Task load scheduling optimization method and system for distributed network nodes

Abstract

The invention provides a task load scheduling optimization method and system for distributed network nodes, and the method comprises the steps: determining a task transmission path between the distributed network nodes according to historical task congestion data and a real-time link load prediction result; adjusting a modulation coding strategy in a task transmission process according to the channel quality parameter on the task transmission path to obtain an adjusted modulation coding strategy; according to an energy consumption demand corresponding to the adjusted modulation coding strategy, combining load distribution of each communication link and each node on the task transmission path to generate a voltage frequency adjustment instruction corresponding to the distributed network node; performing collaborative optimization on the task transmission path, the channel quality parameter and the voltage frequency regulation instruction to generate a task load scheduling strategy; according to the invention, through collaborative optimization of dynamic path selection, adaptive modulation coding adjustment and precise energy consumption regulation, the transmission reliability, energy efficiency balance and system stability of the distributed network in a high dynamic environment are improved.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular, to a method and system for optimizing task load scheduling of distributed network nodes. Background Art

[0002] With the rapid development of technologies such as edge computing clusters and Internet of Things terminal networking, task load scheduling among distributed network nodes needs to meet the dual requirements of high-reliability transmission and dynamic energy efficiency optimization to address challenges such as channel environment interference, strict real-time resource allocation requirements, and balancing energy efficiency and stability.

[0003] In response to the above challenges, a method for step-by-step optimization of paths and energy consumption based on fixed-period channel detection is currently widely used. Existing methods use periodic channel detection methods, static energy consumption configuration, and independent decision-making mechanisms, which have some defects. For example, the periodic detection of existing methods causes path switching to lag behind the actual degradation process, easily leading to task accumulation or transmission interruption; static energy consumption configuration ignores the power consumption fluctuations caused by modulation strategy adjustment, easily resulting in insufficient power supply for high-load nodes or waste of resources for low-load nodes; the separate path and energy consumption decision-making modules lack cross-layer parameter linkage, making it difficult to handle complex scenarios of channel interference and sudden traffic increases, and the global resource utilization efficiency is low, etc. Summary of the Invention

[0004] The present invention provides a method and system for optimizing task load scheduling of distributed network nodes to solve problems such as insufficient foresight in the prior art, which easily leads to task accumulation or transmission interruption, rigid energy efficiency, lack of coordination, and low global resource utilization efficiency.

[0005] In a first aspect, the present invention provides a method for optimizing task load scheduling of distributed network nodes, including:

[0006] Determining a task transmission path among distributed network nodes according to historical task congestion data and real-time link load prediction results;

[0007] Adjusting the modulation and coding strategy during task transmission according to the channel quality parameters on the task transmission path to obtain an adjusted modulation and coding strategy, where the channel quality parameters include the link quality attenuation rate;

[0008] Generating a voltage and frequency adjustment instruction corresponding to the distributed network node according to the energy consumption requirement corresponding to the adjusted modulation and coding strategy, in combination with the load distribution of each communication link and each node on the task transmission path;

[0009] Cooperatively optimizing the task transmission path, the channel quality parameters, and the voltage and frequency adjustment instruction to generate a task load scheduling strategy.

[0010] Optionally, the channel quality parameter further includes external interference intensity, link quality attenuation rate, and bandwidth utilization rate. Co-optimize the task transmission path, the channel quality parameter, and the voltage-frequency adjustment instruction to generate a task load scheduling strategy, including:

[0011] Select multiple target paths from the task transmission path whose link quality attenuation values are lower than a preset threshold. The link quality attenuation value is obtained by comparing the relative attenuation amplitude and the time-dimensional attenuation trend of the channel quality parameter and the historical reference signal quality parameter;

[0012] According to the load capacity, external interference intensity, and bandwidth utilization rate of the target path, calculate the comprehensive priority scores corresponding to the multiple target paths, and select the target path with the highest priority from the comprehensive priority scores as the updated task transmission path;

[0013] Based on the task distribution ratio of the updated task transmission path, combined with the power consumption increment per single node corresponding to the adjusted modulation and coding strategy, generate the expected total power consumption of each node;

[0014] According to the expected total power consumption and the power supply capacity of the distributed network nodes, select the optimal voltage-frequency combination from the voltage-frequency adjustment instructions to obtain the updated energy consumption configuration parameters;

[0015] Generate a task load scheduling strategy according to the mapping relationship between the updated task transmission path and the updated energy consumption configuration parameters.

[0016] Optionally, according to the expected total power consumption and the power supply capacity of the distributed network nodes, select the optimal voltage-frequency combination from the voltage-frequency adjustment instructions to obtain the updated energy consumption configuration parameters, including:

[0017] Construct a power supply capacity database corresponding to each distributed network node. The power supply capacity database includes the rated output power, voltage adjustment range, and frequency adjustment range corresponding to each node;

[0018] Select candidate voltage-frequency combinations that meet the preset conditions from the voltage-frequency adjustment instructions. The preset conditions include that the instantaneous power consumption is lower than the rated output power of the corresponding node and not less than the expected total power consumption, the voltage value is within the voltage adjustment range, and the frequency value is within the frequency adjustment range;

[0019] Select the combination with the smallest product of the voltage value and the frequency value from the candidate voltage-frequency combinations as the optimal voltage-frequency combination;

[0020] Convert the voltage value and frequency value of the optimal voltage-frequency combination into the control instruction format to generate the updated energy consumption configuration parameters.

[0021] Optionally, according to the historical task congestion data and the real-time link load prediction results, determine the task transmission path between distributed network nodes, including:

[0022] According to the topological structure of the distributed network, determine the candidate transmission paths between all distributed network nodes, extract the historical task congestion data of each link in the candidate transmission paths to calculate the historical congestion coefficient;

[0023] Perform a weighted sum of the historical congestion coefficient, the remaining bandwidth in the real-time link load prediction results, and the estimated traffic growth rate according to preset weights to generate the initial path score for each candidate transmission path;

[0024] Select the path that satisfies the real-time link load capacity constraint and has the highest initial path score from the candidate transmission paths as the initial task transmission path;

[0025] According to the dynamic changes in the real-time link load prediction results, periodically recalculate the initial path score of the initial task transmission path to obtain the updated score. If there is a path whose updated score exceeds the initial path score, then use the path with the highest updated score as the finally determined task transmission path.

[0026] Optionally, according to the channel quality parameters on the task transmission path, adjust the modulation and coding strategy during the task transmission process to obtain the adjusted modulation and coding strategy. The channel quality parameters include the link quality attenuation rate, including:

[0027] Obtain the channel quality parameters of each link in the task transmission path. The channel quality parameters include the link quality attenuation rate, the signal strength deviation value, and the multi-band interference component;

[0028] Divide the initial adjustment level according to the superposition result of the short-term change trend and the long-term change trend of the link quality attenuation rate;

[0029] Based on the cumulative effect of the signal strength deviation value and the dominant frequency band distribution of the multi-band interference component, perform dynamic compensation on the initial adjustment level to obtain the dynamically compensated adjustment level and the compensation coefficient;

[0030] According to the dynamically compensated adjustment level and the compensation coefficient, match the strategy parameters in the predefined modulation and coding strategy set to generate the adjusted modulation and coding strategy. The strategy parameters include the modulation order, the coding redundancy ratio, and the frequency band avoidance rule.

[0031] Optionally, dynamically compensating the initial adjustment level based on the cumulative effect of the signal strength deviation value and the dominant frequency band distribution of the multi-band interference component to obtain the dynamically compensated adjustment level and compensation coefficient includes:

[0032] Counting the number of times the signal strength deviation value exceeds a preset deviation threshold within a preset continuous period to generate a cumulative number of exceeding thresholds;

[0033] Extracting the interference intensity ratio of each frequency band from the multi-band interference components, marking a preset number of frequency bands sorted by the interference intensity ratio as dominant interference frequency bands, and calculating the overlap between the dominant interference frequency bands and the frequency bands specified in the predefined modulation and coding strategy;

[0034] Generate a dynamic compensation parameter according to the cumulative number of exceeding the standard and the degree of overlap;

[0035] The dynamic compensation parameter and the initial adjustment level are superimposed and calculated to generate an adjustment level and a compensation coefficient after dynamic compensation.

[0036] Optionally, generating a voltage and frequency adjustment instruction corresponding to the distributed network node according to the energy consumption requirement corresponding to the adjusted modulation and coding strategy and in combination with the load distribution of each communication link and each node on the task transmission path includes:

[0037] Calculating the expected energy consumption requirement of each node based on the RF power consumption increment of each node in the adjusted modulation and coding strategy and the load weight of each node on the task transmission path;

[0038] The energy consumption value corresponding to the expected energy consumption demand and the basic power consumption value of each node in the idle state are weighted and accumulated to generate the total energy consumption demand of each node;

[0039] According to the total energy consumption requirement and the rated output power of each node, an initial voltage-frequency combination that meets the task processing requirements is matched, and the voltage and frequency parameters of the initial voltage-frequency combination are encoded into standardized node control instructions to generate the voltage-frequency adjustment instructions.

[0040] In a second aspect, the present invention provides a task load scheduling optimization system for distributed network nodes, comprising:

[0041] The determination module is used to determine the task transmission path between distributed network nodes based on historical task congestion data and real-time link load prediction results;

[0042] an adjustment module, configured to adjust a modulation and coding strategy during task transmission according to a channel quality parameter on the task transmission path to obtain an adjusted modulation and coding strategy, wherein the channel quality parameter includes a link quality attenuation rate;

[0043] A generation module, configured to generate a voltage-frequency regulation instruction corresponding to the distributed network node according to the energy consumption requirement corresponding to the adjusted modulation and coding strategy and in combination with the load distribution of each communication link and each node on the task transmission path;

[0044] An optimization module, configured to jointly optimize the task transmission path, the channel quality parameter, and the voltage-frequency regulation instruction to generate a task load scheduling strategy.

[0045] In a third aspect, the present invention provides a computing device, including a processor and a memory, where a computer program is stored in the memory, and the processor is configured to run the computer program to execute the task load scheduling optimization method for a distributed network node according to any one of the first aspects.

[0046] In a fourth aspect, the present invention provides a computer storage medium, on which computer program instructions are stored, and when the computer program instructions are executed by a processor, the task load scheduling optimization method for a distributed network node according to any one of the first aspects is implemented.

[0047] Compared with the prior art, the technical solution adopted in the embodiment of the present invention has at least the following technical advantages:

[0048] In the present invention, according to the historical task congestion data and the real-time link load prediction result, the task transmission path between distributed network nodes is determined;

[0049] According to the channel quality parameter on the task transmission path, the modulation and coding strategy in the task transmission process is adjusted to obtain an adjusted modulation and coding strategy, where the channel quality parameter includes a link quality attenuation rate; according to the energy consumption requirement corresponding to the adjusted modulation and coding strategy and in combination with the load distribution of each communication link and each node on the task transmission path, a voltage-frequency regulation instruction corresponding to the distributed network node is generated;

[0050] The task transmission path, the channel quality parameter, and the voltage-frequency regulation instruction are jointly optimized to generate a task load scheduling strategy.

[0051] The technical solution provided by the present invention screens out highly reliable transmission paths by integrating historical congestion rules and real-time load prediction, reducing the risk of transmission delay caused by link congestion; at the same time, based on the dynamic perception of the channel quality attenuation rate, it adaptively adjusts the signal modulation and coding redundancy ratio to improve the data transmission success rate in harsh channel environments; combines the energy consumption characteristics of the modulation strategy and the node load distribution to generate energy consumption regulation instructions that match the task requirements, avoiding node overload or resource idleness; through the real-time linkage of path re-planning and energy consumption instructions, realizing the global dynamic adaptation of network resources is a technical logic with extremely high research value that the existing technology does not consider. After the above technologies are implemented, the system stability in high-load scenarios can be guaranteed. Among them, the task transmission path is dynamically optimized based on channel quality parameters. By screening low-attenuation paths and calculating the comprehensive priority, an updated task transmission path is generated. Combining the expected total power consumption of the node with the power supply capacity, the optimal voltage-frequency combination is matched, and finally a path-energy consumption configuration mapping relationship is established to generate a task load scheduling strategy.

[0052] These aspects or other aspects of the present invention will be more clearly understood in the following description of the embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0054] Figure 1 It is a flowchart of a method for optimizing task load scheduling of a distributed network node provided by an embodiment of the present invention;

[0055] Figure 2 It is a schematic structural diagram of a system for optimizing task load scheduling of a distributed network node provided by an embodiment of the present invention;

[0056] Figure 3 It is a schematic structural diagram of a computing device provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0057] In order to enable those skilled in the art to better understand the solution of the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention.

[0058] In some of the processes described in the specification, claims, and the above-mentioned drawings of the present invention, a plurality of operations appear in a specific order. However, it should be clearly understood that these operations may not be executed in the order in which they appear herein or may be executed in parallel. The operation numbers such as 101, 102, etc. are only used to distinguish different operations, and the numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions such as "first" and "second" herein are used to distinguish different messages, devices, modules, etc., do not represent a sequence, and do not limit that "first" and "second" are of different types.

[0059] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work belong to the scope of protection of the present invention.

[0060] Figure 1 The flowchart of a method for optimizing task load scheduling of a distributed network node is provided for an embodiment of the present invention. As Figure 1 shown, the method includes:

[0061] In view of the compound problem of dynamic channel degradation and energy efficiency load imbalance in a distributed network scenario (such as continuous attenuation of link quality caused by mobile occlusion of edge nodes and sudden increase in node power consumption due to burst traffic), traditional solutions have defects such as path switching lagging behind channel attenuation and energy efficiency strategies being disconnected from task loads due to the adoption of static path planning and discrete energy consumption regulation mechanisms. This solution realizes dynamic path optimization by integrating historical congestion rules and real-time load prediction, adaptively adjusts the modulation and coding strategy by cross-cycle perception of the link quality attenuation rate, and generates voltage and frequency regulation instructions based on the energy consumption characteristics of the modulation strategy and the spatio-temporal correlation of node load distribution. Finally, through the collaborative optimization mechanism of multi-dimensional parameters (path, channel, energy consumption), it breaks through the contradiction between local optimization and global imbalance in the prior art and realizes the joint improvement of transmission reliability and system energy efficiency in a high-dynamic environment. Based on this, the present invention provides a method for optimizing task load scheduling of a distributed network node, as Figure 1 [[ID=,14]]including:

[0062] Step 101: Determine the task transmission path between distributed network nodes according to historical task congestion data and real-time link load prediction results;

[0063] In this step, the historical task congestion data refers to the information on congestion events of each link recorded during the transmission of past tasks by distributed network nodes, including the congestion occurrence time, duration, location of the congested link, and task queue length, etc., which is used to analyze the periodic patterns of link carrying capacity and the distribution of bottleneck nodes. The real-time link load prediction result refers to the estimated data on the load status of each link within a future set time window generated by combining real-time monitoring of network traffic, node processing capacity, and task arrival rate with a prediction model, including the remaining bandwidth, expected queuing delay, and link availability probability.

[0064] In the embodiment of the present invention, the link failure frequency and task queue peak value in the historical congestion data are combined with the remaining bandwidth predicted in real time to construct a link reliability scoring model; the links with historical congestion frequency exceeding the threshold or the remaining bandwidth predicted in real time being lower than the task demand are excluded to generate a candidate path set; the reliability score of the candidate path is calculated using this scoring model (such as calculating according to the weight ratio of historical stability weight 60% and real-time load weight 40%), and the path with the highest comprehensive score is selected as the task transmission path, and this task transmission path is initial and needs to be further optimized and updated.

[0065] For example, in an edge computing scenario, a certain node needs to transmit real-time monitoring data to a central server. The system analyzes the historical congestion data and finds that the congestion probability of link A reaches 70% during peak hours, while the real-time link load prediction shows that the remaining bandwidth of link B is sufficient. Therefore, link B is selected as the task transmission path to avoid task accumulation.

[0066] Step 102: Adjust the modulation and coding strategy during the task transmission according to the channel quality parameters on the task transmission path to obtain the adjusted modulation and coding strategy, where the channel quality parameters include the link quality attenuation rate;

[0067] In this step, the channel quality parameters are a set of indicators characterizing the transmission performance of a wireless communication link, including link signal strength, frequency band utilization rate, external interference strength, link quality attenuation rate, etc. In this step, it specifically refers to using the link quality attenuation rate in the channel quality parameters for adjustment. The modulation and coding strategy refers to the combination rule of the signal modulation method and error correction coding ratio adopted during the transmission of task data, which directly affects the transmission rate, anti-interference ability, and radio frequency power consumption. The link quality attenuation rate is a dynamic parameter calculated by continuously monitoring the change of link signal strength, reflecting the deterioration trend of the channel quality (such as the continuous attenuation of the signal caused by environmental occlusion or device movement), and is used to trigger the downscale protection of the modulation and coding strategy or path switching.

[0068] In the embodiments of the present invention, the wireless channel state awareness technology is utilized to collect the link signal strength on the task transmission path in real time, so as to calculate the attenuation rate (the difference between the current link signal strength and the historical reference value is divided by the monitoring time); the modulation and coding strategy during the task transmission is determined according to the attenuation rate. The specific process is as follows: if the attenuation rate < 0.05 / second, high-order modulation (such as 64QAM) and low redundancy coding (1 / 3) are adopted; if 0.05 / second ≤ attenuation rate < 0.1 / second, medium-order modulation (16QAM) and medium redundancy coding (1 / 2) are adopted; if the attenuation rate ≥ 0.1 / second, low-order modulation (QPSK) and high redundancy coding (2 / 3) are adopted; if the signal strength deviation value exceeds the threshold for 3 consecutive cycles, the modulation strategy is forced to be downgraded; finally, the adjusted modulation and coding strategy is obtained.

[0069] For example, during the movement of the UAV node, it is monitored that the link quality attenuation rate suddenly increases from 3% / second to 8% / second (due to flying into an occluded area). The system automatically switches the modulation mode from 64QAM to 16QAM and increases the coding redundancy ratio to 1 / 2 to ensure the complete transmission of data.

[0070] Step 103: Generate a voltage-frequency regulation instruction corresponding to the distributed network node according to the energy consumption requirement corresponding to the adjusted modulation and coding strategy, in combination with the load distribution of each communication link and each node on the task transmission path;

[0071] In this step, the energy consumption requirement refers to the increased radio frequency power consumption of the distributed network node due to the adjustment of the modulation and coding strategy (for example, high-order modulation increases by 2W). The communication link refers to the physical or logical communication channel established between two nodes in the distributed network. Each link has independent bandwidth, delay, and load status attributes. The load distribution of each node refers to the task processing amount currently borne by each node in the distributed network and the traffic ratio of its associated link, including quantization indexes such as CPU / GPU utilization rate, memory occupancy rate, and the length of the task queue to be processed. The voltage-frequency regulation instruction refers to the standardized control instruction generated according to the node energy consumption requirement and the load status, including the target voltage value, working frequency value, and effective time window, and is used to dynamically adjust the power supply parameters of the node's computing unit to achieve energy efficiency optimization.

[0072] In the embodiments of the present invention, first, the load weight is calculated, which is obtained by multiplying the initial weight (the number of links connected to the node) by the average value of the remaining bandwidth ratios of each link to get the load weight; then the energy consumption requirement is calculated, that is, according to the basic power consumption (the energy consumption value when the node is in the idle state) × 1.2 + the load weight × the radio frequency power consumption increment, select the voltage-frequency combination that meets the energy consumption requirement from the predefined configuration table, encode it into a control instruction, and generate a voltage-frequency regulation instruction corresponding to the distributed network node.

[0073] For example, in a smart factory, when a distributed network node processes a sudden control instruction, the load weight increases from 1.5 to 2.3, and the total energy consumption demand increases from 15W to 22W. The system matches a combination of voltage 1.0V / frequency 2.0GHz (power consumption = 1.0 × 2.0 = 2.0W), generates and issues a voltage-frequency adjustment instruction to dynamically increase the power supply parameters.

[0074] Step 104: Co-optimize the task transmission path, the channel quality parameter, and the voltage-frequency adjustment instruction to generate a task load scheduling strategy;

[0075] In this step, the task load scheduling strategy refers to a global optimization plan generated by integrating the path planning results, the channel state, and the energy consumption regulation instruction, which clarifies the task distribution path, the transmission parameter configuration, and the node power consumption limit rules to ensure the coordinated goal of high-reliability transmission and low-energy consumption operation.

[0076] In the embodiment of the present invention, if the deterioration of the channel quality parameter triggers a path switch, the energy consumption configuration of the associated nodes is synchronously updated; the node load distribution of the switched path and the voltage-frequency instruction are bound according to a time window to generate a task load scheduling strategy, so as to distribute and schedule instructions according to the task start timing according to this strategy to ensure the timing consistency of path switching and power consumption adjustment.

[0077] For example, in an urban traffic monitoring network, due to a thunderstorm, the quality attenuation rate of the main path exceeds the threshold. The system switches to the backup path and reduces the modulation order of the nodes on the new path. At the same time, it increases the voltage to 1.1V according to the load distribution to generate a complete task load scheduling strategy to ensure the backhaul of all-weather monitoring data.

[0078] The embodiment of the present invention systematically solves the technical barriers such as path switching lag, energy efficiency and load fragmentation, and global coordination lack in the traditional scheme through the path selection integrating historical laws and real-time prediction, the dynamic modulation adjustment driven by the channel attenuation rate, the generation of the task load scheduling strategy with strong load-energy consumption correlation, and the multi-parameter co-optimization mechanism, significantly improving the transmission reliability, energy efficiency balance, and system robustness of the distributed network in a highly dynamic environment, and is applicable to complex scenarios such as edge computing and industrial Internet of Things.

[0079] The present invention provides a specific embodiment. In step 104, co-optimize the task transmission path, the channel quality parameter, and the voltage-frequency adjustment instruction to generate a task load scheduling strategy, which specifically includes the following steps:

[0080] Step 401: Select multiple target paths from the task transmission path whose link quality attenuation values are lower than a preset threshold, and the link quality attenuation value is obtained by comparing the relative attenuation amplitude and the time dimension attenuation trend of the channel quality parameter and the historical reference signal quality parameter;

[0081] In this step, the link quality attenuation value refers to a dynamic index calculated by comparing the difference between the current channel quality parameters (such as signal strength, bit error rate) and the historical reference signal quality parameters, and combining the attenuation trend per unit time (such as the percentage of signal strength decrease per second), which is used to quantify the degree of link quality deterioration. The preset threshold refers to the upper limit value of the tolerance for link quality deterioration preset according to the network reliability requirements. For example, when the link quality attenuation value exceeds 15%, it is determined as a high-risk path.

[0082] In the embodiment of the present invention, the current signal quality parameters of the target link are monitored in real time through the physical layer channel state information, including the received signal strength and the noise power. At the same time, the historical reference signal quality parameters of the link in the stable state stored in advance are retrieved; the current signal quality parameters are compared with the historical reference parameters to calculate the instantaneous relative attenuation amplitude, and the continuous decrease rate of the signal quality in the recent several transmission cycles is statistically analyzed to obtain the attenuation trend in the time dimension; finally, the instantaneous attenuation amplitude and the time attenuation trend are fused and calculated according to the preset weight ratio to generate the link quality attenuation value for failure risk determination; the links with the link quality attenuation value exceeding the preset threshold (such as 15%) are excluded, and the remaining paths are retained as the target path set.

[0083] Step 402: Calculate the comprehensive priority scores corresponding to multiple target paths according to the load capacity, external interference intensity, and frequency band utilization rate of the target paths, and select the target path with the highest priority from the comprehensive priority scores as the updated task transmission path;

[0084] In this step, the load capacity refers to the maximum task traffic that a single link can carry per unit time, which is jointly determined by the link bandwidth, node processing capacity, and protocol overhead, and is usually expressed in tasks per second or bit rate. The external interference intensity refers to the signal interference power intensity from non-target transmission devices (such as adjacent communication systems, electromagnetic noise), which is measured through spectrum analysis or received signal strength indication, and the unit is decibel milliwatt (dBm). The frequency band utilization rate refers to the effective utilization efficiency of the frequency band occupied by the current link, and the calculation formula is: (actual transmitted data volume / theoretical maximum carrying data volume) × 100%, which reflects the rationality of spectrum resource utilization. The comprehensive priority score refers to the score generated by weighted calculation of the load capacity (weight 40%), external interference intensity (weight 30%), and frequency band utilization rate (weight 30%) of the target path, which is used to quantify the pros and cons of the comprehensive transmission performance of the path.

[0085] In the embodiments of the present invention, the load capacity of the target path, the external interference intensity (taking the reciprocal), and the bandwidth utilization rate are respectively mapped to the interval of 0-1; the comprehensive priority score of each path is calculated according to the preset weights (load capacity 40%, external interference intensity 30%, bandwidth utilization rate 30%); the target paths are arranged from high to low according to the comprehensive priority score, and the path with the highest comprehensive priority score is selected as the updated task transmission path.

[0086] Step 403: Based on the task distribution ratio of the updated task transmission path, combined with the power consumption increment of a single node corresponding to the adjusted modulation and coding strategy, generate the expected total power consumption of each node;

[0087] In this step, the task distribution ratio refers to the proportion of the task processing volume allocated according to the load weights (such as the number of connection links and the remaining bandwidth ratio) of each node in the updated task transmission path. For example, node A undertakes 30% of the total task. The power consumption increment of a single node refers to the increased value of the radio frequency unit power consumption of a single node caused by adjusting the modulation and coding strategy (such as switching from QPSK to 16QAM), which is obtained by matching through a predefined power consumption comparison table. The expected total power consumption refers to the power consumption prediction value of each node during the task processing cycle, and the calculation formula is the base power consumption (the energy consumption value when the node is in the idle state)+(the power consumption increment of a single node×the task distribution ratio×the task duration coefficient).

[0088] In the embodiments of the present invention, first, according to the load capacity and priority weight of each link in the updated task transmission path, calculate the proportion of the task volume that each distributed network node needs to undertake; among them, the load capacity is determined by measuring the available bandwidth and current traffic data of each link, and the priority weight is comprehensively calculated based on the node processing ability, task urgency, and path hop count; finally, through normalization processing, the proportion of the task volume of each node is converted into a standardized task distribution ratio for subsequent power consumption calculation and resource allocation; extract the radio frequency power consumption increment associated with each node in the adjusted modulation and coding strategy, and the increment is obtained by matching the signal modulation method and transmission power level corresponding to the node from a preset power consumption comparison table, and perform incremental superposition calculation on multi-link transmission nodes; finally, calculate the expected total power consumption of each node, and the expected total power consumption = base power consumption (idle state)+∑(the increment of a single node×the task distribution ratio×the task duration coefficient), where ∑ represents for each node in the network, calculate the product result of (the increment of a single node×the task distribution ratio×the task duration coefficient), and then add up the product results of all nodes.

[0089] Step 404: According to the expected total power consumption and the power supply capacity of the distributed network node, select the optimal voltage-frequency combination from the voltage-frequency regulation instructions to obtain the updated energy consumption configuration parameters;

[0090] In this step, the power supply capacity of the distributed network node refers to the sustainable power supply power range supported by the node hardware. For example, the rated output power of a certain node is 100W, the adjustable voltage range is 0.8V - 1.2V, and the frequency range is 1GHz - 3GHz. The optimal voltage-frequency combination refers to the configuration combination with the smallest product of the voltage value and the frequency value (such as 1.0V × 2.0GHz = 2.0W) under the condition of meeting the expected total power consumption and power supply capacity constraints, so as to achieve the optimal energy efficiency. The updated energy consumption configuration parameters refer to the set of the node operating voltage, frequency, and the effective time window after optimization. For example, the voltage is 1.1V, the frequency is 2.5GHz, and the effective time window is 50ms before the task starts.

[0091] In the embodiment of the present invention, first, a power supply capacity database of the distributed network node is established, which stores the maximum output power and the adjustable voltage-frequency range of each node power module; then, according to the expected total power consumption, a matching query is performed in the power supply capacity database to screen out all voltage-frequency combinations that can meet the power consumption requirement; the optimal voltage-frequency combination with the smallest product value of the voltage and the frequency (such as 1.0V × 2.0GHz = 2.0W) is selected from the qualified combinations as the optimal operating point; finally, the voltage value and the frequency value of the optimal operating point are encoded into a standardized control instruction to obtain an updated voltage-frequency adjustment instruction, and the corresponding energy consumption configuration parameters of each node are synchronously updated by using it to generate updated energy consumption configuration parameters.

[0092] Step 405: Generate a task load scheduling strategy according to the mapping relationship between the updated task transmission path and the updated energy consumption configuration parameters.

[0093] In this step, the mapping relationship refers to the binding rule between each node in the updated task transmission path and its corresponding energy consumption configuration parameters. For example, nodes N1 - N3 in path X are respectively associated with the configurations of voltages 1.1V / 1.0V / 1.2V.

[0094] In the embodiment of the present invention, according to the topological connection order and the task distribution ratio of each node in the updated task transmission path, the network address of each node is associated and bound with its corresponding energy consumption configuration parameters to form a complete mapping relationship table of the path and the energy consumption, which is used as the core control basis for the task load scheduling strategy.

[0095] The embodiments of the present invention systematically solve the problems of low transmission reliability, insufficient resource utilization, poor dynamic adaptability, etc. caused by the separation of path planning and energy consumption control in traditional solutions through a collaborative mechanism of dynamic link attenuation evaluation, multi-dimensional path optimization, accurate prediction of load power consumption, optimal energy efficiency configuration, and global strategy synchronization. Specifically, the channel quality attenuation trend is linked with energy consumption demand to avoid task interruption caused by path switching delay; node power consumption is adjusted based on the real-time task distribution ratio to suppress local overload or resource idleness; and end-to-end energy efficiency and stability collaborative optimization in high-interference and high-dynamic scenarios are achieved through the mapping relationship between path and energy consumption and timing synchronization control.

[0096] The present invention provides a specific embodiment, step 404, selecting an optimal voltage-frequency combination from the voltage-frequency adjustment instructions based on the expected total power consumption and the power supply capability of the distributed network node to obtain updated energy consumption configuration parameters, specifically includes the following steps:

[0097] Step 411: constructing a power supply capability database corresponding to each distributed network node, wherein the power supply capability database includes the rated output power, voltage adjustment range, and frequency adjustment range corresponding to each node;

[0098] In this step, the power supply capability database refers to a dedicated data set that stores the power supply capability parameters of each distributed network node. The rated output power refers to the maximum output power of the node power module in a continuous operating state, which is used to limit the upper limit of the power consumption of the voltage-frequency combination (for example, if the rated power of a node is 120W, the instantaneous power consumption of all candidate combinations must be ≤120W). The voltage adjustment range refers to the operating voltage adjustment range allowed by the node hardware. For example, a node supports a voltage configuration of 0.8V to 1.2V. Exceeding this range will trigger the overvoltage / undervoltage protection mechanism. The frequency adjustment range refers to the operating frequency adjustment range supported by the node processor or communication unit. For example, the frequency of a node can be dynamically adjusted between 1GHz and 3GHz. The higher the frequency, the faster the computing rate, but the higher the power consumption.

[0099] In an embodiment of the present invention, the rated output power, voltage adjustment range and frequency adjustment range are extracted from the node hardware specification document. For example, the rated power of a node is 100W, the voltage supports 0.8V-1.2V, and the frequency supports 1GHz-3GHz. The parameters are classified by node ID and stored as a structured power supply capability database, each record of which contains the node ID, rated power, voltage range and frequency range. If the node hardware is upgraded or replaced, the corresponding parameters in the power supply capability database are synchronously updated.

[0100] Step 412: Select candidate voltage-frequency combinations that meet the preset conditions from the voltage-frequency adjustment instructions. The preset conditions include that the instantaneous power consumption is lower than the rated output power of the corresponding node and not less than the expected total power consumption, the voltage value is within the voltage adjustment range, and the frequency value is within the frequency adjustment range;

[0101] In this step, the instantaneous power consumption refers to the real-time power consumption value when the voltage-frequency combination is running. The calculation formula is voltage value × frequency value × load factor (the load factor is determined by the current task volume).

[0102] In the embodiment of the present invention, all possible combinations are traversed from the predefined voltage-frequency adjustment instruction set, and the combinations that meet the preset conditions are selected. The preset conditions are: instantaneous power consumption ≤ node rated output power (hardware safety constraint); instantaneous power consumption ≥ expected total power consumption (task demand constraint); voltage value is within the voltage adjustment range; frequency value is within the frequency adjustment range; the combinations that meet all conditions are retained to form candidate voltage-frequency combinations.

[0103] Step 413: Select the combination with the smallest product of the voltage value and the frequency value from the candidate voltage-frequency combinations as the optimal voltage-frequency combination;

[0104] In the embodiment of the present invention, calculate the voltage value × frequency value for each candidate combination. For example, 1.0V × 2.0GHz = 2.0; sort them in ascending order according to the product value, and select the combination corresponding to the minimum value; if the products of multiple combinations are the same, preferentially select the combination with a lower voltage.

[0105] Step 414: Convert the voltage value and frequency value of the optimal voltage-frequency combination into the control instruction format to generate updated energy consumption configuration parameters.

[0106] In this step, the updated energy consumption configuration parameters refer to the set of node working parameters generated after optimization and screening, including the target voltage value, target frequency value, configuration effective time window, and parameter encoding format.

[0107] In an embodiment of the present invention, first, an instruction structure is predefined. Specifically, a node control instruction format including a voltage setting field, a frequency setting field, and a check field is predefined. The voltage setting field stores the integer and decimal parts of the voltage value. The frequency setting field stores the integer part of the frequency value. The check field is generated by performing a bitwise exclusive OR operation on the binary data of the voltage and frequency fields. Then, numerical conversion and encoding are performed, that is, the voltage value of the optimal voltage-frequency combination is split into integer and decimal parts according to a preset precision and filled into the voltage setting field. The frequency value is rounded to an integer multiple unit and filled into the frequency setting field. The check field is generated based on the values of the two fields. Finally, instruction encapsulation and binding are performed, that is, the fields are concatenated into a complete control instruction in a preset order, and an independent instruction is generated for each node, and the updated energy consumption configuration parameters are formed by binding through the node identifier.

[0108] The embodiment of the present invention accurately restricts the hardware security boundary through the power supply capacity database, ensures task requirements and hardware adaptation through multi-condition screening, optimizes energy efficiency through product minimization selection, and ensures executability through standardized instruction encapsulation, systematically solving problems such as ignoring hardware limitations in energy efficiency optimization and inability of static configuration to dynamically adapt to load changes in traditional solutions, significantly improving the energy efficiency balance, hardware security, and controllability of the scheduling strategy in high-load scenarios, and is applicable to fields such as industrial automation and edge computing that have strict requirements for real-time performance and reliability.

[0109] The present invention provides a specific embodiment. Step 101: Determine the task transmission path between distributed network nodes according to historical task congestion data and real-time link load prediction results, which specifically includes the following steps:

[0110] Step 111: According to the topological structure of the distributed network, determine the candidate transmission paths between all distributed network nodes, extract the historical task congestion data of each link in the candidate transmission paths, and calculate the historical congestion coefficient;

[0111] In this step, the topological structure of the distributed network refers to the physical or logical connection relationship between all nodes in the distributed network, including the location distribution of nodes (such as servers, terminal devices), the connection method of links (communication channels), and the network hierarchical structure (such as star, mesh, tree topologies). The historical congestion coefficient refers to the link congestion severity index calculated based on historical task congestion data.

[0112] In the embodiments of the present invention, all possible candidate transmission paths are generated according to the connection relationships among distributed network nodes (such as nodes [A, B, C], nodes [A, D, C]); for each link in each candidate path, its congestion frequency (such as occurring 3 times a week) and average congestion duration (such as 30 minutes each time) are extracted from the historical database, and the historical congestion coefficient is calculated. The calculation formula is historical congestion coefficient = congestion frequency × weight 1 + average congestion duration × weight 2.

[0113] Step 112: Perform weighted summation on the historical congestion coefficient, the remaining bandwidth in the real-time link load prediction result, and the estimated traffic growth rate according to preset weights to generate an initial path score for each candidate transmission path;

[0114] In this step, the remaining bandwidth refers to the available bandwidth that is not currently occupied by the link and is determined by the difference between the maximum bandwidth that the link can carry and the currently used bandwidth. The estimated traffic growth rate refers to the predicted increase ratio of the link traffic within a set future time window based on the real-time monitored traffic change trend, for example, generated by a sliding window statistic or an exponential smoothing algorithm. The preset weights refer to the proportion distribution of the user-defined historical congestion coefficient, remaining bandwidth, and estimated traffic growth rate in the path score

[0115] In the embodiments of the present invention, the remaining bandwidth of each link is obtained through a traffic monitoring module (such as link X has 60% remaining bandwidth); the estimated traffic growth rate is generated based on a time series prediction model (such as a 10% increase in the next period); according to the calculation formula, the initial path score = historical congestion coefficient × 40% + remaining bandwidth × 35% - estimated traffic growth rate × 25% (negative weights suppress high-growth paths), and the initial path score of each candidate transmission path is generated.

[0116] Step 113: Select the path that satisfies the real-time link load capacity constraint and has the highest initial path score from the candidate transmission paths as the initial task transmission path;

[0117] In this step, the real-time link load capacity constraint refers to the maximum task traffic that the link can carry at the current moment and is determined by the difference between the maximum traffic that the link can carry and the current traffic. If the task demand exceeds the load capacity, the link is determined to be an unavailable link.

[0118] In the embodiments of the present invention, select from the candidate transmission paths those that satisfy the real-time link load capacity constraint and exclude the candidate transmission paths whose remaining capacity does not meet the task requirements (such as link Y has 15 Mbps remaining and the task requires 20 Mbps, then exclude it); sort the remaining transmission paths in descending order according to the initial path score, and select the path with the highest score as the initial task transmission path.

[0119] Step 114: According to the dynamic changes of the real-time link load prediction results, periodically recalculate the initial path score of the initial task transmission path to obtain the updated score. If there is a path whose updated score exceeds the initial path score, then use the path with the highest updated score as the finally determined task transmission path.

[0120] In this step, the periodic recalculation means re-executing the path score calculation at fixed time intervals (such as every 30 seconds) or event triggers (such as traffic mutation) to ensure that the path selection is updated synchronously with the real-time network status.

[0121] In the embodiment of the present invention, according to the dynamic changes of the real-time link load prediction results, obtain the real-time load data again at set intervals (such as 30 seconds), update the remaining bandwidth and the estimated traffic growth rate; recalculate the scores of all candidate paths according to the rules of Step 112 to generate the updated score. If there is a candidate path whose updated score exceeds the current path score and meets the load constraint, then use it as the path with the highest score, that is, this path with the highest score is the determined task transmission path between distributed network nodes.

[0122] For example, in the scenario of transmitting cargo dispatching instructions in a smart logistics park, the system first parses the network topology to determine three candidate paths: [W, A, D], [W, B, D], and [W, C, D]. By analyzing the historical task congestion data, it is found that the congestion coefficient of link WA in the candidate path [W, A, D] is the highest (0.8), while the congestion coefficient of link WC in the candidate path [W, C, D] is moderate (0.5). Combining the real-time link load prediction results, it is concluded that the candidate path [W, C, D] becomes the initial task transmission path with an initial path score of 0.65 (calculation method: historical congestion coefficient × 40% + remaining bandwidth × 35% - estimated growth rate × 25%) with 85% remaining bandwidth and only 8% estimated growth rate. However, after 30 seconds of monitoring, it is found that due to sudden traffic, the remaining bandwidth of the candidate path [W, C, D] drops sharply to 40% and the estimated growth rate rises to 20%, and the score drops to 0.25; at the same time, the remaining bandwidth of the candidate path [W, A, D] increases to 75% and the estimated growth rate drops to 3%, and the initial path score rises to 0.62. Based on this, the system automatically switches the task transmission path to [W, A, D] to ensure that the dispatching instructions are always transmitted efficiently and stably in the dynamically changing network environment.

[0123] In the embodiment of the present invention, through a path scoring mechanism that fuses historical congestion rules and real-time load prediction, a periodic dynamic re-evaluation, and a constraint-driven path switching strategy, technical barriers such as response lag, ignoring traffic growth trends, and the disconnection between historical experience and real-time status in traditional static path planning are systematically solved, significantly improving the reliability, resource utilization efficiency, and adaptive ability of task transmission in a highly dynamic network environment, and being applicable to scenarios such as intelligent logistics and vehicle-to-everything (V2X) that require real-time response to complex environmental changes.

[0124] The present invention provides a specific embodiment. In step 102, according to the channel quality parameters on the task transmission path, the modulation and coding strategy during the task transmission is adjusted to obtain an adjusted modulation and coding strategy. The channel quality parameters include the link quality attenuation rate, and specifically include the following steps:

[0125] Step 201: Obtain the channel quality parameters of each link in the task transmission path. The channel quality parameters include the link quality attenuation rate, the signal strength deviation value, and the multi-band interference component.

[0126] In this step, the signal strength deviation value refers to the absolute difference between the current signal strength and the historical reference signal strength. For example, if the reference strength is -80 dBm and the current strength is -90 dBm, the deviation value is 10 dB. The multi-band interference component refers to the distribution data of the interference signal strength in different frequency bands. For example, the interference power values detected in frequency bands such as 2.4 GHz and 5 GHz (such as the interference intensity in the 2.4 GHz band is -85 dBm and in the 5 GHz band is -90 dBm).

[0127] In the embodiment of the present invention, to obtain the channel quality parameters of each link in the task transmission path, for example, the signal strength and interference spectrum data of each link are collected in real time through a spectrum analyzer, and the reference signal strength stored in the historical database (such as the stable signal strength at the initial link establishment) is retrieved; according to the formula, (current signal strength - historical reference strength) / monitoring time, the link quality attenuation rate is calculated; the signal strength deviation value is the absolute difference between the current signal strength and the historical reference strength; the multi-band interference component is the distribution data of the external interference intensity in different frequency bands.

[0128] Step 202: Divide the initial adjustment level according to the superposition result of the short-term change trend and the long-term change trend of the link quality attenuation rate.

[0129] In this step, the short-term change trend of the link quality attenuation rate refers to the average change rate of the link quality attenuation rate within the recently set time window (such as 10 seconds); the long-term change trend refers to the overall change direction (such as continuous deterioration or gradual recovery) of the link quality attenuation rate within a longer period (such as 10 minutes); the superposition result refers to the comprehensive trend evaluation value obtained by weighted fusion of the short-term change trend and the long-term change trend according to a preset weight (such as 60% for the short term and 40% for the long term).

[0130] In an embodiment of the present invention, for example, the short-term change trend is calculated based on the average attenuation rate per second within the most recent 10 seconds; the long-term change trend is calculated based on the average attenuation rate per second within the past 1 minute; the difference between the short-term change trend and the long-term change trend is calculated (such as the short-term trend is 5% / second, the long-term trend is 3% / second, and the difference is 2% / second) to divide the initial adjustment level. The specific division process includes generating a high-risk level mark indicating that significant order reduction modulation is required when the difference between the short-term change trend and the long-term trend exceeds a first threshold (such as the difference ≥ 2% / second); generating a medium-risk level mark indicating that moderate order reduction is required when the difference is between the first threshold and the second threshold (such as 1% / second ≤ difference < 2% / second); generating a low-risk level mark indicating that modulation needs to be maintained or upgraded when the weighted difference is lower than the second threshold (difference < 1% / second).

[0131] Step 203: Based on the cumulative effect of the signal strength deviation value and the dominant frequency band distribution of the multi-band interference components, dynamically compensate the initial adjustment level to obtain the dynamically compensated adjustment level and the compensation coefficient;

[0132] In this step, the cumulative effect refers to the cumulative value of the number of times the signal strength deviation value exceeds a preset threshold within a continuous plurality of monitoring periods. For example, if the deviation value exceeds 15 dB in 5 consecutive periods, the cumulative effect value is 5. The dominant frequency band distribution refers to the set of frequency bands with the top set number (such as the top 3) in terms of interference intensity in the multi-band interference components and their intensity ratios. For example, 2.4 GHz (ratio 50%), 5 GHz (ratio 30%), 1.8 GHz (ratio 20%). The dynamically compensated adjustment level is a level label (such as high, medium, low risk levels) characterizing the conservativeness of the modulation and coding strategy. The compensation coefficient refers to the increased value of the redundancy ratio calculated based on the interference frequency band coincidence degree or the cumulative effect (such as increasing the coding redundancy by 10%).

[0133] In the embodiments of the present invention, the specific process of dynamic compensation is as follows: if the signal strength deviation value exceeds the preset deviation threshold within a continuous set period, the risk marker level of the initial adjustment level is increased; if the coincidence degree between the dominant frequency band of the multi-band interference component and the frequency band of the current modulation and coding strategy exceeds the preset ratio, the compensation coefficient of the redundancy coding ratio is increased; finally, the adjusted level and the compensation coefficient after dynamic compensation are obtained. For example, the number of times the signal strength deviation value continuously exceeds the threshold (such as 3 dB) (such as 3 times exceeding the standard in the most recent 5 periods); if the number of times exceeding the standard ≥ 3 times, the initial adjustment level is raised by one level (such as medium risk is raised to high risk); the top 2 frequency bands with the highest proportion in the multi-band interference component are extracted (such as 2.4 GHz and 5 GHz); if the coincidence degree between the dominant frequency band and the frequency band used by the current modulation strategy ≥ 50%, the compensation coefficient 0.2 is increased, and the result of raising the level is associated with the compensation coefficient to generate the final adjusted level, that is, the adjusted level after dynamic compensation (such as the high risk level plus the compensation coefficient 0.2, and the final adjusted level is obtained as enhanced high risk).

[0134] Step 204: According to the adjusted level and the compensation coefficient after dynamic compensation, match the policy parameters in the predefined modulation and coding strategy set to generate an adjusted modulation and coding strategy, where the policy parameters include modulation order, coding redundancy ratio, and frequency band avoidance rule.

[0135] In this step, the modulation order refers to the signal modulation method (such as QPSK, 16QAM); the coding redundancy ratio refers to the ratio of the error correction code to the total data length (such as 1 / 3, 1 / 2); the frequency band avoidance rule refers to avoiding the frequency band overlapping with the dominant interference frequency band (such as when the interference is concentrated in 2.4 GHz, switching to the 5 GHz frequency band for transmission).

[0136] In the embodiments of the present invention, a basic strategy is selected from the predefined policy library according to the adjusted level after dynamic compensation (such as the high risk level matches QPSK + 2 / 3 coding redundancy ratio); the compensation coefficient is superimposed to adjust the coding redundancy ratio of the basic strategy (such as the original 2 / 3 coding redundancy ratio + 0.2 compensation coefficient, which is changed to 5 / 6 coding redundancy ratio); if the coincidence degree of the dominant frequency band ≥ 50%, a rule for disabling this frequency band is added to this strategy; the modulation order, coding redundancy ratio, and frequency band rule are integrated to generate an adjustable and issuable modulation and coding strategy.

[0137] The embodiments of the present invention systematically solve the technical barriers such as rigid modulation strategies, lagging interference avoidance, and one-sided parameter adjustment in traditional solutions through a multi-cycle perception of the channel attenuation trend, a cross-layer compensation mechanism for signal deviation accumulation and interference frequency band analysis, and dynamic policy matching rules, significantly improving the data transmission reliability, spectrum utilization efficiency, and system adaptability in high-interference and high-dynamic scenarios, and being applicable to complex wireless environments such as unmanned aerial vehicle communication and emergency communication.

[0138] The present invention provides a specific embodiment. In step 203, based on the cumulative effect of the signal strength deviation value and the dominant frequency band distribution of the multi-band interference components, dynamic compensation is performed on the initial adjustment level to obtain the adjusted level and compensation coefficient after dynamic compensation, which specifically includes the following steps:

[0139] Step 211: Count the number of times the signal strength deviation value exceeds a preset deviation threshold within a preset continuous period to generate a cumulative over-standard count;

[0140] In this step, the preset continuous period refers to the pre-set number of consecutive monitoring time windows, which is used to count the continuous over-standard situation of the signal strength deviation value. For example, it is set to 5 consecutive periods (each period is 1 second), then it is necessary to continuously monitor whether the deviation value exceeds the standard within 5 seconds. The preset deviation threshold refers to the upper limit value of the signal strength deviation tolerance pre-set according to the network reliability requirements. For example, it is allowed that the signal strength deviation does not exceed 5 dB, and if it exceeds this value, it is determined to be over-standard. The cumulative over-standard count refers to the cumulative number of times the signal strength deviation value exceeds the preset deviation threshold within the continuous monitoring period, which is used to quantify the continuous deterioration degree of the signal quality.

[0141] In the embodiment of the present invention, the preset deviation threshold is set according to the network reliability requirements (such as signal strength deviation ≥ 5 dB is over-standard); the continuous monitoring period window is set (such as the last 5 periods); the signal strength deviation values of each period within the window are traversed. If the deviation value of a certain period ≥ the threshold, then the count +1; the cumulative over-standard count is output (for example, there are 3 over-standard cases in the last 5 periods).

[0142] Step 212: Extract the interference intensity ratio of each frequency band from the multi-band interference components, mark the preset number of frequency bands with the sorted interference intensity ratio as the dominant interference frequency bands, and calculate the coincidence degree between the dominant interference frequency bands and the specified frequency bands in the predefined modulation and coding strategy;

[0143] In this step, the preset number of frequency bands refers to the top N frequency bands (such as the first 2 frequency bands) with the highest interference intensity ratio selected from the multi-band interference components, which is used to characterize the frequency domain distribution characteristics of the main interference sources. The interference intensity ratio refers to the ratio of the single-band interference intensity to the total interference intensity of all frequency bands. The coincidence degree refers to the overlapping ratio between the dominant interference frequency bands and the specified frequency bands in the predefined modulation and coding strategy. For example, the dominant interference frequency bands are 2.4 GHz and 5 GHz, and the strategy-specified frequency band is 5 GHz, then the coincidence degree is 50%. The specified frequency band refers to the frequency band range clearly used for task transmission in the predefined modulation and coding strategy (such as 2.4 GHz - 2.4835 GHz), which is usually set based on spectrum license or anti-interference requirements.

[0144] In the embodiment of the present invention, sort the interference intensity ratios of each frequency band from high to low (for example, frequency band A accounts for 60%, frequency band B accounts for 30%, and frequency band C accounts for 10%); select the top N frequency bands (for example, select frequency bands A and B) as the dominant interference frequency bands; count the number of overlaps between the dominant interference frequency bands and the specified frequency bands in the modulation and coding strategy defined according to the current state (for example, this strategy uses specified frequency bands A and C, and the dominant frequency bands are A and B, then the number of overlaps is 1), and calculate the coincidence degree according to the formula: coincidence degree = (number of overlaps / total number of dominant frequency bands) × 100% (for example, 1 / 2 × 100% = 50%).

[0145] Step 213: Generate a dynamic compensation parameter according to the cumulative number of exceedances and the coincidence degree;

[0146] In the embodiment of the present invention, if the cumulative number of exceedances ≥ the set threshold (for example, 3 times), then raise the initial adjustment level by one level (for example, raise the medium risk to high risk); if the cumulative number of exceedances ≥ a threshold higher than the set threshold (for example, 5 times), additionally superimpose an incremental compensation coefficient of 0.1; calculate the dynamic compensation parameter according to the formula. The specific calculation process is as follows. First, calculate the basic compensation coefficient, that is, the product of the coincidence degree and the weight (for example, coincidence degree 50% × 0.4 = 0.2); if both the cumulative number of exceedances ≥ 3 times and the coincidence degree ≥ 50% are satisfied, then calculate the total compensation coefficient (dynamic compensation parameter), that is, the sum of the basic coefficient and the incremental compensation coefficient (for example, 0.2 + 0.1 = 0.3).

[0147] Step 214: Perform a superposition calculation on the dynamic compensation parameter and the initial adjustment level to generate an adjusted level and a compensation coefficient after dynamic compensation.

[0148] In the embodiment of the present invention, combine the initial adjustment level (for example, medium risk) with the promotion level (+1 level) to generate an adjusted level after dynamic compensation (high risk); associate the dynamic compensation parameter (for example, 0.3) with the adjusted level after dynamic compensation to form a final adjusted level of high risk + compensation 0.3.

[0149] The embodiment of the present invention solves the problems of one-sided adjustment of the modulation strategy and lag in interference avoidance in the traditional solution through a joint compensation mechanism of signal degradation cumulative effect perception and interference frequency band coincidence degree analysis, significantly improves the strategy response speed and anti-interference ability in high-interference scenarios, effectively avoids data packet loss and communication interruption caused by continuous signal degradation or frequency band conflict, and is applicable to scenarios with strong interference and high reliability requirements such as smart grids and industrial Internet of Things.

[0150] The present invention provides a specific embodiment. In step 301, according to the energy consumption requirements corresponding to the adjusted modulation and coding strategy, and in combination with the load distribution of each communication link and each node on the task transmission path, a voltage-frequency regulation instruction corresponding to the distributed network node is generated, which specifically includes the following steps:

[0151] Step 311: According to the radio frequency power consumption increment of each node in the adjusted modulation and coding strategy, and in combination with the load weights of each node on the task transmission path, calculate the expected energy consumption requirements of each node;

[0152] In this step, the radio frequency power consumption increment refers to the increased value of the radio frequency unit power consumption caused by the distributed network node adjusting the modulation and coding strategy (such as increasing the modulation order or increasing the coding redundancy), and is obtained by matching the current modulation strategy with a predefined power consumption comparison table. For example, when the node switches from QPSK modulation to 16QAM modulation, the radio frequency power consumption may increase by 2W. The load weight refers to the proportion of the task processing volume borne by the node in the task transmission path. The initial load distribution weight is generated by counting the number of direct connection links of each node in the task transmission path, and then the initial load distribution weight is dynamically corrected according to the real-time remaining bandwidth ratio of each link. The expected energy consumption requirement refers to the dynamic power consumption prediction value of the node within the task processing cycle, which is the product value of the load weight, the radio frequency power consumption increment, and the task duration coefficient (this coefficient is dynamically generated by the historical execution records stored locally by the node and the real-time task monitoring module, without manual configuration, and can adapt to the energy consumption requirement changes of different service scenarios).

[0153] In the embodiment of the present invention, the current modulation order (such as 16QAM) and the coding redundancy ratio (such as 1 / 2) are extracted from the adjusted modulation and coding strategy; the predefined power consumption comparison table is queried to match the corresponding radio frequency power consumption increment (for example, 16QAM corresponds to an increment of 1.5W); the number of links connected to the node is counted as the initial weight (such as the node is connected to 3 links, the initial weight = 3), and in combination with the real-time remaining bandwidth ratio of each link (such as link A remaining 60% and link B remaining 40%), the corrected load weight is calculated (such as 3×(0.6 + 0.4) / 2 = 1.5); finally, according to the formula, the expected energy consumption requirement = load weight × radio frequency power consumption increment × task duration coefficient (for example, 1.5×1.5×1.2 = 2.7W), and the expected energy consumption requirement is calculated.

[0154] Step 312: Perform a weighted accumulation calculation on the energy consumption value corresponding to the expected energy consumption requirement and the basic power consumption value of each node in the idle state to generate the total energy consumption requirement of each node;

[0155] In an embodiment of the present invention, the base power consumption value (e.g., 10W) in the idle state is read from the node hardware manual; according to the preset weight distribution, the total energy consumption requirements of each node are calculated. For example, if the base power consumption weight is set to 1.2 (a fixed value reflecting the necessary power consumption of the basic circuit) and the expected energy consumption requirement weight is 1.0 (adjusted according to the dynamic task ratio), then the total energy consumption requirement = base power consumption value × 1.2 + energy consumption value corresponding to the expected energy consumption requirement × 1.0 (e.g., 10 × 1.2 + 2.7 × 1.0 = 14.7W).

[0156] In this step, the total energy consumption requirement refers to the overall power consumption requirement for the node to complete the task, including the base power consumption and the dynamic increment

[0157] Step 313: According to the total energy consumption requirement and the rated output power of each node, an initial voltage-frequency combination that meets the task processing requirements is matched, and the voltage and frequency parameters of the initial voltage-frequency combination are encoded into a standardized node control instruction to generate the voltage-frequency regulation instruction.

[0158] In an embodiment of the present invention, first, according to the total energy consumption requirement and the rated output power of the node, an initial combination that meets the following conditions is selected from the predefined voltage-frequency configuration table: 1. The instantaneous power consumption of the combination is not lower than the total energy consumption requirement; 2. The voltage value is within the adjustable range supported by the node; 3. The frequency value is within the adjustable range supported by the node. Then, the voltage and frequency parameters of the qualified initial voltage-frequency combination are converted into a standardized node control instruction format to directly generate the initial voltage-frequency regulation instruction

[0159] The embodiment of the present invention systematically solves the technical barriers in traditional energy consumption regulation, such as the rigidity of static strategies, ignoring the hardware safety boundary, and the disconnection between energy efficiency and load, through precise matching of radio frequency increments, dynamic correction of load weights, and screening of the optimal energy efficiency combination, significantly improving the energy efficiency balance and node operation reliability in high-load scenarios, and is applicable to scenarios such as industrial automation and edge computing that require real-time response to complex task changes.

[0160] Figure 2 The following is a schematic structural diagram of a task load scheduling optimization system for a distributed network node provided by an embodiment of the present invention, as Figure 2 shown, the system includes:

[0161] A determination module 21, configured to determine the task transmission path between distributed network nodes according to historical task congestion data and real-time link load prediction results;

[0162] An adjustment module 22, configured to adjust the modulation and coding strategy during the task transmission according to the channel quality parameters on the task transmission path to obtain the adjusted modulation and coding strategy, where the channel quality parameters include the link quality attenuation rate;

[0163] A generation module 23, configured to generate a voltage - frequency regulation instruction corresponding to the distributed network node according to the energy consumption requirement corresponding to the adjusted modulation and coding strategy, in combination with the load distribution of each communication link and each node on the task transmission path;

[0164] An optimization module 24, configured to co - optimize the task transmission path, the channel quality parameter, and the voltage - frequency regulation instruction to generate a task load scheduling strategy.

[0165] Figure 2 The task load scheduling optimization system of a distributed network node shown can execute Figure 1 The task load scheduling optimization method of a distributed network node described in the illustrated embodiment, and its implementation principle and technical effects will not be elaborated. For the task load scheduling optimization system of a distributed network node in the above - mentioned embodiment, the specific manners in which each module and unit perform operations have been described in detail in the embodiment related to the method, and will not be elaborated here.

[0166] In a possible design, Figure 2 The task load scheduling optimization system of a distributed network node in the illustrated embodiment can be implemented as a computing device, such as Figure 3 shown, and this computing device may include a storage component 31 and a processing component 32;

[0167] The storage component 31 stores one or more computer instructions, and among them, the one or more computer instructions are called and executed by the processing component 32.

[0168] The processing component 32 is configured to: determine the task transmission path between distributed network nodes according to historical task congestion data and real - time link load prediction results; adjust the modulation and coding strategy during the task transmission according to the channel quality parameter on the task transmission path to obtain an adjusted modulation and coding strategy, where the channel quality parameter includes the link quality attenuation rate; generate a voltage - frequency regulation instruction corresponding to the distributed network node according to the energy consumption requirement corresponding to the adjusted modulation and coding strategy, in combination with the load distribution of each communication link and each node on the task transmission path; co - optimize the task transmission path, the channel quality parameter, and the voltage - frequency regulation instruction to generate a task load scheduling strategy.

[0169] Among them, the processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above methods. Of course, the processing component may also be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components for executing the above methods.

[0170] The storage component 31 is configured to store various types of data to support the operation of the terminal. The storage component may be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0171] Of course, the computing device may also necessarily include other components, such as input / output interfaces, display components, communication components, etc.

[0172] The input / output interface provides an interface between the processing component and the peripheral interface module, and the above peripheral interface module may be an output device, an input device, etc.

[0173] The communication component is configured to facilitate communication between the computing device and other devices in a wired or wireless manner, etc.

[0174] Among them, the computing device may be a physical device or an elastic computing host provided by a cloud computing platform, etc. At this time, the computing device may refer to a cloud server, and the above processing component, storage component, etc. may be basic server resources leased or purchased from a cloud computing platform.

[0175] The embodiment of the present invention also provides a computer storage medium storing a computer program, and when the computer program is executed by a computer, it can implement the above Figure 1 task load scheduling optimization method for a distributed network node in the illustrated embodiment.

[0176] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the above-described systems, devices, and units can refer to the corresponding processes in the foregoing method embodiments and will not be described herein again.

[0177] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. A person of ordinary skill in the art can understand and implement it without creative effort.

[0178] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the essence of the above technical solution, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0179] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments or equivalently replace some of the technical features. However, these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of each embodiment of the present invention.

Claims

1. A task load scheduling optimization method for distributed network nodes, characterized in that, Including: Determine the task transmission path between distributed network nodes according to historical task congestion data and real-time link load prediction results; Adjust the modulation and coding strategy during the task transmission according to the channel quality parameters on the task transmission path to obtain the adjusted modulation and coding strategy, where the channel quality parameters include the link quality attenuation rate; Generate a voltage-frequency regulation instruction corresponding to the distributed network node by combining the energy consumption requirements corresponding to the adjusted modulation and coding strategy with the load distribution of each communication link and each node on the task transmission path; Cooperatively optimize the task transmission path, the channel quality parameters, and the voltage-frequency regulation instruction to generate a task load scheduling strategy.

2. The method according to claim 1, wherein The channel quality parameters further include external interference intensity, link quality attenuation rate, and bandwidth utilization rate. Cooperatively optimizing the task transmission path, the channel quality parameters, and the voltage-frequency regulation instruction to generate a task load scheduling strategy includes: Select multiple target paths from the task transmission path whose link quality attenuation values are lower than a preset threshold, where the link quality attenuation value is obtained by comparing the relative attenuation amplitude and the time-dimensional attenuation trend of the channel quality parameters with the historical reference signal quality parameters; Calculate the comprehensive priority scores corresponding to the multiple target paths according to the load capacity, external interference intensity, and bandwidth utilization rate of the target paths, and select the target path with the highest priority from the comprehensive priority scores as the updated task transmission path; Generate the expected total power consumption of each node by combining the task distribution ratio of the updated task transmission path with the power consumption increment per node corresponding to the adjusted modulation and coding strategy; Select the optimal voltage-frequency combination from the voltage-frequency regulation instructions according to the expected total power consumption and the power supply capacity of the distributed network node to obtain the updated energy consumption configuration parameters; Generate a task load scheduling strategy according to the mapping relationship between the updated task transmission path and the updated energy consumption configuration parameters.

3. The method according to claim 2, wherein Select the optimal voltage-frequency combination from the voltage-frequency regulation instructions according to the expected total power consumption and the power supply capacity of the distributed network node to obtain the updated energy consumption configuration parameters, including: Construct a power supply capacity database corresponding to each distributed network node, where the power supply capacity database includes the rated output power, voltage regulation range, and frequency regulation range corresponding to each node; Select candidate voltage-frequency combinations that meet preset conditions from the voltage-frequency regulation instructions, where the preset conditions include that the instantaneous power consumption is lower than the rated output power of the corresponding node and not less than the expected total power consumption, the voltage value is within the voltage regulation range, and the frequency value is within the frequency regulation range; Select the combination with the smallest product of the voltage value and the frequency value from the candidate voltage-frequency combinations as the optimal voltage-frequency combination; Convert the voltage value and the frequency value of the optimal voltage-frequency combination into the control instruction format; to generate the updated energy consumption configuration parameters.

4. The method according to claim 1, wherein Determine the task transmission path between distributed network nodes according to historical task congestion data and real-time link load prediction results, including: According to the topology of the distributed network, determine the candidate transmission paths between all distributed network nodes, extract the historical task congestion data of each link in the candidate transmission paths to calculate the historical congestion coefficient; Perform a weighted sum of the historical congestion coefficient, the remaining bandwidth in the real-time link load prediction result, and the estimated traffic growth rate according to preset weights to generate the initial path score for each candidate transmission path; Select the path that satisfies the real-time link load capacity constraint and has the highest initial path score from the candidate transmission paths as the initial task transmission path; According to the dynamic changes in the real-time link load prediction result, periodically recalculate the initial path score of the initial task transmission path to obtain the updated score. If there is a path with an updated score exceeding the initial path score, then use the path with the highest updated score as the finally determined task transmission path.

5. The method according to claim 1, characterized in that According to the channel quality parameters on the task transmission path, adjust the modulation and coding strategy during the task transmission to obtain the adjusted modulation and coding strategy. The channel quality parameters include the link quality attenuation rate, including: Obtain the channel quality parameters of each link in the task transmission path. The channel quality parameters include the link quality attenuation rate, the signal strength deviation value, and the multi-band interference component; Divide the initial adjustment level according to the superposition result of the short-term change trend and the long-term change trend of the link quality attenuation rate; Based on the cumulative effect of the signal strength deviation value and the dominant frequency band distribution of the multi-band interference component, perform dynamic compensation on the initial adjustment level to obtain the dynamically compensated adjustment level and the compensation coefficient; According to the dynamically compensated adjustment level and the compensation coefficient, match the policy parameters in the predefined modulation and coding strategy set to generate the adjusted modulation and coding strategy. The policy parameters include the modulation order, the coding redundancy ratio, and the frequency band avoidance rule.

6. The method according to claim 5, wherein Based on the cumulative effect of the signal strength deviation value and the dominant frequency band distribution of the multi-band interference component, perform dynamic compensation on the initial adjustment level to obtain the dynamically compensated adjustment level and the compensation coefficient, including: Count the number of times the signal strength deviation value exceeds the preset deviation threshold within a preset continuous period to generate the cumulative over-standard times; Extract the interference intensity ratio of each frequency band from the multi-band interference component, mark the preset number of frequency bands with the sorted interference intensity ratio as the dominant interference frequency bands, and calculate the coincidence degree between the dominant interference frequency bands and the specified frequency bands in the predefined modulation and coding strategy; Generate dynamic compensation parameters according to the cumulative over-standard times and the coincidence degree; Perform a superposition calculation on the dynamic compensation parameters and the initial adjustment level to generate the dynamically compensated adjustment level and the compensation coefficient.

7. The method according to claim 1, wherein According to the energy consumption requirements corresponding to the adjusted modulation and coding strategy, combined with the load distribution of each communication link and each node on the task transmission path, generate a voltage and frequency adjustment instruction corresponding to the distributed network node, including: According to the radio frequency power consumption increment of each node in the adjusted modulation and coding strategy, combined with the load weights of each node on the task transmission path, calculate the expected energy consumption requirements of each node; Perform a weighted cumulative calculation on the energy consumption value corresponding to the expected energy consumption demand and the basic power consumption values of each node in the idle state to generate the total energy consumption demand of each node; According to the total energy consumption demand and the rated output power of each node, match the initial voltage-frequency combination that meets the task processing requirements, and encode the voltage and frequency parameters of the initial voltage-frequency combination into a standardized node control instruction to generate the voltage-frequency adjustment instruction.

8. A task load scheduling optimization system for distributed network nodes, characterized in that, Comprising: A determination module, configured to determine the task transmission path between distributed network nodes according to historical task congestion data and real-time link load prediction results; An adjustment module, configured to adjust the modulation and coding strategy during the task transmission according to the channel quality parameters on the task transmission path to obtain the adjusted modulation and coding strategy, where the channel quality parameters include the link quality attenuation rate; A generation module, configured to generate a voltage-frequency adjustment instruction corresponding to the distributed network node according to the energy consumption demand corresponding to the adjusted modulation and coding strategy, in combination with the load distribution of each communication link and each node on the task transmission path; An optimization module, configured to perform collaborative optimization on the task transmission path, the channel quality parameters, and the voltage-frequency adjustment instruction to generate a task load scheduling strategy.

9. A computing device, characterized in that, Comprising a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a method for optimizing task load scheduling of a distributed network node according to any one of claims 1-7.

10. A computer storage medium, characterized in that, Stores a computer program, and when the computer program is executed by a computer, it implements a method for optimizing task load scheduling of a distributed network node according to any one of claims 17.

Images