WO2024000519A1

WO2024000519A1 - Trajectory data characterization method and apparatus

Info

Publication number: WO2024000519A1
Application number: PCT/CN2022/103140
Authority: WO
Inventors: 唐国斌; 周才发; 八华峰; 曾丹丹; 李亮
Original assignee: 华为技术有限公司
Priority date: 2022-06-30
Filing date: 2022-06-30
Publication date: 2024-01-04

Abstract

A trajectory data characterization method and apparatus. The method comprises: acquiring a plurality of pieces of trajectory data, wherein each of the plurality of pieces of trajectory data includes a plurality of time periods, and characteristic signals, which are acquired in all of the plurality of time periods (S310); and generating a plurality of location points and an attribute of each of the plurality of location points on the basis of the plurality of pieces of trajectory data, wherein the plurality of location points include a first location point, the first location point corresponds to one or more time periods in the plurality of pieces of trajectory data, combination identifiers (ID) of all of the one or more time periods are the same, the combination ID of each time period is generated from the ID of at least one characteristic signal corresponding to the time period, and the attribute of the first location point includes the number of characteristic signals of each class and/or a number distribution characteristic of characteristic signals of each class in the one or more time periods (S320). Storage space for trajectory data can be effectively saved on; moreover, the trajectory data can be compatible with the format of outdoor data, such that integration of indoor data and the outdoor data is achieved.

Description

Trajectory data representation methods and devices

Technical field

The present application relates to the field of indoor and outdoor positioning technologies, and in particular, to a method and device for representing trajectory data.

Background technique

The popularity of mobile devices and the development of mobile communication technology provide different possibilities for realizing positioning and navigation. Taking positioning as an example, the Global Navigation Satellite System (GNSS) chip integrated in mobile devices can be used to provide users with accurate outdoor positioning services. Satellite-based outdoor positioning and navigation can meet many outdoor travel needs of users, but due to obstructions such as buildings, satellite signal-based positioning technology cannot provide usable positioning services indoors.

Existing indoor positioning technology can be divided into three types based on data sources and positioning methods: 1) redeployment indoor positioning: indoor positioning is achieved by deploying specific positioning base stations and communication servers in the indoor environment; 2) light deployment indoor positioning: through An indoor multi-floor fingerprint database is constructed through manual collection, and positioning is achieved by initiating a positioning request and matching the currently scanned fingerprint with the fingerprint database; 3) Zero-deployment indoor positioning: Collect sensors collected by users under specific trigger conditions through crowdsourcing. , WIFI, Global Positioning System (GPS) and other data are uploaded to the cloud. The cloud processes the algorithm to obtain an indoor positioning fingerprint database for subsequent indoor positioning. Among them, zero-deployment indoor positioning has gradually become a mainstream indoor positioning solution due to its many advantages such as low cost, high timeliness, and high degree of automation.

However, in the above zero-deployment positioning solution, since most of the indoor trajectory data does not contain location information such as GNSS and can only be stored one by one, it will face the problem of being unable to store when the amount of data is large; at the same time, there is also the inability to store outdoor data. Compatibility issues.

Contents of the invention

The embodiments of the present application provide a method and device for representing trajectory data, which can effectively save the storage space of trajectory data; at the same time, it can be compatible with the format of outdoor data, realize the integration of indoor and outdoor data, and thereby improve the use of trajectory data in the field of indoor and outdoor positioning. of practicality.

In a first aspect, this application provides a method for characterizing trajectory data. The method includes: acquiring multiple pieces of trajectory data, each piece of trajectory data in the multiple pieces of trajectory data including multiple time periods, and the multiple pieces of trajectory data. Characteristic signals collected in each time period in the time period; generating multiple location points and attributes of each of the multiple location points based on the multiple trajectory data; wherein the multiple location points include A first position point, the first position point corresponds to one or more time periods in the plurality of trajectory data, the combination identifier of each time period in the one or more time periods is the same, and each time period The combined identification of the segment is generated by the identification ID of at least one characteristic signal corresponding to each time period, and the first position point is characterized by the combined identification; the characteristic signals collected in one or more time periods include At least one type of characteristic signal, the attribute of the first position point includes the quantity of each type of characteristic signal in the at least one type of characteristic signal and/or the quantity distribution characteristics of each type of characteristic signal.

Wherein, the first position point is the collection position of the characteristic signal in the one or more time periods.

Wherein, at least one characteristic signal corresponding to each time period includes the characteristic signal collected during the time period.

From a technical effect point of view, due to the uniqueness of the characteristic signal combination identification at different location points, for one or more time periods with the same combination identification in the trajectory data, it can be known that the characteristic signal data in the one or more time periods is Collected at the same location. Furthermore, the characteristic signal data in the one or more time periods can be jointly counted and characterized to obtain data such as the quantity and quantity distribution characteristics of each type of characteristic signal in the one or more time periods as attributes of the location point. That is, this application uses position points and attributes of the position points to characterize multiple trajectory data to replace the existing method of directly storing the original trajectory data, thereby achieving the technical effect of saving trajectory data storage space. Furthermore, when the trajectory data reaches a certain amount, storing the newly collected trajectory data will hardly increase the storage overhead, making it possible to store massive trajectory data in the big data era, which can effectively improve the effectiveness of trajectory data for indoor and outdoor positioning. value.

In a feasible implementation, the one or more time periods include a first time period, the first time period includes M characteristic signals, and the M characteristic signals include Class C characteristic signals, M and C are positive integers; the combination identification is generated by the ID of N characteristic signals among the M characteristic signals, the N characteristic signals include D-type characteristic signals, and N is a positive integer less than or equal to M. , D is a positive integer less than or equal to C.

From a technical effect point of view, since the characteristic signals collected at different locations and the corresponding signal strengths are different, at least one of the characteristic signals collected in each time period can be used to generate a combined identification of each time period. The collection location corresponding to each time period is identified, thereby achieving unified representation and storage of subsequent data collected at the same location point, thus replacing the method of storing trajectory data one by one in the existing technology and saving storage space for trajectory data.

In a feasible implementation, the C-type characteristic signals include first-type characteristic signals, the M characteristic signals include A of the first-type characteristic signals, and the N characteristic signals include B The first type of characteristic signals, A of the first type of characteristic signals include the B of the first type of characteristic signals, the signal strength of the B of the first type of characteristic signals is greater than that of the A of the first type of characteristic signals The signal strength of other characteristic signals; where B is a positive integer less than or equal to A.

From a technical effect point of view, since the signal strength of the characteristic signals collected at different locations is different, the characteristic signal intensity can be used to filter out the characteristic signals used to generate the combined identification in each type of characteristic signal. In this way, the characteristic signals collected at the same location can be The combined identification generated by the collected characteristic signals (that is, the characteristic signal data in one or more time periods is collected at the same location point) is unique, and the combination identification generated by the characteristic signals collected at different location points is different; further, using The similarities and differences of the combined identifiers will uniformly represent and store the characteristic signal data of one or more time periods with the same combined identifier, thus replacing the way of storing trajectory data one by one in the existing technology and saving the storage space of trajectory data.

In a feasible implementation, the attributes of the first location point include indoor and outdoor location identifiers, and the indoor and outdoor location identifiers are used to indicate that the first location point is an indoor location point or an outdoor location point; when the When the first position point is an indoor position point, the collection position of the characteristic signal in the one or more time periods is located indoors. When the first position point is an outdoor position point, the collection position of the characteristic signal in the one or more time periods is The signal collection location is located outdoors.

From a technical effect point of view, since the coordinates of the positions indicated by most outdoor position points are already included in the trajectory data, the positions of different position points are distinguished through indoor and outdoor position markers, so that the coordinates of the outdoor position points can be used to calculate the relevant indoor The longitude and latitude of the location point, so that indoor location points can also be represented and stored based on coordinates, and the compatibility of indoor and outdoor location point data can be achieved.

In a feasible implementation, the plurality of position points also includes a second position point, and the combined identifier characterizing the second position point is generated by the ID of E characteristic signals; when the first position point and When the second position point satisfies one or more of the first condition, the second condition or the third condition, the first position point is adjacent to the second position point; wherein the first condition is : The E characteristic signals and the N characteristic signals include common G characteristic signals, the ratio of G and E, and the ratio of G and N are both greater than or equal to the first threshold; and among the E characteristic signals The difference between the average signal strength of the first signal and the average signal strength of the first signal among the N characteristic signals is less than or equal to the second threshold, and the first signal is any one of the G characteristic signals; The second condition is: the first location point corresponds to H time periods in the multiple trajectory data, the second location point corresponds to 1 time period in the multiple trajectory data, and the H time periods The segment contains J time segments, and the ratio of J to H is greater than or equal to the third threshold. For the second time segment among the J time segments, the I time segment includes the third time segment, and the third time segment The second time period and the third time period are located on the same trajectory data, and the interval between the second time period and the third time period is less than or equal to the fourth threshold, and the second time period is the J Any time period within the time period; the third condition is: the distance between the first position point and the second position point is less than or equal to the fifth threshold, the first position point and the third position point are The distance between the two position points is calculated by the coordinates of the first position point and the coordinates of the second position point. The coordinates of the first position point are calculated by comparing the coordinates collected in the one or more time periods. The position coordinates are summed and averaged.

From a technical effect point of view, for any two position points among the above multiple position points, the any two position points can be determined based on the spatial relationship (i.e., the first condition and the third condition) and the temporal relationship (i.e., the second condition). Whether the positions indicated by the position points are adjacent, and then this adjacent relationship can be used in the future, based on the approximation of the same attribute between adjacent position points, through the first attribute included in the attributes of some position points) to calculate this Some position points have the first attribute of other position points (whose attributes do not include the first attribute) with a certain adjacent relationship. Among them, the first condition indicating the spatial relationship is specifically: for any two position points, when the signal intensity of the characteristic signal that generates their combined identification is close, it can be indicated that the two position points are adjacent or close; the temporal relationship is specifically: : For any one of the multiple time periods corresponding to a location point, among the multiple time periods corresponding to another location point, there is a time period with a collection time close to it, and these two The two time periods are located on the same trajectory data. It can be seen from the movement patterns of indoor people or objects that the collection positions corresponding to these two time periods are adjacent.

In a feasible implementation, the attributes of the first position point include a first attribute, and the method further includes: generating a first graph structure based on the adjacent relationship between the plurality of position points; wherein, The first graph structure is used to describe the connection relationship between the plurality of position points, and when the first position point is adjacent to the second position point, the first position point and the third position point Two position points are connected by an edge in the first graph structure. The first graph structure includes a first type of position point and a second type of position point. The position points of each position point in the first type of position point are The attributes include the first attribute, and the attributes of each position point in the second type of position points do not include the first attribute; the first graph structure is cut to obtain a second graph structure; where , the edge between any two first-type position points in the first graph structure is cut, and the edge between any two second-type position points connected to the first-type position point is cut Cutting, each position point in the second graph structure has a connection relationship with at least one position point.

Wherein, the first attribute is any one of the attributes of the first position point.

From a technical effect point of view, the adjacent relationship of each position point is represented through the first graph structure, and then a second graph composed of the first type of position point and the second type of position point adjacent to the first type of position point is generated. structure, and use the second graph structure to calculate the missing or inaccurate first attributes among the attributes of the second type of location points, so as to remove the farther first type of location points that need to be included in the attributes of the second type of location points. The impact of the calculation of the first attribute.

In a feasible implementation, the method further includes: obtaining an adjacency distance matrix based on the connection relationship between each position point in the second graph structure; wherein each element in the adjacency distance matrix is represented by In order to characterize the number of edges between a second type position point and a first type position point in the second graph structure; normalize the adjacency distance matrix to obtain a weight matrix; Multiply the weight matrix and the first matrix to obtain a second matrix; wherein, the elements in the first matrix are used to characterize the attributes of each first type position point in the second graph structure. An attribute, the elements in the second matrix are used to characterize the first attribute included in the attributes of each second type position point in the second graph structure.

From a technical effect point of view, the distance from the second type position point to the first type position point is represented by the number of edges, and then normalized to obtain the weight matrix of all second type position points in the second graph structure, and then Multiply with the first matrix to accurately obtain the first attribute required by each position point in the second type of position point.

In a feasible implementation, the first attribute is the coordinates of the first location point or an indoor or outdoor location identifier, and the indoor or outdoor location identifier is used to indicate that the first location point is an indoor location point or an outdoor location. location point.

From a technical effect point of view, by completing the coordinates and storing them in the attributes of the location points, the coordinates can be used to replace the combined identification of each indoor location point for indexing, achieving format compatibility and unification with outdoor data. In addition, attribute completion in this way can better compensate for flaws and errors in the data collection process, so that each location point contains complete attributes, effectively improving the practicality of the solution.

In a feasible implementation, the method further includes: performing location positioning, navigation, big data statistical analysis or service recommendation based on the coordinates of each location point in the plurality of location points.

From a technical effect point of view, after calculating the coordinates of each location point, various upper-layer application scenarios can be based on the coordinates of the location points, and the application prospects are broad.

In a feasible implementation, the attributes of the first location point also include L first-type characteristic signals, and the L first-type characteristic signals conform to the quantity distribution characteristics of the first-type characteristic signals.

From a technical perspective, some original data that conforms to the distribution rules can also be stored in the attributes of the location points, so that the true characteristics of the original data can be better preserved while discarding most of the original trajectory data.

In a feasible implementation, the characteristic signal includes one or more of GNSS signals, position coordinates, radio frequency signals, optical signals, acoustic signals, sensor signals or geomagnetic signals, wherein, Radio frequency signals include one or more of WIFI signals, Bluetooth signals, cell CELL signals or ultra-wideband UWB signals.

In a feasible implementation, the quantity distribution characteristics include at least one of normal distribution, Poisson distribution, discrete distribution or interval distribution, and the interval distribution is used to describe the number of characteristic signals located in different intervals.

From a technical effect point of view, interval distribution and/or normal distribution can greatly save the storage space of trajectory data. By saving distribution-specific (such as the mean and variance of normal distribution) and partially consistent with distribution-specific original data, it can be achieved Greatly retain the characteristics of the original data while saving storage space.

In a second aspect, the present application provides a device for characterizing trajectory data. The device includes: an acquisition unit configured to acquire multiple pieces of trajectory data. Each piece of trajectory data in the multiple pieces of trajectory data includes multiple time periods, and the characteristic signals collected in each of the plurality of time periods; a processing unit configured to generate a plurality of position points and attributes of each of the plurality of position points based on the plurality of trajectory data; Wherein, the plurality of position points include a first position point, the first position point corresponds to one or more time periods in the plurality of trajectory data, and each time period in the one or more time periods The combination identification is the same, the combination identification of each time period is generated by the identification ID of at least one characteristic signal corresponding to each time period, the first position point is characterized by the combination identification; the one or The characteristic signals collected in multiple time periods include at least one type of characteristic signal, and the attributes of the first location point include the number of each type of characteristic signal in the at least one type of characteristic signal and/or the quantity distribution characteristics of each type of characteristic signal. .

In a feasible implementation, the attributes of the first location point include a first attribute, and the processing unit is further configured to: generate a first graph structure based on the adjacent relationship between the multiple location points; Wherein, the first graph structure is used to describe the connection relationship between the plurality of position points, and when the first position point is adjacent to the second position point, the first position point and the The second position points are connected by an edge in the first graph structure. The first graph structure includes first type position points and second type position points. Each position in the first type of position point The attributes of the points include the first attribute, and the attributes of each position point in the second type of position points do not include the first attribute; cut the first graph structure to obtain the second graph structure ; Wherein, the edge between any two of the first type position points in the first graph structure is cut, and the edge between any two of the second type of position points connected to the first type of position point The edges are cut, and each position point in the second graph structure has a connection relationship with at least one position point.

In a feasible implementation, the processing unit is further configured to: obtain an adjacency distance matrix based on the connection relationship between each location point in the second graph structure; wherein each of the adjacency distance matrix The element is used to represent the number of edges between a second type position point and a first type position point in the second graph structure; normalize the adjacency distance matrix to obtain a weight matrix; Multiply the weight matrix and the first matrix to obtain a second matrix; wherein the elements in the first matrix are used to represent the attributes contained in each of the first type position points in the second graph structure. The first attribute of the second matrix is used to characterize the first attribute included in the attributes of each second type position point in the second graph structure.

In a feasible implementation, the processing unit is further configured to perform location positioning, navigation, big data statistical analysis, or service recommendation based on the coordinates of each of the multiple location points.

In a third aspect, embodiments of the present application provide a chip system. The chip system includes at least one processor, a memory, and an interface circuit. The memory, the interface circuit, and the at least one processor are interconnected through lines. Instructions are stored in the at least one memory; when the instructions are executed by the processor, the method described in any one of the above first aspects is implemented.

In a fourth aspect, embodiments of the present application provide a computer device. The computer device includes at least one processor, a memory, and an interface circuit. The memory, the interface circuit, and the at least one processor are interconnected through lines. Instructions are stored in the at least one memory; when the instructions are executed by the processor, the method described in any one of the above first aspects is implemented.

In a feasible implementation, the computer device is a server or a terminal device, wherein the terminal device includes a mobile phone, a computer, a car machine, or a tablet.

In a fifth aspect, embodiments of the present application provide a computer-readable storage medium that stores a computer program. When the computer program is executed, the method described in any one of the above-mentioned first aspects can be performed. accomplish.

In a sixth aspect, embodiments of the present application provide a computer program product. The computer program product includes program instructions. When the program instructions are run on a computer, the method described in any one of the above first aspects is implemented. .

Description of drawings

The drawings used in the embodiments of this application are introduced below.

Figure 1 is a schematic diagram of a system architecture provided by an embodiment of the present application;

Figure 2 is a schematic diagram of a trajectory data collection scenario provided by an embodiment of the present application;

Figure 3 is a schematic flow chart of a characterization method for trajectory data provided by an embodiment of the present application;

Figure 4 is a schematic structural diagram of trajectory data in an embodiment of the present application;

Figures 5A-5C are schematic diagrams of the generation process of a graph structure provided by the embodiment of the present application;

Figure 6 is a schematic diagram of the calculation process of a second matrix provided by an embodiment of the present application;

Figures 7A-7B are schematic diagrams of indoor and outdoor spatial topological structures extracted based on the trajectory data representation method of this application;

Figure 8 is a schematic structural diagram of a device for characterizing trajectory data provided by an embodiment of the present application;

Figure 9 is a schematic structural diagram of a computer device provided by this application.

Detailed ways

The embodiments of the present application will be described below with reference to the drawings in the embodiments of the present application. Among them, in the description of the embodiments of this application, unless otherwise stated, "/" means or, for example, A/B can mean A or B; "and/or" in the text is only a way to describe related objects. The association relationship means that there can be three relationships. For example, A and/or B can mean: A exists alone, A and B exist simultaneously, and B exists alone. In addition, in the description of the embodiment of the present application , "plurality" means two or more than two.

The terms “first”, “second”, “third” and “fourth” in the description, claims and drawings of this application are used to distinguish different objects, rather than to describe a specific sequence. . Furthermore, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or alternatively Other steps or elements inherent to such processes, methods, products or devices are also included. Reference herein to "an embodiment" means that a particular feature, structure or characteristic described in connection with the embodiment can be included in at least one embodiment of the present application. The appearances of this phrase in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. Those skilled in the art understand, both explicitly and implicitly, that the embodiments described herein may be combined with other embodiments.

First, the relevant terms in this application are explained.

(1) Crowdsourcing: refers to collecting available data such as sensors, radio frequency signals and network signals of users’ smart terminals without the user’s perception through a certain triggering mechanism. Crowdsourcing is a common way to collect indoor and outdoor trajectory data.

(2) Trajectory data: refers to the collection of characteristic signal data collected by the user's intelligent terminal within a continuous period of time. Each trajectory data represents the characteristic signal data collected at different locations by the intelligent terminal as the user moves, including indoor trajectories. data and outdoor trajectory data.

(3) Cell CELL: A wireless coverage area identified by a base station identification code or a Cell Global Identifier (CGI). When using an omnidirectional antenna structure, the cell is the base station area.

(4) Indoor or Outdoor Location Identifier (IOD): A location identifier generated by whether a GNSS signal is received to identify whether the current location is indoors or outdoors. For example, when the above signal is received, it is set to 1, that is, the location point is located outdoors. When the above signal is not received, it is set to 0, that is, the location point is located indoors.

The system architecture and application scenarios of this application will be introduced below.

Please refer to Figure 1, which is a schematic diagram of a system architecture provided by an embodiment of the present application. As shown in Figure 1, the system architecture includes user equipment 101, user equipment 102, cloud 110 and data storage system 120, where cloud 110 and user equipment perform data transmission through a communication network. It should be noted that the user equipment listed in Figure 1 does not constitute a limit to their number.

The following will introduce two specific application scenarios to which the present invention is applicable based on the system architecture shown in Figure 1.

In the first scenario, the user equipment collects data over a period of time to form trajectory data, and then uploads the trajectory data to the cloud 110 through the communication network. Optionally, the trajectory data collection method may be crowdsourcing, manual collection or other collection methods, which is not limited by this application. The trajectory data includes indoor trajectory data and/or outdoor trajectory data.

The cloud 110 characterizes the trajectory data received in a specific area (such as a shopping mall-centered area) within a period of time (such as 3 days or a month) (i.e., the characterization method of trajectory data in this application) to form the Maps in a specific area (including indoor and outdoor), and store the generated position map and representation data of the trajectory data (ie, the position points and attributes of the position points found in the following embodiments) into the data storage system 120 . When the user device subsequently enters a building in the specific area, it may send a positioning request to the cloud 110 .

The cloud 110 provides users with online positioning services based on the generated map, or sends the user a map of the specific area to provide users with offline low-power positioning services. Then, based on the positioning service and other personalized needs of the user (for example, navigation request and service recommendation request), the user is provided with navigation service and personalized recommendation service (for example, based on the user's specific request to recommend the user in the specific area) merchants, etc.).

Optionally, the cloud 110 may include at least one server, and the data storage system 120 may be a device independent of the server, or integrated within the server, which is not limited in this application.

In the second scenario, when the trajectory data is collected from a relatively small range (when the area radius is smaller than the preset radius, it can be a subway station or a bus station, etc.), the user in Figure 1 The device can independently execute the trajectory data characterization method in this application. The trajectory data at this time may be collected by the user equipment and/or transmitted to the user equipment by other equipment, which is not limited in this application.

Optionally, the user equipment is a terminal device with communication and data collection functions such as a mobile phone, computer, smart watch, bracelet, etc. This application is not limited to this.

Please refer to Figure 2. Figure 2 is a schematic diagram of a trajectory data collection scenario provided by an embodiment of the present application. It is used to describe the collection of trajectory data in a specific area centered on a building. The specific area includes the building and the surrounding area. Streets adjacent to buildings.

Figure 2 shows 6 trajectories (ie, 6 trajectories data) within this specific area. Among them, the data collection locations on Track 1 and Track 2 are all located indoors, the data collection locations on Tracks 3 to 5 include both indoors and outdoors, and the data collection locations on Track 6 are all located outdoors.

Please refer to FIG. 3 , which is a schematic flowchart of a method for characterizing trajectory data provided by an embodiment of the present application. As shown in Figure 3, the method includes step S310 and step S320. in,

Step S310: Acquire multiple pieces of trajectory data, each of the multiple pieces of trajectory data includes multiple time periods, and characteristic signals collected in each of the multiple time periods.

Each track data is a characteristic signal collected within a continuous period of time. The length of the continuous period of time is not limited, for example, it can be 30 minutes or 1 hour. Wherein, the characteristic signals collected within a continuous period of time include at least one type of characteristic signals.

Wherein, the characteristic signal includes one or more of GNSS signals, position coordinates, radio frequency signals, optical signals, acoustic signals, sensor signals or geomagnetic signals, where the radio frequency signals include WIFI signals, Bluetooth signal, cell CELL signal or ultra-wideband UWB signal.

Optionally, the above-mentioned position coordinates may be coordinates in different coordinate reference systems. The coordinate reference system may be a geodetic coordinate system, a projected coordinate system, a custom coordinate system, etc. This application is not limited to this. Among them, the position coordinates in the geodetic coordinate system can be characterized by longitude, latitude, elevation, etc.

Among them, Ultra Wide Band (UWB) is a wireless carrier signal with a frequency bandwidth above 1GHz.

Optionally, the above position coordinates may be calculated through collected GNSS signals, or may be obtained through manual collection or direct acquisition from other terminal devices, which is not limited in this application.

Among them, within the above-mentioned continuous period of time, each type of characteristic signal can be collected at fixed or non-fixed time intervals, and the collection time intervals of different types of characteristic signals can be the same or different.

Among them, for a characteristic signal with signal strength, its collection content at a collection moment includes at least one identification ID of this type of characteristic signal and the corresponding signal strength.

For example, for WIFI signals, the collection content at each collection moment includes the ID of the WIFI signal recognized by the smart terminal (for example, Media Access Control (MAC) ID) and the corresponding signal strength.

For another example, for CELL signals, the collection content at each collection moment includes the ID of the CELL signal recognized by the smart terminal (such as the global cell identifier (Cell Global Identifier, CGI) and/or the physical cell identifier (Physical Cell Identifier, PCI) )) and the corresponding signal strength.

Among them, the relationship between the multiple time periods included in the above-mentioned trajectory data and a continuous period of time during which each trajectory data is collected is: the multiple time periods included in each trajectory data are based on a fixed time interval for this continuous period of time. Or divided into non-fixed time intervals, and any two time periods among the multiple divided time periods do not overlap.

Optionally, the method of dividing a continuous period of time for data collection of each trajectory data can be determined according to the specific scenario, which is not limited in this application.

Optionally, for each type of characteristic signal, each divided time period contains only one collection moment of the type of characteristic signal.

Figure 4 is a schematic structural diagram of trajectory data in an embodiment of the present application. Examples of collecting two types of characteristic signals on a trajectory data are listed. As shown in Figure 4, the collection time of this trajectory data (i.e., the above-mentioned continuous time) is from 12:30 to 12:45, which contains three time periods: 12:30-12:35, 12:35- 12:40, 12:40-12:45 is obtained by dividing the continuous time (15 minutes) according to a fixed time interval (5 minutes). The collected characteristic signals include cell CELL signals and WIFI signals. Among them, the collection times of cell signals are: 12:32, 12:37 and 12:43; the collection times of WIFI signals are 12:34, 12:39 and 12:44. That is, the collection time interval of both cell signal and WIFI signal is 5 minutes.

Optionally, the above-mentioned cell signals collected at each moment include a master station signal and at least one neighbor station signal. The WIFI signals collected at each moment include all WIFI signals recognized by the smart terminal. That is, at each collection moment within the above collection time, the number of collected characteristic signals is at least one.

It should be understood that the above-mentioned figure 4 is only an example of collecting trajectory data, and it does not constitute a limitation on the types of characteristic signals collected, the collection time interval of each type of characteristic signal, etc.

Step S320: Generate multiple location points and attributes of each location point in the multiple location points based on the multiple trajectory data; wherein the multiple location points include a first location point, and the first location point Points in the plurality of trajectory data correspond to one or more time periods, the combination identifier of each time period in the one or more time periods is the same, and the combination identifier of each time period is determined by each time period. The identification ID of at least one characteristic signal corresponding to the segment is generated, and the first location point is characterized by the combined identification; the characteristic signals collected in one or more time periods include at least one type of characteristic signal, and the first The attributes of the location points include the number of each type of characteristic signals in the at least one type of characteristic signals and/or the quantity distribution characteristics of each type of characteristic signals.

It should be understood that this application describes the first position point among multiple position points as an object to represent the generation of each position point and its attributes, that is, the first position point is any one of the multiple position points.

After obtaining multiple time periods of each trajectory data according to the aforementioned method, start constructing the combined identifier of each time period. Time periods with the same combination identifier correspond to the same location point, and the same location point is represented by the combination identifier.

Specifically, the combined identifier of each time period is obtained by combining the identifiers of at least one characteristic signal collected within the time period. From a technical effect point of view, since the types and signal strengths of the characteristic signals collected at different locations are different, combination identifiers can be used to indicate the similarities and differences of the collection locations. That is, the time period with the same combination identifier corresponds to the same collection location, and different combinations The identified time periods correspond to different collection locations.

The following describes the construction process of the combined identifier of each time period, taking one or more time periods corresponding to the first position point as an object.

Optionally, the one or more time periods include a first time period, the first time period contains M characteristic signals, the M characteristic signals include C-type characteristic signals, M and C are positive Integer; the combined identification is generated by the ID of N characteristic signals among the M characteristic signals, the N characteristic signals include D type characteristic signals, N is a positive integer less than or equal to M, and D is less than or equal to A positive integer equal to C.

Wherein, the first time period is any time period among the one or more time periods mentioned above. That is, for each time period, its combination identification is generated by a combination of the IDs of at least one characteristic signal collected within the time period. The at least one characteristic signal includes at least one type of characteristic signal.

For example, the combined identification of each time period can be generated by a combination of IDs of multiple CELL signals (such as a master station signal + multiple neighbor station signals); or by a combination of MAC IDs of multiple WIFI signals; or by at least one It is generated by an ID combination of CELL signal, at least one WIFI signal and at least one Bluetooth signal; or it is generated by an ID combination of at least one WIFI signal, at least one geomagnetic signal and at least one sensor signal, which is not limited by this application.

Among them, the method of generating the combined identification from the ID combination of at least one characteristic signal can be directly using the serial combination of IDs, or can be calculated using a specific algorithm (such as a hash algorithm, etc.).

It should be understood that the more types of characteristic signals used to generate the combined identification, the stronger the uniqueness of the combined identification, that is, the collection locations corresponding to the time periods of different combined identifications are different.

Further, when multiple characteristic signals in the same type of characteristic signals are used to generate a combined identifier, the screening method of multiple characteristic signals in the same type of characteristic signals is as follows:

For characteristic signals with signal strength, the same type of characteristic signals in the first time period are sorted from high to low according to the signal strength, and then a preset number of characteristic signals are filtered out according to the signal strength from high to low, and are The filtered characteristic signals are subsequently used to generate a combined identification of the first time period.

Specifically, the above-mentioned type C characteristic signals include first type characteristic signals. The M characteristic signals include A first-type characteristic signals, the N characteristic signals include B first-type characteristic signals, and the A first-type characteristic signals include the B first-type characteristic signals. Class characteristic signals, the signal strength of the B first class characteristic signals is greater than the signal intensity of other characteristic signals in the A first class characteristic signals; where B is a positive integer less than or equal to A.

Among them, the above-mentioned A first-type characteristic signals include the above-mentioned B first-type characteristic signals. B first-type characteristic signals are selected from A first-type characteristic signals based on the signal strength of the characteristic signal. Specifically, the signal strength of the B first-type characteristic signals is greater than the signal strength of other characteristic signals among the A first-type characteristic signals.

For example, assume that the combined identifier in each time period is generated by a combination of the primary station ID and three neighbor station IDs. If the CELL signals collected in the first time period include a master station signal and 10 neighbor station signals, first filter out the top 3 neighbor station signals with higher signal strength according to the signal strength of the neighbor station signals, and then use the main station signal to The station ID (such as CGI) and the three filtered neighbor station IDs (such as PCI) are combined to generate a combined identification of the first time period.

For example, assume that the combination identifier is generated by a combination of 5 WIFI signals in each time period. If the WIFI signals collected in the first time period include 10 WIFI signals, then the top 5 WIIF signals with higher signal strengths are screened out according to the signal strength of the WIFI signals, and then the ID combination of the top 5 WIFI signals is used to generate The combination identifier of the first time period.

Optionally, the attributes of the first location point include indoor and outdoor location identifiers, and the indoor and outdoor location identifiers are used to indicate that the first location point is an indoor location point or an outdoor location point; when the first location point is When the first position point is an indoor position point, the collection position of the characteristic signal in the one or more time periods is located indoors. When the first position point is an outdoor position point, the collection position of the characteristic signal in the one or more time periods is located in outdoor.

Specifically, the indoor and outdoor identification of the first location point can be determined by counting the number of time periods corresponding to the first location point and whether the characteristic signals collected in each time period include GNSS signals. When the number of time periods during which satellite positioning signals are collected is greater than or equal to the preset ratio, the indoor and outdoor location identifier of the first location point can be set to 1 to indicate one or more time periods corresponding to the first location point. The collection location of the characteristic signal is located outdoors; when the number of time periods in which satellite positioning signals are collected is less than the preset ratio, the indoor and outdoor location identifier of the first location point can be set to 0 to indicate a location corresponding to the first location point. Or the collection location of characteristic signals in multiple time periods is located indoors.

After multiple position points are generated using the acquired trajectory data, adjacent relationships between the multiple position points are determined.

Wherein, the plurality of position points also include a second position point, and the combined identifier representing the second position point is generated from the IDs of the E characteristic signals.

The following takes the first position point and the second position point among the above multiple position points as an example to describe how to determine the adjacent relationship between any two position points among the multiple position points, that is, the first position point and the second position point. are any two of multiple location points.

Specifically, when the first position point and the second position point satisfy one or more of the first condition, the second condition or the third condition, the first position point and the second position point are Adjacent.

That is, different rules can be selected according to specific application scenarios to determine whether the first position point and the second position point are adjacent (or have an adjacent relationship). Wherein, the above-mentioned first position point and the second position point are adjacent means that the distance between the first position point and the second position point is less than or equal to the preset distance.

It can be seen from the above description that there are seven rules for determining whether the first position point and the second position point are adjacent: 1) satisfy the first condition; 2) satisfy the second condition; 3) satisfy the third condition; 4) satisfy the third condition at the same time The first condition and the second condition; 5) The first condition and the third condition are met simultaneously; 6) The first condition and the third condition are met simultaneously; 7) The first condition, the second condition and the third condition are met simultaneously.

The above-mentioned first condition, second condition and third condition are introduced in detail below.

(1)First condition

The first condition is: the E characteristic signals and the N characteristic signals include common G characteristic signals, and the ratio of G and E and the ratio of G and N are both greater than or equal to the first threshold; and The difference between the average signal strength of the first signal among the E characteristic signals and the average signal strength of the first signal among the N characteristic signals is less than or equal to the second threshold, and the first signal is the G characteristic signal. any of the signals.

Specifically, for the first position point, the average signal strength of each of the N characteristic signals used to generate the combined identification is first calculated. Specifically, for each of the N characteristic signals, the signal intensity of the characteristic signal in each time period corresponding to the first position point is summed, and then averaged to obtain each characteristic signal. the average signal strength. In the same way, the average signal strength of each of the E characteristic signals can also be obtained by referring to the above method. Finally, calculate the difference between the average signal intensity of the first characteristic signal among the E characteristic signals and the average signal intensity among the N characteristic signals. When the difference is less than or equal to the second threshold, and each of the G characteristic signals When the characteristic signals all meet the above conditions of the first characteristic signal, and the ratio of G and E and the ratio of G and N are both greater than or equal to the first threshold, then the first vertex and the second vertex are deemed to meet the first condition.

Wherein, the first threshold and the second threshold are set based on specific scenarios. The first threshold is to ensure that the same characteristic signals contained in the E characteristic signals and N characteristic signals reach a certain proportion to prove that the first position point and the second position point are close to each other. E, G and N are positive integers.

(2) Second condition

The second condition is: the first location point corresponds to H time periods in the multiple trajectory data, the second location point corresponds to 1 time period in the multiple trajectory data, and the H time periods The segment contains J time segments, and the ratio of J to H is greater than or equal to the third threshold. For the second time segment among the J time segments, the I time segment includes the third time segment, and the third time segment The second time period and the third time period are located on the same trajectory data, and the interval between the second time period and the third time period is less than or equal to the fourth threshold, and the second time period is the J Any time period within the time period.

Specifically, the second condition is also called a timing condition. For J time periods among the H time periods corresponding to the first position point (the ratio of J to H is greater than or equal to the third threshold), for any one of the J time periods (i.e., the above-mentioned second time period) In terms of , there is a third time period in I time periods corresponding to the second position point. This third time period and the second time period are located on the same trajectory data, and the second time period and the third time period When the time interval is less than or equal to the fourth threshold, it is determined that the first position point and the second position point satisfy the above-mentioned second condition.

For example, the collection time of the second time period is 15:35-15:45, and the collection time of the third time period is 15:45-15:55. The end time of the previous time period (i.e., the second time period) can be used Calculate the time interval with the start time of the subsequent time period (ie, the third time period), that is, the time interval between the second time period and the third time period is 0. Of course, other methods can also be used to calculate the time interval between the above two time periods, and this application is not limited to this.

The third threshold and the fourth threshold are set based on specific scenarios. H, I and J are positive integers.

(3) The third condition

The third condition is: the distance between the first position point and the second position point is less than or equal to a fifth threshold, and the distance between the first position point and the second position point passes through The coordinates of the first position point and the coordinates of the second position point are calculated. The coordinates of the first position point are obtained by summing and averaging the position coordinates collected in the one or more time periods. .

Specifically, for position points with coordinates, the distance between two position points can be directly calculated using the coordinates. The coordinates of each position point are obtained by summing and averaging the position coordinates collected in one or more corresponding time periods.

For example, the first position point corresponds to 100 time periods, and position coordinates are collected in 89 of the 100 time periods. Then the coordinates of the first position point can be obtained using the 89 data collected in the 89 time periods. The position coordinates are calculated by averaging. For the collection process of position coordinates in each time period, please refer to the description in the previous embodiment.

The fifth threshold is set according to specific scenarios, which is not limited in this application.

After the adjacent relationships between multiple location points are determined according to the above embodiment, attribute completion of the location points can be performed based on the adjacent relationships. The attributes of the first position point include the first attribute, and the first attribute is any attribute that needs attribute completion. Attribute completion of the first attribute means: for position points whose first attribute is missing or inaccurate, calculate the first attribute of these position points. The process is as follows:

Optionally, the method further includes: generating a first graph structure based on adjacent relationships between the multiple location points; wherein the first graph structure is used to describe connections between the multiple location points. relationship, and when the first position point is adjacent to the second position point, the first position point and the second position point are connected by an edge in the first graph structure, and the third position point A graph structure includes a first type of location point and a second type of location point. The attributes of each location point in the first type of location point include the first attribute. Each location point in the second type of location point includes the first attribute. The attributes of the position points do not contain the first attribute; the first graph structure is cut to obtain a second graph structure; wherein, the distance between any two first-type position points in the first graph structure The edges are cut, and the edges between any two second-type position points connected to the first-type position points are cut, and each position point in the second graph structure has a connection relationship with at least one position point.

Among them, the first type of location points are location points whose attributes include the first attribute, and the above-mentioned second type of location points are location points whose attributes do not include the first attribute. Wherein, not including the first attribute should be understood to mean that the first attribute does not exist in the attributes or the first attribute is inaccurate. Whether the first attribute is accurate depends on the specific scenario, and this application will not elaborate on it in detail.

Specifically, for the plurality of position points mentioned above, two adjacent position points (having an adjacent relationship) are connected with an edge to generate a first graph structure representing the connection relationship between the plurality of position points. Then cut the edges between all first-type position points in the first graph structure, and cut the edges between any two second-type position points that have an edge relationship with the first-type position points (connected by an edge). Carry out cutting to obtain the second picture structure.

It can be seen that the second graph structure is used to represent the connection relationship between the second type of location points, and between the second type of location points and adjacent first type location points.

Please refer to Figures 5A-5C. Figures 5A-5C are schematic diagrams of a generation process of a graph structure provided by an embodiment of the present application.

FIG. 5A is a specific example of a first graph structure, used to describe adjacent relationships between multiple position points generated from trajectory data collected in a specific area. As shown in Figure 5A, it contains a total of 14 position points (i.e. position points 0-13). Among them, the dotted circle represents the first type of location points (i.e., location points 4-6 and location points 8-13), and the solid line circle represents the second type of location points (i.e., location points 0-3 and location point 7), with There is an edge connecting the two position points of the adjacent relationship.

FIG. 5B is a schematic diagram describing the process of cutting the first image structure shown in FIG. 5A. As shown in FIG. 5B , according to the foregoing description of the cutting method, the dotted line segment indicates the edge to be cut. The process of cutting the edges between all first-type position points in Figure 5B includes cutting between position points 6 and 8, between position points 6 and 9, between position points 5 and 10, and between position points 5 and 11. Cut the edges between, between position points 4 and 12, between position points 4 and 13, and between position points 12 and 13. The process of cutting the edge between any two second-type position points that have an edge relationship with the first-type position point includes cutting the edge between position points 7 and 3.

FIG. 5C is used to describe an example of the second graph structure obtained after cutting. As shown in Figure 5C, it contains 4 second-type position points (ie, position points 0-3) and 3 first-type position points (ie, position points 4-6).

After obtaining the above second graph structure, the second type of position can be calculated based on the first attribute contained in the attributes of each position point in the first type of position point in the second graph structure and the connection relationship between each position point. The first attribute that needs to be included in the attributes of each position point in the point. The specific process is as follows:

Optionally, the method further includes: obtaining an adjacency distance matrix based on the connection relationship between each position point in the second graph structure; wherein each element in the adjacency distance matrix is used to characterize the first The number of edges between a second type position point and a first type position point in the two-graph structure; normalize the adjacency distance matrix to obtain a weight matrix; multiply the weight matrix and the first matrix , obtain the second matrix; wherein, the elements in the first matrix are used to characterize the first attributes contained in the attributes of each first type position point in the second graph structure, and the elements in the second matrix The first attribute used to characterize the attributes of each second type position point in the second graph structure.

Specifically, in the adjacency distance matrix, the elements in each row respectively represent the path from a second type location point to each first type location point (or a second type location point and a first type location point). the number of edges contained between). For example, if the second feature map contains three first-category position points, then the path between a second-category position point and these three first-category position points contains the number of edges (such as 2, 3, and 4), is a row element in the adjacency distance matrix. In the same way, each column element in the adjacency distance matrix can also be used to respectively represent the number of edges contained on the path from a second type location point to each first type location point, and this application is not limited to this. That is, each element in the adjacency distance matrix is used to represent the number of edges contained on the path from a second-type location point to a first-type location point.

Further, optionally, the above-mentioned normalization processing refers to normalization processing in units of each row or column of elements in the adjacency distance matrix. When each row element in the adjacency distance matrix represents the number of edges contained on the path from a second-type position point to each first-type position point, normalization is performed in units of each row element; when the adjacency distance matrix When the elements in each column represent the number of edges contained on the path from a second-category position point to each first-category position point, normalization is performed in units of each column element.

For example, when normalizing each row of elements as a unit, if one of the row elements is 2, 3, and 3, then after normalization, the row elements will be 1/4, 3/8, and 3/8 respectively. .

Wherein, each row or column element in the first matrix is used to represent the first attribute contained in the attributes of each first-type position point. Each row or column element in the second matrix is used to represent the first attribute that needs to be included in the attributes of each second type position point. That is, after calculating the first attribute that needs to be included in the attributes of each second type position point, the calculated first attribute is added to the attributes of the position point and stored.

The following is an example based on Figure 6, taking the first attribute as the coordinate, and introduces the calculation process of the second matrix in detail. At this time, the above-mentioned first type of location point is the location point that contains coordinates in the attribute, and the second type of location point is the location point that is not included in the attribute or contains inaccurate coordinates.

As shown in Figure 6, the adjacency distance matrix is obtained based on the second graph structure in the embodiment shown in Figures 5A-5C. It is a 4*3 matrix, in which each row of elements is used to represent a second type of location point. The number of edges contained on the path to each first-type location point. 0, 1, 2 and 3 on the left side of the adjacency distance matrix represent the second type of position points corresponding to the elements in each row, and 4, 5 and 6 above the adjacency distance matrix represent the first type of position points corresponding to the elements in each column. Each element is used to represent the number of edges contained on the path from a second-category location point to a first-category location point. For example, in the adjacency distance matrix, the elements in the first row and first column represent the second-category location. The number of edges contained on the path from point 0 to first-category point 4, that is, 1. The elements in the third row and second column represent the number of edges contained on the path from second-category point 2 to first-category point 5. quantity, which is 2.

Then, normalization is performed with each row element in the adjacency distance matrix as the unit, and the weight matrix shown in Figure 6 is obtained.

Each row of elements in the first matrix is used to represent the coordinates contained in the attributes of each location point in a first-type location point (in this example, represented by longitude and latitude in the geodetic coordinate system, and the elevation is omitted in this example). That is, in this example, the coordinates included in the attributes of the three first-type position points are (21, 28), (42, 56), and (21, 14) respectively.

Finally, multiply the weight matrix with the first matrix to get the second matrix. Wherein, each row element in the second matrix represents the coordinates (ie, longitude and latitude) that will be included in the attributes of each location point in a second type location point. The coordinates that need to be included in the attributes of the four second-type position points finally calculated are (30, 34), (24, 26), (28, 33), and (30, 42) respectively.

Generally, location points that contain coordinates are outdoor location points, and location points that do not contain coordinates are indoor location points. Therefore, indoor and outdoor location maps can be generated through the above-mentioned trajectory data representation method in this application to provide users with indoor and outdoor positioning and other services based on indoor and outdoor positioning.

In the same way, you can refer to the above coordinate completion process to complete other attributes, which will not be described again here.

Optionally, the first attribute is the coordinates of the first location point or an indoor or outdoor location identifier, and the indoor or outdoor location identifier is used to indicate that the first location point is an indoor location point or an outdoor location point.

Specifically, attributes that can be completed include but are not limited to coordinates of location points, indoor and outdoor location identifiers, and collected feature signals.

Optionally, the above method further includes: performing location positioning, navigation, big data statistical analysis or service recommendation based on the coordinates of each location point in the plurality of location points.

Specifically, after obtaining the coordinates of each of the multiple location points, a map (representing indoor and outdoor locations) of a specific area where the multiple trajectory data is located can be generated based on the coordinates of each location point. Then, after the user enters this specific area, the user can send a positioning request to the server, and the server can provide the user with positioning services based on the generated map of the specific area.

Furthermore, in addition to providing positioning services, navigation and service recommendations can also be provided for users based on user requests (for example, recommending merchants closest to the user).

Among them, the above big data statistical analysis can be based on the characterized regional or city trajectory data to construct a city-level or regional-level heat map to analyze the regional or city's flow of people, etc. This application is not limited to this.

Optionally, the quantity distribution characteristics include at least one of normal distribution, Poisson distribution, discrete distribution or interval distribution, and the interval distribution is used to describe the quantity of characteristic signals located in different intervals.

Furthermore, optionally, the index based on counting the quantity distribution characteristics of each type of characteristic signal may be the signal strength of each type of characteristic signal, etc., which is not limited in this application.

Specifically, when interval distribution is used to characterize the quantity distribution characteristics of a type of characteristic signal, the number of characteristic signals of this type in different signal strength intervals can be counted, and then the range and corresponding number of the signal strength interval are added to the attributes of the location point for storage. . When the normal distribution is used to characterize the quantitative distribution characteristics of a type of feature signal, the mean and variance of the normal distribution obtained based on signal strength statistics can be added to the location point attributes for storage.

It should be understood that other mathematical distribution laws (such as Poisson distribution, discrete distribution, Bernoulli distribution, etc.) can also be used to describe/characterize the quantitative distribution characteristics of each type of characteristic signal, and this application is not limited to this.

Optionally, the attributes of the first location point also include L first-type characteristic signals, and the L first-type characteristic signals conform to the quantity distribution characteristics of the first-type characteristic signals. L is a positive integer.

Wherein, the first type of characteristic signal is any type of characteristic signal among at least one type of characteristic signal mentioned above.

Specifically, for each type of characteristic signal, the attributes of the location point also include part of the original data of each type of characteristic signal, and this part of the original data conforms to the quantity distribution characteristics of the type of characteristic signal. As a result, the storage method based on location points and attributes can retain the original data characteristics of multiple trajectory data to the greatest extent.

For example, the first type of characteristic signal is a WIFI signal, and the number of WIFI signals collected in one or more time periods corresponding to the first location point is 5,000. The quantity distribution characteristics of the 5,000 WIFI signals are represented by a normal distribution, that is, the normal distribution is used to represent the normal distribution characteristics of the 5,000 signal strengths corresponding to the 5,000 signals. At this time, according to the specific scene requirements, 100 WIFI signals can be selected from 5,000 WIFI signals, and the signal strength distribution of these 100 WIFI signals conforms to the normal distribution of the signal strengths of the 5,000 WIFI signals, and these 100 WIFI signals The signal is stored together as an attribute of the first position point.

For another example, when using interval distribution to describe the quantity distribution characteristics of the above 5000 WIIF signals, first count the number of WIFI signals in different signal strength intervals. Assume that the signal strength is between [-10, 0]dbm, [0, The quantity proportions in 10]dbm, [10,20]dbm, and [20,30]dbm are 10%, 20%, 40%, and 30% respectively. Then, 100 original data that are added to the attributes of the location point and stored together are selected from 5,000 WIFI signals, and their quantity distribution also conforms to the distribution rules of the above four signal strength intervals.

In the same way, when the first type of characteristic signal is also represented by other quantity distribution characteristics, the first type of characteristic signal that is selected and stored as an attribute of the first position point must also conform to the corresponding other quantity distribution characteristics.

To sum up, the attributes of the first position point may include at least one of the following three types.

(1) The number of each type of characteristic signal in at least one type of characteristic signal mentioned above.

(2) The quantity distribution characteristics of each type of characteristic signal in at least one type of characteristic signal mentioned above.

(3) Part of the characteristic signals of each type of characteristic signal in at least one type of characteristic signal mentioned above, which part of the characteristic signal conforms to the quantitative distribution characteristics of the characteristic signal of the type to which it belongs.

The technical effects brought about by using the trajectory data characterization method in this application for data characterization and storage are described below based on Table 1. Table 1 is used to represent the compression ratio of trajectory data in a certain area of Nanjing within one week and one month after adopting the trajectory data characterization method in this application.

Among them, the number of trajectory points in Table 1 is the number of time periods obtained after dividing the collected multiple trajectory data. The compression ratio is obtained by dividing the number of trajectory points by the number of position points, and M represents the counting unit—million.

As shown in Table 1, for the trajectory data collected within a week, the number of trajectory points among mobile operators is 113.59M. After using the method of this application to characterize the trajectory data, only 0.94M location points need to be stored and compressed. The magnification reached 120.29 times. Similarly, within a week, for China Unicom and telecom operators, after characterization of trajectory data, the compression ratio reached 79.55 times and 128.44 times respectively.

In addition, it can be seen from Table 1 that as the acquisition time increases, the compression ratio will continue to increase. Within a month, for the trajectory data collected by China Mobile, China Unicom and telecom operators, after characterization of the trajectory data, the compression ratio reached 238.64 times, 142.44 times and 215.02 times respectively.

The main reasons for using the trajectory data representation method in this application to achieve data compression technical effects are:

(1) Position point—the representation method of attributes. Trajectory data is prone to over-accumulation, especially for trajectory data collected through crowdsourcing. Crowdsourcing data itself has a large amount of data. When the data is accumulated to a certain extent, the accumulated data has fully reflected the capabilities of the crowdsourcing data itself. For all the data characteristics expressed, increasing the amount of data at this time has no impact on the data distribution and characteristics. This has reached the boundary of crowdsourcing data accumulation. At this time, the continuous accumulation of increasing data is redundant. In the characterization method of this application, attributes (quantity distribution characteristics and quantities, etc.) are used to characterize multiple types of characteristic signals in a local range, and limited storage of part of the original data not only balances the crowdsourcing data, but also avoids data crowdsourcing. Transition accumulation over data localities and signal categories. Thus achieving the compression effect.

(2) Limitation of location points. The number of location points in an area is limited, and as the collection time continues to increase, the data will also increase linearly. The linearly increasing data will be converted into attributes such as quantity distribution characteristics and quantity, and then compressed and stored into limited location points, so the compression ratio will continue to increase.

Table 1: Compression ratio table of trajectory data in different collection time periods

4G运营商4G operator	累计天数Cumulative days	位置点数量(M)Number of location points (M)	轨迹点数量(M)Number of track points (M)	压缩倍率Compression ratio
移动move	77	944292944292	113592680113592680	120.29120.29
联通 China Unicom	77	319097319097	2538551025385510	79.5579.55
电信 telecommunications	77	587243587243	7542682875426828	128.44128.44
移动 move	3030	23631362363136	508115826508115826	215.02215.02
联通 China Unicom	3030	835072835072	118943509118943509	142.44142.44
电信 telecommunications	3030	14146251414625	337584951337584951	238.64238.64

See Figures 7A-7B. Figures 7A-7B are schematic diagrams of indoor and outdoor spatial topological structures extracted based on the trajectory data representation method of the present application.

As shown in Figure 7A, Figure 7A shows the extracted outdoor spatial topology. The left side of Figure 7A shows the distribution of all trajectory points included in the collected trajectory data (that is, the position points corresponding to each time period included in the trajectory data) on the map. The denser the trajectory points are, the higher the density of human flow is. Usually, roads have the highest density of people flow, and the entire outdoor trajectory data will essentially tend to represent the road network. The left image in Figure 7A shows the location information of outdoor track points, which is consistent with the road network. The right side of Figure 7A shows the outdoor heat statistics map after rasterization of the left image, showing the overall road network situation.

As shown in Figure 7B, Figure 7B shows the extracted indoor spatial topology. Based on the distribution of indoor trajectory points in each area, hotspot indoor areas with dense human flow can be found. These hotspot areas provide the basis for subsequent indoor positioning and floor identification.

Please refer to Figure 8. Figure 8 is a schematic structural diagram of a trajectory data characterization device provided by an embodiment of the present application, which can be used to perform each step in the foregoing method embodiment. As shown in Figure 8, the device includes an acquisition unit 810 and a processing unit 820. in,

The acquisition unit 810 is configured to acquire multiple pieces of trajectory data, each of the multiple pieces of trajectory data including multiple time periods, and the characteristic signals collected in each of the multiple time periods.

The processing unit 820 is configured to generate multiple location points and attributes of each location point in the multiple location points based on the multiple trajectory data.

Wherein, the plurality of position points include a first position point, the first position point corresponds to one or more time periods in the plurality of trajectory data, and each time period in the one or more time periods The combination identification is the same, the combination identification of each time period is generated by the identification ID of at least one characteristic signal corresponding to each time period, the first position point is characterized by the combination identification; the one or The characteristic signals collected in multiple time periods include at least one type of characteristic signal, and the attributes of the first location point include the number of each type of characteristic signal in the at least one type of characteristic signal and/or the quantity distribution characteristics of each type of characteristic signal. .

In a feasible implementation, the C-type characteristic signals include first-type characteristic signals; the M characteristic signals include A of the first-type characteristic signals, and the N characteristic signals include B The first type of characteristic signals, A of the first type of characteristic signals include the B of the first type of characteristic signals, the signal strength of the B of the first type of characteristic signals is greater than that of the A of the first type of characteristic signals The signal strength of other characteristic signals; where B is a positive integer less than or equal to A.

In a feasible implementation, the attributes of the first location point include a first attribute, and the processing unit 820 is further configured to: generate a first graph structure based on the adjacent relationship between the multiple location points. ; Wherein, the first graph structure is used to describe the connection relationship between the plurality of position points, and when the first position point is adjacent to the second position point, the first position point and The second location points are connected by an edge in the first graph structure. The first graph structure includes first type location points and second type location points. Each of the first type location points The attributes of the position points include the first attribute, and the attributes of each position point in the second type of position points do not include the first attribute; cut the first graph structure to obtain the second graph Structure; wherein, the edge between any two of the first type position points in the first graph structure is cut, and the edge between any two of the second type of position points connected to the first type of position point is The edges are cut, and each position point in the second graph structure has a connection relationship with at least one position point.

In a feasible implementation, the processing unit 820 is further configured to: obtain an adjacency distance matrix based on the connection relationship between each location point in the second graph structure; wherein each element in the adjacency distance matrix elements are used to represent the number of edges between a second type position point and a first type position point in the second graph structure; normalize the adjacency distance matrix to obtain a weight matrix ; Multiply the weight matrix and the first matrix to obtain a second matrix; wherein the elements in the first matrix are used to characterize the attributes of each first type position point in the second graph structure. The elements in the second matrix are used to characterize the first attributes included in the attributes of each second type position point in the second graph structure.

In a feasible implementation, the processing unit 820 is further configured to perform location positioning, navigation, big data statistical analysis, or service recommendation based on the coordinates of each location point in the plurality of location points.

Specifically, the specific execution process of each unit in the trajectory data characterization device can be referred to the specific steps in the method embodiment of FIG. 3, and will not be described again here.

Please refer to FIG. 9 . FIG. 9 is a schematic structural diagram of a computer device provided by the present application. It may be a server or user device in the cloud 110 in the embodiment of FIG. 1 . The computer device includes a memory 901, one or more (only one is shown in the figure) processors 902, an interface circuit 903 and a bus 904. Among them, the memory 901, the processor 902, and the interface circuit 903 implement communication connections between each other through the bus 904.

The processor 902 is configured to obtain multiple pieces of trajectory data through the interface circuit 903. Each piece of trajectory data in the multiple pieces of trajectory data includes multiple time periods, and the characteristic signals collected in each of the multiple time periods. . And generate a plurality of position points and attributes of each position point in the plurality of position points based on the plurality of trajectory data; wherein the plurality of position points include a first position point, and the first position point is at The multiple pieces of trajectory data correspond to one or more time periods, the combination identifier of each time period in the one or more time periods is the same, and the combination identifier of each time period is determined by the number of times in each time period. The identification ID of at least one characteristic signal collected is generated, and the first location point is characterized by the combined identification; the characteristic signal collected in one or more time periods includes at least one type of characteristic signal, and the first location point is characterized by the combined identification. The attributes of the location points include the number of each type of characteristic signals in the at least one type of characteristic signals and/or the quantity distribution characteristics of each type of characteristic signals.

The memory 901 may be used to store the above-mentioned plurality of location points and attributes of each of the plurality of location points.

Specifically, the specific process of the computer device performing the characterization method of trajectory data can be referred to the various steps in the method embodiment of FIG. 3, which will not be described again here.

The memory 901 can be any one of random access memory (Random Access Memory, RAM), read-only memory (Read-Only Memory, ROM) or flash memory (Flash Memory); wherein, RAM includes static random access memory (Static random access memory). RAM, SRAM) and dynamic random access memory (Dynamic RAM, DRAM), etc. ROM includes erasable programmable ROM (Erasable Programmable ROM, EPROM) and electrically erasable programmable read-only memory (Electrically Erasable Programmable ROM, EEPROM), etc. The memory 901 can store programs. When the program stored in the memory 901 is executed by the processor 902, the processor 902 and the interface circuit 903 are used to execute various steps of the trajectory data characterization method in the embodiment of the present application.

The processor 902 may be a general central processing unit (CPU), a microprocessor, an application specific integrated circuit (ASIC), a graphics processing unit (GPU), or one or more Integrated circuits, etc., are used to execute relevant programs to realize the functions required to be performed by each unit in the trajectory data characterization device according to the embodiment of the present application, or to execute the trajectory data characterization method according to the method embodiment of the present application.

The interface circuit 903 uses a transceiver device such as, but not limited to, a transceiver to implement communication between the computer device and other devices or communication networks. For example, trajectory data can be obtained from a smart terminal (user equipment) through the interface circuit 903.

Bus 904 may include a path for transmitting information between various components (eg, memory 901, processor 902, interface circuit 903) on the computer device shown in Figure 9.

The embodiment of the present application provides a chip system. The chip system includes at least one processor, a memory and an interface circuit. The memory, the interface circuit and the at least one processor are interconnected through lines. The at least one memory Instructions are stored in; when the instructions are executed by the processor, some or all of the steps described in any of the above method embodiments can be realized.

Embodiments of the present application provide a computer storage medium. The computer storage medium stores a computer program. When the computer program is executed, some or all of the steps described in any of the above method embodiments can be realized.

Embodiments of the present application provide a computer program product. The computer program product includes program instructions. When the program instructions are run on a computer, some or all of the steps described in any of the above method embodiments can be implemented.

In the above embodiments, each embodiment is described with its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments. It should be noted that for the sake of simple description, the foregoing method embodiments are expressed as a series of action combinations. However, those skilled in the art should know that the present application is not limited by the described action sequence. Because according to this application, certain steps may be performed in other orders or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily necessary for this application.

In the several embodiments provided in this application, it should be understood that the disclosed device can be implemented in other ways. For example, the device embodiments described above are only illustrative. For example, the division of the above units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated. to another system, or some features can be ignored, or not implemented. On the other hand, the coupling or direct coupling or communication connection between each other shown or discussed may be through some interfaces, and the indirect coupling or communication connection of the devices or units may be in electrical or other forms.

The units described above as separate components may or may not be physically separated. The components shown as units may or may not be physical units, that is, they may be located in one place, or they may be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

As mentioned above, the above embodiments are only used to illustrate the technical solution of the present application, but not to limit it. Although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still make the foregoing technical solutions. The technical solutions described in each embodiment may be modified, or some of the technical features may be equivalently replaced; however, these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions in each embodiment of the present application.

Claims

A method for characterization of trajectory data, characterized in that the method includes:

Acquire multiple pieces of trajectory data, each of the multiple pieces of trajectory data includes multiple time periods, and characteristic signals collected in each of the multiple time periods;

Generate multiple location points and attributes of each location point in the multiple location points based on the multiple trajectory data;

Wherein, the plurality of position points include a first position point, the first position point corresponds to one or more time periods in the plurality of trajectory data, and each time period in the one or more time periods The combination identification of each time period is the same, the combination identification of each time period is generated by the identification ID of at least one characteristic signal corresponding to each time period, the first position point is characterized by the combination identification; the one or The characteristic signals collected in multiple time periods include at least one type of characteristic signal, and the attributes of the first location point include the number of each type of characteristic signal in the at least one type of characteristic signal and/or the quantity distribution characteristics of each type of characteristic signal. .
The method according to claim 1, characterized in that:

The one or more time periods include a first time period, the first time period includes M characteristic signals, the M characteristic signals include C-type characteristic signals, and M and C are positive integers;

The combined identification is generated by the ID of N characteristic signals among the M characteristic signals. The N characteristic signals include D type characteristic signals. N is a positive integer less than or equal to M, and D is less than or equal to C. positive integer.
The method according to claim 2, characterized in that the type C characteristic signal includes a first type characteristic signal,

The M characteristic signals include A first-type characteristic signals, the N characteristic signals include B first-type characteristic signals, and the A first-type characteristic signals include the B first-type characteristic signals. Class characteristic signals, the signal strength of the B first class characteristic signals is greater than the signal intensity of other characteristic signals in the A first class characteristic signals; where B is a positive integer less than or equal to A.
The method according to claim 3, characterized in that:

The attributes of the first location point also include L first-type characteristic signals, and the L first-type characteristic signals conform to the quantity distribution characteristics of the first-type characteristic signals.
The method according to any one of claims 2-4, characterized in that,

The plurality of position points also includes a second position point, and the combined identifier characterizing the second position point is generated by the ID of the E characteristic signals;

When the first position point and the second position point satisfy one or more of the first condition, the second condition or the third condition, the first position point is adjacent to the second position point;

Wherein, the first condition is: the E characteristic signals and the N characteristic signals include common G characteristic signals, and the ratio of G and E and the ratio of G and N are both greater than or equal to the first threshold; And the difference between the average signal strength of the first signal among the E characteristic signals and the average signal strength of the first signal among the N characteristic signals is less than or equal to the second threshold, and the first signal is the G Any one of the characteristic signals;

The second condition is: the first location point corresponds to H time periods in the multiple trajectory data, the second location point corresponds to 1 time period in the multiple trajectory data, and the H time periods The segment contains J time segments, and the ratio of J to H is greater than or equal to the third threshold. For the second time segment among the J time segments, the I time segment includes the third time segment, and the third time segment is included in the segment. The second time period and the third time period are located on the same trajectory data, and the interval between the second time period and the third time period is less than or equal to the fourth threshold, and the second time period is the J Any time period within the time period;

The third condition is: the distance between the first position point and the second position point is less than or equal to a fifth threshold, and the distance between the first position point and the second position point passes through The coordinates of the first position point and the coordinates of the second position point are calculated. The coordinates of the first position point are obtained by summing and averaging the position coordinates collected in the one or more time periods. .
The method of claim 5, wherein the attributes of the first location point include a first attribute, and the method further includes:

A first graph structure is generated based on the adjacent relationship between the multiple location points; wherein the first graph structure is used to describe the connection relationship between the multiple location points, and when the first location point When adjacent to the second position point, the first position point and the second position point are connected by an edge in the first graph structure, and the first graph structure contains the first type of position point. and a second type of position point, the attributes of each position point in the first type of position point include the first attribute, and the attributes of each position point in the second type of position point do not include the first attribute;

Cut the first graph structure to obtain a second graph structure; wherein, the edge between any two first-type position points in the first graph structure is cut, and is connected to the first-type position point. The edges between any two connected second type position points are cut, and each position point in the second graph structure has a connection relationship with at least one position point.
The method of claim 6, further comprising:

Based on the connection relationship between each position point in the second graph structure, an adjacency distance matrix is obtained; wherein each element in the adjacency distance matrix is used to characterize one of the second categories in the second graph structure The number of edges between a location point and one of the first-type location points;

Normalize the adjacency distance matrix to obtain a weight matrix;

Multiply the weight matrix and the first matrix to obtain a second matrix; wherein the elements in the first matrix are used to represent the attributes contained in each of the first type position points in the second graph structure. The first attribute of the second matrix is used to characterize the first attribute included in the attributes of each second type position point in the second graph structure.
The method according to claim 6 or 7, characterized in that,

The first attribute is the coordinates of the first location point or an indoor or outdoor location identifier, and the indoor or outdoor location identifier is used to indicate that the first location point is an indoor location point or an outdoor location point.
The method of claim 8, further comprising:

Position positioning, navigation, big data statistical analysis or service recommendation are performed based on the coordinates of each of the plurality of position points.
The method according to any one of claims 1-9, characterized in that,

The quantity distribution characteristics include at least one of normal distribution, Poisson distribution, discrete distribution or interval distribution, and the interval distribution is used to describe the number of characteristic signals located in different intervals.
The method according to any one of claims 1-10, characterized in that,

The characteristic signal includes one or more of GNSS signals, position coordinates, radio frequency signals, optical signals, acoustic signals, sensor signals or geomagnetic signals, where the radio frequency signals include WIFI signals, Bluetooth signals, One or more of cell CELL signals or ultra-wideband UWB signals.
A device for characterizing trajectory data, characterized in that the device includes:

An acquisition unit, configured to acquire multiple pieces of trajectory data, each of the multiple pieces of trajectory data including multiple time periods, and characteristic signals collected in each of the multiple time periods;

A processing unit configured to generate a plurality of position points and attributes of each position point in the plurality of position points based on the plurality of trajectory data;

Wherein, the plurality of position points include a first position point, the first position point corresponds to one or more time periods in the plurality of trajectory data, and each time period in the one or more time periods The combination identification is the same, the combination identification of each time period is generated by the identification ID of at least one characteristic signal corresponding to each time period, the first position point is characterized by the combination identification; the one or The characteristic signals collected in multiple time periods include at least one type of characteristic signal, and the attributes of the first location point include the number of each type of characteristic signal in the at least one type of characteristic signal and/or the quantity distribution characteristics of each type of characteristic signal. .
The device according to claim 12, characterized in that:

The one or more time periods include a first time period, the first time period includes M characteristic signals, the M characteristic signals include C-type characteristic signals, and M and C are positive integers;

The combined identification is generated by the ID of N characteristic signals among the M characteristic signals. The N characteristic signals include D type characteristic signals. N is a positive integer less than or equal to M, and D is less than or equal to C. positive integer.
The device according to claim 13, characterized in that the type C characteristic signal includes a first type characteristic signal;

The M characteristic signals include A first-type characteristic signals, the N characteristic signals include B first-type characteristic signals, and the A first-type characteristic signals include the B first-type characteristic signals. Class characteristic signals, the signal strength of the B first class characteristic signals is greater than the signal intensity of other characteristic signals in the A first class characteristic signals; where B is a positive integer less than or equal to A.
The device according to claim 14, characterized in that:

The attributes of the first location point also include L first-type characteristic signals, and the L first-type characteristic signals conform to the quantity distribution characteristics of the first-type characteristic signals.
The device according to any one of claims 12-15, characterized in that,

The plurality of position points also includes a second position point, and the combined identifier characterizing the second position point is generated by the ID of the E characteristic signals;

When the first position point and the second position point satisfy one or more of the first condition, the second condition or the third condition, the first position point is adjacent to the second position point;

Wherein, the first condition is: the E characteristic signals and the N characteristic signals include common G characteristic signals, and the ratio of G and E and the ratio of G and N are both greater than or equal to the first threshold; And the difference between the average signal strength of the first signal among the E characteristic signals and the average signal strength of the first signal among the N characteristic signals is less than or equal to the second threshold, and the first signal is the G Any one of the characteristic signals;

The second condition is: the first location point corresponds to H time periods in the multiple trajectory data, the second location point corresponds to 1 time period in the multiple trajectory data, and the H time periods The segment contains J time segments, and the ratio of J to H is greater than or equal to the third threshold. For the second time segment among the J time segments, the I time segment includes the third time segment, and the third time segment The second time period and the third time period are located on the same trajectory data, and the interval between the second time period and the third time period is less than or equal to the fourth threshold, and the second time period is the J Any time period within the time period;

The third condition is: the distance between the first position point and the second position point is less than or equal to a fifth threshold, and the distance between the first position point and the second position point passes through The coordinates of the first position point and the coordinates of the second position point are calculated. The coordinates of the first position point are obtained by summing and averaging the position coordinates collected in the one or more time periods. .
The device according to claim 16, wherein the attributes of the first location point include a first attribute, and the processing unit is further configured to:

A first graph structure is generated based on the adjacent relationship between the multiple location points; wherein the first graph structure is used to describe the connection relationship between the multiple location points, and when the first location point When adjacent to the second position point, the first position point and the second position point are connected by an edge in the first graph structure, and the first graph structure contains the first type of position point. and a second type of position point, the attributes of each position point in the first type of position point include the first attribute, and the attributes of each position point in the second type of position point do not include the first attribute;

Cut the first graph structure to obtain a second graph structure; wherein, the edge between any two first-type position points in the first graph structure is cut, and is connected to the first-type position point. The edges between any two connected second type position points are cut, and each position point in the second graph structure has a connection relationship with at least one position point.
The device according to claim 17, characterized in that the processing unit is also used for:

Based on the connection relationship between each position point in the second graph structure, an adjacency distance matrix is obtained; wherein each element in the adjacency distance matrix is used to characterize one of the second categories in the second graph structure The number of edges between a location point and one of the first-type location points;

Normalize the adjacency distance matrix to obtain a weight matrix;

Multiply the weight matrix and the first matrix to obtain a second matrix; wherein the elements in the first matrix are used to represent the attributes contained in each of the first type position points in the second graph structure. The first attribute of the second matrix is used to characterize the first attribute included in the attributes of each second type position point in the second graph structure.
The device according to claim 17 or 18, characterized in that,

The first attribute is the coordinates of the first location point or an indoor or outdoor location identifier, and the indoor or outdoor location identifier is used to indicate that the first location point is an indoor location point or an outdoor location point.
The device according to claim 19, characterized in that the processing unit is also used for:

Position positioning, navigation, big data statistical analysis or service recommendation are performed based on the coordinates of each of the plurality of position points.
The device according to any one of claims 12-20, characterized in that:

The quantity distribution characteristics include at least one of normal distribution, Poisson distribution, discrete distribution or interval distribution, and the interval distribution is used to describe the number of characteristic signals located in different intervals.
The device according to any one of claims 12-21, characterized in that,

The characteristic signal includes one or more of GNSS signals, position coordinates, radio frequency signals, optical signals, acoustic signals, sensor signals or geomagnetic signals, where the radio frequency signals include WIFI signals, Bluetooth signals, One or more of cell CELL signals or ultra-wideband UWB signals.
A computer device, characterized in that the computer device includes at least one processor, a memory and an interface circuit, the memory, the interface circuit and the at least one processor are interconnected through lines, and the at least one memory stores There are instructions; when the instructions are executed by the processor, the method of any one of the above claims 1-11 is implemented.
The device according to claim 23, characterized in that the computer device is a server or a terminal device, wherein the terminal device includes a mobile phone, a computer, a car machine or a tablet.
A chip system, the chip system includes at least one processor, a memory and an interface circuit, the memory, the interface circuit and the at least one processor are interconnected through lines, and instructions are stored in the at least one memory; When the instructions are executed by the processor, the method of any one of the above claims 1-11 is implemented.
A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program. When the computer program is executed, any one of the methods in claims 1-11 can be implemented.
A computer program product, characterized in that the computer program product includes program instructions. When the program instructions are run on a computer, the method of any one of the above claims 1-11 is implemented.