A low-power wearable smart watch

Abstract

This solution belongs to the field of wearable devices, and specifically relates to a low-power wearable smart watch. It includes: an acquisition module, a processing module and a control module. The acquisition module is used to collect the signal strength of the current mobile network signal as the first data according to a preset acquisition period, and store the acquisition time of the first data in association with the first data; the processing module initializes the number of weakening times and the weakening measurement, compares the first data currently acquired with the first data acquired in the previous acquisition period, and if the first data becomes weaker, increases the number of weakening times, and accumulates the change value of the first data into the weakening measurement; if the first data becomes stronger, determines whether the change value of the first data is greater than the weakening measurement. If it is greater than the weakening measurement, initializes the number of weakening times and the weakening measurement. If the change value of the first data is not greater than the weakening measurement. This solution solves the problems of high power consumption and low user experience of smart watches in a network environment with a weak signal.

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 widespread adoption of smartwatches, their use in environments with weak signals (such as basements, remote areas, and inside buildings) is increasing. These environments often feature uneven network coverage and strong signal interference, resulting in smartwatches requiring higher power consumption to maintain connectivity during data transmission and communication. Existing smartwatches rely on cellular networks (such as 2G, 3G, 4G, 5G, and NB-IoT) or Wi-Fi for data transmission, enabling functions such as positioning, calls, and social interaction.

[0003] However, power consumption is particularly problematic for smartwatches in weak signal environments. Due to hardware limitations (such as weak antenna performance), network connection strategies (such as the cellular network module's need to switch base stations more frequently or increase transmit power), and inadequate power management (such as the lack of effective optimization mechanisms for weak signal environments), smartwatches face high power consumption challenges in weak signal environments. This not only affects device battery life but also degrades 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 poor user experience.

[0005] To achieve the above objectives, this solution provides a low-power wearable smart watch, including:

[0006] a collection module, configured to collect the signal strength of the current mobile network signal as first data according to a preset collection period, and associate the collection time of the first data with the first data for storage;

[0007] The processing module initializes the number of weakening times and the weakening measurement, compares the currently acquired first data with the first data acquired in the previous acquisition cycle, and if the first data becomes weaker, increases the number of weakening times and accumulates the change value of the first data into the weakening measurement; if the first data becomes stronger, determines whether the change value of the first data is greater than the weakening measurement, and if so, initializes the number of weakening times and the weakening measurement; if the change value of the first data is not greater than the weakening measurement, reduces the weakening measurement according to the change value of the first data; when the number of weakening times is greater than the preset warning number, determines whether the weakening measurement is greater than the preset warning measurement, and if so, generates a first connection request, initializes the number of weakening times and the weakening measurement; if not, continues to accumulate the number of weakening times and the weakening measurement according to the change of the first data; the processing module is further used to generate the first connection request when the currently acquired first data is less than the preset minimum strength;

[0008] The control module is used to obtain a first connection request, and after obtaining the first connection request, turn on Bluetooth and obtain historical connection devices in the Bluetooth list, select one of the historical connection devices as the main device according to the signal strength of each historical connection device, and send a Bluetooth connection request to the main device.

[0009] The principles and technical benefits of this solution are as follows: First, it uses a periodic signal acquisition mechanism to collect the current mobile network signal strength at preset intervals. The acquisition time is associated with the signal strength and stored to create a historical record of signal strength. Each signal acquisition is timestamped, providing a data foundation for subsequent signal trend analysis. This approach not only reduces the power consumption associated with continuous signal monitoring, but also reduces energy consumption through intermittent acquisition, avoiding frequent wake-ups of the smartwatch's main processor, thereby reducing overall power consumption.

[0010] Secondly, this solution initiates deep monitoring for a preset duration based on signal trends. If the signal strength fluctuation during the monitoring period exceeds a preset threshold (the difference between the maximum and minimum values) or falls below the minimum strength threshold, the smartwatch connects to the host device via Bluetooth for data transmission. This approach allows for a proactive response to signal deterioration, further reducing the smartwatch's power consumption. Furthermore, establishing a Bluetooth connection with the host device in advance when signal deterioration persists ensures stable data transmission, thus enhancing the user experience and ensuring communication quality during periods of persistent signal deterioration.

[0011] In summary, this solution solves the problems of high power consumption and poor user experience of smart watches in a network environment with weak signals.

[0012] Furthermore, the processing module processes the collection cycle The processing of the first data includes the following steps:

[0013] S10: Each acquisition cycle Internal The instantaneous signal strength sampling is recorded as { }, where The original signal strength value of the sampling time, the sampling interval Need to meet ;

[0014] S20: Sampling sequence Perform sliding average filtering and output the filtered signal The expression formula is as follows:

[0015] (1),

[0016] Or use windowed finite impulse response filtering, the filtered signal The expression formula is as follows formula (2):

[0017] (2);

[0018] S30: The first data of the current collection cycle It is defined as the energy mean of the filtered signal, The calculation formula is as follows:

[0019] (3),

[0020] or when When the fluctuation is less than 3dB, it is simplified to the arithmetic mean, and the calculation formula is shown in the following formula (4):

[0021] (4);

[0022] S40: If a certain sampling value satisfy is the standard deviation of samples within the period), then remove the sample value and recalculate .

[0023] This solution introduces a time domain filtering model (sliding average / FIR filtering) to suppress noise in the original sampled data, thereby solving the measurement error problem that may be caused by transient signal interference; at the same time, the dynamic selection strategy of logarithmic energy mean and arithmetic mean is adopted, which significantly improves the data representation accuracy in signal fluctuation scenarios compared to a single data recording method. Furthermore, this solution increases The outlier removal condition embeds the data cleaning mechanism in the collection process, making the final output The value has stronger robustness; this improvement is directly related to the subsequent weakening times and the weakening measurement calculation module, so more stable input data can reduce the probability of misjudgment of the processing module. Finally, the mathematical modeling of this scheme (such as the filter coefficient and sampling interval The constraints of the proposed scheme provide a quantifiable implementation standard for the “preset acquisition cycle”, which enables the low power consumption characteristics of this scheme (which relies on cycle optimization) and the signal monitoring reliability (which relies on data processing) to form a technical closed loop.

[0024] Further, the first data currently obtained The first data obtained in the previous collection cycle The comparison includes the following steps:

[0025] S50: Define adaptive contrast threshold , The definition formula is as follows:

[0026] (5),

[0027] in, is the minimum allowed threshold; For the front cycles The standard deviation of The calculation formula is as follows:

[0028] (6),

[0029] in, is the sensitivity coefficient;

[0030] S60: Signal strength change trend Determined by the following piecewise function formula (7):

[0031] (7),

[0032] S70: Weakened meter The update introduces the trend weight factor, The calculation formula is as follows:

[0033] (8),

[0034] in is the weight coefficient.

[0035] This solution significantly improves the performance and adaptability of the technical solution by introducing 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 fluctuation of the signal through the dynamic threshold formula (5), so that the threshold is 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 judgment of this solution not only distinguishes between significant / minor changes, but also identifies stable states. This refined processing not only reduces invalid operations caused by small fluctuations, but also reduces the number of invalid operations caused by small fluctuations. The weighted coefficient enhances the response speed to key signals, ultimately forming a complete closed-loop control from data acquisition, dynamic processing to early warning decision-making, while maintaining low power consumption characteristics and improving the reliability of connection switching decisions.

[0036] 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 as a weak signal position in association with the collected signal strength of the mobile network; 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 when the distance is less than the preset first distance, generate a Bluetooth start instruction and send it to the Bluetooth historical connection device.

[0037] Furthermore, the processing module is further configured to generate a Bluetooth activation instruction and send it to the Bluetooth historical connection device when the number of weakening times is greater than a preset warning number or the weakening measurement is greater than a preset warning measurement.

[0038] First, by establishing a weak signal location database, real-time signal detection is upgraded to a dual perception method of "historical record + spatial prediction", enabling smart watches to predict signal strength based on geographic location; second, when it detects that it is close to a known weak signal area (distance < first distance), Bluetooth connection preparation is activated in advance. Compared with the passive recovery mechanism, this predictive strategy significantly reduces the connection interruption time; more importantly, this method is deeply integrated with the device collaboration mechanism, and the handshake process between devices is started in advance through location triggering, which wins a valuable time window for pre-transmission; at the same time, this intelligent prediction method based on geographic fences enables rate optimization decisions and connection fault tolerance mechanisms to obtain more sufficient response time; finally, by integrating the spatial location dimension with the original signal strength, transmission rate, and device collaboration dimensions, a complete low-power management system of "time and space integration" is constructed, further improving the environmental adaptability of this solution.

[0039] Furthermore, the acquisition module obtains the signal strength of the Bluetooth signal as the second data when the smart watch is connected to the main device via Bluetooth, and stores the second data in association with the acquisition time; the processing module initializes the number of times the Bluetooth signal becomes weak and Bluetooth signal weakening measurement , the second data currently obtained The second data of the previous acquisition cycle In contrast, if , then increase and the attenuation of the second data Accumulated to ;like , then judge Is it greater than If it is greater than, reset and ,like No more than ,but reduce ;when Exceeds the preset Bluetooth weakening threshold When judging Whether the preset Bluetooth weakening measurement threshold is exceeded If it exceeds, a Bluetooth disconnect instruction is generated; the control module is used to obtain the Bluetooth disconnect instruction. When the control module obtains the Bluetooth disconnect instruction, it generates a disconnect Bluetooth message and sends it to the main device, and then turns off Bluetooth.

[0040] First, compared to simply monitoring the mobile network signal strength, this solution applies dynamic threshold determination and cumulative metering to Bluetooth signal monitoring during Bluetooth connection with the host device. This allows the device to proactively disconnect when the Bluetooth connection quality continues to deteriorate, thus avoiding inefficient power consumption (such as energy loss from repeatedly attempting to maintain a weak connection). Furthermore, unlike location-based strategies, this approach makes decisions based entirely on real-time trends in Bluetooth signal strength (rather than geographic location), providing a more direct reflection of current connection quality. This is particularly useful in scenarios where the Bluetooth signal is subject to dynamic interference (such as human obstruction and multipath effects) during user movement. This improvement is directly linked to the mobile network monitoring module: when the Bluetooth connection is disconnected due to signal degradation, this solution immediately triggers the mobile network monitoring process, enabling a smooth transition to a new network connection mode. Ultimately, by generating a closed-loop control process of "disconnect command → send disconnect message → disable Bluetooth", this not only solves the problem of traditional devices constantly consuming power when the Bluetooth connection is weak, but also complements the Bluetooth reconnection mechanism. Once the Bluetooth signal stabilizes, the device can re-initiate an efficient connection based on the mobile network signal strength monitoring rules, thereby ensuring communication reliability while maximizing energy savings.

[0041] Furthermore, when the acquisition module waits to receive a Bluetooth connection request, the control module stores the data to be uploaded locally, and then re-uploads the data when the signal strength of the mobile network signal is greater than the minimum strength or a Bluetooth connection request is received.

[0042] First, by temporarily storing the data to be transmitted locally during connection interruption, the risk of data loss during the waiting period for recovery is effectively solved, ensuring the integrity of data transmission; secondly, the dual triggering strategy of "signal strength priority" and "Bluetooth connection priority" provides more reliable data transmission guarantee compared to the simple rate switching mechanism; more importantly, this caching mechanism is deeply coordinated with motion state perception - local storage is automatically enabled when the connection is unstable due to high-speed movement, and transmission is resumed when the movement is stable or enters a good signal area, greatly improving the success rate of data transmission; at the same time, this intelligent buffering strategy provides data security for the device handover process, ensuring that key information is not lost when switching the main device; in addition, through the selective upload mechanism, it avoids the high energy consumption of forced transmission in weak signal environments, and overcomes the passivity of simply waiting for connection recovery; it is particularly noteworthy that this method forms a perfect match with the dynamic acquisition strategy, intelligently adjusting the data upload timing according to network conditions and motion status to maximize energy efficiency.

[0043] Furthermore, when the processing module is used to compare the Bluetooth signal strength and the mobile network signal strength, if the Bluetooth signal strength and the mobile network signal strength are both less than the preset minimum signal strength, a connection suspension message is generated and sent to the main device, and then the Bluetooth connection with the main device is disconnected, and the Bluetooth connection is waited for to receive a Bluetooth connection request.

[0044] First, by setting a minimum speed threshold and monitoring the transmission performance of two channels (mobile network and Bluetooth), the single signal strength comparison is upgraded to a "dual-channel quality assessment" method, enabling the smartwatch to identify completely unavailable network environments; secondly, when it is detected that both channels do not meet the standards, the connection is actively suspended and the smartwatch enters standby mode. Compared with the passive caching strategy, this further reduces the energy loss caused by invalid connection attempts; more importantly, this protection mechanism forms a perfect match with the data caching function - data is saved locally during the suspension of the connection, and transmission is automatically resumed after the network is restored (location trigger or motion status improved), which not only ensures data security but also optimizes energy consumption.

[0045] Furthermore, a dynamic priority scoring mechanism is introduced when selecting the master device. The calculation formula is as follows:

[0046] (9),

[0047] in, Connecting devices for history Bluetooth signal strength, The maximum signal strength among the devices currently scanned; For equipment The remaining battery percentage; For equipment CPU utilization, It is the preset upper limit threshold of CPU utilization; 、 and is the weight coefficient, satisfying ;

[0048] The control module will The highest device is the master device; if there are multiple devices The difference is less than the preset tolerance , then give priority to signal strength Higher equipment.

[0049] This solution's priority scoring method simultaneously considers signal strength (RSSI), device battery (Battery), and CPU load (CPU). This eliminates the need to rely solely on instantaneous signal strength when selecting the primary device, thus avoiding misjudgments caused by a single metric. For example, if the phone's signal strength is slightly stronger than the car's Bluetooth signal (RSSI_phone = -60dBm vs RSSI_car = -65dBm), but the phone's battery is only 15% (Battery_phone = 15) and the CPU load is 70% (CPU_phone = 70), the dynamic scoring method will lower the phone's priority (Score_phone) and instead prioritize the car's device with sufficient battery (Battery_car = 100) and idle CPU (CPU_car = 10). This significantly reduces connection interruptions caused by sudden phone shutdowns or lags. Furthermore, the introduction of a tolerance threshold ensures that even when multiple devices have similar scores, the solution still prioritizes the device with the stronger signal, thus ensuring multi-dimensional optimization while avoiding excessive sacrifice of signal quality.

[0050] Further, priority scoring The weight coefficient 、 and Dynamic adjustment, adjustment rules are based on Markov decision process, state space It is defined as shown in the following formula (10):

[0051] (10),

[0052] Action Space is the adjustment step of the weight coefficient, the reward function It is shown in the following formula (11):

[0053] (11),

[0054] 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 ;

[0055] The control module iteratively updates the weight coefficient through the Q-learning algorithm, so that the long-term reward maximize, is the discount factor.

[0056] The Q-learning algorithm in this solution uses a reward function to penalize frequent switching, so this solution will learn a strategy of "maintaining a stable connection for a long time." For example, in an office scenario, even if the mobile phone signal fluctuates briefly (such as from -70dBm to -75dBm), the algorithm will not immediately switch to other devices unless the score difference exceeds the tolerance, thereby reducing the energy consumption of Bluetooth re-handshake. Conversely, if it detects that the battery level of the current primary device continues to decline (such as a 5% decrease per minute), the dynamic weight adjustment will gradually increase the battery weight, prompting this solution to switch to a high-battery device in advance, thereby avoiding the data retransmission overhead caused by sudden power outages. This mechanism complements the Bluetooth disconnection rule: when the Bluetooth signal and battery level deteriorate simultaneously, the connection is actively disconnected and mobile network monitoring is triggered, rather than maintaining an invalid connection that consumes a lot of energy. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 The figure is a schematic diagram of the module structure of a low-power wearable smart watch in an embodiment of the present invention. DETAILED DESCRIPTION

[0058] The following will clearly and completely describe the concept and technical effects of the present invention in conjunction with the embodiments to fully understand the purpose, features and effects of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, other embodiments obtained by those skilled in the art without creative work are all within the scope of protection of the present invention: Example

[0059] like Figure 1 As shown, a low-power wearable smart watch includes: an acquisition module, a processing module and a control module.

[0060] a collection module, configured to collect the signal strength of the current mobile network signal as first data according to a preset collection period, and associate the collection time of the first data with the first data for storage;

[0061] The processing module initializes the number of weakening times and the weakening measurement, compares the currently acquired first data with the first data acquired in the previous acquisition cycle, and if the first data becomes weaker, increases the number of weakening times and accumulates the change value of the first data into the weakening measurement; if the first data becomes stronger, determines whether the change value of the first data is greater than the weakening measurement, and if so, initializes the number of weakening times and the weakening measurement; if the change value of the first data is not greater than the weakening measurement, reduces the weakening measurement according to the change value of the first data; when the number of weakening times is greater than the preset warning number, determines whether the weakening measurement is greater than the preset warning measurement, and if so, generates a first connection request, initializes the number of weakening times and the weakening measurement; if not, continues to accumulate the number of weakening times and the weakening measurement according to the change of the first data; the processing module is further used to generate the first connection request when the currently acquired first data is less than the preset minimum strength;

[0062] The control module is used to obtain a first connection request, and after obtaining the first connection request, turn on Bluetooth and obtain historical connection devices in the Bluetooth list, select one of the historical connection devices as the main device according to the signal strength of each historical connection device, and send a Bluetooth connection request to the main device.

[0063] Among them, the processing module is responsible for the collection cycle The processing of the first data includes the following steps:

[0064] S10: Each acquisition cycle Internal The instantaneous signal strength sampling is recorded as { }, where The original signal strength value of the sampling time, the sampling interval Need to meet ;

[0065] S20: Sampling sequence Perform sliding average filtering and output the filtered signal The expression formula is as follows:

[0066] (1),

[0067] Or use windowed finite impulse response filtering, the filtered signal The expression formula is as follows formula (2):

[0068] (2);

[0069] S30: The first data of the current collection cycle It is defined as the energy mean of the filtered signal, The calculation formula is as follows:

[0070] (3),

[0071] or when When the fluctuation is less than 3dB, it is simplified to the arithmetic mean, and the calculation formula is shown in the following formula (4):

[0072] (4);

[0073] S40: If a certain sampling value satisfy is the standard deviation of samples within the period), then remove the sample value and recalculate .

[0074] Among them, the first data currently obtained The first data obtained in the previous collection cycle The comparison includes the following steps:

[0075] S50: Define adaptive contrast threshold , The definition formula is as follows:

[0076] (5),

[0077] in, is the minimum allowed threshold (in dBm, ); For the front cycles The standard deviation (in dBm) is the dimensionless sensitivity coefficient (1.2≤β≤2.0), The calculation formula is as follows:

[0078] (6),

[0079] in, is the sensitivity coefficient ( );

[0080] S60: Signal strength change trend Determined by the following piecewise function formula (7):

[0081] (7),

[0082] S70: Weakened meter The update introduces the trend weight factor, The calculation formula is as follows:

[0083] (8),

[0084] in is the weight coefficient ( when ,otherwise ).Will )middle The coefficient is set to 0.5 and the logic distinction is "significantly stronger". When "significantly stronger", the signal change exceeds the adaptive threshold , indicating that the environmental signal quality has substantially improved, and the design is to completely offset the weakening measurement (i.e. = , to quickly reset the degradation trend. The change in “slightly stronger” does not exceed , which falls within the normal fluctuation range. If it were completely offset, it would cause frequent oscillations in the weakened measurement. The coefficient of 0.5 was chosen because the amplitude of a "slightly stronger" change is typically within 50% of a "significantly stronger" change (based on signal statistics). The 0.5 coefficient creates a linear mapping between the response amplitude and the degree of change, avoiding overreaction.

[0085] Among them, 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 as a weak signal position in association with the collected signal strength of the mobile network; 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 when the distance is less than the preset first distance, generate a Bluetooth start instruction and send it to the Bluetooth historical connection device.

[0086] The processing module is further configured to generate a Bluetooth start instruction and send it to the Bluetooth historical connection device when the number of weakening times is greater than the preset warning number or the weakening measurement is greater than the preset warning measurement.

[0087] The acquisition module obtains the signal strength of the Bluetooth signal as the second data when the smart watch is connected to the main device via Bluetooth, and stores the second data in association with the acquisition time. The processing module initializes the number of times the Bluetooth signal becomes weak and Bluetooth signal weakening measurement , the second data currently obtained The second data of the previous acquisition cycle In contrast, if , then increase and the attenuation of the second data Accumulated to ;like , then judge Is it greater than If it is greater than, reset and ,like No more than ,but reduce ;when Exceeds the preset Bluetooth weakening threshold When, judge Whether the preset Bluetooth weakening measurement threshold is exceeded If it exceeds, a Bluetooth disconnect instruction is generated; the control module is used to obtain the Bluetooth disconnect instruction. When the control module obtains the Bluetooth disconnect instruction, it generates a disconnect Bluetooth message and sends it to the main device, and then turns off Bluetooth.

[0088] Among them, when the acquisition module is waiting to receive a Bluetooth connection request, the control module will store the data to be uploaded locally, and then upload the data again after the signal strength of the mobile network signal is greater than the minimum strength or a Bluetooth connection request is received.

[0089] Among them, when the processing module is used to compare the Bluetooth signal strength and the mobile network signal strength, if the Bluetooth signal strength and the mobile network signal strength are both less than the preset minimum signal strength, it generates a connection suspension message and sends it to the main device, then disconnects the Bluetooth connection with the main device, and waits to receive a Bluetooth connection request.

[0090] Among them, a dynamic priority scoring mechanism is introduced when selecting the master device. The calculation formula is as follows:

[0091] (9),

[0092] in, Connecting devices for history Bluetooth signal strength (unit: dBm), The maximum signal strength among the devices currently scanned; For equipment The remaining battery percentage; For equipment CPU utilization (unit: %), The preset upper threshold of CPU utilization (such as 80%). 、 and is the weight coefficient, satisfying , the default value is =0.5, =0.3, =0.2;

[0093] The control module will The highest device is the master device; if there are multiple devices The difference is less than the preset tolerance (like = 0.1), the signal strength is preferred Higher equipment.

[0094] Specifically, the weight coefficient Dynamic adjustment, whose adjustment rules are based on Markov decision process (MDP), state space It is defined as shown in the following formula (10):

[0095] (10),

[0096] Action Space is the adjustment step of the weight coefficient (such as ), the reward function It is shown in the following formula (11):

[0097] (11),

[0098] 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 ;

[0099] The control module iteratively updates the weight coefficient through the Q-learning algorithm, so that the long-term reward maximize( is the discount factor).

[0100] initialization Table, Status Discrete classification of current signal, power, and CPU load; after each master device selection, update based on connection performance feedback The specific value is shown in the following formula (12):

[0101] (12);

[0102] Among them, the weight coefficient is based on the optimal strategy Dynamic adjustment, here represents any strategy, represents the optimal strategy.

[0103] In practice, a user enters a shopping mall's underground parking lot with smartwatch D and prepares to drive away. Due to the complex structure of the parking lot, the mobile network signal (4G) coverage is extremely unstable. Smartwatch D's acquisition module begins collecting mobile network signal strength at a preset interval (every 3 seconds).

[0104] During a 5-minute monitoring period, the acquisition module detects that the signal strength has weakened five times (exceeding the preset minimum of three times), and the cumulative signal weakening value of the five consecutive fluctuations reaches a maximum of 58dBm (> the preset difference of 50dBm). The processing module generates a first connection request. The control module immediately turns on Bluetooth and scans the list of historically connected devices. It finds the in-vehicle Bluetooth device E (signal strength -60dBm) and the user's mobile phone F (signal strength -75dBm). Based on the signal strength, it prioritizes the in-vehicle Bluetooth device E as the primary device and sends a Bluetooth connection request.

[0105] When the signal strength is less than -100dBm, the acquisition module records the current location (parking lot B2, Area G) as a weak signal location and associates the signal strength data. The next day, when the user drives into the parking lot entrance again (less than 80 meters from the weak signal location, a preset first distance), the acquisition module automatically generates a Bluetooth activation command, preemptively wakes up the Bluetooth module, and sends the Bluetooth activation command to phone F and in-car Bluetooth device E. After receiving the command, phone F and in-car Bluetooth device E wake up the Bluetooth module and wait for a connection request from watch D.

[0106] After parking the car, the user leaves the car and walks away from the in-car Bluetooth device. As the user walks away, the Bluetooth signal between smartwatch D and in-car Bluetooth device E becomes weaker. 5 times (exceeding the preset , 4 times), judge ( 51dBm) exceeds the preset Bluetooth weakening measurement threshold (The default setting is 50dBm), smart watch D generates a Bluetooth disconnect message and sends it to the main device, and turns off Bluetooth.

[0107] After turning off the Bluetooth of smartwatch D, the user does not leave the garage. At this time, the smartwatch detects that the current mobile network signal strength is -105dBm (lower than the preset minimum strength of -100dBm), so it turns on Bluetooth again to obtain nearby historical connected devices. The smartwatch obtains the nearby mobile phone F, and then uses mobile phone F as the master device and sends a Bluetooth connection request to mobile phone F.

[0108] After the user leaves the parking lot, the mobile network signal strength returns to -85dBm (> the minimum strength of -100dBm). The control module automatically triggers data retransmission and synchronizes the parking data to the parking payment application on mobile phone F to ensure that data is not lost.

[0109] One day, a user wearing smartwatch D entered a shopping mall's underground parking lot. Due to unstable mobile network signals, the acquisition module monitored signal strength every three seconds. Within five minutes, the signal weakened five times, reaching a cumulative 58dBm, triggering the first connection request. The control module activated a dynamic priority scoring mechanism and detected in-car Bluetooth device E (signal -60dBm, battery 80%, CPU utilization 30%) and mobile phone F (signal -75dBm, battery 90%, CPU utilization 20%). E was calculated to have a higher priority score and was selected as the primary device for the connection.

[0110] When the signal strength is less than -100dBm, the watch records area G on the B2 floor of the parking lot as a weak signal location. The next day, when the user reenters the parking lot, the watch activates Bluetooth in advance. At this point, the weight coefficient has been dynamically adjusted through a Markov decision process, prioritizing signal strength.

[0111] After the user leaves the vehicle, the signal between the watch and the car's Bluetooth E continues to weaken, triggering a disconnect condition. The processing module compares the Bluetooth and mobile network signals and finds that both are below the preset strength. It then sends a pause message and disconnects. Later, because the mobile network signal remains weak, the watch rescans and selects phone F based on the score.

[0112] 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 switching times to improve the subsequent connection strategy.

[0113] The above is only an embodiment of the present invention, and the common knowledge such as the specific structure and characteristics of the scheme is not described in detail here. It should be pointed out that for those skilled in the art, without departing from the structure of the present invention, several variations and improvements can be made, which should also be regarded as the scope of protection of the present invention, and these will not affect the effect of the implementation of the present invention and the practicality of the patent. The scope of protection required by this application shall be based on the content of its claims, and the specific implementation methods and other records in the specification can be used to interpret the content of the claims.

Claims

1. A low-power wearable smart watch, characterized in that: include: a collection module, configured to collect the signal strength of the current mobile network signal as first data according to a preset collection period, and associate the collection time of the first data with the first data for storage; a processing module, initializing the number of weakenings and the weakening measurement, comparing the currently acquired first data with the first data acquired in the previous acquisition cycle, and if the first data weakens, increasing the number of weakenings and accumulating the change value of the first data into the weakening measurement; if the first data strengthens, determining whether the change value of the first data is greater than the weakening measurement; if so, initializing the number of weakenings and the weakening measurement; and if the change value of the first data is not greater than the weakening measurement, reducing the weakening measurement according to the change value of the first data; When the number of weakening times is greater than the preset warning number, determining whether the weakening metric is greater than the preset warning metric, and if so, generating a first connection request to initialize the number of weakening times and the weakening metric; If it is not greater than the warning measurement, then continue to accumulate the number of weakening times and the weakening measurement according to the change of the first data; the processing module is further used to generate a first connection request when the currently collected first data is less than the preset minimum intensity; The control module is used to obtain a first connection request, and after obtaining the first connection request, turn on Bluetooth and obtain historical connection devices in the Bluetooth list, select one of the historical connection devices as the main device according to the signal strength of each historical connection device, and send a Bluetooth connection request to the main device.

2. A low-power wearable smart watch according to claim 1, characterized in that: Processing module for collection cycle The processing of the first data includes the following steps: S10: Each acquisition cycle Internal The instantaneous signal strength sampling is recorded as { }, where The original signal strength value of the sampling time, the sampling interval Need to meet ; S20: Sampling sequence Perform sliding average filtering and output the filtered signal The expression formula is as follows: （1）， Or use windowed finite impulse response filtering, the filtered signal The expression formula is as follows formula (2): （2）； S30: The first data of the current collection cycle It is defined as the energy mean of the filtered signal, The calculation formula is as follows: （3）， or when When the fluctuation is less than 3dB, it is simplified to the arithmetic mean, and the calculation formula is shown in the following formula (4): （4）； S40: If a certain sampling value satisfy If the standard deviation of the samples in the period is removed, the sample value is recalculated. .

3. A low-power wearable smart watch according to claim 2, characterized in that: The first data currently obtained The first data obtained in the previous collection cycle The comparison includes the following steps: S50: Define adaptive contrast threshold , The definition formula is as follows: （5）， in, is the minimum allowed threshold; For the front cycles The standard deviation of The calculation formula is as follows: （6）， in, is the sensitivity coefficient; S60: Signal strength change trend Determined by the following piecewise function formula (7): （7）， S70: Weakened meter The update introduces the trend weight factor, The calculation formula is as follows: （8）， in is the weight coefficient.

4. A low-power wearable smart watch according to claim 3, characterized in that: The acquisition module is further configured to acquire the current position of the watch as a first position when acquiring the signal strength of the mobile network, and when the signal strength of the mobile network is less than a preset minimum strength, store the first position as a weak signal position in association with the acquired signal strength of the mobile network; The acquisition module is used to obtain the current location, compare the current location with the weak signal location, obtain the distance between the current location and the weak signal location, and generate a Bluetooth start instruction when the distance is less than a preset first distance and send it to the Bluetooth historical connection device.

5. A low-power wearable smart watch according to claim 4, characterized in that: The processing module is further configured to generate a Bluetooth activation instruction and send it to the Bluetooth historical connection device when the number of weakening times is greater than the preset warning number or the weakening measurement is greater than the preset warning measurement.

6. The low-power wearable smart watch according to claim 5, characterized in that: When the smartwatch is connected to the main device via Bluetooth, 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 times the Bluetooth signal becomes weak and Bluetooth signal weakening measurement , the second data currently obtained The second data of the previous acquisition cycle In contrast, if , then increase and the attenuation of the second data Accumulated to ;like , then judge Is it greater than If it is greater than, reset and ,like No more than ,but reduce ;when Exceeds the preset Bluetooth weakening threshold When, judge Whether the preset Bluetooth weakening measurement threshold is exceeded If it exceeds, a Bluetooth disconnect instruction is generated; the control module is used to obtain the Bluetooth disconnect instruction. When the control module obtains the Bluetooth disconnect instruction, it generates a disconnect Bluetooth message and sends it to the main device, and then turns off Bluetooth.

7. The low-power wearable smart watch according to claim 6, characterized in that: When the acquisition module waits to receive a Bluetooth connection request, the control module stores the data to be uploaded locally, and then re-uploads the data when the signal strength of the mobile network signal is greater than the minimum strength or a Bluetooth connection request is received.

8. The low-power wearable smart watch according to claim 7, characterized in that: The processing module is used to compare the Bluetooth signal strength and the mobile network signal strength. If the Bluetooth signal strength and the mobile network signal strength are both less than the preset minimum signal strength, it generates a connection suspension message and sends it to the main device, then disconnects the Bluetooth connection with the main device, and waits for receiving a Bluetooth connection request.

9. The low-power wearable smart watch according to claim 8, characterized in that: A dynamic priority scoring mechanism is introduced when selecting the master device. The calculation formula is as follows: （9）， in, Connecting devices for history Bluetooth signal strength, The maximum signal strength among the devices currently scanned; For equipment The remaining battery percentage; For equipment CPU utilization, It is the preset upper limit threshold of CPU utilization; 、 and is the weight coefficient, satisfying ; The control module will The highest device is the master device; if there are multiple devices The difference is less than the preset tolerance , then give priority to signal strength Higher equipment.

10. The low-power wearable smart watch according to claim 9, characterized in that: Priority Scoring The weight coefficient 、 and Dynamic adjustment, adjustment rules are 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.

Images