Dynamic weighted edge calculation scheduling method based on 5G RedCap module

Abstract

The invention belongs to the technical field of edge computing, and particularly relates to a dynamic weighted edge computing scheduling method based on a 5G RedCap module, and the method comprises the steps: 1, carrying out the node resource sensing of each RedCap module node in a network, and calculating the effective computing power of each RedCap module node at each moment; step 2, by analyzing the dynamic characteristics of the RedCap link of each RedCap module node in the network, extracting the RedCap link feature vector of each RedCap module node; 3, modeling is carried out on the resource demand of each task, and the execution time on different RedCap module nodes is estimated; 4, constructing a multi-dimensional decision matrix, and quantifying the execution benefit of each task on each RedCap module node; and 5, based on the multi-dimensional decision matrix, solving an optimal task allocation scheme by using a Lagrange relaxation method. According to the method, tasks can be scheduled in an optimal mode, and the task execution success rate is increased.

Description

Technical Field

[0001] The present invention belongs to the field of edge computing technology, and specifically relates to a dynamic weighted edge computing scheduling method based on a 5G RedCap module. Background Art

[0002] With the rapid development of 5G communication technology, edge computing, as a technology that can perform computing tasks at the edge of the network close to users, has gradually become an important means to solve the computing pressure of cloud computing centers and reduce data transmission delays. 5G networks have the characteristics of high bandwidth, low latency and large-scale connections, which enable edge computing to carry more applications with high real-time and high computing requirements, such as intelligent transportation, industrial automation, augmented reality (AR) / virtual reality (VR), etc. In this context, the 5GRedCap module, as an important technology in the 5G standard, is optimized for low-power and lightweight computing scenarios such as the Internet of Things (IoT) and smart terminals, so that 5G can better support the access of large-scale low-power devices. However, the existing 5G RedCap module still has many challenges in edge computing task scheduling, especially in terms of computing resource management, communication overhead optimization, energy consumption control and dynamic environment adaptability of task execution. The existing technology has not yet provided an efficient task scheduling solution.

[0003] At present, research on edge computing task scheduling mainly focuses on two types of methods: static task scheduling based on computing resource perception and task offloading strategy based on dynamic network status. The first type of method focuses on the computing power of edge computing nodes. For example, based on the shortest task execution time or minimum completion time strategy, tasks are assigned to nodes with the strongest computing power to minimize the execution time of tasks. However, such methods usually assume that the computing power of edge computing nodes is stable, while ignoring the dynamic changes in computing power of RedCap modules due to factors such as CPU frequency adjustment, temperature changes, and load fluctuations. Therefore, in practical applications, static task scheduling strategies often cannot adapt to the volatility of computing resources of RedCap modules, resulting in overload of some nodes and low computing resource utilization of other nodes, thereby reducing the computing efficiency of the entire system.

[0004] The second type of method mainly considers the dynamic characteristics of wireless communication networks and adopts task offloading strategies based on factors such as channel status, bandwidth occupancy, and packet error rate. Since the communication capability of the RedCap module is limited compared to traditional 5G devices, task offloading usually needs to consider the quality of the communication link to reduce data transmission delay and energy consumption. For example, existing task offloading strategies usually use signal-to-noise ratio, channel quality indicator (CQI) or round-trip time (RTT) to evaluate the quality of the network link and decide whether the task should be executed locally or offloaded to other computing nodes. However, this type of method often has the following problems: (1) Lack of coordinated optimization of computing resources and communication resources, only focusing on the quality of the communication link, while ignoring the computing power and energy consumption characteristics of the RedCap module, so that the task may be offloaded to a node with weaker computing power, resulting in reduced task execution efficiency; (2) The characteristics of the task itself are not considered, such as the computational complexity of the task, data input and output requirements, etc. Different types of tasks have different requirements for computing resources and network resources, so a single task offloading strategy cannot meet all application scenarios. Summary of the invention

[0005] In view of this, the main purpose of the present invention is to provide a dynamic weighted edge computing scheduling method based on the 5G RedCap module. The dynamic weighted edge computing scheduling method of the present invention ensures that tasks can be scheduled in the optimal way and improves the success rate of task execution through precise computing power modeling, network quality analysis, task execution time prediction, energy consumption optimization, load balancing and intelligent optimization solution.

[0006] The technical solution adopted by the present invention is as follows: A dynamic weighted edge computing scheduling method based on a 5G RedCap module, the method comprising: Step 1: Perform node resource perception on each RedCap module node in the network and calculate the effective computing capacity of each RedCap module node at each moment; Step 2: By analyzing the dynamic characteristics of the RedCap link of each RedCap module node in the network, extract the RedCap link feature vector of each RedCap module node; Step 3: Model the resource requirements of each task and estimate the execution time on different RedCap module nodes; Step 4: Construct a multi-dimensional decision matrix to quantify the execution benefits of each task on each RedCap module node; Step 5: Based on the multi-dimensional decision matrix, use the Lagrangian relaxation method to solve the optimal task allocation solution.

[0007] Furthermore, in step 1, the effective computing capacity of each RedCap module node at each moment is calculated by the following formula: ; in, For RedCap module nodes At the moment Effective computing power; For RedCap module nodes CPU frequency; For RedCap module nodes At the moment The CPU utilization rate is in the range of (0,1). For RedCap module nodes At the moment The system overhead factor represents the proportion of computing resources consumed by system maintenance; For RedCap module nodes At the moment Temperature; The temperature threshold is the value above which computing performance begins to decline significantly. The value range is 70 to 85. is the temperature scaling factor, which controls the steepness of the effect of temperature on performance; For RedCap module nodes At the moment Available memory; For RedCap module nodes The total memory capacity.

[0008] Furthermore, in step 2, the RedCap link feature vector of each RedCap module node is extracted by the following formula: ; in, For RedCap module nodes At the moment The RedCap link feature vector contains 4 dimensions; For RedCap module nodes At the moment The effective bandwidth of For RedCap module nodes At the moment jitter; is the reference jitter value; For RedCap module nodes At the moment The round trip delay of is the maximum round-trip delay acceptable to the system; For RedCap module nodes At the moment The physical movement speed of the RedCap module node is 0. is the reference physical movement speed value; For RedCap module nodes At the moment signal-to-noise ratio; is the system noise floor signal-to-noise ratio; For RedCap module nodes At the moment The channel quality index ranges from 0 to 15; For RedCap module nodes At the moment Packet error rate; For RedCap module nodes At the moment Channel switching frequency; is the packet error impact coefficient, ranging from 5 to 100; is the influence coefficient of error packet switching frequency, ranging from 2 to 50.

[0009] Furthermore, in step 3, the execution time on different RedCap module nodes is estimated by the following formula: ; in, For the task In the RedCap module node At the moment Execution time; For the task The number of computational operations required; For the task The computational complexity coefficient of ; For the task The amount of input data; For the task The amount of output data.

[0010] Further, the task The computational complexity coefficient of Calculated using the following formula: ; in, For the task The basic complexity coefficient is determined by the asymptotic complexity of the task: for O(1) complexity tasks, ; For O(logn) complexity tasks, ; For O(n) complexity tasks, ; For O(nlogn) complexity tasks, ; For O(n²) complexity tasks, ; For O(n³) complexity tasks, ; For the task The working set memory size, that is, the amount of memory frequently accessed during task execution; For the task Total memory requirements; is the memory impact factor, with a value ranging from 0.2 to 0.5, reflecting the impact of memory access on computing performance; For the task The data dependency of is 0 or 1, which indicates the dependency between computing units within the task: 0 indicates complete parallelism and no dependency; 1 indicates complete serialization and strong dependency; is the dependency impact factor, with a value range of 0.5 to 1.5; is the instruction mode coefficient, which is determined according to the instruction characteristics of the task: for computationally intensive tasks, The value range is 0.8 to 1.0; for memory-intensive tasks, The value range is 1.2 to 1.5.

[0011] Furthermore, in step 4, a multi-dimensional decision matrix is ​​constructed through the following formula to quantify the execution benefit of each task on each RedCap module node: ; in, For the task In the RedCap module node At the moment implementation benefits; For the task The priority weight of For the task Deadline constraints; For the task In the RedCap module node Calculate the energy required; For the task In the RedCap module node At the moment Energy required for communication; For RedCap module nodes At the moment The load variance reflects the load stability; is the link quality feature vector The impact factor of dimension; Furthermore, Calculated using the following formula: ; in, For RedCap module nodes Energy efficiency coefficient.

[0012] Furthermore, Calculated using the following formula:

[0013] in, For RedCap module nodes Energy efficiency coefficient; For RedCap module nodes The transmission power of For RedCap module nodes The received power.

[0014] Furthermore, in step 5, the optimal task allocation solution is solved using the Lagrangian relaxation method based on the multi-dimensional decision matrix through the following formula: ; in, is the objective function; Assign the matrix to the task, Indicates the task Assigned to RedCap module nodes ,otherwise ; For RedCap module nodes The Lagrange multiplier constrained by the computing resource capacity, is the Lagrange multiplier vector of resource capacity constraints; For the task The Lagrange multiplier of the distribution constraint, is the Lagrange multiplier vector of the allocation constraint; is the total number of tasks; is the total number of RedCap module nodes; the final scheduling decision is obtained by solving the following optimization problem: ; in, and are the Lagrange multiplier vector of the optimal computing resource capacity constraint and the Lagrange multiplier vector of the optimal allocation constraint solved iteratively by the subgradient method, respectively; is the optimal task assignment matrix.

[0015] By adopting the above technical scheme, the present invention produces the following beneficial effects: In the traditional edge computing scheduling method, the computing power evaluation of the task usually only depends on the CPU frequency or the idle state of the computing node, and fails to fully consider the dynamic changes of computing resources. The present invention constructs a computing power model of the RedCap module, introduces multiple factors such as CPU utilization, system overhead, temperature influence, memory occupancy, etc., and accurately models the computing power of the RedCap module, so that task scheduling can accurately evaluate the computing power of each computing node. For example, when the temperature is high, the computing power of the RedCap module will decrease, and the traditional method cannot perceive this, which may cause the task to be assigned to the high-temperature node, thereby affecting the execution efficiency. The computing power model of the present invention can dynamically adjust the computing power evaluation, so that the task scheduling can avoid the nodes with reduced computing power, thereby improving the utilization of computing resources. The computing tasks of the RedCap module often involve data transmission, and most traditional task scheduling methods only consider computing power, but ignore the dynamic changes of network status, resulting in failure or long execution time during task execution due to insufficient bandwidth, high latency, and increased packet error rate. The present invention constructs a RedCap link quality feature vector, extracts multiple key network features such as bandwidth, delay, jitter, signal-to-noise ratio, packet error rate, channel switching frequency, etc., and uses mathematical modeling methods such as exponential decay, normalization and logarithmic transformation, so that the task scheduling system can accurately evaluate the communication quality of different computing nodes, ensure that tasks are preferentially assigned to RedCap module nodes with better communication conditions, thereby reducing the failure rate of task execution and improving the stability of task execution. In traditional task scheduling methods, tasks tend to be assigned to nodes with the strongest computing power, which may cause some RedCap modules to be overloaded, while the computing resources of other nodes are not fully utilized, affecting the overall throughput of the system. The present invention adjusts the task scheduling scheme by constructing a load variance model of computing nodes, so that the load variance is minimized during the task allocation process, ensuring that the computing tasks are evenly distributed between different RedCap modules, and avoiding overloading of some nodes and causing the system computing power to decrease. In this way, the scheduling method of the present invention can improve the stability of the entire computing system and reduce the failure rate caused by overload during task scheduling. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A method flow chart of a dynamic weighted edge computing scheduling method based on a 5G RedCap module provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0017] All features disclosed in this specification, or steps in all methods or processes disclosed, except mutually exclusive features and / or steps, can be combined in any manner.

[0018] Any feature disclosed in this specification (including any additional claims and abstract), unless otherwise stated, may be replaced by other alternative features that are equivalent or have similar purposes. That is, unless otherwise stated, each feature is only an example of a series of equivalent or similar features.

[0019] Example 1: Reference Figure 1 , a dynamic weighted edge computing scheduling method based on a 5G RedCap module, the method comprising: Step 1: Perform node resource perception on each RedCap module node in the network and calculate the effective computing capacity of each RedCap module node at each moment; In the process of evaluating the computing power of the RedCap module, multiple influencing factors need to be considered, including the CPU's main frequency, the current CPU utilization, system maintenance overhead, node temperature, and available memory. First, the CPU's main frequency determines the basic computing power of the computing node, but since the RedCap module usually adopts a low-power architecture and its main frequency is low, it is inaccurate to measure the computing power only by the main frequency. The computing power is also affected by the current CPU utilization. The higher the utilization, the less idle computing resources the node has, so the computing power will be reduced accordingly. At the same time, in the actual operation process, system maintenance tasks (such as operating system background processes, network protocol stack management, etc.) will also consume certain computing resources, so it is necessary to introduce system maintenance overhead parameters to reflect the impact of these overheads on computing power. In addition, temperature is also a key factor. Since the hardware design of the RedCap module usually emphasizes low power consumption and high energy efficiency, its heat dissipation capacity is weaker than that of traditional edge computing devices when running under high load. When the module temperature rises, the chip may start dynamic frequency scaling (DVFS) to reduce power consumption and temperature, which directly affects the computing power. To this end, the present invention introduces a temperature impact factor when modeling computing power to quantify the impact of temperature changes on computing power, thereby avoiding the problem of computing performance degradation caused by ignoring temperature factors during scheduling. Specifically, when the temperature exceeds a certain threshold, the computing power will drop significantly, so a nonlinear function is used in the computing model to simulate this effect, so that as the temperature rises, the decline in computing power presents a gradual characteristic. In addition to temperature, the availability of memory is also an important factor affecting computing power. Since the tasks in edge computing often involve a large amount of data processing, and the memory resources of the RedCap module are limited, if the available memory of the node is insufficient, the task execution may be restricted, or even cause the task to fail. Therefore, when modeling computing power, the present invention introduces the ratio of available memory to total memory as a correction factor to reflect the impact of the storage resources of the current node on computing power. When there is more available memory, the computing power is relatively strong; conversely, when the available memory is insufficient, the computing power will be reduced accordingly, so that the scheduling algorithm can avoid possible memory bottleneck problems when allocating tasks.

[0020] Step 2: By analyzing the dynamic characteristics of the RedCap link of each RedCap module node in the network, extract the RedCap link feature vector of each RedCap module node; In order to accurately measure the network link quality of the RedCap module, we first need to focus on the core indicator of effective bandwidth. Effective bandwidth is a key parameter that determines the data transmission rate, which directly affects the transmission efficiency of task data in the network. However, bandwidth is not constant, but is affected by network load, interference and changes in channel quality. Therefore, in actual measurement, we cannot rely solely on the current bandwidth value, but need to combine historical data for smoothing to obtain a more representative network status. At the same time, considering that RedCap module devices are mostly used in low-power scenarios, their bandwidth usage mode may also have certain limitations. Therefore, when measuring effective bandwidth, it is also necessary to combine the power consumption management strategy of RedCap devices to ensure that the measurement results can reflect the transmission capacity of the device in actual operation. In addition to bandwidth, network jitter is another factor that cannot be ignored. Jitter refers to the change in the delay of data packet transmission, which determines the stability of the task data flow. When the jitter is large, even if the average delay is low, the actual task execution may be affected, especially in delay-sensitive application scenarios such as industrial control or telemedicine. Therefore, in the link quality modeling of the present invention, jitter is introduced as an important characteristic indicator and processed in an exponential decay manner to avoid drastic fluctuations in the calculation results due to short-term jitter anomalies. In addition, in order to more intuitively measure the impact of jitter on data flow, the system also introduces a reference jitter threshold so that the impact of jitter can be dynamically adjusted within a certain range to adapt to different types of computing tasks.

[0021] Another key factor of the network link is the round-trip delay, which is an important criterion for measuring whether the task scheduling can meet the low latency requirements. For edge computing tasks, low latency is the basis for ensuring real-time performance, and although the RedCap module uses 5G communication technology, due to its simplified design, its round-trip delay may be affected by greater uncertainty. Therefore, in the network feature extraction process of the present invention, it is necessary to normalize the round-trip delay of the RedCap module and dynamically compare it with the maximum acceptable delay to ensure that high-latency nodes can be effectively avoided during task scheduling. In addition, considering the mobility of the RedCap module, the network delay may also be affected by the moving speed of the device. Therefore, when calculating the delay, a speed parameter is also introduced so that the system can dynamically adjust the task allocation strategy according to the mobile state of the device, thereby reducing the delay fluctuation caused by the movement of the device. The signal-to-interference ratio is also an important indicator affecting the link quality. In a wireless communication environment, the ratio of signal strength to environmental interference directly determines the stability of data transmission. For the RedCap module, since it mainly works in a low-power mode, its signal power is usually low, making the signal-to-interference ratio more susceptible to external interference. In the method of the present invention, the signal-to-interference ratio is processed by a logarithmic function so that the change in signal quality can be more intuitively reflected during task scheduling. In addition, in order to further enhance the accuracy of channel quality assessment, a channel quality indication parameter is also introduced, which is provided by a 5G base station and can reflect the channel environment of the current RedCap module in real time. By combining the signal-to-interference ratio with the channel quality indication, the system can more accurately evaluate the communication stability of the node, so that nodes with better communication quality are given priority when assigning tasks to reduce the risk of task execution failure. In the actual communication process, the packet error rate and the channel switching frequency are also key factors affecting the stability of data transmission. The packet error rate refers to the probability of data packet loss during data transmission. Too high a packet error rate will cause data retransmission, thereby increasing network delay and energy consumption. The channel switching frequency reflects the frequency of device switching between different channels. When the switching frequency is too high, the stability of data transmission will be affected. Therefore, in the network link quality assessment model of the present invention, the packet error rate and the channel switching frequency are combined into a comprehensive factor, which is normalized by a nonlinear function so that its impact can better meet the actual needs of the edge computing scenario. Specifically, during the task scheduling process, when the packet error rate or channel switching frequency of a RedCap module is too high, the link quality score of the node will be reduced accordingly, thereby reducing the probability of assigning tasks to the node.

[0022] Step 3: Model the resource requirements of each task and estimate the execution time on different RedCap module nodes; In the process of modeling the resource requirements of tasks, multiple factors such as the amount of computation of the task, the size of data input and output, and the computational complexity need to be considered. The amount of computation of a task is usually measured by the number of floating-point operations required to perform the task, which is the basic parameter for determining the execution time of the task, while the computational complexity is determined by the algorithmic characteristics of the task, reflecting the execution efficiency of the task on different computing architectures. Due to the limited computing power of the RedCap module, the impact of computational complexity on the task execution time is particularly obvious, and the execution time of different tasks on the same computing resources may vary greatly. Therefore, in the method of the present invention, the task computational complexity is modeled separately so that the computational requirements of different tasks can be distinguished during the scheduling process, and more accurate allocation decisions can be made based on the computing power of the RedCap module. In addition to the computational requirements, the execution time of the task is also affected by the data transmission delay. Since the RedCap module is usually deployed in an edge computing environment, the data input and output of the task often need to be transmitted through a wireless network, and the communication capability of the RedCap module is limited compared to the full-featured 5G device, so the proportion of data transmission overhead in the task execution time may be high. In order to estimate the task execution time more accurately, it is necessary to consider the task data input size, data output size and the network bandwidth of the RedCap module. At the same time, the packet error rate is an important factor affecting the data transmission time. When the packet error rate is high, data retransmission will cause additional transmission delay. Therefore, when calculating the task transmission delay, the packet error rate needs to be corrected to reflect its actual impact on the task execution time.

[0023] In addition, the data transmission of the task is also affected by the round-trip delay. In the task scenario with a large amount of data, a higher round-trip delay will significantly increase the transmission time of the task. Therefore, in the method of the present invention, the execution time estimation of the task not only takes into account the computational overhead, but also combines the dynamic characteristics of network transmission to ensure the accuracy of the estimation result. In the calculation process of the task execution time, it is necessary to comprehensively consider the calculation time and the data transmission time to obtain the complete execution delay of the task on different RedCap modules. First, the calculation time is directly related to the task calculation amount and the computing power of the RedCap module. When the computing power of the RedCap module is strong, the calculation time is short, otherwise the calculation time will be longer. Therefore, when scheduling tasks, it is necessary to give priority to RedCap modules with strong computing power and low current load to reduce the calculation time. On the other hand, the data transmission time is affected by factors such as bandwidth, packet error rate, and round-trip delay. In order to reduce the data transmission time, the system needs to select RedCap modules with higher bandwidth, lower packet error rate, and shorter round-trip delay as much as possible to perform tasks. Especially in large-scale data processing tasks, data transmission time often occupies an important part of the total task execution time. Therefore, in the task scheduling process, it is necessary to dynamically adjust the weights of computing and communication to ensure a balance between computing and communication overheads so that the task can be completed within the optimal time range.

[0024] Step 4: Construct a multi-dimensional decision matrix to quantify the execution benefits of each task on each RedCap module node; In the process of task scheduling, we first need to clarify the core factors that affect the efficiency of task execution, including task execution delay, computing energy consumption, communication energy consumption, node load balancing, and network link quality. Execution delay is one of the key indicators that affect task scheduling. In the RedCap module, due to its limited computing power, the task execution delay may be subject to dual constraints of computing resources and network bandwidth. Therefore, in the process of building the decision matrix, the task execution delay is modeled as an exponential decay function to ensure that the scheduling scheme with shorter delay obtains a higher benefit value. At the same time, in order to reduce the energy consumption of the system while ensuring computing efficiency, the decision matrix also needs to consider the impact of computing energy consumption and communication energy consumption. In the RedCap module, computing energy consumption is proportional to the frequency of the CPU, while communication energy consumption is closely related to the data transmission time and transmission power. Therefore, when building the decision matrix, it is necessary to normalize the computing energy consumption and communication energy consumption so that it can maintain the same dimension as indicators such as execution delay, so as to achieve fair comparison between different indicators. In addition, the load balancing of the RedCap module is also an important factor affecting task scheduling. Due to the limited computing resources of the RedCap module, if the load of some nodes is too high, it may cause the task queuing waiting time to increase, or even resource overflow. Therefore, in the construction process of the decision matrix, it is necessary to model the load variance of the node to ensure that the task can be evenly distributed between different RedCap module nodes, thereby avoiding the unbalanced use of computing resources and improving the overall computing efficiency of the system. At the same time, in order to further improve the accuracy of task scheduling, the method of the present invention also introduces the influencing factors of network link quality, including parameters such as bandwidth, delay, packet error rate and channel switching frequency, and integrates these network characteristics into the decision matrix by weighted product, so that the RedCap module nodes with better network conditions can obtain higher scores in the scheduling process to improve the execution success rate of the task and the data transmission efficiency.

[0025] Step 5: Based on the multi-dimensional decision matrix, use the Lagrangian relaxation method to solve the optimal task allocation solution.

[0026] In the optimization process of task scheduling, we first need to define the distribution relationship between tasks and nodes, and build an optimization goal to maximize the computing efficiency of the overall system while ensuring the rational use of computing resources and communication resources. Since the computing power of the RedCap module is relatively limited, if all tasks are scheduled to the node with the strongest computing power, the node may reduce the overall task throughput due to overload. Therefore, computing resource constraints need to be introduced in the optimization goal to ensure that tasks can be evenly distributed on different RedCap module nodes. In addition, since the execution of tasks is not only affected by computing power, but also by network factors such as network bandwidth, round-trip delay, and packet error rate, communication resource constraints need to be added to the optimization goal to ensure that the scheduling of tasks can meet the needs of both computing and communication, thereby improving the success rate of task execution and reducing task delays and failures caused by network bottlenecks.

[0027] In the actual solution process, since the task scheduling problem is usually a discrete optimization problem, direct solution may lead to excessive computational complexity, especially when the number of tasks is large, the traditional exhaustive search method is difficult to obtain the optimal solution within a reasonable time. In order to solve this problem, the present invention adopts the Lagrange relaxation method, which relaxes the constraints of the task scheduling problem into an unconstrained optimization problem by introducing Lagrange multipliers, so that the task allocation decision can be gradually converged to the optimal solution through iterative solution. During the solution process, the Lagrange multipliers will be dynamically adjusted according to the current task allocation situation, so as to ensure that the computing resources and communication resource constraints of the task can be met, while maximizing the computing efficiency of the system. Compared with the traditional integer programming solution method, the Lagrange relaxation method can effectively reduce the computational complexity, making the optimization process of task scheduling more efficient, and is suitable for edge computing environments with limited computing power such as the RedCap module.

[0028] In the optimization process, the priority of tasks also needs to be considered. Due to the different urgency of different tasks, in some scenarios, high-priority tasks need to be preferentially scheduled to RedCap module nodes with strong computing power and better network conditions, while low-priority tasks can be appropriately delayed for execution. Therefore, in the optimization target of the Lagrangian relaxation method, the present invention adopts a dynamic weight adjustment mechanism, that is, according to the priority of the task, the benefit value of different tasks is weighted, so that high-priority tasks can obtain greater weights during the scheduling process, thereby ensuring that key tasks can be completed in the shortest time. At the same time, in order to prevent low-priority tasks from being in a waiting state for a long time, the method of the present invention also introduces a priority improvement strategy based on task waiting time, that is, when the waiting time of a task exceeds a certain threshold, its priority will gradually increase over time, thereby ensuring that all tasks can be scheduled and executed within a reasonable time range, and preventing some tasks from being unprocessed for a long time due to low priority. In the optimization process of task scheduling, the load balancing of the system also needs to be considered. If the computing load of some RedCap module nodes is too high, and the computing resources of other nodes are still in an idle state, the computing efficiency of the overall system will be affected. Therefore, in the optimization process of the Lagrangian relaxation method, the present invention adopts a load balancing constraint, that is, in the task scheduling process, the load variance of the computing node needs to be kept within a certain range to ensure that the computing tasks can be evenly distributed on different RedCap module nodes, thereby avoiding some nodes from reducing the computing efficiency due to overload, and improving the computing throughput of the entire system. In addition, in order to enhance the stability of task scheduling, the method of the present invention also adopts an iterative optimization strategy based on the subgradient method, that is, in the optimization process, the system will dynamically adjust the task allocation scheme according to the current task allocation, so that the computing resources and communication resource utilization of the task can be continuously optimized, thereby improving the overall efficiency of task scheduling.

[0029] Example 2: In step 1, the effective computing capacity of each RedCap module node at each moment is calculated by the following formula: ; in, For RedCap module nodes At the moment Effective computing power; For RedCap module nodes CPU frequency; For RedCap module nodes At the moment The CPU utilization rate is in the range of (0,1). For RedCap module nodes At the moment The system overhead factor represents the proportion of computing resources consumed by system maintenance; For RedCap module nodes At the moment Temperature; The temperature threshold is the value above which computing performance begins to decline significantly. The value range is 70 to 85. is the temperature scaling factor, which controls the steepness of the effect of temperature on performance; For RedCap module nodes At the moment Available memory; For RedCap module nodes The total memory capacity.

[0030] Specifically, the numerator part of the formula Mainly used to describe the theoretical computing power of a node. Represents the CPU frequency, usually in Hz, which determines the number of computing instructions that can be executed per unit time. In most traditional computing power modeling methods, CPU frequency is one of the most core parameters that determine computing power. However, in the RedCap device environment, it is not enough to only consider the CPU frequency, because RedCap devices are often in a state of dynamic load changes, and some computing resources may be occupied by system processes. Therefore, this formula introduces , that is, CPU utilization, its value range is It represents the available ratio of CPU resources at the current moment. A higher CPU utilization means that the computing resources of the node have been occupied by more tasks, so the actual available computing power will be reduced accordingly. In addition, RedCap devices usually need to perform system maintenance tasks, such as network protocol management, data cache update, etc. These tasks will consume some computing resources, so the formula also introduces , which is the proportion of computing resources occupied by system maintenance. When it is high, it means that the computing resources of the node are occupied by more system tasks, thus reducing the actual capacity available for task computing. Therefore, the formula adopts As a correction factor to reflect the impact of system maintenance tasks on computing power.

[0031] Secondly, the denominator of the formula Mainly used for temperature attenuation modeling. In the actual operation of RedCap equipment, temperature is one of the key factors affecting computing power. High temperature will cause the power consumption of the chip to increase, which will trigger the dynamic frequency adjustment mechanism (DVFS), causing the CPU frequency to drop, and may even trigger the thermal protection mechanism, causing the computing power to drop sharply. Therefore, in the computing power modeling of the present invention, a temperature influence factor based on the Logistic function is used, so that the computing power is almost unaffected when the temperature is low, and when the temperature exceeds a certain threshold (usually in arrive ℃), the computing power will decrease exponentially as the temperature increases. The steepness of this temperature effect is controlled. The smaller the value, the more sensitive the computing power is to temperature changes. The larger the value, the gentler the temperature effect. In this way, the system can accurately evaluate the computing power of the RedCap device under different temperature conditions and avoid high-temperature nodes when scheduling tasks to improve the stability of the overall system. Finally, the last term of the formula Used to characterize the impact of memory availability on computing power. In the computing environment of RedCap devices, tasks usually involve data access, and the availability of memory directly affects the execution efficiency of tasks. When there is less available memory, the computing process of the task may be affected by memory swapping, and may even cause the task to fail. Therefore, in the computing power modeling of the present invention, the memory usage ratio is introduced. As a correction factor. In order to avoid the computing power from fluctuating dramatically with the memory usage, the formula uses the square root function for smoothing, so that the computing power is slightly enhanced when high memory is available, and the computing power is significantly reduced when low memory is available, thereby ensuring that the task scheduling can give priority to the RedCap nodes with more available memory to improve the successful execution rate of the task.

[0032] Example 3: In step 2, the RedCap link feature vector of each RedCap module node is extracted by the following formula: ; in, For RedCap module nodes At the moment The RedCap link feature vector contains 4 dimensions; For RedCap module nodes At the moment The effective bandwidth of For RedCap module nodes At the moment jitter; is the reference jitter value; For RedCap module nodes At the moment The round trip delay of is the maximum round-trip delay acceptable to the system; For RedCap module nodes At the moment The physical movement speed of the RedCap module node is 0. is the reference physical movement speed value; For RedCap module nodes At the moment signal-to-noise ratio; is the system noise floor signal-to-noise ratio; For RedCap module nodes At the moment The channel quality index ranges from 0 to 15; For RedCap module nodes At the moment Packet error rate; For RedCap module nodes At the moment Channel switching frequency; is the packet error impact coefficient, ranging from 5 to 100; is the influence coefficient of error packet switching frequency, ranging from 2 to 50.

[0033] Specifically, the first item of the link feature vector Mainly used to describe the bandwidth status of RedCap devices and their jitter impact. Representative equipment At the moment The effective bandwidth determines how fast the device can transmit data. However, in a wireless communication environment, considering only the bandwidth is not enough, because network jitter may cause instability in the transmission time of data packets, which in turn affects the execution time of the task. Therefore, the formula introduces an exponential decay factor , which is used to describe the impact of jitter on bandwidth. When the jitter is small, this term has little effect on bandwidth attenuation, while when the jitter is large, the exponential term approaches zero, thereby reducing the link quality score of the node. In this way, the formula can more accurately reflect the effectiveness of bandwidth under different jitter levels and improve the preference of task scheduling for high-stability communication paths. Secondly, the second term of the link feature vector It is mainly used to describe the impact of round-trip time (RTT) and the physical movement speed (V) of the device. Represents the round-trip time of a data packet between the RedCap device and the remote server. It is a key indicator for measuring network latency. In order to make this indicator more adaptable, the formula is normalized to make it relative to the maximum acceptable round-trip delay of the system. Compare the time intervals between the two links to ensure that the task prioritizes the low-latency link. At the same time, considering that the RedCap device may be in a mobile state, the physical movement speed It will also affect the stability of data transmission. For example, when the device is in high-speed motion, the channel quality will become unstable and the volatility of the round-trip delay will increase. Therefore, the formula introduces an exponential decay factor. ,ensuring that devices with faster movement speeds are assigned lower link scores during task scheduling, so as to reduce the impact of highly dynamic environments on task execution.

[0034] The third dimension It is mainly used to measure the impact of signal quality on link performance. In a wireless communication environment, the signal-to-interference-plus-noise ratio (SINR) is an important factor in determining communication quality. It reflects the ratio of signal strength to environmental noise. When SINR is too low, the error rate of data transmission will increase, resulting in increased retransmission and delay. Therefore, this formula uses As a normalized signal quality indicator, SINR can be reasonably quantified under different background noises. In addition, the channel quality index (CQI) is also an important parameter for measuring the quality of wireless communication. Its value range is 0 to 15. The higher the value, the better the channel state. Therefore, this formula will Normalized to This allows the item to accurately reflect the signal quality of the device and give priority to selecting RedCap devices with better signal quality for data transmission during task scheduling. Finally, the fourth item of the link feature vector It is mainly used to characterize the stability of data transmission, that is, the impact of packet error rate (PER) and channel switching frequency (H). It reflects the probability of a device losing data packets during wireless communication. A higher packet error rate means that more data retransmissions are required, which increases the task execution time and increases the energy consumption of the system. Therefore, the formula adopts This nonlinear decay function makes the term close to 1 when the packet error rate is low, and drops rapidly when the packet error rate is high, thereby reducing the link score of the device. Similarly, the channel switching frequency It reflects the number of times the device switches between different channels. A higher switching frequency usually means the instability of the communication link, which is prone to packet loss or delay. Therefore, the formula is adopted As a correction item, it ensures that devices with stable channels are more likely to be selected to perform computing tasks during task scheduling.

[0035] Example 4: In step 3, the execution time on different RedCap module nodes is estimated by the following formula: ; in, For the task In the RedCap module node At the moment Execution time; For the task The number of computational operations required; For the task The computational complexity coefficient of ; For the task The amount of input data; For the task The amount of output data.

[0036] Specifically, the first term of the formula The pure computing time of the task on the RedCap device is calculated. Representative tasks The number of computational operations required, usually measured in floating point operations (FLOPs), reflects the computational effort of the task itself. Tasks with a higher number of computational operations usually take longer to compute, so It plays a role of linear growth in the calculation time. In order to more accurately characterize the calculation requirements of different types of tasks, the calculation complexity coefficient is introduced into the formula , which is used to describe the algorithmic complexity of a task on a RedCap device. Even if some tasks have the same number of computational operations, their computational efficiency may vary greatly due to complex data dependencies, matrix operations, or nonlinear transformations. As an adjustable parameter, the calculation time estimation can be more in line with the actual calculation cost of the task. The computing power of the RedCap device is a dynamic value, which is affected by factors such as CPU frequency, system load, temperature, and memory usage. Therefore, this formula uses As a computing power indicator, it represents the computing power of the RedCap device at the moment The actual available computing power. In the computation time estimation, As the denominator, the RedCap device with higher computing power can provide shorter computing time under the same computing task, thereby improving the efficiency of task execution. The computing power value has been accurately defined in the computing power modeling of step 1, so it can ensure the rational use of computing resources in the task execution time estimation, so that task scheduling can fully consider the computing power differences of different nodes and avoid overloaded nodes affecting computing efficiency.

[0037] Next, the second part of the formula The data transmission time of the task on the RedCap device is calculated. In edge computing scenarios, tasks usually require data transmission between the RedCap device and the cloud or other computing nodes, so the impact of communication resources on task execution time cannot be ignored. Here, Representative tasks The amount of input data, Representative tasks The sum of the two reflects the total data transmission requirements of the task. In 5G RedCap devices, due to limited bandwidth, data transmission may become a bottleneck for task execution. Therefore, when estimating execution time, bandwidth needs to be considered comprehensively. and Packet Error Rate Impact on data transfer rate. The first term of data transfer time Indicates the theoretical data transmission duration of the task, where Represents the current moment RedCap Equipment The available bandwidth determines the rate at which data can be transmitted. However, in a wireless communication environment, bit errors and packet loss may occur during data transmission, requiring data retransmission. Therefore, this formula introduces the bandwidth term As a correction factor, it ensures that the RedCap device with a higher packet error rate shows a higher communication delay in the task execution time estimation, so that the task scheduling can automatically avoid the link with a high bit error rate to improve the success rate of task execution. It is used to correct the impact of network round-trip time (RTT) on data transmission. In the actual task execution process, the data transmission time depends not only on the bandwidth, but also on the network delay. The network link quality of the RedCap device may be affected by factors such as channel interference, device mobility, and network load, resulting in longer round-trip delays for data packets, thereby reducing the overall efficiency of data transmission. Therefore, this formula introduces an RTT correction term when calculating the data transmission time to describe the impact of the round-trip time of the data packet between the RedCap device and the server on the data transmission rate. When it is small, the correction term is close to 1, and the impact on the transmission time is small; When it is large, the correction term will increase significantly, making the task execution time calculation more in line with the actual network environment.

[0038] Example 5: Task The computational complexity coefficient of Calculated using the following formula: ; in, For the task The basic complexity coefficient is determined by the asymptotic complexity of the task: for O(1) complexity tasks, ; For O(logn) complexity tasks, ; For O(n) complexity tasks, ; For O(nlogn) complexity tasks, ; For O(n²) complexity tasks, ; For O(n³) complexity tasks, ; For the task The working set memory size, that is, the amount of memory frequently accessed during task execution; For the task Total memory requirements; is the memory impact factor, with a value ranging from 0.2 to 0.5, reflecting the impact of memory access on computing performance; For the task The data dependency of is 0 or 1, which indicates the dependency between computing units within the task: 0 indicates complete parallelism and no dependency; 1 indicates complete serialization and strong dependency; is the dependency impact factor, with a value range of 0.5 to 1.5; is the instruction mode coefficient, which is determined according to the instruction characteristics of the task: for computationally intensive tasks, The value range is 0.8 to 1.0; for memory-intensive tasks, The value range is 1.2 to 1.5.

[0039] Specifically, the basic part of the computational complexity is given by The parameter is set according to the asymptotic time complexity of the task. Since the computational complexity of the task hardly increases with the input size, we set , that is, the computational complexity is the lowest. As for the logarithmic time complexity For tasks such as binary search or hash lookup, the computational complexity is slightly higher than constant time, so we set To appropriately improve the computational complexity estimate. For linear time complexity For tasks such as simple traversal or linear search, the computational complexity grows proportionally with the input size, so we set For linear logarithmic time complexity For tasks such as quick sort or merge sort, the computational complexity is between linear and quadratic complexity, so set For higher order complexities, such as quadratic complexity , cubic complexity And higher-order complexity tasks, since the amount of computation increases dramatically with the increase of input size, we set to 3.0 and 5.0 to more accurately reflect the consumption of computing resources by these tasks.

[0040] However, it is not enough to rely solely on the asymptotic complexity of the algorithm to estimate the computational complexity of the task, because in the actual execution process, the computational complexity of the task is also affected by the memory access pattern. Therefore, the present invention introduces a memory impact factor, the second term of the formula The impact of the task's memory access overhead on computational complexity is quantified. represents the working set size of the task, that is, the size of the memory area frequently accessed during the calculation process, and Represents the total memory requirement of the task. If near , which means that the calculation of this task involves a lot of memory access, which may cause cache misses, thus increasing the calculation time. Therefore, this formula uses a normalized ratio As a correction factor for computational complexity, the parameter The weights are adjusted from 0.2 to 0.5 to ensure that the memory access pattern does not have too much impact on the computational complexity, but is sufficient to reflect the differences in computational efficiency of different tasks at the CPU-cache-memory level.

[0041] In addition to memory access patterns, data dependencies within tasks are also important factors affecting computational complexity. Some tasks can be fully parallelized during computation, while others have strong dependencies and need to be executed in a strict order, which directly affects the computational efficiency of the tasks. Therefore, the third term of the formula The impact of task parallelism on computational complexity is quantified. Data Dependency The value is 0 or 1, where 0 means no data dependency and can be fully parallelized, such as matrix addition, vector operations, etc., while 1 means the task is completely serial and has strong data dependency, such as matrix Gaussian elimination, some dynamic programming tasks, etc. In the computational complexity estimation, if the task dependency is high, its computational complexity will increase accordingly, so this item is passed (value range 0.5 to 1.5) is adjusted so that the computational complexity of tasks with strong data dependence is higher, while the computational complexity of fully parallel tasks is lower, thereby optimizing the execution efficiency of tasks on the RedCap device.

[0042] Finally, the fourth term of the formula It is used to consider the instruction mode of the task, that is, how the calculation type of the task affects the computational complexity. Different tasks may tend to be computationally intensive or memory intensive, and these tasks have large differences in the consumption of computing resources during actual execution. Computationally intensive tasks mainly rely on the CPU to perform floating-point operations or integer operations, and have a higher instruction execution efficiency, so their instruction mode coefficient The value range is 0.8 to 1.0. For memory-intensive tasks, such as database query and large-scale data sorting, the execution time of the task is mainly affected by the memory access overhead, and the instruction execution efficiency is low. The value range is relatively high (1.2 to 1.5) to accurately reflect the computational complexity characteristics of this type of task.

[0043] Example 6: In step 4, a multi-dimensional decision matrix is ​​constructed by the following formula to quantify the execution benefit of each task on each RedCap module node: ; in, For the task In the RedCap module node At the moment implementation benefits; For the task The priority weight of For the task Deadline constraints; For the task In the RedCap module node Calculate the energy required; For the task In the RedCap module node At the moment Energy required for communication; For RedCap module nodes At the moment The load variance reflects the load stability; is the link quality feature vector The impact factor of dimension.

[0044] Specifically, the numerator part of the formula It is mainly used to describe the time sensitivity of tasks and the impact of task priority weight on execution efficiency. Representative tasks The priority weight is used to distinguish the importance of different tasks. For example, in edge computing scenarios, some tasks (such as telemedicine and unmanned driving control instructions) have high real-time requirements, while some tasks (such as data batch processing and log analysis) have low execution latency requirements. Therefore, this formula allows the scheduling system to set different priority weights for different tasks, so that high-priority tasks can get more computing resource support during the scheduling process. Task execution time It is determined by both computing time and data transmission time, and is one of the key factors affecting the efficiency of task execution. In order to make the deadline constraint of the task have an appropriate impact on the scheduling decision, this formula uses an exponential decay function. , which is used to quantify the impact of task execution time on benefits. ), the item is close to 1, indicating that the task can be completed within a sufficient time frame without affecting the overall execution efficiency; and when the execution time of the task is close to or even exceeds its deadline (i.e. ), which approaches 0, means that the value of executing the task on the current node is low, thereby reducing the probability of selecting the node in task allocation. Through this time decay mechanism, the present invention can effectively avoid the execution failure of timed tasks and improve the overall completion rate of tasks.

[0045] Secondly, the denominator of the formula It is mainly used to describe the energy consumption of tasks and the load stability of nodes to ensure low energy consumption and load balance of task scheduling. Representative tasks In RedCap Equipment The computing energy consumption on the RedCap is usually proportional to the square of the CPU frequency, that is, devices with high computing power tend to consume more energy. Therefore, during the task scheduling process, if the computing energy consumption of a RedCap device is too high, the system will reduce the number of tasks scheduled on the device to reduce the energy consumption of the entire system and improve the sustainability of computing. At the same time, represents the communication energy consumption of the task, which is mainly affected by bandwidth, data volume, packet error rate and transmission distance. In the edge computing environment of 5G RedCap devices, the data transmission cost of the task cannot be ignored, especially when the bandwidth is limited, the communication energy consumption may be much higher than the computing energy consumption. Therefore, this formula takes into account the computing energy consumption and communication energy consumption to ensure that the execution efficiency of the task depends not only on the computing power, but also on the communication conditions. In addition, Reflects the load balancing status of the RedCap device. Representative equipment At the moment The load variance is used to measure the load stability of the current RedCap device. A higher load variance means that the computing resource usage of the device fluctuates greatly, and there may be resource shortages at certain times, resulting in task execution failure. Therefore, in this formula, As a load balancing penalty item, devices with large load fluctuations are given lower execution benefits during task scheduling, so that RedCap devices with more stable loads are given priority to execute tasks. This method can effectively improve the global load balancing of task scheduling and avoid overloading of individual devices and affecting the overall computing efficiency of the system. Finally, the last part of the formula The impact of the link quality of the RedCap device on the performance of the mission is quantified. Represents RedCap Equipment At the moment The link feature vector includes multiple key indicators such as bandwidth, delay, signal quality, and packet error rate. Different link quality characteristics have different impacts on task scheduling, so the weight factor is introduced into the formula , which is used to adjust the contribution of different network parameters to the performance of the task. For example, in some tasks (such as high-throughput data transmission tasks), bandwidth may be the most important factor, so the bandwidth-related parameters For low-latency tasks (such as real-time video analysis and remote control tasks), the round-trip delay may be the decisive factor, so the delay-related parameters Through the weighted product of the link feature vectors, the present invention can ensure that the influence of the communication network is fully considered during task scheduling, and give priority to RedCap devices with low latency, high bandwidth, low packet error rate and good signal quality, thereby improving the execution success rate of the task and the communication efficiency of the overall system.

[0046] Embodiment 7: Calculated using the following formula: ; in, For RedCap module nodes Energy efficiency coefficient.

[0047] Specifically, the core part of the formula is , which characterizes the computing energy consumption characteristics of the RedCap device. Represents RedCap Equipment The energy efficiency coefficient is a parameter related to the hardware characteristics of the device, which is used to describe the energy consumption sensitivity of the device. Different RedCap devices may have different processor architectures and power management strategies, so May vary depending on device design. For example, a high-performance RedCap device may use a more advanced, low-power processor that Relatively small, while some older or less energy-efficient devices may have higher , which means that it consumes more energy when calculating the same task. During the task scheduling process, the system can The energy consumption is optimized and adjusted according to the difference in energy consumption, and nodes with lower energy consumption are selected to perform computing tasks first, thereby reducing the energy consumption of the entire computing network. In addition, the formula It reflects the impact of CPU frequency on computing energy consumption. In most computing devices (including RedCap devices), computing energy consumption is usually proportional to the square of CPU frequency, that is, . This is due to the effect of the dynamic voltage and frequency adjustment mechanism. When the CPU runs at a high frequency, not only the computing performance is improved, but its power consumption will also increase exponentially. Therefore, during the task scheduling process, if you want to reduce computing energy consumption, the system can select a low-frequency RedCap device to execute the task to reduce energy consumption. However, too low a CPU frequency may cause the task execution time to be too long. Therefore, the method of the present invention needs to comprehensively weigh the impact of CPU frequency and computing energy consumption during scheduling to ensure that the task can be executed with low energy consumption and that the task execution will not fail due to too low a CPU frequency.

[0048] In addition to the computing characteristics of the device itself, the computing requirements of the task are also the determining factors of computing energy consumption. Therefore, the second half of this formula Quantifies the computational effort of the task. Here, Representative tasks The number of computational operations, usually measured in floating point operations (FLOPs), represents the basic amount of computation required during the execution of a task. The larger the number of computational operations, the more computational steps the task needs to perform, and therefore the higher its computational energy consumption. The computational complexity coefficient of the task is determined by factors such as the algorithmic complexity, data dependency, and memory access pattern of the task, and has been defined in detail in Example 5. Different tasks have different computational complexities. Even if the number of computational operations is the same, the computational energy consumption of the tasks may vary greatly. For example, computationally intensive tasks (such as deep learning reasoning and matrix operations) usually consume more computational energy than I / O intensive tasks (such as database queries and log processing). The value of may vary significantly in different types of tasks. Incorporating computational energy consumption models makes energy consumption estimation more accurate and helps the task scheduling system make more intelligent decisions when allocating tasks.

[0049] Embodiment 8: Calculated using the following formula:

[0050] in, For RedCap module nodes Energy efficiency coefficient; For RedCap module nodes The transmission power of For RedCap module nodes The received power.

[0051] Specifically, the formula consists of two parts. The first term The uplink transmission energy consumption of the task is calculated, that is, the data output of the task The energy consumed by transmitting from the RedCap device to the remote server or other computing node; the second The downlink transmission energy consumption of the task is calculated, that is, the data input of the task The energy consumed by transmitting from an external server or other computing node to the RedCap device. The sum of the two parts is the total communication energy consumption of the task. The core part of the formula is and Respectively represent the transmit power and receive power of the RedCap device, usually in watts (W). Determines the energy consumption required when a device transmits data on a wireless link, while the receiving power Reflects the power consumption of the device when receiving data transmitted by the remote server. Since RedCap devices usually use low-power wireless communication protocols, and Compared with traditional 5G devices, the energy consumption is lower, but when running for a long time or transmitting large amounts of data, its cumulative energy consumption may still be high. Therefore, in the task scheduling process, it is necessary to accurately evaluate the data transmission requirements of the task and ensure that the task can be assigned to the device with lower communication energy consumption first to reduce the overall energy consumption of the system. Secondly, the energy consumption in the formula and They are used to calculate the uplink and downlink data transmission time of the task respectively. and They represent the amount of output data and input data of the task, respectively, which determines the amount of data that needs to be transmitted between the RedCap device and the remote server or other computing nodes. Represents the RedCap device at time The available communication bandwidth determines the rate at which data can be transmitted, and Represents the packet error rate, which is used to characterize the reliability of the wireless channel. Since data transmission in a wireless communication environment has a certain packet loss rate and needs to be retransmitted, this formula introduces a correction factor in the bandwidth term. , which is used to compensate for the impact of packet errors on transmission time. When the packet error rate is high, the actual data throughput is reduced, resulting in an increase in data transmission time, thereby increasing communication energy consumption. Therefore, during task scheduling, the system should give priority to devices with lower packet error rates for task scheduling to reduce the energy consumption caused by data retransmission.

[0052] Embodiment 9: In step 5, the optimal task allocation solution is solved using the Lagrangian relaxation method based on the multi-dimensional decision matrix through the following formula: ; in, is the objective function; Assign the matrix to the task, Indicates the task Assigned to RedCap module nodes ,otherwise ; For RedCap module nodes The Lagrange multiplier constrained by the computing resource capacity, is the Lagrange multiplier vector of resource capacity constraints; For the task The Lagrange multiplier of the distribution constraint, is the Lagrange multiplier vector of the allocation constraint; is the total number of tasks; is the total number of RedCap module nodes; the final scheduling decision is obtained by solving the following optimization problem: ; in, and are the Lagrange multiplier vector of the optimal computing resource capacity constraint and the Lagrange multiplier vector of the optimal allocation constraint solved iteratively by the subgradient method, respectively; is the optimal task assignment matrix.

[0053] Specifically, the core goal of the method of the present invention is to maximize the execution benefit of the task, that is, to ensure that the task is assigned to the most suitable RedCap module node so that it can be completed in the shortest time and to reduce computing energy consumption and communication overhead as much as possible. When constructing the optimization target, the execution benefit obtained when each task is executed on different nodes is first considered. Execution benefit is a comprehensive indicator that measures the execution efficiency of a task on a specific node. It not only depends on the priority of the task, but also is affected by the execution time, computing energy consumption, communication energy consumption and device load balancing of the task. In order to enable task scheduling to adapt to the characteristics of different tasks, the present invention introduces a task execution benefit function in the optimization target, so that the task scheme with high priority, low execution delay and low energy consumption obtains a greater weight in the optimization process. However, task scheduling must not only pursue the highest execution benefit, but also must meet the constraints of computing resources. Since the computing power of each RedCap module node is limited, when assigning tasks, a certain node cannot be allowed to undertake tasks that exceed its computing power, otherwise it will cause task execution failure or computing resource congestion. Therefore, the present invention introduces computing resource capacity constraints to ensure that the task computing requirements assigned to each node will not exceed its currently available computing power. This constraint is related to the number of computational operations, computational complexity coefficient and available computing power of the node, ensuring that the task scheduling system always considers the constraints of computing resources when optimizing the task allocation scheme, thereby avoiding the unbalanced use of computing resources. In order to enable the optimization problem to be solved under the constraint conditions, the present invention adopts the Lagrange multiplier method to introduce the computing resource constraints into the optimization objective in a relaxed form, so that the task load distribution can be dynamically adjusted during the optimization process to ensure the reasonable use of computing resources.

[0054] In addition to the limitation of computing resources, task scheduling must also meet the constraint of unique task allocation, that is, each task must and can only be assigned to a RedCap module node for execution. This constraint is crucial in the task scheduling process, because if a task is not assigned to any node, the task execution fails, and if the task is assigned to multiple nodes, it will cause a waste of computing resources. Therefore, the method of the present invention introduces the constraint of unique task allocation in the optimization target, and relaxes it through the Lagrange multiplier method, so that task scheduling can dynamically adjust the task allocation scheme during the optimization process, ensuring that each task can obtain a unique computing node for execution, while maximizing the execution efficiency of the overall task. When solving the optimization problem, the present invention adopts the subgradient method for iterative solution. Since the task scheduling problem is a discrete optimization problem, the traditional integer programming method has a high computational complexity and is difficult to efficiently solve under the limited computing power of the RedCap module. The advantage of the Lagrange relaxation method is that it can convert the original discrete optimization problem into a continuous optimization problem by introducing Lagrange multipliers, making the optimization calculation more efficient. The sub-gradient method gradually approaches the optimal solution by continuously adjusting the value of the Lagrange multiplier, and finally converges to the optimal task allocation scheme that satisfies the unique allocation constraints of computing resources and tasks. During the optimization process, the Lagrange multiplier will be dynamically adjusted according to the computing requirements of the current task, the available computing resources of the node, and the priority of the task, so as to ensure that the optimization result can not only ensure the reasonable allocation of computing resources, but also make the task scheduling strategy meet the overall optimization goal of the system. Finally, through the construction of the optimization target, the introduction of the Lagrange relaxation method, and the solution of the sub-gradient method, the method of the present invention can realize the optimal scheduling scheme of the task under the premise of ensuring the reasonable allocation of computing resources. Compared with the traditional task scheduling method, the advantage of this method is that it can optimize multiple targets at the same time, so that the task scheduling scheme can not only ensure the real-time performance of task execution, but also can balance computing resources, communication resources, and energy consumption. In addition, since this method adopts a dynamic optimization strategy, it can be dynamically adjusted according to the computing state, network environment, and priority of the task of the RedCap module, so as to ensure that the task scheduling strategy can adapt to different computing environments and improve the overall throughput and computing efficiency of the system.

[0055] Although the specific embodiments of the present invention are described above, it should be understood by those skilled in the art that these specific embodiments are only illustrative, and those skilled in the art may omit, replace, and change the details of the above methods and systems in various ways without departing from the principles and essence of the present invention. For example, merging the above method steps so as to perform substantially the same functions in substantially the same manner to achieve substantially the same results is within the scope of the present invention. Therefore, the scope of the present invention is limited only by the appended claims.

Claims

1. A dynamic weighted edge computing scheduling method based on 5G RedCap module, characterized in that: The method comprises: Step 1: Perform node resource perception on each RedCap module node in the network and calculate the effective computing capacity of each RedCap module node at each moment; Step 2: By analyzing the dynamic characteristics of the RedCap link of each RedCap module node in the network, extract the RedCap link feature vector of each RedCap module node; Step 3: Model the resource requirements of each task and estimate the execution time on different RedCap module nodes; Step 4: Construct a multi-dimensional decision matrix to quantify the execution benefits of each task on each RedCap module node; Step 5: Based on the multi-dimensional decision matrix, use the Lagrangian relaxation method to solve the optimal task allocation solution.

2. The dynamic weighted edge computing scheduling method based on the 5G RedCap module according to claim 1, characterized in that: In step 1, the effective computing capacity of each RedCap module node at each moment is calculated using the following formula: ； in, For RedCap module nodes At the moment Effective computing power; For RedCap module nodes CPU frequency; For RedCap module nodes At the moment The CPU utilization rate is in the range of (0,1). For RedCap module nodes At the moment The system overhead factor represents the proportion of computing resources consumed by system maintenance; For RedCap module nodes At the moment Temperature; The temperature threshold is the value above which computing performance begins to decline significantly. The value range is 70 to 85. is the temperature scaling factor, which controls the steepness of the effect of temperature on performance; For RedCap module nodes At the moment Available memory; For RedCap module nodes The total memory capacity.

3. The dynamic weighted edge computing scheduling method based on the 5G RedCap module as claimed in claim 2, characterized in that: In step 2, the RedCap link feature vector of each RedCap module node is extracted using the following formula: ； in, For RedCap module nodes At the moment The RedCap link feature vector contains 4 dimensions; For RedCap module nodes At the moment The effective bandwidth of For RedCap module nodes At the moment jitter; is the reference jitter value; For RedCap module nodes At the moment The round trip delay of is the maximum round-trip delay acceptable to the system; For RedCap module nodes At the moment The physical movement speed of the RedCap module node is 0. is the reference physical movement speed value; For RedCap module nodes At the moment signal-to-noise ratio; is the system noise floor signal-to-noise ratio; For RedCap module nodes At the moment The channel quality index ranges from 0 to 15; For RedCap module nodes At the moment Packet error rate; For RedCap module nodes At the moment The channel switching frequency; is the packet error impact coefficient, ranging from 5 to 100; is the influence coefficient of error packet switching frequency, ranging from 2 to 50.

4. The dynamic weighted edge computing scheduling method based on the 5G RedCap module as claimed in claim 3, characterized in that: In step 3, the execution time on different RedCap module nodes is estimated using the following formula: ； in, For the task In the RedCap module node At the moment Execution time; For the task The number of computational operations required; For the task The computational complexity coefficient of ; For the task The amount of input data; For the task The amount of output data.

5. The dynamic weighted edge computing scheduling method based on the 5G RedCap module as claimed in claim 4, characterized in that: Task The computational complexity coefficient of Calculated using the following formula: ； in, For the task The basic complexity coefficient is determined by the asymptotic complexity of the task: for O(1) complexity tasks, ; For O(logn) complexity tasks, ; For O(n) complexity tasks, ; For O(nlogn) complexity tasks, ; For O(n²) complexity tasks, ; For O(n³) complexity tasks, ; For the task The working set memory size, that is, the amount of memory frequently accessed during task execution; For the task Total memory requirements; is the memory impact factor, with a value ranging from 0.2 to 0.5, reflecting the impact of memory access on computing performance; For the task The data dependency of is 0 or 1, which indicates the dependency between computing units within the task: 0 indicates complete parallelism and no dependency; 1 indicates complete serialization and strong dependency; is the dependency impact factor, with a value range of 0.5 to 1.5; is the instruction mode coefficient, which is determined according to the instruction characteristics of the task: for computationally intensive tasks, The value range is 0.8 to 1.0; for memory-intensive tasks, The value range is 1.2 to 1.

5.

6. The dynamic weighted edge computing scheduling method based on the 5G RedCap module as claimed in claim 5, characterized in that: In step 4, a multi-dimensional decision matrix is ​​constructed using the following formula to quantify the execution benefit of each task on each RedCap module node: ； in, For the task In the RedCap module node At the moment implementation benefits; For the task The priority weight of For the task Deadline constraints; For the task In the RedCap module node Calculate the energy required; For the task In the RedCap module node At the moment Energy required for communication; For RedCap module nodes At the moment The load variance reflects the load stability; is the link quality feature vector The impact factor of dimension.

7. The dynamic weighted edge computing scheduling method based on the 5G RedCap module as claimed in claim 5, characterized in that: Calculated using the following formula: ； in, For RedCap module nodes Energy efficiency coefficient.

8. The dynamic weighted edge computing scheduling method based on the 5G RedCap module as claimed in claim 7, characterized in that: Calculated using the following formula: in, For RedCap module nodes Energy efficiency coefficient; For RedCap module nodes The transmission power of For RedCap module nodes The received power.

9. The dynamic weighted edge computing scheduling method based on the 5G RedCap module as claimed in claim 8, characterized in that: In step 5, the optimal task allocation solution is solved using the Lagrangian relaxation method based on the multi-dimensional decision matrix using the following formula: ； in, is the objective function; Assign the matrix to the task, Indicates the task Assigned to RedCap module nodes ,otherwise ; For RedCap module nodes The Lagrange multiplier constrained by the computing resource capacity, is the Lagrange multiplier vector of resource capacity constraints; For the task The Lagrange multiplier of the distribution constraint, is the Lagrange multiplier vector of the allocation constraint; is the total number of tasks; is the total number of RedCap module nodes; the final scheduling decision is obtained by solving the following optimization problem: ； in, and are the Lagrange multiplier vector of the optimal computing resource capacity constraint and the Lagrange multiplier vector of the optimal allocation constraint solved iteratively by the subgradient method, respectively; is the optimal task assignment matrix.

Images