CN118215002A - Mobile phone signaling and AOI data-based travel OD accurate identification method - Google Patents
Mobile phone signaling and AOI data-based travel OD accurate identification method Download PDFInfo
- Publication number
- CN118215002A CN118215002A CN202410089980.0A CN202410089980A CN118215002A CN 118215002 A CN118215002 A CN 118215002A CN 202410089980 A CN202410089980 A CN 202410089980A CN 118215002 A CN118215002 A CN 118215002A
- Authority
- CN
- China
- Prior art keywords
- point
- travel
- points
- data
- stay
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 54
- 230000011664 signaling Effects 0.000 title claims abstract description 47
- 238000005070 sampling Methods 0.000 claims abstract description 23
- 230000008569 process Effects 0.000 claims abstract description 17
- 238000012545 processing Methods 0.000 claims abstract description 11
- 230000007613 environmental effect Effects 0.000 claims abstract description 8
- 238000007781 pre-processing Methods 0.000 claims description 5
- 230000002776 aggregation Effects 0.000 claims description 2
- 238000004220 aggregation Methods 0.000 claims description 2
- 208000013715 atelosteogenesis type I Diseases 0.000 claims description 2
- 238000012216 screening Methods 0.000 claims description 2
- 238000012163 sequencing technique Methods 0.000 claims description 2
- 238000007689 inspection Methods 0.000 claims 1
- 230000003287 optical effect Effects 0.000 claims 1
- 230000000875 corresponding effect Effects 0.000 description 7
- 238000010586 diagram Methods 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 2
- 239000006185 dispersion Substances 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 230000002123 temporal effect Effects 0.000 description 2
- 102100029469 WD repeat and HMG-box DNA-binding protein 1 Human genes 0.000 description 1
- 101710097421 WD repeat and HMG-box DNA-binding protein 1 Proteins 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000003542 behavioural effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000011002 quantification Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/02—Services making use of location information
- H04W4/029—Location-based management or tracking services
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/20—Services signaling; Auxiliary data signalling, i.e. transmitting data via a non-traffic channel
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
The invention discloses a travel OD accurate identification method based on mobile phone signaling and AOI data. The method utilizes the ultra-large sample mobile phone signaling data of urban scale, comprehensively utilizes data processing, modeling and geographic informatics methods, combines the characteristics of AOI data, base station distribution, sampling frequency and the like in the mobile phone signaling sampling process, and calculates and obtains accurate trip OD information of a user. The invention combines AOI data as geographic information characteristics, and improves the accuracy and the scientificity of stay point identification. The invention also considers the influence of the base station distribution density, uniformity, sampling frequency and other environmental characteristic variables involved in the mobile phone signaling sampling process, can adaptively adjust the space-time parameters, and is suitable for different places, environments and crowds.
Description
Technical Field
The invention relates to a travel OD accurate identification method based on mobile phone signaling and AOI data, which is used for supporting urban traffic planning and urban traffic travel structure analysis based on big data, and belongs to the field of intelligent traffic and traffic planning.
Background
In recent years, with popularization of smart phones, a resident trip information acquisition method based on mobile phone signaling data has been widely focused because of the advantages of large sample size, low cost, high real-time performance, complete trip space-time coverage and the like. The mobile phone signaling data contains geographic position information, so that travel tracks, travel OD (OD) and the like of residents can be effectively described. And the information can be further applied to analyzing urban population occupancy distribution, identifying travel modes and the like. The above research can describe the current situation of urban resident traveling and traffic demand in a more detailed and real way, and provides support for urban traffic planning and management.
Existing approaches mainly identify coarse-grained OD matrices that aggregate with grids or traffic cells as statistical units. In recent years, a method of analyzing a combination of a cell phone signaling and geographic information data has appeared, and point of interest (Point of Interest, POI) data and the like have been mainly used, but since it is point coordinate data, there is a limitation in recognition accuracy. And AOI (Area of Interest), i.e. an interest surface, is more than POI (Point of Interest), and is mainly used for expressing regional geographic entities in a map, such as a residential district, a school or a scenic spot, etc., which can better describe the geographic information.
Therefore, it is necessary to provide an accurate recognition method for urban resident travel OD based on mobile phone signaling data and combining the AOI data and the environmental variables such as base station density, uniformity, sampling frequency and the like related in the sampling process, so as to realize accurate depiction of resident travel demands and urban traffic structures and provide scientific quantification basis for scientific management decisions.
Disclosure of Invention
The invention aims to provide a travel OD accurate identification method based on mobile phone signaling and AOI data.
The invention utilizes the ultra-large sample mobile phone signaling data of urban scale, comprehensively utilizes data processing, modeling and geographic informatics methods, combines the characteristics of AOI data, base station distribution and sampling frequency in the mobile phone signaling sampling process and the like, and calculates and obtains the accurate trip OD information of the user.
In order to achieve the above object, the travel OD precise identification method based on mobile phone signaling and AOI data provided by the invention comprises the following steps: preprocessing mobile phone signaling data, combining variable parameter sliding window stay point identification such as AOI data and the like, travel segment division, stay point clustering, OD identification and calibration.
The basic steps of the invention are as follows:
s1: pre-processing mobile phone signaling data, including processing drift data, ping-pong data and the like;
S2: combining the AOI data with surrounding base station distribution, sampling frequency and other environmental variables, and carrying out stay point identification based on a variable parameter sliding window;
s3: clustering the stay points to form a 'normally-going place', and dividing travel sections;
S4: the travel OD is identified and calibrated.
Due to the adoption of the technical scheme, the invention has the following beneficial effects:
1. and the accuracy and the scientificity of the stay point identification are improved by combining the AOI data as the geographic information characteristics.
2. The influence of the environmental characteristic variables such as the distribution density, uniformity, sampling frequency and the like of the base stations involved in the mobile phone signaling sampling process is considered, so that the space-time parameters can be adaptively adjusted, and the mobile phone signaling sampling method is suitable for different places, environments and crowds.
3. The complete recognition model from the mobile phone signaling data to the travel chain is constructed, and the accuracy of recognizing the resident travel OD by using the mobile phone signaling data is improved.
Drawings
Figure 1 is an overall flow chart of the inventive solution.
Fig. 2 is a flow chart for stay point identification.
Fig. 3 shows a schematic diagram of a distribution uniformity calculation of a base station.
Fig. 4 illustrates a handset signaling data location profile.
FIG. 5 is a schematic diagram of the distribution of AOI in a main urban area portion of a city.
Fig. 6 illustrates an OD profile identified by the data.
Fig. 7 is a schematic diagram of travel OD and related information of a user.
Detailed Description
The invention will be further described with reference to FIG. 1
The basic steps of the invention are as follows:
s1: pre-processing mobile phone signaling data, including processing drift data, ping-pong data and the like;
S2: combining the AOI data with surrounding base station distribution, sampling frequency and other environmental variables, and carrying out stay point identification based on a variable parameter sliding window;
s3: clustering the stay points to form a 'normally-going place', and dividing travel sections;
S4: the travel OD is identified and calibrated.
The process of step S1 includes:
And processing repeated data, drifting data and ping-pong switching data in the mobile phone signaling data.
The repeated data refers to data that is identical or sampled multiple times at the same time.
The drift data refers to data generated by accidental handover to a base station far away from each other when a signal is connected,
The ping-pong handover data refers to data repeatedly handed over between two or more base stations.
The specific pretreatment process is as follows: and for repeated data, after the signaling data are ordered according to time, only one piece of signaling data of the same user at the same moment is reserved, and the rest is deleted. And for drift data, a distance speed threshold method is adopted to identify the drift data in signaling, and whether the intermediate point is the drift data is judged according to the relative relation between the sampling interval distance and the speed (namely, the sampling interval distance is divided by the sampling interval time) of three continuous track points. Assuming A, B, C as three consecutive track points, if the distance between AB and BC is greater than the threshold and the distance between AC is less than the threshold, then the B point is determined to be drift data, and the B point is eliminated. The same applies to the speed determination method. And for ping-pong switching data, identifying the ping-pong switching data by adopting a rule method and deleting the ping-pong switching data.
The process of step S2 includes:
The corresponding space-time threshold value is obtained by calculation of environmental variables (including surrounding base station distribution, sampling frequency and the like) of each track point, so that the space-time threshold value can be dynamically adjusted, and a sliding window is set by the variable space-time threshold value parameter. And selecting track points which are adjacent in time and space through a sliding window for aggregation, and utilizing the AOI of each track point to assist in judgment, so that individual stay point identification is realized, and the process is shown in figure 2.
S21: and sequencing the signaling track points according to the recording time, and calculating the environment variable of each track point. The spatial threshold sp_dist and the temporal threshold sp_time (variable parameters) corresponding to each point are calculated according to the environment variable.
The environment variables specifically include:
(1) Base station density: refers to the number of base stations (x 1) within 400 meters per locus point square.
(2) Base station direction distribution uniformity: refers to the variance (x 2) of the relative distribution direction of the base stations within 400 meters of the square circle of each locus point.
All base stations within 400 meters of each locus point square are distributed to a unit circle centering on the locus point according to the azimuth thereof with respect to the locus point, as shown in fig. 3 (a). The direction of each base station relative to the locus point can be regarded as a unit vector z, which has an angle q with respect to the x-axis, and the coordinates of the vector z are (cosq, sinq).
The corresponding angle q 1,q2,…,qn is expressed by the unit vector z 1,z2,…,zn of all base stations near a certain locus point, then the average direction of q 1,q2,…,qn That is, the direction indicated by the average vector z of z 1,z2,…,zn, an average vector/>Is/>Then:
Average vector Length/>The method comprises the following steps:
The value of (a) is between 0 and1, which can reflect the degree of dispersion of the data, as shown in fig. 3 (b). Specifically, if q 1,q2,…,qn are all clustered together, then/> Then it is close to 1 and vice versa it is close to 0. /(I)The smaller the data, the more dispersed and the more uniform the distribution.
The uniformity of the base station distribution is measured using x 2 as the variance, so that the dispersion (uniformity) is positively correlated with the variance.
(3) Sampling frequency: refers to the number of samples (x 3,x4) 15min before and 15min after each trace point.
In order to enable the space-time threshold to be dynamically adjusted according to the change of the environmental variable, the present embodiment further constructs a linear relationship:
sp_dist＝a0+a1*x1+a2*x2+a3*x3+a4*x4 (5)
sp_time＝b0+b1*x1+b2*x2+b3*x3+b4*x4 (6)
Wherein the characteristic array is { x 1,x2,x3,x4 } independent variable, the threshold sp_dist (m) and sp_time (min) are dependent variables, { a 0,a1,a2,a3,a4 } and { b 0,b1,b2,b3,b4 } are coefficient parameters to be determined, and the 10 values can be set empirically or can be calibrated additionally.
S22: a sequence of stay points is constructed based on rule recognition.
(1) Checking the time interval between two continuous track points one by one, if the time interval is larger than the maximum stay time threshold value max_time_gap, the time interval between the two track points is too long, the travel state between the two points cannot be judged, the two points are broken, and the stay points cannot be formed.
(2) Sequentially traversing the subsequent track points serving as points x i by taking the initial point as a reference point x s, calculating the distance and time interval between the two points x s and x i, comparing the distance and time interval with a space threshold sp_dist and a time threshold sp_time, and considering the AOI of the two points; if the corresponding condition is met, it is considered that the movement occurred at point x i and updated to be reference point x s (with the current x i point as the new x s point). During traversal, all the trace points between x s and x i (i.e., x s～xi-1) are merged to form a dwell point. The process is then repeated starting with a new reference point x s.
Further, the AOI information of the track point is mainly used to help determine whether a new stay point is formed, and if the track point moves only in the same AOI area, the track point is not considered as a new trip, and still stays in the stay process.
Specifically, even if the time interval and the space interval of x s and x i satisfy the stay point discrimination condition, if the AOI of the two points are the same, all points between x s～xi-1 are not recognized as new stay points although they are in a stay state at this time. In this case, x i needs to continue the backward traversal until the space-time condition is satisfied and the two-point AOI is different, at which point x i-1 is the last track point of the stay state.
In addition, because different AOI areas may overlap, matching corresponding AOI information according to track point coordinates may occur when one point corresponds to multiple AOIs. To solve this problem, the matching principle is to perform one-to-one matching, and to preferentially match to the AOI with larger area and higher level.
(3) When traversing to the last, no new movement occurs, and if the duration meets the requirement, the rest track points are summarized as the last stay point. The algorithm traverses all the individual track points through the steps, and identifies the stay points, so that the individual stay point sequence is constructed.
The process of step S3 includes:
s31: the stay points are clustered to form a 'normally-going place'.
For a specific location where a user stays multiple times, the base station is connected differently when the user stays each time, so that the identified stay point is deviated and is identified as two locations. To avoid this problem, the identified stay points are spatially clustered using the DBSCAN method to form "frequented places". In the algorithm, a neighborhood radius parameter e needs to be set, and the minimum dwell point number of the clusters is set to be 1, so that offset dwell points with multiple dwells are aggregated into one, and single dwell points are reserved.
S32: and dividing the travel sections.
After the stay points are identified, all the track points which are not identified as stay states are classified as travel states, and a path formed by all the travel track points between two continuous stay points is divided into travel sections. To further segment the continuous travel track, short waiting and transfer actions need to be identified. If the time interval of two continuous travel track points is larger than the parameter g, the original travel section is interrupted at the time, and two new travel sections are formed.
The process of step S4 includes:
the row OD is formed and calibrated.
All travel segments between two stay points can be combined into one trip (one OD). The starting and stopping time of the travel is the starting time of the first travel section and the ending time of the last travel section of the travel respectively, and the coordinates of the point O and the point D of the travel are the coordinates of the stop points at the two ends respectively.
And (3) performing calibration: for the travel which is not identified as a stop point due to shorter stop time at a destination in part of real travel, the scene of 'going to a station for receiving and delivering people' is more typical, the starting and ending positions of the identified travel are the same, so that the travel with the same point position O, D is split.
The method comprises the following steps: and screening out travel with the same common travel place of O, D points and at least two travel sections in the middle, and finding out the travel section with the D point farthest from the O point in each travel (the travel section cannot be the last travel section of travel). Finally, splitting one trip into two trips: the first time is from the original O point to the end point of the travel section farthest from the O point, and the second time is from the start point of the next travel section of the travel section farthest from the original O point to the original D point.
Further calibration, for travel with shorter travel duration, a travel time threshold m (units/second) needs to be set, and travel OD with duration smaller than the threshold is deleted. The track points originally belonging to the travel are classified into the previous stay points, and the process is equivalent to indirectly processing some noise data.
Examples:
The invention is further described by taking volunteer mobile phone signaling data collected in a certain city as an example and combining partial AOI and other data of the city, and by applying the invention, the travel OD of the volunteer mobile phone signaling data is accurately identified.
The specific process is as follows:
1. Preprocessing data;
In this case, the mobile phone signaling data of 31 users in a city of 2022 and 12 months for 28 continuous days are selected, and the original data is about 7.3 ten thousand, and the distribution of the original data is shown in fig. 4. Ten thousand of the AOI data of the city are obtained through the hundred-degree map API method, as shown in fig. 5. In addition, there are several pieces of position distribution data of the city part base station.
Processing redundant data, duplicate data, ping-pong data, drift data, etc. The drift data processing sets a distance threshold of 3000 meters and a speed threshold of 120 km/h. And for ping-pong switching data, identifying the ping-pong switching data by adopting a rule method and deleting the ping-pong switching data. After pretreatment, the total amount of signaling data is reduced by about 3%, and about 7.1 ten thousand pieces remain.
2. Combining with AOI data and the like, carrying out stay point identification based on a variable parameter sliding window;
① Sorting the track points according to the recording time, and calculating the environment variable of each track point: base station density x 1, base station directional distribution uniformity x 2, sampling frequency x 3、x4.
② The spatial threshold sp_dist and the temporal threshold sp_time (variable parameters) corresponding to each point are calculated according to the environment variable. The values of the coefficient parameters are shown in table 1.
sp_dist＝a0+a1*x1+a2*x2+a3*x3+a4*x4
sp_time＝b0+b1*x1+b2*x2+b3*x3+b4*x4
Table 1 reference values of coefficient parameters
③ The time interval between two successive trace points is checked one by one, and if it is greater than the maximum dwell time threshold max_time_gap (which can be set to 24 h), the two points are broken and they cannot form a dwell point.
④ The initial point is taken as a reference point x s, the subsequent track points are sequentially traversed as points x i, the distance and the time interval between the two points of x s and x i are calculated, the distance and the time interval are compared with a space threshold sp_dist and a time threshold sp_time, the AOI of the two points is considered, if the corresponding conditions are met, the point x i is considered to move, and the reference point x s is updated (the current point x i is taken as a new point x s). During traversal, all the trace points between x s and x i (i.e., x s～xi-1) are merged to form a dwell point. The process is then repeated starting with a new reference point x s.
⑤ When traversing to the last, no new movement occurs, and if the duration meets the requirement, the rest track points are summarized as the last stay point. Traversing the track points of each individual through the steps, identifying the stay points, and further constructing the stay point sequence of the individual. Based on the 7.1 ten thousand pieces of signaling data after the pretreatment, 3089 stay points are identified in total in this step.
3. Clustering the stay points to form a 'normally-going place', and dividing travel sections;
And performing spatial clustering on the identified stay points by using a DBSCAN method to form a 'frequently-going place'. Setting the neighborhood radius parameter e to be 200 meters, and setting the minimum stay point number of the clusters to be 1. All track points which are not identified as stay states are classified as travel states, and a path consisting of all travel track points between two consecutive stay points is divided into travel segments. Setting a parameter g to be 12.6 minutes, and if the time interval of two continuous travel track points is larger than g, interrupting the original travel section to form two new travel sections.
In this example, 547 "normally-going places" and 2002 travel sections are obtained by the co-processing.
4. Identifying a travel OD and calibrating the travel OD;
All travel segments between two stay points are combined into one trip (one OD). The starting and stopping time of the travel is the starting time of the first travel section and the ending time of the last travel section of the travel respectively, and the coordinates of the point O and the point D of the travel are the coordinates of the stop points at the two ends respectively. And splitting the travel with the same O, D point positions. The travel time threshold m is set to 200 seconds, and the travel OD having a duration less than the threshold is deleted.
Up to this point, all trips OD 1624 for 28 days for 31 users were obtained, the distribution is shown in fig. 6. Further experiments prove that the result is in high agreement with the actual travel situation.
Fig. 7 is a schematic diagram of original signaling data which is complete for a certain user on a certain day and trip OD and related information obtained after model processing, wherein three stay points, two trip sections, two normally-going places and two trip OD are identified in total. Wherein the first and third stay points are clustered into the same constant-going place, and the two travel ODs respectively comprise one travel section. Table 2 shows the user's finished stay and travel OD information. This information can be used to make behavioral portraits.
TABLE 2 stay and travel OD information for a user on a day
The foregoing description of the formulas and examples is merely a general description for the purpose of facilitating the understanding and application of the present invention by those skilled in the art. It will be apparent to those skilled in the art that the formulas can be easily modified without the need for inventive labor, and therefore, all modifications and improvements made within the scope of the claims of the present invention shall fall within the scope of the present invention.
Claims (8)
1. A travel OD accurate identification method based on mobile phone signaling and AOI data is characterized by comprising the following steps:
S1: pre-processing mobile phone signaling data, including processing drift data and ping-pong data;
S2: combining the AOI data and the environment variables, carrying out stay point identification based on the variable parameter sliding window, wherein the stay point identification specifically comprises the following steps:
calculating the environment variable of each track point to obtain a corresponding space-time threshold value, so that the space-time threshold value can be dynamically adjusted;
setting a sliding window with the variable spatiotemporal threshold parameter;
selecting track points which are adjacent in time and space through a sliding window for aggregation, and utilizing AOI (automatic optical inspection) auxiliary judgment of each track point to realize individual stay point identification;
s3: clustering the stay points to form a 'normally-going place', and dividing travel sections, specifically:
performing spatial clustering on the identified stay points by using a DBSCAN method to form a 'frequently-going place';
s32: dividing a travel section;
after the stay points are identified, all the track points which are not identified as stay states are classified as travel states, and a path formed by all travel track points between two continuous stay points is divided into travel sections;
s4: the trip OD is identified, specifically:
All travel sections between the two stay points are combined into one travel, namely one OD; the starting and stopping time of the travel is the starting time of the first travel section and the ending time of the last travel section of the travel respectively, and the coordinates of the point O and the point D of the travel are the coordinates of the stop points at the two ends respectively.
2. The trip OD precise identification method based on mobile phone signaling and AOI data according to claim 1 is characterized by comprising the following steps: the step S1 specifically comprises the following steps:
for repeated data, after the signaling data are ordered according to time, only one piece of signaling data of the same user at the same moment is reserved, and the rest is deleted;
for drift data, a distance speed threshold method is adopted to identify the drift data in signaling, and whether the intermediate point is the drift data is judged according to the relative relation between the sampling interval distances and the speeds of three continuous track points;
And for ping-pong switching data, identifying the ping-pong switching data by adopting a rule method and deleting the ping-pong switching data.
3. The trip OD precise identification method based on mobile phone signaling and AOI data according to claim 1 is characterized by comprising the following steps: the step S2 specifically comprises the following steps:
S21: sequencing the signaling track points according to the recording time, and calculating the environment variable of each track point; calculating a space threshold sp_dist and a time threshold sp_time corresponding to each point according to the environment variable;
The environment variables include:
(1) Base station density: refers to the number x of base stations within 400 meters per track point square circle 1
(2) Base station direction distribution uniformity: the variance x 2 of the relative distribution direction of the base stations in 400 meters of each track point square circle is indicated;
(3) Sampling frequency: refers to the sampling number x 3 in the front 15min and the sampling number x 4 in the back 15min of each track point;
S22: constructing a stay point sequence based on rule identification;
(1) Checking the time interval between two successive track points one by one;
(2) Sequentially traversing the subsequent track points serving as points x i by taking the initial points as reference points x s, calculating the distance and time interval between the two points x s and x i, comparing the distance and time interval with a space threshold sp_dist and a time threshold sp_time, considering the AOI of the two points, and if the corresponding conditions are met, considering that the point x i moves and updating the point as the reference point x s;
(3) When traversing to the last, no new movement occurs, and if the duration meets the requirement, the rest track points are summarized as the last stay point;
traversing all the individual track points through the steps, identifying the stay points, and further constructing an individual stay point sequence.
4. The trip OD precise identification method based on mobile phone signaling and AOI data according to claim 3, wherein the trip OD precise identification method is characterized by comprising the following steps of: to enable the spatiotemporal threshold to be dynamically adjusted as the environmental variable changes, a linear relationship is constructed:
sp_dist＝a0+a1*x1+a2*x2+a3*x3+a4*x4
sp_time＝b0+b1*x1+b2*x2+b3*x3+b4*x4
Where the feature array is { x 1,x2,x3,x4 } argument, the thresholds sp_dist and sp_time are arguments, { a 0,a1,a2,a3,a4 } and { b 0,b1,b2,b3,b4 } are coefficient parameters to be determined.
5. The trip OD precise identification method based on mobile phone signaling and AOI data according to claim 3, wherein the trip OD precise identification method is characterized by comprising the following steps of: the AOI information of the track points is mainly used for helping to judge whether a new stay point is formed or not; if the mobile station moves only in the same AOI area, the mobile station is not considered as a new trip and still stays in the stay process;
specifically, even if the time interval and the space interval of x s and x i satisfy the stay point discrimination condition, if the AOIs of the two points are the same, all points between x s～xi-1 are not identified as new stay points although they are in a stay state; in this case, x i needs to continue the backward traversal until the space-time condition is satisfied and the two-point AOI is different, at which point x i-1 is the last track point of the stay state.
6. The trip OD precise identification method based on mobile phone signaling and AOI data according to claim 1 is characterized by comprising the following steps: in step S32, in order to further segment the continuous travel track, a short waiting and transferring action needs to be identified; if the time interval of two continuous travel track points is larger than the parameter g, the original travel section is interrupted at the time, and two new travel sections are formed.
7. The trip OD precise identification method based on mobile phone signaling and AOI data according to claim 1 is characterized by comprising the following steps: the method further comprises the step of calibrating the identified travel OD, and specifically comprises the following steps:
screening out travel with at least two travel sections in the middle, wherein the common travel places of O, D points are the same;
Finding out a travel section with the D point farthest from the O point in each travel;
Splitting one trip into two trips: the first time is from the original O point to the end point of the travel section farthest from the O point, and the second time is from the start point of the next travel section of the travel section farthest from the original O point to the original D point.
8. The trip OD precise identification method based on mobile phone signaling and AOI data according to claim 7, wherein the trip OD precise identification method is characterized by comprising the following steps: setting a travel time threshold m for travel with shorter travel duration, and deleting a travel OD with the duration smaller than the threshold m; and classifying the track points originally belonging to the travel into the previous stay point.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410089980.0A CN118215002A (en) | 2024-01-22 | 2024-01-22 | Mobile phone signaling and AOI data-based travel OD accurate identification method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410089980.0A CN118215002A (en) | 2024-01-22 | 2024-01-22 | Mobile phone signaling and AOI data-based travel OD accurate identification method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN118215002A true CN118215002A (en) | 2024-06-18 |
Family
ID=91447810
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410089980.0A Pending CN118215002A (en) | 2024-01-22 | 2024-01-22 | Mobile phone signaling and AOI data-based travel OD accurate identification method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118215002A (en) |
-
2024
- 2024-01-22 CN CN202410089980.0A patent/CN118215002A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110047277B (en) | Urban road traffic jam ranking method and system based on signaling data | |
| CN110111575B (en) | Urban traffic flow network analysis method based on complex network theory | |
| CN110012428B (en) | Indoor positioning method based on WiFi | |
| CN112512020B (en) | Traffic state weak signal perception studying and judging method based on multi-source data fusion | |
| Fang et al. | FTPG: A fine-grained traffic prediction method with graph attention network using big trace data | |
| CN108109382A (en) | A kind of congestion points based on composite network, congestion line, the discovery method of congestion regions | |
| CN113709660B (en) | Method for accurately extracting travel path by using mobile phone signaling data | |
| CN108171993B (en) | Highway vehicle speed calculation method based on mobile phone signaling big data | |
| CN108320501A (en) | Public bus network recognition methods based on user mobile phone signaling | |
| CN109729518B (en) | Mobile phone signaling-based urban traffic early peak congestion source identification method | |
| Sohn et al. | Space-based passing time estimation on a freeway using cell phones as traffic probes | |
| CN110149595B (en) | HMM-based heterogeneous network user behavior prediction method | |
| CN103116990A (en) | Traffic speed vehicle-mounted acquisition system and method based on mobile phone switch position | |
| CN110598917B (en) | Destination prediction method, system and storage medium based on path track | |
| US6954650B2 (en) | Directivity characteristics of mobile terminals | |
| Cui et al. | Improved genetic algorithm to optimize the Wi-Fi indoor positioning based on artificial neural network | |
| CN113993172A (en) | Ultra-dense network switching method based on user movement behavior prediction | |
| CN114973657A (en) | Urban traffic noise pollution analysis and evaluation method based on trajectory data | |
| CN112000755A (en) | Regional trip corridor identification method based on mobile phone signaling data | |
| Li et al. | Estimating crowd flow and crowd density from cellular data for mass rapid transit | |
| CN114638434A (en) | Variable bus area station optimization method, device and computer storage medium | |
| Zhou et al. | Method for judging parking status based on yolov2 target detection algorithm | |
| CN118215002A (en) | Mobile phone signaling and AOI data-based travel OD accurate identification method | |
| CN110933601B (en) | Target area determination method, device, equipment and medium | |
| CN116668958A (en) | Urban resident dynamic OD matrix extraction method based on mobile phone signaling data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |