A Resource Allocation Method for Multi-Device Cooperative Tasks in an Industrial Digital Twin Edge Network

Abstract

The present invention discloses a resource allocation method for multi-device collaborative tasks in an industrial digital twin edge network, including: constructing a two-layer digital twin edge network task offloading framework, including a physical entity layer and a digital twin layer; establishing a comprehensive system including a multi-device collaborative task model, a sensing model, a communication and computing model, and a delay and energy consumption model in the physical entity layer; based on the comprehensive system, establishing an SCC-DT model for sensors, edge servers, and actuators in the digital twin layer based on sensing resource deviation, computing resource deviation, and communication resource deviation; and jointly optimizing the computing frequency, transmission power, transmission bandwidth, and offloading factor based on the two-layer digital twin edge network task offloading framework to minimize the end-to-end delay of multi-device collaborative tasks. The present invention solves the problem that most existing task offloading and resource allocation schemes are difficult to be directly applied, and effectively reduces the end-to-end delay of MDC tasks.

Description

Technical Field

[0001] The present invention belongs to the technical fields of wireless communication and industrial Internet of things, and particularly relates to a resource allocation method for multi-device collaborative tasks in an industrial digital twin edge network. Background Art

[0002] With the rapid development of wireless communication and industrial Internet of things (IIoT), application requirements such as unmanned driving and industrial automation have increased sharply. These applications have huge demands for computing resources and strict requirements for time delay. To address these challenges, mobile edge computing (MEC) technology has been introduced into IIoT. By providing cloud computing capabilities on the edge side of the network and migrating resources to a location closer to users, MEC offloads tasks to the edge through task offloading technology. In addition, the concept of digital twin edge network (DITEN) combining digital twin (DT) and edge network (EN) has also been proposed. By collecting real-time information through edge nodes, DT models are established and maintained, and task offloading and resource allocation are optimized in the digital domain.

[0003] Although the prior art has made some progress in IIoT through MEC and DITEN, there are still some deficiencies, specifically including the following:

[0004] 1. When establishing the DT model, the existing DITEN architecture only considers the deviation of computing resources and ignores the equally important deviation of communication and sensing resources.

[0005] 2. Existing research mainly focuses on how to offload tasks from terminal devices to edge servers for execution, without fully considering the requirements of multi-device data collaboration in the IIoT scenario. The formation and execution logics of multi-device collaborative tasks (MDC) are different from those of traditional local independent tasks, making it difficult to directly apply the current task offloading and resource allocation schemes.

[0006] 3. Most of the existing work only considers the case where the task initiating device and the result receiving device are the same device, ignoring the scenario of task and result transmission between different devices, which is an important consideration in actual industrial unmanned operations. Summary of the Invention

[0007] To solve the above technical problems, the present invention proposes a resource allocation method for multi-device collaborative tasks in an industrial digital twin edge network, combining MEC and DT technologies to study the multi-device collaborative task offloading and resource allocation problems in IIoT to address the limitations of the prior art.

[0008] To achieve the above object, the present invention provides a resource allocation method for multi-device collaborative tasks in an industrial digital twin edge network, including:

[0009] Construct a two-layer digital twin edge network task offloading framework, including a physical entity layer and a digital twin layer;

[0010] Establish a comprehensive system including a multi-device collaborative task model, a sensing model, a communication and computing model, and a delay and energy consumption model in the physical entity layer;

[0011] Based on the comprehensive system, establish an SCC-DT model for sensors, edge servers, and actuators in the digital twin layer based on sensing resource deviation, computing resource deviation, and communication resource deviation;

[0012] Based on the two-layer digital twin edge network task offloading framework, jointly optimize the computing frequency, transmission power, transmission bandwidth, and offloading factor to minimize the end-to-end delay of multi-device collaborative tasks.

[0013] Preferably, the physical entity layer includes: a sensing layer, an MEC control center layer, and an execution layer;

[0014] The digital twin layer is composed of the digital twins of physical entities and the mirror image of the entire wireless communication environment.

[0015] Preferably, the SCC-DT model includes:

[0016] The DT model of sensor k is expressed as:

[0017] DT k ={ST k ,ΔST k};

[0018] Wherein, ST k represents the state information of sensor k, and ΔST k represents the DT deviation, and ΔST k ={Δf k ,Δp k ,Δb k ,Δs k} respectively represent the computing frequency deviation, transmission power deviation, bandwidth deviation, and sensing ability deviation between sensor k and its DT;

[0019] The DT model of actuator m is expressed as:

[0020] DT m ={ST m , ΔST m};

[0021] Among them, ST m represents the S state information of the actuator m, and ΔST m ={Δp m , Δb m} represents the transmission power deviation and bandwidth deviation between the actuator and its DT;

[0022] The DT model of the MEC control center layer is expressed as:

[0023] DT ME3C ={ST ME3C , DS ME3C};

[0024] Among them, ST ME3C represents the real-time state data of ME3C, and DS ME3C represents the system information stored in ME3C, and ME3C is the MEC control center layer.

[0025] Preferably, the multi-device collaborative task model includes data perception, raw data offloading, data preprocessing, preprocessed data offloading, task synthesis, edge computing, and result downloading.

[0026] Preferably, the multi-device collaborative task is expressed as:

[0027] J MT =(D MT , K MT , M MT , C MT , T max );

[0028] Among them, D MT , K MT and M MT respectively represent the set of data items required for the task, the set of sensors, and the set of actuators, and C MT and T max represent the number of CPU cycles required for ME3C to calculate unit data and the maximum tolerable delay of the task.

[0029] Preferably, the perception model includes:

[0030] The amount of sensed data of sensor k is:

[0031]

[0032] Among them, s kDenote the sensing ability, as the sensing time;

[0033] The actual sensed data volume of sensor k is:

[0034]

[0035] where, Δs k is the sensing deviation of sensor k;

[0036] The sensing energy consumption is:

[0037] Preferably, the joint optimization problem includes: delay constraint, energy consumption constraint, bandwidth constraint, transmit power constraint, local computing frequency constraint, offloading factor constraint.

[0038] Preferably, the solution process of the optimization problem:

[0039] Decouple the original problem into four sub-problems;

[0040] Based on the solutions of the four sub-problems, combined with the alternating optimization algorithm of the internal convex approximation method and the Lagrangian dual method, obtain the ICA-LD-AO algorithm;

[0041] Based on the ICA-LD-AO algorithm, solve the problem of multi-device collaborative task computing offloading and resource allocation.

[0042] Preferably, the alternating optimization algorithm includes local resource optimization, data transmission optimization, offloading decision optimization, download process optimization.

[0043] The present invention also provides a computer storage medium, on which computer program instructions are stored, and when the computer program instructions are executed by a processor, the resource allocation method of the industrial digital twin edge network multi-device collaborative task as described is implemented.

[0044] Compared with the prior art, the present invention has the following advantages and technical effects:

[0045] The present invention provides a resource allocation method for an industrial digital twin edge network multi-device collaborative task, including: first constructing a two-layer digital twin edge network task offloading framework, including a physical entity layer and a digital twin layer; secondly establishing a comprehensive system including a multi-device collaborative task model, a sensing model, a communication and computing model, and a delay and energy consumption model in the physical entity layer; then based on the comprehensive system, establishing an SCC-DT model for sensors, edge servers, and actuators in the digital twin layer based on sensing resource deviation, computing resource deviation, and communication resource deviation; finally jointly optimizing the computing frequency, transmit power, transmission bandwidth, and offloading factor to minimize the end-to-end delay of the multi-device collaborative task.

[0046] The present invention constructs a two - layer digital twin edge network task offloading framework including a physical layer and a digital layer. This method can more comprehensively grasp the operating state of industrial devices. This two - layer structure enables resource allocation and task offloading strategies to be directly optimized in the digital domain, reducing waste of physical resources and avoiding frequent communication between the edge server and industrial devices.

[0047] The present invention considers the requirements of multi - device data collaboration in the IIoT scenario, such as cross - camera tracking, multi - workshop intelligent logistics distribution, etc. Such tasks are formed by multiple sensors collecting data and the edge server aggregating multiplexed data, which is different from traditional local independent tasks in terms of formation and execution logic. The present invention is specifically designed for such multi - device collaborative tasks (MDC), solving the problem that most existing task offloading and resource allocation schemes are difficult to be directly applied.

[0048] The present invention effectively reduces the end - to - end delay of MDC tasks by jointly optimizing the computing frequency, transmission power, transmission bandwidth, and offloading factor. Brief Description of the Drawings

[0049] The drawings forming a part of this application are used to provide a further understanding of this application. The schematic embodiments of this application and their descriptions are used to explain this application and do not constitute an improper limitation of this application. In the drawings:

[0050] Figure 1 It is the DITEN system model in the IIoT scenario of the embodiment of the present invention;

[0051] Figure 2 It is a schematic diagram of the execution process of multi - device collaborative tasks in the embodiment of the present invention;

[0052] Figure 3 It is a schematic diagram of the system delay constraint analysis in the embodiment of the present invention;

[0053] Figure 4 It is a schematic diagram of the relationship between transmission bandwidth and end - to - end delay in the embodiment of the present invention;

[0054] Figure 5 It is a schematic diagram of the relationship between local computing frequency and end - to - end delay in the embodiment of the present invention;

[0055] Figure 6 It is a schematic diagram of the influence of different data deviation rates on end - to - end delay in the embodiment of the present invention;

[0056] Figure 7 It is a schematic diagram of the influence of different numbers of sensors on end - to - end delay in the embodiment of the present invention. Detailed Embodiments

[0057] It should be noted that, without conflict, the embodiments in the present application and the features in the embodiments can be combined with each other. The present application will be described in detail below with reference to the drawings and in combination with the embodiments.

[0058] It should be noted that the steps shown in the flowchart of the drawings can be executed in a computer system such as a set of computer-executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order than here.

[0059] Embodiment 1

[0060] This embodiment provides a resource allocation method for multi-device collaborative tasks in an industrial digital twin edge network, including:

[0061] S1. Construct a two-layer digital twin edge network task offloading framework, including a physical entity layer and a digital twin layer;

[0062] Specifically, as Figure 1 , in this embodiment, a digital twin edge network (DITEN) system is constructed in the IIoT scenario, and the DITEN system includes a physical entity layer and a digital twin layer.

[0063] S2. Establish a comprehensive system including a multi-device collaborative task model, a sensing model, a communication and computing model, and a delay and energy consumption model in the physical entity layer;

[0064] Specifically, the physical entity layer is composed of a sensing layer, an MEC control center (ME3C) layer, and an execution layer. The sensing layer consists of a large number of sensors, which are responsible for sensing and collecting environmental data and have local preprocessing functions. The ME3C layer consists of a base station (BS) equipped with an MEC server, which is responsible for receiving the offloading data of the sensors, integrating them into computing tasks and processing them. The execution layer consists of multiple actuators, which are responsible for receiving the calculation results and performing corresponding operations.

[0065] S3. Based on the comprehensive system, establish an SCC-DT model for sensors, edge servers, and actuators in the digital twin layer based on sensing resource deviation, computing resource deviation, and communication resource deviation;

[0066] Specifically, S31. SCC-DT model;

[0067] The DT layer is composed of the digital twins of physical entities and the mirror image of the entire wireless communication environment, and is responsible for monitoring the operating state of the physical system and optimizing the task offloading and resource allocation schemes through real-time interaction. K sensors and M actuators in the system are represented by sets and respectively.

[0068] The DT model is constructed and maintained at the ME3C layer, and optimizes resource allocation by monitoring the system status and simulating network operation. Due to factors such as the time-varying nature of the channel and interference, there will inevitably be data deviations between the DT model and its physical entities, and it is impossible to maintain complete data synchronization. These deviations exist not only in computing resources, but also in communication and sensing resources. Therefore, the SCC-DT model established in this embodiment comprehensively considers these three resource deviations. The DT model of sensor k is expressed as:

[0069] DT k ={ST k , ΔST k}(1)

[0070] Where ST k represents the status information (computing frequency, power, etc.) of sensor k, and ΔST k represents the DT deviation. ΔST k ={Δf k , Δp k , Δb k , Δs k} respectively represent the computing frequency deviation, transmission power deviation, bandwidth deviation, and sensing ability deviation between sensor k and its DT.

[0071] The DT model of actuator m is expressed as:

[0072] DT m ={ST m , ΔST m}(2)

[0073] Where ST m represents the S status information (bandwidth, channel status, etc.) of actuator m, and ΔST m ={Δp m , Δb m} represents the transmission power deviation and bandwidth deviation between the actuator and its DT.

[0074] The DT model of ME3C is expressed as:

[0075] DT ME3C ={ST ME3C , DS ME3C}(3)

[0076] Where ST ME3C represents the real-time status data of ME3C (CPU computing frequency, task generation status), and DS ME3C represents the system information stored in ME3C, such as the sensing range of the sensor, the position of the actuator, and the task data volume, etc.

[0077] S32. Multi-device collaborative task model;

[0078] This embodiment considers tasks such as cross-camera trajectory tracking in the IIoT scenario. Such tasks are different from traditional time-driven tasks. Time-driven tasks generate and require the completion of a task in each fixed time slot. However, the MDC tasks in this embodiment are event-driven tasks, which are generated by a specific event. For example, a task is generated only when a person to be tracked enters the monitoring area and does not have to follow a fixed time slot. Therefore, compared with the time-driven scheme, event-driven tasks have lower implementation complexity, especially when the task completion time is much greater than the time slot duration. As Figure 2 shown, the event-driven MDC task requires the collaboration of sensors, ME3C, and actuators to complete. The task execution process consists of seven stages: data perception, raw data offloading, data preprocessing, preprocessed data offloading, task synthesis, edge computing, and result downloading.

[0079] The MDC task initiated by ME3C is denoted as J MT =(D MT ,K MT ,M MT ,C MT ,T max ), where D MT , K MT and M MT represent the set of data items required for the task, the set of sensors, and the set of actuators respectively, satisfying where represents the set of all data items, N k and N m represent the number of sensors and actuators related to the task. C MT and T max represent the number of CPU cycles required for ME3C to calculate unit data and the maximum tolerable delay of the task.

[0080] Since each sensor is located in a different position and has different sensing capabilities, the data it senses is also different. Assume that the set of all data that can be sensed in the system is The data that sensor k can sense is In addition, assume that the sensors are reasonably distributed and there will be no situation where different sensors sense the same data.

[0081] S33. Sensing model;

[0082] The set of sensing capabilities of the sensors is S = {s1,..., s K}, where s k represents the sensing capability, is the sensing time, then the amount of sensed data of sensor k is:

[0083]

[0084] According to the DT model, the sensing deviation of sensor k is Δs k , then the actual sensed data volume of sensor k is:

[0085]

[0086] Therefore, the sensing energy consumption is:

[0087]

[0088] S34. Communication model;

[0089] (1) Uplink transmission. represents the bandwidth allocated when sensor k communicates with ME3C. represents the channel gain between sensor k and ME3C, where β0 represents the path loss at the reference distance d0, represents the actual distance from sensor k to ME3C, and ε represents the path loss coefficient. In this embodiment, the FDMA protocol is adopted to avoid co-channel interference. Therefore, the uplink transmission rate is:

[0090]

[0091] where, N0 and p k represent the noise power spectral density and the transmission power of sensor k, respectively.

[0092] α k is the offloading factor of sensor k, that is The data needs to be preprocessed locally, The data is directly offloaded to ME3C. Then the DT estimated delay for sensor k to offload part of the original data is:

[0093]

[0094] The bandwidth deviation Δb k and the transmission power deviation Δp k can be obtained from the SCC-DT model. Therefore, the delay deviation for sensor k to offload the original data is:

[0095]

[0096] In the formula represents considering the actual signal-to-interference-plus-noise ratio. Therefore, the actual delay for sensor k to offload the original data is:

[0097]

[0098] Assume β kis the ratio of the amount of data after preprocessing to that before preprocessing for sensor k. Then, the DT estimation delay for sensor k to offload preprocessed data is:

[0099]

[0100] Similar to Equation (9), the delay deviation for sensor k to offload preprocessed data to ME3C is:

[0101]

[0102] Therefore, the actual delay for sensor k to offload preprocessed data is:

[0103]

[0104] From the delays of raw data offloading and preprocessed data offloading, the energy consumption during the data offloading process can be calculated:

[0105]

[0106] (2) Downlink transmission. After receiving multi-channel sensor data, ME3C synthesizes and calculates the tasks and finally sends the calculation results to some actuators. The bandwidth allocated to actuator m is b m , satisfying Similar to the uplink, represents the channel gain between ME3C and actuator m. p m represents the power allocated to actuator m, satisfying Then, the downlink transmission rate from ME3C to actuator m is:

[0107]

[0108] where N0 represents the noise power spectral density.

[0109] Assuming that the ratio of the calculation result to the task volume is γ, the calculation result can be obtained as where represents the size of the data volume sent from sensor k to ME3C. Therefore, the download delay from ME3C to actuator m is:

[0110]

[0111] From the SCC-DT model, the bandwidth deviation Δp m and power deviation Δb m in the downlink process can be obtained. Therefore, the download delay deviation of actuator m is:

[0112]

[0113] In the formula Indicates the signal-to-interference-plus-noise ratio (SINR) in the downlink process. Therefore, the delay for actuator m to download the ME3C calculation result is:

[0114]

[0115] S35. Computational model;

[0116] (1) Local preprocessing. The sensor performs preprocessing operations such as denoising and screening on part of the data locally. The amount of preprocessed data of sensor k is f k represents the local processing rate, and C k represents the number of CPU cycles required to preprocess one unit of bit data. Therefore, the local preprocessing delay and energy consumption of sensor k are:

[0117]

[0118] where Δf k represents the calculation frequency deviation in the DT model, and ξ k represents a constant related to the device hardware structure.

[0119] (2) Edge computing. After ME3C receives the offloaded data from the sensor, it first integrates the data to form a computational task, and the task volume is The synthesis time is extremely short and can be ignored. Therefore, the delay for ME3C to process the task is:

[0120]

[0121] S36. Delay and energy consumption model;

[0122] (1) Delay model. Since the preprocessed data offloading and the raw data offloading cannot be carried out simultaneously, there will be Figure 3 the two situations shown. Figure 3 In case (a) in it means that the raw data offloading delay is greater than the preprocessing delay, that is Figure 3 At this time, the preprocessed data needs to wait for a period of time before being transmitted. In case (b) in

[0123] In the multi-device collaborative task considered in this embodiment, the offloading data of any sensor cannot be discarded, because the absence of any piece of data will lead to the failure of task synthesis. Therefore, the data offloading delay is determined by the offloading delay of the slowest sensor. Similarly, the download delay is determined by the download delay of the slowest actuator. Therefore, the total delay is jointly determined by the maximum upload delay and the maximum download delay in 3(b). So, the total end-to-end delay is expressed as:

[0124]

[0125] (2) Energy consumption model. Since ME3C has sufficient energy supply, the energy consumption of edge computing and result download can be ignored. Therefore, the total energy consumption consists of sensing energy consumption, preprocessing energy consumption, and data offloading energy consumption, and can be expressed as:

[0126]

[0127] S4. Jointly optimize the computing frequency, transmission power, transmission bandwidth, and offloading factor to minimize the end-to-end delay of the multi-device collaborative task.

[0128] Specifically, it includes: S41. Propose an optimization problem;

[0129] This embodiment considers seven stages of the MDC task from initiation to result download. Since the delay composition is complex and there are delay conflicts as shown in Fig. 3. Therefore, the goal of this embodiment is to jointly optimize the offloading factor α = {α k}, the local computing frequency f = {f k}, and the transmission power uplink bandwidth the transmission power p allocated by ME3C = {p m}, and the downlink bandwidth b = {b m} to minimize the E2E delay of the MDC task. The optimization problem P1 is expressed as:

[0130]

[0131] In problem P1, (24b) and (24c) represent the delay and energy consumption constraints respectively; (24d) constrains the bandwidth value; (24e) represents the power constraint; (24f) constrains the value of the local computing frequency; (24g) constrains the value of the offloading factor; (24h) represents some non-zero constraints. Since problem P1 is a non-convex problem and is difficult to solve directly. Therefore, this embodiment first decouples the original problem into four independent sub-problems, and then uses the internal convex approximation method (ICA) and the Lagrangian dual method (LD) to convert the non-convex problem into a convex problem, and then adopts the alternating optimization (AO) method to solve it iteratively.

[0132] S42. Solve the computational offloading and resource allocation problems of multi-device collaborative tasks based on the ICA-HA-AO algorithm;

[0133] In problem P1, the sensing and edge computing delays do not involve optimization variables and can be regarded as known quantities. Therefore, we use and to represent the delays of the uplink process and the download process respectively. Then, the E2E delay can be rewritten as:

[0134]

[0135] Optimizing the uplink transmission delay actually means ensuring that all sensors can complete data transmission within the time duration τ up , that is, satisfying Similarly, it is also necessary to ensure that all actuators can complete result reception within the time duration τ down , that is, satisfying Therefore, let τ = {τ up , τ down}, then the original problem P1 can be rewritten as problem P2:

[0136]

[0137] Problem P2 is still non-convex. Next, we solve it by the method of problem decomposition.

[0138] S4201. Local resource optimization;

[0139] This sub-problem finds the next optimal local computing and communication resources by fixing and can be expressed as SP1: In SP1, constraints (26c) and (24c) are non-convex constraints. Next, we use the ICA method to transform SP1 into a convex problem. Among them, constraint (26c) can be split into the following two constraints:

[0140]

[0141] In practice, due to the limited local computing power, the proportion of the preprocessing part is usually very small, that is, β is much less than 1. Therefore,

[0142]

[0143] can be reasonably ignored. Furthermore, (28) and (29) can be simplified to:

[0144]

[0145] Equation (30) is still non-convex. Therefore, we introduce a variable satisfying Then equation (30) is equivalent to:

[0146]

[0147] At this time, (32) is a convex constraint. Substituting the variable into (24b), this constraint can be equivalently expressed as:

[0148]

[0149] At this time, constraint (33a) is still non-convex. According to the ICA method, when x > 0 and y > 0, the upper bound convex function of f(x, y) = xy at the point can be given by the following inequality:

[0150]

[0151] According to (34), let x = p k , (33a) can be iteratively expressed as the following convex constraint:

[0152]

[0153] Based on the above transformation, the non-convex constraint (33a) is converted into constraints (33b) and (35), and the non-convex constraint (30) is converted into (32). Therefore, the non-convex problem SP1 is transformed into the convex problem SP1-C:

[0154]

[0155] Using convex optimization methods such as the interior point method can effectively solve SP1-C. Therefore, the algorithm for solving problem SP1 is summarized as Algorithm 1.

[0156]

[0157]

[0158] S4202, Data transmission optimization;

[0159] In this sub-problem, fix the value and solve for the next optimal uplink bandwidth which is expressed as problem SP2:

[0160]

[0161] Problem SP2 is a convex problem and can be solved using convex optimization methods such as Newton's method

[27] .

[0162] S4203, Offloading decision optimization;

[0163] In this sub-problem, fix the Value, solve for the next optimal offloading decision (α), expressed as problem SP3:

[0164]

[0165] Problem SP3 is a problem with all linear constraints and can be easily solved using linear programming methods.

[0166] S4204. Download process optimization;

[0167] In this sub-problem, fix to find the next optimal downlink computing and communication resources ( b , p ). Expressed as problem SP4:

[0168]

[0169] For ease of calculation, first equivalently consider (b m -Δb m ) and (p m -Δp m ) as b m and p m , and then consider the DT deviation value after obtaining the optimal value. It can be seen that is a concave function of p m , and is the perspective function of the function g m (·). Therefore, the function is a joint concave function of b m and p m . So, problem SP4 can be solved using the Lagrangian duality method. Assume η≥0 represents the dual variable of constraint (39a), γ≥0 represents the dual variable of constraint (39b), and μ m ≥0 represents the dual variable of the m-th constraint in constraint (39c). Therefore, the Lagrangian function of problem SP4 is:

[0170]

[0171] Therefore, the Lagrangian even function of problem SP-4 is:

[0172]

[0173] Its dual problem can be defined as:

[0174]

[0175] SP4 satisfies the Slater condition, so there is strong duality between this problem and its dual problem. Therefore, solving SP4 is equivalent to solving problem (42). Thus, first solve problem (41) to obtain the dual function of SP4, and then maximize the dual function to obtain the solution of problem (42). When a set of {μ m} ≥ 0, η ≥ 0 and γ ≥ 0 is given, problem (41) can be decomposed into the following N m +1 subproblems:

[0176]

[0177] For subproblem (43), it is easy to obtain that the optimal solution is given by:

[0178]

[0179] For problem (44), it can be easily proved that when η = γ = μ m = 0, any b m > 0, p m > 0 are the optimal solutions of the problem; when η > 0, γ = μ m = 0, b m = 0, p m > 0 are the optimal solutions; when γ > 0, η = μ m = 0, p m = 0 is the optimal solution. Obviously, the above cases are of no practical significance. Therefore, the following focuses on solving the optimal solution of problem (44) when γ > 0, μ m > 0.

[0180] Theorem 1: When η > 0, γ > 0, {μ m > 0} are given, the optimal solutions of {b m} and {p m} satisfy:

[0181]

[0182] where the meaning of the symbol |·| * is

[0183] When and are obtained, problem (42) can be solved. Problem (42) is convex but not necessarily differentiable. Therefore, consider solving this problem by the ellipsoid method. The subgradients of the function f(η, γ, {μ m}) with respect to the variables {μ m > 0},

[0184] η > 0 and γ > 0 are respectively and When the optimal solution of the dual problem is obtained η * and γ * then, what is obtained from Equation (45) is the optimal solution of the original problem SP4. It is easy to prove that when the optimal solution is reached, the following conditions must be satisfied:

[0185]

[0186] Combining Equation (46) and Equation (48), the optimal solution of the bandwidth can be obtained:

[0187]

[0188] Furthermore, the optimal solution of the power can be obtained:

[0189]

[0190] Therefore, the algorithm for solving sub-problem SP4 is summarized as Algorithm 2.

[0191]

[0192]

[0193] S4205, ICA-LD-AO algorithm for solving the MTOCCRA problem;

[0194] Based on the solution of the four sub-problems, an alternating optimization (AO) algorithm based on the interior convex approximation method (ICA) and the Lagrangian dual method (LD) is proposed, named the ICA-LD-AO algorithm. Assume that the optimal solution of problem P1 is The algorithm is summarized as Algorithm 3.

[0195]

[0196] S43, Simulation results;

[0197] Consider an IIoT edge computing system consisting of a control center equipped with an MEC server, 10 sensor devices, and 6 actuator devices. The service radius of the MCC is 100 m, and the sensors and actuators are randomly distributed in a 50 m × 50 m area. The simulation parameters are given in Table 1.

[0198] Table 1

[0199]

[0200] To prove the effectiveness of this patent solution, it is compared with the following three benchmark solutions:

[0201] (1) Communication efficiency - priority scheme: Fix the CPU frequency and use the semi - definite relaxation method to solve the problem, aiming to prioritize the optimization of the communication process.

[0202] (2) Computation efficiency - priority scheme: Given the bandwidth and power, use the convex difference algorithm to minimize the E2E delay, prioritizing the optimization of the computation process.

[0203] (3) No - pre - processing scheme: All data in this scheme is offloaded to ME3C to study the impact of the offloading strategy on the delay.

[0204] Figure 4 It shows the impact of the transmission bandwidth on the end - to - end delay. Compared with the other three benchmark schemes, the joint optimization scheme adopted in this embodiment can achieve lower delay. In the case of relatively limited bandwidth resources, the scheme in this embodiment has a more significant performance improvement. When the bandwidth resources are relatively sufficient, although the performance improvement slows down, the delay is still reduced by about 0.25 s compared with the worst - case scheme.

[0205] Figure 5 It shows the relationship between the end - to - end delay and the local computing frequency. When the local computing frequency is high, data tends to be pre - processed locally. Therefore, the pre - processing ratio in the figure increases with the increase of the local computing frequency. In addition, although the enhancement of local computing power will speed up the pre - processing, it will significantly increase the energy consumption, thus resulting in a trade - off problem between pre - processing and energy consumption. Therefore, to balance the delay and energy consumption, the offloading amount must be increased, which also explains why the performance of all schemes decreases with the increase of the local computing frequency, but the scheme in this embodiment is always better than the benchmark schemes.

[0206] Figure 6 It shows the impact of different data deviation rates on the end - to - end delay. As the data deviation rate increases, the delay also shows an increasing trend. It can be seen that data deviation has an important impact on the system performance. Therefore, when constructing the DT model, various data deviations must be fully considered and the deviation rate must be reasonably set to ensure the accuracy of the model. Therefore, it is very meaningful to consider various resource deviations when establishing the SCC - DT model in this embodiment, which can provide reference value for the further optimization of the DT model in the future.

[0207] Figure 7 It shows the performance of the four schemes when the number of sensors is different. The increase in the number of sensors causes an increase in the computational load, resulting in an increase in the delay of all schemes, but the scheme in this embodiment has the best performance. The reason is that the scheme in this embodiment can reasonably allocate computing and communication resources and balance the amount of data for pre - processing and offloading. Therefore, the scheme in this embodiment can be applied to scenarios with large amounts of data and large device scales.

[0208] The above are only the preferred specific embodiments of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present application should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A resource allocation method for multi-device collaborative tasks in an industrial digital twin edge network, characterized in that: The following steps are involved: Build a two-layer digital twin edge network task offloading framework, including the physical entity layer and the digital twin layer; A comprehensive system including a multi-device collaborative task model, a perception model, a communication and computing model, and a delay and energy consumption model is established in the physical entity layer; the physical entity layer includes: a perception layer, a MEC control center layer, and an execution layer; the digital twin layer is composed of a digital twin of the physical entity and a mirror image of the entire wireless communication environment; Event-driven multi-device collaborative tasks require the collaboration of sensors, ME3C and actuators; the multi-device collaborative task model includes data perception, raw data unloading, data preprocessing, preprocessed data unloading, task synthesis, edge computing and result downloading; The multi-device collaborative task is expressed as: J MT =(D MT ,K MT ,M MT ,C MT ,T max ); Among them, D MT , K MT and M MT They represent the data item set, sensor set, and actuator set required for the task, respectively. MT and T max Indicates the number of CPU cycles required by ME3C to calculate unit data and the maximum tolerable latency of the task; Based on the comprehensive system, an SCC-DT model is established for sensors, edge servers, and actuators at the digital twin layer based on perception resource deviation, computing resource deviation, and communication resource deviation; the three resource deviations of perception, computing, and communication are used to reflect the data deviation between the physical entity layer and the digital twin layer; The SCC-DT model includes: The DT model of sensor k is expressed as: DT k ={ST k ,ΔST k }; Among them, ST k Indicates the status information of sensor k, ΔST k Indicates DT deviation, ΔST k ={Δf k ,Δp k ,Δb k ,Δs k } respectively represent the calculation frequency deviation, transmission power deviation, bandwidth deviation and perception capability deviation between sensor k and its DT; The DT model of actuator m is expressed as: DT m ={ST m ,ΔST m }; Among them, ST m Indicates the S state information of actuator m, ΔST m ={Δp m ,Δb m } represents the transmission power deviation and bandwidth deviation between the actuator and its DT; The DT model of the MEC control center layer is expressed as: DT ME3C ={ST ME3C ,DS ME3C }; Among them, ST ME3C Indicates the real-time status data of ME3C, DS ME3C Indicates system information stored in ME3C, ME3C is the MEC control center layer; Based on the two-layer digital twin edge network task offloading framework, the computing frequency, transmission power, transmission bandwidth and offloading factor are jointly optimized to minimize the end-to-end delay of multi-device collaborative tasks.

2. The resource allocation method for multi-device collaborative tasks in an industrial digital twin edge network according to claim 1 is characterized in that: The perception model includes: The amount of sensor k’s perception data is: Among them, s k Indicates the ability to perceive, To perceive time; The actual amount of sensor k’s perception data is: Among them, Δs k is the perception bias of sensor k; The perceived energy consumption is:

3. The resource allocation method for multi-device collaborative tasks in an industrial digital twin edge network according to claim 1 is characterized in that: The joint optimization problems include: delay constraint, energy consumption constraint, bandwidth constraint, transmission power constraint, local computing frequency constraint, and offloading factor constraint.

4. The resource allocation method for multi-device collaborative tasks in an industrial digital twin edge network according to claim 1 is characterized in that: The solution process of the joint optimization problem is: Decouple the original problem into four sub-problems; Based on the solution of four sub-problems, the ICA-LD-AO algorithm is obtained by combining the alternating optimization algorithm of the interior convex approximation method and the Lagrangian dual method; Based on the ICA-LD-AO algorithm, the problem of multi-device collaborative task computing offloading and resource allocation is solved.

5. The resource allocation method for multi-device collaborative tasks in an industrial digital twin edge network according to claim 4 is characterized in that: The alternating optimization algorithm includes local resource optimization, data transmission optimization, offloading decision optimization, and download process optimization.

6. A computer storage medium, characterized in that: The computer storage medium stores computer program instructions, which, when executed by a processor, implement the resource allocation method for multi-device collaborative tasks in an industrial digital twin edge network as described in any one of claims 1 to 5.

Images