delay, the second processing delay, and the third processing delay. The problem-solving module is used to: solve the optimization problem using the differential evolution algorithm to obtain a cloud-edge-device resource scheduling scheme.

[0013] The present invention also provides a cloud-edge-device collaborative computing system, including a cloud server, an edge server and multiple terminal devices, characterized in that the cloud-edge-device collaborative computing system uses a cloud-edge-device resource scheduling scheme generated by any of the cloud-edge-device resource scheduling scheme generation methods described above to process tasks.

[0014] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the cloud-edge-device resource scheduling scheme generation method described above.

[0015] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the cloud-edge-device resource scheduling scheme generation method as described above.

[0016] The present invention also provides a computer program product, including a computer program, which, when executed by a processor, implements the cloud-edge-device resource scheduling scheme generation method described above.

[0017] The present invention provides a cloud-edge-device resource scheduling scheme generation method, apparatus, device, and storage medium. Based on the Rayleigh fading model, it calculates a first transmission rate and a second transmission rate. The first transmission rate is the transmission rate at which the terminal device offloads tasks to the edge server, and the second transmission rate is the transmission rate at which the terminal device offloads tasks to the cloud server. Based on the computing power of the terminal device, the computing density of the computing tasks, the first transmission rate, and the second transmission rate, it determines a first processing delay, a second processing delay, and a third processing delay. The first processing delay is the time required for the edge server to process the task, the second processing delay is the time required for the cloud server to process the task, and the third processing delay is the time required for the terminal device to process the task. Based on the first processing delay, the second processing delay, and the third processing delay, it generates an optimization problem. The differential evolution algorithm is used to solve the optimization problem to obtain the cloud-edge-device resource scheduling scheme. This solution addresses the high-speed movement characteristics of trains by modeling the small-scale fading of the vehicle-to-ground channel using the Rayleigh fading model. This allows for the accurate calculation of the vehicle-to-ground channel communication transmission rate, which in turn generates an optimization problem. Solving this optimization problem yields a cloud-edge-device resource scheduling scheme that is well-suited for high-speed movement scenarios. This reduces the task processing latency of cloud-edge-device collaborative computing in high-speed movement scenarios and improves task processing efficiency. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0019] Figure 1This is a flowchart illustrating the cloud-edge-device resource scheduling scheme generation method provided by the present invention.

[0020] Figure 2 This is the average task latency performance curve under the influence of terminal device bandwidth provided by the present invention.

[0021] Figure 3 This is the average task latency performance curve under the influence of the number of terminal devices provided by the present invention.

[0022] Figure 4 This is a schematic diagram of the cloud-edge-device collaborative computing system provided by the present invention.

[0023] Figure 5 This is a schematic diagram of the cloud-edge-device resource scheduling scheme generation device provided by the present invention.

[0024] Figure 6 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0026] It should be noted that in the description of the embodiments of the present invention, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element. The terms "upper," "lower," etc., indicating orientation or positional relationships based on the orientation or positional relationships shown in the accompanying drawings, are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the present invention. Unless otherwise expressly specified and limited, the terms "installed," "connected," and "linked" should be interpreted broadly, for example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two elements. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0027] The terms "first," "second," etc., used in this invention are used to distinguish similar objects, not to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the invention can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first," "second," etc., are generally of the same class, without limiting the number of objects; for example, a first object can be one or more. Furthermore, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0028] The following is combined with Figures 1-6 This invention describes the cloud-edge-device resource scheduling scheme generation method, apparatus, equipment, and storage medium provided in embodiments of the present invention.

[0029] Figure 4 This is a schematic diagram of the cloud-edge-device collaborative computing system provided by the present invention, as shown below. Figure 4As shown, the cloud-edge-device collaborative computing system includes a cloud server, an edge server, and multiple terminal devices. The cloud-edge-device collaborative computing system processes tasks using a cloud-edge-device resource scheduling scheme generated by the cloud-edge-device resource scheduling scheme generation method described in any of the following embodiments. That is, after generating the cloud-edge-device resource scheduling scheme using the cloud-edge-device resource scheduling scheme generation method described in any of the following embodiments, the cloud server, the edge server, and multiple terminal devices transmit data according to the parameters in the generated cloud-edge-device resource scheduling scheme and collaboratively complete the task processing.

[0030] Figure 4 An exemplary architecture diagram is shown. For descriptive purposes, the architecture depicted is merely an example of a suitable environment and does not imply any limitation on the scope or functionality of the invention.

[0031] like Figure 4 As shown in this embodiment of the invention, the cloud server and edge server are deployed on the ground side and connected via a wired network. A computing terminal (terminal device) and A sensor array is deployed at In different carriages, each sensor group is composed of Composed of several sensors, making Indicates the relationship with the first A sensor group for wireless communication with a computing terminal. Reception complete. After processing the data from all sensors, the first... The computing terminal can generate the first... Each computing task requires offloading it to an edge server or cloud server with abundant computing resources via a vehicle-to-ground channel to reduce latency due to the relatively low computing power of the computing terminal.

[0032] The cloud-edge-device collaborative computing system provided in this embodiment of the invention completes the functional construction of each entity module (sensor group, computing terminal, edge server, cloud server) in the high-speed train cloud-edge-device collaborative computing system, and derives the generation and unloading process of computing tasks.

[0033] The cloud-edge-device resource scheduling scheme generation method provided in the embodiments of the present invention is described below. The cloud-edge-device resource scheduling scheme generation method described below can be referred to in correspondence with the cloud-edge-device collaborative computing system described above.

[0034] Figure 1 This is a flowchart illustrating the cloud-edge-device resource scheduling scheme generation method provided by the present invention, as follows: Figure 1 As shown, the method includes the following: S110, based on the Rayleigh fading model, calculates the first transmission rate and the second transmission rate; S120, based on the computing power of the terminal device, the computing density of the computing task, the first transmission rate and the second transmission rate, determine the first processing delay, the second processing delay and the third processing delay; S130, an optimization problem is generated based on the first processing delay, the second processing delay, and the third processing delay; S140, The differential evolution algorithm is used to solve the optimization problem to obtain a cloud-edge-device resource scheduling scheme.

[0035] It should be noted that the execution subject of the cloud-edge-device resource scheduling scheme generation method provided in the embodiments of the present invention can be a server or computer device, such as a tablet computer, a laptop computer, a handheld computer, an ultra-mobile personal computer (UMPC), a netbook, etc.

[0036] Furthermore, it should be understood that this cloud-edge-device resource scheduling scheme generation method can be used in any scenario where terminal devices are moving at high speed, and is not limited to high-speed trains.

[0037] In S110, the first transmission rate is the transmission rate at which the terminal device offloads the task to the edge server, and the second transmission rate is the transmission rate at which the terminal device offloads the task to the cloud server. Rayleigh fading is a small-scale fading phenomenon that describes the random variation in signal strength caused by multipath propagation. Small-scale fading in the vehicle-to-ground channel can be modeled as Rayleigh fading. In the first... In the wireless transmission channel between the terminal device and the edge server, the first The portion of a computing task offloaded to the edge server and the portion offloaded to the cloud server are transmitted simultaneously. The communication rate between the terminal devices in the carriage and the edge server can be calculated based on the conditional probability density function of Rayleigh fading.

[0038] In S120, the first processing latency is the time required for the edge server to process the task, i.e., the processing latency of the portion of data processed by the computing task offloaded to the edge server; the second processing latency is the time required for the cloud server to process the task, i.e., the processing latency of the portion of data processed by the computing task offloaded to the cloud server; and the third processing latency is the time required for the terminal device to process the task, i.e., the processing latency of the portion of data processed locally on the terminal device. The computational density of the computing task is the number of CPU cycles required to process each bit of data.

[0039] In an optional embodiment: No. The amount of data for each computation task is represented as follows: ; in, express The Middle The amount of data generated by each sensor This indicates the percentage change in the size of the aggregated data.

[0040] make and They represent the first The ratio of the number of computing tasks offloaded to edge servers and cloud servers, then the number of tasks offloaded to edge servers and cloud servers. The amount of data processed by a computing task at the edge server can be represented as: ; No. The amount of data computed by a computing task on a cloud server can be represented as: ; No. The amount of data processed by a computing task at the terminal device can be expressed as: .

[0041] The third processing latency (the processing latency of the portion of data computed locally on the terminal device) can then be expressed as: ; in, This represents the computing power of each terminal device. Representing the Computational density of a computing task; First processing delay (the first) The processing latency of a portion of the data computed by a computing task on the edge server can be expressed as: ; in, Assign the edge server to the first Computational resources for each computational task The first transmission rate; The second processing latency (the processing latency of the portion of data that is unloaded from the computing task and computed on the cloud server) can be expressed as: ; in, Assign cloud servers to the first Computational resources for each computational task This refers to the wired transmission rate between the edge server and the cloud server. This is the second transmission rate.

[0042] The cloud-edge-device resource scheduling scheme generation method provided in this embodiment of the invention derives the processing latency expressions of each computing task at the local, edge server, and cloud server by constructing a computing and task offloading model of the high-speed train cloud-edge-device collaborative computing system.

[0043] In S130, after determining the first processing delay, the second processing delay, and the third processing delay, an optimization problem is generated with the goal of minimizing the average delay of the computing tasks of all terminal devices.

[0044] The cloud-edge-device resource scheduling scheme generation method provided in this embodiment of the invention is based on the Rayleigh fading model. It calculates a first transmission rate and a second transmission rate, where the first transmission rate is the rate at which the terminal device offloads tasks to the edge server, and the second transmission rate is the rate at which the terminal device offloads tasks to the cloud server. Based on the computing power of the terminal device, the computing density of the computing tasks, the first transmission rate, and the second transmission rate, it determines a first processing delay, a second processing delay, and a third processing delay, where the first processing delay is the time required for the edge server to process the task, the second processing delay is the time required for the cloud server to process the task, and the third processing delay is the time required for the terminal device to process the task. Based on the first processing delay, the second processing delay, and the third processing delay, it generates an optimization problem. The differential evolution algorithm is used to solve the optimization problem to obtain the cloud-edge-device resource scheduling scheme. This solution addresses the high-speed movement characteristics of trains by modeling the small-scale fading of the vehicle-to-ground channel using the Rayleigh fading model. This allows for the accurate calculation of the vehicle-to-ground channel communication transmission rate, which in turn generates an optimization problem. Solving this optimization problem yields a cloud-edge-device resource scheduling scheme that is well-suited for high-speed movement scenarios. This reduces the task processing latency of cloud-edge-device collaborative computing in high-speed movement scenarios and improves task processing efficiency.

[0045] In an optional embodiment, the first transmission rate is calculated based on the following formula: ; in, Indicates the first transmission rate. This indicates the transmission bandwidth for the terminal device to offload tasks to the edge server. This represents the fading coefficient of the fast fading in the vehicle-to-ground channel between the terminal device and the edge server. This represents the fading coefficient of the slow fading in the vehicle-to-ground channel between the terminal device and the edge server. This indicates the transmit power of the edge server. Represents the noise power spectral density. express The conditional probability density function.

[0046] The second transmission rate is calculated based on the following formula: ; in, Indicates the second transmission rate. This indicates the transmission bandwidth for the terminal device to offload tasks to the cloud server. This represents the fading coefficient of the fast fading in the vehicle-to-ground channel between the terminal device and the edge server. This represents the fading coefficient of the slow fading in the vehicle-to-ground channel between the terminal device and the edge server. This indicates the transmit power of the edge server. Represents the noise power spectral density. express The conditional probability density function.

[0047] In the In the wireless transmission channel between the terminal device and the edge server, the first The portion of the computing task offloaded to the edge server and the portion offloaded to the cloud server are transmitted simultaneously, with the transmission bandwidth allocated to these two portions of the task being respectively... and Then its average transmission rate can be calculated using the above formula.

[0048] Furthermore, in the Rayleigh fading model The formula for calculating the conditional probability density function is as follows: ; in, , ; Indicates the delay of channel estimation. This indicates expired channel state information. Represents the time-domain correlation coefficient. Denotes the zeroth-order Bessel function of the first kind. Indicates the channel estimation error. This represents the first kind of modified zeroth-order Bessel function. This indicates the maximum Doppler frequency shift of the vehicle-to-ground channel.

[0049] Specifically, due to the high mobility of high-speed trains, the vehicle-to-ground channel is a fast time-varying channel, and its maximum Doppler frequency shift can be expressed as: ; in, This refers to the speed at which the high-speed train travels. At the speed of light, This indicates the carrier frequency used in the vehicle-to-ground channel.

[0050] Small-scale fading in the train-to-ground channel can be modeled as Rayleigh fading. Due to the high mobility of the train, the channel coherence time is shorter than the mission transmission time, and there is an unavoidable delay in the channel estimation process. Perfect Channel State Information (CSI) cannot be obtained; therefore, the temporal correlation of the channel is used to model it. Thus, the... n The small-scale fading of the vehicle-to-ground channel between a terminal device and an edge server can be represented as... ; in, , The delay for channel estimation, This indicates an expired CSI. The time-domain correlation coefficient, Denotes the zeroth-order Bessel function of the first kind. This indicates the channel estimation error.

[0051] when and When known, The conditional probability density function is: ; in This represents a modified zeroth-order Bessel function of the first kind.

[0052] Furthermore, the first Each terminal device simultaneously interacts with the sensor group It communicates with edge servers to complete data integration and offloading, thus facing the problem of communication bandwidth resource allocation, and must meet the following constraints: ; That is, the first Terminal devices and sensor groups The sum of the communication bandwidth allocated to the terminal device, the communication bandwidth allocated to the edge server, and the communication bandwidth allocated to the cloud server shall not exceed the communication bandwidth of the terminal device.

[0053] For edge servers, the bandwidth constraints are as follows: ; That is, the first The sum of the communication bandwidth allocated to the edge server and the communication bandwidth allocated to the cloud server by each terminal device shall not exceed the communication bandwidth of the edge server.

[0054] No. Each terminal device completes receiving middle After all the data from the sensors are transmitted in parallel, the data can be aggregated to generate the first... The computational task, therefore the first The generation time of a computational task can be defined as: ; in, ,express The Middle The sensor sends to the first The data transmission time of each terminal device. express The first in The sensor and the first The transmission rate between terminal devices.

[0055] Multiple sensor nodes in the first part are orthogonally connected to the second part. Each terminal device, therefore The first in The sensor and the first The transmission rate between terminal devices can be expressed as: ; in, express The first in The sensor and the first Channel gain between terminal devices This represents the bandwidth allocated to this channel. express The first in The transmit power of each sensor.

[0056] No. The total processing time for each computational task can be expressed as: .

[0057] To minimize the average latency of all computational tasks, the optimization problem is formulated as follows: .

[0058] The cloud-edge-device resource scheduling scheme generation method provided in this embodiment of the invention models the total task processing latency and the system's computing resource constraints based on expressions for the first processing latency, the second processing latency, and the third processing latency, thereby deriving the problem of minimizing the average task latency.

[0059] Further, optionally, the objective function of the differential evolution algorithm is as follows: ; in, , ; Describe the objective function. This represents the total processing time of the computational task. N Indicates the number of computational tasks. J This indicates the number of constraints in the optimization problem. Represents the penalty function. This represents the penalty factor for each constraint. Indicates the first j One constraint condition.

[0060] Specifically, when At that time, the first Generation time of each computational task It can reach the minimum value. Therefore, let The first sensor and the second sensor can be obtained. Communication bandwidth between terminal devices and the The sensor and the first Communication bandwidth between terminal devices The relationship between them: ; in, ; For Lambert-W functions, i.e., functions The inverse function of .

[0061] Similarly, the remaining bandwidth can also be expressed as The function. Therefore, when the first... The acquisition bandwidth of each terminal device Given the information, the optimal acquisition bandwidth allocation strategy can be obtained. Based on this, the number of variables in the optimization problem can be reduced.

[0062] Differential Evolution (DE) is a population-based optimization technique that finds the optimal solution to a function by simulating the processes of natural selection and genetic mutation. It works by maintaining a population of candidate solutions and iteratively updating these solutions to gradually approach the global optimum.

[0063] In this embodiment of the invention, each individual in the population, i.e., each solution vector It contains variables .make express The dimension of is the number of variables in the optimization problem. Let . and express The upper and lower limits of the value.

[0064] Considering the constraints The objective function for algorithm optimization is defined as: ; in , This indicates the number of constraints in the optimization problem. For the penalty function, It is for each constraint condition The designed penalty factor Defined as: ; In each iteration of the algorithm, the population is updated using a set of differential mutation and crossover operators, and new candidate solutions are generated by combining existing candidate solutions. Let... , , They represent the first time. The first iteration in the round There are three vectors: the original vector, the mutated vector, and the cross vector. The mutated vector... It can be represented as: ; in, For the variation factor with values ​​in the range [0,2], Given three randomly selected distinct integers, For the first Three different individuals in the round of iteration Population size.

[0065] The The element can be derived from the following formula: ; in, For crossover probability, Let be a random number that follows a normal distribution. It is in Random integers within the range, which can be guaranteed At least one element's value has changed.

[0066] The cloud-edge-device resource scheduling scheme generation method provided in this embodiment of the invention is based on the differential evolution algorithm. It solves the optimal task unloading and resource allocation scheme through a set objective function to obtain the cloud-edge-device resource scheduling scheme and reduce the latency of computing tasks in the cloud-edge-device collaborative system.

[0067] Without loss of generality, to verify that the cloud-edge-device resource scheduling scheme generation method provided by the present invention can reduce the task processing latency, the effect of the invention example is illustrated using the parameter configuration in Table 1.

[0068] Table 1 Simulation Parameter Table

[0069] like Figure 4 As shown, the cloud-edge-device collaborative computing system provided by this invention deploys a computing terminal in each carriage to integrate data sent by the sensor group and connect to the edge server. The computing tasks of the high-speed train are generated at each terminal device and can be offloaded to the edge server or cloud server through the vehicle-to-ground channel.

[0070] Under the simulation parameter settings shown in Table 1, three comparison schemes—local computing, edge computing, and cloud computing—were set up to compare their performance with the cloud-edge-device resource scheduling scheme provided by this invention. Through simulation calculations, the average task latency performance curves of the proposed scheme under the influence of terminal device bandwidth and the number of terminal devices were obtained.

[0071] Figure 2 This invention provides an average task latency performance curve under the influence of terminal device bandwidth, such as... Figure 2 As shown, when the terminal device bandwidth As bandwidth increases, terminal devices have more communication resources to collect sensor information and offload computational tasks. Therefore, in the proposed solution, local computing solution, edge computing solution, and cloud computing solution, the average task latency decreases with the increase of terminal device bandwidth. Furthermore, it can be observed that the proposed solution achieves a lower average task latency compared to the local computing solution, edge computing solution, and cloud computing solution.

[0072] Figure 3 This is the average task latency performance curve provided by the present invention under the influence of the number of terminal devices, such as... Figure 3 As shown, when the number of terminal devices As the number of devices increases, the average task latency in the present invention, local computing, edge computing, and cloud computing solutions all increase accordingly. This is because the computing and communication resources allocated to each task by edge servers and cloud servers decrease, leading to more tasks being computed locally. Furthermore, the present invention consistently achieves lower average task latency as the number of terminal devices increases.

[0073] In summary, the cloud-edge-device resource scheduling scheme generation method provided by this invention models the small-scale fading of the vehicle-to-ground channel based on the high-speed movement characteristics of trains, models the communication transmission rate of the vehicle-to-ground channel based on its conditional probability density function, designs corresponding task offloading and resource allocation schemes, realizes the optimal resource allocation scheme and task offloading algorithm, and efficiently reduces the average task processing latency.

[0074] The cloud-edge-device resource scheduling scheme generation device provided in the embodiments of the present invention is described below. The cloud-edge-device resource scheduling scheme generation device described below and the cloud-edge-device resource scheduling scheme generation method described above can be referred to in correspondence with each other.

[0075] Figure 5 This is a schematic diagram of the cloud-edge-device resource scheduling scheme generation device provided by the present invention, as shown below. Figure 5 As shown, the cloud-edge-device resource scheduling scheme generation device may include, but is not limited to; The rate calculation module 510 is used to: calculate a first transmission rate and a second transmission rate based on the Rayleigh fading model, wherein the first transmission rate is the transmission rate at which the terminal device offloads the task to the edge server, and the second transmission rate is the transmission rate at which the terminal device offloads the task to the cloud server. The latency calculation module 520 is used to: determine a first processing latency, a second processing latency, and a third processing latency based on the computing power of the terminal device, the computing density of the computing task, the first transmission rate, and the second transmission rate, wherein the first processing latency is the time required for the edge server to process the task, the second processing latency is the time required for the cloud server to process the task, and the third processing latency is the time required for the terminal device to process the task. Problem generation module 530 is used to: generate optimization problems based on the first processing delay, the second processing delay and the third processing delay; The problem-solving module 540 is used to: solve the optimization problem using the differential evolution algorithm to obtain a cloud-edge-device resource scheduling scheme.

[0076] It should be noted that the cloud-edge-device resource scheduling scheme generation device provided in this embodiment of the invention can execute the cloud-edge-device resource scheduling scheme generation method described in any of the above embodiments during specific operation, and this embodiment will not elaborate on this.

[0077] Figure 6 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 6As shown, the electronic device may include: a processor 610, a communications interface 620, a memory 630, and a communications bus 640, wherein the processor 610, the communications interface 620, and the memory 630 communicate with each other through the communications bus 640. The processor 610 can call logical instructions in the memory 630 to execute a cloud-edge-device resource scheduling scheme generation method. This method includes: calculating a first transmission rate and a second transmission rate based on the Rayleigh fading model, where the first transmission rate is the transmission rate at which the terminal device offloads tasks to the edge server, and the second transmission rate is the transmission rate at which the terminal device offloads tasks to the cloud server; determining a first processing delay, a second processing delay, and a third processing delay based on the computing power of the terminal device, the computing density of the computing tasks, the first transmission rate, and the second transmission rate, where the first processing delay is the time required for the edge server to process the task, the second processing delay is the time required for the cloud server to process the task, and the third processing delay is the time required for the terminal device to process the task; generating an optimization problem based on the first processing delay, the second processing delay, and the third processing delay; and solving the optimization problem using a differential evolution algorithm to obtain a cloud-edge-device resource scheduling scheme.

[0078] Furthermore, the logical instructions in the aforementioned memory 630 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0079] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the cloud-edge-device resource scheduling scheme generation method provided by the above methods. The method includes: calculating a first transmission rate and a second transmission rate based on the Rayleigh fading model, wherein the first transmission rate is the transmission rate at which the terminal device offloads tasks to the edge server, and the second transmission rate is the transmission rate at which the terminal device offloads tasks to the cloud server; determining a first processing delay, a second processing delay, and a third processing delay based on the computing power of the terminal device, the computing density of the computing tasks, the first transmission rate, and the second transmission rate, wherein the first processing delay is the time required for the edge server to process the task, the second processing delay is the time required for the cloud server to process the task, and the third processing delay is the time required for the terminal device to process the task; generating an optimization problem based on the first processing delay, the second processing delay, and the third processing delay; and solving the optimization problem using a differential evolution algorithm to obtain a cloud-edge-device resource scheduling scheme.

[0080] In another aspect, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon. When executed by a processor, the computer program implements a cloud-edge-device resource scheduling scheme generation method provided by the above methods. This method includes: calculating a first transmission rate and a second transmission rate based on a Rayleigh fading model, wherein the first transmission rate is the transmission rate at which a terminal device offloads a task to an edge server, and the second transmission rate is the transmission rate at which the terminal device offloads a task to a cloud server; determining a first processing delay, a second processing delay, and a third processing delay based on the computing power of the terminal device, the computing density of the computing task, the first transmission rate, and the second transmission rate, wherein the first processing delay is the time required for the edge server to process the task, the second processing delay is the time required for the cloud server to process the task, and the third processing delay is the time required for the terminal device to process the task; generating an optimization problem based on the first processing delay, the second processing delay, and the third processing delay; and solving the optimization problem using a differential evolution algorithm to obtain a cloud-edge-device resource scheduling scheme.

[0081] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0082] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0083] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for generating a cloud-edge-device resource scheduling scheme, characterized in that, Applied to high-speed trains, including: Based on the Rayleigh fading model, a first transmission rate and a second transmission rate are calculated. The first transmission rate is the transmission rate at which the terminal device offloads tasks to the edge server in the wireless transmission channel between the terminal device and the edge server. The second transmission rate is the transmission rate at which the terminal device offloads tasks to the cloud server in the wireless transmission channel between the terminal device and the edge server. Based on the computing power of the terminal device, the computing density of the computing task, the first transmission rate and the second transmission rate, a first processing delay, a second processing delay and a third processing delay are determined. The first processing delay is the time required for the edge server to process the task, the second processing delay is the time required for the cloud server to process the task, and the third processing delay is the time required for the terminal device to process the task. An optimization problem is generated based on the first processing delay, the second processing delay, and the third processing delay; The differential evolution algorithm is used to solve the optimization problem and obtain a cloud-edge-device resource scheduling scheme. The first transmission rate is calculated based on the following formula: ； in, Indicates the first transmission rate. This indicates the transmission bandwidth for the terminal device to offload tasks to the edge server. This represents the fading coefficient of the fast fading in the vehicle-to-ground channel between the terminal device and the edge server. This represents the fading coefficient of the slow fading in the vehicle-to-ground channel between the terminal device and the edge server. This indicates the transmit power of the terminal device. Represents the noise power spectral density. express The conditional probability density function; The second transmission rate is calculated based on the following formula: ； in, Indicates the second transmission rate. This indicates the transmission bandwidth for the terminal device to offload tasks to the cloud server. This represents the fading coefficient of the fast fading in the vehicle-to-ground channel between the terminal device and the edge server. This represents the fading coefficient of the slow fading in the vehicle-to-ground channel between the terminal device and the edge server. This indicates the transmit power of the terminal device. Represents the noise power spectral density. express The conditional probability density function; In the Rayleigh fading model The formula for calculating the conditional probability density function is as follows: ； in, , ; Indicates the delay of channel estimation. This indicates expired channel state information. Represents the time-domain correlation coefficient. Denotes the zeroth-order Bessel function of the first kind. Indicates the channel estimation error. This represents the first kind of modified zeroth-order Bessel function. This indicates the maximum Doppler frequency shift of the vehicle-to-ground channel.

2. The cloud-edge-device resource scheduling scheme generation method according to claim 1, characterized in that, The objective function of the differential evolution algorithm is as follows: ； in, , ; Describe the objective function. This represents the total processing time of the computational task. N Indicates the number of computational tasks. J This indicates the number of constraints in the optimization problem. Represents the penalty function. This represents the penalty factor for each constraint. Indicates the first j One constraint condition.

3. A cloud-edge-device resource scheduling scheme generation device, characterized in that, Applied to high-speed trains, the cloud-edge-device resource scheduling scheme generation method as described in claim 1 includes: The rate calculation module is used to: calculate a first transmission rate and a second transmission rate based on the Rayleigh fading model, wherein the first transmission rate is the transmission rate at which the terminal device offloads the task to the edge server, and the second transmission rate is the transmission rate at which the terminal device offloads the task to the cloud server. The latency calculation module is used to: determine a first processing latency, a second processing latency, and a third processing latency based on the computing power of the terminal device, the computing density of the computing task, the first transmission rate, and the second transmission rate, wherein the first processing latency is the time required for the edge server to process the task, the second processing latency is the time required for the cloud server to process the task, and the third processing latency is the time required for the terminal device to process the task. The problem generation module is used to generate optimization problems based on the first processing delay, the second processing delay, and the third processing delay. The problem-solving module is used to: solve the optimization problem using the differential evolution algorithm to obtain a cloud-edge-device resource scheduling scheme.

4. A cloud-edge-device collaborative computing system, comprising a cloud server, an edge server, and multiple terminal devices, characterized in that, The cloud-edge-device collaborative computing system uses the cloud-edge-device resource scheduling scheme generated by the cloud-edge-device resource scheduling scheme generation method as described in any one of claims 1 to 2 to process tasks.

5. An electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the cloud-edge-device resource scheduling scheme generation method as described in any one of claims 1 to 2.

6. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the cloud-edge-device resource scheduling scheme generation method as described in any one of claims 1 to 2.

7. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the cloud-edge-device resource scheduling scheme generation method as described in any one of claims 1 to 2.

Images