Dynamic power consumption optimization method, module and system and storage medium

Abstract

The invention relates to the technical field of child smart watch power consumption optimization, and discloses a dynamic power consumption optimization method, module and system and a storage medium. The method comprises the following steps: acquiring behavior states of a user, such as indoor rest, indoor motion, outdoor rest, outdoor walking, outdoor riding, subway mode and the like; establishing a mapping relation between each behavior state and a corresponding positioning power consumption strategy, wherein different states correspond to different positioning module working modes, starting frequencies and auxiliary component cooperation modes; based on behavior state changes, a power consumption positioning strategy is dynamically switched through the steps of triggering, pre-judging, smooth transition and error correction, and a probability model is established in combination with a time rule to assist optimization. The problem of power consumption waste caused by a traditional fixed frequency positioning strategy is solved, dynamic balance of positioning precision and power consumption is achieved, the endurance time of the child smart watch is effectively prolonged, the use experience of a user is improved, and the method is suitable for a positioning power consumption control scene of the child smart watch.

Description

Technical Field

[0001] This application relates to the technical field of power consumption optimization technology for smart devices, and in particular to a dynamic power consumption optimization method, module, system, and storage medium. Background Technology

[0002] The children's smartwatch industry focuses on child safety monitoring and convenient interaction. A key pain point is the limited battery capacity due to device size constraints. The GPS positioning module, as a core functional component, accounts for over 60% of power consumption, and battery life directly impacts user experience. As parents demand higher accuracy in tracking their children's movements, balancing power consumption and performance in positioning functions has become a key technological challenge for the industry.

[0003] Currently, there are two improved technologies in the industry: the first is power consumption adjustment technology based on geofencing, which reduces the positioning frequency within the preset safe zones such as home and school, and increases the frequency outside the zones; the second is adaptive positioning technology based on pedometer data, which judges the movement status according to the changes in steps and dynamically adjusts the positioning interval.

[0004] The former has obvious drawbacks: geofencing relies on preset locations and cannot cope with temporary activity scenarios such as playing in a park, and positioning gaps or power waste are likely to occur when switching area boundaries; the latter is limited by relying on only single step count data and cannot distinguish non-walking movement states such as taking a car or subway, resulting in insufficient targeting of the positioning strategy. At the same time, in static scenarios with no change in step count, it is difficult to accurately judge the indoor and outdoor environment, resulting in limited power consumption optimization effects. Summary of the Invention

[0005] To address the insufficient adaptability of traditional fixed-frequency positioning strategies, this application provides a dynamic power consumption optimization method, module, system, and storage medium.

[0006] In a first aspect, this application provides a dynamic power consumption optimization method, comprising the following steps: The user's behavioral state is obtained by collecting environmental signal data and motion data and performing multi-dimensional analysis and identification. A mapping relationship is established between the behavioral state and the positioning power consumption strategy, which is determined by adjusting the positioning mode, the start frequency, and the coordination method of the auxiliary positioning components. The system monitors changes in user behavior in real time. When a change in behavior is detected, the system dynamically switches the positioning power consumption strategy based on the mapping relationship to achieve a balance between power consumption optimization and trajectory tracking quality.

[0007] By adopting the above technical solution, a technical effect of achieving full-scene adaptive power consumption optimization and precise balance of trajectory tracking quality is achieved. Its working principle is as follows: First, through multi-source data acquisition and multi-dimensional analysis, the user's current behavioral state is identified in real time; then, based on a preset mapping relationship, the behavioral state is dynamically associated with the corresponding positioning power consumption strategy; finally, by continuously monitoring changes in the behavioral state, a strategy switching mechanism is triggered to ensure trajectory continuity in motion scenarios and minimize power consumption in stationary scenarios, forming a closed-loop optimization logic of "state recognition - strategy matching - dynamic adjustment," fundamentally solving the problem of insufficient adaptability of traditional fixed-frequency positioning strategies.

[0008] Furthermore, the step of obtaining the user's behavior state specifically includes: Collect environmental signal data and motion data. The environmental signal data includes GPS signal strength, WiFi hotspot distribution, and base station block code. The motion data includes step count, movement speed, and motion trajectory continuity. Feature extraction and threshold judgment are performed on the environmental signal data and motion data, and multi-dimensional analysis is conducted on the data to identify behavioral states; Output the behavioral state and the corresponding confidence score.

[0009] By adopting the above technical solution, high accuracy in behavioral state recognition and reliable confidence quantification are achieved. Its working principle is as follows: Simultaneously, environmental signal data such as GPS signal strength, WiFi hotspot distribution, and base station block codes are collected, along with motion data such as step count, movement speed, and trajectory continuity. Through feature extraction and threshold judgment, six typical behavioral states are accurately distinguished. Simultaneously, a confidence score is calculated based on feature matching consistency, providing a reliable basis for subsequent strategy switching and avoiding power waste or positioning failure due to misjudgment of the state.

[0010] Furthermore, the step of establishing the mapping relationship between behavioral states and positioning power consumption strategies specifically includes: The data is analyzed from multiple dimensions to identify behavioral states, including at least one of the following: indoor stillness, indoor movement, outdoor stillness, outdoor walking, outdoor riding in a vehicle, and subway mode. For indoor static conditions, a deep sleep strategy is configured to execute the operation of turning off GPS and WiFi, and control the base station signal reception and step counting functions to work at a preset low frequency cycle, so that the processor enters a deep sleep mode. And / or, for indoor motion states, configure a low-frequency positioning strategy, perform GPS shutdown operation, and control WiFi and base station signal reception to scan intermittently in a first preset period; And / or, for stationary outdoor conditions, configure a low-frequency positioning strategy to control GPS to start intermittently at a second preset cycle, and combine auxiliary data from WiFi and base station signal reception to confirm location; And / or, for outdoor walking conditions, configure a high-frequency positioning strategy, continuously turn on GPS, and perform positioning at a third preset cycle; And / or, for outdoor vehicle travel, configure a medium-to-high frequency positioning strategy, continuously turn on GPS, perform positioning in the fourth preset cycle, and combine movement speed data to help determine the positioning frequency; And / or, for subway mode, configure an alternative positioning strategy, turn off GPS, and rely on base station block code changes and WiFi to scan in a fifth preset cycle to achieve trajectory tracking; Among them, the first preset period is greater than the second preset period, the second preset period is greater than the fourth preset period, the fourth preset period is greater than the third preset period, and the fifth preset period is between the first preset period and the second preset period.

[0011] By adopting the above technical solution, we achieve the technical effect of highly adaptable positioning strategy and behavior state, and refined power consumption classification. Its working principle is as follows: Differentiated strategy parameters are set for different behavior states. Through periodic priority configuration, high-frequency strategies are used for high trajectory demand scenarios, and low-frequency strategies are used for low power consumption scenarios, thus achieving a precise trade-off between power consumption and performance.

[0012] Furthermore, the step of dynamically switching the positioning power consumption strategy based on changes in behavioral state specifically includes a triggering step: When a change in behavior state is detected, a state change event is generated, which includes the new behavior state and the corresponding confidence score. Based on the aforementioned state change event, the positioning power consumption strategy switching process is initiated.

[0013] By adopting the above technical solution, the system achieves timely response to state changes and a rigorous policy switching triggering mechanism. Its working principle is as follows: the behavior state recognition module continuously monitors changes in data characteristics and instantly generates a state change event containing the new state and confidence score; this event triggers the policy switching process through a system interrupt mechanism, ensuring the synchronization of state changes and policy adjustments, avoiding positioning gaps or power waste caused by response delays, and guaranteeing a consistent user experience.

[0014] Furthermore, the steps of the dynamic switching positioning power consumption strategy also include a prediction step: After receiving the status change event, determine whether the confidence score has reached a preset confidence threshold; When the confidence score reaches the preset confidence threshold, the target positioning power consumption strategy corresponding to the new behavior state is preloaded, and the environment adaptability verification is started at the same time. The environmental adaptability verification includes collecting corresponding environmental parameters for feasibility testing based on the target positioning power consumption strategy. If the test results meet the preset adaptability conditions, the positioning power consumption strategy is switched. If the test results do not meet the preset adaptability conditions, the transitional positioning strategy is entered.

[0015] By adopting the above technical solution, the technical effects of accurate policy switching prediction and avoidance of unnecessary power consumption are achieved. Its working principle is as follows: After receiving a state change event, the confidence score is first compared with a preset confidence threshold; if the score meets the standard, the target policy is preloaded and environmental adaptability verification is initiated; if the verification passes, the switching is executed; if it fails, a transitional policy is entered, forming a logical closed loop of "threshold judgment - policy preloading - environmental verification - accurate execution," preventing policy mismatch and power waste caused by environmental incompatibility.

[0016] Furthermore, the environmental adaptability verification specifically includes: When the target positioning power consumption strategy is a high-frequency positioning strategy, a signal pre-scan is performed for a preset duration to collect GPS signal strength data. If the GPS signal strength data is greater than the preset signal strength threshold, it is determined that the preset adaptation conditions are met. If the GPS signal strength data is less than or equal to a preset signal strength threshold, it is determined that the preset adaptation conditions are not met, and an outdoor pending confirmation transitional positioning strategy is entered, controlling the GPS to intermittently start positioning during the transition period. The transition period is greater than the third preset period and less than the second preset period, and the preset duration is 3-5 seconds.

[0017] By adopting the above technical solution, the high-frequency positioning strategy achieves strong environmental adaptability and avoids power consumption in signal blind spots. Its working principle is as follows: When the target strategy is high-frequency positioning, a 3-5 second GPS signal pre-scan is performed to collect signal strength data; if the strength is greater than a preset threshold, environmental adaptability is determined and a switch is executed; if the strength is insufficient, a transitional strategy is entered, which can attempt positioning at 2-minute intervals, avoiding power waste caused by blindly starting high-frequency positioning in areas with weak signals and improving strategy execution efficiency.

[0018] Furthermore, the steps of the dynamic switching positioning power consumption strategy also include a smooth transition step: After confirming the switching of the positioning power consumption strategy, the operating parameters of the positioning component are gradually adjusted using a progressive adjustment method. The gradual adjustment method specifically involves: gradually adjusting the startup frequency, operating power, and data transmission rate of the positioning component according to a preset adjustment gradient, with the duration of each adjustment gradient being a preset transition time, until the operating parameter standard corresponding to the target positioning power consumption strategy is reached.

[0019] By adopting the above technical solution, the technical effects of smooth strategy switching process and avoidance of hardware damage and instantaneous power consumption spikes are achieved. Its working principle is as follows: a gradual adjustment method is adopted to decompose the strategy switching into multiple gradient steps; through gradient transition buffering, instantaneous current surges caused by parameter mutations are avoided, protecting the stability of hardware components, while reducing additional energy consumption during the switching process.

[0020] Furthermore, the step of dynamically switching the positioning power consumption strategy also includes an error correction step: After switching to the target positioning power consumption strategy, start the effectiveness monitoring timer and set the monitoring duration; During the monitoring period, real-time feedback data of the positioning component is collected, including positioning effectiveness, signal stability, and power consumption. If the positioning effectiveness is lower than the preset effective threshold or the signal stability is lower than the preset stable threshold during the monitoring period, the current positioning power consumption strategy is determined to be abnormal and will automatically revert to the positioning power consumption strategy before the switch. The monitoring duration is set to 60 seconds, the preset effective threshold is 3 consecutive successful positioning attempts, and the preset stable threshold is a signal fluctuation amplitude of no more than 20%.

[0021] By adopting the above technical solution, the system achieves real-time error correction for strategy operation anomalies and avoidance of prolonged high-power consumption due to ineffectiveness. Its working principle is as follows: after switching to the target strategy, a 60-second effectiveness monitoring timer is started to collect real-time data on positioning effectiveness and signal stability. If any indicator fails to reach the threshold, a rollback mechanism is triggered to restore the original strategy, preventing continuous high-power operation due to sudden environmental changes or misjudgments, thus forming a closed-loop fault-tolerant control.

[0022] Furthermore, it also includes auxiliary optimization steps that incorporate time patterns: Record user's historical behavior state data and establish a time-state probability model, which includes the matching probability of different time periods and corresponding behavior states; Based on the current time information, the time-state probability model is queried to obtain the high-probability behavioral state for the current time period; The activation conditions and parameter configuration of the high-probability behavior state fine-tuning positioning power consumption strategy.

[0023] By adopting the above technical solution, we have achieved the technical effect of strong forward-looking strategy optimization and high adaptability to children's daily routines. Its working principle is as follows: by recording historical behavioral data, a time-state probability model is established; based on the current time, the model is queried to obtain the high-probability state, and the strategy activation conditions are fine-tuned to reduce unnecessary strategy switching, thereby improving the accuracy of power consumption optimization and user experience.

[0024] Furthermore, the step of performing multi-dimensional analysis of the data to identify behavioral states specifically includes: Feature extraction is performed on the collected environmental signal data and motion data to obtain signal strength features, motion frequency features, position change features, and velocity features; The features are input into a preset behavior state recognition model for classification and recognition, and the corresponding behavior state and confidence score are output. And / or, the data is analyzed by using preset feature threshold judgment rules, feature threshold ranges corresponding to each behavioral state are set, the extracted features are matched with the feature threshold ranges, and the behavioral state and confidence score are determined based on the matching results.

[0025] Furthermore, the first preset period is set to 10 minutes, the second preset period is set to 5 minutes, the third preset period is set to 30 seconds, the fourth preset period is set to 1 minute, and the fifth preset period is set to 2-3 minutes. And / or, the first preset period, the second preset period, the third preset period, the fourth preset period and the fifth preset period can be adjusted through user-defined configuration, and the start frequency of the positioning power consumption strategy can be updated according to the period parameters input by the user.

[0026] By adopting the above technical solution, we achieve a flexible behavioral state recognition method with strong adaptability to complex scenarios. Its working principle is as follows: it supports a dual-path approach of model recognition and threshold judgment. Model recognition processes feature data through machine learning algorithms, making it suitable for complex data patterns; threshold judgment quickly matches data within a preset feature range, making it suitable for simple scenarios. These two methods complement each other to improve recognition efficiency and accuracy.

[0027] Furthermore, the preset confidence threshold is set to 80 points, and the confidence score adopts a quantitative standard of 0-100 points; And / or, the preset confidence threshold is dynamically adjusted based on the time-state probability model. When the current time period corresponds to a high-probability behavior state, the preset confidence threshold is reduced to 70 points, and when the current time period corresponds to a low-probability behavior state, the preset confidence threshold is increased to 85 points.

[0028] By adopting the above technical solution, the system achieves flexible positioning cycle configuration and strong adaptability to personalized needs. Its working principle is as follows: the default cycle meets general scenario requirements, and users can adjust the cycle parameters through a custom interface; the system updates the strategy frequency based on user configuration, while simultaneously recording data to optimize the time-state model, making the cycle setting more aligned with actual usage habits.

[0029] Furthermore, the preset adjustment gradient is set to 3-5 levels; And / or, the preset adjustment gradient and preset transition time are dynamically configured according to the hardware performance parameters of the positioning component, including the maximum startup rate, power consumption adjustment range and signal response delay, and the optimal adjustment gradient and transition time are calculated according to the hardware performance parameters.

[0030] By adopting the above technical solution, the technical effects of dynamic adaptation of confidence threshold and improved accuracy of state judgment are achieved. Its working principle is as follows: a fixed threshold ensures a consistent judgment benchmark, while a dynamic threshold is adjusted based on a time-state model, enabling state judgment to balance efficiency and accuracy, and avoiding abnormal power consumption caused by erroneous switching.

[0031] Furthermore, the step of establishing the time-state probability model includes: Machine learning algorithms are used to train the matching probability of the time-state probability model on the user's historical behavior state data. And / or, using statistical analysis methods, frequency statistics are performed on historical behavioral states within the same time period, and the behavioral state with the highest frequency is taken as the high-probability behavioral state for that time period, and the time-state probability model is updated periodically.

[0032] By adopting the above technical solution, the technical effects of strong hardware adaptability of transition parameters and guaranteed switching stability are achieved. Its working principle is as follows: the default gradient and duration are adapted to conventional hardware; if hardware performance parameters are abnormal, the optimal parameters are dynamically calculated to ensure smooth switching between hardware with different performance levels and avoid hardware compatibility issues.

[0033] Furthermore, the positioning validity is determined by the longitude and latitude deviation values ​​of the positioning data. When the deviation value is less than a preset deviation threshold, the positioning is determined to be valid. And / or, the positioning validity is determined by the number of satellite receptions; when the number of satellite receptions is greater than a preset satellite number threshold, the positioning is deemed valid.

[0034] By adopting the above technical solution, the technical effects of comprehensive positioning validity judgment and avoidance of power consumption due to invalid positioning are achieved. Its working principle is as follows: positioning accuracy is ensured by judging the deviation value, and signal reliability is ensured by judging the number of satellites; the dual standards complement each other to cover different environments, avoid power waste caused by invalid positioning, and ensure the accuracy of trajectory data.

[0035] Secondly, this application provides a dynamic power consumption optimization module, comprising: A behavior state acquisition unit is used to acquire the user's behavior state, which is obtained by collecting environmental signal data and motion data and performing multi-dimensional analysis and identification. The strategy mapping unit is used to establish a mapping relationship between the behavior state and the positioning power consumption strategy, which is determined by adjusting the positioning mode, the start frequency and the coordination method of the auxiliary positioning components. The dynamic switching unit is used to monitor changes in user behavior status in real time. When a change in behavior status is detected, the positioning power consumption strategy is dynamically switched based on the mapping relationship to achieve a balance between power consumption optimization and trajectory tracking quality.

[0036] Thirdly, this application provides a dynamic power consumption optimization system, comprising: Data acquisition device, used to collect environmental signal data and motion data; The processor is communicatively connected to the data acquisition device; The memory stores a computer program that, when executed by the processor, implements a dynamic power consumption optimization method.

[0037] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the dynamic power consumption optimization method as described above. Attached Figure Description

[0038] Figure 1 A schematic diagram of the dynamic power consumption optimization method steps of this embodiment of the first aspect of this application is shown.

[0039] Figure 2 A schematic diagram of a dynamic power consumption optimization system provided in the second aspect of this application is shown.

[0040] Figure 3 A schematic diagram of a dynamic power consumption optimization module provided in the third aspect of this application is shown.

[0041] Figure label: 201. Data acquisition device; 2011. Environmental signal acquisition module; 2012. Motion data acquisition module; 202. Processor; 203. Memory; 204. Positioning component; 2041. GPS module; 2042. Wi-Fi module; 2043. Base station signal receiving module; 205. Barometric pressure sensor; Bluetooth module; 207. Power consumption sensor; 301. Behavior state acquisition unit; 302. Strategy mapping unit; 303. Dynamic switching unit Detailed Implementation The present application will be further described in detail below with reference to specific embodiments. It should be noted that the embodiments are only used to explain the present application and are not intended to limit the scope of protection of the present application. The core of the dynamic power consumption optimization method described in this application lies in accurately identifying the user's behavior state, establishing a dynamic mapping relationship between the state and the positioning power consumption strategy, achieving a balance between power consumption and positioning accuracy, and extending the watch's battery life. The following embodiments will disclose in detail the complete process, core components, parameter configurations, and alternative solutions for implementing this method, ensuring that those skilled in the art can replicate the method based on the description.

[0042] In the description of this application, it should be noted that the terms "connection" and "communication connection" can refer to wired connections such as flexible circuit board connections, or wireless connections such as Bluetooth and near-field communication; any connection method that can realize data transmission and signal interaction is acceptable. "Preset parameters" and "thresholds" can be adjusted according to the actual application scenario, including application scenarios such as the hardware performance of different brands of watches and the user's geographical location, and are not fixed values. "Positioning components" include core positioning-related components such as a Global Positioning System (GPS) module, a WiFi (Wireless Fidelity) module, and a base station signal receiving module. "Processor" can be a chip with data processing capabilities, such as a microcontroller unit (MCU) or a central processing unit (CPU).

[0043] Firstly, this application provides a dynamic power consumption optimization method.

[0044] Example 1 Reference Figure 1 The dynamic power consumption optimization method in this embodiment includes the following steps: S1: Obtain User Behavior Status. This step achieves accurate status identification through multi-source data collection and analysis. Specifically, it involves collecting environmental signal data and motion data, performing multi-dimensional analysis and identification, and obtaining the user's behavior status and corresponding confidence score. This includes the following sub-steps: S11: Collect environmental signal data and motion data.

[0045] The following environmental signal data and motion data are collected synchronously. Environmental signal data includes GPS signal strength, WiFi hotspot distribution, and base station block codes. In one embodiment, the GPS signal strength collection frequency can be 1 time / second, the WiFi hotspot scanning frequency can be 1 time / 5 seconds, and the base station block code collection frequency can be 1 time / 10 seconds. In another embodiment, the frequency can be adjusted according to hardware performance: the GPS signal strength collection frequency can be 1 time / 2 seconds, the WiFi hotspot scanning frequency can be 1 time / 8 seconds, the base station block code collection frequency can be 1 time / 15 seconds, and the step count collection frequency can be once per minute, as long as data timeliness is guaranteed. The collected data is transmitted to the processor via UART and I2C interfaces and temporarily stored in memory. Specifically, GPS signal strength can be collected by the GPS module, WiFi hotspot distribution can be obtained by scanning the service set identifiers of surrounding WiFi hotspots using the WiFi module, and base station block codes can be collected by the cell identification code of the currently accessed base station using the base station signal receiving module.

[0046] Motion data includes steps, movement speed, and trajectory continuity. Steps can be collected by a six-axis sensor; movement speed can be calculated from acceleration data collected by the six-axis sensor; trajectory continuity can be determined by continuously collected position data, and is expressed as the ratio of the length of the continuous trajectory segment to the total collection time.

[0047] This step, which includes "environmental signal data (GPS signal strength, WiFi hotspot distribution, base station block codes)," "motion data (step count, movement speed, and trajectory continuity)," "feature extraction and threshold judgment," and "output confidence score," ensures the comprehensiveness and accuracy of behavioral state recognition. Environmental signal data is used to determine the indoor and outdoor environment, while motion data is used to determine the motion state. Feature extraction and threshold judgment are the core methods for recognition, and the confidence score provides a reliable basis for subsequent strategy switching. Environmental signal data acquisition can support simultaneous parsing of data from two satellite navigation systems, such as GPS+GLONASS (Global Navigation Satellite System) or GPS+BDS (BeiDou Navigation Satellite System), to improve the availability and accuracy of positioning signals. Motion data acquisition can use a three-axis sensor instead of a six-axis sensor; any device capable of performing the corresponding data acquisition function is acceptable. Behavioral state recognition can use neural network models, clustering algorithms, etc., instead of support vector machine models or decision tree models; any method capable of behavioral state recognition is acceptable.

[0048] S12: Perform feature extraction and threshold judgment on the environmental signal data and motion data, and analyze and identify the behavior state from multiple dimensions.

[0049] Feature extraction and threshold judgment are performed on collected environmental signal data and motion data to identify at least one behavioral state among indoor stillness, indoor movement, outdoor stillness, outdoor walking, outdoor riding in a vehicle, and subway mode. Specifically, this includes feature extraction and threshold judgment. Feature extraction includes extracting signal strength features, motion frequency features, position change features, and speed features. In one embodiment, a sliding window method is used for feature extraction, with a window length of 10 seconds; in another embodiment, the window length can be set to 15 seconds, as long as stable feature data can be extracted. Signal strength features can be the average GPS signal strength or the average WiFi signal strength; motion frequency features can be the step rate of change or the acceleration frequency of change; position change features can be the position offset per unit time; and speed features can be the average movement speed or the speed fluctuation amplitude.

[0050] Threshold determination involves setting preset feature threshold ranges for each behavior state, matching the extracted features against these threshold ranges, and then determining the behavior state. Typical threshold range settings for each behavior state are as follows: Indoor stationary state usually means no effective GPS signal and multiple WiFi hotspots are usually present indoors. At this time: GPS signal strength ≤ -120dBm, number of WiFi hotspots ≥ 3, step change rate = 0, moving speed ≤ 1m / s; Indoor movement typically means no effective GPS signal, multiple WiFi hotspots are usually present indoors, and there is a change in step count, but the indoor movement speed is relatively slow. At this time: GPS signal strength ≤ -120dBm, number of WiFi hotspots ≥ 3, step count change rate > 0, and movement speed 1m / s < v ≤ 1.5m / s; An outdoor stationary state usually means that there is a valid GPS signal, few outdoor WiFi hotspots, and no change in step count. At this time: GPS signal strength > -120dBm, number of WiFi hotspots < 3, step count change rate = 0, and moving speed ≤ 1m / s; Outdoor walking status typically means there is a valid GPS signal, few outdoor WiFi hotspots, changes in step count, and a wide range of walking speed. At this time: GPS signal strength > -120dBm, number of WiFi hotspots < 3, step count change rate > 0, and walking speed 1.5 < v ≤ 5m / s; Outdoor vehicle travel typically means there is a valid GPS signal, few outdoor WiFi hotspots, no significant step count while traveling, no change in step count, and travel speed is usually greater than walking speed. At this time, the GPS signal strength is > -120dBm, the number of WiFi hotspots is <3, the step count change rate is 0, and the moving speed is >5m / s. Subway mode typically means that there is no GPS signal inside the subway, there are usually no external WiFi hotspots inside the subway, the frequency of change is greater than 1 time / minute, and the subway operating speed range is relatively small. At this time, the GPS signal strength is ≤-130dBm, the number of WiFi hotspots is 0, the base station block code changes frequently, and the moving speed is 5<v≤30m / s.

[0051] In another embodiment, classification and identification can be performed using a preset behavior state recognition model. The extracted features are input into the model, and the model outputs the corresponding behavior state. The feature input model can be a support vector machine model or a decision tree model. For example, when using a decision tree model, GPS signal strength, movement speed, and step change rate are used as decision nodes to progressively determine the behavior state. Any method capable of achieving behavior state recognition is acceptable. The aforementioned GPS signal strength, number of WiFi hotspots, step change rate, and movement speed can be adjusted according to the actual application scenario.

[0052] S13: Output behavioral status and confidence score.

[0053] The processor calculates a confidence score based on the degree of matching between features and threshold ranges. The score uses a quantification standard of 0-100 points, with higher matching degrees resulting in higher scores. In one embodiment, if all features meet the threshold range for a certain behavioral state, the confidence score is 100 points; for each feature that does not meet the threshold range, the score decreases by 20 points, down to 0 points. In another embodiment, a weighted scoring method can be used, assigning higher weights to important features and lower weights to secondary features, and calculating the confidence score based on the weighted matching degree. The processor outputs the identified behavioral state and its corresponding confidence score and stores it in memory. The aforementioned important features can be GPS signal strength or moving speed, with higher weights of 0.3 or 0.4; secondary features can be the number of WiFi hotspots or the step change rate, with lower weights of 0.1 or 0.2, as long as the higher weight is greater than the lower weight.

[0054] S2: Establish the mapping relationship between behavioral state and location power consumption strategy.

[0055] This step determines the positioning power consumption strategy by adjusting the positioning mode, startup frequency, and the coordination method of auxiliary positioning components, and establishes its mapping relationship with the behavioral state. Specifically, it configures differentiated positioning power consumption strategies for different behavioral states, and the parameter configurations for each strategy are as follows: For indoor stationary states: Configure a deep sleep strategy. The processor shuts down the GPS and WiFi modules, controls the base station signal receiving module and pedometer function to operate at a preset low-frequency cycle, and simultaneously enters deep sleep mode. Specifically, in deep sleep mode, the processor power consumption is ≤2mA. In one embodiment, the preset low-frequency cycle is set to 10 minutes, meaning the base station signal receiving module starts collecting base station block codes every 10 minutes, and the pedometer function starts collecting step counts every 10 minutes. In another embodiment, the preset low-frequency cycle can be set to 15 minutes, or any cycle that can guarantee basic positioning needs and has low power consumption is acceptable. The basic positioning needs can specifically be obtaining a general location in an emergency.

[0056] For indoor movement: Configure a low-frequency positioning strategy. The processor disables the GPS module and controls the WiFi module and base station signal receiving module to scan intermittently at a first preset period. In one embodiment, the first preset period is set to 10 minutes; in another embodiment, it can be set to 8 minutes, as long as low-frequency positioning is satisfied and a general trajectory can be tracked.

[0057] For stationary outdoor conditions: a low-frequency positioning strategy is configured. The processor controls the GPS module to start intermittently at a second preset period, combining auxiliary data collected by the WiFi module and the base station signal receiving module to confirm the location. In one embodiment, the second preset period is set to 5 minutes; in another embodiment, it can be set to 6 minutes, but any period that ensures accurate positioning and low power consumption in a stationary state is acceptable.

[0058] For outdoor walking: Configure a high-frequency positioning strategy. The processor controls the GPS module to remain continuously active, performing positioning at a third preset cycle. In one embodiment, the third preset cycle is set to 1 minute; in another embodiment, it can be set to 30 seconds, as long as the walking trajectory can be continuously and accurately tracked.

[0059] For outdoor vehicle travel: a medium-to-high frequency positioning strategy is configured. The processor controls the GPS module to remain continuously active, performing positioning at a fourth preset cycle. Simultaneously, it incorporates movement speed data to assist in determining the positioning frequency; for example, the faster the speed, the higher the positioning frequency can be. In one embodiment, the fourth preset cycle is set to 1 minute; in another embodiment, it can be set to 45 seconds, but any cycle that ensures accurate vehicle trajectory and reasonable power consumption is acceptable.

[0060] For subway mode: Configure an alternative positioning strategy. The processor disables the GPS module and relies on changes in base station block codes collected by the base station signal receiving module and the results of scanning by the WiFi module at a fifth preset cycle to achieve trajectory tracking. Although there is no GPS signal in the subway, the base station block codes change frequently with the subway's movement, allowing location determination through these changes. In one embodiment, the fifth preset cycle is set to 2-3 minutes, specifically 2.5 minutes; in another embodiment, it can be set to 3-4 minutes, or any cycle sufficient to ensure subway trajectory tracking. The order of the preset cycles is as follows: the first preset cycle is greater than the second, the second is greater than the fourth, the fourth is greater than the third, and the fifth is between the first and second. For example, a first preset cycle of 10 minutes > a second preset cycle of 5 minutes > a fourth preset cycle of 2 minutes > a third preset cycle of 30 seconds, and a fifth preset cycle of 2.5 minutes is between 10 minutes and 5 minutes. Furthermore, each preset cycle can be adjusted through user-defined configuration. Users can input cycle parameters via the watch app, and the processor updates the activation frequency of the positioning power consumption strategy after receiving the parameters.

[0061] The processor stores the mapping relationship between the above behavioral states and the positioning power consumption strategy in memory to form a mapping table, which facilitates quick subsequent lookup and retrieval.

[0062] The steps of "obtaining user behavior status" and "establishing a mapping relationship between behavior status and positioning power consumption strategy" are the foundation and core of this application. "Obtaining user behavior status" achieves accurate status identification through multi-source data collection and analysis; "establishing a mapping relationship" achieves accurate matching between status and strategy. The aforementioned "cycle length relationship corresponding to each behavior status and positioning power consumption strategy parameters such as positioning mode, startup frequency, and auxiliary component collaboration method" realizes the hierarchical refinement of power consumption strategy, configuring differentiated strategies for different scenarios. The cycle length relationship ensures that high-frequency strategies are used for high-demand scenarios, and low-frequency strategies are used for low-power scenarios, achieving a precise balance between power consumption and accuracy. Regarding the expansion of the protection scope, various preset cycles, thresholds, durations, and other parameters can be arbitrarily adjusted according to the actual application scenario; any parameter that can achieve a balance between power consumption optimization and positioning accuracy is acceptable.

[0063] S3: Real-time monitoring of changes in user behavior. When a change is detected, the positioning power consumption strategy is dynamically switched based on the above mapping relationship to achieve a balance between power consumption optimization and trajectory tracking quality. This includes triggering steps, prediction steps, smooth transition steps, and error correction steps. S31: The processor monitors the behavioral state output by S1 in real time. When a change in behavioral state is detected, such as changing from a static indoor state to a moving indoor state, a state change event is generated. This event includes the new behavioral state and the corresponding confidence score. Based on this state change event, the processor initiates the positioning power consumption strategy switching process through a system interrupt mechanism. The interrupt mechanism ensures timely response to strategy switching and avoids delays. In one embodiment, the behavioral state monitoring frequency is consistent with the data acquisition frequency of S1; in another embodiment, a dedicated monitoring cycle can be set, such as 1 second / time or 2 seconds / time, as long as the state change can be detected in a timely manner.

[0064] The "State Change Event" and "Initiate Policy Switching Process" steps in this section trigger the switch via a state change event, ensuring the synchronization of state changes and policy adjustments, and avoiding power waste or location gaps caused by delays. This is the fundamental mechanism for achieving dynamic response.

[0065] S32: After receiving a state change event, the processor performs the following prediction operation: confidence threshold judgment. The preset confidence threshold is set to 80 points. The processor judges whether the confidence score of the new behavior state reaches 80 points. In another embodiment, the preset confidence threshold can be dynamically adjusted based on a time-state probability model. When the current time period corresponds to a high-probability behavior state, the threshold is reduced to 70 points; when the current time period corresponds to a low-probability behavior state, the threshold is increased to 85 points to improve the accuracy of state judgment.

[0066] Strategy preloading and environmental adaptability verification: When the confidence score reaches a preset confidence threshold, the processor queries the memory mapping table for the target positioning power consumption strategy corresponding to the new behavior state and preloads the strategy, that is, reads the parameter configuration corresponding to the strategy into the processor cache in advance to shorten the switching time. At the same time, environmental adaptability verification is started, and corresponding environmental parameters are collected according to the target positioning power consumption strategy to perform feasibility testing. The specific method of environmental adaptability verification is determined according to the target positioning power consumption strategy. For example, when the target strategy is a high-frequency positioning strategy corresponding to the outdoor walking state, the processor controls the GPS module to perform a signal pre-scan for a preset duration, which can be 5-10 seconds, specifically 8 seconds, to collect GPS signal strength data. If the collected GPS signal strength data is greater than the preset signal strength threshold, such as -110dBm, it is determined that it meets the preset adaptation conditions; if it is less than or equal to the preset signal strength threshold, it is determined that it does not meet the adaptation conditions, and enters the outdoor pending confirmation transitional positioning strategy. The GPS module is controlled to intermittently start positioning attempts with a transition period, where the transition period is greater than the third preset period and less than the second preset period, such as the transition period being set to 2 minutes. In another embodiment, the preset duration can be set to 7 seconds, the preset signal strength threshold can be set to -105dBm, and the transition period can be set to 1.5 minutes. Any parameter that can achieve environmental adaptability detection is acceptable.

[0067] If the environmental adaptability verification result meets the preset adaptability conditions, the positioning power consumption strategy will be switched; otherwise, the transitional positioning strategy will be entered. The transitional strategy is a temporary strategy that can ensure basic positioning needs while avoiding power waste.

[0068] The "confidence threshold judgment," "strategy preloading," "environmental adaptability verification," and "transitional positioning strategy" steps in this section predict strategy switching, avoiding unnecessary power consumption due to misjudgment of the state or environmental incompatibility, and improving the accuracy of switching. The "specific methods for environmental adaptability verification of high-frequency positioning strategies," such as signal pre-scanning, GPS signal strength judgment, and transitional strategies, address the high power consumption characteristics of high-frequency strategies. Pre-scanning verifies environmental adaptability, avoiding blindly starting high-frequency positioning in areas with weak signals and reducing unnecessary power consumption. Pre-set confidence thresholds, signal strength thresholds, and other parameters can be dynamically adjusted according to the actual scenario, and the verification methods can also be extended to other positioning strategies.

[0069] S33: After confirming the switching of the positioning power consumption strategy, a gradual adjustment method is used to progressively adjust the operating parameters of the positioning components to avoid instantaneous power consumption spikes or hardware damage caused by parameter mutations. Specifically, the startup frequency, operating power, and data transmission rate of the positioning components are gradually adjusted according to a preset adjustment gradient, which can be 3-5 levels, specifically 4 levels. The duration of each adjustment gradient is a preset transition time, which can be 5-10 seconds, specifically 8 seconds, until the operating parameter standard corresponding to the target positioning power consumption strategy is reached. For example, when switching from an indoor static state to an indoor active state, the deep sleep strategy also switches to a low-frequency positioning strategy. The base station signal reception cycle is set to 10 minutes, and the adjustment gradient is set to 4 levels, each lasting 8 seconds: Level 1, the base station signal reception cycle is adjusted to 8 minutes, the WiFi module starts but does not scan; Level 2, the base station signal reception cycle is adjusted to 9 minutes, the WiFi module scans at a 15-minute cycle; Level 3, the base station signal reception cycle is adjusted to 10 minutes, the WiFi module scans at a 12-minute cycle; Level 4, the base station signal reception cycle is maintained at 10 minutes, the WiFi module scans at a 10-minute cycle, and the target parameters are reached. In another embodiment, the preset adjustment gradient can be set to 3 levels, and the preset transition duration can be set to 10 seconds. Any gradient and duration that can achieve a smooth transition is acceptable.

[0070] The "gradual adjustment method," "preset adjustment gradient," and "preset transition duration" in this step avoid instantaneous power consumption spikes and hardware damage caused by sudden parameter changes, ensuring the stability of the switching process. The adjustment gradient and transition duration can be dynamically configured according to the hardware performance parameters of the positioning component, such as calculating the optimal parameters based on the maximum startup rate and power consumption adjustment range of the GPS module, to ensure that watches with different hardware performance can achieve smooth switching.

[0071] S34: After switching to the target positioning power consumption strategy, start the effectiveness monitoring timer and set the monitoring duration, such as 60 seconds. During the monitoring duration, collect real-time feedback data from the positioning component, including positioning effectiveness, signal stability, and power consumption values. Positioning accuracy is determined by the horizontal precision factor (HDOP) parameter of the positioning data. If the HDOP value is less than the preset accuracy threshold, such as HDOP≤10, the positioning is considered valid. Alternatively, the positioning accuracy can be determined by the number of satellites received. If the number of satellites received is greater than the preset satellite number threshold, such as 4, the positioning is considered valid. The signal strength fluctuation amplitude is used to represent the stability of the signal. If the fluctuation amplitude does not exceed a preset stability threshold, such as 20%, the signal is considered stable. Data is collected via the processor's built-in power monitoring module or via an external power sensor.

[0072] If, within the monitoring period, the positioning effectiveness is lower than a preset effective threshold (e.g., three consecutive successful positioning attempts constitute an effective threshold, while failure to reach this threshold indicates ineffectiveness), or if the signal stability is lower than a preset stability threshold, then the current positioning power consumption strategy is deemed to be abnormally adapted. The processor automatically reverts to the positioning power consumption strategy before the switch and records the abnormal information, such as the switch time, environmental parameters, and abnormal type, in memory for subsequent optimization. In another embodiment, the monitoring period can be set to 90 seconds, the preset effective threshold can be set to four consecutive successful positioning attempts, and the preset stability threshold can be set to 15%. Any parameters that can achieve strategy effectiveness monitoring are acceptable.

[0073] The "effectiveness monitoring timer," "work feedback data," and "strategy rollback mechanism" in this step, through monitoring and rollback mechanisms, promptly correct strategy adaptation anomalies, avoid prolonged ineffective high-power operation, and form a closed-loop fault-tolerant control. Positioning effectiveness judgment can be achieved through a combination of methods, such as simultaneously using deviation values ​​and satellite counts to improve the comprehensiveness of the judgment; monitoring parameters can be flexibly adjusted according to the actual application scenario.

[0074] S4: This step establishes a time-state probability model by recording historical user behavior data, assisting in fine-tuning the activation conditions and parameter configuration of the positioning power consumption strategy. Specifically, it includes a data recording step where the processor continuously records the historical user behavior data output in S1, including behavior state, corresponding time, environmental parameters, etc., and stores it in memory. The recording cycle can be set to 1 day / data processing, and the data retention period can be set to 7 days. Old data exceeding 7 days can be automatically deleted to save storage space. The corresponding time can be date, hour, or minute. In another embodiment, the recording cycle can be set to 2 days / time, and the data retention period can be set to 7 days; any recording cycle and retention period that ensures model accuracy is acceptable.

[0075] The model building steps involve using statistical analysis methods to process historical behavioral state data and establish a time-state probability model. Specifically, a day is divided into multiple time periods, such as one hour per period, for a total of 24 time periods. The frequency of each behavioral state within each time period is counted, and the behavioral state with the highest frequency is taken as the high-probability behavioral state for that time period. The matching probability of each behavioral state within that time period is calculated, where the matching probability = frequency of a behavioral state / total number of records in that time period. For example, if statistics show that the user's "outdoor riding in a vehicle" state has the highest frequency between 8:00 and 9:00 on weekdays, with a matching probability of 85%, then the high-probability behavioral state for that time period is "outdoor riding in a vehicle". In another embodiment, machine learning algorithms (such as Bayesian algorithms) can be used to train historical data and update the matching probability of the time-state probability model; any method that can establish a correlation between time and behavioral state is acceptable.

[0076] The strategy fine-tuning step involves the processor querying the time-state probability model based on the current time information to obtain the high-probability behavior state for the current time period. Based on this high-probability behavior state, the activation conditions and parameter configurations of the positioning power consumption strategy are fine-tuned. For example, if the current time period is weekday 8:00-9:00, and the high-probability behavior state is "outdoor vehicle riding," then the confidence threshold for "outdoor vehicle riding" is reduced from 80 points to 70 points, and the fourth preset period is adjusted from 1 minute to 45 seconds to improve the timeliness of strategy switching and positioning accuracy. In another embodiment, the operating power of the positioning component can be fine-tuned, such as appropriately reducing the operating power in high-probability states to further save power consumption. Any fine-tuning method that can assist in optimizing the power consumption strategy is acceptable.

[0077] The aforementioned behavioral state recognition can use neural network models, clustering algorithms, etc., to replace support vector machine models and decision tree models; the time-state probability model can use hidden Markov models to replace statistical analysis methods, and any algorithm that can achieve the corresponding function is acceptable.

[0078] This step, "Time-State Probability Model," "Historical Behavioral State Data," and "Fine-tuning Strategy Activation Conditions and Parameters," leverages the temporal regularity of user behavior to improve the foresight and accuracy of strategy optimization, adapting to children's daily routines. "Feature Extraction," "Behavioral State Recognition Model," and "Feature Threshold Judgment Rules" provide two behavioral state recognition methods suitable for scenarios of varying complexity, enhancing the flexibility and adaptability of recognition. The time-state probability model can use a Hidden Markov Model instead of statistical analysis methods, and the model update cycle can be dynamically adjusted according to the data volume; the behavioral state recognition model can utilize a more advanced deep learning model to improve recognition accuracy.

[0079] Example 2 This embodiment enhances the method steps based on Embodiment 1 to further improve power consumption optimization and positioning accuracy, as detailed below: S1: Behavioral state recognition through multi-sensor fusion.

[0080] In S11, air pressure data acquisition is added, specifically through air pressure sensors. This air pressure data is then used as part of the environmental signal data for analysis. For example, when the fluctuation range is <5 hPa, indoor air pressure is generally relatively stable; when the fluctuation range is >5 hPa, outdoor air pressure fluctuates significantly. This helps in judging the indoor and outdoor environment and improves the accuracy of behavioral state recognition.

[0081] In S12, feature extraction adds air pressure features, such as average air pressure and air pressure fluctuation amplitude. Threshold judgment also incorporates threshold ranges for air pressure features, further refining the judgment conditions for behavioral states. For example, the threshold range for indoor static states is expanded to "air pressure fluctuation amplitude < 5 hPa", and the threshold range for outdoor static states is expanded to "air pressure fluctuation amplitude > 5 hPa", reducing misjudgments of states caused by abnormal GPS signals.

[0082] S2: Positioning power consumption strategy with adaptive parameter configuration.

[0083] When establishing the mapping relationship, the processor combines historical power consumption data collected by the power consumption sensor to adaptively adjust the parameters of each positioning power consumption strategy. For example, if statistics show that the power consumption of a certain positioning component is too high at the current startup frequency, exceeding a preset power consumption threshold, such as 5mAh / h, the startup frequency of that component will be automatically reduced, such as adjusting the third preset cycle from 30 seconds to 40 seconds, to ensure that power consumption is controlled within a reasonable range. If the positioning accuracy is insufficient, and the deviation exceeds a preset deviation threshold, such as 15 meters, the startup frequency will be appropriately increased to achieve a dynamic balance between power consumption and accuracy. At the same time, users can customize the parameter configuration of each strategy through the Bluetooth module of the mobile APP. After receiving the customized parameters, the processor updates the strategy parameters in the mapping table.

[0084] S4: Multidimensional time-state probability model.

[0085] When building a time-state probability model, adding dimensions such as weekday / weekend and seasonal dimensions improves model accuracy. For example, the high-probability behavior state during weekdays (8:00-9:00) is "outdoor transportation," while the high-probability behavior state during weekends (8:00-9:00) is "outdoor walking." Similarly, the high-probability behavior state during summer (18:00-19:00) is "outdoor exercise," while the high-probability behavior state during winter (18:00-19:00) is "indoor stillness." The processor queries the model based on the dimensions corresponding to the current time, such as weekdays and summer, to obtain high-probability behavior states that better reflect users' actual behavioral habits, further improving the accuracy of strategy optimization.

[0086] Example 3 This embodiment targets children's smartwatches with lower hardware costs, simplifying system composition and method steps. While ensuring core power consumption optimization, it reduces implementation costs, as detailed below: The simplified method retains the core functionality, as follows: S1 Simplified: Only collects GPS signal strength, environmental signal data, step count, and movement speed. Movement speed is calculated from the step count collected by the pedometer module. It identifies four core behavioral states: indoor stationary, outdoor stationary, outdoor walking, and outdoor riding in a vehicle. It omits the identification of indoor exercise and subway mode. The confidence score uses a simple matching item counting method. 25 points are awarded for each matching feature. There are four features in total, with a maximum score of 100 points.

[0087] S2 Simplification: Configure positioning power consumption strategies for four core behavior states, simplifying cycle parameters. The first preset cycle is 10 minutes, the second preset cycle is 5 minutes, the third preset cycle is 30 seconds, and the fourth preset cycle is 1 minute. The user-defined configuration function is eliminated, and fixed parameter configuration is adopted.

[0088] S3 Simplification: The triggering step, prediction step, and error correction step are retained, while the smooth transition step is omitted, simplifying it to a direct switching strategy, reducing the processor's computational pressure. In the prediction step, only the confidence threshold is judged, and the threshold is fixed at 80 points, eliminating the need for environmental adaptability verification. The monitoring duration of the error correction step is set to 30 seconds, and the preset effective threshold is set to two consecutive successful positioning attempts, simplifying the monitoring parameters.

[0089] S4 Simplification: This implementation omits the establishment and application of a time-state probability model, switching strategies solely based on real-time behavioral states, thus reducing memory usage and processor computation. While this embodiment simplifies some functions, it still achieves core dynamic power consumption optimization, making it suitable for low-cost children's smartwatch scenarios.

[0090] The scope of protection of this application is not limited to the parameters, materials, and implementation methods disclosed in the above embodiments, but also includes the following alternative solutions: Environmental signal data acquisition can be supplemented or replaced by BeiDou Navigation Satellite System (BDS) signal strength acquisition. Motion data acquisition can be replaced by a three-axis sensor instead of a six-axis sensor. Any device that can perform the corresponding data acquisition function is acceptable.

[0091] The processor can be any chip with data processing capabilities, such as a digital signal processor (DSP) or a field-programmable gate array (FPGA), as long as it can execute the algorithm logic of this application.

[0092] Each preset period, threshold, duration, and other parameters can be adjusted arbitrarily according to the actual application scenario, as long as the parameters can achieve a balance between power consumption optimization and positioning accuracy.

[0093] The aforementioned preset periods can be the first to the fifth preset periods; the threshold can be the confidence threshold or the signal strength threshold; the duration can be parameters such as monitoring duration or transition duration, and the actual application scenarios include watch hardware performance, user region, and usage habits.

[0094] For behavioral state recognition, neural network models and clustering algorithms can be used to replace support vector machine models and decision tree models; for time-state probability models, hidden Markov models can be used to replace statistical analysis methods. Any algorithm that can achieve the corresponding function is acceptable.

[0095] This application is not only applicable to children's smartwatches, but also to other wearable devices that require power optimization for positioning, such as smart bracelets and positioning watches for the elderly. It is applicable to any wearable device scenario that has positioning function and requires power optimization.

[0096] Secondly, embodiments of this application provide a dynamic power consumption optimization system.

[0097] Example 1 Reference Figure 2 The dynamic power consumption optimization system of this embodiment includes a data acquisition device 201, a processor 202, a memory 203, and a positioning component 204. The connection relationship and function of each component are as follows: The data acquisition device 201 and the processor 202 are connected via a flexible circuit board and include an environmental signal acquisition module 2011 for acquiring environmental signal data and a motion data acquisition module 2012 for acquiring motion data. The environmental signal acquisition module 2011 uses a GPS module 2041 (model UBLOXNEO-6M) for acquiring GPS signal strength; a WiFi module 2042 (model ESP8266 or UWS6137) for acquiring WiFi hotspot distribution; a cellular communication module (model SIM800C or UWS6137) for acquiring base station block codes; and a motion data acquisition module 2012 (model MPU6050 or QMI8658B) for acquiring step count, movement speed, and motion trajectory continuity. In another embodiment, the environmental signal acquisition module 2011 may be an UBLOXNEO-7M GPS module 2041 or an RTL8188FTV Wi-Fi module 2042, and the motion data acquisition module 2012 may be a BMI160 or QMI8658B six-axis sensor, or any module that can perform the corresponding data acquisition function.

[0098] The processor 202 is an STM32L476RGT6 microcontroller unit (MCU), which features low power consumption, supports multi-module collaborative control, and establishes communication connections with the data acquisition device 201, positioning component 204, and memory 203. It is used to execute core algorithm logic such as behavior state recognition, strategy mapping, and dynamic switching. In another embodiment, the processor 202 can be a low-power MCU such as an STM32L496VGT6 or Nordic RF52840, as long as it has equivalent data processing capabilities and low power consumption. In yet another embodiment, a digital signal processor (DSP) or field-programmable gate array (FPGA) with data processing capabilities can also be used, as long as it can execute the algorithm logic of this application.

[0099] The aforementioned environmental signal data acquisition can be supplemented or replaced by BeiDou Navigation Satellite System (BDS) signal strength acquisition. The motion data acquisition module 2012 can use a three-axis sensor instead of a six-axis sensor. Any device that can achieve the corresponding data acquisition function is acceptable.

[0100] The memory 203 uses a W25Q64JV flash memory chip and is connected to the processor 202 via a Serial Peripheral Interface (SPI, a high-speed serial communication interface). It is used to store computer programs, historical behavior state data, time-state probability models, and preset parameters such as preset periods or thresholds. In another embodiment, the memory 203 can be a W25Q128JV flash memory chip, an SD card, or any other storage medium capable of data storage and retrieval.

[0101] The positioning component 204 includes the aforementioned GPS module 2041, Wi-Fi module 2042, and base station signal receiving module 2043. It is communicatively connected to the processor 202, receives operating parameter instructions from the processor 202, such as start or stop, operating frequency, etc., and performs positioning data acquisition and transmission.

[0102] Example 2 Based on the system composition of Embodiment 1, a barometric pressure sensor 205, a Bluetooth module 206, and a power consumption sensor 207 are added. The barometric pressure sensor 205 is a BMP280 model, connected to the processor 202, used to collect ambient barometric pressure data to assist in judging indoor and outdoor environments. Since there is a difference in barometric pressure between indoors and outdoors, this improves the accuracy of environmental identification. In another embodiment, an MS5611 barometric pressure sensor 205 can be used; however, any sensor capable of collecting barometric pressure data can be employed.

[0103] The Bluetooth module 206 is selected as model HC-05, which is connected to the processor 202. It is used to receive custom parameters, such as preset periods and thresholds, sent by the user through a mobile APP. At the same time, it uploads the watch's working status, such as current behavior and power consumption, to the mobile APP, enabling remote monitoring and configuration by the user. In another embodiment, an nRF8001 Bluetooth module 206 can be used, or any Bluetooth module 206 capable of data interaction can be used.

[0104] The power consumption sensor 207 is an INA219 model, connected to the processor 202, used to accurately collect the power consumption values ​​of each positioning component, providing more accurate power consumption data for strategy optimization. In another embodiment, an INA226 power consumption sensor 207 can be used, but any sensor capable of power consumption acquisition can be used.

[0105] Example 3 The simplified system consists of a data acquisition device, a processor, a memory, and a positioning component. The data acquisition device retains only the GPS module, pedometer module, and base station signal receiving module; the processor uses a low-cost MCU, such as the STM32F103C8T6, and the memory uses a small-capacity flash memory chip, such as the W25Q16JV; the positioning component shares the GPS module and base station signal receiving module with the data acquisition device. The WiFi module, six-axis sensor, barometric pressure sensor, Bluetooth module, and power consumption sensor are omitted, reducing hardware costs.

[0106] Thirdly, this application provides a dynamic power consumption optimization module.

[0107] Example 1 Reference Figure 3 A dynamic power consumption optimization module includes: The behavior state acquisition unit 301 is used to acquire the user's behavior state, which is obtained by collecting environmental signal data and motion data and performing multi-dimensional analysis and identification. The strategy mapping unit 302 is used to establish a mapping relationship between the behavior state and the positioning power consumption strategy, wherein the positioning power consumption strategy is determined by adjusting the positioning mode, the start frequency and the coordination method of the auxiliary positioning components. The dynamic switching unit 303 is used to monitor changes in user behavior status in real time. When a change in behavior status is detected, the positioning power consumption strategy is dynamically switched based on the mapping relationship to achieve a balance between power consumption optimization and trajectory tracking quality.

[0108] This application solves the power consumption waste problem caused by traditional fixed-frequency positioning strategies through the core logic of "obtaining user behavior status - establishing mapping relationship - dynamically switching positioning power consumption strategy - time-based optimization". The above embodiments disclose in detail the complete process, system composition, parameter configuration, and alternative solutions for implementing this method, covering all technical features of the claims. The explanations of each technical feature are clear and specific, and the positional and connection relationships of the components are explicit, ensuring that those skilled in the art can replicate the implementation based on the description. Furthermore, by expanding alternative solutions such as parameters, materials, and algorithms, the scope of protection of this application is broadened, making it applicable to children's smartwatches and other wearable devices with different hardware costs and application scenarios.

Claims

1. A dynamic power consumption optimization method, characterized in that, Includes the following steps: The user's behavioral state is obtained by collecting environmental signal data and motion data and performing multi-dimensional analysis and identification. A mapping relationship is established between the behavioral state and the positioning power consumption strategy, which is determined by adjusting the positioning mode, the start frequency, and the coordination method of the auxiliary positioning components. The system monitors changes in user behavior in real time. When a change in behavior is detected, the system dynamically switches the positioning power consumption strategy based on the mapping relationship to achieve a balance between power consumption optimization and trajectory tracking quality.

2. The method according to claim 1, characterized in that, The steps for obtaining user behavior status specifically include: Collect environmental signal data and motion data. The environmental signal data includes GPS signal strength, WiFi hotspot distribution, and base station block code. The motion data includes step count, movement speed, and motion trajectory continuity. Feature extraction and threshold judgment are performed on the environmental signal data and motion data, and multi-dimensional analysis is conducted on the data to identify behavioral states; Output the behavioral state and the corresponding confidence score.

3. The method according to claim 1, characterized in that, The steps for establishing the mapping relationship between behavioral states and positioning power consumption strategies specifically include: The data is analyzed from multiple dimensions to identify behavioral states, including at least one of the following: indoor stillness, indoor movement, outdoor stillness, outdoor walking, outdoor riding in a vehicle, and subway mode. For indoor static conditions, a deep sleep strategy is configured to execute the operation of turning off GPS and WiFi, and controlling the base station signal reception and step counting functions to work at a preset low frequency cycle, so that the processor enters a deep sleep mode. And / or, for indoor movement, configure a low-frequency positioning strategy, perform GPS shutdown, and control WiFi and base station signal reception to scan intermittently in a first preset period; And / or, for stationary outdoor conditions, configure a low-frequency positioning strategy to control GPS to start intermittently at a second preset cycle, and combine auxiliary data from WiFi and base station signal reception to confirm location; And / or, for outdoor walking conditions, configure a high-frequency positioning strategy, continuously turn on GPS, and perform positioning at a third preset cycle; And / or, for outdoor vehicle travel, configure a medium-to-high frequency positioning strategy, continuously turn on GPS, perform positioning in the fourth preset cycle, and combine movement speed data to help determine the positioning frequency; And / or, for subway mode, configure an alternative positioning strategy, turn off GPS, and rely on base station block code changes and WiFi to scan in a fifth preset cycle to achieve trajectory tracking; Among them, the first preset period is greater than the second preset period, the second preset period is greater than the fourth preset period, the fourth preset period is greater than the third preset period, and the fifth preset period is between the first preset period and the second preset period.

4. The method according to claim 1, characterized in that, The steps for dynamically switching the positioning power consumption strategy based on changes in behavioral state specifically include a triggering step: When a change in behavior state is detected, a state change event is generated, which includes the new behavior state and the corresponding confidence score. Based on the aforementioned state change event, the positioning power consumption strategy switching process is initiated.

5. The method according to claim 4, characterized in that, The steps of the dynamic switching positioning power consumption strategy also include a prediction step: After receiving the status change event, determine whether the confidence score has reached a preset confidence threshold; When the confidence score reaches the preset confidence threshold, the target positioning power consumption strategy corresponding to the new behavior state is preloaded, and the environment adaptability verification is started at the same time. The environmental adaptability verification includes collecting corresponding environmental parameters for feasibility testing based on the target positioning power consumption strategy. If the test results meet the preset adaptability conditions, the positioning power consumption strategy is switched. If the test results do not meet the preset adaptability conditions, the transitional positioning strategy is entered.

6. The method according to claim 5, characterized in that, The environmental adaptability verification specifically includes: When the target positioning power consumption strategy is a high-frequency positioning strategy, a signal pre-scan is performed for a preset duration to collect GPS signal strength data. If the GPS signal strength data is greater than the preset signal strength threshold, it is determined that the preset adaptation conditions are met. If the GPS signal strength data is less than or equal to a preset signal strength threshold, it is determined that the preset adaptation conditions are not met, and an outdoor pending confirmation transitional positioning strategy is entered, controlling the GPS to intermittently start positioning during the transition period. The transition period is greater than the third preset period and less than the second preset period.

7. The method according to claim 5, characterized in that, The steps of the dynamic switching positioning power consumption strategy also include a smooth transition step: After confirming the switching of the positioning power consumption strategy, the operating parameters of the positioning component are gradually adjusted using a progressive adjustment method. The gradual adjustment method specifically involves: gradually adjusting the startup frequency, operating power, and data transmission rate of the positioning component according to a preset adjustment gradient, with the duration of each adjustment gradient being a preset transition time, until the operating parameter standard corresponding to the target positioning power consumption strategy is reached.

8. The method according to claim 5, characterized in that, The steps of dynamically switching the positioning power consumption strategy also include an error correction step: After switching to the target positioning power consumption strategy, start the effectiveness monitoring timer and set the monitoring duration; During the monitoring period, real-time feedback data of the positioning component is collected, including positioning effectiveness, signal stability, and power consumption. If the positioning effectiveness is lower than the preset effective threshold or the signal stability is lower than the preset stable threshold during the monitoring period, the current positioning power consumption strategy is determined to be abnormal and will automatically revert to the positioning power consumption strategy before the switch.

9. The method according to claim 1, characterized in that, It also includes auxiliary optimization steps that incorporate time patterns: Record user's historical behavior state data and establish a time-state probability model, which includes the matching probability of different time periods and corresponding behavior states; Based on the current time information, the time-state probability model is queried to obtain the high-probability behavioral state for the current time period; The activation conditions and parameter configuration of the high-probability behavior state fine-tuning positioning power consumption strategy.

10. The method according to claim 2, characterized in that, The steps for multi-dimensional analysis of data to identify behavioral states specifically include: Feature extraction is performed on the collected environmental signal data and motion data to obtain signal strength features, motion frequency features, position change features, and velocity features; The features are input into a preset behavior state recognition model for classification and recognition, and the corresponding behavior state and confidence score are output. And / or, the data is analyzed by using preset feature threshold judgment rules, feature threshold ranges corresponding to each behavioral state are set, the extracted features are matched with the feature threshold ranges, and the behavioral state and confidence score are determined based on the matching results.

Images