Wearable smart watch with low power consumption

Abstract

The invention belongs to the field of wearable equipment, and particularly relates to a low-power-consumption wearable smart watch. Comprising an acquisition module, a processing module and a regulation and control module. The acquisition module is used for acquiring the signal strength of the current mobile network signal as first data according to a preset acquisition period, and associatively storing the acquisition time of the first data and the first data; the processing module is used for initializing the weakening times and weakening measurement, comparing the currently acquired first data with the first data acquired in the previous acquisition period, if the first data is weakened, increasing the weakening times, and accumulating the change value of the first data into the weakening measurement; if the first data becomes strong, whether the change value of the first data is larger than the weakening measurement or not is judged, if the change value of the first data is larger than the weakening measurement, the weakening frequency and the weakening measurement are initialized, and if the change value of the first data is not larger than the weakening measurement, the weakening measurement is started. According to the scheme, the problems of high power consumption and low user experience of the smart watch in a network environment with relatively weak signals are solved.

Description

Technical Field

[0001] This solution belongs to the field of wearable devices, and specifically relates to a low-power wearable smart watch. Background Art

[0002] With the wide application of smart watches, their usage scenarios in environments with weak signals (such as basements, remote areas, inside buildings, etc.) are increasing day by day. These environments usually have characteristics such as uneven network coverage and strong signal interference, resulting in the need for higher power consumption for smart watches to maintain connections during data transmission and communication. Existing smart watches rely on cellular networks (such as 2G, 3G, 4G, 5G, NB-IoT) or Wi-Fi for data transmission to achieve functions such as positioning, calling, and social interaction.

[0003] However, in environments with weak signals, the power consumption problem of smart watches is particularly prominent. Due to hardware limitations (such as weak antenna performance), network connection strategies (such as the cellular network module needs to switch base stations more frequently or increase the transmission power), and insufficient power consumption management (such as the lack of an effective optimization mechanism for environments with weak signals), smart watches face high-power consumption challenges in environments with weak signals, which not only affects the battery life of the device but also reduces the user experience. Summary of the Invention

[0004] The purpose of this solution is to provide a low-power wearable smart watch to solve the problem of high power consumption of smart watches in a network environment with weak signals, resulting in a low user experience.

[0005] To achieve the above objective, this solution provides a low-power wearable smart watch, including: An acquisition module, configured to acquire the signal strength of the current mobile network signal as first data according to a preset acquisition period, and store the acquisition time of the first data in association with the first data; A processing module, initializes the weakening times and the weakening measurement, compares the currently obtained first data with the first data obtained in the previous acquisition period. If the first data weakens, the weakening times are increased, and the change value of the first data is accumulated into the weakening measurement; if the first data strengthens, it is determined whether the change value of the first data is greater than the weakening measurement. If it is greater than the weakening measurement, the weakening times and the weakening measurement are initialized. If the change value of the first data is not greater than the weakening measurement, the weakening measurement is reduced according to the change value of the first data; when the weakening times are greater than the preset warning times, it is determined whether the weakening measurement is greater than the preset warning measurement. If it is greater than the warning measurement, a first connection request is generated, and the weakening times and the weakening measurement are initialized; if it is not greater than the warning measurement, the weakening times and the weakening measurement are continuously accumulated according to the change of the first data; the processing module is also configured to generate a first connection request when the currently acquired first data is less than the preset minimum strength; A control module is used to obtain a first connection request, turn on Bluetooth after obtaining the first connection request, obtain historical connected devices in the Bluetooth list, select one of the historical connected devices as the master device according to the signal strength of each historical connected device, and send a Bluetooth connection request to the master device.

[0006] The principle and technical effect of this solution are as follows: First, this solution adopts a periodic signal acquisition mechanism, collects the current mobile network signal strength at preset time intervals, associates and stores the acquisition time with the signal strength, and establishes a historical record of the signal strength. Each signal acquisition is marked with a timestamp, providing a data basis for subsequent signal trend analysis. This method can not only reduce the power consumption caused by continuous signal monitoring and lower the energy consumption through intermittent acquisition, but also avoid frequently waking up the main processor of the smart watch, thereby reducing the overall power consumption.

[0007] Secondly, this solution starts a deep listening for a preset duration according to the signal change trend. When the signal strength fluctuates more than a preset threshold (the difference between the maximum value and the minimum value) or the signal strength is lower than the lowest strength threshold during the listening period, the smart watch connects to the master device via Bluetooth for data transmission. This method can respond in advance when the signal deteriorates, further reducing the power consumption of the smart watch; at the same time, it can establish a Bluetooth connection with the master device in advance when the signal continues to deteriorate, ensuring the stability of data transmission, thereby improving the user experience of the smart watch when the signal continues to deteriorate and ensuring the communication quality.

[0008] In summary, this solution solves the problems of high power consumption and low user experience of smart watches in a network environment with weak signals.

[0009] Further, the processing module processes the first data within the acquisition period including the following steps: S10: Conduct instantaneous signal strength samplings within each acquisition period , denoted as { }, where is the original signal strength value of the th sampling, and the sampling interval needs to satisfy ; S20: Perform a moving average filter on the sampling sequence , and the expression formula of the output filtered signal is shown in the following formula (1): (1), Or adopt a windowed finite impulse response filter, and the expression formula of the filtered signal is as follows formula (2): (2); S30: Define the first data of the current acquisition period as the energy mean of the filtered signal. The calculation formula is as shown in formula (3) below: (3), Or when the fluctuation is less than 3 dB, it is simplified to an arithmetic mean. The calculation formula is as shown in formula (4) below: (4); S40: If a certain sampling value satisfies (where is the sample standard deviation within the period), then eliminate this sampling value and recalculate

[0010] This solution suppresses noise in the original sampling data by introducing a time-domain filtering model (moving average / FIR filtering), thus solving the measurement error problem that may be caused by instantaneous signal interference. At the same time, by adopting a dynamic selection strategy of logarithmic energy mean and arithmetic mean, compared with a single data recording method, it significantly improves the data representation accuracy in the signal fluctuation scenario. Further, by adding an outlier rejection condition, a data cleaning mechanism is embedded in the acquisition process, making the finally output value more robust. This improvement is directly related to the subsequent weakening times and weakening measurement calculation module, so more stable input data can reduce the misjudgment probability of the processing module. Finally, the mathematical modeling of this solution (such as the constraints of the filtering coefficient and the sampling interval ) provides a quantifiable implementation standard for the "preset acquisition period", enabling the low-power characteristic (relying on period optimization) and signal monitoring reliability (relying on data processing) of this solution to form a technical closed-loop.

[0011] Furthermore, when comparing the currently obtained first data with the first data obtained in the previous acquisition period, it includes the following steps: S50: Define the adaptive comparison threshold , The definition formula of is as shown in formula (5) below: where is the minimum allowable threshold; is the standard deviation of the previous cycles , The calculation formula of (6), wherein is the sensitivity coefficient; S60: the trend of signal intensity change is determined by the following piecewise function formula (7): (7), S70: the weakening measurement introduces a trend weight factor in its update, and its calculation formula is as shown in the following formula (8): (8), where is the weight coefficient.

[0012] This solution significantly improves the performance and adaptability of the technical solution by introducing a dynamic threshold adjustment, multi-level trend determination, and weighted measurement correction mechanism. Specifically, this solution realizes the adaptive adjustment of the threshold with the historical signal fluctuations through the dynamic threshold formula (5), enabling the threshold to be automatically relaxed in high-fluctuation scenarios and tightened in stable scenarios, thereby further improving the environmental adaptability while inheriting the advantages of data preprocessing. In addition, compared with simple binary judgment, the four-state trend determination of this solution not only distinguishes significant / slight changes but also identifies stable states. This refined processing not only reduces invalid operations caused by minor fluctuations but also strengthens the response speed to key signals through the weighted coefficient, ultimately forming a complete closed-loop control from data acquisition, dynamic processing to early warning decision-making, improving the reliability of connection switching decisions while maintaining the low-power consumption characteristics.

[0013] Furthermore, the acquisition module is also used to collect the current position of the watch as the first position when collecting the signal strength of the mobile network. When the signal strength of the mobile network is less than the preset minimum strength, the first position is stored in association with the collected signal strength of the mobile network as the weak signal position; the acquisition module is used to obtain the current position, compare the current position with the weak signal position, obtain the distance between the current position and the weak signal position, and generate a Bluetooth activation instruction to send to the Bluetooth historical connection device when the distance is less than the preset first distance.

[0014] Furthermore, the processing module is also used to generate a Bluetooth activation instruction to send to the Bluetooth historical connection device when the number of weakening times is greater than the preset warning times or the weakening measurement is greater than the preset warning measurement.

[0015] First, by establishing a weak signal position database, the real-time signal detection is upgraded to a dual perception method of "historical record + spatial prediction", enabling the smart watch to predict the signal strength based on the geographical location. Secondly, when it is detected that the device is approaching a known weak signal area (distance < first distance), the Bluetooth connection preparation is activated in advance. This predictive strategy significantly reduces the connection interruption time compared with the passive recovery mechanism. More prominently, such a method is deeply integrated with the device cooperation mechanism, and the handshake process between devices is started in advance through position triggering, winning a precious time window for pre-transmission. At the same time, this intelligent prediction method based on the geofence enables both the rate optimization decision and the connection fault tolerance mechanism to obtain a more sufficient response time. Finally, through the integration of the spatial position dimension with the original signal strength, transmission rate, and device cooperation dimension, a complete low-power management system of "space-time integration" is constructed, further enhancing the environmental adaptability of this solution.

[0016] Furthermore, when the smart watch is Bluetooth-connected to the master device, the acquisition module obtains the signal strength of the Bluetooth signal as the second data, and stores the second data in association with the acquisition time; the processing module initializes the number of times the Bluetooth signal weakens and the Bluetooth signal weakening measurement . The currently obtained second data is compared with the second data of the previous acquisition cycle . If , then it increases , and accumulates the attenuation amount of the second data to ; if , then it judges whether is greater than . If it is greater, then it resets and . If is not greater than , then it decreases ; when exceeds the preset Bluetooth weakening times threshold , it judges whether exceeds the preset Bluetooth weakening measurement threshold . If it exceeds, then it generates a Bluetooth disconnection instruction; the regulation module is used to obtain the Bluetooth disconnection instruction. When the regulation module obtains the Bluetooth disconnection instruction, it generates a disconnection Bluetooth message and sends it to the master device, and then closes the Bluetooth.

[0017] First, compared with only monitoring the strength of mobile network signals, this solution applies a dynamic threshold judgment and cumulative metering mechanism to Bluetooth signal monitoring when connected to the master device via Bluetooth. This enables the device to actively disconnect when the Bluetooth connection quality deteriorates continuously, thus avoiding ineffective power consumption (such as energy loss caused by repeatedly attempting to maintain a weak connection). Further, different from the location association strategy, such a method makes decisions entirely based on the real-time change trend of Bluetooth signal strength (rather than geographical location), so it can more directly reflect the current connection quality, especially suitable for scenarios where Bluetooth signals are dynamically interfered (such as human body occlusion, multipath effect) during user movement. This improvement is directly related to the mobile network monitoring module because when the Bluetooth connection is disconnected due to signal deterioration, this solution can immediately trigger the mobile network monitoring process to achieve a smooth switch of the network connection method. Finally, by generating a closed-loop control process of "disconnection instruction → sending disconnection information → turning off Bluetooth", it not only solves the problem of continuous power consumption of traditional devices in the weak Bluetooth connection state but also forms a complement with the Bluetooth reconnection mechanism. When the Bluetooth signal resumes stability, the device can re-initiate an efficient connection according to the monitoring rules of the mobile network signal strength, thus maximizing the energy-saving effect while ensuring communication reliability.

[0018] Further, when the acquisition module is waiting to receive a Bluetooth connection request, the regulation module stores the data to be uploaded locally and then uploads the data again when the signal strength of the mobile network signal is greater than the minimum strength or a Bluetooth connection request is received.

[0019] First, by temporarily storing the data to be transmitted locally during the connection interruption, it effectively solves the risk of data loss during the waiting for recovery period and ensures the integrity of data transmission. Second, adopting a dual-trigger strategy of "signal strength priority" and "Bluetooth connection priority" provides a more reliable data transmission guarantee compared with a simple rate switching mechanism. More importantly, this caching mechanism deeply collaborates with the motion state perception - automatically enabling local storage in the case of unstable connections caused by high-speed movement and resuming transmission when the movement is stable or entering an area with good signals, greatly improving the data transmission success rate. At the same time, this intelligent buffering strategy provides data security guarantee for the device handover process to ensure that key information will not be lost when switching the master device. In addition, through the opportunistic upload mechanism, it not only avoids the high energy consumption of forced transmission in a weak signal environment but also overcomes the passivity of simply waiting for the connection to resume. Particularly noteworthy is that this method forms a perfect cooperation with the dynamic acquisition strategy, intelligently adjusting the data upload timing according to network conditions and motion states to achieve maximum energy efficiency.

[0020] Further, when the processing module is used to compare the Bluetooth signal strength and the mobile network signal strength, if both the Bluetooth signal strength and the mobile network signal strength are less than a preset minimum signal strength, it generates information to suspend the connection and sends it to the master device, then disconnects the Bluetooth connection with the master device, and waits to receive a Bluetooth connection request.

[0021] First, by setting a minimum speed threshold and monitoring the transmission performance of the dual channels (mobile network and Bluetooth), the single signal strength comparison is upgraded to a "dual-channel quality assessment" method, enabling the smart watch to identify a completely unavailable network environment; second, when it detects that both channels do not meet the standards, it actively suspends the connection and enters the standby mode, which further reduces the energy consumption caused by invalid connection attempts compared with the passive caching strategy; more importantly, this protection mechanism forms a perfect cooperation with the data caching function - saving data locally during the suspended connection and automatically resuming transmission after the network is restored (triggered by location or improved motion state), which not only ensures data security but also optimizes energy consumption.

[0022] Further, a dynamic priority scoring mechanism is introduced when selecting the master device, and the priority score is calculated as shown in the following formula (9): (9), where is the Bluetooth signal strength of the historical connected device , is the maximum signal strength among the currently scanned devices; is the remaining battery percentage of device ; is the CPU utilization rate of device , is the preset upper threshold of the CPU utilization rate; , and are weight coefficients, satisfying ; The regulation module takes the device with the highest as the master device; if the difference between multiple devices is less than the preset tolerance , then it preferentially selects the device with a higher signal strength .

[0023] The priority scoring method of this solution takes into account the three dimensions of signal strength (RSSI), device power (Battery) and CPU load (CPU). When selecting the main device, it no longer relies solely on the instantaneous signal strength, avoiding misjudgment caused by a single indicator. For example, when the signal strength of the mobile phone is slightly higher than that of the car Bluetooth (RSSI_mobile phone = -60dBm vs RSSI_car = -65dBm), but the mobile phone has only 15% power left (Battery_mobile phone = 15) and the CPU load reaches 70% (CPU_mobile phone = 70), the dynamic scoring will reduce the priority of the mobile phone (Score_mobile phone) and select the car device with sufficient power (Battery_car = 100) and idle CPU (CPU_car = 10), thereby significantly reducing the connection interruption caused by sudden shutdown or freezing of the mobile phone. At the same time, the introduction of the tolerance threshold means that when the scores of multiple devices are close, this solution still gives priority to the device with stronger signal, thereby ensuring multi-dimensional optimization while avoiding excessive sacrifice of signal quality.

[0024] Further, priority scoring The weight coefficient , and Dynamic adjustment, adjustment rules based on Markov decision process, state space It is defined as shown in the following formula (10): (10), Action Space is the adjustment step of the weight coefficient, the reward function It is shown in the following formula (11): (11), is the minimum and maximum value of the connection success rate; : The minimum and peak data transmission rates; : The minimum observed value and the actual maximum observed value of the switching number; , is the weight coefficient, satisfying ; The control module iteratively updates the weight coefficient through the Q-learning algorithm, so that the long-term reward maximize, is the discount factor.

[0025] In the Q-learning algorithm of this solution, frequent switching is punished through the reward function. Therefore, this solution will learn the strategy of "maintaining a stable connection in the long term". For example, in an office scenario, even if the mobile phone signal fluctuates briefly (such as dropping from -70 dBm to -75 dBm), the algorithm will not immediately switch to other devices unless the scoring difference exceeds the tolerance, thereby reducing the energy consumption of Bluetooth re-handshaking. On the contrary, if it is detected that the battery level of the current master device continues to decline (such as decreasing by 5% per minute), the dynamic weight adjustment will gradually increase the battery weight, prompting this solution to switch to a device with a high battery level in advance, thereby avoiding the data retransmission overhead caused by sudden power-off of the device. This mechanism complements the Bluetooth disconnection rule: when both the Bluetooth signal and the battery level deteriorate, actively disconnect the connection and trigger mobile network monitoring instead of maintaining an ineffective connection with high energy consumption. Brief Description of the Drawings

[0026] Figure 1 It is a schematic diagram of the module structure of a low-power wearable smart watch in an embodiment of the present invention. Detailed Embodiments

[0027] The following will clearly and completely describe the concept and technical effects generated by the present invention in combination with the embodiments to fully understand the purpose, features, and effects of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all embodiments. Other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts shall fall within the scope of protection of the present invention: Embodiment

[0028] As Figure 1 shown, a low-power wearable smart watch includes: a collection module, a processing module, and a regulation module.

[0029] The collection module is used to collect the signal strength of the current mobile network signal as the first data according to a preset collection period, and store the collection time of the first data in association with the first data; The processing module initializes the weakening times and the weakening measurement, compares the currently acquired first data with the first data acquired in the previous acquisition cycle. If the first data weakens, it increases the weakening times and accumulates the change value of the first data into the weakening measurement; if the first data strengthens, it determines whether the change value of the first data is greater than the weakening measurement. If it is greater than the weakening measurement, it initializes the weakening times and the weakening measurement. If the change value of the first data is not greater than the weakening measurement, it reduces the weakening measurement according to the change value of the first data; when the weakening times is greater than the preset warning times, it determines whether the weakening measurement is greater than the preset warning measurement. If it is greater than the warning measurement, it generates a first connection request and initializes the weakening times and the weakening measurement; if it is not greater than the warning measurement, it continues to accumulate the weakening times and the weakening measurement according to the change of the first data; the processing module is also used to generate a first connection request when the first data currently acquired is less than the preset minimum intensity; The regulation module is used to obtain the first connection request, and after obtaining the first connection request, it turns on the Bluetooth and obtains the historical connection devices in the Bluetooth list, and takes one of the historical connection devices as the master device according to the signal strength of each historical connection device, and sends a Bluetooth connection request to the master device.

[0030] Among them, the processing module for the acquisition cycle The processing of the first data within includes the following steps: S10: In each acquisition cycle Perform times of instantaneous signal strength sampling, denoted as { }, where is the original signal strength value of the th sampling, and the sampling interval needs to satisfy ; S20: Perform moving average filtering on the sampling sequence , and the expression formula of the output filtered signal is shown in the following formula (1): (1), Or use windowed finite impulse response filtering, and the expression formula of the filtered signal is as follows formula (2): (2); S30: Define the first data of the current acquisition cycle as the energy mean value of the filtered signal, and the calculation formula of is shown in the following formula (3): Or when the fluctuation is less than 3dB, it is simplified to arithmetic mean, and the calculation formula is shown in the following formula (4): (4); S40: If a certain sampled value satisfies is the sample standard deviation within the period), then reject this sampled value and recalculate .

[0031] Among them, when comparing the currently obtained first data with the first data obtained in the previous acquisition period , the following steps are included: S50: Define the adaptive comparison threshold , The definition formula of is as shown in the following formula (5): (5), Among them, is the minimum allowable threshold (unit: dBm, ); is the standard deviation of the previous cycles (unit: dBm), is a dimensionless sensitivity coefficient (1.2 ≤ β ≤ 2.0), The calculation formula of is as shown in the following formula (6): (6), Among them, is the sensitivity coefficient ( ); S60: The signal strength change trend is determined by the following piecewise function formula (7): (7), S70: The update of the weakening measurement introduces a trend weight factor, The calculation formula of is as shown in the following formula (8): (8), Among them is the weight coefficient ( When , otherwise ). Set the ) coefficient in to 0.5 for logical distinction from "significantly getting stronger". When "significantly getting stronger", the signal change amount exceeds the adaptive threshold , indicating that the environmental signal quality has been substantially improved. At this time, it is designed to completely offset the weakening measurement (that is = = , to quickly reset the deterioration trend. The change amount of "slightly getting stronger" does not exceed , which belongs to the normal fluctuation range. If it is directly completely offset, it will cause frequent oscillations in the weakening measurement. Selecting 0.5 as the coefficient is because the change amplitude of "slightly getting stronger" is usually within 50% of that of "significantly getting stronger" (based on signal statistical laws), and the 0.5 coefficient can form a linear mapping between the response amplitude and the degree of change to avoid overreaction.

[0032] Among them, the acquisition module is also used to acquire the current position of the watch as the first position when acquiring the signal strength of the mobile network. When the signal strength of the mobile network is less than the preset minimum strength, the first position is stored in association with the acquired signal strength of the mobile network as the weak signal position; the acquisition module is used to obtain the current position, compare the current position with the weak signal position, obtain the distance between the current position and the weak signal position, and generate a Bluetooth enabling instruction to send to the Bluetooth historical connection device when the distance is less than the preset first distance.

[0033] Among them, the processing module is also used to generate a Bluetooth enabling instruction to send to the Bluetooth historical connection device when the number of weakening times is greater than the preset warning times or the weakening measurement is greater than the preset warning measurement.

[0034] Among them, when the smart watch is Bluetooth-connected to the main device, the acquisition module acquires the signal strength of the Bluetooth signal as the second data and stores the second data in association with the acquisition time. The processing module initializes the number of Bluetooth signal weakening times and the Bluetooth signal weakening measurement , compares the currently acquired second data with the second data of the previous acquisition period . If , then increase , and accumulate the attenuation amount of the second data to ; if , then judge whether it is greater than . If it is greater, then reset and . If is not greater than , then decrease ; when exceeds the preset Bluetooth weakening times threshold , judge whether it exceeds the preset Bluetooth weakening measurement threshold . If it exceeds, then generate a Bluetooth disconnection instruction; the regulation module is used to obtain the Bluetooth disconnection instruction. When the regulation module obtains the Bluetooth disconnection instruction, it generates a disconnection Bluetooth information to send to the main device and then turns off the Bluetooth.

[0035] Among them, when the acquisition module waits to receive a Bluetooth connection request, the control module stores the data to be uploaded locally, and then uploads the data again when the signal strength of the mobile network signal is greater than the minimum strength or a Bluetooth connection request is received.

[0036] Among them, when the processing module is used to compare the Bluetooth signal strength and the mobile network signal strength, if both the Bluetooth signal strength and the mobile network signal strength are less than the preset minimum signal strength, it generates information to pause the connection and sends it to the master device, then disconnects the Bluetooth connection with the master device, and waits to receive a Bluetooth connection request.

[0037] Among them, a dynamic priority scoring mechanism is introduced when selecting the master device, and the priority score is calculated as shown in the following formula (9): (9), Among them, is the Bluetooth signal strength (unit: dBm) of the historical connected device , is the maximum signal strength among the currently scanned devices; is the remaining battery percentage of device ; is the CPU utilization rate (unit: %) of device ; is the preset upper limit threshold of CPU utilization rate (such as 80%); , and are weight coefficients, satisfying , and the default value is = 0.5, = 0.3, = 0.2; The control module takes the device with the highest as the master device; if the difference between multiple devices is less than the preset tolerance (such as = 0.1), then the device with a higher signal strength is preferentially selected.

[0038] Specifically, the weight coefficient is dynamically adjusted, and its adjustment rule is based on the Markov decision process (MDP). The state space is defined as shown in the following formula (10): (10), The action space is the adjustment step size of the weight coefficient (such as ), and the reward function As shown in the following formula (11): (11), are the minimum and maximum values of the connection success rate; : the minimum value and peak value of the data transmission rate; : the minimum observed value and the actual maximum observed value of the number of handovers; , is the weight coefficient, satisfying ; The regulation module iteratively updates the weight coefficient through the Q-learning algorithm, making the long-term reward maximize ( is the discount factor).

[0039] Initialize table, the state is the discretized classification of the current signal, power, and CPU load; after each selection of the master device, update the value according to the connection performance feedback, as shown in the following formula (12): (12); Among them, the weight coefficient is dynamically adjusted according to the optimal policy .

[0040] Specifically, when the user enters the underground parking lot of the shopping mall with the smart watch D and is about to drive away. Due to the complex building structure in the parking lot, the mobile network signal (4G) coverage is extremely unstable, and the acquisition module of the smart watch D starts to acquire the mobile network signal strength according to a preset period (every 3 seconds).

[0041] Within a 5-minute monitoring period, the acquisition module detects that the signal strength weakens 5 times (exceeding the preset minimum number of times 3 times), and the cumulative weakening value of the adjacent 5 signal fluctuations reaches a maximum of 58 dBm (> the preset difference 50 dBm), and the processing module generates a first connection request. The regulation module immediately turns on the Bluetooth, scans the in-vehicle Bluetooth device E (signal strength -60 dBm) and the user's mobile phone F (signal strength -75 dBm) in the historical connection device list, and preferentially selects the in-vehicle Bluetooth device E as the master device according to the signal strength and sends a Bluetooth connection request.

[0042] When the signal strength < -100 dBm, the acquisition module records the current location (area G on the B2 floor of the parking lot) as a weak signal location and associates the signal strength data. When the user drives into the parking lot entrance again the next day (distance from the weak signal location < 80 meters, the preset first distance), the acquisition module automatically generates a Bluetooth activation instruction, wakes up the Bluetooth module in advance, and sends Bluetooth activation instructions to the mobile phone F and the in-vehicle Bluetooth device E. After receiving the instructions, the mobile phone F and the in-vehicle Bluetooth device E wake up the Bluetooth module and wait for the connection request from the watch D.

[0043] After the user parks the car and leaves the vehicle, they go home in a direction away from the in-vehicle Bluetooth device. As the user moves farther away, the Bluetooth signal between the smartwatch D and the in-vehicle Bluetooth device E continuously weakens. When the number of times the Bluetooth signal weakens is 5 times (exceeding the preset , which is 4 times), it is determined that ( which is 51dBm) exceeds the preset Bluetooth weakening measurement threshold (preset to 50dBm), the smartwatch D generates a Bluetooth disconnection message and sends it to the master device, and then turns off the Bluetooth.

[0044] After turning off the Bluetooth of the smartwatch D, the user has not left the garage yet. At this time, the smartwatch detects that the signal strength of the current mobile network is -105dBm (lower than the preset minimum strength of -100dBm), so it turns on the Bluetooth again to obtain nearby historical connected devices. The smartwatch obtains the nearby mobile phone F, so it takes the mobile phone F as the master device and sends a Bluetooth connection request to the mobile phone F.

[0045] After the user leaves the parking lot, the mobile network signal strength rises to -85dBm (> the minimum strength of -100dBm), and the control module automatically triggers data retransmission to synchronize the parking data to the parking payment application of the mobile phone F to ensure that the data is not lost.

[0046] One day, the user enters the underground parking lot of the shopping mall with the smartwatch D. Due to the unstable mobile network signal, the acquisition module acquires the signal strength every 3 seconds. Within 5 minutes, the signal weakens 5 times and the cumulative weakening value reaches 58dBm, triggering the first connection request. The control module enables the dynamic priority scoring mechanism and scans the in-vehicle Bluetooth device E (signal -60dBm, battery power 80%, CPU utilization rate 30%) and the mobile phone F (signal -75dBm, battery power 90%, CPU utilization rate 20%). After calculation, the priority score of E is higher and it is selected as the connected master device.

[0047] When the signal strength < -100dBm, the watch records the B2 floor G area of the parking lot as a weak signal location. The next day when the user enters the parking lot again, the watch wakes up the Bluetooth in advance. At this time, the weight coefficient has been dynamically adjusted through the Markov decision process and pays more attention to the signal strength.

[0048] When the user leaves the vehicle, the signal between the watch and the in-vehicle Bluetooth E continues to weaken. After the disconnection condition is triggered, the processing module compares and finds that both the Bluetooth and the mobile network signals are lower than the preset strength, and then sends a pause connection message and disconnects. Then, due to the still weak mobile network signal, the watch scans again and selects the mobile phone F to connect according to the score.

[0049] When the user leaves the parking lot and the signal recovers, data retransmission is triggered. The watch optimizes the weight coefficient through the Q-learning algorithm based on the connection success rate, transmission rate, and number of handovers, and improves the subsequent connection strategy.

[0050] The above are only embodiments of the present invention, and common knowledge such as specific structures and characteristics well known in the art are not described in detail herein. It should be noted that for those skilled in the art, without departing from the structure of the present invention, several modifications and improvements can be made, which should also be regarded as the protection scope of the present invention, and these will not affect the implementation effect of the present invention and the practicality of the patent. The protection scope required by this application shall be subject to the content of its claims, and the specific implementation manners described in the specification can be used to interpret the content of the claims.

Claims

1. A low-power wearable smartwatch, characterized in that, Including: A collection module, which is used to collect the signal strength of the current mobile network signal as the first data according to a preset collection period, and associate and store the collection time of the first data with the first data; A processing module, which initializes the weakening times and weakening measurement, compares the currently obtained first data with the first data obtained in the previous collection period. If the first data weakens, it increases the weakening times and accumulates the change value of the first data into the weakening measurement; if the first data strengthens, it determines whether the change value of the first data is greater than the weakening measurement. If it is greater than the weakening measurement, it initializes the weakening times and weakening measurement. If the change value of the first data is not greater than the weakening measurement, it reduces the weakening measurement according to the change value of the first data; When the weakening times is greater than the preset warning times, it determines whether the weakening measurement is greater than the preset warning measurement. If it is greater than the warning measurement, it generates a first connection request and initializes the weakening times and weakening measurement; If it is not greater than the warning measurement, it continues to accumulate the weakening times and weakening measurement according to the change of the first data; the processing module is also used to generate a first connection request when the first data collected currently is less than the preset minimum strength; A regulation module, which is used to obtain the first connection request, and after obtaining the first connection request, turn on the Bluetooth and obtain the historical connection devices in the Bluetooth list, and use one of the historical connection devices as the master device according to the signal strength of each historical connection device, and send a Bluetooth connection request to the master device.

2. The low-power wearable smart watch according to claim 1, wherein: The processing module processes the first data within the acquisition period The processing includes the following steps: S10: During each acquisition period perform instantaneous signal strength samplings, denoted as { }, where is the original signal strength value of the -th sampling, and the sampling interval needs to satisfy ; S20: Perform moving average filtering on the sampling sequence and output the filtered signal The expression formula is shown in formula (1) below: （1）， Or use windowed finite impulse response filtering, and the filtered signal is expressed by the following formula (2): （2）； S30: Define the first data of the current acquisition period as the energy mean value of the filtered signal, and its calculation formula is as shown in formula (3) below: （3）， Or when When the fluctuation is less than 3 dB, it is simplified to the arithmetic mean, and the calculation formula is as shown in formula (4) below: （4）； S40: If a certain sampling value satisfies the sample standard deviation within the period, then eliminate this sampling value and recalculate .

3. The low-power wearable smart watch according to claim 2, characterized in that: The currently acquired first data is compared with the first data acquired in the previous acquisition cycle and includes the following steps: S50: Define an adaptive contrast threshold , The definition formula is as shown in the following formula (5): （5）， wherein, is the minimum allowable threshold; is the first cycles standard deviation of, The calculation formula of is shown in formula (6) as follows: （6）， Among them, is the sensitivity coefficient; S60: Signal strength change trend Determined by the following piecewise function formula (7): （7）， S70: Diminishing measurement The update of introduces a trend weight factor, and its calculation formula is as shown in the following formula (8): （8）， wherein is a weight coefficient.

4. A low-power wearable smartwatch according to claim 3, characterized in that: The collection module is also used to collect the current position of the watch as the first position when collecting the signal strength of the mobile network. When the signal strength of the mobile network is less than the preset lowest strength, it associates and stores the first position as the weak signal position with the signal strength of the collected mobile network; The collection module is used to obtain the current position, compare the current position with the weak signal position, obtain the distance between the current position and the weak signal position, and generate a Bluetooth turn-on instruction and send it to the Bluetooth historical connection device when the distance is less than the preset first distance.

5. The low-power wearable smart watch according to claim 4, wherein: The processing module is also used to generate a Bluetooth turn-on instruction and send it to the Bluetooth historical connection device when the weakening times is greater than the preset warning times or the weakening measurement is greater than the preset warning measurement.

6. A low-power wearable smartwatch according to claim 5, characterized in that: When the collection module makes a Bluetooth connection between the smart watch and the master device, it obtains the signal strength of the Bluetooth signal as the second data, and associates and stores the second data with the collection time; The processing module initializes the number of times the Bluetooth signal weakens and the Bluetooth signal weakening measurement , and compares the currently obtained second data with the second data of the previous acquisition cycle . If , then increment , and accumulate the attenuation amount of the second data to . If , then determine whether it is greater than . If it is greater, then reset and . If is not greater than , then decrement . When exceeds the preset Bluetooth weakening times threshold , determine whether it exceeds the preset Bluetooth weakening measurement threshold . If it exceeds, then generate a Bluetooth disconnection instruction; the regulation module is used to obtain the Bluetooth disconnection instruction. When the regulation module obtains the Bluetooth disconnection instruction, it generates a disconnection Bluetooth message and sends it to the master device, and then turns off the Bluetooth.

7. A low-power wearable smartwatch according to claim 6, characterized in that: When the collection module waits to receive a Bluetooth connection request, the regulation module stores the data to be uploaded locally, and then uploads the data again when the signal strength of the mobile network signal is greater than the lowest strength or a Bluetooth connection request is received.

8. A low-power wearable smartwatch according to claim 7, characterized in that: When the processing module is used to compare the Bluetooth signal strength and the mobile network signal strength, if both the Bluetooth signal strength and the mobile network signal strength are less than the preset lowest signal strength, it generates an information of pausing the connection and sends it to the master device, then disconnects the Bluetooth connection with the master device, and waits to receive a Bluetooth connection request.

9. A low-power wearable smartwatch according to claim 8, characterized in that: Introduce a dynamic priority scoring mechanism when selecting the master device. The priority score is calculated as shown in the following formula (9): （9）， Among them, is the Bluetooth signal strength of the historical connection device , is the maximum signal strength among the currently scanned devices; is the remaining battery percentage of device ; is the CPU utilization rate of device , is the preset upper threshold of CPU utilization rate; , and are weight coefficients, satisfying ; The control module will select the device with the highest value as the master device; if the difference between multiple devices is less than the preset tolerance , the device with a higher signal strength will be preferentially selected.

10. A low-power wearable smartwatch according to claim 9, characterized in that: Priority score weight coefficient and and dynamically adjusted, and the adjustment rule is based on the Markov decision process, and the state space is defined as shown in the following formula (10): （10）， Action space is the adjustment step size of the weight coefficient, and the reward function is as shown in the following formula (11): （11）， are the minimum and maximum values of the connection success rate; : the minimum value and peak value of the data transfer rate; : the minimum observed value and the actual maximum observed value of the number of handovers; , is the weight coefficient, satisfying ; The control module iteratively updates the weight coefficients through the Q-learning algorithm to maximize the long-term reward and is the discount factor.

Images