Energy acquisition network-oriented double-layer dynamic information updating method and product

Abstract

The invention provides an energy harvesting network-oriented double-layer dynamic information updating method and product, and relates to the technical field of edge computing. According to the technical scheme provided by the embodiment of the invention, an internal and external double-layer dynamic transmission and power regulation mechanism of sensor group scheduling and intra-group sensor scheduling based on a KKT (Karush-Kuhn-Tucker) condition can be realized, and the system performance is remarkably improved on the premise of ensuring the quality of updated data. The strategy has the following advantages: (1) an accurate sensor selection mechanism: establishing a multi-dimensional evaluation model, comprehensively considering indexes such as sensor energy states, channel quality, data importance and the like, optimizing and selecting an optimal sensor combination through a KKT condition, and ensuring that the updating efficiency of a system is maximized; (2) an efficient power regulation strategy: dynamically regulating transmission power, adapting to channel conditions and real-time change of energy supply to realize optimal allocation of energy resources, and avoiding invalid energy consumption; and (3) the performance is obviously improved: the invalid updating times are effectively reduced, the updating efficiency is obviously improved, and the overall efficiency of the system is maximized while the data quality and the AoI requirement are ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of edge computing, and particularly to a two-layer dynamic information update method and product for an energy harvesting network. Background Art

[0002] With the rapid development of 5G communication, Internet of Things (IoT) and edge computing technologies, real-time application systems have been widely used in fields such as industrial automation, intelligent transportation and smart cities. These systems rely on data continuously generated by large-scale sensor networks for intelligent analysis and decision-making. However, the timeliness and quality of data directly affect the accuracy of the analysis results. Outdated or low-quality data may lead to decision-making biases, thus affecting the overall system performance. Therefore, ensuring the timeliness and quality in the data collection process has become a key issue for improving the accuracy of real-time decision-making.

[0003] To accurately characterize the timeliness of information in real-time applications, the academic community has proposed a new metric called Age of Information (AoI). AoI is defined as the time elapsed from data generation to reception at the destination. Different from traditional network performance metrics, AoI not only measures the data transmission delay, but also comprehensively considers the data generation frequency, and can more accurately describe the dynamic changes in the data generation interval. This metric is particularly important in highly dynamic real-time applications because it directly reflects the "freshness" of the information relied on by the system, thus affecting the accuracy of system decision-making.

[0004] Mobile edge computing effectively reduces the data transmission delay and improves the timeliness and quality of data by deploying computing resources at edge nodes close to the data sources. However, in practical applications, the bandwidth resources of wireless sensors are limited and often restricted by battery power. Frequent battery replacement increases the maintenance cost and may cause system interruptions. Although some source sensors adopt energy harvesting technology to collect energy from the environment to maintain device operation, the limitations of energy and bandwidth resources still pose great challenges to ensuring data timeliness and quality. Summary of the Invention

[0005] The present invention provides a two-layer dynamic information update method and product for an energy harvesting network to at least partially solve the above problems.

[0006] In a first aspect of the present invention, there is provided a two-layer dynamic information update method for an energy harvesting network, which is applied to an edge node. The edge node is used to schedule K sensor groups for information uploading. The K sensor groups include N sensors. The method includes: Traverse the K sensor groups from 1 to K according to the following steps: For the k-th sensor group traversed, traverse the n-th sensor belonging to the k-th sensor group, and calculate the power of the n-th sensor belonging to the k-th sensor group ; When the power of the n-th sensor belonging to the k-th sensor group meets the power constraint condition, regard the n-th sensor belonging to the k-th sensor group as a primary selected sensor, and determine the k-th sensor group as a primary selected sensor group; When multiple primary selected sensor groups are determined, with the goal of minimizing the transmission energy consumption of the scheduled sensors and the age of information of the information uploaded by the scheduled sensor groups, determine the scheduled sensor groups, the scheduled sensors within the scheduled sensor groups, and the transmission power of the scheduled sensors; Schedule the scheduled sensors to upload information at the transmission power.

[0007] Optionally, calculating the power of the n-th sensor belonging to the k-th sensor group , includes: According to the cumulative throughput debt of the k-th sensor group in time slot t , the channel gain during the transmission of the n-th sensor and the length of time slot t T d , calculate the power of the n-th sensor belonging to the k-th sensor group .

[0008] Optionally, after calculating the power of the n-th sensor belonging to the k-th sensor group , further includes: Verify whether the power of the n-th sensor belonging to the k-th sensor group meets the transmission power constraint of this sensor, and the transmission power constraint indicates that the transmission power of the scheduled sensor is not greater than the maximum power of the scheduled sensor and not greater than the power that the actual energy of the scheduled sensor can provide; When the transmission power constraint of this sensor is met, regard the n-th sensor belonging to the k-th sensor group as a primary selected sensor; When the k-th sensor group is a scheduled sensor group, determine the power of the n-th sensor of the k-th sensor group as the transmission power of the scheduled sensor.

[0009] Optionally, the method further includes: When the transmission power constraint of this sensor is not met, adjust the power of the n-th sensor of the k-th sensor group to the power boundary value of this sensor, and the power boundary value ; where represents the remaining energy of the nth sensor in the previous time slot t - 1, represents the energy collected by the nth sensor in time slot t, represents the basic energy consumption of the nth sensor; When the adjusted power is greater than zero, the nth sensor belonging to the kth sensor group is used as a primary selected sensor; When the kth sensor group is the scheduled sensor group, the power boundary value of the nth sensor in the kth sensor group is determined as the transmission power of the scheduled sensor; When the adjusted power is equal to zero, it is determined that the sensor does not meet the transmission power constraint of the sensor.

[0010] Optionally, with the goal of minimizing the transmission energy consumption of the scheduled sensor and the age of information of the information uploaded by the scheduled sensor group, determining the scheduled sensor group includes: According to the energy consumption of the information uploaded by the primary selected sensor group in time slot , the cumulative throughput debt of the primary selected sensor group in time slot t , the total amount of data uploaded by all primary selected sensors in the primary selected sensor group in time slot t , the importance weight parameter of the age of information of the primary selected sensor group , the age of information of the information transmitted by the primary selected sensor group in time slot t , calculate the target value of the primary selected sensor group , which is defined as ; ; Perform an ascending sort on the target values of multiple primary selected sensor groups , and select the first smaller value corresponding sensor groups as the scheduled sensor groups; Among them, represents the energy consumption of the nth sensor belonging to the kth sensor group in transmitting information at time sequence t , is the set of sensors belonging to sensor group ; ; represents the transmission rate of sensor ; , represents whether the nth sensor in the kth sensor group is a primary selected sensor; represents a non - negative control parameter used to balance the goal and throughput debt.

[0011] Optionally, the method further includes: Before the next round of scheduling, update the following parameter values: The remaining energy of each sensor at time slot t ; , indicating the maximum stored energy of the sensor; The energy consumption of the uploaded information of each sensor group at time slot ; ; The age of information of the information transmitted by each sensor group within time slot t ; The cumulative throughput debt of each sensor group within time slot t .

[0012] The second aspect of the present invention provides a two - layer dynamic information update device for an energy - harvesting network, which is applied to an edge node. The edge node is used to schedule K sensor groups for information uploading. The K sensor groups include N sensors. The device includes: A traversal module, configured to: traverse the K sensor groups from 1 to K according to the following steps: For the k - th sensor group traversed, traverse the n - th sensor belonging to the k - th sensor group, and calculate the power of the n - th sensor belonging to the k - th sensor group ; When the power of the n - th sensor belonging to the k - th sensor group meets the power constraint condition, regard the n - th sensor belonging to the k - th sensor group as a primary - selected sensor, and determine the k - th sensor group as a primary - selected sensor group; An optimization module, configured to, when multiple primary - selected sensor groups are determined, aim at minimizing the transmission energy consumption of the scheduled sensors and the age of information of the information uploaded by the scheduled sensor groups, and determine the scheduled sensor groups, the scheduled sensors within the scheduled sensor groups, and the transmission power of the scheduled sensors; A scheduling module, configured to schedule the scheduled sensors to upload information with the transmission power.

[0013] The third aspect of the present invention provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes, it implements the two - layer dynamic information update method for an energy - harvesting network as described in the first aspect of the present invention.

[0014] The fourth aspect of the present invention provides a computer - readable storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the two - layer dynamic information update method for an energy - harvesting network as described in the first aspect of the present invention.

[0015] The fifth aspect of the present invention provides a computer program product, including computer programs / instructions, which are implemented by a processor to perform the steps in the double-layer dynamic information update method for an energy harvesting network as described in the first aspect of the present invention.

[0016] The technical solution provided by the embodiments of the present invention can implement an internal and external double-layer dynamic transmission and power adjustment mechanism for sensor group scheduling and in-group sensor scheduling based on the KKT (Karush-Kuhn-Tucker) conditions, and significantly improve the system performance on the premise of ensuring the quality of updated data. This strategy has the following advantages: (1) Precise sensor selection mechanism: Establish a multi-dimensional evaluation model, comprehensively consider indicators such as sensor energy status, channel quality, and data importance, and optimize the selection of the optimal sensor combination through the KKT conditions to ensure the maximization of the system update efficiency; (2) Efficient power adjustment strategy: Dynamically adjust the transmission power to adapt to the real-time changes of channel conditions and energy supply, realize the optimal allocation of energy resources, and avoid ineffective energy consumption; (3) Significant performance improvement: Effectively reduce the number of ineffective updates, and significantly improve the update efficiency. While ensuring data quality and AoI requirements, maximize the overall system efficiency.

[0017] Through innovative algorithm design and system architecture, the present invention realizes the following comprehensive performance advantages: (1) Strong adaptability: It can dynamically adapt to fluctuations in energy supply and changes in channel conditions; (2) High resource utilization rate: Realize the optimal allocation of resources such as energy and bandwidth; (3) Good scalability: Suitable for real-time systems of different scales and application scenarios; (4) Low implementation cost: No additional hardware support is required, and it can be realized through software upgrade.

[0018] These advantages make the present invention have broad application prospects and significant practical value in real-time application scenarios such as industrial Internet of Things and smart cities. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the present invention, the drawings required for the description of the present invention will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0020] Figure 1 is a flowchart of the steps of the double-layer dynamic information update method for an energy harvesting network provided by the present invention; Figure 2 is a schematic structural diagram of the double-layer dynamic information update system for an energy harvesting network provided by the present invention; Figure 3 is a block diagram of the structure of the double-layer dynamic information update device for an energy harvesting network provided by the present invention. Detailed implementation manners

[0021] To make the above objects, features, and advantages of the present invention more obvious and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific implementation manners.

[0022] In edge computing, AoI-based data collection mainly focuses on reducing the AoI of the system to improve the accuracy of real-time decision-making. Existing research classifies these methods into two categories according to energy constraints: energy-sufficient systems (such as stable power supplies) and energy-constrained systems (such as batteries or energy harvesting devices). The former emphasizes the efficient allocation of resources to optimize AoI, while the latter finds a balance between energy efficiency and AoI minimization to meet real-time requirements. For the status update of energy-unconstrained devices, centralized and distributed scheduling strategies can be applied. The centralized method uses the complete system status information to preferentially transmit data with higher AoI or critical data, minimizing the overall AoI of the system.

[0023] Currently, the research on information update for AoI optimization mainly uses centralized or distributed strategies to improve the overall AoI performance, usually assuming sufficient energy resources. However, these methods often ignore the changes in data volume requirements in real-time applications. Key sensors or regions require frequent updates or higher data volumes to maintain analysis accuracy, while non-critical regions can tolerate fewer updates. Currently, techniques such as random scheduling, maximum weight, or drift plus penalty can balance throughput and AoI with low complexity, but fail to address the demand for high-quality and high-timeliness data in real-time monitoring, especially in energy harvesting systems.

[0024] In the field of energy-constrained data updates, related technologies mainly adopt energy-aware update strategies and update methods based on the sleep-wake mechanism. These methods aim to achieve the dual goals of efficient energy utilization and minimization of the age of information (AoI) by dynamically adjusting the transmission timing, transmission frequency, and data priority.

[0025] The core mechanism of the energy-aware scheduling strategy is that when the age of information reaches a preset threshold and the device has sufficient energy, the system triggers the data transmission process. This strategy optimizes the utilization efficiency of energy and transmission resources through a dual-condition judgment mechanism. Specifically, the system needs to monitor two key parameters in real time: (1) whether the current age of information exceeds the preset threshold; (2) whether the available energy of the device meets the transmission requirements. However, this strategy has significant technical limitations: its basic assumption is that the energy collected per time slot is a fixed unit or can be accurately predicted, which deviates significantly from the actual application scenario. In actual deployment, due to the dynamic change of the relative position between the energy source and the receiving source, and the uncontrollability of environmental factors (such as weather conditions), the energy that can be collected per time slot shows significant randomness characteristics. This randomness of energy collection severely limits the applicability of the scheduling method based on the fixed energy assumption in the actual data update scenario.

[0026] To reduce the system energy consumption, researchers have proposed a transmission scheduling strategy based on the sleep-wake mechanism. This strategy adopts an adaptive sleep mechanism. The device dynamically selects to enter the sleep state to save energy according to the time sensitivity and importance indicators of the data, and wakes up the device when detecting the need for critical data transmission. Although this mechanism can significantly reduce the energy consumption when the device is idle, it has the following technical defects: (1) State transition energy consumption: Frequent switching between the sleep state and the working state of the device will generate additional energy consumption, and this transition energy consumption increases significantly with the increase of the state switching frequency; (2) Wake-up delay: It takes a certain amount of time to recover from the sleep state to the working state, and this delay will cause the lag of data transmission and affect the real-time performance of the system; (3) Decrease in response ability: The device in the sleep state cannot respond to sudden data transmission requirements in time, and may miss the opportunity to transmit critical data.

[0027] It can be seen that the current research on transmission scheduling for AoI optimization has the following limitations: (1) Unstable energy at the source: The energy supply of sensor devices has high volatility, especially for sensors relying on energy harvesting technology, whose energy harvesting is significantly affected by environmental condition changes. In addition, the dynamics of the wireless network during the transmission process increase the uncertainty of energy consumption.

[0028] (2) In energy-constrained sensor networks, factors such as channel fluctuations, network topology changes, and external interference result in large time-varying characteristics of the bandwidth resources between the source and the destination.

[0029] (3) Under the conditions of limited energy and bandwidth, achieving a balance among data throughput, energy consumption, and AoI is a complex multi-objective optimization problem. Increasing the data transmission power can increase the throughput, but it will also significantly increase the energy consumption, which may cause the sensor to lack sufficient energy in subsequent transmissions, thereby increasing the system AoI.

[0030] The above three limitations make it difficult to guarantee the quality of information and AoI performance required for current real-time applications with energy harvesting. Based on this, the present invention specifically proposes a high-timeliness information update method for energy harvesting networks, which can significantly reduce the age of information and the energy required for updating information in real-time application systems while ensuring the quality of data required for updating, thereby ensuring the safe and efficient operation of real-time application systems. Specifically, as Figure 1 shown, it shows a flowchart of steps of a two-layer dynamic information update method for an energy harvesting network provided by an embodiment of the present invention. The method includes the following steps: S101, from 1 to K, traverse K sensor groups according to the following steps: For the traversed k-th sensor group, traverse the n-th sensor belonging to the k-th sensor group, and calculate the power of the n-th sensor belonging to the k-th sensor group ; When the power of the n-th sensor belonging to the k-th sensor group meets the power constraint condition, regard the n-th sensor belonging to the k-th sensor group as a primary selected sensor, and determine the k-th sensor group as a primary selected sensor group.

[0031] S102, when multiple primary selected sensor groups are determined, with the goal of minimizing the transmission energy consumption of the scheduled sensors and the age of information of the information uploaded by the scheduled sensor groups, determine the scheduled sensor groups, the scheduled sensors within the scheduled sensor groups, and the transmission power of the scheduled sensors.

[0032] S103, schedule the scheduled sensors to upload information at the transmission power.

[0033] The two-layer dynamic information update method for an energy harvesting network provided by an embodiment of the present invention is applied to an edge node, and the edge node is used to schedule K sensor groups to upload information. The K sensor groups include N sensors. Specifically, the two-layer dynamic information update system for an energy harvesting network in an embodiment of the present invention is as Figure 2As shown in the figure, the edge node is used to implement analysis and decision-making to determine the sensor group scheduling decision (scheduled sensor group), the sensor scheduling decision (scheduled sensor), and the power adjustment of the scheduled sensor. Each observation area includes a plurality of sensors, and the plurality of sensors included in the observation area form a sensor group. When the sensor group is scheduled, the scheduled sensors within the sensor group update information at the transmission power determined through analysis.

[0034] In the embodiments of the present invention, relevant parameters in the information transmission scheduling process of the two-layer dynamic information update for the energy harvesting network are defined, including: observation area (sensor group), sensor group scheduling decision (scheduled sensor group), sensor scheduling decision (scheduled sensor), energy harvesting, AoI (age of information) status, energy status, throughput status, and other parameters. According to the data quality required by the information update system, AoI, channel status, and the energy status of each sensing device, appropriate sensors are selected for data upload to ensure the minimum AoI and energy consumption of the updated data at the destination end during the update process, and further ensure the long-term reliable operation of the system. By setting different weight parameters, the overall AoI performance of the system is improved.

[0035] In the embodiments of the present invention, the problem is first abstractly modeled based on the information update process. A network model, an energy model, a minimum throughput requirement model for each group, and an AoI model are constructed based on the transmission process of the sensing devices, and the entire optimization problem model can be constructed according to the above models.

[0036] Specifically, the sensors and different sets of observation areas are defined as and . In the embodiments of the present invention, all the sensors deployed in the same observation area are regarded as a sensor group. Define as the set of sensors deployed in the observation environment area (i.e., the sensor group) , then we can obtain: . The edge node assigns a sub-channel to each group. Due to the limitation of bandwidth resources, no more than groups are allowed to transmit their update packets without conflict simultaneously (such as through orthogonal channels). Indicates that the edge node schedules the sensor group to transmit data to the space station at time slot , then indicates that the edge node does not schedule the sensor group for data transmission. Based on this, there are the following group scheduling constraints: (1) In the embodiment of the present invention, in step S102, when determining the scheduled sensor group, the group scheduling constraint condition in formula (1) needs to be satisfied, and it is required that the number of determined scheduled sensor groups is not greater than the number of sensor groups allowed to transmit updated data simultaneously.

[0037] In the embodiment of the present invention, it is set that the sensors used for environmental perception use solar energy as their energy source. Therefore, the energy change process of each sensor is as follows: (2) Wherein, represents the remaining energy of the nth sensor at time slot t, represents the remaining energy of the nth sensor at the previous time slot t - 1, represents the energy collected by the nth sensor at time slot t, represents the basic energy consumption of the nth sensor, represents the energy consumption of the nth sensor for transmitting information at time sequence t; represents the maximum storage energy of the sensor.

[0038] To maximize the success probability of data transmission, at each time slot , the edge node will select sensors with sufficient battery energy to transmit updated data, such as: (3) Wherein, ; T d represents the length of time slot t, represents the transmission power of the nth sensor.

[0039] At each time slot , in addition to determining whether to schedule sensor group k and whether to schedule sensor n of sensor group (i.e., and ), the controller on the edge node also needs to adjust the transmission power of the scheduled sensors, and the power cannot exceed its maximum value, such as: (4) By combining inequalities (3) and (4), the range of can be obtained as: (5) In the embodiment of the present invention, the above formula (5) is the power constraint condition of the sensor, indicating that the transmission power of the scheduled sensor is not greater than the power that the actual energy of the sensor can support, nor greater than the maximum power of the sensor.

[0040] The magnitude of the transmission power directly affects the amount of data transmitted. Specifically, when in a time slot Adjust the sensor has a power of , the energy consumption of the sensor in the corresponding time slot is . Based on this, the energy consumption of each sensor group in the time slot can be deduced as: (6) According to the above formula, the average transmission energy consumption of all sensors in the system can be expressed as: (7) When the sensor is selected for data transmission in the time slot , its signal-to-noise ratio at this moment can be expressed as: (8) where , and represent the Gaussian white noise power spectral density, the link bandwidth, and the channel gain, respectively. Based on Shannon's theorem, the transmission rate (or throughput) of the sensor is: (9) It represents the amount of data uploaded by the sensor to the edge node in the time slot . Therefore, the total amount of data uploaded by all selected sensors in this time slot group can be obtained as: (10) For each group , its long-term throughput needs to meet the following minimum requirements: (11) where is a strictly positive real number, representing the minimum requirement for the throughput of the sensor group .

[0041] Let be the AoI of the edge node receiving the th area observation information at the beginning of the time slot . Its evolution process is: (12) Based on the evolution of the single-region AoI in the above formula, further, the AoI expression of all observation regions on the edge node uploading updated data within consecutive time slots can be obtained as: (13) in, It means The predefined weight parameters of the AoI importance of each observation area.

[0042] Set by each time slot The energy consumption of the transmission node and the AoI of the observation area information are minimized by selecting the scheduled sensor group, the scheduled sensors in the scheduled sensor group, and adjusting the power of the selected sensors. The goal is to minimize the average total transmission energy of all sensors and the AoI of the observation area on the edge node side by jointly optimizing the scheduled sensor group, the scheduled sensors, and the transmission power adjustment of the sensors in each time slot, while meeting the constraints of bandwidth, throughput requirements of each sensor group, and energy collection of each sensor. The mathematical model can be expressed as: (14) in It is a weight parameter that balances AoI and energy consumption. It aims to design strategies such as the scheduled sensor group, scheduled sensor selection, and power regulation to achieve the best trade-off among energy consumption, AoI, and throughput.

[0043] Problem P1 is still difficult to solve, so the present invention proposes to transform it based on Lyapunov optimization technology. First, in each time slot For each sensor group The following virtual queue is defined: : (15) Virtual Queues Indicates sensor group exist The accumulated throughput debt in the time slots is initialized as . Introduction The purpose is to ensure that inequality (11) satisfies the minimum throughput constraint, but it does not necessarily require that the throughput of each time slot is strictly greater than the threshold. Based on the above analysis of problem P1, we can first define the quadratic Lyapunov function as , and the one-step conditional Lyapunov drift function is defined as: (16) in Indicates time slot The network state at the beginning, where the use To ensure that the throughput is positive. According to the random network optimization technology, The smaller, The stability of the virtual queue is more easily guaranteed. On the premise of ensuring the stability of the virtual queue, it is also necessary to consider the optimal sensor group scheduling, sensor selection, and power adjustment strategies to minimize the AoI and sensor energy consumption. To solve this problem, the embodiment of the present invention introduces the Lyapunov drift-plus-penalty function , where represents a non-negative control parameter used to balance the optimization objective and the throughput debt queue length. According to 's definition and substituting the objective function into the penalty term, the upper bound of the drift-plus-penalty can be deduced as: (17) where . It can be found that only depends on and is independent of the group scheduling decision , the sensor selection decision , and the power adjustment decision . Therefore, in each time slot, only the second term on the right side of inequality (17) needs to be optimized, and problem P1 can be transformed into: (18) In problem P2, due to the mutual coupling relationship between the scheduled sensor group decision , the selection of the scheduled sensors , and the adjustment of the power of the selected sensors , it is difficult to solve directly. Therefore, the embodiment of the present invention uses the idea of the drift-plus-penalty algorithm to sequentially set the sensor group scheduling decision as in each time slot. After the sensor group scheduling decision is given, based on the relationship between the decision variables and , that is, when , ; otherwise . Therefore, problem P2 is further simplified to a power adjustment sub-problem: (19) After solving the sub-problem P3 by designing a better algorithm, the optimal objective value of the sensor group at any time for problem P2 can be obtained, and its definition is . Then, the objective values are sorted in ascending order, and the first smaller value-corresponding sensor groups are selected for scheduling to ensure that there is no conflict during simultaneous scheduling. Based on the above analysis, the group scheduling strategy set can be obtained as . Next, the convexity and concavity of problem P3 are analyzed to solve this sub-problem.

[0044] By substituting equations (6), (8), (9), and (10) into P3, define as the objective function of problem P3, then it can be expressed as (20) According to the convex optimization theory, the points that satisfy the KKT conditions are the optimal solutions of this optimization problem. Based on this, the embodiments of the present invention use the method of Lagrangian relaxation to find the point values that satisfy the KKT conditions, so as to achieve the purpose of solving this problem. First, define the Lagrangian function as: (21) where and are Lagrange multipliers, and is the upper bound of inequality (5), defined as . The KKT conditions of problem P3 can be expressed as: ① Stationarity:

[0045] ② Primal feasibility:

[0046] ③ Dual feasibility:

[0047] ④ Complementary slackness:

[0048]

[0049] where ④ is derived from the inequality constraint of problem P3. To evaluate the stationarity, it is necessary to calculate the partial derivative of the Lagrangian function with respect to . According to this partial derivative, we can get: (22) Two possible cases can be obtained from the equation of (22): 1) and , then , indicating that the optimal value of power regulation is within the feasible region; 2) or , then or , indicating that the optimal value of power regulation is on the boundary. Therefore, by setting the two Lagrange multipliers to 0, that is and , the within the feasible region can be initially deduced. Then, by checking the feasibility of the deduced , and finding the Finally, for all set up and , rearrange equation (22) and solve the equation to obtain The power regulation formula is: (twenty three) Based on formula (23) , it is also necessary to verify whether it satisfies If If it falls outside the feasible region, it needs to be adjusted to the nearest boundary value: (twenty four) Therefore, the embodiment of the present invention proposes: calculating the power of the nth sensor belonging to the kth sensor group The steps include: According to the cumulative throughput debt of the kth sensor group in time slot t , the channel gain during the transmission of the nth sensor and the length of time slot t T d , calculate the power of the nth sensor belonging to the kth sensor group Specifically, the power of the nth sensor is calculated by the above formula (23): .

[0050] In the embodiment of the present invention, after obtaining the calculation result of the power of the nth sensor, it is also necessary to perform power constraint verification on the calculation result. Specifically, it includes: Verify the power of the nth sensor belonging to the kth sensor group Whether the transmission power constraint of the sensor is met, where the transmission power constraint indicates that the transmission power of the scheduled sensor is not greater than the maximum power of the scheduled sensor and is not greater than the power that can be provided by the actual energy of the scheduled sensor; If the transmission power constraint of the sensor is met, the nth sensor belonging to the kth sensor group is used as the primary sensor. If the kth sensor group is the scheduled sensor group, the power of the nth sensor in the kth sensor group is set as The transmission power of the scheduled sensors is determined.

[0051] If the transmission power constraint of the sensor is not met, the power of the nth sensor in the kth sensor group is reduced. Adjust to the power boundary value of the sensor, the power boundary value ;in, represents the remaining energy of the nth sensor in the previous time slot t-1, denotes the energy collected by the n-th sensor in time slot t, denotes the basic energy consumption of the n-th sensor.

[0052] When the adjusted power is greater than zero, the n-th sensor belonging to the k-th sensor group is used as the primary selected sensor.

[0053] When the k-th sensor group is the scheduled sensor group, the power boundary value of the n-th sensor in the k-th sensor group is determined as the transmission power of the scheduled sensor.

[0054] When the adjusted power is equal to zero, it is determined that the sensor does not meet the transmission power constraint of the sensor.

[0055] In the embodiment of the present invention, after determining the primary selected sensor, the sensor group to which the primary selected sensor belongs is used as the primary selected sensor group, and further, with the goal of minimizing the transmission energy consumption of the scheduled sensor and the age of information of the information uploaded by the scheduled sensor group, the primary selected sensor group is optimized. Specifically, according to the energy consumption of the information uploaded by the primary selected sensor group in the time slot and the cumulative throughput debt of the primary selected sensor group in time slot t and the total amount of data uploaded by all the primary selected sensors in the primary selected sensor group in time slot t and the age of information importance weight parameter of the primary selected sensor group and the age of information of the information transmitted by the primary selected sensor group in time slot t the target value of the primary selected sensor group is calculated , which is defined as ; ; The target values of multiple primary selected sensor groups are sorted in ascending order, and the first smaller value corresponding sensor groups are used as the scheduled sensor groups; Among them, represents the energy consumption of the n-th sensor belonging to the k-th sensor group when uploading information at time sequence t , is the set of sensors belonging to the sensor group ; ; represents the transmission rate of the sensor ; , represents whether the n-th sensor in the k-th sensor group is a primary selected sensor; represents a non-negative control parameter used to balance the target and the throughput debt.

[0056] In the embodiments of the present invention, the target value of the primary sensor group comprehensively considers the energy consumption, throughput, and age of information of the primary sensor group.

[0057] In the embodiments of the present invention, in each time slot t, after determining the primary sensors and the target value of the primary sensor group, based on the degree scheduling constraint (Equation (1)), the first M primary sensor groups with smaller target values can be selected as the scheduled sensor groups.

[0058] In the embodiments of the present invention, before the next round of scheduling (i.e., before the start of the next time slot), the following parameter values are updated: The remaining energy of each sensor in time slot t ; , where represents the maximum storage energy of the sensor; The energy consumption of the uploaded information of each sensor group in time slot ; ; The age of information of the information transmitted by each sensor group in time slot t ; The cumulative throughput debt of each sensor group in time slot t .

[0059] Based on the updated parameter values, the scheduled sensor groups, the scheduled sensors, and the transmission power of the scheduled sensors are re-determined.

[0060] In the embodiments of the present invention, the dynamic transmission scheduling and power regulation algorithm based on KKT can be expressed as follows:

[0061] where, in step 7, the feasibility of is checked according to the power constraint condition in Equation (5).

[0062] Thus, in the embodiments of the present invention, through the analysis and modeling of the dynamic transmission and power regulation problems of the inner and outer layers, the dynamic transmission and power regulation problems of the inner and outer layers are abstracted. Based on the Lyapunov optimization technique, the original problem is transformed to facilitate the solution of the problem. A dynamic transmission and power regulation method for the inner and outer layers is designed.

[0063] The present invention innovatively designs an inner and outer double-layer dynamic transmission and power regulation mechanism for sensor group scheduling and in-group sensor scheduling based on the KKT (Karush-Kuhn-Tucker) conditions, achieving a significant improvement in system performance while ensuring the quality of updated data. Specifically: In terms of AoI optimization: Compared with traditional AoI-aware scheduling methods, the AoI reduction of the present invention exceeds 20%; compared with the benchmark polling scheduling method, the AoI reduction can reach more than 30%. In terms of energy efficiency: While maintaining excellent AoI performance, the energy efficiency of the present invention is comparable to that of the optimal energy-efficient scheduling method, and through the dynamic power regulation mechanism, the best balance between energy consumption and system performance is achieved. Compared with the benchmark polling scheduling method, the AoI reduction can reach more than 30%. The inner and outer double-layer dynamic transmission and power regulation method designed by the present invention based on the KKT conditions can reduce the system AoI by more than 20% compared with the AoI-aware scheduling method while ensuring the quality of the required updated data, and can reduce the system AoI by more than 30% compared with the traditional update method (polling scheduling method). For the system energy consumption, compared with the above two scheduling methods, it can be close to the scheduling method with the optimal energy efficiency.

[0064] The intelligent sensor selection and power regulation strategy based on the KKT conditions proposed by the present invention has the following advantages: (1) Precise sensor selection mechanism: Establish a multi-dimensional evaluation model, comprehensively consider indicators such as sensor energy status, channel quality, and data importance, and optimize the selection of the optimal sensor combination through the KKT conditions to ensure the maximization of system update efficiency; (2) Efficient power regulation strategy: Dynamically adjust the transmission power to adapt to the real-time changes of channel conditions and energy supply, realize the optimal allocation of energy resources, and avoid ineffective energy consumption; (3) Significant performance improvement: Effectively reduce the number of ineffective updates, and the update efficiency is significantly improved. While ensuring data quality and AoI requirements, the overall system efficiency is maximized.

[0065] Through innovative algorithm design and system architecture, the present invention achieves the following comprehensive performance advantages: (1) Strong adaptability: It can dynamically adapt to fluctuations in energy supply and changes in channel conditions; (2) High resource utilization rate: Realize the optimal allocation of resources such as energy and bandwidth; (3) Good scalability: Suitable for real-time systems of different scales and application scenarios; (4) Low implementation cost: No additional hardware support is required, and it can be realized through software upgrade.

[0066] These advantages enable the present invention to have broad application prospects and significant practical value in real-time application scenarios such as industrial Internet of Things and smart cities.

[0067] Based on the same inventive concept, the present invention also provides a double-layer dynamic information update device for an energy harvesting network, as Figure 3As shown, it shows a structural block diagram of a two - layer dynamic information update device for an energy - harvesting network. The two - layer dynamic information update device 300 for the energy - harvesting network includes: A traversal module 301, configured to: traverse K sensor groups from 1 to K according to the following steps: For the k - th sensor group traversed, traverse the n - th sensor belonging to the k - th sensor group, and calculate the power of the n - th sensor belonging to the k - th sensor group ; When the power of the n - th sensor belonging to the k - th sensor group meets the power constraint condition, regard the n - th sensor belonging to the k - th sensor group as a primary - selected sensor, and determine the k - th sensor group as a primary - selected sensor group; An optimization module 302, configured to, when multiple primary - selected sensor groups are determined, aim at minimizing the transmission energy consumption of the scheduled sensors and the age of information of the information uploaded by the scheduled sensor groups, and determine the scheduled sensor groups, the scheduled sensors within the scheduled sensor groups, and the transmission power of the scheduled sensors; A scheduling module 303, configured to schedule the scheduled sensors to upload information at the transmission power.

[0068] Optionally, calculating the power of the n - th sensor belonging to the k - th sensor group , includes: According to the cumulative throughput debt of the k - th sensor group in time slot t , the channel gain during the transmission of the n - th sensor , and the length of time slot t T d , calculate the power of the n - th sensor belonging to the k - th sensor group .

[0069] Optionally, after calculating the power of the n - th sensor belonging to the k - th sensor group , it further includes: Verify whether the power of the n - th sensor belonging to the k - th sensor group meets the transmission power constraint of this sensor. The transmission power constraint indicates that the transmission power of the scheduled sensor is not greater than the maximum power of the scheduled sensor and not greater than the power that the actual energy of the scheduled sensor can provide; When meeting the transmission power constraint of this sensor, regard the n - th sensor belonging to the k - th sensor group as a primary - selected sensor; When the k - th sensor group is a scheduled sensor group, determine the power of the n - th sensor of the k - th sensor group as the transmission power of the scheduled sensor.

[0070] Optionally, when the transmission power constraint of the sensor is not satisfied, the power of the n-th sensor in the k-th sensor group is adjusted to the power boundary value of the sensor, and the power boundary value ; where represents the remaining energy of the n-th sensor in the previous time slot t-1, represents the energy collected by the n-th sensor in the time slot t, represents the basic energy consumption of the n-th sensor; When the adjusted power is greater than zero, the n-th sensor belonging to the k-th sensor group is used as a primary selected sensor; When the k-th sensor group is the scheduled sensor group, the power boundary value of the n-th sensor in the k-th sensor group is determined as the transmission power of the scheduled sensor; When the adjusted power is equal to zero, it is determined that the sensor does not satisfy the transmission power constraint of the sensor.

[0071] Optionally, the optimization module 302 is configured to: According to the energy consumption of the uploaded information of the primary selected sensor group in the time slot , the cumulative throughput debt of the primary selected sensor group in the time slot t , the total amount of data uploaded by all primary selected sensors in the primary selected sensor group in the time slot t , the importance weight parameter of the age of information of the primary selected sensor group , the age of information of the information transmitted by the primary selected sensor group in the time slot t , calculate the target value of the primary selected sensor group , which is defined as ; ; Perform an ascending sort on the target values of multiple primary selected sensor groups , and select the first smaller value corresponding sensor groups as the scheduled sensor groups; where represents the energy consumption of the n-th sensor belonging to the k-th sensor group when uploading information in the time sequence t , is the set of sensors belonging to the sensor group ; ; represents the transmission rate of the sensor ; , represents whether the n-th sensor in the k-th sensor group is a primary selected sensor; A non - negative control parameter used to balance the target and throughput debt.

[0072] Optionally, the device further includes an update module for: Before the next round of scheduling, update the following parameter values: The remaining energy of each sensor at time slot t ; , Denotes the maximum stored energy of the sensor; The energy consumption of the uploaded information of each sensor group at time slot ; ; The age of information of the information transmitted by each sensor group within time slot t ; The cumulative throughput debt of each sensor group within time slot t .

[0073] Based on the same inventive concept, the present invention further provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes, it implements the steps in the double - layer dynamic information update method for an energy - harvesting network described in any of the above - mentioned embodiments.

[0074] Based on the same inventive concept, the present invention further provides a computer - readable storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the steps in the double - layer dynamic information update method for an energy - harvesting network described in any of the above - mentioned embodiments.

[0075] Based on the same inventive concept, the present invention provides a computer program product, including a computer program / instructions. When the computer program / instructions are executed by a processor, it implements the steps in the double - layer dynamic information update method for an energy - harvesting network described in any of the above - mentioned embodiments.

[0076] Each embodiment in this specification is described in a progressive manner. The key point of each embodiment is to illustrate the differences from other embodiments. The same or similar parts among the embodiments can be referred to each other.

[0077] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a device, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer - usable storage media (including but not limited to disk memories, CD - ROMs, optical memories, etc.) containing computer - usable program codes.

[0078] The present invention is described with reference to the flowcharts and / or block diagrams of methods, terminal devices (apparatuses), and computer program products according to the present invention. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and combinations of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processors of general-purpose computers, special-purpose computers, embedded processors, or other programmable terminal devices to generate a machine, such that the instructions executed by the processors of the computer or other programmable terminal devices produce means for implementing the functions specified in one Figure 1 flow or multiple flows and / or blocks Figure 1 block or multiple blocks.

[0079] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable terminal device to work in a specific manner, such that the instructions stored in the computer-readable memory produce a manufactured article including instruction means that implement the functions specified in one Figure 1 flow or multiple flows and / or blocks Figure 1 block or multiple blocks.

[0080] These computer program instructions can also be loaded onto a computer or other programmable terminal device, such that a series of operation steps are executed on the computer or other programmable terminal device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable terminal device provide steps for implementing the functions specified in one Figure 1 flow or multiple flows and / or blocks Figure 1 block or multiple blocks.

[0081] Although the preferred embodiments of the present invention have been described, those skilled in the art can make additional changes and modifications to these embodiments once they learn the basic creative concepts. Therefore, the appended claims are intended to be construed to include the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0082] Finally, it should also be noted that in the present invention, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or terminal device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or terminal device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, method, article or terminal device comprising the element.

[0083] The above has introduced in detail a two-layer dynamic information update method for an energy harvesting network provided by the present invention. Specific examples are used in the present invention to elaborate on the principle and implementation manner of the present invention. The description of the above embodiments is only used to help understand the method and its core idea of the present invention; at the same time, for those of ordinary skill in the art, according to the idea of the present invention, there will be changes in the specific implementation manner and application scope. In summary, the content of this specification should not be construed as a limitation to the present invention.

Claims

1. A two-layer dynamic information update method for an energy harvesting network, applied to an edge node, wherein the edge node is used to schedule K sensor groups to upload information, the K sensor groups comprising N sensors, characterized in that: The method comprises: From 1 to K, traverse the K sensor groups according to the following steps: For the kth sensor group traversed, traverse the nth sensor belonging to the kth sensor group and calculate the power of the nth sensor belonging to the kth sensor group ; The power of the nth sensor belonging to the kth sensor group When the power constraint condition is met, the nth sensor belonging to the kth sensor group is used as the primary sensor, and the kth sensor group is determined as the primary sensor group; When multiple preliminary sensor groups are determined, the scheduled sensor groups, the scheduled sensors in the scheduled sensor groups, and the transmission powers of the scheduled sensors are determined with the goal of minimizing the transmission energy consumption of the scheduled sensors and the information age of the information uploaded by the scheduled sensor groups; The scheduled sensor is scheduled to upload information at the transmission power.

2. The double-layer dynamic information updating method for energy harvesting network according to claim 1 is characterized in that: Calculate the power of the nth sensor belonging to the kth sensor group ,include: According to the cumulative throughput debt of the kth sensor group in time slot t , the channel gain during the transmission of the nth sensor and the length of time slot t T d , calculate the power of the nth sensor belonging to the kth sensor group .

3. The double-layer dynamic information updating method for energy harvesting network according to claim 2 is characterized in that: When calculating the power of the nth sensor belonging to the kth sensor group After that, it also includes: Verify the power of the nth sensor belonging to the kth sensor group Whether the transmission power constraint of the sensor is met, where the transmission power constraint indicates that the transmission power of the scheduled sensor is not greater than the maximum power of the scheduled sensor and is not greater than the power that can be provided by the actual energy of the scheduled sensor; In the case where the transmission power constraint of the sensor is met, the nth sensor belonging to the kth sensor group is used as the primary sensor; When the kth sensor group is the scheduled sensor group, the power of the nth sensor of the kth sensor group is The transmission power of the scheduled sensors is determined.

4. The double-layer dynamic information updating method for energy harvesting network according to claim 3 is characterized in that: The method further comprises: If the transmission power constraint of the sensor is not met, the power of the nth sensor in the kth sensor group is reduced. Adjust to the power boundary value of the sensor, the power boundary value ;in, represents the remaining energy of the nth sensor in the previous time slot t-1, represents the energy collected by the nth sensor in time slot t, represents the basic energy consumption of the nth sensor; When the adjusted power is greater than zero, the nth sensor belonging to the kth sensor group is used as the primary sensor; When the kth sensor group is the scheduled sensor group, the power boundary value of the nth sensor of the kth sensor group is Determining a transmission power of a scheduled sensor; In a case where the adjusted power is equal to zero, it is determined that the sensor does not satisfy the transmission power constraint of the sensor.

5. The double-layer dynamic information updating method for energy harvesting network according to claim 1 is characterized in that: With the goal of minimizing the transmission energy consumption of the scheduled sensors and the information age of the information uploaded by the scheduled sensor group, the scheduled sensor group is determined, including: According to the initial selection of sensor groups in the time slot Energy consumption of uploading information , the cumulative throughput debt of the primary sensor group in time slot t , the total amount of data uploaded by all the preliminary selected sensors in the preliminary selected sensor group within time slot t , Information age importance weight parameter of the primary sensor group , the information age of the information transmitted by the primary sensor group in time slot t , calculate the target value of the preliminary sensor group , which is defined as ; Target values ​​for multiple pre-selected sensor groups Sort in ascending order, select the first Smaller The sensor group corresponding to the value is the scheduled sensor group; in, represents the energy consumption of the nth sensor belonging to the kth sensor group uploading information at time sequence t , Belongs to the sensor group A collection of sensors; ; Indicates sensor The transmission rate; , Indicates whether the nth sensor of the kth sensor group is a primary selected sensor; Represents a non-negative control parameter used to balance the target and throughput debt.

6. The double-layer dynamic information updating method for energy harvesting network according to claim 5 is characterized in that: The method further comprises: Before the next round of scheduling, update the following parameter values: The remaining energy of each sensor in time slot t ; , Indicates the maximum storage energy of the sensor; Each sensor group has a time slot Energy consumption of uploading information ; The information age of each sensor group transmitting information in time slot t ; The cumulative throughput debt of each sensor group in time slot t .

7. A two-layer dynamic information update device for an energy harvesting network, applied to an edge node, the edge node is used to schedule K sensor groups to upload information, the K sensor groups include N sensors, the device comprises: The traversal module is used to traverse K sensor groups from 1 to K according to the following steps: For the kth sensor group traversed, traverse the nth sensor belonging to the kth sensor group and calculate the power of the nth sensor belonging to the kth sensor group ; The power of the nth sensor belonging to the kth sensor group When the power constraint condition is met, the nth sensor belonging to the kth sensor group is used as the primary sensor, and the kth sensor group is determined as the primary sensor group; an optimization module, for determining the scheduled sensor group, the scheduled sensors in the scheduled sensor group, and the transmission power of the scheduled sensors with the goal of minimizing the transmission energy consumption of the scheduled sensors and the information age of the information uploaded by the scheduled sensor group when multiple preliminary sensor groups are determined; The scheduling module is used to schedule the scheduled sensor to upload information with the transmission power.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the double-layer dynamic information updating method for energy harvesting network described in any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the double-layer dynamic information updating method for an energy harvesting network as described in any one of claims 1 to 6 is implemented.

10. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instruction is executed by a processor, the steps in the double-layer dynamic information updating method for energy harvesting network described in any one of claims 1 to 6 are implemented.

Images