A back clip terminal with BeiDou positioning and low power management functions

Abstract

This invention relates to the field of data processing technology, and in particular to a clip-on terminal with BeiDou positioning and low-power management functions. The terminal includes a processor and a memory. The processor executes a computer program in the memory to perform the following steps: acquiring positioning data of the target user's clip-on terminal at each positioning time; acquiring the activity intensity at each positioning time based on the difference between the positioning data at each positioning time and the previous positioning time; analyzing the activity pattern of the target user based on the changing pattern of the activity intensity at each positioning time; obtaining a final predicted value of the activity intensity for future times based on the activity pattern and the activity intensity at the current positioning time, thereby improving the accuracy of predicting the activity intensity of the target user at future times; and acquiring the adaptive positioning frequency of the clip-on terminal based on the final predicted value of the activity intensity, thus realizing low-power management of the BeiDou positioning module in the clip-on terminal.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, and particularly relates to a back clamping terminal with Beidou positioning and low power consumption management functions. BACKGROUND

[0002] The back clamping terminal is a kind of intelligent expansion device fixed on a mobile device and working cooperatively with the mobile device through Bluetooth, and is mainly used for adding additional professional functions to the mobile device, such as high-precision positioning, data acquisition, identity recognition, etc. Among them, the Beidou positioning function is one of the core functions of the back clamping terminal, and compared with the positioning system in the mobile terminal (such as a mobile phone), the back clamping terminal can usually continuously position in a complex environment, and the positioning precision and stability are relatively high. However, the Beidou positioning module usually consumes high power when continuously running, and the back clamping terminal usually has small size and limited battery capacity, so reducing the power consumption of the positioning module is crucial to improve the running continuity of the back clamping terminal.

[0003] At present, the back clamping terminal mainly adopts a fixed positioning frequency to perform positioning, but since the activity intensity of the user is not fixed, using a fixed positioning frequency may cause unnecessary energy consumption when the user's position changes little, or the positioning accuracy is insufficient when the user's position changes quickly, thereby affecting the power utilization efficiency of the back clamping terminal.

[0004] Therefore, how to adaptively set the positioning frequency of the Beidou positioning module in the back clamping terminal according to different activity intensities of the user, and then realize low power consumption management of the Beidou positioning module in the back clamping terminal has become a problem to be solved. SUMMARY

[0005] Therefore, the embodiments of the present application provide a back clamping terminal with Beidou positioning and low power consumption management functions to solve the problem of how to adaptively set the positioning frequency of the Beidou positioning module in the back clamping terminal according to different activity intensities of the user, and then realize low power consumption management of the Beidou positioning module in the back clamping terminal.

[0006] In the embodiments of the present application, a back clamping terminal with Beidou positioning and low power consumption management functions is provided, which comprises a memory, a processor and a computer program stored in the memory and running on the processor, and the processor implements the following steps when executing the computer program:

[0007] According to the Beidou positioning module in the back clamping terminal of the target user, the positioning data of each positioning time in a preset historical period containing the current positioning time of the back clamping terminal is acquired;

[0008] Based on the location data difference between each location time and its previous location time within a preset historical period, the activity intensity of the target user at each location time within the preset historical period is obtained. Based on all activity intensities, a preset prediction algorithm is used to obtain the initial activity intensity prediction value of the target user at future times.

[0009] Obtain the fitted curves of all activity intensities, divide the fitted curves into at least two curve segments, analyze the periodicity between each curve segment based on the data differences and length differences between each curve segment, and obtain the degree of activity regularity of the target user, which is used to characterize the credibility of the initial activity intensity prediction value.

[0010] Based on the activity patterns of the target user, the initial activity intensity prediction value, and the activity intensity of the target user at the current positioning time, the final activity intensity prediction value of the target user at future times is obtained. Based on the final activity intensity prediction value, the adaptive positioning frequency of the target user's clip-on terminal at future times is obtained, which is used to realize low power management of the Beidou positioning module in the clip-on terminal.

[0011] Preferably, the step of obtaining the activity intensity of the target user at each location time within the preset historical period based on the location data difference between each location time and its previous location time includes:

[0012] For any location within a preset historical time period, the distance between the location data at that location and the location data at the previous location is obtained. The distance is used as the numerator, and the time interval between the location and the previous location is used as the denominator to obtain the activity intensity of the target user at that location.

[0013] Preferably, the step of analyzing the periodicity between each curve segment based on the data and length differences between each curve segment to obtain the degree of activity regularity of the target user includes:

[0014] For any two curve segments, the time interval between each two adjacent positioning times in the two curve segments is obtained, and the average value of all time intervals is recorded as the reference interval.

[0015] Based on the reference interval and the time period corresponding to each curve segment in any two curve segments, obtain the reference time within the time period corresponding to each curve segment in any two curve segments, obtain the data corresponding to each reference time in each of the two curve segments, obtain the reference data sequence for each curve segment in any two curve segments, and obtain the periodic similarity between the two curve segments based on the difference between two data at the same position in the two reference data sequences corresponding to the two curve segments and the length difference of the time period corresponding to the two curve segments.

[0016] Obtain the periodic similarity between any two curve segments, sort all periodic similarities in descending order to obtain a periodic similarity sequence, obtain the fitted curve of the periodic similarity sequence, use the Kneedle algorithm to obtain the first inflection point on the fitted curve of the periodic similarity sequence, and denot it as the periodic similarity threshold. Form a set of curve segments corresponding to two curve segments with periodic similarity greater than or equal to the periodic similarity threshold. If there is no intersection between any two sets of curve segments, then each set of curve segments is denoted as a similar curve segment group, and at least one set of similar curve segment groups is obtained. If there is an intersection between at least two sets of curve segments, then the sets of curve segments with intersection are merged to obtain at least one set of similar curve segment groups.

[0017] The degree of activity regularity of the target user is obtained based on the time interval between the start times of each curve segment in each group of similar curve segments.

[0018] Preferably, obtaining the periodic similarity between any two curve segments based on the difference between two data points at the same position in the two reference data sequences corresponding to any two curve segments, and the difference in the length of the time intervals corresponding to the two curve segments, includes:

[0019] Obtain the sequence lengths of the two reference data sequences corresponding to any two curve segments respectively. Record the reference data sequence with the shortest sequence length as the first reference data sequence and the other reference data sequence as the second reference data sequence. For any data in the first reference data sequence, according to the position of any data in the first reference data sequence, obtain the data with the same position in the second reference data sequence and record it as the target data of any data. Obtain the target data of each data in the first reference data sequence. Calculate the absolute value of the difference between each data in the first reference data sequence and its target data. Linearly normalize the sum of all the absolute values ​​of the differences with the reciprocal of the sum of the sums of the absolute values ​​of the differences and the preset constant to obtain the first similarity between any two curve segments.

[0020] Obtain the time length corresponding to any two curve segments, calculate the absolute value of the difference between the time lengths corresponding to any two curve segments, and linearly normalize the reciprocal of the sum of the absolute value of the difference and a preset constant to obtain the second similarity between any two curve segments.

[0021] The product of the first similarity and the second similarity is calculated to obtain the periodic similarity between any two curve segments.

[0022] Preferably, the step of obtaining the degree of activity regularity of the target user based on the time interval between the start times of each curve segment in each group of similar curve segments includes:

[0023] For any group of similar curve segments, sort the start times of each curve segment in the group of similar curve segments in ascending order to obtain the start time sequence;

[0024] Obtain the time interval between every two adjacent start times in the start time sequence, calculate the standard deviation of all time intervals, and denote the reciprocal of the sum of the standard deviation and a preset constant as the periodic intensity of any group of similar curve segments;

[0025] Obtain the periodic intensity of each group of similar curve segments, and linearly normalize the mean of all periodic intensities to obtain the degree of activity regularity of the target user.

[0026] Preferably, the step of obtaining the final predicted activity intensity value of the target user at a future time based on the activity pattern of the target user, the initial predicted activity intensity value, and the activity intensity of the target user at the current location time includes:

[0027] The degree of activity regularity of the target user is used as the weight of the initial activity intensity prediction value. The difference between the constant 1 and the degree of activity regularity of the target user is used as the weight of the activity intensity of the target user at the current location time. The initial activity intensity prediction value and the activity intensity of the target user at the current location time are weighted and summed to obtain the final activity intensity prediction value of the target user at future time.

[0028] Preferably, obtaining the adaptive positioning frequency of the target user's clip-on terminal at future moments based on the final predicted value of activity intensity includes:

[0029] Using a preset clustering algorithm, the predicted final activity intensity of the target user at a future time, as well as the activity intensity of the target user at each location time within a preset historical period, are clustered to obtain a preset number of clusters, the number of which is the same as the number of preset activity intensity levels.

[0030] The average value of all data in each cluster is calculated to obtain the average activity intensity of each cluster. The average activity intensities of each cluster are sorted in ascending order to obtain the average activity intensity sequence. The target cluster in which the predicted final activity intensity value of the target user is located at a future time is obtained. Based on the position of the target cluster in the average activity intensity sequence, the activity intensity level of the target cluster is obtained. The preset positioning frequency of the activity intensity level of the target cluster is used as the adaptive positioning frequency of the target user's back clip terminal at a future time.

[0031] Preferably, dividing the fitted curve into at least two curve segments includes:

[0032] Obtain all peaks in the fitted curve and use all peaks to divide the fitted curve into at least two curve segments.

[0033] The beneficial effects of the embodiments of the present invention compared with the prior art are as follows:

[0034] This invention predicts the activity intensity of a target user at each location time within a preset historical period using a preset prediction algorithm, obtaining an initial activity intensity prediction value. Considering that the target user's activity is easily affected by random events, the initial activity intensity prediction value may be inaccurate. Therefore, by analyzing the activity intensity of the target user at each location time within the preset historical period, the degree of activity regularity is analyzed to characterize the reliability of the initial activity intensity prediction value. Then, by combining the initial activity intensity prediction value, the degree of activity regularity, and the target user's activity intensity at the current location time, a final activity intensity prediction value is obtained, effectively improving the reliability of the prediction results. Based on the final activity intensity prediction value, the positioning frequency of the BeiDou positioning module in the clip-on terminal is adaptively adjusted. While obtaining a relatively accurate user movement route, this greatly reduces the power consumption of the BeiDou positioning module, improves the applicability of the positioning frequency to actual needs, achieves low-power management of the BeiDou positioning module in the clip-on terminal, and effectively improves the battery life of the clip-on terminal. Attached Figure Description

[0035] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0036] Figure 1 This is a flowchart of an implementation method for a back clip terminal with BeiDou positioning and low power management functions provided in Embodiment 1 of the present invention. Detailed Implementation

[0037] Embodiments of this disclosure are described in detail below, with examples of these embodiments illustrated in the accompanying drawings. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this disclosure, and should not be construed as limiting it.

[0038] It should be noted that the terms "first," "second," etc., used in this disclosure and the accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this disclosure described herein can be implemented in orders other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure.

[0039] To illustrate the technical solution of the present invention, specific embodiments are described below.

[0040] See Figure 1 This is a flowchart of an implementation method for a back clip terminal with BeiDou positioning and low-power management functions provided in Embodiment 1 of the present invention. Figure 1 As shown, the method may include:

[0041] Step S101: Based on the BeiDou positioning module in the target user's clip-on terminal, obtain the positioning data of the clip-on terminal for each positioning time within a preset historical period including the current positioning time.

[0042] The clip-on terminal primarily uses a fixed positioning frequency for positioning. However, since user activity levels are not constant, a fixed positioning frequency may result in frequent positioning when the user's location changes little, leading to unnecessary energy consumption. Conversely, when the user's location changes rapidly, a fixed positioning frequency may suffer from insufficient positioning accuracy, affecting the clip-on terminal's power efficiency. Therefore, in this embodiment of the invention, the user's activity intensity is analyzed using historical positioning data from the clip-on terminal to predict future activity levels. Based on the prediction results, the positioning frequency of the BeiDou positioning module in the clip-on terminal is dynamically adjusted to address the low power efficiency issue caused by the fixed positioning frequency method. This enhances the clip-on terminal's battery life and achieves low-power management.

[0043] Any user to be analyzed is designated as the target user. Based on the historical positioning frequency of the BeiDou positioning module in the target user's clip-on terminal, the positioning data (latitude and longitude) of the target user's clip-on terminal at each positioning time within three days prior to the current positioning time (including the current positioning time) is obtained. At the same time, the preset duration of the historical period is set to three days, which is not limited here. The implementer can set it according to the specific application scenario of the target user's clip-on terminal. This is used to analyze the pattern of the target user's activity intensity through the positioning data within three days and predict the activity intensity of the target user at future times.

[0044] Step S102: Based on the location data difference between each location time and its previous location time within the preset historical period, obtain the activity intensity of the target user at each location time within the preset historical period. Based on all activity intensities, use a preset prediction algorithm to obtain the initial activity intensity prediction value of the target user at future times.

[0045] After obtaining the location data of the target user's clip-on device for each location within the three days prior to the current location time, the activity intensity of the target user at each location within the three days prior to the current location time is first analyzed. Specifically:

[0046] For any location within three days prior to the current location time, the distance between the location data at that location time and the location data at the previous location time is obtained. The distance is used as the numerator, and the time interval between the location time and the previous location time is used as the denominator to obtain the activity intensity of the target user at that location time.

[0047] In one implementation, taking the i-th location time within three days prior to the current location time as an example, the formula for calculating the activity intensity of the target user at the i-th location time is:

[0048]

[0049] in, This represents the activity intensity of the target user at the i-th location time. This represents the distance between the location data at the i-th location time and the location data at the (i-1)-th location time within the three days prior to the current location time. This represents the time interval between the i-th and (i-1)-th positioning times within three days prior to the current positioning time.

[0050] It should be noted that, The larger the value, the farther the target user's movement distance between location time (i-1) and location time (i), meaning the greater the activity intensity of the target user at location time (i). The larger.

[0051] Similarly, the activity intensity of the target user at each location time within the three days prior to the current location time is obtained. Then, based on all activity intensities, the SARIMA algorithm is used to predict the target user's activity intensity at future times, obtaining an initial predicted activity intensity value, denoted as . The choice of prediction algorithm is not restricted; implementers can set it according to the specific scenario, such as LSTM. The SARIMA algorithm is an existing technology and will not be described in detail here.

[0052] Step S103: Obtain the fitted curves of all activity intensities, divide the fitted curves into at least two curve segments, analyze the periodicity between each curve segment based on the data differences and length differences between each curve segment, and obtain the degree of activity regularity of the target user, which is used to characterize the credibility of the initial activity intensity prediction value.

[0053] Since the activity intensity of the target user is affected by both the target user's internal physiological factors and external environmental factors, some random events may disrupt the target user's activity patterns. The initial activity intensity prediction value obtained in step S102 is obtained using a prediction algorithm. The prediction algorithm cannot handle random events well, which may make the initial activity intensity prediction value inaccurate. Therefore, in this embodiment of the invention, it is also necessary to analyze the activity intensity of the target user at each positioning time in the three days before the current positioning time to analyze the degree of activity pattern of the target user, which is used as the credibility of the initial activity intensity prediction value, improves the accuracy of the prediction result of the target user's activity intensity at future times, and thus improves the accuracy of the adaptive adjustment of the positioning frequency of the Beidou positioning module in the target user's back clip terminal.

[0054] The steps for obtaining the degree of regularity in the target user's activities are as follows:

[0055] (1) Using the cubic spline fitting method, the activity intensity of the target user in the three days prior to the current location time is curve fitted to obtain a fitted curve. The data on the fitted curve are the fitted values ​​of the target user's activity intensity. The fitted curve is divided into at least two curve segments using all the peak points on the fitted curve. The similarity between each curve segment is used to determine whether the activity intensity of the target user in the three days prior to the current location time has a high degree of regularity. The cubic spline fitting method is an existing technology and will not be described in detail here.

[0056] (2) Based on the data difference and length difference between each two curve segments, obtain the periodic similarity between each two curve segments.

[0057] Taking the a-th curve segment and the b-th curve segment as an example, firstly, based on the data differences between the a-th curve segment and the b-th curve segment, the first similarity between the a-th curve segment and the b-th curve segment is obtained. Specifically:

[0058] Since the time interval between any two adjacent positioning times may be inconsistent within the three days prior to the current positioning time, in order to facilitate the comparison of data differences between the a-th curve segment and the b-th curve segment, the time interval between any two adjacent positioning times in the a-th and b-th curve segments is obtained respectively. The average value of all time intervals is recorded as the reference interval. The starting time of the a-th and b-th curve segments is taken as the first reference time. Based on the reference interval, all reference times within the corresponding time period of the a-th and b-th curve segments are obtained. Data corresponding to each reference time (i.e., the fitted value of the target user's activity intensity at each reference time) is obtained in the a-th and b-th curve segments respectively, and the reference data sequences corresponding to the a-th and b-th curve segments are obtained respectively.

[0059] Since the lengths of the a-th curve segment and the b-th curve segment may be different, the number of data in their corresponding reference data sequences may also be different. Therefore, the sequence lengths of the reference data sequences corresponding to the a-th curve segment and the b-th curve segment are obtained respectively. The reference data sequence with the shortest sequence length is denoted as the first reference data sequence, and the other reference data sequence is denoted as the second reference data sequence. For any data in the first reference data sequence, according to the position of the data in the first reference data sequence, the data with the same position in the second reference data sequence is obtained and denoted as the target data of the data. The target data of each data in the first reference data sequence is obtained, and the absolute value of the difference between each data in the first reference data sequence and its target data is calculated. The sum of all the absolute values ​​of the difference is linearly normalized to the reciprocal of the sum of the sums of the absolute values ​​of the differences and the preset constant to obtain the first similarity between the a-th curve segment and the b-th curve segment.

[0060] In one embodiment, the formula for calculating the first similarity between the a-th curve segment and the b-th curve segment is:

[0061]

[0062] in, This represents the first similarity between the a-th curve segment and the b-th curve segment. This represents the number of data points in the shortest reference data sequence among the reference data sequences corresponding to the a-th and b-th curve segments. This represents the j-th data point in the reference data sequence corresponding to the a-th curve segment. This represents the j-th data point in the reference data sequence corresponding to the b-th curve segment. This represents a preset constant used to ensure that the fraction is meaningful. In this embodiment of the invention, it is set... There are no restrictions here; implementers can set them according to the specific scenario. Represents a linear normalization function. This represents the absolute value function.

[0063] It should be noted that, The smaller the value, the more similar the a-th curve segment and the b-th curve segment are, indicating that the activity intensity of the target user is more regular, and thus... The larger.

[0064] The first similarity between the a-th and b-th curve segments is based solely on the data differences between the two segments, ignoring the length differences between them. The more consistent the lengths of the a-th and b-th curve segments, the more pronounced the periodicity between them, indicating a more regular activity intensity of the target user. Therefore, after obtaining the first similarity between the a-th and b-th curve segments, it is necessary to obtain the corresponding time lengths and calculate the absolute value of the difference between their corresponding time lengths. The reciprocal of the sum of this absolute difference and a preset constant is then linearly normalized to obtain the second similarity between the a-th and b-th curve segments, thus comprehensively reflecting their similarity.

[0065] The formula for calculating the second similarity between the a-th curve segment and the b-th curve segment is as follows:

[0066]

[0067] in, This represents the second similarity between the a-th curve segment and the b-th curve segment. This represents the time length corresponding to the a-th curve segment. This represents the time length corresponding to the b-th curve segment. This represents a preset constant used to ensure that the fraction is meaningful. In this embodiment of the invention, it is set... There are no restrictions here; implementers can set them according to the specific scenario. Represents a linear normalization function. This represents the absolute value function.

[0068] It should be noted that, The smaller the value, the more consistent the lengths of the a-th and b-th curve segments are, and the more pronounced the periodicity between them. The larger the value, the more regular the activity intensity of the target users.

[0069] The first similarity between the a-th curve segment and the b-th curve segment is obtained. Second similarity Next, calculate the first similarity. Second similarity The product of the two curve segments is used as the periodic similarity between the a-th and b-th curve segments, denoted as . ,Right now .

[0070] Similarly, obtain the periodic similarity between every two curve segments.

[0071] (3) Based on the periodic similarity between any two curve segments, divide all curve segments into at least one group of similar curve segments.

[0072] Specifically: All periodic similarities are sorted in descending order to obtain a periodic similarity sequence. A cubic spline fitting method is used to obtain the fitted curve of the periodic similarity sequence. The Kneedle algorithm is used to obtain the first inflection point on the fitted curve of the periodic similarity sequence, denoted as the periodic similarity threshold. Two curve segments corresponding to periodic similarities greater than or equal to the periodic similarity threshold are grouped into a curve segment set. If any two curve segment sets have no intersection, each curve segment set is denoted as a similar curve segment group, resulting in at least one similar curve segment group. If at least two curve segment sets have an intersection, the curve segment sets with the intersection are merged to obtain at least one similar curve segment group. The Kneedle algorithm is existing technology and will not be elaborated upon here.

[0073] (4) Based on the time interval between the start times of each curve segment in each group of similar curve segments, obtain the degree of activity regularity of the target user.

[0074] Specifically: For any group of similar curve segments, sort the start times of each curve segment in the group of similar curve segments in ascending order to obtain a sequence of start times; obtain the time interval between every two adjacent start times in the sequence of start times, calculate the standard deviation of all time intervals, and denote the reciprocal of the sum of the standard deviation and a preset constant as the periodic intensity of the group of similar curve segments.

[0075] Obtain the periodic intensity of each group of similar curve segments, and linearly normalize the mean of all periodic intensities to obtain the degree of activity regularity of the target user.

[0076] In one implementation, the formula for calculating the degree of regularity of the target user's activities is:

[0077]

[0078] Where k represents the degree of regularity of the target user's activity, and N represents the number of similar curve segments. This represents the standard deviation of the time interval between any two adjacent starting times in the starting time sequence corresponding to the u-th group of similar curve segments. This represents a preset constant used to ensure that the fraction is meaningful. In this embodiment of the invention, it is set... There are no restrictions here; implementers can set them according to the specific scenario. This represents the linear normalization function.

[0079] It should be noted that, The smaller the value, the more consistent the time interval between each two adjacent starting moments is in the starting moment sequence corresponding to the u-th group of similar curve segments. The stronger the periodicity of the curve segments in the u-th group of similar curve segments, that is, the more regular the activity intensity of the target user is, and thus the larger k is.

[0080] This gives us an idea of ​​the degree of regularity in the target users' activities.

[0081] Step S104: Based on the activity pattern of the target user, the initial activity intensity prediction value, and the activity intensity of the target user at the current positioning time, the final activity intensity prediction value of the target user at future times is obtained. Based on the final activity intensity prediction value, the adaptive positioning frequency of the target user's back clip terminal at future times is obtained to realize low power management of the Beidou positioning module in the back clip terminal.

[0082] The higher the degree of activity regularity of the target user obtained in step S103, the higher the reliability of the initial activity intensity prediction value obtained in step S102, that is, the higher the accuracy of the initial activity intensity prediction value; conversely, the lower the degree of activity regularity of the target user, the lower the reliability of the initial activity intensity prediction value obtained in step S102, that is, the lower the accuracy of the initial activity intensity prediction value. At this time, the prediction of the target user's activity intensity in the future also needs to rely on the target user's activity intensity at the current moment, so as to improve the accuracy of the prediction result of the target user's activity intensity in the future moment, obtain the final activity intensity prediction value of the target user in the future moment, and thus improve the accuracy of the adaptive adjustment of the positioning frequency of the Beidou positioning module in the target user's back clip terminal.

[0083] The specific method for obtaining the predicted final activity intensity value of the target user at a future time is as follows:

[0084] The degree of activity regularity of the target user is used as the weight of the initial activity intensity prediction value. The difference between the constant 1 and the degree of activity regularity of the target user is used as the weight of the activity intensity of the target user at the current location time. The initial activity intensity prediction value and the activity intensity of the target user at the current location time are weighted and summed to obtain the final activity intensity prediction value of the target user at future time.

[0085] In one embodiment, the formula for calculating the predicted final activity intensity of the target user at a future time is:

[0086]

[0087] Where M represents the predicted final activity intensity of the target user at a future time. This represents the initial predicted activity intensity. This indicates the activity intensity of the target user at the current location time, and k indicates the regularity of the target user's activity.

[0088] It should be noted that the larger the value of k, the more regular the activity intensity of the target users, and the higher the accuracy of the initial activity intensity prediction. The value of M is related to... The closer the value of M is to k, the less regular the activity intensity of the target user, and the lower the accuracy of the initial activity intensity prediction. In this case, the prediction result of the target user's activity intensity at future times should depend to a large extent on the target user's activity intensity at the current time, that is, the value of M is close to the value of k. The closer the values ​​are.

[0089] After obtaining the predicted value of the target user's final activity intensity at a future time, the k-means clustering algorithm is used to cluster the predicted value of the target user's final activity intensity at a future time, as well as the activity intensity of the target user at each location time within a preset historical period. Here, the k value in the k-means clustering algorithm is set to 4, that is, 4 clusters are obtained. There is no limit here, and the implementer can set the number of clusters according to the specific scenario. Then, the average value of all data in each cluster is calculated to obtain the average activity intensity of each cluster. The average activity intensities of each cluster are sorted in ascending order to obtain the average activity intensity sequence. Based on the position of each average activity intensity in the average activity intensity sequence, the activity intensity level of the corresponding cluster and its corresponding positioning frequency are set. For example, the first data in the average activity intensity sequence is the smallest, meaning that the activity intensity in the cluster corresponding to the first data is relatively low. Therefore, the activity intensity level of the cluster corresponding to the first data in the average activity intensity sequence can be set to "stationary", and its corresponding positioning frequency can be set to 1 time / minute. The activity intensity level of the cluster corresponding to the second data in the average activity intensity sequence is set to "slow movement", and its corresponding positioning frequency is set to 1 time / 10 seconds. The activity intensity level of the cluster corresponding to the third data in the average activity intensity sequence is set to "medium speed movement", and its corresponding positioning frequency is set to 1 time / 5 seconds. The fourth data in the average activity intensity sequence is the largest, meaning that the activity intensity in the cluster corresponding to the fourth data is relatively high. Therefore, the activity intensity level of the cluster corresponding to the fourth data can be set to "high speed movement", and its corresponding positioning frequency is set to 1 time / 1 second. There are no restrictions here; implementers can set the activity intensity level and its corresponding positioning frequency according to the application scenario of the target user's clip-on terminal.

[0090] Based on the cluster where the target user's final activity intensity prediction value belongs in the future, the adaptive positioning frequency of the target user's clip-on terminal in the future is obtained to achieve low-power management of the Beidou positioning module in the clip-on terminal: If the final activity intensity prediction value is in the cluster corresponding to the first data in the average activity intensity sequence, that is, the target user's activity intensity level in the future is "stationary", then the adaptive positioning frequency of the target user's clip-on terminal in the future is 1 time / minute, that is, the clip-on terminal will reposition itself 1 minute after the current positioning time; if the final activity intensity prediction value is in the cluster corresponding to the second data in the average activity intensity sequence, that is, the target user's activity intensity level in the future is "slow movement", then the adaptive positioning frequency of the target user's clip-on terminal in the future is... The adaptive positioning frequency for the target user's clip-on terminal is 1 time / 10 seconds, meaning the target user's clip-on terminal will reposition itself 10 seconds after the current positioning time. If the final activity intensity prediction value is in the cluster corresponding to the third data in the average activity intensity sequence, meaning the target user's activity intensity level in the future is "medium speed movement", then the adaptive positioning frequency for the target user's clip-on terminal in the future is 1 time / 5 seconds, meaning the target user's clip-on terminal will reposition itself 5 seconds after the current positioning time. If the final activity intensity prediction value is in the cluster corresponding to the fourth data in the average activity intensity sequence, meaning the target user's activity intensity level in the future is "high speed movement", then the adaptive positioning frequency for the target user's clip-on terminal in the future is 1 time / 1 second, meaning the target user's clip-on terminal will reposition itself 1 second after the current positioning time.

[0091] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A back clamping terminal with Beidou positioning and low power consumption management function, comprising a memory, a processor and a computer program stored in the memory and running on the processor, characterized in that, The processor implements the following steps when executing the computer program: According to the Beidou positioning module in the back-clamped terminal of the target user, obtain the positioning data of the back-clamped terminal at each positioning time within a preset historical period containing the current positioning time; According to the difference between the positioning data of each positioning time within the preset historical period and the positioning data of the time point above it, obtain the activity intensity of the target user at each positioning time within the preset historical period, and according to all the activity intensities, use a preset prediction algorithm to obtain an initial activity intensity prediction value of the target user at a future time; Obtain a fitting curve of all the activity intensities, divide the fitting curve into at least two curve segments, analyze the periodicity between each curve segment according to the data difference and length difference between each curve segment, and obtain the activity regularity degree of the target user, which is used to represent the credibility of the initial activity intensity prediction value; According to the activity regularity degree of the target user, the initial activity intensity prediction value, and the activity intensity of the target user at the current positioning time, obtain a final activity intensity prediction value of the target user at a future time, and according to the final activity intensity prediction value, obtain an adaptive positioning frequency of the back-clamped terminal of the target user at a future time, which is used to realize low-power management of the Beidou positioning module in the back-clamped terminal; According to the activity regularity degree of the target user, the initial activity intensity prediction value, and the activity intensity of the target user at the current positioning time, obtain a final activity intensity prediction value of the target user at a future time, and according to the final activity intensity prediction value, obtain an adaptive positioning frequency of the back-clamped terminal of the target user at a future time, which is used to realize low-power management of the Beidou positioning module in the back-clamped terminal; According to the activity regularity degree of the target user, the initial activity intensity prediction value, and the activity intensity of the target user at the current positioning time, obtain a final activity intensity prediction value of the target user at a future time, and according to the final activity intensity prediction value, obtain an adaptive positioning frequency of the back-clamped terminal of the target user at a future time, which is used to realize low-power management of the Beidou positioning module in the back-clamped terminal; 2. The back clamping terminal with Beidou positioning and low power consumption management function according to claim 1, characterized in that, According to the difference between the positioning data of each positioning time within the preset historical period and the positioning data of the time point above it, obtain the activity intensity of the target user at each positioning time within the preset historical period, and according to all the activity intensities, use a preset prediction algorithm to obtain an initial activity intensity prediction value of the target user at a future time; According to the difference between the positioning data of each positioning time within the preset historical period and the positioning data of the time point above it, obtain the activity intensity of the target user at each positioning time within the preset historical period, and according to all the activity intensities, use a preset prediction algorithm to obtain an initial activity intensity prediction value of the target user at a future time; 3. The back clamping terminal with Beidou positioning and low power consumption management function according to claim 1, characterized in that, According to the difference between the positioning data of each positioning time within the preset historical period and the positioning data of the time point above it, obtain the activity intensity of the target user at each positioning time within the preset historical period, and according to all the activity intensities, use a preset prediction algorithm to obtain an initial activity intensity prediction value of the target user at a future time; ​ According to the reference interval and the time period corresponding to each of the two arbitrary curve segments, a reference moment in the time period corresponding to each of the two arbitrary curve segments is obtained, data corresponding to each reference moment in each of the two arbitrary curve segments is obtained, a reference data sequence of each of the two arbitrary curve segments is obtained, a difference between two data at the same position in the two reference data sequences corresponding to the two arbitrary curve segments is obtained, and a length difference of the time period corresponding to the two arbitrary curve segments is obtained, so as to obtain a period similarity between the two arbitrary curve segments; A period similarity between each two curve segments is obtained, all period similarities are sorted in descending order to obtain a period similarity sequence, a fitting curve of the period similarity sequence is obtained, a first inflection point on the fitting curve of the period similarity sequence is obtained by using a Kneedle algorithm, and the first inflection point is recorded as a period similarity threshold value, two curve segments corresponding to a period similarity greater than or equal to the period similarity threshold value are combined to form a curve segment set, if there is no intersection between any two curve segment sets, each curve segment set is recorded as a similar curve segment group, and at least one similar curve segment group is obtained, if there is an intersection between at least two curve segment sets, the curve segment sets with the intersection are combined to obtain at least one similar curve segment group. According to a time interval between starting moments of each curve segment in each of the similar curve segment groups, an activity regularity degree of the target user is obtained.

4. The back clamping terminal with Beidou positioning and low power consumption management function according to claim 3, characterized in that, The period similarity between the two arbitrary curve segments is obtained according to the difference between the two data at the same position in the two reference data sequences corresponding to the two arbitrary curve segments and the length difference of the time period corresponding to the two arbitrary curve segments, and includes: The sequence lengths of the two reference data sequences corresponding to the two arbitrary curve segments are obtained respectively, the reference data sequence with the shortest sequence length is recorded as a first reference data sequence, and the other reference data sequence is recorded as a second reference data sequence, for any data in the first reference data sequence, a data at the same position in the second reference data sequence is obtained according to the position of the any data in the first reference data sequence, and the data is recorded as a target data of the any data, the target data of each data in the first reference data sequence is obtained, an absolute value of a difference between each data in the first reference data sequence and the target data thereof is calculated, a reciprocal of a sum of all difference absolute values and a preset constant is linearly normalized to obtain a first similarity between the two arbitrary curve segments; The time lengths corresponding to the two arbitrary curve segments are obtained, an absolute value of a difference between the time lengths corresponding to the two arbitrary curve segments is calculated, and a reciprocal of a sum between the absolute value of the difference and a preset constant is linearly normalized to obtain a second similarity between the two arbitrary curve segments; A product between the first similarity and the second similarity is calculated to obtain the period similarity between the two arbitrary curve segments.

5. The back clamping terminal with Beidou positioning and low power consumption management function according to claim 3, characterized in that, The activity regularity degree of the target user is obtained according to time intervals between starting time points of each curve segment in each of the groups of similar curve segments, and the method comprises the following steps: For any group of similar curve segments, starting time points of each curve segment in the any group of similar curve segments are sorted in ascending order to obtain a sequence of starting time points; A standard deviation of all time intervals is calculated, and an inverse of a sum between the standard deviation and a preset constant is recorded as a periodic intensity of the any group of similar curve segments; The periodic intensities of each group of similar curve segments are obtained, and a mean value of all periodic intensities is linearly normalized to obtain the activity regularity degree of the target user.

6. The back clamping terminal with Beidou positioning and low power consumption management function according to claim 1, characterized in that, The adaptive positioning frequency of the back-clamp terminal of the target user at the future time point is obtained according to the final activity intensity prediction value, and the method comprises the following steps: A preset clustering algorithm is used to cluster the final activity intensity prediction value of the target user at the future time point and the activity intensity of the target user at each positioning time point in a preset historical period to obtain a preset number of clusters, and the number of the clusters is the same as the number of preset activity intensity levels; An average value of all data in each cluster is calculated to obtain an average activity intensity of each cluster, the average activity intensities of each cluster are sorted in ascending order to obtain an average activity intensity sequence, a target cluster in which the final activity intensity prediction value of the target user at the future time point is located is obtained, an activity intensity level of the target cluster is obtained according to a position of the target cluster in the average activity intensity sequence, and a preset positioning frequency corresponding to the activity intensity level of the target cluster is taken as the adaptive positioning frequency of the back-clamp terminal of the target user at the future time point.

7. The back clamping terminal with Beidou positioning and low power consumption management function according to claim 1, characterized in that, The fitted curve is divided into at least two curve segments, and the method comprises the following steps: All peak values in the fitted curve are obtained, and the fitted curve is divided into at least two curve segments by using all the peak values.

Images