Spatial rhythm decomposition method and device

Abstract

The invention provides a spatial rhythm decomposition method and device, and the method is characterized in that the method comprises the following steps: S1, carrying out the calculation according to corresponding mobile phone signaling data, and obtaining a first set; s2, normalizing each first set to obtain a crowd gathering rhythm curve of the corresponding plot; s3, processing the resident time by using the survey data to obtain a second data set; s4, normalizing each second data set to obtain an initial activity rhythm curve of the corresponding initial activity class; s5, aggregating all the initial activity rhythm curves to obtain a plurality of activity rhythm curves; and S6, for each land parcel, fitting a corresponding crowd gathering rhythm curve according to all the activity rhythm curves to obtain various activity analysis results corresponding to the land parcel. In a word, the method can effectively analyze the space mixed utilization condition of each land parcel.

Description

Technical Field

[0001] The invention belongs to the field of space planning, and in particular relates to a method and device for decomposing space rhythm. Background Art

[0002] With the continuous advancement of urbanization, urban space faces multiple challenges such as slowing population growth and limited construction land. In this context, urban planning practice has gradually shifted from extensive "incremental planning" to "stock planning" to improve the quality of stock. How to achieve efficient mixed utilization of urban space functions within a limited space has become a key issue that needs to be urgently addressed in current urban renewal, which is of great significance to improving the efficiency of spatial resource utilization and the quality of life of residents.

[0003] At present, when evaluating the mixed use of urban space, the mainstream technical means mainly rely on the current land use status, building boundaries and heights, and point of interest (POI) data to reflect the degree of functional mixing of different spaces from the static perspective of the urban built environment. However, this method is difficult to reflect the dynamics of urban space utilization and the diversity of human activities. Under the current demand for stock renewal and refined design, it is difficult to deeply explore the utilization potential of urban space.

[0004] Although mobile phone signaling data can dynamically capture the temporal and spatial distribution patterns of urban populations and approximate full sample statistics, it is difficult to accurately identify specific activity types.

[0005] Therefore, there are still many problems in how to conduct a systematic analysis of the mixed spatial utilization of various urban areas. Summary of the invention

[0006] The present invention is made to solve the above-mentioned problems, and its purpose is to provide a spatial rhythm decomposition method and device.

[0007] The present invention provides a spatial rhythm decomposition method and device, which is used to obtain the analysis results of various activities of each plot according to the time utilization survey data of urban residents and the mobile phone signaling data of each plot in the city, and has the following characteristics, including the following steps: step S1, for each plot, according to the corresponding mobile phone signaling data, calculate the number of people staying in each time period in a day as a first set; step S2, normalize each first set respectively, and obtain the corresponding normalized result as the crowd gathering rhythm curve of the corresponding plot; step S3, process the time utilization survey data of residents, and obtain the number of participants in each time period in a day of each initial activity class as a first set. is the second data set; step S4, normalizing each second data set respectively, and obtaining the corresponding normalized result as the initial activity rhythm curve of the corresponding initial activity class; step S5, aggregating all the initial activity rhythm curves to obtain multiple activity rhythm curves; step S6, for each plot, fitting the corresponding crowd gathering rhythm curve according to all the activity rhythm curves, and obtaining the fitting weight corresponding to each activity rhythm curve as the analysis result of each type of activity corresponding to the plot, wherein each activity rhythm curve corresponds to an activity type respectively, and the fitting weight corresponding to the activity rhythm curve in the plot is used as the proportion of the activity type in the plot.

[0008] The spatial rhythm decomposition method provided by the present invention may also have the following features: wherein the mobile phone signaling data of the plot is collected by each base station in the plot, and the mobile phone signaling data includes the length of time the user stays within the range of the base station,

[0009] In step S1 , the number of people staying in a certain period of time is the total number of users whose stay time in the period of time is greater than a preset stay threshold.

[0010] The spatial rhythm decomposition method provided by the present invention may also have the following characteristics: wherein, in step S2, the normalized calculation expression is: Where v i is the number of people staying in the i-th period of a certain plot, k is the total number of periods, V i It is the normalized crowd gathering intensity of the i-th time period of the plot. The crowd gathering rhythm curve is composed of the crowd gathering intensities of each plot arranged in chronological order.

[0011] The spatial rhythm decomposition method provided by the present invention may also have the following characteristics: wherein, the residents' time utilization survey data includes multiple initial activity categories performed by residents in different time periods of a day. In step S3, the number of participants in the initial activity category in a certain time period of a day is the sum of the number of people performing the initial activity category in the residents' time utilization survey data during that time period.

[0012] The spatial rhythm decomposition method provided by the present invention may also have the following characteristics: in step S5, clustering is performed using the KMeans algorithm, and the similarity between the initial activity rhythm curves is calculated using the DTW distance, the number of clusters is evaluated using the silhouette coefficient, and the activity rhythm curve is the curve of the corresponding cluster center point.

[0013] The spatial rhythm decomposition method provided by the present invention may also have the following characteristics: wherein, in step S6, the similarity r is used as the optimization target, and the fitting weight is calculated according to the fitting formula, and the expression of the fitting formula is: Where S i is the crowd gathering rhythm curve corresponding to the ith plot, n is the total number of activity rhythm curves, w i is the fitting weight corresponding to the ith activity rhythm curve in the ith plot, B i is the i-th activity rhythm curve, ε is the decomposition error, and the calculation expression of the similarity r is: A i =S i .

[0014] The present invention also provides a spatial rhythm decomposition device, which is used to obtain various activity analysis results of each plot based on the city's resident time utilization survey data and the mobile phone signaling data of each plot in the city, and has the following characteristics, including: a signaling data processing module, which is used to calculate the number of people staying in each time period of a day for each plot according to the corresponding mobile phone signaling data as a first set; a first normalization module, which is used to normalize each first set respectively, and obtain the corresponding normalized result as the crowd gathering rhythm curve of the corresponding plot; a survey data processing module, which is used to process the resident time utilization survey data to obtain the number of participants in each initial activity category in each time period of a day as the second data set; a second normalization module, used to normalize each second data set respectively, and obtain the corresponding normalized result as the initial activity rhythm curve of the corresponding initial activity class; an aggregation module, used to aggregate all the initial activity rhythm curves to obtain multiple activity rhythm curves; a fitting module, used to fit the corresponding crowd gathering rhythm curve for each plot according to all the activity rhythm curves, and obtain the fitting weight corresponding to each activity rhythm curve as the analysis result of each type of activity corresponding to the plot, wherein each activity rhythm curve corresponds to an activity type respectively, and the fitting weight corresponding to the activity rhythm curve in the plot is used as the proportion of the activity type in the plot.

[0015] The spatial rhythm decomposition device provided by the present invention may also have the following features: it also includes: a dynamic feature characterization module, which is used to obtain a visualized spatial distribution map of the number of activity participants and a time change map of the number of activity participants based on the total number of people in each plot as dynamic features of spatial mixed utilization of the city.

[0016] The spatial rhythm decomposition device provided by the present invention may also have the following features: wherein the dynamic feature characterization module includes: a calculation unit, for each plot, multiplying the total number of people corresponding to each time period by the corresponding fitting weights, to obtain the number of participants in each time period of each activity type of the plot; a spatial distribution map generation unit, for obtaining a spatial distribution map of the number of activity participants in each activity type in the city based on the number of participants in each time period of each activity type of all plots; a time change map generation unit, for obtaining a time change map of the number of activity participants in each activity type of all plots based on the number of participants in each time period of each activity type.

[0017] Functions and Effects of the Invention

[0018] According to the spatial rhythm decomposition method and device involved in the present invention, first, the mobile phone signaling data is processed by the signaling data processing module and the first normalization module to obtain the crowd gathering rhythm curve; then, the residents' time utilization survey data is processed by the survey data processing module, the second normalization module and the aggregation module to obtain the activity rhythm curve; finally, the crowd gathering rhythm curve and the activity rhythm curve are fitted by the fitting module to obtain the analysis results of various activities of each plot. Therefore, the spatial rhythm decomposition method and device of the present invention can effectively analyze the spatial mixed utilization status of each plot. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 is a block diagram of a spatial rhythm decomposition device in an embodiment of the present invention;

[0020] Figure 2 is a schematic diagram of part of the data in the first set in an embodiment of the present invention;

[0021] Figure 3 is a schematic diagram of the concentration intensity of some people in an embodiment of the present invention;

[0022] Figure 4 is a schematic diagram of initial activity rhythm curves of each initial activity class in an embodiment of the present invention;

[0023] Figure 5 is a schematic diagram of each activity rhythm curve in an embodiment of the present invention;

[0024] Figure 6 is a schematic diagram of six types of activities and their contents in an embodiment of the present invention;

[0025] Figure 7 is a schematic diagram of the relationship between some plots and activity types in an embodiment of the present invention;

[0026] Figure 8 It is a scatter plot comparing the number of workers generated by two methods in each block in the embodiment of the present invention;

[0027] Fig. 9 is a spatial distribution diagram of the number of participants in each activity type in Shanghai on a day in an embodiment of the present invention;

[0028] Fig.10 is a time variation diagram of the number of participants in work and study activities in Shanghai every four hours in an embodiment of the present invention;

[0029] Fig.11 It is a flowchart of the spatial rhythm decomposition method in an embodiment of the present invention. DETAILED DESCRIPTION

[0030] In order to make the technical means, creative features, objectives and effects achieved by the present invention easy to understand, the following embodiments and the accompanying drawings specifically illustrate the spatial rhythm decomposition method and device of the present invention.

[0031] In this embodiment, a spatial rhythm decomposition device is provided, which is used to obtain various activity analysis results of each plot in the city based on the time utilization survey data of urban residents and the mobile phone signaling data of each plot in the city.

[0032] Since residents have different activity characteristics on weekdays and weekends, it is necessary to analyze the various activity analysis results of each plot on weekdays and weekends separately, and the corresponding mobile phone signaling data and resident time utilization survey data also need to be collected on the corresponding weekdays or weekends. In this embodiment, the various activity analysis results of each plot on weekdays are analyzed, and the analysis process on weekends is the same as that on weekdays.

[0033] The mobile phone signaling data of the plot is collected by each base station in the plot, and the mobile phone signaling data includes the length of time the user stays within the range of the base station.

[0034] The residents' time utilization survey data includes multiple initial activity categories performed by residents in different time periods of a day. In this embodiment, the residents surveyed need to classify the activities within 15 minutes in a few days and fill in the corresponding activity locations, for example, activity B at location A at 4:00 and activity C at location A at 4:15.

[0035] In this embodiment, the city is divided into several 500m*500m grids as the basic spatial analysis units, i.e., plots. The boundaries of the plots are then spatially matched with the geographical coordinates of each base station in the city, i.e., the base stations set in the plots are used as the base stations corresponding to the plots, and the base station numbers are matched with the plot numbers to construct a corresponding relationship table, thereby achieving the unification of the spatial scales of the base stations and the plots.

[0036] Figure 1 It is a block diagram of a spatial rhythm decomposition device in an embodiment of the present invention.

[0037] like Figure 1 As shown, the spatial rhythm decomposition device 100 includes a signaling data processing module 10, a first normalization module 20, a survey data processing module 30, a second normalization module 40, an aggregation module 50, a fitting module 60, a dynamic feature characterization module 70 and a control module 80 for controlling the operation of the above modules.

[0038] The signaling data processing module 10 is used to calculate the number of people staying in each time period of a day for each plot according to the corresponding mobile phone signaling data as a first set.

[0039] The number of users staying in a certain period of time is the total number of users whose stay time in the period of time is greater than the preset stay threshold.

[0040] In this embodiment, a single working day is divided into 24 time periods according to the 24-hour system, for example, 00:00-01:59 is a time period, and so on. And the preset stay threshold is set to 30 minutes. The mobile phone signaling data of this embodiment is the data of 5 consecutive working days from Monday to Friday, then the first set includes the average of the number of people staying in each time period of each plot within 5 days, that is, when the machine signaling data does not contain data for only one day, the average is used to calculate the number of people staying. The number of people staying in each plot in a certain time period on a certain day is the sum of the number of people staying in all corresponding base stations in that time period.

[0041] Figure 2 It is a schematic diagram of part of the data in the first set in an embodiment of the present invention.

[0042] like Figure 2 As shown, the first column is the number of each plot, namely nid, and the second to 25th columns are the number of people staying in the corresponding plots in 24 consecutive time periods.

[0043] The first normalization module 20 is used to normalize each first set respectively, and obtain a corresponding normalization result as the crowd gathering rhythm curve of the corresponding plot.

[0044] The normalized calculation expression is:

[0045]

[0046] Where v i is the number of people staying in the i-th period of a certain plot, k is the total number of periods, V i is the normalized crowd gathering intensity of the plot in the i-th period.

[0047] In this embodiment, the crowd gathering rhythm curve is composed of the crowd gathering intensities of the plot arranged in chronological order.

[0048] Figure 3 It is a schematic diagram of the concentration intensity of some people in an embodiment of the present invention.

[0049] like Figure 3 As shown, the first column is the number of each plot, namely nid, and the second to 25th columns are the crowd gathering intensity obtained by normalizing the number of people staying in the corresponding plots in 24 consecutive time periods.

[0050] In this embodiment, the above normalization processing can effectively eliminate the differences in population size between different plots, highlight the relative changes in aggregation intensity, and reflect the unique spatial utilization rhythm of each plot.

[0051] The survey data processing module 30 is used to process the residents' time utilization survey data to obtain the number of participants in each initial activity category in each time period of the day as the second data set.

[0052] The number of participants in a certain period of time in the initial activity class is the sum of the number of people who performed the initial activity class in the period in the residents' time utilization survey data. In this embodiment, since the residents' time utilization survey data is data of multiple days, the sum of the number of people is also the average of the sum of the number of people corresponding to each survey day.

[0053] In this embodiment, the initial activity categories include "sleep + health care + leisure", "food and dining", "employment", "agricultural production + non-agricultural operations", "education and learning", "housework + care", "purchase of goods or services", "medical treatment", "fitness and leisure offline outdoor entertainment" and "visiting and tourism offline indoor activities", a total of ten categories.

[0054] The second normalization module 40 is used to normalize each second data set respectively, and obtain a corresponding normalization result as the initial activity rhythm curve of the corresponding initial activity class.

[0055] Figure 4 Schematic diagram of initial activity rhythm curves of various initial activity classes in an embodiment of the present invention.

[0056] like Figure 4 As shown, the horizontal axis is 24 hours and the vertical axis is the normalized value. It can be seen that the ten initial activity classes have different active characteristics during different time periods within a day.

[0057] The aggregation module 50 is used to aggregate all initial activity rhythm curves to obtain multiple activity rhythm curves.

[0058] Among them, clustering is performed using the KMeans algorithm, and the similarity between the initial activity rhythm curves is calculated using the DTW distance. The number of clusters is evaluated using the silhouette coefficient, and the activity rhythm curve is the curve of the corresponding cluster center point.

[0059] In this embodiment, since the initial activity rhythm curve corresponding to "sleep + hygiene care + leisure" is completely inconsistent with the shape of other initial activity rhythm curves, only the nine curves other than "sleep + hygiene care + leisure" are clustered, and it is calculated that the best effect is achieved when the number of clusters is 5.

[0060] Figure 5 Schematic diagram of each activity rhythm curve in the embodiment of the present invention.

[0061] like Figure 5 As shown, (a) to (e) are the activity rhythm curves corresponding to the five clustered classes, the horizontal axis is 24 hours, and the vertical axis is the normalized value. The light-colored curves in (c) are the initial activity rhythm curves clustered in the class, and the points of the dark-colored curves represent the central values ​​of the values ​​corresponding to all the initial activity rhythm curves in the class.

[0062] In this embodiment, an analysis is performed based on the clustering results, and the activity types corresponding to the five activity rhythm curves are defined. Then, combined with "sleep+hygiene care+leisure", a total of 6 activity types and their contents are obtained.

[0063] Figure 6 It is a schematic diagram of six types of activities and their contents in an embodiment of the present invention.

[0064] like Figure 6 As shown, the first column is the various activity types, and the second column is the content breakdown corresponding to each activity type.

[0065] The fitting module 60 is used to fit the corresponding crowd gathering rhythm curve according to all activity rhythm curves for each plot, and obtain the fitting weight corresponding to each activity rhythm curve as the analysis result of each type of activity corresponding to the plot.

[0066] Among them, the similarity r is taken as the optimization target, and the fitting weight is calculated according to the fitting formula.

[0067] The expression of the fitting formula is:

[0068]

[0069] Where S iis the crowd gathering rhythm curve corresponding to the ith plot, n is the total number of activity rhythm curves, w i is the fitting weight corresponding to the ith activity rhythm curve in the ith plot, B i is the i-th activity rhythm curve, and ε is the decomposition error.

[0070] The calculation expression of similarity r is:

[0071]

[0072] A i =S i .

[0073] Among them, each activity rhythm curve corresponds to an activity type, and the fitting weight corresponding to the activity rhythm curve in the plot is used as the proportion of the activity type in the plot.

[0074] Figure 7 It is a schematic diagram of the relationship between some plots and activity types in an embodiment of the present invention.

[0075] like Figure 7 As shown, the first column is the number of each plot, namely nid, and the second to seventh columns are the fitting weights of each plot on each activity type, namely the proportion of each activity type on the plot.

[0076] In this embodiment, the accuracy of the above-mentioned analysis results of various activities is verified by the scale of employed population. On the one hand, according to the total number of people in each plot and the weight of work and study activities, i.e., fitting weight, the number of people engaged in such activities in each plot and their trend throughout the day are estimated. On the other hand, the actual number of people working in each plot at different times of the weekday is identified using mobile phone signaling data. Then, the total number of employed people obtained by the two methods is compared and a scatter plot is drawn.

[0077] Figure 8 It is a scatter plot comparing the number of workers generated by the two methods in each block in the embodiment of the present invention.

[0078] like Figure 8 As shown in the figure, the horizontal axis is the working population inferred from the mobile phone signaling data, and the vertical axis is the number of workers calculated by time decomposition, that is, the fitting weight. Each point in the figure corresponds to a plot, indicating the number of workers inferred from the mobile phone signaling data and the number of workers calculated by the fitting weight on the plot. It can be seen that the two are significantly positively correlated, with a correlation coefficient as high as 0.96, which shows the effectiveness and reliability of the fitting weight.

[0079] The dynamic feature characterization module 70 is used to obtain a visualized spatial distribution diagram of the number of activity participants and a time variation diagram of the number of activity participants according to the total number of people in each plot as a dynamic feature of the spatial mixed utilization of the city.

[0080] The dynamic feature characterization module 70 includes a calculation unit 701 , a spatial distribution graph generation unit 702 , and a time variation graph generation unit 703 .

[0081] The calculation unit 701 is used to multiply the total number of people corresponding to each time period by the corresponding fitting weights for each plot, so as to obtain the number of participants in each time period for each activity type of the plot.

[0082] The spatial distribution map generating unit 702 is used to obtain a spatial distribution map of the number of participants in each activity type in the city according to the number of participants in each activity type and each time period in all plots.

[0083] Fig. 9 It is a spatial distribution diagram of the number of participants in each activity type in Shanghai on one day in an embodiment of the present invention.

[0084] like Fig. 9 As shown in the figure, the darker the color, the more people participate in work and study activities in the area during the current period. Basic activities such as sleeping and resting, working and studying, eating and dining, and evening fitness have the largest number of participants in the main urban area, followed by secondary clusters in central towns such as Songjiang, Minhang, Nanhui, and Jiading. In general towns, they are distributed in small groups, and some are connected to central towns. This shows that the spatial distribution of the above activities is highly correlated with the overall urban system and population distribution pattern of Shanghai. The spatial distribution pattern of shopping activities is similar to the above activities, but areas with a large number of participants often also gather a large number of shopping facilities, indicating that shopping activities are affected by both population size and facility layout. The scattered points of medical morning exercise activities are evenly distributed, and the peak of participation often occurs in the morning. Combined with map comparison, it is found that areas with higher activity intensity are usually close to hospitals, green spaces, and residential areas. It is speculated that this pattern may be related to the concept of the "15-minute living circle", which reflects the spatial organization logic of meeting leisure and medical needs nearby.

[0085] The time variation graph generating unit 703 is used to obtain a time variation graph of the number of participants in each activity type in the city according to the number of participants in each activity type in each time period of all plots.

[0086] Fig.10 This is a time variation diagram of the number of participants who participate in work and study activities every four hours in Shanghai in an embodiment of the present invention.

[0087] like Fig.10 As shown in the figure, the darker the color, the more people in the area are participating in work and study activities during the current period. It can be seen that the number of people participating in work and study during the daytime is generally large, especially in the central urban area, where a clear gathering area is formed, while the number of people working and studying in various places during the noon and night stages decreases significantly. This dynamic change process is consistent with the work and rest time of most residents, and objectively reflects the typical rhythm of urban life.

[0088] The control module 80 stores a control program for controlling the operation of each module.

[0089] The following describes the process of performing a spatial rhythm decomposition method using the spatial rhythm decomposition device 100 in conjunction with the accompanying drawings.

[0090] Fig.11 It is a flowchart of the spatial rhythm decomposition method in an embodiment of the present invention.

[0091] like Fig.11 As shown, the spatial rhythm decomposition method includes the following steps:

[0092] Step S1, using the signaling data processing module 10 to calculate the number of people staying in each time period of a day for each plot according to the corresponding mobile phone signaling data as a first set.

[0093] Step S2, using the first normalization module 20 to normalize each first set respectively, and obtain the corresponding normalization result as the crowd gathering rhythm curve of the corresponding plot.

[0094] Step S3, using the survey data processing module 30 to process the residents' time utilization survey data, and obtaining the number of participants in each initial activity category in each time period of the day as the second data set.

[0095] Step S4, using the second normalization module 40 to normalize each second data set respectively, and obtain a corresponding normalization result as the initial activity rhythm curve of the corresponding initial activity class.

[0096] Step S5: using the aggregation module 50 to aggregate all the initial activity rhythm curves to obtain multiple activity rhythm curves.

[0097] Step S6, using the fitting module 60 to fit the corresponding crowd gathering rhythm curve according to all activity rhythm curves for each plot, and obtaining the fitting weight corresponding to each activity rhythm curve as the analysis result of each type of activity corresponding to the plot.

[0098] In short, the spatial rhythm decomposition device 100 can obtain the fitting weights of each plot for different activity types, and then quantitatively reveal the distribution characteristics and evolution laws of urban population activities in different regions and time periods, which helps to depict the mixed functional utilization of various plots in the city from a spatiotemporal dynamic perspective.

[0099] Functions and Effects of the Embodiments

[0100] According to the spatial rhythm decomposition method and device involved in this embodiment, first, the mobile phone signaling data is processed by the signaling data processing module and the first normalization module to obtain the crowd gathering rhythm curve; then, the residents' time utilization survey data is processed by the survey data processing module, the second normalization module and the aggregation module to obtain the activity rhythm curve; finally, the crowd gathering rhythm curve and the activity rhythm curve are fitted by the fitting module to obtain the analysis results of various activities of each plot. In short, this method can effectively analyze the spatial mixed utilization status of each plot.

[0101] Those skilled in the art should understand that the present invention is not limited to the above embodiments, and the above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, and these changes and improvements fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A spatial rhythm decomposition method for obtaining various activity analysis results of each plot based on the time utilization survey data of urban residents and the mobile phone signaling data of each plot in the city, characterized in that: The following steps are involved: Step S1, for each of the plots, the number of people staying in each time period in a day is calculated according to the corresponding mobile phone signaling data as a first set; Step S2, normalizing each of the first sets respectively, and obtaining a corresponding normalized result as a crowd gathering rhythm curve of the corresponding plot; Step S3, processing the residents' time utilization survey data to obtain the number of participants in each initial activity category in each time period of the day as a second data set; Step S4, normalizing each of the second data sets respectively to obtain a corresponding normalized result as an initial activity rhythm curve of the corresponding initial activity class; Step S5, aggregating all the initial activity rhythm curves to obtain multiple activity rhythm curves; Step S6, for each of the plots, fitting the corresponding crowd gathering rhythm curve according to all the activity rhythm curves, and obtaining the fitting weight corresponding to each of the activity rhythm curves as the analysis result of each type of activity corresponding to the plot, Each of the activity rhythm curves corresponds to an activity type, and the fitting weight corresponding to the activity rhythm curve in the plot is used as the proportion of the activity type in the plot.

2. The spatial rhythm decomposition method according to claim 1, characterized in that: in, The mobile phone signaling data of the plot is collected by each base station in the plot, The mobile phone signaling data includes the length of time the user stays within the range of the base station, In step S1, the number of people staying in a certain period of time is the total number of users whose stay time in the period of time is greater than a preset stay threshold.

3. The spatial rhythm decomposition method according to claim 1, characterized in that: in, In step S2, the normalized calculation expression is: Where v i is the number of people staying in the i-th period of a certain plot, k is the total number of periods, V i is the normalized crowd gathering intensity of the plot in the i-th period, The crowd gathering rhythm curve is composed of the crowd gathering intensities of the plot arranged in chronological order.

4. The spatial rhythm decomposition method according to claim 1, characterized in that: in, The residents' time utilization survey data includes the initial activity categories performed by a plurality of residents at different time periods in a day. In step S3, the number of participants in the initial activity class in a certain period of time within a day is the total number of people who perform the initial activity class in the period in the residents' time utilization survey data.

5. The spatial rhythm decomposition method according to claim 1, characterized in that: in, In step S5, clustering is performed using the KMeans algorithm, and the similarity between the initial activity rhythm curves is calculated using the DTW distance. The number of clusters of the clusters is evaluated by the silhouette coefficient, The activity rhythm curve is the curve of the corresponding cluster center point.

6. The spatial rhythm decomposition method according to claim 1, characterized in that: in, In step S6, the similarity r is used as the optimization target, and the fitting weight is calculated according to the fitting formula. The expression of the fitting formula is: Where S i is the crowd gathering rhythm curve corresponding to the ith plot, n is the total number of activity rhythm curves, w i is the fitting weight corresponding to the ith activity rhythm curve in the ith plot, B i is the i-th activity rhythm curve, ε is the decomposition error, The calculation expression of the similarity r is: A i =S i 。 7. A spatial rhythm decomposition device, used to obtain various activity analysis results of each plot based on the time utilization survey data of urban residents and the mobile phone signaling data of each plot in the city, characterized in that: include: A signaling data processing module, used for calculating, for each of the plots, the number of people staying in each time period of a day according to the corresponding mobile phone signaling data as a first set; A first normalization module, used for normalizing each of the first sets respectively, and obtaining a corresponding normalization result as a crowd gathering rhythm curve of the corresponding plot; A survey data processing module is used to process the residents' time utilization survey data to obtain the number of participants in each initial activity category in each time period of the day as a second data set; A second normalization module, used for normalizing each of the second data sets respectively, and obtaining a corresponding normalization result as an initial activity rhythm curve of the corresponding initial activity class; an aggregation module, used for aggregating all the initial activity rhythm curves to obtain multiple activity rhythm curves; A fitting module is used to fit the corresponding crowd gathering rhythm curve according to all the activity rhythm curves for each of the plots, and obtain the fitting weight corresponding to each of the activity rhythm curves as the analysis result of the various activities corresponding to the plot, Each of the activity rhythm curves corresponds to an activity type, and the fitting weight corresponding to the activity rhythm curve in the plot is used as the proportion of the activity type in the plot.

8. The spatial rhythm decomposition device according to claim 7, characterized in that: Also includes: The dynamic feature characterization module is used to obtain a visualized spatial distribution map of the number of activity participants and a time variation map of the number of activity participants based on the total number of people in each of the plots as the dynamic features of spatial mixed utilization of the city.

9. The spatial rhythm decomposition device according to claim 8, Features: Wherein, the dynamic feature characterization module includes: A calculation unit, for each of the plots, multiplying the total number of people corresponding to each time period by the corresponding fitting weights to obtain the number of participants in each time period of each activity type of the plot; A spatial distribution map generating unit, configured to obtain a spatial distribution map of the number of participants in each activity type in the city according to the number of participants in each activity type in each time period of all the plots; The time variation graph generating unit is used to obtain the time variation graph of the number of participants in each activity type in the city according to the number of participants in each activity type in each time period of all the plots.

Images