Low-Power Optimization Method, Device, Equipment and Storage Medium for MEMS Sensors

Abstract

The present invention relates to a low-power optimization method, device, equipment and storage medium for a MEMS sensor, including the following steps: adaptively threshold-compressing the output signal of the MEMS sensor to obtain a compressed data stream, and extracting time-domain features from the compressed data stream to obtain a signal activity feature set; performing energy consumption point analysis on the MEMS sensor based on the signal activity feature set to obtain a power consumption distribution map; performing hierarchical energy efficiency configuration mapping on the MEMS sensor based on the power consumption distribution map to obtain an energy efficiency configuration matrix; performing non-linear optimization solution on the energy efficiency configuration matrix to obtain an optimal power consumption control sequence; and performing dynamic power consumption adjustment on the MEMS sensor based on the optimal power consumption control sequence to obtain the target power consumption state of the MEMS sensor, solving the technical problem that some existing power consumption management schemes are too simplistic and fail to fully consider the dynamic change factors in actual operation, resulting in limited energy-saving effects in specific usage scenarios.

Description

Technical Field

[0001] The present invention relates to the technical field of MEMS sensors, and particularly to a low-power optimization method, device, equipment, and storage medium for MEMS sensors. Background Art

[0002] In the context of the rapid development of the current Internet of Things and intelligent devices, MEMS (Micro-Electro-Mechanical Systems) sensors, as a bridge connecting the physical world and the digital world, have an expanding range of applications, from smartphones, wearable devices to smart homes and industrial automation. However, with the increasing complexity of the functions of these devices and the continuous improvement of the requirements for portability and continuous operation time, how to effectively reduce the energy consumption of sensors has become an urgent problem to be solved. Traditional optimization methods often focus on the hardware level, such as improving manufacturing processes or selecting low-power materials, but this method faces problems such as high costs and great technical implementation difficulties.

[0003] On the other hand, optimization strategies at the software level, such as data compression and the application of signal processing algorithms, although can reduce the energy consumption of sensors to a certain extent, the existing methods lack flexibility and adaptability, and it is difficult to make effective adjustments for different application scenarios and working conditions. In addition, some existing power management schemes are too simplistic and fail to fully consider the dynamic change factors in actual operations, resulting in limited energy-saving effects in specific usage scenarios. Therefore, a more intelligent and refined method is needed to optimize the energy consumption performance of MEMS sensors.

[0004] Based on the above challenges, researchers have begun to explore new methods to improve the energy efficiency of MEMS sensors, especially to maintain high-efficiency operation while minimizing energy consumption under complex environmental conditions. This new method not only needs to consider the collaborative optimization of hardware and software, but also needs to introduce advanced algorithms and technologies, such as adaptive threshold compression, time-domain feature extraction, and non-linear optimization, so as to achieve precise control and dynamic adjustment of the sensor energy consumption. This provides a theoretical basis and technical support for the development of more intelligent and efficient MEMS sensors, and at the same time promotes the research in related fields to develop in a deeper and broader direction. Summary of the Invention

[0005] The main purpose of the present invention is to provide a low-power optimization method, device, equipment, and storage medium for MEMS sensors, which solves the technical problem that some existing power management schemes are too simplistic and fail to fully consider the dynamic change factors in actual operations, resulting in limited energy-saving effects in specific usage scenarios.

[0006] To achieve the above purpose, the present invention provides a low-power optimization method for MEMS sensors, including the following steps:

[0007] Adaptive threshold compression is performed on the output signal of the MEMS sensor to obtain a compressed data stream, and time-domain feature extraction is performed on the compressed data stream to obtain a signal activity feature set;

[0008] Based on the signal activity feature set, energy consumption point analysis is performed on the MEMS sensor to obtain a power consumption distribution map;

[0009] Based on the power consumption distribution map, hierarchical energy efficiency configuration mapping is performed on the MEMS sensor to obtain an energy efficiency configuration matrix;

[0010] Nonlinear optimization solution is performed on the energy efficiency configuration matrix to obtain an optimal power consumption control sequence;

[0011] Based on the optimal power consumption control sequence, dynamic power consumption regulation is performed on the MEMS sensor to obtain the target power consumption state of the MEMS sensor.

[0012] Further, the adaptive threshold compression of the output signal of the MEMS sensor to obtain a compressed data stream includes:

[0013] Wavelet packet decomposition is performed on the output signal of the MEMS sensor to obtain multi-level wavelet coefficients, and energy entropy calculation is performed based on the multi-level wavelet coefficients to obtain a signal energy entropy feature vector;

[0014] Threshold segmentation is performed on the signal energy entropy feature vector through a dynamic threshold adjustment algorithm to obtain a binary feature sequence, and run-length encoding is performed based on the binary feature sequence to obtain a compressed data stream.

[0015] Further, the energy consumption point analysis of the MEMS sensor based on the signal activity feature set to obtain a power consumption distribution map includes:

[0016] The signal activity feature set is divided into time sequence segments to obtain an activity time sequence segment group, and energy consumption feature calculation is performed on the activity time sequence segment group to obtain an energy consumption feature point set;

[0017] Energy efficiency correlation analysis is performed on the energy consumption feature point set through a preset voltage-frequency mapping table to obtain an energy efficiency correlation matrix, and energy consumption feature classification is performed on the MEMS sensor based on the energy efficiency correlation matrix to obtain an energy consumption type sequence;

[0018] Spatio-temporal correlation analysis is performed on the energy consumption type sequence to obtain an energy consumption correlation map, spatial distribution statistics is performed on the energy consumption correlation map through energy consumption density calculation to obtain an energy consumption density map, and energy consumption hot spot positioning is performed based on the energy consumption density map to obtain an energy consumption hot spot set;

[0019] Perform time-domain evolution analysis on the energy consumption hot spot set to obtain an energy consumption evolution sequence, and perform energy consumption trend prediction based on the energy consumption evolution sequence to obtain an energy consumption trend graph, and generate a power consumption spectrum map based on the energy consumption trend graph to obtain a power consumption distribution spectrum map.

[0020] Further, hierarchically mapping the energy efficiency configuration of the MEMS sensor based on the power consumption distribution spectrum map to obtain an energy efficiency configuration matrix, including:

[0021] Perform feature level division based on the power consumption distribution spectrum map to obtain a level feature table, and perform power consumption level classification based on the level feature table to obtain a power consumption level sequence;

[0022] Perform mapping conversion on the power consumption level sequence through a preset voltage frequency configuration table to obtain a configuration parameter set, and perform working mode matching based on the configuration parameter set to obtain a working mode group;

[0023] Perform state transition analysis on the working mode group to obtain a state transition graph, and calculate the transition cost based on the state transition graph to obtain a transition cost matrix;

[0024] Perform cost balancing processing on the transition cost matrix through a preset multi-objective weight table to obtain a balance coefficient set, and perform parameter optimization configuration on the balance coefficient set to obtain an optimized configuration group;

[0025] Verify the constraint conditions for the optimized configuration group to obtain an effective configuration sequence, and perform configuration item mapping based on the effective configuration sequence to obtain a mapping configuration table;

[0026] Generate an energy efficiency matrix based on the mapping configuration table to obtain an energy efficiency configuration matrix; wherein, the energy efficiency configuration matrix includes static energy efficiency configuration, dynamic energy efficiency configuration, and transient energy efficiency configuration.

[0027] Further, perform non-linear optimization solution on the energy efficiency configuration matrix to obtain an optimal power consumption control sequence, including:

[0028] Analyze the constraint conditions for the energy efficiency configuration matrix to obtain a constraint condition set, and construct a feasible domain based on the constraint condition set to obtain a feasible solution space;

[0029] Construct an objective function for the feasible solution space through the Lagrange multiplier method to obtain an optimized objective function, and calculate the gradient of the optimized objective function to obtain a gradient vector group;

[0030] Determine the search direction for the gradient vector group to obtain a search direction sequence, and calculate the step size for the search direction sequence to obtain an iteration step size table;

[0031] The parameter iteration optimization of the iteration step table is carried out through nonlinear programming to obtain an iteration solution sequence, and the convergence of the iteration solution sequence is verified to obtain a convergence solution set;

[0032] Based on the convergence solution set, multi-objective trade-off analysis is carried out on the MEMS sensor to obtain a trade-off coefficient matrix, and based on the trade-off coefficient matrix, the solution space is reconstructed to obtain a reconstructed solution sequence;

[0033] The control instruction mapping is carried out on the reconstructed solution sequence through a preset control sequence generator to obtain a control instruction group, and based on the control instruction group, the timing scheduling plan of the MEMS sensor is carried out to obtain an optimal power consumption control sequence, where the optimal power consumption control sequence includes a voltage control sequence, a frequency control sequence and a mode control sequence.

[0034] Further, the timing scheduling plan of the MEMS sensor based on the control instruction group to obtain an optimal power consumption control sequence includes:

[0035] The instruction dependence analysis is carried out on the control instruction group to obtain an instruction dependence relationship diagram, and based on the instruction dependence relationship diagram, the critical path is extracted to obtain a critical instruction linked list;

[0036] The execution timing analysis is carried out on the critical instruction linked list through timing constraint analysis to obtain an execution timing diagram, and based on the execution timing diagram, it is detected whether there is resource competition in the control instruction group;

[0037] If there is, the conflict resolution process is carried out on the control instruction group to obtain a conflict-free instruction group, and the parallelism analysis is carried out on the conflict-free instruction group to obtain a parallel scheduling table;

[0038] The time slice allocation is carried out on the parallel scheduling table to obtain a time slice allocation scheme, and based on the time slice allocation scheme, the timing scheduling plan of the MEMS sensor is carried out to obtain an optimal power consumption control sequence.

[0039] Further, the dynamic power consumption regulation of the MEMS sensor based on the optimal power consumption control sequence to obtain the target power consumption state of the MEMS sensor includes:

[0040] The timing parsing is carried out on the optimal power consumption control sequence to obtain a control timing table, and based on the control timing table, the state transition analysis is carried out to obtain a transition path set;

[0041] The voltage regulation parameter configuration is carried out on the transition path set through a preset voltage regulator to obtain a voltage regulation sequence, and based on the voltage regulation sequence, the frequency dynamic compensation is carried out to obtain a frequency compensation vector;

[0042] Decode the working mode of the frequency compensation vector to obtain a mode switching instruction set, and perform clock gating configuration on the mode switching instruction set to obtain a gating configuration table;

[0043] Through the power consumption monitoring unit, perform real-time power consumption sampling on the MEMS sensor based on the gating configuration table to obtain a power consumption sampling sequence, and perform deviation analysis on the power consumption sampling sequence to obtain deviation characteristic data;

[0044] Calculate compensation parameters for the deviation characteristic data to obtain a compensation vector group, and reconstruct adjustment instructions based on the compensation vector group to obtain a reconstructed instruction sequence;

[0045] Through a dynamic feedback mechanism, adjust the power consumption state of the data based on the reconstructed instruction sequence to obtain the target power consumption state of the MEMS sensor.

[0046] The present invention also provides a low-power optimization device for a MEMS sensor, including:

[0047] An extraction module, configured to perform adaptive threshold compression on the output signal of the MEMS sensor to obtain a compressed data stream, and perform time-domain feature extraction on the compressed data stream to obtain a signal activity feature set;

[0048] An analysis module, configured to perform energy consumption point analysis on the MEMS sensor based on the signal activity feature set to obtain a power consumption distribution map;

[0049] A mapping module, configured to perform hierarchical energy efficiency configuration mapping on the MEMS sensor based on the power consumption distribution map to obtain an energy efficiency configuration matrix;

[0050] A solution module, configured to perform non-linear optimization solution on the energy efficiency configuration matrix to obtain an optimal power consumption control sequence;

[0051] An adjustment module, configured to perform dynamic power consumption adjustment on the MEMS sensor based on the optimal power consumption control sequence to obtain the target power consumption state of the MEMS sensor.

[0052] The present invention also provides a computer device, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the steps of the method described in any one of the above are implemented.

[0053] The present invention also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the method described in any one of the above are implemented.

[0054] A low-power optimization method for a MEMS sensor provided by the present invention includes the following steps: adaptively compressing the output signal of the MEMS sensor to obtain a compressed data stream, and extracting time-domain features from the compressed data stream to obtain a signal activity feature set; analyzing the energy consumption points of the MEMS sensor based on the signal activity feature set to obtain a power consumption distribution map; performing hierarchical energy efficiency configuration mapping on the MEMS sensor based on the power consumption distribution map to obtain an energy efficiency configuration matrix; performing non-linear optimization solution on the energy efficiency configuration matrix to obtain an optimal power consumption control sequence; and performing dynamic power consumption adjustment on the MEMS sensor based on the optimal power consumption control sequence to obtain the target power consumption state of the MEMS sensor, solving the technical problem that some existing power management schemes are too simplified and fail to fully consider the dynamic change factors in actual operation, resulting in limited energy-saving effects in specific usage scenarios, and realizing solving the energy efficiency configuration matrix through a non-linear optimization algorithm to obtain an optimal power consumption control sequence. This method takes into account the complex interactions and dynamic change factors within the system, ensuring that the proposed control strategy can achieve the best energy-saving effect under various working conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 is a schematic diagram of the steps of a low-power optimization method for a MEMS sensor in an embodiment of the present invention;

[0056] Figure 2 is a structural block diagram of a low-power optimization device for a MEMS sensor in an embodiment of the present invention;

[0057] Figure 3 is a schematic structural block diagram of a computer device in an embodiment of the present invention.

[0058] The implementation, functional features, and advantages of the object of the present invention will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0059] In order to make the object, technical solution, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0060] As Figure 1 shown, Figure 1 is a schematic diagram of the steps of a low-power optimization method for a MEMS sensor in an embodiment of the present invention;

[0061] An embodiment of the present invention provides a low-power optimization method for a MEMS sensor, including the following steps:

[0062] Step S1, perform adaptive threshold compression on the output signal of the MEMS sensor to obtain a compressed data stream, and extract time-domain features from the compressed data stream to obtain a signal activity feature set.

[0063] Specifically, in the low-power optimization method of the MEMS sensor, it is first necessary to perform adaptive threshold compression on the sensor output signal to obtain a compressed data stream, and further extract time-domain features from the compressed data stream to obtain a signal activity feature set. The core of this process is to automatically adjust the threshold value through an intelligent algorithm to achieve data compression while ensuring that key information is not lost. Specifically, the adaptive threshold compression technology will adjust the compression threshold in real time according to the dynamic range and characteristics of the input signal, so that only when the signal exceeds a certain threshold will it be recorded, while the signal below the threshold is regarded as background noise or redundant information and is ignored. The advantage of doing this is that it can significantly reduce the amount of data and reduce the burden of subsequent processing and transmission. Once the adaptive threshold compression is completed, the next step is to extract time-domain features from the compressed data stream. The so-called time-domain features here refer to the characteristics of the signal changing with time, such as peak value, frequency, amplitude change, etc. By extracting these features, important activity patterns in the signal can be identified, such as periodic fluctuations, burst pulses, etc., which may be related to the target events monitored by the sensor. For example, in a smart home environment, a MEMS accelerometer is used to detect whether there is someone moving in the room. In this case, the sensor continuously collects vibration data in the environment, but not all data is helpful for judging whether someone is present. By applying the above method, only when the vibration intensity exceeds the set adaptive threshold will the corresponding data be recorded and analyzed, which helps to filter out the interference signals brought by daily environmental noise and focus on capturing the truly meaningful human activity signals. Then, using the time-domain feature extraction technology, a series of parameters describing the signal behavior characteristics can be extracted from the compressed data stream to form a signal activity feature set. These feature sets not only contain important information about the signal itself, but also provide basic data for subsequent energy consumption point analysis. For example, in the above smart home scenario, if the system detects a series of short and high-frequency vibration signals, it may mean that someone is quickly passing through the room; while slow and continuous vibration may mean that someone is walking in the room. In this way, the system can more accurately understand the environmental state, and then make more reasonable decisions. The whole process not only improves the data processing efficiency, but also enhances the response speed and accuracy of the system, and finally realizes the efficient operation and low-power operation of the MEMS sensor in complex application scenarios.

[0064] Step S2, perform energy consumption point analysis on the MEMS sensor based on the signal activity feature set to obtain a power consumption distribution map.

[0065] Specifically, the process of analyzing the energy consumption points of the MEMS sensor based on the signal activity feature set to obtain the power consumption distribution map aims to identify the energy consumption patterns of the sensor in different operating states by deeply analyzing the dynamic characteristics of the signal. First, use the signal activity feature set extracted in the previous steps. These features include, but are not limited to, peaks, frequency changes, amplitude fluctuations, etc., which can comprehensively depict the behavior pattern of the sensor during actual operation. Next, by analyzing the relationship between these features and the energy consumption of the sensor, it is possible to determine which specific signal activities will result in higher energy consumption and clarify the conditions and environments in which these high-energy-consuming activities occur. For example, when using a MEMS accelerometer in a smart home environment to monitor the movement of people in the room, the system will conduct a detailed analysis based on the detected different types of vibration signals (such as short and high-frequency pulses or slow and continuous fluctuations) and their corresponding energy consumption situations. Specifically, when the sensor captures a short and high-frequency vibration signal generated by someone quickly passing through the room, due to the need for a higher sampling rate and more complex signal processing algorithms to accurately identify these signals, this process may result in higher energy consumption. In contrast, for slow and continuous vibration signals, although they also need to be processed, because they are relatively simple and require fewer computing resources, the energy consumption is lower. Through such analysis, a detailed power consumption distribution map can be generated. This map not only shows the energy consumption of the sensor throughout the operation cycle but also reveals the specific impact of different signal activities on energy consumption. This provides an important basis for further optimizing the energy efficiency configuration of the sensor, enabling designers to adjust the working parameters of the sensor according to actual needs to achieve the best energy consumption performance. For example, reducing the sensitivity or data acquisition frequency of the sensor during low-activity periods at night to achieve energy-saving effects. This can not only improve the overall efficiency of the system but also extend the service life of the device.

[0066] Step S3: Based on the power consumption distribution map, perform hierarchical energy efficiency configuration mapping on the MEMS sensor to obtain an energy efficiency configuration matrix.

[0067] Specifically, the process of hierarchical energy efficiency configuration mapping for MEMS sensors based on the power consumption distribution map to obtain the energy efficiency configuration matrix is to analyze the energy consumption characteristics under different operating modes and assign the optimal energy management strategy to each working state. First, use the power consumption distribution map generated in the previous step, which details the energy consumption of the sensor under various signal activities. Next, based on this information, it can be identified which operating modes or signal processing tasks require high energy support and which can operate at lower energy consumption. For example, in the application scenario of using a MEMS accelerometer to monitor the movement of people in a smart home environment, the system can understand from the power consumption distribution map that higher energy is required when rapid movement is detected, while the energy consumption is lower in the stationary state. Then, through hierarchical classification of these different energy consumption modes, the working states of the sensor are divided into multiple levels, and each level corresponds to a specific energy requirement and performance goal. For example, the operations of the sensor can be divided into several levels such as high-sensitivity mode, standard mode, and low-power mode. For each level, further determine its corresponding energy efficiency configuration parameters, such as sampling rate, data transmission frequency, and the complexity of the signal processing algorithm. In this way, an energy efficiency configuration matrix can be constructed, which not only clarifies the specific configuration requirements for each level but also provides guidelines on how to optimize energy use in different application scenarios. For example, at night or during periods when no one is present, the system can automatically switch to the low-power mode to reduce unnecessary data collection and processing, thus significantly reducing the overall energy consumption; while during the day or when activities are detected, it switches back to the high-sensitivity mode to ensure accurate capture of all important signal changes. This hierarchical energy efficiency configuration method enables MEMS sensors to achieve optimal energy management while meeting performance requirements, not only improving the efficiency of the system but also extending the service life of the device.

[0068] Step S4: Perform non-linear optimization and solution on the energy efficiency configuration matrix to obtain the optimal power consumption control sequence.

[0069] Specifically, the process of performing non - linear optimization on the energy - efficiency configuration matrix to obtain the optimal power - consumption control sequence aims to find a control strategy that minimizes the overall energy consumption of the sensor under different operating modes through mathematical optimization methods. First, based on the energy - efficiency configuration matrix constructed in the previous steps, this matrix details the energy requirements and performance objectives at each level, including parameters such as sampling rate, data - transmission frequency, and signal - processing algorithm complexity. Next, non - linear optimization algorithms are used. These algorithms can handle complex multi - variable problems and find global or local optimal solutions. Specifically, the non - linear optimization process takes into account the dynamic changes of the sensor in different application scenarios. For example, when using a MEMS accelerometer in a smart - home environment to monitor the movement of people in the room, the system needs to adjust its operating mode according to real - time activities to achieve the best energy - consumption performance. In this process, the non - linear optimization algorithm simulates the operating state of the sensor under different configurations and calculates the energy consumption corresponding to each configuration. By continuously iterating and adjusting parameter settings, the algorithm can gradually approach a solution that minimizes the total energy consumption. For example, during low - activity periods at night, the system can save energy by reducing the sampling rate and data - transmission frequency; while during high - activity periods during the day, it is necessary to increase sensitivity and data - processing capabilities to ensure accurate capture of all important signals. In addition, the non - linear optimization algorithm can also consider changes in external environmental factors, such as the impact of temperature fluctuations on sensor performance, and dynamically adjust the optimization scheme accordingly. Finally, after a series of complex calculations and optimizations, the algorithm will generate an optimal power - consumption control sequence, which contains specific control measures to be taken at different time periods and conditions, such as when to switch to the energy - saving mode and when to resume the high - sensitivity mode. For example, in a smart - home scenario, the system can automatically adjust the operating mode of the accelerometer according to different times of the day: using the high - sensitivity mode during frequent activities in the early morning and evening, and switching to the low - power mode at night to extend battery life. This non - linear - optimization - based method not only improves the energy - utilization efficiency of the system but also enhances the flexibility and adaptability of the system, ensuring the best energy - consumption management effect in various complex application scenarios. In this way, the MEMS sensor can minimize energy consumption while meeting functional requirements, thereby enhancing the reliability and sustainability of the overall system.

[0070] Step S5: Based on the optimal power - consumption control sequence, perform dynamic power - consumption regulation on the MEMS sensor to obtain the target power - consumption state of the MEMS sensor.

[0071] Specifically, the process of dynamically adjusting the power consumption of MEMS sensors based on the optimal power consumption control sequence to reach the target power consumption state aims to achieve the best energy management by adjusting the working parameters of the sensors in real time. First, the optimal power consumption control sequence generated in the previous steps is utilized. This sequence details the specific control measures to be taken under different time periods and conditions, including parameter settings such as sampling rate, data transmission frequency, and signal processing algorithm complexity. Next, the system gradually implements the corresponding adjustment strategies according to this sequence, thereby ensuring that the sensors can minimize energy consumption while meeting performance requirements. For example, in the application scenario of using a MEMS accelerometer to monitor the movement of people in a smart home environment, the system can automatically adjust the working mode of the accelerometer according to different times of the day. Specifically, when the system detects that the current time period is a low-activity period (such as late at night), it reduces the sampling rate and data transmission frequency of the accelerometer and simplifies the signal processing algorithm according to the instructions in the optimal power consumption control sequence to reduce energy consumption. On the contrary, during high-activity periods (such as during the day or in the evening), the system increases the sampling rate and data processing capabilities to ensure that all important signals can be accurately captured. This dynamic adjustment mechanism not only takes into account the changes in daily activity patterns but also can respond to emergencies, such as a temporarily increased activity event. For example, if an abnormal vibration signal is suddenly detected late at night, the system can quickly switch to a high-sensitivity mode, increase the sampling rate, and enable a more complex signal processing algorithm to promptly and accurately identify potential security threats. In addition, the dynamic power consumption adjustment process also takes into account the influence of external environmental factors, such as the impact of temperature changes on sensor performance. By monitoring these factors in real time, the system can further optimize its power consumption control strategy to ensure optimal performance under various working conditions. Finally, after a series of fine dynamic adjustments, the MEMS sensors will reach their target power consumption state, which not only improves the energy utilization efficiency of the system but also extends the service life of the device. For example, in smart home applications, through this intelligent power management method, not only can the accelerometer provide high-precision data when needed, but also the overall energy consumption can be significantly reduced, thereby enhancing the user experience and the sustainability of the system. In this way, the MEMS sensors achieve a perfect balance between high performance and low power consumption in complex and changing application scenarios.

[0072] In a specific embodiment, the adaptive threshold compression of the output signal of the MEMS sensor to obtain a compressed data stream includes:

[0073] Performing wavelet packet decomposition on the output signal of the MEMS sensor to obtain multi-level wavelet coefficients, and calculating the energy entropy based on the multi-level wavelet coefficients to obtain a signal energy entropy feature vector;

[0074] The signal energy entropy feature vector is threshold segmented by a dynamic threshold adjustment algorithm to obtain a binary feature sequence, and run-length encoding is performed based on the binary feature sequence to obtain a compressed data stream.

[0075] Specifically, the process of adaptively compressing the output signal of a MEMS sensor to obtain a compressed data stream first involves wavelet packet decomposition of the sensor output signal. Specifically, wavelet packet decomposition is a multi-resolution analysis method that can decompose the original signal into multiple frequency bands to obtain wavelet coefficients in different frequency segments. These wavelet coefficients contain rich signal information and can reflect the energy distribution of the signal in each frequency band. Next, based on these multi-level wavelet coefficients, energy entropy calculation is performed to obtain a signal energy entropy feature vector. Energy entropy is an important indicator for measuring signal complexity. It evaluates the randomness and irregularity of the signal by quantifying the energy distribution of wavelet coefficients in each frequency band. For example, when using a MEMS accelerometer to monitor the movement of people in a smart home environment, the signal energy entropy feature vector can help distinguish different types of activity patterns, such as fast movement, slow walking, or stationary state. Once the signal energy entropy feature vector is obtained, the next step is to perform threshold segmentation on it through a dynamic threshold adjustment algorithm. The key to this process is to dynamically adjust the threshold according to the current signal characteristics and system requirements, so that only those features exceeding a specific threshold are retained, and the rest are regarded as redundant information and ignored. In this way, the data volume can be effectively reduced while retaining key information. In specific implementation, first, an initial threshold is set according to the signal energy entropy feature vector, and then this threshold is adjusted through iterative optimization until the optimal segmentation point is found, thereby generating a binary feature sequence. This binary feature sequence not only simplifies the representation of the original signal but also provides a basis for further data compression. Subsequently, run-length encoding (RLE) is performed based on the generated binary feature sequence to obtain the final compressed data stream. Run-length encoding is a simple but effective lossless data compression technique that reduces storage space by recording the length of consecutive identical symbols. In this step, the system scans the binary feature sequence, identifies the runs of consecutive '1's or '0's, and represents each run with a pair of values indicating the starting position and length of the run. For example, in the above smart home application scenario, when rapid movement is detected in the room, since the signal changes violently at this time, there may be more '1's in the binary feature sequence, while in the stationary state, it is mainly composed of '0's. Through RLE compression, the system can significantly reduce the amount of data that needs to be transmitted and processed, improving the overall efficiency. The entire process not only effectively compresses the output signal of the MEMS sensor but also greatly reduces the burden of subsequent processing and transmission on the premise of ensuring that key information is not lost. For example, in a smart home system, by performing the above processing on the accelerometer signal, not only can the user's activity patterns be identified more efficiently, but also the device battery life can be extended and the user experience can be improved. In addition, this adaptive threshold compression method has good flexibility and adaptability, and can dynamically adjust the parameter settings according to the needs of the actual application scenario to ensure the best compression effect and energy consumption performance under various working conditions.Therefore, this method provides a practical solution for achieving a balance between high performance and low power consumption. Through this series of steps, the MEMS sensor can maintain high precision in complex and changing application scenarios while significantly reducing energy consumption, further promoting the development and application of intelligent sensing technology.

[0076] In a specific embodiment, the energy consumption point analysis of the MEMS sensor based on the signal activity feature set to obtain a power consumption distribution map includes:

[0077] Dividing the signal activity feature set into time sequence segments to obtain a set of active time sequence segments, and calculating energy consumption characteristics for the set of active time sequence segments to obtain a set of energy consumption characteristic points;

[0078] Performing energy efficiency correlation analysis on the set of energy consumption characteristic points through a preset voltage-frequency mapping table to obtain an energy efficiency correlation matrix, and classifying the energy consumption characteristics of the MEMS sensor based on the energy efficiency correlation matrix to obtain an energy consumption type sequence;

[0079] Performing spatio-temporal correlation analysis on the energy consumption type sequence to obtain an energy consumption correlation map, performing spatial distribution statistics on the energy consumption correlation map through energy consumption density calculation to obtain an energy consumption density map, and locating energy consumption hotspots based on the energy consumption density map to obtain a set of energy consumption hotspots;

[0080] Performing time-domain evolution analysis on the set of energy consumption hotspots to obtain an energy consumption evolution sequence, predicting the energy consumption trend based on the energy consumption evolution sequence to obtain an energy consumption trend map, and generating a power consumption map based on the energy consumption trend map to obtain a power consumption distribution map.

[0081] Specifically, in the process of analyzing the energy consumption points of the MEMS sensor based on the signal activity feature set to obtain the power consumption distribution map, it is first necessary to divide the signal activity feature set into time segments to obtain the activity time segment group. This process is achieved by dividing the continuous signal data into several paragraphs with specific time intervals, and each paragraph represents the sensor activity pattern within a certain period of time. For example, when using a MEMS accelerometer to monitor the movement of people in a smart home environment, the data for 24 hours a day can be divided into 96 time periods of 15 minutes each, and each time period represents an activity time segment. By calculating the signal activity features within these time periods, such as peaks, frequency changes, and amplitude fluctuations, a set of energy consumption feature points can be obtained. Suppose during the daytime (8:00 - 20:00), due to frequent activities, there may be 3 - 5 significant energy consumption feature points per hour; while during the nighttime (20:00 - 8:00), due to reduced activities, there may be only 1 - 2 energy consumption feature points per hour. Next, use the preset voltage - frequency mapping table to perform energy efficiency correlation analysis on the energy consumption feature point set to generate an energy efficiency correlation matrix. The core of this step is to evaluate the energy consumption of the sensor in different operating states according to different voltage and frequency combinations. For example, when the sensor is in the high - sensitivity mode, its operating voltage is 3.3V and the frequency is 1MHz, while in the low - power mode, the operating voltage drops to 1.8V and the frequency decreases to 250kHz. Through this mapping relationship, the system can identify which operating modes or signal processing tasks require higher energy support and which can operate with lower energy consumption. Based on this information, the system can further classify the energy consumption feature points to generate an energy consumption type sequence. Suppose in a typical home environment, the sensor is in the high - sensitivity mode about 70% of the time during the daytime, and in the low - power mode about 90% of the time during the nighttime. Then, perform spatio - temporal correlation analysis on the energy consumption type sequence to obtain an energy consumption correlation map, and perform spatial distribution statistics on the energy consumption correlation map through energy consumption density calculation to generate an energy consumption density map. In this process, the system not only considers the energy consumption changes in the time dimension but also combines the influencing factors of spatial location. For example, in a smart home, if a certain room is close to the entrance, then the sensors in that room may record more activity data, resulting in a higher energy consumption density. By calculating the average energy consumption density of each room, it can be found that the room close to the entrance consumes about 0.5 watt - hours of energy per square meter per day on average, while the room far from the entrance consumes only 0.2 watt - hours. Based on these data, the system can further locate the energy consumption hot - spot set, that is, those regions or time periods with significantly higher energy consumption than the average level. Next, perform time - domain evolution analysis on the energy consumption hot - spot set to obtain an energy consumption evolution sequence, and based on this sequence, perform energy consumption trend prediction to generate an energy consumption trend map. This process helps predict future energy consumption patterns by analyzing the changing trend of energy consumption hot - spots over time.For example, in the above smart home scenario, the system can analyze the data of the past month and find that the energy consumption hot spot increases significantly every Friday night, which may be due to family members gathering on weekends. Through this long-term trend analysis, the system can make adjustments in advance, optimize the working mode of the sensor to cope with the upcoming high energy consumption period. Suppose that in the past month, the energy consumption on Friday nights is about 40% higher than usual. Then the system can predict the energy consumption growth in future similar time periods based on this trend and take corresponding energy-saving measures. Finally, an energy consumption distribution map is generated based on the energy consumption trend graph. This map not only shows the energy consumption of the sensor during the entire operation cycle, but also reveals the specific impact of different signal activities on energy consumption. For example, in a smart home application, the system can generate a detailed power consumption distribution map showing the energy consumption of each room and each time period. Suppose that in a week, the total energy consumption of the living room is 10 watt-hours, the bedroom is 5 watt-hours, and the kitchen is 8 watt-hours. Through this map, users can clearly see which areas and time periods have high energy consumption and adjust their living habits or device settings accordingly to achieve energy-saving goals. In addition, this power consumption distribution map can also help designers and users better understand the operating efficiency of the system and provide a basis for further optimization, enabling the MEMS sensor to achieve the best energy management effect while meeting performance requirements, not only improving the overall efficiency of the system but also extending the service life of the device. Through this series of steps, the MEMS sensor achieves a balance between high performance and low power consumption in complex and changing application scenarios, promoting the development and application of intelligent sensing technology.

[0082] In a specific embodiment, hierarchically mapping the energy efficiency configuration of the MEMS sensor based on the power consumption distribution map to obtain an energy efficiency configuration matrix includes:

[0083] Dividing the feature levels based on the power consumption distribution map to obtain a hierarchical feature table, and classifying the power consumption levels based on the hierarchical feature table to obtain a power consumption level sequence;

[0084] Performing mapping conversion on the power consumption level sequence through a preset voltage-frequency configuration table to obtain a set of configuration parameters, and performing working mode matching based on the set of configuration parameters to obtain a group of working modes;

[0085] Performing state transition analysis on the group of working modes to obtain a state transition diagram, and calculating the transition cost based on the state transition diagram to obtain a transition cost matrix;

[0086] Performing cost balancing processing on the transition cost matrix through a preset multi-objective weight table to obtain a set of balance coefficients, and performing parameter optimization configuration on the set of balance coefficients to obtain an optimized configuration group;

[0087] Verify the constraint conditions for the optimized configuration group to obtain a valid configuration sequence, and perform configuration item mapping based on the valid configuration sequence to obtain a mapping configuration table;

[0088] Generate an energy efficiency matrix based on the mapping configuration table to obtain an energy efficiency configuration matrix; wherein, the energy efficiency configuration matrix includes static energy efficiency configuration, dynamic energy efficiency configuration, and transient energy efficiency configuration.

[0089] Specifically, in the process of hierarchically mapping the energy efficiency configuration of MEMS sensors based on the power consumption distribution map to obtain the energy efficiency configuration matrix, it is first necessary to start from the power consumption distribution map and obtain the hierarchical feature table through feature level division. This step aims to divide the entire system into several levels with similar energy consumption characteristics according to the characteristics of energy consumption. For example, in a smart home scenario, considering the differences in usage frequency and activity levels in different areas such as the living room, bedroom, and kitchen, these areas can be divided into high-energy consumption areas (such as the living room), medium-energy consumption areas (such as the kitchen), and low-energy consumption areas (such as the bedroom) according to their average daily energy consumption. Suppose the living room consumes 0.5 watt-hours of energy per square meter per day, while the bedroom consumes 0.2 watt-hours. In this way, each area can be classified according to the actual energy consumption situation to form a hierarchical feature table. Then, based on the above hierarchical feature table, power consumption level classification is performed to generate a power consumption level sequence. This process involves determining the energy consumption range of each level and allocating the corresponding power consumption level accordingly. For example, it can be set that higher than 0.4 watt-hours / square meter·day is a high level, between 0.2 and 0.4 watt-hours is a medium level, and lower than 0.2 watt-hours is a low level. In the smart home application example, the living room is classified as a high level, the kitchen as a medium level, and the bedroom as a low level. This step not only helps to clarify the energy consumption status of each area but also provides basic data support for further optimization. Subsequently, the power consumption level sequence is mapped and transformed using a preset voltage-frequency configuration table to obtain a set of configuration parameters, and based on this set, a working mode matching is performed to generate a working mode group. The key here is to select the appropriate voltage and frequency combination according to different power consumption levels to achieve the balance between the best performance and energy consumption. For example, for the high-level living room area, a higher working voltage (such as 3.3V) and frequency (such as 1MHz) can be selected, while in the low-level bedroom, a lower working voltage (such as 1.8V) and frequency (such as 250kHz) are used. In this way, the working mode of the sensor can be adjusted according to the actual needs of each area to ensure that while meeting the functional requirements, the energy consumption is minimized as much as possible. Next is the analysis of the state transition of the working mode group to obtain the state transition diagram and calculate the transition cost to generate the transition cost matrix. The core of this stage lies in evaluating the resource and time costs required for switching between different working modes. For example, in a smart home environment, when the user leaves the room, the system may need to switch from a high-sensitivity mode to a low-power mode, and vice versa. Suppose each mode switch consumes approximately an additional 0.01 watt-hours of energy and causes a response delay of about 1 second. By analyzing all possible state transitions, the system can quantify the cost of each transition, thereby guiding subsequent optimization decisions. Then, the transition cost matrix is processed for cost balance using a preset multi-objective weight table to obtain a set of balance coefficients, and parameter optimization configuration is performed on these coefficients to generate an optimized configuration group.In this process, the system will consider multiple optimization goals, such as reducing the total energy consumption, improving the response speed, etc., and assign different weights according to their respective importance. For example, if the weight of the goal of reducing energy consumption is 0.7 and the weight of improving the response speed is 0.3, then the system will search for an optimal solution between the two. In the case of smart home, this means ensuring sufficient sensitivity to capture user actions while minimizing unnecessary energy consumption. Finally, the constraint conditions of the optimized configuration group are verified to obtain an effective configuration sequence, and based on this sequence, configuration item mapping is performed, and finally a mapping configuration table is generated; a performance configuration matrix is generated based on this table. During this period, the system must ensure that all configuration options comply with the hardware and software limitations, such as the maximum working voltage, minimum working frequency, etc. For example, in smart home devices, all configuration schemes need to ensure that they do not exceed the maximum working voltage of 3.3V and the minimum working frequency of 250kHz of the sensor. Once the verification is completed, the system can establish a detailed energy efficiency configuration matrix based on the effective configuration scheme, including static energy efficiency configuration (such as fixed working modes), dynamic energy efficiency configuration (such as working modes that automatically adjust over time), and transient energy efficiency configuration (such as temporary adjustments in response to emergencies). In summary, through this series of steps, the MEMS sensor can efficiently manage energy use during its life cycle, not only improving the overall performance of the system but also extending the service life of the device. For example, in the application of smart home, through a carefully designed hierarchical energy efficiency configuration mapping strategy, the sensor network in the entire home environment can be made more intelligent and energy-saving. For example, by reasonably arranging the working modes of sensors in different rooms, the monthly energy consumption of the entire home can be reduced by about 10% to 15%, which is of great significance for improving the user experience and saving energy. At the same time, this strategy also provides valuable reference experience for the application of MEMS sensors in other fields, promoting the development and progress of related technologies.

[0090] In a specific embodiment, the non-linear optimization solution of the energy efficiency configuration matrix to obtain an optimal power consumption control sequence includes:

[0091] Analyze the constraint conditions of the energy efficiency configuration matrix to obtain a constraint condition set, and construct a feasible region based on the constraint condition set to obtain a feasible solution space;

[0092] Construct an objective function for the feasible solution space by the Lagrange multiplier method to obtain an optimized objective function, and calculate the gradient of the optimized objective function to obtain a gradient vector group;

[0093] Determine the search direction of the gradient vector group to obtain a search direction sequence, and calculate the step size of the search direction sequence to obtain an iteration step size table;

[0094] The parameter iteration optimization is carried out on the iteration step table through nonlinear programming to obtain an iteration solution sequence, and the convergence verification is carried out on the iteration solution sequence to obtain a convergence solution set;

[0095] Based on the convergence solution set, the multi-objective trade-off analysis is carried out on the MEMS sensor to obtain a trade-off coefficient matrix, and based on the trade-off coefficient matrix, the solution space reconstruction is carried out to obtain a reconstructed solution sequence;

[0096] The control instruction mapping is carried out on the reconstructed solution sequence through a preset control sequence generator to obtain a control instruction group, and based on the control instruction group, the timing scheduling plan is carried out on the MEMS sensor to obtain an optimal power consumption control sequence, where the optimal power consumption control sequence includes a voltage control sequence, a frequency control sequence, and a mode control sequence.

[0097] Specifically, in the process of non-linearly optimizing and solving the energy efficiency configuration matrix to obtain the optimal power consumption control sequence, it is first necessary to analyze the constraint conditions of the energy efficiency configuration matrix to obtain the constraint condition set. This process aims to identify various limiting factors in the system operation, such as the maximum working voltage, the minimum working frequency, and the physical limits of the hardware. For example, when using a MEMS accelerometer to monitor the movement of people in a smart home environment, the working voltage of the sensor cannot exceed 3.3V, and the minimum working frequency is 250kHz. These constraint conditions not only ensure the stability and reliability of the system but also provide boundary conditions for subsequent optimization. Based on these constraint conditions, a feasible solution space can be constructed, which contains all possible solutions that meet these conditions. Next, the Lagrange multiplier method is used to construct the objective function for the feasible solution space, generate the optimization objective function, and calculate the gradient of this function to obtain the gradient vector group. The Lagrange multiplier method is a commonly used mathematical tool for finding the optimal solution under constraint conditions. In this process, the system will transform each constraint condition into the corresponding Lagrange multiplier and incorporate it into the objective function. For example, in a smart home application, assuming the goal is to minimize the total energy consumption while ensuring a certain response speed, then the system will construct an objective function that comprehensively considers these two factors. By calculating the gradient of this objective function, the change direction of each parameter, that is, the gradient vector group, can be determined, which provides a basis for subsequent search and optimization. Then, determine the search direction for the gradient vector group to obtain the search direction sequence, and calculate the step size for the search direction sequence to generate the iteration step size table. Determining the search direction is one of the key steps in the optimization process, which determines how to gradually approach the optimal solution. Specifically, the system will select a suitable search direction based on the gradient information in the current state, so that the value of the objective function can decrease fastest. For example, in a smart home scenario, if the current working mode of the sensor results in high energy consumption, the system will select a direction to reduce energy consumption for adjustment. Once the search direction is determined, the next step is to calculate the step size for each step, that is, the specific amplitude of each adjustment. Assuming the step size for each adjustment is 0.1V or 0.05MHz, in this way, the system can gradually approach the optimal solution within a limited time. Subsequently, non-linear programming is used to iteratively optimize the parameters of the iteration step size table to obtain the iteration solution sequence, and the convergence of the iteration solution sequence is verified to obtain the convergence solution set. Non-linear programming is a powerful optimization technique that can find the global or local optimal solution in complex multi-variable problems. In this process, the system will continuously adjust each parameter, gradually narrow the search range, until a set of optimal solutions is found. For example, in a smart home application, the system can iteratively adjust the working voltage and frequency of the sensor multiple times to finally find a set of parameter combinations that can ensure sufficient sensitivity and minimize energy consumption to the greatest extent.To ensure that the solution found is indeed the optimal one, the system also needs to perform convergence verification, that is, to check whether the solution after each iteration tends to be stable and shows no significant changes. Next, based on the convergent solution set, a multi-objective trade-off analysis is carried out on the MEMS sensor to generate a trade-off coefficient matrix, and based on this matrix, the solution space is reconstructed to obtain a reconstructed solution sequence. The core of the multi-objective trade-off analysis lies in balancing the relationships between different optimization objectives, such as energy consumption and response speed. For example, in a smart home scenario, the system not only has to consider how to reduce energy consumption but also ensure that the sensor can capture the user's activity signals in a timely and accurate manner. By assigning different weights to each objective, the system can find a compromise solution to optimize the overall performance. Suppose the weight for reducing energy consumption is 0.7 and the weight for improving response speed is 0.3. The system will search for an optimal solution between the two and generate a trade-off coefficient matrix accordingly. Based on this matrix, the system can further reconstruct the solution space to find a set of better solution sequences. Finally, through a preset control sequence generator, the control instruction mapping is performed on the reconstructed solution sequence to generate a control instruction set, and based on this set, the timing scheduling plan for the MEMS sensor is carried out to obtain the optimal power consumption control sequence. Here, the control sequence generator is a software module responsible for converting the optimization results into specific control instructions. For example, in a smart home application, the system can generate a series of voltage control instructions (such as from 3.3V to 1.8V), frequency control instructions (such as from 1MHz to 250kHz), and mode control instructions (such as switching from high-sensitivity mode to low-power mode) according to the optimization results. Through these control instructions, the system can dynamically adjust the working mode of the sensor to achieve the best energy management effect. Suppose that at different times of the day, the system will automatically adjust the working mode of the sensor according to the user's activities: high-sensitivity mode during the day and low-power mode at night, which can not only ensure the normal operation of the system but also significantly reduce energy consumption. In summary, through this series of steps, the MEMS sensor can achieve a balance between high performance and low power consumption in complex and changing application scenarios. For example, in a smart home environment, through a carefully designed non-linear optimization method, the sensor network in the entire home environment can be made more intelligent and energy-efficient. Suppose that through this optimization strategy, the monthly household energy consumption is reduced by about 10% to 15%, which is of great significance for improving the user experience and saving energy. In addition, this method also provides valuable reference experiences for the application of MEMS sensors in other fields, promoting the development and progress of related technologies. Through continuous iterative optimization and real-time adjustment, the system can minimize energy consumption while meeting the functional requirements, extend the service life of the device, and further improve the reliability and sustainability of the system.

[0098] In a specific embodiment, the timing scheduling and planning of the MEMS sensor based on the control instruction group to obtain an optimal power consumption control sequence includes:

[0099] Perform instruction dependency analysis on the control instruction group to obtain an instruction dependency relationship graph, and extract critical paths based on the instruction dependency relationship graph to obtain a critical instruction linked list;

[0100] Perform execution timing analysis on the critical instruction linked list through timing constraint analysis to obtain an execution timing graph, and detect whether there is resource competition in the control instruction group based on the execution timing graph;

[0101] If there is, perform conflict resolution processing on the control instruction group to obtain a conflict-free instruction group, and perform parallelism analysis on the conflict-free instruction group to obtain a parallel scheduling table;

[0102] Perform time slice allocation on the parallel scheduling table to obtain a time slice allocation scheme, and perform timing scheduling and planning on the MEMS sensor based on the time slice allocation scheme to obtain an optimal power consumption control sequence.

[0103] Specifically, in the process of performing timing scheduling planning on the MEMS sensor based on the control instruction set to obtain the optimal power consumption control sequence, it is first necessary to perform instruction dependency analysis on the control instruction set, generate an instruction dependency graph, and extract the critical path based on this graph to form a critical instruction linked list. This process aims to identify the dependency relationships between individual control instructions and ensure the correctness of the execution order. For example, when using a MEMS accelerometer to monitor the movement of people in a smart home environment, the system may include multiple control instructions, such as adjusting the working voltage, changing the sampling frequency, and switching the working mode. By performing dependency analysis on these instructions, it is possible to determine which instructions must be executed before or after other instructions, thereby constructing a clear instruction dependency graph. Then, through the critical path extraction method, identify the instruction chains that have the greatest impact on the overall performance, that is, the critical instruction linked list. These critical instructions determine the response speed and energy consumption level of the system, so special attention needs to be paid. Next, perform execution timing analysis on the critical instruction linked list through timing constraint analysis, generate an execution timing diagram, and based on this diagram, detect whether there is resource contention in the control instruction set. The core of this step is to evaluate the execution time of each instruction and its time relationship with others to ensure that all instructions can be completed within the specified time window. For example, in a smart home application, assuming that the system needs to quickly adjust the working mode of the sensor to capture activity signals within a few seconds after the user enters the room, it is necessary to ensure that the relevant instructions can be executed in a timely manner. At the same time, the system also needs to check whether there is a resource contention problem, that is, whether multiple instructions attempt to access the same hardware resource simultaneously, resulting in conflicts. For example, if two instructions simultaneously request to change the working voltage of the sensor, resource contention may occur, affecting the normal operation of the system. If resource contention is detected, it is necessary to perform conflict resolution processing on the control instruction set, generate a conflict-free instruction set, and perform parallelism analysis on this set to generate a parallel scheduling table. The goal of conflict resolution is to rearrange the execution order of instructions or allocate different resources to avoid conflicts. For example, in the above smart home scenario, the resource contention problem can be solved by adjusting the priority of instructions or allocating different time periods. Once the conflict is resolved, the system can further analyze the parallelism of instructions to determine which instructions can be executed simultaneously without interfering with each other. In this way, a detailed parallel scheduling table can be generated, specifying the execution time and resource requirements of each instruction, improving the overall efficiency of the system. Subsequently, perform time slice allocation on the parallel scheduling table, generate a time slice allocation scheme, and based on this scheme, perform timing scheduling planning on the MEMS sensor to finally obtain the optimal power consumption control sequence. The key to time slice allocation lies in reasonably dividing the execution time interval of each instruction to ensure that system resources are fully utilized while avoiding overload. For example, in a smart home environment, assuming that there are different activity patterns in different time periods of a day, the system can dynamically adjust the working mode of the sensor according to the user's activity pattern.During the daytime, due to frequent activities, the system can allocate more resources for the high-sensitivity mode; while at night, when activities decrease, the system can switch to the low-power mode to save energy. Through this fine time-slot allocation, the system can not only meet the functional requirements but also significantly reduce energy consumption. Throughout the process, the system will continuously optimize the control instruction set and the timing scheduling scheme according to the requirements of the actual application scenario. For example, in smart home applications, by real-time monitoring the user's activities and environmental changes, the system can dynamically adjust the working mode of the sensors to ensure the best performance and the lowest energy consumption in any situation. Specifically, when the system detects that the user has left the room, it can automatically reduce the sensitivity of the sensors or decrease the data acquisition frequency, thereby extending the battery life; while when the user returns to the room, the system can quickly return to the high-sensitivity mode to ensure accurate capture of all important activity signals. This flexible timing scheduling strategy enables the MEMS sensor to achieve a balance between high performance and low power consumption in complex and changing application scenarios. In summary, through this series of steps, the MEMS sensor can efficiently manage energy usage during its life cycle, not only improving the overall efficiency of the system but also extending the service life of the device. For example, in smart home applications, through a carefully designed timing scheduling and planning strategy, the sensor network in the entire home environment can be made more intelligent and energy-efficient. By reasonably arranging the working modes of sensors in different rooms, the system can not only ensure sufficient sensitivity to capture the user's actions but also minimize unnecessary energy consumption. In addition, this method also provides valuable reference experience for the application of MEMS sensors in other fields, promoting the development and progress of related technologies. Through continuous iterative optimization and real-time adjustment, the system can minimize energy consumption while meeting the functional requirements, improving the user experience and saving energy. This intelligent scheduling mechanism not only enhances the reliability and sustainability of the system but also lays a solid foundation for future smart home and other Internet of Things applications.

[0104] In a specific embodiment, the dynamically adjusting the power consumption of the MEMS sensor based on the optimal power consumption control sequence to obtain the target power consumption state of the MEMS sensor includes:

[0105] Performing timing analysis on the optimal power consumption control sequence to obtain a control timing table, and performing state transition analysis based on the control timing table to obtain a set of transition paths;

[0106] Configuring voltage regulation parameters for the set of transition paths through a preset voltage regulator to obtain a voltage regulation sequence, and performing frequency dynamic compensation based on the voltage regulation sequence to obtain a frequency compensation vector;

[0107] Decode the working mode of the frequency compensation vector to obtain a mode switching instruction set, and perform clock gating configuration on the mode switching instruction set to obtain a gating configuration table;

[0108] Through the power consumption monitoring unit, based on the gating configuration table, perform real-time power consumption sampling on the MEMS sensor to obtain a power consumption sampling sequence, and perform deviation analysis on the power consumption sampling sequence to obtain deviation characteristic data;

[0109] Calculate compensation parameter for the deviation characteristic data to obtain a compensation vector group, and based on the compensation vector group, reconstruct adjustment instructions to obtain a reconstructed instruction sequence;

[0110] Through a dynamic feedback mechanism, based on the reconstructed instruction sequence, adjust the power consumption state of the data to obtain the target power consumption state of the MEMS sensor.

[0111] Specifically, in the process of dynamically adjusting the power consumption of the MEMS sensor based on the optimal power consumption control sequence to obtain the target power consumption state, it is first necessary to perform timing analysis on the optimal power consumption control sequence to generate a control timing table, and perform state transition analysis based on this table to obtain a set of transition paths. This process aims to convert abstract control instructions into specific execution steps to ensure that each instruction can be executed at the correct time point. For example, when using a MEMS accelerometer to monitor the movement of people in a smart home environment, the system will dynamically adjust the working mode of the sensor according to the user's activity pattern and environmental changes. By performing timing analysis on the optimal power consumption control sequence, the specific execution time and sequence of each control instruction can be determined to form a detailed control timing table. Then, through state transition analysis, the system can determine the paths required to transition from the current working state to the target state, that is, the set of transition paths. These paths not only clarify the conversion relationship between each state but also provide a basis for subsequent voltage and frequency regulation. Next, the voltage regulation parameters of the set of transition paths are configured through a preset voltage regulator to generate a voltage regulation sequence, and frequency dynamic compensation is performed based on this sequence to obtain a frequency compensation vector. Voltage regulation is one of the key steps in achieving dynamic power consumption regulation. By precisely controlling the working voltage of the sensor, energy consumption can be minimized while meeting performance requirements. Specifically, the system will configure corresponding voltage regulation parameters according to each state transition in the set of transition paths to generate a voltage regulation sequence. For example, in a smart home application, when the user leaves the room, the system can save energy by reducing the working voltage of the sensor; when the user returns to the room, it quickly returns to the normal working voltage to ensure sensitivity. At the same time, to maintain the stability and response speed of the system, the system also needs to perform dynamic compensation on the frequency to generate a frequency compensation vector. In this way, even when the voltage changes, the sensor can still maintain a high working efficiency. Then, the working mode of the frequency compensation vector is decoded to generate a set of mode switching instructions, and clock gating configuration is performed on this set of instructions to generate a gating configuration table. The goal of working mode decoding is to convert the frequency compensation vector into specific mode switching instructions to ensure that the system can dynamically adjust the working mode according to actual needs. For example, in a smart home scenario, the system can automatically switch the working mode of the sensor according to the user's activity, such as switching from a high-sensitivity mode to a low-power mode. To further optimize energy consumption management, the system also needs to perform clock gating configuration on the set of mode switching instructions to generate a gating configuration table. Clock gating is an effective energy-saving technology that reduces dynamic power consumption by turning off unnecessary clock signals. In this way, the system can significantly reduce energy consumption without affecting performance. Subsequently, through the power consumption monitoring unit, real-time power consumption sampling of the MEMS sensor is performed based on the gating configuration table to generate a power consumption sampling sequence, and deviation analysis is performed on this sequence to obtain deviation characteristic data.The power consumption monitoring unit plays a crucial role throughout the process. It can monitor the power consumption of the sensor in real time to ensure that the system is always in the optimal power consumption state. For example, in smart home applications, the system can regularly collect the power consumption data of the sensor through the power consumption monitoring unit to generate a power consumption sampling sequence. By performing deviation analysis on this data, the difference between the actual power consumption and the expected value can be identified, and deviation characteristic data can be generated. These data not only help to discover potential problems but also provide a basis for the subsequent calculation of compensation parameters. Then, compensation parameters are calculated based on the deviation characteristic data to generate a compensation vector group, and based on this group, a regulation instruction reconstruction is carried out to generate a reconstructed instruction sequence. The core of the compensation parameter calculation lies in adjusting the control strategy of the system according to the deviation characteristic data to ensure that it can maintain the best performance under various working conditions. For example, in a smart home scenario, if the system detects that the actual power consumption of the sensor is higher than the expected value, the power consumption can be reduced by adjusting the voltage and frequency parameters. Specifically, the system will calculate a series of compensation parameters based on the deviation characteristic data to generate a compensation vector group, and based on this group, a regulation instruction sequence is reconstructed. These instructions not only include the new voltage and frequency settings but may also include other optimization measures, such as further refining the clock gating strategy, etc. Finally, through a dynamic feedback mechanism, the power consumption state of the data is adjusted based on the reconstructed instruction sequence to obtain the target power consumption state of the MEMS sensor. The dynamic feedback mechanism is a closed-loop control system that continuously monitors and adjusts the operating state of the system to ensure that it always remains in the optimal power consumption state. For example, in smart home applications, the system can dynamically adjust parameters such as the working mode, voltage, and frequency of the sensor according to the real-time power consumption data and compensation parameters, and finally reach the target power consumption state. This dynamic adjustment mechanism not only improves the energy efficiency of the system but also enhances its adaptability and flexibility, enabling it to achieve a balance between high performance and low power consumption in complex and changing application scenarios. In summary, through this series of steps, the MEMS sensor can achieve efficient energy management in complex and changing application scenarios. For example, in a smart home environment, through a carefully designed dynamic power consumption regulation strategy, the sensor network in the entire home environment can be made more intelligent and energy-efficient. The system can not only dynamically adjust the working mode of the sensor according to the user's activity pattern but also monitor and optimize its power consumption state in real time to ensure that the best performance and the lowest energy consumption can be provided under any circumstances. In addition, this method also provides valuable reference experience for the application of MEMS sensors in other fields, promoting the development and progress of related technologies. Through continuous iterative optimization and real-time adjustment, the system can minimize energy consumption while meeting the functional requirements, improving the user experience and saving energy. This intelligent scheduling mechanism not only enhances the reliability and sustainability of the system but also lays a solid foundation for future smart home and other Internet of Things applications.

[0112] The above describes the low power consumption optimization method of the MEMS sensor in the embodiment of the present invention. The following describes the low power consumption optimization device of the MEMS sensor in the embodiment of the present invention. Figure 2 , an embodiment of a low power consumption optimization device for a MEMS sensor in an embodiment of the present invention includes:

[0113] An extraction module 21 is used to perform adaptive threshold compression on the output signal of the MEMS sensor to obtain a compressed data stream, and perform time domain feature extraction on the compressed data stream to obtain a signal activity feature set;

[0114] An analysis module 22, configured to perform energy consumption point analysis on the MEMS sensor based on the signal activity feature set to obtain a power consumption distribution map;

[0115] A mapping module 23, configured to perform hierarchical energy efficiency configuration mapping on the MEMS sensor based on the power consumption distribution map to obtain an energy efficiency configuration matrix;

[0116] A solution module 24 is used to perform nonlinear optimization on the energy efficiency configuration matrix to obtain an optimal power consumption control sequence;

[0117] The adjustment module 25 is used to dynamically adjust the power consumption of the MEMS sensor based on the optimal power consumption control sequence to obtain a target power consumption state of the MEMS sensor.

[0118] In this embodiment, for the specific implementation of each unit in the above device embodiment, please refer to the above method embodiment, which will not be repeated here.

[0119] Reference Figure 3 The present invention also provides a computer device in an embodiment, wherein the internal structure of the computer device can be as follows: Figure 3 As shown. The computer device includes a processor, a memory, a display screen, an input device, a network interface and a database connected through a system bus. Among them, the processor designed by the computer is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the corresponding data in this embodiment. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the above method is implemented.

[0120] Those skilled in the art will understand that Figure 3 The structure shown in the figure is merely a block diagram of a portion of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied.

[0121] An embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the above method is implemented. It can be understood that the computer-readable storage medium in this embodiment can be a volatile readable storage medium or a non-volatile readable storage medium.

[0122] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium provided by the present invention and used in the embodiments can include non-volatile and / or volatile memories. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM, etc.

[0123] It should be noted that in this article, the term "including", "comprising", or any other variant thereof is intended to cover a non-exclusive inclusion, so that a process, device, article, or method including 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, device, article, or method. Without further limitation, an element defined by the statement "including one..." does not exclude the existence of another identical element in the process, device, article, or method including that element.

[0124] The above are only the preferred embodiments of the present invention, and do not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the content of the specification and drawings of the present invention, or directly or indirectly applied in other related technical fields, shall be equally included in the patent protection scope of the present invention.

Claims

1. A low power consumption optimization method for a MEMS sensor, characterized in that: The following steps are involved: Performing adaptive threshold compression on the output signal of the MEMS sensor to obtain a compressed data stream, and performing time domain feature extraction on the compressed data stream to obtain a signal activity feature set; Performing energy consumption point analysis on the MEMS sensor based on the signal activity feature set to obtain a power consumption distribution map; Based on the power consumption distribution map, the MEMS sensor is mapped in a hierarchical energy efficiency configuration to obtain an energy efficiency configuration matrix; Performing nonlinear optimization on the energy efficiency configuration matrix to obtain an optimal power consumption control sequence; Dynamically adjust the power consumption of the MEMS sensor based on the optimal power consumption control sequence to obtain a target power consumption state of the MEMS sensor; Dividing the signal activity feature set into time segments to obtain an activity time segment group, and calculating energy consumption features of the activity time segment group to obtain an energy consumption feature point set; Performing energy efficiency correlation analysis on the energy consumption feature point set through a preset voltage-frequency mapping table to obtain an energy efficiency correlation matrix, and classifying energy consumption characteristics of the MEMS sensor based on the energy efficiency correlation matrix to obtain an energy consumption type sequence; Performing spatiotemporal correlation analysis on the energy consumption type sequence to obtain an energy consumption correlation graph, performing spatial distribution statistics on the energy consumption correlation graph through energy consumption density calculation to obtain an energy consumption density graph, and locating energy consumption hot spots based on the energy consumption density graph to obtain an energy consumption hot spot set; Perform time domain evolution analysis on the energy consumption hotspot set to obtain an energy consumption evolution sequence, perform energy consumption trend prediction based on the energy consumption evolution sequence to obtain an energy consumption trend graph, and generate a power consumption spectrum based on the energy consumption trend graph to obtain a power consumption distribution spectrum.

2. The low power consumption optimization method of a MEMS sensor according to claim 1, characterized in that: The step of performing adaptive threshold compression on the output signal of the MEMS sensor to obtain a compressed data stream includes: Performing wavelet packet decomposition on the output signal of the MEMS sensor to obtain multi-level wavelet coefficients, and performing energy entropy calculation based on the multi-level wavelet coefficients to obtain a signal energy entropy feature vector; The signal energy entropy feature vector is threshold segmented by a dynamic threshold adjustment algorithm to obtain a binary feature sequence, and run-length encoding is performed based on the binary feature sequence to obtain a compressed data stream.

3. The low power consumption optimization method of a MEMS sensor according to claim 1, characterized in that: The step of performing hierarchical energy efficiency configuration mapping on the MEMS sensor based on the power consumption distribution map to obtain an energy efficiency configuration matrix includes: Based on the power consumption distribution map, feature hierarchical division is performed to obtain a hierarchical feature table, and based on the hierarchical feature table, power consumption level classification is performed to obtain a power consumption level sequence; Mapping and converting the power consumption level sequence through a preset voltage and frequency configuration table to obtain a configuration parameter set, and performing working mode matching based on the configuration parameter set to obtain a working mode group; Performing state transition analysis on the working mode group to obtain a state transition diagram, and performing transition cost calculation based on the state transition diagram to obtain a transition cost matrix; Performing cost balancing processing on the conversion cost matrix through a preset multi-objective weight table to obtain a balance coefficient set, and performing parameter optimization configuration on the balance coefficient set to obtain an optimized configuration group; Verifying the constraints of the optimized configuration group to obtain a valid configuration sequence, and mapping configuration items based on the valid configuration sequence to obtain a mapping configuration table; An energy efficiency matrix is ​​generated based on the mapping configuration table to obtain an energy efficiency configuration matrix; wherein the energy efficiency configuration matrix includes static energy efficiency configuration, dynamic energy efficiency configuration and transient energy efficiency configuration.

4. The low power consumption optimization method of a MEMS sensor according to claim 1, characterized in that: The nonlinear optimization solution of the energy efficiency configuration matrix to obtain an optimal power consumption control sequence includes: Performing constraint analysis on the energy efficiency configuration matrix to obtain a constraint set, and constructing a feasible domain based on the constraint set to obtain a feasible solution space; Constructing an objective function for the feasible solution space by using the Lagrange multiplier method to obtain an optimized objective function, and performing gradient calculation on the optimized objective function to obtain a gradient vector group; Determining a search direction for the gradient vector group to obtain a search direction sequence, and calculating a step length for the search direction sequence to obtain an iterative step length table; Iteratively optimizing the parameters of the iterative step table through nonlinear programming to obtain an iterative solution sequence, and verifying the convergence of the iterative solution sequence to obtain a converged solution set; Performing a multi-objective trade-off analysis on the MEMS sensor based on the converged solution set to obtain a trade-off coefficient matrix, and reconstructing a solution space based on the trade-off coefficient matrix to obtain a reconstructed solution sequence; The reconstructed solution sequence is mapped to control instructions by a preset control sequence generator to obtain a control instruction group, and the MEMS sensor is time-scheduled based on the control instruction group to obtain an optimal power consumption control sequence; wherein the optimal power consumption control sequence includes a voltage control sequence, a frequency control sequence and a mode control sequence.

5. The low power consumption optimization method of the MEMS sensor according to claim 4, characterized in that: The performing timing scheduling planning on the MEMS sensor based on the control instruction group to obtain an optimal power consumption control sequence includes: Performing instruction dependency analysis on the control instruction group to obtain an instruction dependency graph, and performing key path extraction based on the instruction dependency graph to obtain a key instruction linked list; Performing execution timing analysis on the key instruction chain list through timing constraint analysis to obtain an execution timing diagram, and detecting whether there is resource competition in the control instruction group based on the execution timing diagram; If so, performing conflict resolution processing on the control instruction group to obtain a conflict-free instruction group, and performing parallelism analysis on the conflict-free instruction group to obtain a parallel scheduling table; The parallel scheduling table is allocated with time slices to obtain a time slice allocation scheme, and the MEMS sensor is scheduled with timing based on the time slice allocation scheme to obtain an optimal power consumption control sequence.

6. The low power consumption optimization method of a MEMS sensor according to claim 1, characterized in that: The dynamically adjusting the power consumption of the MEMS sensor based on the optimal power consumption control sequence to obtain a target power consumption state of the MEMS sensor includes: Performing timing analysis on the optimal power consumption control sequence to obtain a control timing table, and performing state migration analysis based on the control timing table to obtain a migration path set; Performing voltage regulation parameter configuration on the migration path set through a preset voltage regulator to obtain a voltage regulation sequence, and performing frequency dynamic compensation based on the voltage regulation sequence to obtain a frequency compensation vector; Decoding the frequency compensation vector according to the working mode to obtain a mode switching instruction set, and performing clock gating configuration on the mode switching instruction set to obtain a gating configuration table; By means of a power consumption monitoring unit, real-time power consumption sampling is performed on the MEMS sensor based on the gating configuration table to obtain a power consumption sampling sequence, and deviation analysis is performed on the power consumption sampling sequence to obtain deviation characteristic data; Calculating compensation parameters for the deviation characteristic data to obtain a compensation vector group, and reconstructing adjustment instructions based on the compensation vector group to obtain a reconstructed instruction sequence; Through a dynamic feedback mechanism, the power consumption state of the data is adjusted based on the reconstruction instruction sequence to obtain a target power consumption state of the MEMS sensor.

7. A low power consumption optimization device for a MEMS sensor, characterized in that: include: An extraction module, used to perform adaptive threshold compression on the output signal of the MEMS sensor to obtain a compressed data stream, and perform time domain feature extraction on the compressed data stream to obtain a signal activity feature set; An analysis module, configured to perform energy consumption point analysis on the MEMS sensor based on the signal activity feature set to obtain a power consumption distribution map; A mapping module, used to perform hierarchical energy efficiency configuration mapping on the MEMS sensor based on the power consumption distribution map to obtain an energy efficiency configuration matrix; A solution module, used for performing nonlinear optimization on the energy efficiency configuration matrix to obtain an optimal power consumption control sequence; An adjustment module, configured to dynamically adjust the power consumption of the MEMS sensor based on the optimal power consumption control sequence to obtain a target power consumption state of the MEMS sensor; Dividing the signal activity feature set into time segments to obtain an activity time segment group, and calculating energy consumption features of the activity time segment group to obtain an energy consumption feature point set; Performing energy efficiency correlation analysis on the energy consumption feature point set through a preset voltage-frequency mapping table to obtain an energy efficiency correlation matrix, and classifying energy consumption characteristics of the MEMS sensor based on the energy efficiency correlation matrix to obtain an energy consumption type sequence; Performing spatiotemporal correlation analysis on the energy consumption type sequence to obtain an energy consumption correlation graph, performing spatial distribution statistics on the energy consumption correlation graph through energy consumption density calculation to obtain an energy consumption density graph, and locating energy consumption hot spots based on the energy consumption density graph to obtain an energy consumption hot spot set; Perform time domain evolution analysis on the energy consumption hotspot set to obtain an energy consumption evolution sequence, perform energy consumption trend prediction based on the energy consumption evolution sequence to obtain an energy consumption trend graph, and generate a power consumption spectrum based on the energy consumption trend graph to obtain a power consumption distribution spectrum.

8. A computer device comprising a memory and a processor, wherein a computer program is stored in the memory, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are 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 steps of the method according to any one of claims 1 to 6 are implemented.

Images