State updating method and device of low-power-consumption embedded system and storage medium

Abstract

The invention provides a state updating method and device of a low-power-consumption embedded system and a storage medium. The method comprises the following steps: monitoring the running load of the low-power-consumption embedded system in real time by using a particle swarm optimization algorithm, and dynamically adjusting the working parameters of a processor according to the running load of the low-power-consumption embedded system; calculating the power consumption of the low-power-consumption embedded system according to the dynamically adjusted working parameters of the processor, and updating the running state of the low-power-consumption embedded system according to the power consumption; and continuously updating the power consumption of the low-power-consumption embedded system by using a particle swarm optimization algorithm, and switching the running state of the low-power-consumption embedded system among a full working mode, a low-power-consumption mode and an ultra-low-power-consumption mode according to the updated power consumption. According to the method, the calculation complexity can be reduced, the algorithm robustness is improved, and the system performance and stability are improved.

Description

Technical Field

[0001] The present application relates to the technical field of embedded systems, and in particular to a state updating method, device and storage medium for a low-power embedded system. Background Art

[0002] With the increasing application of embedded systems, how to reduce power consumption while ensuring performance has become a key technical requirement. Low-power embedded systems usually face the challenges of limited resources and strict energy consumption control. Therefore, the application of intelligent optimization technology in system design becomes very important. Existing intelligent optimization technologies, such as genetic algorithms, particle swarm optimization algorithms, and simulated annealing algorithms, have been widely used to optimize the design of embedded systems to improve system performance and reduce energy consumption.

[0003] Although these intelligent optimization algorithms have improved the performance of the system to a certain extent, there are still many problems in practical applications. First, the problem of high computational complexity is particularly prominent. Especially when dealing with large-scale problems, these algorithms often require a lot of computing resources and time, which limits their efficiency in practical applications. Secondly, the difficulty in setting algorithm parameters is also a major challenge. The performance of some intelligent optimization algorithms is highly dependent on the precise adjustment of parameters. It is difficult for non-professionals to effectively configure these parameters, making it difficult to achieve the best results of the algorithm.

[0004] In addition, existing algorithms are prone to falling into local optimal solutions, especially when dealing with complex problems, and it is difficult to find the global optimal solution, which affects the overall performance of the system. Some intelligent optimization algorithms are also less robust and more sensitive to external noise and interference, which may affect the stability and reliability of embedded systems. Finally, the versatility of existing intelligent optimization algorithms is limited and cannot adapt to the design requirements of low-power embedded systems of different types and sizes. Summary of the invention

[0005] In view of this, the embodiments of the present application provide a state update method, device and storage medium for a low-power embedded system to solve the problems of high computational complexity, difficulty in effectively configuring algorithm parameters, existence of local optimal solutions, poor algorithm robustness, and reduced performance and stability of the low-power embedded system in the prior art.

[0006] According to a first aspect of an embodiment of the present application, a state update method for a low-power embedded system is provided, comprising: using a particle swarm optimization algorithm to monitor the operating load of the low-power embedded system in real time, and dynamically adjusting the operating parameters of a processor according to the operating load of the low-power embedded system; calculating the power consumption of the low-power embedded system according to the dynamically adjusted operating parameters of the processor, and updating the operating state of the low-power embedded system according to the power consumption; using a particle swarm optimization algorithm to continuously update the power consumption of the low-power embedded system, and switching the operating state of the low-power embedded system between a full working mode, a low power consumption mode, and an ultra-low power consumption mode according to the updated power consumption.

[0007] According to a second aspect of an embodiment of the present application, a state updating device for a low-power embedded system is provided, comprising: a monitoring module, configured to use a particle swarm optimization algorithm to monitor the operating load of the low-power embedded system in real time, and dynamically adjust the operating parameters of the processor according to the operating load of the low-power embedded system; an updating module, configured to calculate the power consumption of the low-power embedded system according to the dynamically adjusted operating parameters of the processor, and update the operating state of the low-power embedded system according to the power consumption; a switching module, configured to use a particle swarm optimization algorithm to continuously update the power consumption of the low-power embedded system, and switch the operating state of the low-power embedded system between a full working mode, a low power consumption mode and an ultra-low power consumption mode according to the updated power consumption.

[0008] According to a third aspect of an embodiment of the present application, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the steps of the above method are implemented when the processor executes the computer program.

[0009] According to a fourth aspect of an embodiment of the present application, a readable storage medium is provided, which stores a computer program, and when the computer program is executed by a processor, the steps of the above method are implemented.

[0010] At least one of the above technical solutions adopted in the embodiments of the present application can achieve the following beneficial effects:

[0011] The operating load of the low-power embedded system is monitored in real time by using a particle swarm optimization algorithm, and the operating parameters of the processor are dynamically adjusted according to the operating load of the low-power embedded system; the power consumption of the low-power embedded system is calculated according to the dynamically adjusted operating parameters of the processor, and the operating state of the low-power embedded system is updated according to the power consumption; the power consumption of the low-power embedded system is continuously updated by using a particle swarm optimization algorithm, and the operating state of the low-power embedded system is switched between full working mode, low power mode and ultra-low power mode according to the updated power consumption. This application can reduce the computational complexity, improve the robustness of the algorithm, and enhance the system performance and stability. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0013] Figure 1 It is a flowchart of a state updating method of a low-power embedded system provided by an embodiment of the present application;

[0014] Figure 2 It is a structural schematic diagram of a state updating device for a low-power embedded system provided in an embodiment of the present application;

[0015] Figure 3 It is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0016] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0017] In the current low-power embedded system design, intelligent optimization technology is widely used to improve system performance and reduce energy consumption. Intelligent optimization technology includes genetic algorithm, particle swarm optimization algorithm, simulated annealing algorithm, etc. These methods can optimize system design by simulating natural evolution or physical process. However, these existing intelligent optimization methods have some significant shortcomings in practical applications.

[0018] The design of low-power embedded systems often faces challenges such as limited resources and strict power consumption control. In order to optimize system performance under these constraints, researchers have introduced intelligent optimization techniques. These techniques imitate evolution, group behavior, or physical annealing processes in nature and try to find the optimal solution in the design space. Although these methods have brought some performance improvements to the design of low-power embedded systems, they are not without problems in practical applications.

[0019] Existing intelligent optimization algorithms for low-power embedded system optimization still have the following shortcomings:

[0020] 1. High computational complexity: When dealing with large-scale problems, existing intelligent optimization algorithms, such as genetic algorithms and particle swarm optimization algorithms, often consume a lot of computing resources and time. This is because these algorithms usually involve a large number of iterative calculations, especially in the process of searching for the optimal solution, the amount of calculation increases sharply as the size of the problem increases. This high computational complexity limits their practical application in resource-constrained embedded systems.

[0021] 2. Difficulty in setting algorithm parameters: The performance of intelligent optimization algorithms usually depends on the correct setting of multiple parameters. For example, the mutation rate and crossover rate in genetic algorithms, the inertia weight and learning factor in particle swarm algorithms, etc. The setting of these parameters requires certain experience and expertise. For non-professionals, it may be very difficult to adjust these parameters for optimal performance, and incorrect parameter settings may lead to inefficient algorithms.

[0022] 3. Local optimal solution problem: In complex problems, intelligent optimization algorithms may stop at the local optimal solution and fail to find the global optimal solution. This means that the algorithm may be trapped in a local area during the search process and cannot continue to find a better solution, thus affecting the performance optimization effect of the final system.

[0023] 4. Poor algorithm robustness: Some intelligent optimization algorithms are sensitive to noise and interference in the environment. This means that in actual applications, when the system is disturbed by external factors, the algorithm may not be able to maintain stable performance, thus affecting the reliability of the entire embedded system.

[0024] 5. Algorithm versatility limitations: Some intelligent optimization algorithms are only applicable to specific types or scales of problems. Due to the different design requirements of different embedded systems, the versatility of these algorithms is limited and they cannot adapt to the design challenges of all low-power embedded systems. Therefore, in the face of various complex application scenarios, existing intelligent optimization methods may not provide the best solutions.

[0025] Although the existing intelligent optimization methods have certain application value in the design of low-power embedded systems, their high computational complexity, complex parameter setting, local optimal solution problems, poor robustness and limited versatility limit their wide application. Therefore, it is urgent to develop an improved intelligent optimization method that can reduce the computational complexity, simplify the parameter setting process, enhance the global search capability and robustness of the algorithm, and improve its applicability in embedded systems of different types and sizes, so as to comprehensively improve the performance and stability of low-power embedded systems.

[0026] Before introducing the specific embodiments of the technical solution of the present application in detail, first of all, the design optimization direction of the current low-power embedded system is summarized. The design optimization of the existing low-power embedded system is mainly carried out in the following aspects:

[0027] 1. Hardware level:

[0028] 1) Microprocessor selection and optimization: Select an embedded processor that is suitable for application requirements and has low power consumption characteristics, such as low-power chips of the ARM Cortex-M series or RISC-V architecture.

[0029] 2) Dynamic Voltage and Frequency Scaling (DVFS): Reduces power consumption by adjusting the operating voltage and frequency to match system load requirements in real time.

[0030] 3) Power management module: Design an efficient power management system, including multiple working modes (active, hibernation, deep sleep, etc.), and automatically switch to low power state when there is no operation or low load.

[0031] 2. Software level:

[0032] 1) Algorithm optimization: For specific application scenarios, low-complexity, low-energy consumption algorithms are used, and the algorithms are further optimized by combining intelligent optimization algorithms (such as genetic algorithms, particle swarm optimization, etc.).

[0033] 2) Task Scheduling and Parallel Processing: Rationally schedule tasks on multi-core / multi-processor systems, reduce unnecessary wake-ups and context switches, and use intelligent scheduling algorithms to balance the relationship between task execution and power consumption.

[0034] 3) Energy-aware programming: Energy consumption factors are fully considered during the development process, and energy consumption during program execution is monitored and controlled through energy-aware programming technology.

[0035] 3. Intelligent optimization method:

[0036] 1) Machine learning prediction and control: Use machine learning to predict load changes and adjust system status in advance to reduce power consumption; at the same time, it can also be used to adaptively optimize system parameter settings.

[0037] 2) Meta-heuristic optimization algorithm: applied to system-level optimization, such as determining the optimal operating voltage, frequency allocation, and the best task division and scheduling scheme.

[0038] 4. Cross-level optimization:

[0039] 1) Hardware-software collaborative design: Starting from the system level, comprehensively consider the optimization of multiple levels such as hardware structure, operating system kernel, and application program to minimize the overall power consumption.

[0040] 2) Sensor networks and data processing: In large-scale embedded sensor networks, distributed intelligent algorithms are used to reduce data transmission volume and local processing load, thereby reducing the energy consumption of the entire network.

[0041] Through the integration and implementation of the above technologies and methods, it is possible to effectively design embedded systems that meet low power requirements and save energy to the greatest extent while ensuring performance.

[0042] The contents of the technical solution of the present application are described in detail below with reference to the accompanying drawings and specific embodiments.

[0043] Figure 1 1 is a flow chart of a method for updating the state of a low-power embedded system provided in an embodiment of the present application. Figure 1 As shown, the state update method of the low-power embedded system may specifically include:

[0044] S101, using a particle swarm optimization algorithm to monitor the operating load of the low-power embedded system in real time, and dynamically adjusting the operating parameters of the processor according to the operating load of the low-power embedded system;

[0045] S102, calculating the power consumption of the low-power embedded system according to the dynamically adjusted working parameters of the processor, and updating the operating state of the low-power embedded system according to the power consumption;

[0046] S103, continuously updating the power consumption of the low-power embedded system by using a particle swarm optimization algorithm, and switching the operation state of the low-power embedded system between a full working mode, a low power consumption mode and an ultra-low power consumption mode according to the updated power consumption.

[0047] Particle Swarm Optimization (PSO) is an optimization technique based on group collaboration, which is often used to solve continuous and discrete optimization problems. In low-power embedded systems, PSO can be used to optimize multiple system parameters to improve performance and reduce energy consumption.

[0048] Energy management: Particle swarm optimization can be used to optimize power management strategies in embedded systems, such as dynamic voltage frequency scaling (DVFS), sleep mode, etc. By intelligently adjusting these parameters, the system can minimize energy consumption while meeting performance requirements.

[0049] Task Scheduling: In a multi-tasking environment, PSO can be used to optimize the scheduling strategy of tasks to balance processor load and reduce energy consumption. The algorithm can consider the priority, execution time and energy consumption of tasks to find the best scheduling solution.

[0050] System parameter configuration: For system parameters that need to be determined during design (such as cache size, processor type, etc.), PSO can help designers select the best configuration solution by simulating the performance of different configurations.

[0051] Network optimization: In embedded systems that require network communication, PSO can optimize network parameters such as transmission power, routing protocols, etc. to reduce energy consumption during communication.

[0052] Sensor networks: For distributed sensor networks, PSO can be used to optimize the deployment location and transmission strategy of sensors to extend the life cycle of the network and improve the efficiency of data collection.

[0053] The key advantages of particle swarm optimization in these applications are its simplicity and flexibility, which can be adapted to a variety of optimization problems and is easy to implement. In addition, PSO does not require gradient information, making it very suitable for problems where it is difficult to directly calculate the gradient.

[0054] In some embodiments, before using the particle swarm optimization algorithm to monitor the running load of the low-power embedded system in real time, the method further includes:

[0055] When the low-power embedded system is initially powered on, the running state of the low-power embedded system is set to a full working mode, and the parameters of the particle swarm optimization algorithm run by the low-power embedded system are initialized.

[0056] Specifically, when the low-power embedded system is initially powered on, the system's operating state is set to full working mode. In full working mode, the system can run at the highest performance, providing a stable basic state for the subsequent optimization process. In this state, all core functional modules are activated to support the initialization and operation of the particle swarm optimization algorithm.

[0057] In some examples, in order to ensure that the particle swarm optimization algorithm can accurately and efficiently evaluate and optimize the system's operating load, its parameters need to be initialized before the algorithm runs. For example, the following initialization operations may be included:

[0058] First, the number of particles is determined to ensure that the particle swarm has sufficient coverage in the search space.

[0059] Next, set the upper and lower limits of the particle position and velocity to constrain the search range of the particles and prevent unreasonable state configuration.

[0060] Secondly, an initial inertia weight factor is assigned to control the continuity of particle velocity. Larger weight values ​​enhance global search capabilities, while smaller weight values ​​enhance local search capabilities.

[0061] Then, the individual learning factor and the group learning factor are defined to balance the tendency of particles to the individual historical optimal position and the group global optimal position.

[0062] Finally, the initial position and velocity are randomly assigned to each particle to ensure the diversity of the particle swarm and provide more possible solutions for the optimization process.

[0063] Furthermore, after completing the initialization operation, the particle swarm optimization algorithm is in an operational state, and the initial load data of the embedded system is passed as input to the algorithm for calculating the optimization target. At the same time, the system remains in full working mode to provide sufficient computing power for load monitoring and parameter adjustment.

[0064] Through the method of the present embodiment described above, before using the particle swarm optimization algorithm to perform real-time monitoring of the operating load of the low-power embedded system, the system ensures the stability of the operating environment and the efficiency of the optimization algorithm through initialization, providing a reliable foundation for subsequent power consumption optimization.

[0065] In some embodiments, initializing parameters of a particle swarm optimization algorithm executed by a low-power embedded system includes:

[0066] Set the function dimension, number of particles and number of iterations of the particle swarm optimization algorithm, set the inertia weight of the particle velocity, and configure the strategy for the inertia weight to gradually decay with the number of iterations;

[0067] Set the self-learning factor and social learning factor used to guide particle movement, and define the upper and lower limits of particle position and particle change speed;

[0068] Within the preset range, the initial position and initial velocity of each particle are randomly initialized, and the initial position of each particle is taken as the individual historical optimal position of the corresponding particle.

[0069] Specifically, the embodiment of the present application divides the working modes of the low-power embedded system into the following three types: Normal, Standby, and ultra-low power mode; among them, Normal mode is a full working mode, Standby mode is a low-power mode, and ultra-low power mode only supports background SMS wake-up, CAN wake-up, charging wake-up, etc. The low-power embedded system of the present application can automatically shut down a certain function according to the current functional requirements, and update the system status in real time according to the current system power consumption through the particle swarm optimization algorithm.

[0070] Furthermore, when the low-power embedded system is initially powered on, the low-power embedded system is in full working state, namely Normal mode. In this mode, the various parameters of the particle swarm optimization algorithm are first initialized. The following is a detailed description of the process of initializing the parameters of the particle swarm optimization algorithm in conjunction with examples in specific scenarios, which may include the following:

[0071] At the time of initialization, the dimension of the function is set to D = 10, the number of particles is N = 50, the number of iterations is G = 600, the speed weight w = 0.8, the speed weight attenuation coefficient: h = (0.8-0.2) / G, the self-learning factor c1 = 2, the social learning factor c2 = 2, the lower bound of the variable: Xmin = -3, the upper bound of the variable: Xmax = 3, the lower bound of the particle change speed: Vmin = -3, the upper bound of the particle change speed: Vmax = 3, x = Xmin + rand (N, D) * (Xmax-Xmin); % given Each particle in the population is initialized to position v=Vmin+rand(N,D)*(Vmax-Vmin); % Initialize the velocity p=x for each particle in the population; % Initialize particle pbest=fitness(x); % Fitness value of contemporary particles [gbest,index]=max(pbest); % Highest fitness value g=x(index,:); % Optimal individual, gplot=zeros(G,1); % Record the highest fitness value of each generation gplot(1)=gbest.

[0072] In some embodiments, a particle swarm optimization algorithm is used to monitor the operating load of a low-power embedded system in real time, including:

[0073] Calculate the fitness of each particle and compare the current fitness of each particle with the fitness of the corresponding individual historical optimal position. When the current fitness of the particle is better than the fitness of the individual historical optimal position, update the individual historical optimal position to the current position of the particle.

[0074] Compare the current fitness of each particle with the fitness of the corresponding historical optimal position of the group. When the current fitness of the particle is better than the fitness of the historical optimal position of the group, the current position corresponding to the particle with the highest current fitness is taken as the global optimal position of the group.

[0075] In each iteration, the speed and position of each particle are updated according to the individual's historical optimal position, the group's global optimal position and the inertia weight factor, and the operating load of the low-power embedded system is determined in real time based on the updated speed and position.

[0076] Specifically, the embodiment of the present application utilizes a particle swarm optimization algorithm to monitor the operating load of a low-power embedded system in real time, and dynamically adjusts the system state through the particle swarm optimization algorithm, thereby optimizing the overall power consumption performance of the system.

[0077] First, when the embedded system is initialized, each particle of the particle swarm optimization algorithm is assigned an initial position and velocity. Each particle represents a possible state configuration of the embedded system, and the particle's fitness function is used to measure the system's operating load and power consumption under this configuration.

[0078] In each iteration step, the fitness of each particle is first calculated. The fitness is calculated based on the objective function, which usually reflects the current power consumption state of the system. For example, when the system is under high load, the fitness value is low; while under low load or optimization state, the fitness value is high. This process is used to evaluate the effect of the current position of the particle, thus providing a basis for subsequent adjustments.

[0079] Further, update the individual historical optimal position Pi: after calculating the fitness, compare the current fitness of each particle with the fitness of its individual historical optimal position. The individual historical optimal position refers to the best position that the particle has ever reached in all iterations. If the fitness of the current particle is better than the fitness of its individual historical optimal position, the current position of the particle is updated to its new individual historical optimal position. In this way, the particle can remember and retain its own optimal state.

[0080] Furthermore, the global optimal position Pg of the group is updated: the current fitness of each particle is compared with the fitness of the historical optimal position of the group. The historical optimal position of the group refers to the position of the particle with the highest fitness in the entire particle group. If the current fitness of a particle is better than the fitness of the historical optimal position of the group, the current position of the particle is set as the new global optimal position of the group. This process ensures that the entire particle group can evolve towards the optimal state.

[0081] Furthermore, the particle's speed and position are updated: In each iteration, the particle's speed and position are updated based on the inertia weight factor, the individual's historical optimal position, and the group's global optimal position. Specifically, the particle's position update depends on its current speed and is affected by both the individual extreme value and the group extreme value, which enables the particle to explore and develop more effectively in the search space. Under the influence of the inertia weight factor, the particle's search behavior can strike a balance between global search and local search, avoiding premature convergence to the local optimal solution while ensuring that the global optimal solution is eventually reached.

[0082] Through the steps of the above embodiment, after each iteration, the speed and position of the particle will be updated in real time, and the operating load of the system will be determined in real time. The updated system operating load will be used to adjust the operating parameters such as voltage and frequency of the embedded system to achieve the purpose of optimizing power consumption. The whole process is continuously iterated to ultimately achieve the best energy efficiency performance of the system under different load conditions.

[0083] In the embodiment of the present application, the real-time monitoring and adjustment of the operating load of the embedded system using the particle swarm optimization algorithm effectively improves the overall performance and stability of the system while reducing power consumption, which is particularly suitable for resource-constrained and energy-sensitive embedded system design.

[0084] In some embodiments, the velocity and position of each particle are updated using the following formula:

[0085]

[0086] Among them, X represents the current position of the particle in the search space, P represents the position of the particle with the optimal fitness f(k), V(k) represents the search direction of the particle, w represents the inertia weight factor, c1 and c2 represent acceleration constants, and r1 and r2 represent random numbers with uniform distribution in the range.

[0087] Specifically, this embodiment also discloses a specific application of the particle swarm optimization algorithm in a low-power embedded system, especially how particles update their own speed and position through individual extreme values ​​and group extreme values ​​in each iteration process to optimize the operating state of the system.

[0088] First, in the initial stage of the particle swarm optimization algorithm, each particle is assigned an initial speed and position. The initial position of the particle represents a possible state configuration of the embedded system, while the initial speed determines the direction and amplitude of the particle's movement in the search space. In each iteration, the particle continuously adjusts its speed and position through the following formula to find the global optimal solution.

[0089] In one example, the velocity of a particle is a key factor in its movement in the search space. The velocity is updated based on the following formula:

[0090]

[0091] In the above formula, w is the inertia weight factor, which controls the continuity of the particle's current velocity. If the w value is large, the particle has a stronger ability in global search; if the w value is small, the particle is more likely to converge to the local optimal solution.

[0092] c1 and c2 are acceleration constants, representing the individual learning factor and the group learning factor, respectively. c1 controls the speed at which particles approach their individual historical optimal positions, while c2 controls the speed at which particles approach the group's global optimal position.

[0093] r1 and r2 are random numbers with uniform distribution in the range, and the maximum value is the corresponding control parameter, which ensures that the search of particles has a certain randomness, thereby avoiding premature convergence to the local optimal solution.

[0094] In one example, the particle's position update depends on its current velocity, and the update formula is as follows:

[0095]

[0096] In the above formula, X represents the current position of the particle in the search space, and V(k) represents the search direction of the particle.

[0097] In this process, the particle moves in the search space and determines the new position in the next iteration based on its updated speed. The update of position is the key to the particle's exploration in the search space, which determines the search range and direction of the particle in subsequent iterations.

[0098] Furthermore, by adjusting the inertia weight factor w and the values ​​of the acceleration constants c1 and c2, the global search ability and local search ability of the particles can be dynamically balanced. In the early stages of the algorithm, the w value is usually large to enhance the global search ability; as the iteration proceeds, the w value gradually decreases, prompting the particles to gradually concentrate their search and thus find the global optimal solution.

[0099] Furthermore, as the particle swarm optimization algorithm iterates, the speed and position of the particles are continuously updated, allowing the embedded system to adapt to changes in the operating load in real time. This continuous adjustment process helps the system always maintain the best balance between power consumption and performance under different operating conditions.

[0100] Through the above method of this embodiment, the application of particle swarm optimization algorithm in low-power embedded systems is explained in detail, especially how particles update their own speed and position through individual extreme values ​​and group extreme values ​​to optimize the system state. This method effectively balances global search and local search, ensuring that the system can achieve optimal performance in a complex and changing operating environment.

[0101] In some embodiments, after updating the speed and position of each particle, the method further includes:

[0102] When the termination condition of the particle swarm optimization algorithm is met, the particle swarm optimization algorithm is terminated and the result is output; when the termination condition is not met, the fitness of each particle is recalculated and the next round of iteration is performed until the termination condition is met; wherein the termination condition includes reaching a preset error or number of iterations.

[0103] Specifically, this embodiment also discloses the application of the particle swarm optimization algorithm in a low-power embedded system, especially how to judge the termination condition of the algorithm after updating the speed and position of each particle, and further operations when the termination condition is not met.

[0104] First, in each iteration, the particle swarm optimization algorithm evaluates the state of each particle by calculating the fitness of the particle, and updates the speed and position of the particle according to the individual historical optimal position and the global optimal position of the group. After the update is completed, the system operation state is adjusted according to the new position to achieve the optimization purpose.

[0105] Furthermore, after updating the speed and position of each particle, the particle swarm optimization algorithm will immediately determine whether the preset termination condition is met. The termination condition usually includes the following two aspects:

[0106] 1) Reaching the preset error: that is, the error between the current global optimal solution and the target value is less than the preset threshold. If the fitness of any particle in the current particle swarm reaches or exceeds the preset accuracy requirement, it is considered that the algorithm has found a sufficiently good solution and the algorithm can be terminated.

[0107] 2) Reaching the maximum number of iterations: that is, the number of iterations of the algorithm reaches the preset maximum value. If a solution that satisfies the preset error is still not found after reaching the maximum number of iterations, the algorithm will also terminate and output the current optimal solution as the result.

[0108] Furthermore, when the particle swarm optimization algorithm meets any of the above termination conditions, the algorithm stops further iterations and outputs the current global optimal solution. This solution represents the best state configuration of the embedded system in all iterations, including the optimal processor voltage, frequency and other parameter settings to achieve a balance between low power consumption and high efficiency.

[0109] Furthermore, if the termination condition is not met, the algorithm enters the next round of iteration. In this process, the fitness of each particle is first recalculated to reflect the impact of the latest position on the system performance. Then, the algorithm continues to update the speed and position of the particle based on these new fitness values, and re-evaluates the individual historical optimal position and the group global optimal position. This process is repeated until the termination condition is met.

[0110] This process ensures that the particle swarm optimization algorithm can dynamically adapt to the load changes that may occur during the operation of the embedded system and continuously optimize the system state. Through multiple iterations and recalculation of fitness, the algorithm gradually approaches the optimal solution, ensuring that the system can maintain low power consumption and high efficiency in different working modes.

[0111] Through the above method of this embodiment, this embodiment describes in detail how to determine the termination condition after updating the particle speed and position in the particle swarm optimization algorithm, and continue to the next round of iteration when the condition is not met. This process ensures that the algorithm can effectively converge to the global optimal solution, providing reliable support for the optimization of low-power embedded systems.

[0112] In some embodiments, dynamically adjusting the operating parameters of the processor according to the operating load of the low-power embedded system includes:

[0113] Monitor the operating load of low-power embedded systems and use particle swarm optimization algorithm to calculate the optimized system operating parameters;

[0114] Based on the optimized system operating parameters, the voltage and frequency of the processor are dynamically adjusted; the voltage and frequency are reduced when the operating load decreases, and the voltage and frequency are increased when the operating load increases.

[0115] Specifically, this embodiment also discloses how to dynamically adjust the operating parameters of the processor based on the operating load of the low-power embedded system to achieve optimized power consumption management of the system. The adjustment process is implemented through a particle swarm optimization algorithm to ensure that the voltage and frequency of the processor are dynamically adjusted according to changes in the system load.

[0116] Furthermore, during the operation of the system, the embedded system will continuously monitor its current operating load. The operating load can be measured by multiple parameters, such as the current utilization of the processor, the length of the task queue, and the frequency of input and output operations. These load parameters will be recorded in real time and used as input data for the particle swarm optimization algorithm.

[0117] Furthermore, after receiving the system's operating load data, the particle swarm optimization algorithm gradually approaches the optimal solution by simulating the iterative process of the particle group in the search space. Specifically, the algorithm calculates the fitness of each particle, evaluates the system power consumption performance in its current state, and dynamically adjusts the speed and position of each particle based on the individual historical optimal position and the group's global optimal position, thereby calculating a set of optimized system operating parameters.

[0118] Furthermore, according to the optimized system operating parameters calculated by the particle swarm optimization algorithm, the voltage and frequency of the processor are dynamically adjusted:

[0119] When the system is running at a lower load, the processor voltage and frequency are reduced to reduce power consumption. This adjustment enables the system to save energy at low loads, extend battery life, and reduce unnecessary power consumption.

[0120] When the system load increases, the processor voltage and frequency will be increased to ensure that the system can handle higher computing demands. At this time, the system responds to the higher load by increasing the processor's working capacity, thereby ensuring smooth operation of applications and timely processing of tasks.

[0121] Furthermore, the dynamic adjustment process is continuous, that is, in each period of system operation, the particle swarm optimization algorithm will calculate the current operating load in real time and continuously update the processor voltage and frequency settings based on the calculation results. Through this cyclic adjustment mechanism, the embedded system can find the best balance between high performance and low power consumption.

[0122] Furthermore, through this dynamic adjustment mechanism based on the particle swarm optimization algorithm, the overall power consumption of the embedded system has been significantly optimized. Under low load conditions, the system can minimize energy consumption, while under high load conditions, the system can ensure that the necessary computing performance is not affected, thus achieving the dual goals of low power consumption and high efficiency.

[0123] This embodiment describes in detail how to dynamically adjust the voltage and frequency settings of the processor based on the real-time operating load in a low-power embedded system through a particle swarm optimization algorithm to achieve the effect of power consumption optimization. Through this method, the system can adapt to different operating states and effectively balance the relationship between performance and power consumption.

[0124] In some embodiments, updating the running state of the low power embedded system according to the power consumption includes:

[0125] When it is determined that the power consumption meets the requirement of the preset sleep condition, the low power embedded system is switched from the full working mode to the low power mode, so that the low power embedded system enters the low power state.

[0126] Specifically, this embodiment also discloses how to calculate the power consumption of a low-power embedded system based on dynamically adjusted processor operating parameters, and update the system's operating status accordingly, switching from a full operating mode to a low-power mode to achieve effective management of system power consumption.

[0127] Furthermore, during the operation of the embedded system, the particle swarm optimization algorithm is used to dynamically adjust the operating parameters of the processor, such as voltage and frequency, according to the real-time load conditions. The purpose of these parameter adjustments is to optimize the power consumption performance of the system under different load conditions.

[0128] Furthermore, after the processor voltage and frequency are adjusted, the system will calculate the current power consumption based on these new operating parameters. For example, after reducing the processor voltage and frequency, the system power consumption will decrease accordingly; while when increasing the voltage and frequency to cope with higher loads, the power consumption will increase accordingly.

[0129] Furthermore, the calculated power consumption value is compared with the system's preset sleep condition, which is usually a power consumption threshold. When the system power consumption is lower than the threshold, it means that the system's current workload is low and no longer needs to be maintained in full working mode.

[0130] Furthermore, when it is determined that the system power consumption meets the sleep conditions, the system will automatically switch from full working mode to low power mode (Standby mode). In low power mode, the system shuts down or reduces unnecessary functions and only retains basic background tasks, such as basic wake-up mechanisms (SMS wake-up, CAN wake-up, etc.), to further reduce energy consumption.

[0131] Through the mechanism of the above embodiment, when the system operation load is reduced and the power consumption drops to a certain level, the embedded system can enter the low power consumption mode in time, significantly reducing unnecessary energy consumption. This intelligent switching mechanism ensures that the system can achieve maximum power consumption savings while meeting actual needs, extending the battery life of the system, and improving overall energy efficiency.

[0132] According to the technical solution provided in the embodiments of the present application, the technical solution of the present application has at least the following advantages:

[0133] 1. Improve accuracy: Intelligent optimization technologies such as machine learning and data analysis can improve the accuracy of embedded systems when handling complex tasks. For example, in self-driving cars, low-power embedded systems can process large amounts of sensor data in real time, optimize decision-making processes through intelligent algorithms, and improve driving safety and efficiency.

[0134] 2. Enhanced safety: In safety-critical applications, such as medical equipment or industrial control systems, smart optimization can enable more stringent safety monitoring and early warning systems. These systems can identify potential faults or abnormal conditions in real time, intervene or correct them in time, and greatly reduce the risk of accidents.

[0135] 3. Improve sealing and reliability: In applications that require strict environmental control, such as satellites or deep-sea probes, intelligently optimized low-power embedded systems can continuously monitor device status, optimize its operating conditions, and ensure the stability and reliability of the device in harsh environments.

[0136] 4. Enhanced real-time and fault tolerance: Intelligent optimization technology can improve the system's ability to process real-time data, enabling it to respond instantly to external changes and internal failures. In addition, intelligent algorithms can be designed to be adaptive and self-healing, improving the system's fault tolerance and keeping it running even when some components fail.

[0137] 5. Improve compatibility: Through intelligent optimization, embedded systems can be more compatible with other devices and systems and support a wider range of standards and interfaces. This makes it easier to integrate devices into the existing technology ecosystem and enhances the flexibility and scalability of their applications.

[0138] 6. Save cost, space, time and energy: Intelligently optimized low-power design enables the system to reduce energy consumption and physical size without sacrificing performance, thereby reducing manufacturing and operation costs. In addition, optimized processing speed and automation functions can reduce manpower requirements and operating time, further reducing overall costs.

[0139] In general, intelligent optimization of low-power embedded systems can not only improve technical performance, but also bring significant positive effects on economic and environmental levels. These characteristics make it have important value and potential in future technological development and application.

[0140] The following is an embodiment of the device of the present application, which can be used to execute the embodiment of the method of the present application. For details not disclosed in the embodiment of the device of the present application, please refer to the embodiment of the method of the present application.

[0141] Figure 2 Schematic diagram of the structure of the state update device of the low-power embedded system provided by the embodiment of the present application. Figure 2 As shown, the state updating device of the low-power embedded system includes:

[0142] The monitoring module 201 is configured to use a particle swarm optimization algorithm to monitor the operating load of the low-power embedded system in real time, and dynamically adjust the operating parameters of the processor according to the operating load of the low-power embedded system;

[0143] The updating module 202 is configured to calculate the power consumption of the low-power embedded system according to the dynamically adjusted operating parameters of the processor, and update the operating state of the low-power embedded system according to the power consumption;

[0144] The switching module 203 is configured to continuously update the power consumption of the low-power embedded system using a particle swarm optimization algorithm, and switch the operating state of the low-power embedded system between the full working mode, the low power consumption mode and the ultra-low power consumption mode according to the updated power consumption.

[0145] In some embodiments, Figure 2 Before using the particle swarm optimization algorithm to monitor the operating load of the low-power embedded system in real time, the initialization module 204 sets the operating state of the low-power embedded system to the full working mode when the low-power embedded system is initially powered on, and initializes the parameters of the particle swarm optimization algorithm running by the low-power embedded system.

[0146] In some embodiments, Figure 2 The initialization module 204 sets the function dimension, number of particles and number of iterations of the particle swarm optimization algorithm, sets the inertia weight of the particle speed, and configures the strategy of gradually decaying the inertia weight with the number of iterations; sets the self-learning factor and social learning factor for guiding the movement of particles, and defines the upper and lower limits of the particle position and the particle change speed; within a preset range, randomly initializes the initial position and initial speed of each particle, and uses the initial position of each particle as the individual historical optimal position of the corresponding particle.

[0147] In some embodiments, Figure 2 The monitoring module 201 calculates the fitness of each particle, and compares the current fitness of each particle with the fitness of the corresponding individual historical optimal position. When the current fitness of the particle is better than the fitness of the individual historical optimal position, the individual historical optimal position is updated as the current position of the particle; the current fitness of each particle is compared with the fitness of the corresponding group historical optimal position. When the current fitness of the particle is better than the fitness of the group historical optimal position, the current position corresponding to the particle with the highest current fitness is used as the group global optimal position; in each iteration process, the speed and position of each particle are updated according to the individual historical optimal position, the group global optimal position and the inertia weight factor, and the running load of the low-power embedded system is determined in real time according to the updated speed and position.

[0148] In some embodiments, Figure 2After updating the speed and position of each particle, the monitoring module 201 terminates the particle swarm optimization algorithm and outputs the result if the termination condition of the particle swarm optimization algorithm is met; if the termination condition is not met, the fitness of each particle is recalculated and the next round of iteration is performed until the termination condition is met; wherein the termination condition includes reaching a preset error or number of iterations.

[0149] In some embodiments, Figure 2 The monitoring module 201 monitors the operating load of the low-power embedded system and calculates the optimized system operating parameters using the particle swarm optimization algorithm; based on the optimized system operating parameters, the voltage and frequency of the processor are dynamically adjusted; when the operating load decreases, the voltage and frequency are reduced, and when the operating load increases, the voltage and frequency are increased.

[0150] In some embodiments, Figure 2 When determining that the power consumption meets the requirement of the preset sleep condition, the update module 202 switches the low power embedded system from the full working mode to the low power mode, so that the low power embedded system enters the low power state.

[0151] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0152] Figure 3 Schematic diagram of the structure of the electronic device 3 provided in the embodiment of the present application. Figure 3 As shown, the electronic device 3 of this embodiment includes: a processor 301, a memory 302, and a computer program 303 stored in the memory 302 and executable on the processor 301. When the processor 301 executes the computer program 303, the steps in the above-mentioned method embodiments are implemented. Alternatively, when the processor 301 executes the computer program 303, the functions of the modules / units in the above-mentioned device embodiments are implemented.

[0153] Exemplarily, the computer program 303 may be divided into one or more modules / units, which are stored in the memory 302 and executed by the processor 301 to complete the present application. The one or more modules / units may be a series of computer program instruction segments capable of completing specific functions, which are used to describe the execution process of the computer program 303 in the electronic device 3.

[0154] The electronic device 3 may be a desktop computer, a notebook, a PDA, a cloud server, or other electronic device. The electronic device 3 may include, but is not limited to, a processor 301 and a memory 302. Those skilled in the art will appreciate that Figure 3It is only an example of the electronic device 3 and does not constitute a limitation of the electronic device 3. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the electronic device may also include input and output devices, network access devices, buses, etc.

[0155] The processor 301 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may be any conventional processor, etc.

[0156] The memory 302 may be an internal storage unit of the electronic device 3, for example, a hard disk or memory of the electronic device 3. The memory 302 may also be an external storage device of the electronic device 3, for example, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the electronic device 3. Further, the memory 302 may also include both an internal storage unit and an external storage device of the electronic device 3. The memory 302 is used to store computer programs and other programs and data required by the electronic device. The memory 302 may also be used to temporarily store data that has been output or is to be output.

[0157] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0158] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0159] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0160] In the embodiments provided in the present application, it should be understood that the disclosed devices / computer equipment and methods can be implemented in other ways. For example, the device / computer equipment embodiments described above are only schematic. For example, the division of modules or units is only a logical function division. There may be other division methods in actual implementation. Multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0161] 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 on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0162] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0163] If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. The computer program may include computer program code, which may be in source code form, object code form, executable file or some intermediate form. Computer-readable media may include: any entity or device capable of carrying computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc.

[0164] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A state update method for a low-power embedded system, characterized in that: include: Using a particle swarm optimization algorithm to monitor the operating load of a low-power embedded system in real time, and dynamically adjusting the operating parameters of a processor according to the operating load of the low-power embedded system; Calculating the power consumption of the low-power embedded system according to the dynamically adjusted operating parameters of the processor, and updating the operating state of the low-power embedded system according to the power consumption; The particle swarm optimization algorithm is used to continuously update the power consumption of the low-power embedded system, and according to the updated power consumption, the operating state of the low-power embedded system is switched between the full working mode, the low power consumption mode and the ultra-low power consumption mode.

2. The method according to claim 1, characterized in that Before using the particle swarm optimization algorithm to monitor the running load of the low-power embedded system in real time, the method further includes: When the low-power embedded system is initially powered on, the operating state of the low-power embedded system is set to a full working mode, and the parameters of the particle swarm optimization algorithm run by the low-power embedded system are initialized.

3. The method according to claim 2, characterized in that Initializing the parameters of the particle swarm optimization algorithm run by the low-power embedded system includes: The function dimension, number of particles and number of iterations of the particle swarm optimization algorithm are set, the inertia weight of the particle velocity is set, and the strategy of gradually decaying the inertia weight with the number of iterations is configured; Set the self-learning factor and social learning factor used to guide particle movement, and define the upper and lower limits of particle position and particle change speed; Within the preset range, the initial position and initial velocity of each particle are randomly initialized, and the initial position of each particle is taken as the individual historical optimal position of the corresponding particle.

4. The method according to claim 1, characterized in that: The method of using a particle swarm optimization algorithm to monitor the operating load of a low-power embedded system in real time includes: The fitness of each particle is calculated, and the current fitness of each particle is compared with the fitness of the corresponding individual historical optimal position. When the current fitness of the particle is better than the fitness of the individual historical optimal position, the individual historical optimal position is updated to the current position of the particle; Compare the current fitness of each particle with the fitness of the corresponding historical optimal position of the group. When the current fitness of the particle is better than the fitness of the historical optimal position of the group, the current position corresponding to the particle with the highest current fitness is taken as the global optimal position of the group. In each iteration process, the speed and position of each particle are updated according to the individual historical optimal position, the group global optimal position and the inertia weight factor, and the operating load of the low-power embedded system is determined in real time according to the updated speed and position.

5. The method according to claim 3, characterized in that: After the speed and position of each particle are updated, the method further includes: When the termination condition of the particle swarm optimization algorithm is met, the particle swarm optimization algorithm is terminated and a result is output; When the termination condition is not met, the fitness of each particle is recalculated and the next round of iteration is performed until the termination condition is met; wherein the termination condition includes reaching a preset error or number of iterations.

6. The method according to claim 1, characterized in that The dynamically adjusting the operating parameters of the processor according to the operating load of the low-power embedded system includes: Monitoring the operating load of the low-power embedded system and calculating optimized system operating parameters using a particle swarm optimization algorithm; Based on the optimized system operating parameters, the voltage and frequency of the processor are dynamically adjusted; wherein, when the operating load decreases, the voltage and frequency are reduced, and when the operating load increases, the voltage and frequency are increased.

7. The method according to claim 1, characterized in that The updating of the running state of the low-power embedded system according to the power consumption includes: When it is determined that the power consumption meets the requirement of the preset sleep condition, the low power consumption embedded system is switched from the full working mode to the low power consumption mode, so that the low power consumption embedded system enters the low power consumption state.

8. A state updating device for a low-power embedded system, characterized in that: include: A monitoring module is configured to monitor the operating load of the low-power embedded system in real time using a particle swarm optimization algorithm, and dynamically adjust the operating parameters of the processor according to the operating load of the low-power embedded system; An updating module, configured to calculate the power consumption of the low-power embedded system according to the dynamically adjusted operating parameters of the processor, and update the operating state of the low-power embedded system according to the power consumption; The switching module is configured to continuously update the power consumption of the low-power embedded system using the particle swarm optimization algorithm, and switch the operating state of the low-power embedded system between the full working mode, the low power consumption mode and the ultra-low power consumption mode according to the updated power consumption.

9. 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 steps of the method according to any one of claims 1 to 7 are implemented.

10. A readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images