CN108955693B - Road network matching method and system - Google Patents
Road network matching method and system Download PDFInfo
- Publication number
- CN108955693B CN108955693B CN201810872312.XA CN201810872312A CN108955693B CN 108955693 B CN108955693 B CN 108955693B CN 201810872312 A CN201810872312 A CN 201810872312A CN 108955693 B CN108955693 B CN 108955693B
- Authority
- CN
- China
- Prior art keywords
- track
- base station
- section
- matched
- determining
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/20—Instruments for performing navigational calculations
Landscapes
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Automation & Control Theory (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
The invention discloses a road network matching method and a road network matching system. The method comprises the following steps: acquiring a road section mapping relation between a base station and a coverage area; acquiring mobile phone signaling data of a user; cleaning the signaling data of the mobile phone; determining all the stop points of the user; determining a multi-segment track of a user; extracting the characteristics of each section of track of the user; determining a travel mode corresponding to each section of track by adopting a decision tree algorithm; extracting all tracks of the vehicle in a travel mode, and generating a track set to be matched; determining a similar historical track corresponding to each section of track to be matched in the track set to be matched; interpolating the tracks to be matched according to the similar historical tracks to obtain an updated track set to be matched; converting the base station sequence corresponding to each section of track into a section sequence; and for each section of track, determining a matched road section sequence by adopting a hidden Markov model to obtain a road network matching result. The method or the system can improve the coverage of road network matching and improve the road network matching precision.
Description
Technical Field
The invention relates to the field of traffic, in particular to a road network matching method and a road network matching system.
Background
The road network matching is one of key steps of track data processing, generally, the obtained track data has certain deviation with the actual position of a user due to equipment, signals, precision and the like, the obtained track data can be matched into the actual road network through a road network matching algorithm, the road network data of the user is determined, and the follow-up research on problems of urban traffic, population distribution and the like is facilitated. At present, most of research is carried out on road network matching based on vehicle GPS data, such as data uploaded to a server by GPS equipment installed in a taxi and GPS data uploaded by mobile phone software. Although these data can solve certain problems, there are still certain disadvantages, such as data skew, small coverage, etc., and thus the road network matching effect is not ideal.
Disclosure of Invention
The invention aims to provide a road network matching method and a road network matching system, which are used for improving the coverage range of road network matching and improving the accuracy of road network matching.
In order to achieve the purpose, the invention provides the following scheme:
a method of road network matching, the method comprising:
acquiring a road section mapping relation between a base station and a coverage area;
acquiring mobile phone signaling data of a user based on all base stations;
cleaning the mobile phone signaling data to obtain cleaned signaling data;
determining all the stop points of the user according to the cleaned signaling data; the dwell point is a base station coordinate of the user whose dwell time in a signal coverage area of a base station exceeds a dwell threshold;
determining a multi-section track of the user according to all the stop points of the user; each track corresponds to a base station sequence;
extracting the characteristics of each section of track of the user to obtain the track characteristics corresponding to each section of track; the trajectory features include: track length, track radius, number of switching base stations, average speed and travel time period;
determining a travel mode corresponding to each section of track by adopting a decision tree algorithm according to the track characteristics of each section of track; the trip mode includes: pedestrian, non-motor, vehicular, and subway vehicles;
extracting all tracks of the vehicle in a travel mode, and generating a track set to be matched; the track set to be matched comprises a plurality of sections of tracks to be matched;
determining a similar historical track corresponding to each section of track to be matched in the track set to be matched according to parameters of the historical track in a track database;
interpolating the track to be matched according to the similar historical track to obtain an updated track set to be matched;
converting the base station sequence corresponding to each track in the updated track set to be matched into a track sequence according to the road section mapping relation between the base station and the signal coverage area to obtain a road section sequence set, wherein the base station sequence corresponding to each track corresponds to a plurality of road section sequences;
for each section of track in the updated track set to be matched, determining a matched section sequence in the section sequence set by adopting a hidden Markov model;
and sequentially determining the matching road section sequences corresponding to all the tracks in the updated track set to be matched to obtain the road network matching result of the user.
Optionally, the obtaining a road segment mapping relationship between the base station and the signal coverage area further includes:
acquiring a maximum threshold and a minimum threshold;
judging whether the distance between the second base station and the first base station is smaller than a minimum threshold value or not to obtain a first judgment result;
when the first judgment result shows that the distance between the second base station and the first base station is not less than the minimum threshold, judging whether the distance between the third base station and the first base station is less than the minimum threshold;
when the first judgment result shows that the distance between the second base station and the first base station is smaller than the minimum threshold value, combining the second base station into a first base station set; the first base station set comprises a first base station;
determining coordinates of a center position of the first base station set, wherein the latitude of the coordinates of the center position is the average value of the latitudes of all base stations in the first base station set, and the longitude of the coordinates of the center position is the average value of the longitudes of all base stations in the first base station set;
judging whether the distance from each base station in the first base station set to the center position is smaller than a maximum threshold value to obtain a second judgment result;
when the second judgment result indicates that the distance from each base station in the first base station set to the center position is not smaller than the maximum threshold, removing the second base station from the first base station set, and judging whether the distance from a third base station to the first base station is smaller than the minimum threshold;
when the second judgment result shows that the distance from each base station in the first base station set to the center position is smaller than the maximum threshold, judging whether the distance between a third base station and the second base station is smaller than the minimum threshold;
until all base stations are traversed, an updated first base station set is obtained;
sequentially obtaining all base station sets;
determining coordinates of a center position of each base station set;
and determining the coordinate of the central position of each base station set as the updated coordinate of the first base station, and sequentially obtaining the coordinates of all the updated base stations.
Optionally, the obtaining of the mapping relationship between the base station and the road segment in the coverage area specifically includes:
acquiring a plurality of set radius of coverage areas; each coverage area radius is greater than the maximum threshold;
and acquiring a multi-level mapping relation between the base station and the road sections in the coverage area according to the radius of the coverage area, wherein each radius of the coverage area corresponds to one level mapping relation.
Optionally, the cleaning the signaling data of the mobile phone to obtain the cleaned signaling data specifically includes:
filtering invalid data and redundant data; the invalid data comprises data with missing fields and data with failed bill states; the redundant data comprises repeated data and a data sequence which is continuously positioned in the same base station;
filtering ping-pong data; the ping-pong data are: after passing through the first base station, returning the signaling data of the first base station again through other base stations within the set threshold time;
filtering drift data; the drift data is signaling data which sequentially reaches the second base station and the third base station after passing through the first base station, the ratio of the distance between the first base station and the second base station to the time is larger than a set ratio threshold, and the distance between the third base station and the first base station is smaller than a first set distance threshold.
The interrupt signaling data is partitioned.
Optionally, the determining the multi-segment trajectory of the user according to all the stop points of the user specifically includes:
arranging all the stop points of the user according to a time sequence;
and for the ith track, taking the ith stop point as the starting point of the ith track, taking the (i + 1) th stop point as the end point of the ith track, extracting the ith track, and sequentially segmenting all signaling data of the user to obtain all tracks of the user.
Optionally, the determining, according to the parameters of the historical tracks in the track database, a similar historical track corresponding to each section of track to be matched in the set of tracks to be matched specifically includes:
acquiring a track section to be matched in a track to be matched; the track section to be matched is a track section which points from the base station A to the base station B, the base stations A and B are two continuous base stations in the track to be matched, and the distance between the base station A and the base station B is greater than a second set distance threshold;
acquiring historical tracks of other base stations between the base station A and the base station B, wherein the historical tracks successively pass through the base station A and the base station B in the track database;
calculating the ratio of the time from the base station A to the base station B in the historical track to the time from the base station A to the base station B in the track segment to be matched;
judging whether the ratio is not greater than a set ratio threshold value or not to obtain a third judgment result;
when the third judgment result shows that the ratio is not greater than a set ratio threshold, determining the historical track as a similar historical track corresponding to the track to be matched;
and when the third judgment result shows that the ratio is greater than a set ratio threshold, acquiring the next historical track for judgment.
Optionally, the interpolating the track to be matched according to the similar historical track specifically includes:
and inserting the track section from the base station A to the base station B in the similar historical track into the track section to be matched in the corresponding track to be matched to obtain the updated track to be matched.
Optionally, for each track segment in the updated track set to be matched, determining a matching road segment sequence in the road segment sequence set by using a hidden markov model, specifically including:
for the j section of track in the updated track set to be matched, determining a base station sequence corresponding to the j section of track as an observation state of a hidden Markov model;
determining the road section actually passed by the user as a hidden state of a hidden Markov model;
determining the observation state probability of each road section sequence corresponding to the jth section of track by adopting the hidden Markov model;
determining the state transition probability of each road section sequence corresponding to the jth section of track by adopting the hidden Markov model;
determining the initial state probability of each road section sequence corresponding to the jth section of track by adopting the hidden Markov model;
determining the total probability of each road section sequence corresponding to the jth section of track according to the observation state probability, the state transition probability and the initial state probability of each road section sequence corresponding to the jth section of track;
and determining the road section sequence with the maximum overall probability as the matching road section sequence of the jth track.
A system for road network matching, said system comprising:
the mapping relation acquisition module is used for acquiring the mapping relation between the base station and the road section in the coverage area;
the mobile phone signaling data acquisition module is used for acquiring mobile phone signaling data of a user based on all base stations;
the cleaning module is used for cleaning the mobile phone signaling data to obtain cleaned signaling data;
a stop point determining module, configured to determine all stop points of the user according to the cleaned signaling data; the dwell point is a base station coordinate of the user whose dwell time in a signal coverage area of a base station exceeds a dwell threshold;
the track determining module is used for determining a multi-section track of the user according to all the stop points of the user; each track corresponds to a base station sequence;
the characteristic extraction module is used for extracting the characteristics of each section of track of the user to obtain the track characteristics corresponding to each section of track; the trajectory features include: track length, track radius, number of switching base stations, average speed and travel time period;
the travel mode determining module is used for determining a travel mode corresponding to each section of track by adopting a decision tree algorithm according to the track characteristics of each section of track; the trip mode includes: pedestrian, non-motor, vehicular, and subway vehicles;
the track extraction module is used for extracting all tracks of the vehicle in a travel mode and generating a track set to be matched; the track set to be matched comprises a plurality of sections of tracks to be matched;
the similar historical track determining module is used for determining a similar historical track corresponding to each section of track to be matched in the track set to be matched according to the parameters of the historical track in the track database;
the interpolation module is used for interpolating the track to be matched according to the similar historical track to obtain an updated track set to be matched;
the road section sequence conversion module is used for converting the base station sequence corresponding to each section of track in the updated track set to be matched into a road section sequence according to the road section mapping relation between the base station and the signal coverage area to obtain a road section sequence set, wherein the base station sequence corresponding to each section of track corresponds to a plurality of road section sequences;
a matching road section sequence determining module, configured to determine, for each section of track in the updated track set to be matched, a matching road section sequence in the road section sequence set by using a hidden markov model;
and the road network matching result determining module is used for sequentially determining the matching road section sequences corresponding to all the tracks in the updated track set to be matched to obtain the road network matching result of the user.
Optionally, the matching road segment sequence determining module specifically includes:
an observation state determining unit, configured to determine, for a jth track in the updated set of tracks to be matched, a base station sequence corresponding to the jth track as an observation state of a hidden markov model;
a hidden state determining unit, configured to determine a road segment actually passed by the user as a hidden state of a hidden markov model;
an observation state probability determining unit, configured to determine, by using the hidden markov model, an observation state probability of each road segment sequence corresponding to the jth track;
a state transition probability determining unit, configured to determine a state transition probability of each road segment sequence corresponding to the jth segment of track by using the hidden markov model;
an initial state probability determining unit, configured to determine an initial state probability of each road segment sequence corresponding to the jth segment of track by using the hidden markov model;
the overall probability determining unit is used for determining the overall probability of each road section sequence corresponding to the jth section of track according to the observation state probability, the state transition probability and the initial state probability of each road section sequence corresponding to the jth section of track;
and the matching road section sequence determining unit is used for determining the road section sequence with the maximum overall probability as the matching road section sequence of the jth track.
According to the specific embodiment provided by the invention, the invention discloses the following technical effects:
the invention adopts the mobile phone signaling data as the information source of road network matching, thereby enlarging the coverage. The data cleaning method provided by the invention can effectively solve the problems of ping-pong data and drifting data of the mobile phone signaling, and greatly improves the accuracy of mobile phone signaling positioning; a large number of similar paths can be found through the similar path matching algorithm, so that interpolation is carried out on the path with sparse track points, and the accuracy of road network matching is improved; the method uses a Hidden Markov Model (HMM) and a shortest path combination method to carry out road network matching, and by establishing multi-level mapping between a base station and a road and adding two factors of road level and road straight-going degree into the HMM model, experimental results show that the method has good effect of realizing road network matching on mobile phone signaling data with coarse granularity and low frequency, and has positive significance on urban road planning and traffic management.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without inventive exercise.
FIG. 1 is a flow chart of a road network matching method according to the present invention;
FIG. 2 is a schematic diagram of a road network matching system according to the present invention;
FIG. 3 is a schematic diagram of a raw base station trajectory in an embodiment of the present invention;
FIG. 4 is a trace after data cleansing and similar trace interpolation processing in accordance with an embodiment of the present invention;
fig. 5 shows a final road network matching result in the embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in further detail below.
FIG. 1 is a flow chart of a road network matching method according to the present invention. As shown in fig. 1, the method includes:
step 101: and acquiring a road section mapping relation between the base station and the coverage area.
Before determining the mapping relationship, the base stations with close distances need to be combined first. The method comprises the following specific steps:
acquiring a set maximum threshold and a set minimum threshold;
judging whether the distance between the second base station and the first base station is smaller than a minimum threshold value or not to obtain a first judgment result;
when the first judgment result shows that the distance between the second base station and the first base station is not less than the minimum threshold, judging whether the distance between the third base station and the first base station is less than the minimum threshold; the third base station is any other base station different from the first base station and the second base station;
when the first judgment result shows that the distance between the second base station and the first base station is smaller than the minimum threshold value, combining the second base station into a first base station set; the first base station set comprises a first base station;
determining coordinates of a center position of the first base station set, wherein the latitude of the coordinates of the center position is the average value of the latitudes of all base stations in the first base station set, and the longitude of the coordinates of the center position is the average value of the longitudes of all base stations in the first base station set;
judging whether the distance from each base station in the first base station set to the center position is smaller than a maximum threshold value to obtain a second judgment result;
when the second judgment result indicates that the distance from each base station in the first base station set to the center position is not smaller than the maximum threshold, removing the second base station from the first base station set, and judging whether the distance from a third base station to the first base station is smaller than the minimum threshold;
when the second judgment result shows that the distance from each base station in the first base station set to the center position is smaller than the maximum threshold, judging whether the distance between a third base station and the second base station is smaller than the minimum threshold;
until all base stations are traversed, an updated first base station set is obtained;
sequentially obtaining all base station sets;
determining coordinates of a center position of each base station set;
and determining the coordinate of the central position of each base station set as the updated coordinate of the first base station, and sequentially obtaining the coordinates of all the updated base stations.
And combining the closer base stations to obtain new base stations corresponding to all the base station sets, and taking the new base stations as the basis of subsequent signaling data processing.
When determining the mapping relation between a base station and a road section in a coverage area, firstly, acquiring a plurality of set radius of the coverage area; each coverage area radius is greater than the maximum threshold; and acquiring a multi-level mapping relation between the base station and the road sections in the coverage area according to the radius of the coverage area, wherein each radius of the coverage area corresponds to one level mapping relation.
Step 102: and acquiring mobile phone signaling data of the user based on all base stations.
Each city is distributed with a plurality of mobile phone signal base stations, each base station can record mobile phone user information connected with the base station, and each time a mobile phone user moves from the coverage area of one base station to the coverage area of another base station, a record is generated. At present, with the popularization of smart phones, basically everyone has one smart phone, so that mobile phone signaling data has high coverage rate close to a full sample. Each base station has accurate longitude and latitude information, and the mobile phone signaling data can be uniquely associated with one base station through a large area number (LAC) and a cell number (CI), so that the mobile track of each user can be extracted from the signaling data, and each mobile track is represented by a series of base stations. Compared with the common track data, the mobile phone signaling data has the advantages of wide user number and high coverage rate, and the data is acquired in a passive mode without being actively uploaded by the user, so that the hotspot problems of urban traffic, population distribution and the like can be better researched from the angle close to a full sample through the mobile phone signaling data.
Step 103: and cleaning the mobile phone signaling data to obtain the cleaned signaling data. The method for cleaning the mobile phone signaling data comprises the following steps:
1. filtering invalid data and redundant data: defining data with missing fields and data with failed bill state as invalid data; repeated data and sequences successively located in the same base station are regarded as redundant data. Invalid data and redundant data are deleted.
2. Filtering ping-pong data: setting a time threshold Tp, traversing a track of a user, if a track passes through the base station A and returns to the point A again after passing through a plurality of other base stations within the threshold time Tp, regarding the track as a section of ping-pong switching data, and deleting the middle base station sequence.
3. And (3) filtering drift data: sometimes, the base station connected by the user is suddenly switched from a to a base station B which is far away due to reflection and refraction, etc., the processing method is to set a threshold value V, when the ratio of the distance between A, B two points to the time exceeds the threshold value V, and the next position returns to the position which is close to a, then the next position is regarded as drift data, and the point B is deleted.
4. Splitting interrupt signaling data: due to poor mobile phone signals or shutdown of a user and other reasons, a phenomenon that data are seriously lost for a long time in the middle of a part of tracks can be generated, so that two adjacent base stations are far away from each other and cannot be subjected to road network matching, therefore, a distance threshold value D is set, if the distance between two adjacent base stations in one track exceeds the distance threshold value D, the track is considered to be interrupted, and the interrupted track is divided into two tracks.
Step 104: and determining all the stop points of the user according to the cleaned signaling data. The dwell point is the base station coordinate of the dwell time of the user in the signal coverage area of one base station exceeding the dwell threshold.
Step 105: and determining the multi-section track of the user according to all the stop points of the user. Each track corresponds to a base station sequence. The method comprises the following specific steps:
arranging all the stop points of the user according to a time sequence;
and for the ith track, taking the ith stop point as the starting point of the ith track, taking the (i + 1) th stop point as the end point of the ith track, extracting the ith track, and sequentially segmenting all signaling data of the user to obtain all tracks of the user.
Step 106: and extracting the characteristics of each section of track of the user to obtain the track characteristics corresponding to each section of track. The trajectory features include: track length, track radius, number of switching base stations, average speed and travel time period.
Step 107: and determining the travel mode corresponding to each section of track by adopting a decision tree algorithm according to the track characteristics of each section of track. The trip mode includes: pedestrian, non-motor, vehicular, and subway cars.
Step 108: and extracting all tracks of the vehicle in the travel mode to generate a track set to be matched. The track set to be matched comprises a plurality of sections of tracks to be matched.
Step 109: and determining a similar historical track corresponding to each section of track to be matched in the track set to be matched according to the parameters of the historical tracks in the track database. The method specifically comprises the following steps:
acquiring a track section to be matched in a track to be matched; the track section to be matched is a track section which points from the base station A to the base station B, the base stations A and B are two continuous base stations in the track to be matched, and the distance between the base station A and the base station B is greater than a second set distance threshold;
acquiring historical tracks of other base stations between the base station A and the base station B, wherein the historical tracks successively pass through the base station A and the base station B in the track database;
calculating the ratio of the time from the base station A to the base station B in the historical track to the time from the base station A to the base station B in the track segment to be matched;
judging whether the ratio is not greater than a set ratio threshold value or not to obtain a third judgment result;
when the third judgment result shows that the ratio is not greater than a set ratio threshold, determining the historical track as a similar historical track corresponding to the track to be matched;
and when the third judgment result shows that the ratio is greater than a set ratio threshold, acquiring the next historical track for judgment.
Step 1010: and interpolating the track to be matched according to the similar historical track to obtain an updated track set to be matched. Specifically, a track segment from the base station a to the base station B in the similar historical track is inserted into a track segment to be matched in the corresponding track to be matched, so as to obtain an updated track to be matched.
Step 1011: and converting the base station sequence corresponding to each track in the updated track set to be matched into a track sequence according to the road section mapping relation between the base station and the signal coverage area, so as to obtain a road section sequence set. The base station sequence corresponding to each section of track corresponds to a plurality of section sequences. And (4) converting according to the base station-road network mapping relation established in the step (101), and selecting the mapping relation with different grades according to the distance between adjacent base stations during conversion. Finally, a base station sequence is converted into a sequence of road section sets.
Step 1012: and for each section of track in the updated track set to be matched, determining a matched section sequence in the section sequence set by adopting a hidden Markov model. The hidden markov model mainly comprises 5 elements: for the j-th track in the updated track set to be matched, the process of determining the sequence of the matched road section specifically comprises the following steps:
1. observation state O: and determining the base station sequence corresponding to the jth section of track as an observation state of a hidden Markov model.
2. Hidden state S: and determining the road section actually passed by the user as the hidden state of the hidden Markov model.
3. Observation state probability B: and determining the observation state probability of each road section sequence corresponding to the j section track by adopting the hidden Markov model. The distance between the base station and the nearest point of the link is gaussian distribution as the probability of the observation state, and the observation probability is calculated by using a gaussian distribution model with mu equal to 0 and sigma equal to 1 (or the maximum distance from the road set to the base station). The observation probability calculation formula is as follows:
wherein, bj(k) Represents the kth base station okThe jth road section s in the corresponding road section setjA matching probability value; ot、stRespectively representing a base station connected with a user at the time t and a road section to be matched; | ot-sjThe | | represents the shortest distance between the base station and the road section; and sigma is 1 when the maximum distance between the base station and all the road sections in the road section set is equal to 0, otherwise, the maximum distance is taken.
4. State transition probability a: and determining the state transition probability of each road section sequence corresponding to the j section track by adopting the hidden Markov model. Most of the existing research and invention aims at the sampled GPS data, and because of the particularity of the mobile phone signaling data, the interval between two adjacent base stations is often far, and the corresponding road sections can not be directly connected. Therefore, when the probability transition matrix is constructed, the shortest path of two road sections is acquired firstly. Mainly considering the directional and weighted shortest path length factors.
The weighted shortest path adopts Dijkstra algorithm, and the weight calculation formula is as follows:
wherein D (m → n) represents the weight of the road section m to n; direct (m → n) is 1 when a turn occurs, otherwise it is 0; level (m → n) is 1 when the road grade is changed, and is not changed to 0; omega1，ω2The weights of the first two; lnIs the length of the section n. Obtaining a section s by Dijkstrai→sjThe shortest path L of (a).
Calculating s after obtaining shortest pathi→sjThe transition probability is determined mainly by the factors of angle, shortest path length, turning times and road grade switching times, and the specific calculation formula is as follows:
aij＝P(st+1＝sj|st＝si)＝Fθ(si→sj)*Fd(si→sj)*Fc(si→sj)
wherein, the base station corresponds to the road section s at the time tiThe corresponding road section of the base station at the time of t +1 is sj，aijDenotes siTo sjThe transition probability of (2); fθ(si→sj) Representing cosine similarity; fd(si→sj) Representing the ratio of the distance between two adjacent base stations to the weighted shortest path; fc(si→sj) And a normal distribution showing the number of turns and the number of road rank switching.
5. Initial state probability pi: and determining the initial state probability of each road section sequence corresponding to the j section track by adopting the hidden Markov model, wherein the probability of the road section corresponding to the first base station is represented by using the observation state probability of the base station.
And finally, determining the overall probability of each road section sequence corresponding to the jth section of track according to the observation state probability, the state transition probability and the initial state probability of each road section sequence corresponding to the jth section of track, for example, multiplying the observation state probability, the state transition probability and the initial state probability of each road section sequence to obtain the overall probability. And determining the road section sequence with the maximum overall probability as a matching road section sequence of the jth track, namely converting each base station into a mapping road section and weighted shortest path road section sequence, completing the matching of all the base stations, and finally forming a continuous road section sequence to obtain a final matching result.
Step 1013: and sequentially determining the matching road section sequences corresponding to all the tracks in the updated track set to be matched to obtain the road network matching result of the user.
FIG. 2 is a schematic structural diagram of a road network matching system according to the present invention. As shown in fig. 2, the system includes:
a mapping relation obtaining module 201, configured to obtain a mapping relation between a base station and a road segment in a coverage area;
a mobile phone signaling data obtaining module 202, configured to obtain mobile phone signaling data of a user based on all base stations;
a cleaning module 203, configured to clean the mobile phone signaling data to obtain cleaned signaling data;
a stop point determining module 204, configured to determine all stop points of the user according to the cleaned signaling data; the dwell point is a base station coordinate of the user whose dwell time in a signal coverage area of a base station exceeds a dwell threshold;
a track determining module 205, configured to determine a multi-segment track of the user according to all the stop points of the user; each track corresponds to a base station sequence;
a feature extraction module 206, configured to perform feature extraction on each segment of track of the user to obtain a track feature corresponding to each segment of track; the trajectory features include: track length, track radius, number of switching base stations, average speed and travel time period;
a travel mode determining module 207, configured to determine, according to the trajectory characteristics of each section of trajectory, a travel mode corresponding to each section of trajectory by using a decision tree algorithm; the trip mode includes: pedestrian, non-motor, vehicular, and subway vehicles;
the trajectory extraction module 208 is configured to extract all trajectories of the vehicle in the travel mode, and generate a set of trajectories to be matched; the track set to be matched comprises a plurality of sections of tracks to be matched;
a similar historical track determining module 209, configured to determine, according to parameters of historical tracks in a track database, a similar historical track corresponding to each section of track to be matched in the set of tracks to be matched;
an interpolation module 2010, configured to interpolate the to-be-matched trajectory according to the similar historical trajectory to obtain an updated to-be-matched trajectory set;
a road section sequence conversion module 2011, configured to convert, according to a road section mapping relationship between the base station and a road section in a signal coverage area, a base station sequence corresponding to each section of track in the updated set of tracks to be matched into a road section sequence, so as to obtain a road section sequence set, where the base station sequence corresponding to each section of track corresponds to multiple road section sequences;
a matching road section sequence determining module 2012, configured to determine, for each section of track in the updated set of tracks to be matched, a matching road section sequence in the road section sequence set by using a hidden markov model;
and the road network matching result determining module 2013 is configured to sequentially determine matching road segment sequences corresponding to all the tracks in the updated set of tracks to be matched, so as to obtain a road network matching result of the user.
The matching section sequence determining module 2012 specifically includes:
an observation state determining unit, configured to determine, for a jth track in the updated set of tracks to be matched, a base station sequence corresponding to the jth track as an observation state of a hidden markov model;
a hidden state determining unit, configured to determine a road segment actually passed by the user as a hidden state of a hidden markov model;
an observation state probability determining unit, configured to determine, by using the hidden markov model, an observation state probability of each road segment sequence corresponding to the jth track;
a state transition probability determining unit, configured to determine a state transition probability of each road segment sequence corresponding to the jth segment of track by using the hidden markov model;
an initial state probability determining unit, configured to determine an initial state probability of each road segment sequence corresponding to the jth segment of track by using the hidden markov model;
the overall probability determining unit is used for determining the overall probability of each road section sequence corresponding to the jth section of track according to the observation state probability, the state transition probability and the initial state probability of each road section sequence corresponding to the jth section of track;
and the matching road section sequence determining unit is used for determining the road section sequence with the maximum overall probability as the matching road section sequence of the jth track.
The following gives the implementation of one embodiment of the invention:
using desensitized changchun city base station data and cell phone signaling data provided by jilin province division, unicom system integration, ltd, the base station data includes unique identifications BTS _ ID, LAC (large cell number), CI (cell number), longitude and latitude of the base station, as shown in table 1. The cell phone signaling data includes a user ID, time of connecting to the base station, LAC (large cell number), CI (cell number), and event type, as shown in table 2. The two tables are related through the LAC and the CI, each signaling uniquely determines the connected base station through the LAC and the CI, a record is generated when a mobile phone user performs position updating, power on and off or calls are made, and the records of one user are arranged according to the time sequence to form a travel track of one person.
TABLE 1 base station data sample
| JID | LAC | CI | Longitude (G) | |
| 12679 | 45831 | 10231 | 125.3330315 | 43.876557 |
| 12677 | 45831 | 24932 | 125.33183 | 43.8784815 |
| 10292 | 37128 | 63563 | 125.3306003 | 43.87540033 |
Table 2 sample cell phone signaling data
| ID | Time stamp | LAC | CI | Event type |
| 725E781255570F39642906102C81E70B213 | 20170705025954 | 45831 | 10231 | Location update |
| C09A4DECDCFDCAF15DC969AB79F01B | 20170705030028 | 45831 | 24932 | On-off machine |
| 624DFC89E314632CE52AFA0C6A738E20 | 20170705025956 | 37128 | 63563 | Calling |
The first step is as follows: base station data processing and mobile phone signaling data cleaning
Base station data processing
1. Setting a maximum threshold value of 500 meters and a minimum threshold value of 100 meters, combining base stations of which the base station distances are less than 100 meters and the distances between each base station and the center position of the base station set do not exceed 500 meters, and taking the center position of the base station set as the position coordinates of the combined base stations.
2. And respectively taking 100 meters and 300 meters as radiuses to establish a multi-level mapping relation between the base station and the road section covered in the circular area.
(II) Signaling data scrubbing
1. Filtering invalid data and redundant data: deleting the field missing, the bill state failing, the repeated data and the data continuously in the same base station.
2. Filtering ping-pong data: setting the time threshold value to be 10 minutes, and deleting the sequence which returns to the same base station again after passing through a plurality of other base stations within the threshold time.
3. And (3) filtering drift data: a threshold value 120 is set, and when the ratio of the distance between the two points A, B to the time exceeds the threshold value and the next position returns to the position closer to the distance A, the point B is regarded as drift data and deleted.
4. Dividing an interruption track: and setting a distance threshold value of 1000 meters, if the distance between two adjacent base stations in one section of track exceeds the distance threshold value, considering that the track is interrupted, and dividing the interrupted section of track into two sections of tracks.
The second step is that: user O, D point extraction and trajectory feature extraction
Point 1, point O, D extraction: when the residence time of a user in the coverage range of a base station exceeds 1 hour, the user is considered to be a stop point, the first stop point is found and recorded as a first O point, then more than one stop point of the next stop point is found to be O points, the current stop point is taken as D points, and a series of O, D pairs are finally obtained until all the stop points are extracted.
2. Track segmentation: and dividing the travel track of each user according to the extracted O, D, and dividing the travel track of each user for one day into a plurality of sections of tracks.
3. Extracting track characteristics: extracting features for each segment of the track, including: track length, track radius, number of switching base stations, average speed, travel time period, etc.
The third step: travel mode is judged and vehicle travel track is extracted
1. Judging a track trip mode: according to the extracted track characteristics, the travel mode of each section of track is judged through a decision tree algorithm, and the method mainly comprises the following steps: pedestrian, non-motor vehicle, subway.
2. Extracting a vehicle track: the invention only extracts the travel track of the vehicle to carry out road network matching.
The fourth step: and establishing a track database to realize similar track matching.
1. Establishing a track database:
a database is established, and the trajectory table and the base station passing representation data are shown in tables 3 and 4:
TABLE 3 track table structure
Table 4 table structure via base station
| PID | JID | GID | TIME |
| 1 | 10009 | 2 | 20170705082344 |
| 2 | 10191 | 2 | 20170705083118 |
| 3 | 10183 | 2 | 20170705084305 |
2. And matching similar tracks, and performing interpolation according to the historical track and the simultaneous track.
Similar trajectory matching is performed using trajectory 1 as an example, where the distance between base stations 10009 and 10183 exceeds 100 meters, then
Similar track selection is performed from the database:
(1) firstly, base stations 10009 and 10183 are selected to pass through, and the track of other base stations exists between the two points;
(2) and the time from A to B is not more than (1 +/-0.5) times of the time from A to B of the track to be matched.
(3) The final matching track is 10009 → 10191 → 10183, and 10191 is inserted into track 1.
3. Converting the base station sequence into a road section set sequence: and establishing a base station-road network mapping relation for the base station sequence after the interpolation is completed, and converting. Fig. 3 and 4 are schematic diagrams of the original base station trajectory in the embodiment of the present invention, and fig. 4 is a trajectory after data cleaning and similar trajectory interpolation processing in the embodiment of the present invention. Finally, a base station ID sequence is converted into a sequence of a road section ID set.
12679→{2621,2622,2623,2624,2625,2627,4874,6142,6143,6153,6154,7084,10635,10636}
10275→{24867,4868,4869,6010,6011,6012,6039,6040,13740,15896,17768}
10009→{3394,3395,3396,5629,6019,9928,11964,18748,18970,19946}
10191→{4159,8940,12062,16648,16662,18985}
10183→{6056,6066,6067,6717,8564,8565,8566,9175,11919,12095,16648,19516}
10178→{7247,7582,8929,8932,8933,8934,8935,9042,9043,9044,18659,18660,18663}
10170→{1654,1676,5923,6508,7354,8539,8673,9019,9020,9026,15237,15241,16139}
The fifth step: and (3) establishing a hidden Markov (HMM) model by combining the weighted shortest path, and carrying out road network matching on the vehicle travel track:
1. observation state O: and (4) the base station sequence to be matched is 12679,10275,10009,10191,10183,10178,10170.
2. Hidden state S: the road section actually passed by the mobile phone user is in a hidden state.
3. Observation state probability B:
calculating the maximum distance max (| | o) between each base station and all the road sections in the mapping sett-sj| |), the observation probability calculation formula is as follows:
4. probability of state transition
Calculating the transition probability by using a weighted shortest path Dijkstra algorithm, wherein a weight calculation formula is as follows:
calculating s after obtaining shortest pathi→sjThe transition probability is determined mainly by the factors of angle, shortest path length, turning times and road grade switching times, and the specific calculation formula is as follows:
aij＝P(st+1＝sj|st＝si)＝Fθ(si→sj)*Fd(si→sj)*Fc(si→sj)
wherein, the base station corresponds to the road section s at the time tiAnd the section corresponding to the base station at the time of t +1Is s isj，aijDenotes siTo sjThe transition probability of (2); fθ(si→sj) Representing cosine similarity; fd(si→sj) Representing the ratio of the distance between two adjacent base stations to the weighted shortest path; fc(si→sj) And a normal distribution showing the number of turns and the number of road rank switching.
5. Initial state probability pi: the probability of the road section corresponding to the first base station is represented by the probability of the observation state of the base station.
And (3) final matching result: the path with the highest probability is selected as the final matching result, as shown in fig. 5, fig. 5 is the final road network matching result in the embodiment of the present invention.
The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. For the system disclosed by the embodiment, the description is relatively simple because the system corresponds to the method disclosed by the embodiment, and the relevant points can be referred to the method part for description.
The principles and embodiments of the present invention have been described herein using specific examples, which are provided only to help understand the method and the core concept of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, the specific embodiments and the application range may be changed. In view of the above, the present disclosure should not be construed as limiting the invention.
Claims (10)
1. A method of road network matching, said method comprising:
acquiring a road section mapping relation between a base station and a signal coverage area;
acquiring mobile phone signaling data of a user based on all base stations;
cleaning the mobile phone signaling data to obtain cleaned signaling data;
determining all the stop points of the user according to the cleaned signaling data; the dwell point is a base station coordinate of the user whose dwell time in a signal coverage area of a base station exceeds a dwell threshold;
determining a multi-section track of the user according to all the stop points of the user; each track corresponds to a base station sequence;
extracting the characteristics of each section of track of the user to obtain the track characteristics corresponding to each section of track; the trajectory features include: track length, track radius, number of switching base stations, average speed and travel time period;
determining a travel mode corresponding to each section of track by adopting a decision tree algorithm according to the track characteristics of each section of track; the trip mode includes: pedestrian, non-motor, vehicular, and subway vehicles;
extracting all tracks of the vehicle in a travel mode, and generating a track set to be matched; the track set to be matched comprises a plurality of sections of tracks to be matched;
determining a similar historical track corresponding to each section of track to be matched in the track set to be matched according to parameters of the historical track in a track database;
interpolating the track to be matched according to the similar historical track to obtain an updated track set to be matched;
converting the base station sequence corresponding to each track in the updated track set to be matched into a track sequence according to the road section mapping relation between the base station and the signal coverage area to obtain a road section sequence set, wherein the base station sequence corresponding to each track corresponds to a plurality of road section sequences;
for each section of track in the updated track set to be matched, determining a matched section sequence in the section sequence set by adopting a hidden Markov model;
and sequentially determining the matching road section sequences corresponding to all the tracks in the updated track set to be matched to obtain the road network matching result of the user.
2. The method of claim 1, wherein obtaining the mapping relationship between the base stations and the road segments in the signal coverage area further comprises:
acquiring a maximum threshold and a minimum threshold;
judging whether the distance between the second base station and the first base station is smaller than a minimum threshold value or not to obtain a first judgment result;
when the first judgment result shows that the distance between the second base station and the first base station is not less than the minimum threshold, judging whether the distance between the third base station and the first base station is less than the minimum threshold;
when the first judgment result shows that the distance between the second base station and the first base station is smaller than the minimum threshold value, combining the second base station into a first base station set; the first base station set comprises a first base station;
determining coordinates of a center position of the first base station set, wherein the latitude of the coordinates of the center position is the average value of the latitudes of all base stations in the first base station set, and the longitude of the coordinates of the center position is the average value of the longitudes of all base stations in the first base station set;
judging whether the distance from each base station in the first base station set to the center position is smaller than a maximum threshold value to obtain a second judgment result;
when the second judgment result indicates that the distance from each base station in the first base station set to the center position is not smaller than the maximum threshold, removing the second base station from the first base station set, and judging whether the distance from a third base station to the first base station is smaller than the minimum threshold;
when the second judgment result shows that the distance from each base station in the first base station set to the center position is smaller than the maximum threshold, judging whether the distance between a third base station and the second base station is smaller than the minimum threshold;
until all base stations are traversed, an updated first base station set is obtained;
sequentially obtaining all base station sets;
determining coordinates of a center position of each base station set;
and determining the coordinate of the central position of each base station set as the updated coordinate of the first base station, and sequentially obtaining the coordinates of all the updated base stations.
3. The method according to claim 2, wherein the obtaining the mapping relationship between the base station and the road segment in the coverage area specifically comprises:
acquiring a plurality of set radius of coverage areas; each coverage area radius is greater than the maximum threshold;
and acquiring a multi-level mapping relation between the base station and the road sections in the coverage area according to the radius of the coverage area, wherein each radius of the coverage area corresponds to one level mapping relation.
4. The method according to claim 1, wherein the cleaning the signaling data of the mobile phone to obtain the cleaned signaling data specifically comprises:
filtering invalid data and redundant data; the invalid data comprises data with missing fields and data with failed bill states; the redundant data comprises repeated data and a data sequence which is continuously positioned in the same base station;
filtering ping-pong data; the ping-pong data are: after passing through the first base station, returning the signaling data of the first base station again through other base stations within the set threshold time;
filtering drift data; the drift data is signaling data which sequentially reaches the second base station and the third base station after passing through the first base station, the ratio of the distance between the first base station and the second base station to the time is larger than a set ratio threshold, and the distance between the third base station and the first base station is smaller than a first set distance threshold;
the interrupt signaling data is partitioned.
5. The method according to claim 1, wherein the determining the multi-segment trajectory of the user according to all the stop points of the user specifically comprises:
arranging all the stop points of the user according to a time sequence;
and for the ith track, taking the ith stop point as the starting point of the ith track, taking the (i + 1) th stop point as the end point of the ith track, extracting the ith track, and sequentially segmenting all signaling data of the user to obtain all tracks of the user.
6. The method according to claim 1, wherein the determining, according to the parameters of the historical trajectories in the trajectory database, the similar historical trajectory corresponding to each section of the trajectory to be matched in the set of trajectories to be matched specifically includes:
acquiring a track section to be matched in a track to be matched; the track section to be matched is a track section which points from the base station A to the base station B, the base stations A and B are two continuous base stations in the track to be matched, and the distance between the base station A and the base station B is greater than a second set distance threshold;
acquiring historical tracks of other base stations between the base station A and the base station B, wherein the historical tracks successively pass through the base station A and the base station B in the track database;
calculating the ratio of the time from the base station A to the base station B in the historical track to the time from the base station A to the base station B in the track segment to be matched;
judging whether the ratio is not greater than a set ratio threshold value or not to obtain a third judgment result;
when the third judgment result shows that the ratio is not greater than a set ratio threshold, determining the historical track as a similar historical track corresponding to the track to be matched;
and when the third judgment result shows that the ratio is greater than a set ratio threshold, acquiring the next historical track for judgment.
7. The method according to claim 6, wherein the interpolating the trajectory to be matched according to the similar historical trajectory to obtain an updated trajectory set to be matched specifically comprises:
and inserting the track section from the base station A to the base station B in the similar historical track into the track section to be matched in the corresponding track to be matched to obtain the updated track to be matched.
8. The method according to claim 1, wherein the determining, by using a hidden markov model, a matching road segment sequence in the road segment sequence set for each section of track in the updated set of tracks to be matched specifically includes:
for the j section of track in the updated track set to be matched, determining a base station sequence corresponding to the j section of track as an observation state of a hidden Markov model;
determining the road section actually passed by the user as a hidden state of a hidden Markov model;
determining the observation state probability of each road section sequence corresponding to the jth section of track by adopting the hidden Markov model;
determining the state transition probability of each road section sequence corresponding to the jth section of track by adopting the hidden Markov model;
determining the initial state probability of each road section sequence corresponding to the jth section of track by adopting the hidden Markov model;
determining the total probability of each road section sequence corresponding to the jth section of track according to the observation state probability, the state transition probability and the initial state probability of each road section sequence corresponding to the jth section of track;
and determining the road section sequence with the maximum overall probability as the matching road section sequence of the jth track.
9. A system for road network matching, said system comprising:
the mapping relation acquisition module is used for acquiring the mapping relation between the base station and the road section in the signal coverage area;
the mobile phone signaling data acquisition module is used for acquiring mobile phone signaling data of a user based on all base stations;
the cleaning module is used for cleaning the mobile phone signaling data to obtain cleaned signaling data;
a stop point determining module, configured to determine all stop points of the user according to the cleaned signaling data; the dwell point is a base station coordinate of the user whose dwell time in a signal coverage area of a base station exceeds a dwell threshold;
the track determining module is used for determining a multi-section track of the user according to all the stop points of the user; each track corresponds to a base station sequence;
the characteristic extraction module is used for extracting the characteristics of each section of track of the user to obtain the track characteristics corresponding to each section of track; the trajectory features include: track length, track radius, number of switching base stations, average speed and travel time period;
the travel mode determining module is used for determining a travel mode corresponding to each section of track by adopting a decision tree algorithm according to the track characteristics of each section of track; the trip mode includes: pedestrian, non-motor, vehicular, and subway vehicles;
the track extraction module is used for extracting all tracks of the vehicle in a travel mode and generating a track set to be matched; the track set to be matched comprises a plurality of sections of tracks to be matched;
the similar historical track determining module is used for determining a similar historical track corresponding to each section of track to be matched in the track set to be matched according to the parameters of the historical track in the track database;
the interpolation module is used for interpolating the track to be matched according to the similar historical track to obtain an updated track set to be matched;
the road section sequence conversion module is used for converting the base station sequence corresponding to each section of track in the updated track set to be matched into a road section sequence according to the road section mapping relation between the base station and the signal coverage area to obtain a road section sequence set, wherein the base station sequence corresponding to each section of track corresponds to a plurality of road section sequences;
a matching road section sequence determining module, configured to determine, for each section of track in the updated track set to be matched, a matching road section sequence in the road section sequence set by using a hidden markov model;
and the road network matching result determining module is used for sequentially determining the matching road section sequences corresponding to all the tracks in the updated track set to be matched to obtain the road network matching result of the user.
10. The system according to claim 9, wherein the matching section sequence determining module specifically includes:
an observation state determining unit, configured to determine, for a jth track in the updated set of tracks to be matched, a base station sequence corresponding to the jth track as an observation state of a hidden markov model;
a hidden state determining unit, configured to determine a road segment actually passed by the user as a hidden state of a hidden markov model;
an observation state probability determining unit, configured to determine, by using the hidden markov model, an observation state probability of each road segment sequence corresponding to the jth track;
a state transition probability determining unit, configured to determine a state transition probability of each road segment sequence corresponding to the jth segment of track by using the hidden markov model;
an initial state probability determining unit, configured to determine an initial state probability of each road segment sequence corresponding to the jth segment of track by using the hidden markov model;
the overall probability determining unit is used for determining the overall probability of each road section sequence corresponding to the jth section of track according to the observation state probability, the state transition probability and the initial state probability of each road section sequence corresponding to the jth section of track;
and the matching road section sequence determining unit is used for determining the road section sequence with the maximum overall probability as the matching road section sequence of the jth track.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810872312.XA CN108955693B (en) | 2018-08-02 | 2018-08-02 | Road network matching method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810872312.XA CN108955693B (en) | 2018-08-02 | 2018-08-02 | Road network matching method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108955693A CN108955693A (en) | 2018-12-07 |
| CN108955693B true CN108955693B (en) | 2020-06-16 |
Family
ID=64467284
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810872312.XA Active CN108955693B (en) | 2018-08-02 | 2018-08-02 | Road network matching method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108955693B (en) |
Families Citing this family (25)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109766642A (en) * | 2019-01-15 | 2019-05-17 | 电子科技大学 | One kind is from evolution traffic network topological modelling approach |
| CN111488414B (en) * | 2019-01-29 | 2023-05-12 | 阿里巴巴集团控股有限公司 | Road task matching method, device and equipment |
| CN109974718B (en) * | 2019-04-09 | 2021-10-22 | 百度在线网络技术（北京）有限公司 | Map matching method, apparatus, device and medium |
| CN110290536B (en) * | 2019-06-14 | 2022-02-08 | 南京拾柴信息科技有限公司 | Method for establishing wireless base station and line segment type geographic terrain incidence matrix |
| CN110493728B (en) * | 2019-08-14 | 2021-06-25 | 深圳市名通科技股份有限公司 | Path identification method, device, equipment and medium based on mobile communication network |
| CN112399348B (en) * | 2019-08-15 | 2022-07-01 | 中移(苏州)软件技术有限公司 | Method and device for determining moving track and storage medium |
| CN110599768A (en) * | 2019-09-07 | 2019-12-20 | 北京智数时空科技有限公司 | Highway traffic flow estimation method based on telecommunication big data |
| CN112815955A (en) * | 2019-10-31 | 2021-05-18 | 荣耀终端有限公司 | Method for prompting trip scheme and electronic equipment |
| CN112857378A (en) * | 2019-11-26 | 2021-05-28 | 北京初速度科技有限公司 | Track matching method and device |
| CN111090642B (en) * | 2019-12-02 | 2023-07-14 | 杭州诚智天扬科技有限公司 | Method for cleaning signaling data of mobile phone |
| CN111132046A (en) * | 2019-12-27 | 2020-05-08 | 江苏欣网视讯软件技术有限公司 | Heuristic signaling data noise point filtering algorithm |
| CN111343585B (en) * | 2020-02-28 | 2021-11-02 | 重庆邮电大学 | Mobile user track map matching method based on hidden Markov model |
| CN111521191A (en) * | 2020-04-20 | 2020-08-11 | 中国农业科学院农业信息研究所 | Mobile phone user moving path map matching method based on signaling data |
| CN111597282B (en) * | 2020-04-24 | 2024-03-19 | 汉海信息技术(上海)有限公司 | Method and device for mining road network data and electronic equipment |
| CN113923602B (en) * | 2020-07-10 | 2023-09-19 | ***通信集团浙江有限公司 | Method and device for filtering signaling drift data |
| CN114071347B (en) * | 2020-07-28 | 2024-04-09 | 中移(苏州)软件技术有限公司 | Space-time matching method and device for multiple signaling tracks |
| CN112131233B (en) * | 2020-08-28 | 2022-11-15 | 北京百度网讯科技有限公司 | Method, apparatus, device and computer storage medium for identifying updated road |
| CN112561948B (en) * | 2020-12-22 | 2023-11-21 | 中国联合网络通信集团有限公司 | Space-time trajectory-based accompanying trajectory recognition method, device and storage medium |
| CN113053110A (en) * | 2020-12-23 | 2021-06-29 | 沈阳世纪高通科技有限公司 | Method and device for calculating road flow based on 4g signaling data |
| CN112653997A (en) * | 2020-12-29 | 2021-04-13 | 西安九索数据技术股份有限公司 | Position track calculation method based on base station sequence |
| CN113205700B (en) * | 2021-03-26 | 2022-08-30 | 福建新大陆软件工程有限公司 | High-speed vehicle position identification method based on mobile phone signaling road network matching |
| CN113587944B (en) * | 2021-06-24 | 2024-03-29 | 深圳市跨越新科技有限公司 | Quasi-real-time vehicle driving route generation method, system and equipment |
| CN114374941B (en) * | 2021-12-17 | 2022-09-20 | 智慧足迹数据科技有限公司 | Track matching method based on signaling data and related device |
| CN114245329B (en) * | 2021-12-21 | 2023-04-07 | 北京红山信息科技研究院有限公司 | Traffic mode identification method, device, equipment and storage medium |
| CN116561437A (en) * | 2023-07-10 | 2023-08-08 | 荣耀终端有限公司 | User behavior prediction method, terminal equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH02138813A (en) * | 1988-11-18 | 1990-05-28 | Honda Motor Co Ltd | Running path display device |
| CN105333874A (en) * | 2015-11-12 | 2016-02-17 | 广东欧珀移动通信有限公司 | Indoor navigation path generating method and system and intelligent terminal |
| CN105387864A (en) * | 2015-10-15 | 2016-03-09 | 深圳市城市交通规划设计研究中心有限公司 | Route planning device and method |
| CN107607121A (en) * | 2017-09-26 | 2018-01-19 | 重庆市智权之路科技有限公司 | Travel line and travel line recommended work method are obtained by geographical location information |
| CN107747947A (en) * | 2017-10-23 | 2018-03-02 | 电子科技大学 | A kind of collaboration itinerary based on user's history GPS track recommends method |
-
2018
- 2018-08-02 CN CN201810872312.XA patent/CN108955693B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH02138813A (en) * | 1988-11-18 | 1990-05-28 | Honda Motor Co Ltd | Running path display device |
| CN105387864A (en) * | 2015-10-15 | 2016-03-09 | 深圳市城市交通规划设计研究中心有限公司 | Route planning device and method |
| CN105333874A (en) * | 2015-11-12 | 2016-02-17 | 广东欧珀移动通信有限公司 | Indoor navigation path generating method and system and intelligent terminal |
| CN107607121A (en) * | 2017-09-26 | 2018-01-19 | 重庆市智权之路科技有限公司 | Travel line and travel line recommended work method are obtained by geographical location information |
| CN107747947A (en) * | 2017-10-23 | 2018-03-02 | 电子科技大学 | A kind of collaboration itinerary based on user's history GPS track recommends method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108955693A (en) | 2018-12-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108955693B (en) | Road network matching method and system | |
| CN111862659B (en) | GPS track data matching and complementing method | |
| CN108320501B (en) | Bus route identification method based on user mobile phone signaling | |
| CN107045673B (en) | Public bicycle flow variation prediction method based on stack model fusion | |
| CN104050817B (en) | Speed limiting information base generation and speed limiting information detection method and system | |
| CN102521973B (en) | A kind of mobile phone switches the road matching method of location | |
| CN105243128B (en) | A kind of user behavior method of trajectory clustering based on data of registering | |
| CN104197945B (en) | Global voting map matching method based on low-sampling-rate floating vehicle data | |
| CN107018493B (en) | Geographic position prediction method based on continuous time sequence Markov model | |
| CN111985710A (en) | Bus passenger trip station prediction method, storage medium and server | |
| CN110972073B (en) | Living circle defining method based on mobile phone big data and walking accessibility | |
| CN105809962A (en) | Traffic trip mode splitting method based on mobile phone data | |
| CN105608505A (en) | Cellular signaling data based track traffic travel mode identification method for resident | |
| CN109189949B (en) | A kind of population distribution calculation method | |
| CN106781478A (en) | A kind of trace tracking method based on LTE signaling datas | |
| CN106997666A (en) | A kind of method that utilization mobile phone signaling data position switching obtains traffic flow speed | |
| CN106408124A (en) | Moving path hybrid forecasting method oriented to data sparse environment | |
| CN103440772B (en) | Method for calculating moving speed of user by means of mobile phone location data | |
| CN109729518B (en) | Mobile phone signaling-based urban traffic early peak congestion source identification method | |
| CN108510008B (en) | Road network extraction method based on floating car track point spatial relationship and distribution | |
| CN110160538B (en) | Map matching method based on mobile phone signaling data | |
| CN110414795B (en) | Newly-increased high-speed rail junction accessibility influence method based on improved two-step mobile search method | |
| CN114866969A (en) | Subway clearing method and system based on mobile phone signaling big data | |
| CN112309118A (en) | Vehicle trajectory calculation method based on space-time similarity | |
| CN109520499A (en) | Region isochronal method in real time is realized based on vehicle GPS track 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 | ||
| CB03 | Change of inventor or designer information | ||
| CB03 | Change of inventor or designer information |
Inventor after: Wang En Inventor after: Huang Qiuyang Inventor after: Yang Yongjian Inventor after: Huang Liping Inventor before: Yang Yongjian Inventor before: Huang Qiuyang Inventor before: Wang En Inventor before: Huang Liping |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |