CN106997666A - A kind of method that utilization mobile phone signaling data position switching obtains traffic flow speed - Google Patents
A kind of method that utilization mobile phone signaling data position switching obtains traffic flow speed Download PDFInfo
- Publication number
- CN106997666A CN106997666A CN201710114066.7A CN201710114066A CN106997666A CN 106997666 A CN106997666 A CN 106997666A CN 201710114066 A CN201710114066 A CN 201710114066A CN 106997666 A CN106997666 A CN 106997666A
- Authority
- CN
- China
- Prior art keywords
- base station
- switching
- sequence
- user
- section
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0108—Measuring and analyzing of parameters relative to traffic conditions based on the source of data
- G08G1/012—Measuring and analyzing of parameters relative to traffic conditions based on the source of data from other sources than vehicle or roadside beacons, e.g. mobile networks
Landscapes
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Mobile Radio Communication Systems (AREA)
- Traffic Control Systems (AREA)
Abstract
The present invention, in way monitoring technical field, more particularly to the method that a kind of switching of utilization mobile phone signaling data position obtains traffic flow speed, comprises the following steps towards the road traffic speed of service：S1：Coverage of the Thiessen polygon formed using base station as the base station；S2：The positive switching sequence of target road section and reverse switching sequence are built, section distance set and reverse distance set under single base station range are obtained using GIS platform；S3：Label and switching time based on the base station extract user base station switching track；S4：Path matching degree checking is carried out to user trajectory sequence and standard switching sequence；S5：Speed acquisition is carried out to the user that the match is successful；S6：Step S3 to S5 is performed to all users, the traffic flow speed of target road section is asked for.By detecting traffic flow speed comparison example with microwave detector, road traffic delay speed can effectively be obtained by demonstrating the traffic flow speed acquisition methods proposed by the invention switched based on position.
Description
Technical field
The present invention utilizes mobile phone signaling towards the road traffic speed of service in way monitoring technical field, more particularly to one kind
The method that Data Position switching obtains traffic flow speed.
Background technology
As urban transportation constantly develops growing with people's trip requirements, facilitating control of traffic and road
Meanwhile, the also acquisition to road traffic speed stream proposes higher requirement.Traditional monitoring method mainly has GPS Floating Cars
Detection, microwave detection, video detection and Coil Detector etc., but various detection methods have certain deficiency：GPS Floating Cars
Detect that accuracy of detection is high, but because installation GPS taxi main activities are in urban central zone, overlay area is affected；
Microwave detector detection is influenceed larger by road conditions, and needs certain installing space；Video detection is influenceed by weather and light
It is larger；Coil Detector accuracy of detection when the magnitude of traffic flow is excessive can be reduced.Mobile phone is carried as " detector " for traffic information collection
The mode at brand-new visual angle has been supplied, has been obtained for transport information and provides new application space.
The data display announced according to the Ministry of Industry and Information Technology of China (MIIT), ends in May, 2016, and China's mobile phone is moved
Dynamic communication user has reached 12.96 hundred million, and the diffusion rate of mobile phones is up to 94%, and these tracking exchages can be used as traffic per family
Information wireless collection sensor, facility basis and data basis are provided for traffic information collection.Mobile communications network can be right
Customer location under its coverage is regularly updated, and the difficult point of this method is to need to mark section by GIS platform
It is fixed, user is matched into respective stretch using different map-matching algorithms, and then carry out by traffic flow speed evaluation method
Velocity estimation.This method will be demarcated to target road section switching sequence, and propose user trajectory and target road section switching sequence
Using asking the method for longest common subsequence to carry out map match, estimated using switching point distance with the ratio of journey time
Traffic flow speed.
The content of the invention
In order to solve traffic jam issue, the real-time monitoring to traffic flow speed is realized, the present invention uses Thiessen polygon
Base station range is demarcated, and then target road section standard switching sequence is extracted by generalized information system, is introduced into biology
Needleman-Wunsch algorithms path adaptation is carried out to user trajectory and target road section, utilize correspondence function to determine
With effect, speed is estimated using traffic flow speed evaluation method.Traffic problems, which are solved, for data in mobile phone proposes new think of
Road.Specific method of the present invention is as follows：
A kind of method that utilization mobile phone signaling data position switching obtains traffic flow speed, comprises the following steps：
S1：Coverage of the Thiessen polygon formed using base station as the base station；
S2：The positive switching sequence of target road section and reverse switching sequence are built, obtaining single base station using GIS platform covers
Section distance set and reverse distance set under scope；Wherein, the positive switching sequence is that target road section origin-to-destination covers base
The label stood, the reverse switching sequence is label of the target road section terminal to starting point coverage base station；
S3：Label and switching time based on the base station extract user base station switching track；
S4：Path matching degree checking is carried out to user trajectory sequence and standard switching sequence；
S5：Speed acquisition is carried out to the user that the match is successful；
S6：Step S3 to S5 is performed to all users, the traffic flow speed of target road section is asked for.
The path matching degree is verified：Ask for the user trajectory sequence and the standard switching sequence most
Long common subsequence and its similarity, above-mentioned similarity and similarity threshold are compared, then matched into by threshold value
Work(, otherwise fails.
The computational methods of the longest common subsequence are specially：The user trajectory sequence and the standard are switched
Sequence is matched, and the length matrix of longest common subsequence is built according to recurrence relation, and updates matrix element, according to returning
Longest common subsequence is tried to achieve in path of tracing back.
The computational methods of the similarity are specially：The length of length/standard sequence of similarity=longest common subsequence
Degree.
The user base station switching track is specially mobile communications network user in a communication network due to location updating
The base station switching sequence of generation.
The standard switching sequence is the base station switching set of target road section, and the standard switching sequence of target road section is entered
Row extract method be specially：
S41：The Thiessen polygon surface layer set up by GIS platform to base station is changed, and is converted into surface layer boundary line
Layer；
S42：Target road section is intercepted out by section to be measured, and sets up new target track route layer；
S43：Surface layer boundary line layer is taken with target track route layer and ships calculation, intersection point is extracted and obtains target road section key section
Point layer, the key node is the switching point switched during target road section traveling；
S44：The key node is extracted to form standard switching sequence successively according to road direction.
The path matching degree is specially user's switching track and the data of the similitude of target road section standard switching track
Change metric.
The matching includes randomness matching and directionality matching.
The beneficial effects of the present invention are：The present invention is based on mobile communication subscriber switch data, with reference to GIS systems,
Unique user is extracted, matched, velocity estimation then is carried out to the user that the match is successful, traffic flow speed is finally tried to achieve,
Whole flow process figure is as shown in Figure 3.The precision subsequently calculated has carried out threshold values δ settings, wherein Sim to track similarity>δ use
Family track is retained.In addition, the present invention is based on the traffic flow speed estimation under mobile communication network base station coverage, lead to
Cross and base station range is determined, choose behind section, base station range point can be extracted by generalized information system
Calculated with link length.
Brief description of the drawings
Fig. 1 Beijing base station range Thiessen polygon；
Fig. 2 path matching schematic diagrames；
Fig. 3 traffic flow speeds estimate flow chart；
Fig. 4 sections base station switches schematic diagram；
Choose figure in Fig. 5 researchs section；
Fig. 6 West 5th Ring Roads perfume spring bridge-apricot stone mouthful bridge section base station labeled graph again；
The forward direction speed comparison diagram of Fig. 7 sections 1；
The inverted speed comparison diagram of Fig. 8 sections 1；
Fig. 9 supernatants bridge-extensively seek bridge section base station labeled graph again；
The forward direction speed comparison diagram of Figure 10 sections 2；
The inverted speed comparison diagram of Figure 11 sections 2.
Embodiment
Below in conjunction with the accompanying drawings, embodiment is elaborated.
In order to solve traffic jam issue, the real-time monitoring to traffic flow speed is realized, the present invention uses Thiessen polygon
Base station range is demarcated, and then target road section standard switching sequence is extracted by generalized information system, is introduced into biology
Needleman-Wunsch algorithms path adaptation is carried out to user trajectory and target road section, utilize correspondence function to determine
With effect, speed is estimated using traffic flow speed evaluation method.Traffic problems, which are solved, for data in mobile phone proposes new think of
Road.Specific method of the present invention is as follows：
(1) base station range is determined
Mobile communications network is also commonly referred to as cellular communications networks, ideally, the covering of a mobile network base station
Scope is with certainly as the inscribed regular hexagon in the center of circle.And as base station is constantly set up, and in actual application " focus "
Origin base station coverage density is larger, and regular hexagon can not cover whole region without overlapping.The present invention will be formed with base station
Thiessen polygon as the base station coverage.
Thiessen polygon (Thiessen Polygon) be by the perpendicular bisector of adjacent two-point defined line constitute it is continuous many
Side shape.This polygon has following characteristics：
1. each discrete point is pertaining only to a Thiessen polygon；
2. other point distances are most short in Thiessen polygon belonging to for discrete point；
3. the distance of the Bian Shangdian of Thiessen polygon to two neighboring discrete point is equal.
When some point is dropped into a Thiessen polygon, at this moment the point and central point are closest.This spatial relationship with
COO (Cell of Origin) positioning rule is exactly like, therefore present invention application Thiessen polygon is entered to base station range
Row processing has reasonability.
Draw base station range Thiessen polygon during, present invention uses GIS software to base station at
Reason, is read out to base station data first, and carrying out base sites layer by the latitude and longitude coordinates of base station draws.It is with base sites layer
Basis, is made using GIS instruments to Thiessen polygon, and each polygonal coverage substantially represents a base
Stand the coverage of network signal, the Thiessen polygon drawn by Beijing mobile communication base station is as shown in Figure 1.
(2) target road section switching sequence is demarcated
Mobile subscriber can produce base station switching in moving process, will be to user position when meeting location updating condition
Put and uploaded, therefore just obtain after being extracted to the interior for a period of time location updating of user the movement locus of the user.For
Determination user is specific to be also needed to subsequently to be matched in the motion of which section road, and the base station of target road section switches to gather and just cried
Do standard switching sequence.Standard switching sequence is divided into positive standard switching sequence L (1, n) { x again1,x2,…,xnAnd reverse mark
Quasi- switching sequence L (n, 1) { xn,xn-1,…,x1, wherein positive standard switching sequence by target road section starting point along road driving to
All base stations number that may be switched in target road section end point process, reverse standard switching sequence is then by target road section edge
Road terminal along road driving to target road section starting point during all base stations number that may switch.
In order to obtain section standard switching sequence, mainly by carrying out substantial amounts of drive test to target road section, base station is obtained
" the drive test car signal " taken is analyzed, then determine maximum probability switching sequence according to probability theory theories and methods with reach compared with
Excellent switching sequence.
The first step：By GIS platform, the Thiessen polygon surface layer set up to base station is changed, and is converted into surface layer side
Layer boundary；
Second step：Research section is intercepted, target road section is intercepted out and sets up new target track route layer；
3rd step：Surface layer boundary line layer is taken with target track route layer and ships calculation, intersection point is extracted, target road section pass is obtained
Key node layer, these key nodes are it is believed that the switching point switched during target road section traveling；
4th step：Key node is extracted to form standard switching sequence successively according to road direction.
(3) path matching algorithm
Path matching algorithm is one kind of map match, and the present invention, will be complex based on user's switching track
Map-matching algorithm is converted into the fairly simple sequence alignment algorithms in data mining.The core of sequence alignment algorithms is exactly
Compare the difference between two given sequences (text).Track is carried out by seeking the longest common subsequence of two sequences
Match somebody with somebody, wherein the specific algorithm that the present invention is used is Needleman-Wunsch algorithms.
Needleman-Wunsch algorithm basic thoughts are, when matching two sequences, to find out in the two sequences all
In the presence of and most long identical subsequence.Longest common subsequence does not need element continuously to occur, but requires the order of appearance
Unanimously, such as sequence X={ P1,P2,P3,P4, Y={ P1,P3,P2, then their most normal common subsequence is { P1,P3}.Make
For the basic algorithm of sequence alignment, Needleman-Wunsch algorithms belong to dynamically seeking most to grow public son in global scope
In sequence, the present invention sequence X and sequence Y longest common subsequence are represented using LCS (X, Y).If Len (X) represents sequence X
Length, then have lenLCS (X, Y)=Len (LCS (X, Y)), as lenLCS (X, Y)=0, represent two sequences without public affairs
Common part.Longest common subsequence has following structure：
If sequence X={ x1,x2,…,xmAnd sequence Y={ y1,y2,…,ynA longest common subsequence be Z, Z
=LCS (X, Y)={ z1,z2,…,zk, there is following property：
Property 1：If xm=yn, then zk=xm=ynAnd Zk-1It is Xm-1And Yn-1Longest common subsequence；
Property 2：If xm≠ynAnd zk≠xm, then Z is Xm-1With Y longest common subsequence；
Property 3：If xm≠ynAnd zk≠yn, then Z is X and Yn-1Longest common subsequence.
Wherein, Xm-1={ x1,x2,…,xm-1, Yn-1={ y1,y2,…,yn-1, Zk-1={ z1,z2,…,zk-1}。
It can be drawn by the structural property of longest common subsequence problem, it is desirable to sequence X={ x1,x2,…,xmAnd sequence
Y={ y1,y2,…,ynLongest common subsequence LCS (X, Y) mainly have following two situations:：
First, work as xm=ynWhen, LCS (X, Y)={ LCS (Xm-1,Ym-1),xm}。
Second, work as xm≠ynWhen, LCS (X, Y)={ LCS (Xi,Yj)|max(lenLCS(Xm-1,Y),lenLCS(X,
Yn-1)), i=m-1 or j=n-1 }.
It can be seen from this recurrence relation, the 2nd kind of situation has subproblem overlapping, as calculating Xm-1With Y and X and Yn-1's
During longest common subsequence, it is intended to calculate Xm-1With Yn-1Longest common subsequence.Ask sequence longest common subsequence it
It is preceding, it is necessary to set up the recurrence relation of subproblem optimal value.When there is two sequence X to be matched={ x1,x2,……,xi, Y=
{y1,y2,......,yj, calculating X and Y length has Len (X)=i, Len (Y)=j, defines lenLCS (i, j)=lenLCS
(Xi,Yj).As i=0 or j=0, XiWith YjLongest common subsequence be empty sequence, lenLCS (i, j)=0.Thus, build
Vertical recurrence relation is as follows：
During longest common subsequence is sought, Needleman-Wunsch algorithms are the bases in above recurrence relation
On plinth, each element in sequence to be matched is compared, sequence is represented by setting up the rating matrix based on lenLCS
Matching length.It is now assumed that there is two track sets T1、T2, T1={ P1,P2,P4,P5,P6,P7,P9, T2={ P1,P3,P5,P6,
P7, two track sets are matched, that is, seek LCS (T1, T2), it is necessary first to which most long public affairs are built according to recurrence relation
Sub-sequence length lenLCS matrixes, then update matrix element altogether, final to try to achieve longest common subsequence LCS using backtracking.
Step 1：Build lenLCS matrixes.
The first step：Create and initialize lenLCS matrixes.T1, T2 sequential element number are respectively 7 and 5, so building 6
× 8 matrix, such as following table, unnecessary row and column are initial LCS numbers.
| T1 | P1 | P2 | P4 | P5 | P6 | P7 | P9 | |
| T2 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| P1 | 0 | |||||||
| P3 | 0 | |||||||
| P5 | 0 | |||||||
| P6 | 0 | |||||||
| P7 | 0 |
Second step：Utilize core formula, calculating matrix the first row.
| T1 | P1 | P2 | P4 | P5 | P6 | P7 | P9 | |
| T2 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| P1 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
| P3 | 0 | |||||||
| P5 | 0 | |||||||
| P6 | 0 | |||||||
| P7 | 0 |
3rd step：Using core formula, other rows of calculating matrix obtain final lenLCS (T1,T2)。
According to lenLCS matrixes, it can be deduced that lenLCS (T1,T2)=lenLCS (T2,T1)=lenLCS (5,7)=4.
T has been drawn by Needleman-Wunsch the algorithm matrixes1With T2Longest common subsequence length be 4.In order to most long
Common subsequence LCS finds the sequence accordingly matched, it is necessary to recall to matrix above.
Step 2：Recall matrix, find longest common subsequence.
The first step：Position lenLCS matrixes lower right corner element.
| T1 | P1 | P2 | P4 | P5 | P6 | P7 | P9 | |
| T2 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| P1 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
| P3 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
| P5 | 0 | 1 | 1 | 1 | 2 | 2 | 2 | 2 |
| P6 | 0 | 1 | 1 | 1 | 2 | 3 | 3 | 3 |
| P7 | 0 | 1 | 1 | 1 | 2 | 3 | 4 | 4 |
Second step：Trace unit lattice are to the matrix upper left corner.According to following backtracking rule, matrix unit is recalled.
If T1(i)=T2(j) upper left corner cell of current cell, is then dateed back；
If T1(i)≠T2(j) the current cell upper left corner, top, the cell of left side intermediate value maximum, are then dateed back, such as
There is the cell of identical maximum in fruit, priority is chosen successively according to the order in top, the left side, the upper left corner；
If current cell is the first row in matrix, the cell on the left side is dateed back；
If current cell is the first row in matrix, the cell of top is dateed back.
| T1 | P1 | P2 | P4 | P5 | P6 | P7 | P9 | |
| T2 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| P1 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
| P3 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
| P5 | 0 | 1 | 1 | 1 | 2 | 2 | 2 | 2 |
| P6 | 0 | 1 | 1 | 1 | 2 | 3 | 3 | 3 |
| P7 | 0 | 1 | 1 | 1 | 2 | 3 | 4 | 4 |
3rd step：According to backtracking path, longest common subsequence is write out.
If tracing back to upper left corner cell, the cell element belongs to longest common subsequence element, complete of search
Coupling path, matching sequence LCS (T1,T2)={ P1,P5,P6,P7Complete.
(3) path matching algorithm is realized
According to the recurrence relation of algorithm, the length for calculating longest common subsequence is modeled.The algorithm is with two sequences
X={ x1,x2,…,xmAnd Y={ y1,y2,…,ynAs input, two array Z [0 ... m, 0 ... n] of output and L [1 ... m,
1 ... n], the middle storages of wherein Z [i, j] are XiWith YjLongest common subsequence length, L [i, j] then records current Z [i, j]
With the relation of a upper subproblem, for subsequent builds longest common subsequence.Therefore, the length of X and Y longest common subsequence
Degree is then in storage Z [m, n].Calculating false code is carried out to lenLCS as follows：
By this process, sequence X and the length of sequence Y longest common subsequence are resulted in.Continue to use
The thought of Needleman-Wunsch algorithms backtracking, by L arrays, starts from L [m, n], is searched along arrow direction
Rope, finds longest common subsequence.
WhenRepresent LCS (Xi,Yj)={ LCS (Xi-1,Yj-1),xi, that is to say, that xi=yj；
When L [i, j]=" ↑ ", LCS (X are representedi,Yj)=LCS (Xi-1,Yj)；
When L [i, j]=" ← ", LCS (X are representedi,Yj)=LCS (Xi,Yj-1)。
This method is equally that LCS is found according to inverted sequence, and in order to obtain LCS, it is only necessary to meeting the first situation
Carry out extraction output, extraction process false code is as follows：
Longest common subsequence LCS can be tried to achieve by this process.The example of upper one section is calculated, as a result such as
Under：
Started by L (5,7), searched for successively according to the direction of arrow, record eachCorresponding element, final
To LCS (T1,T2)={ P1,P5,P6,P7, shown result is consistent with analysis.
(3) matching degree is calculated
Sequences match degree refers to that user's switching track is similar compared with target road section standard switching track in the present invention
Property digitization measurement, that is, calculate the similarity (Similarity) of two sequences.It is now assumed that two information A, B to be matched,
Both similarities represent that usually, in the case of A, both B status identical, similarity function Sim should expire with Sim (A, B)
It is enough lower two attributes：
Property one：0≤Sim(A,B)≤1；
Property two：Sim (A, B)=Sim (B, A).
But in the present invention, it is desirable to be that user's switching track matches with target road section standard switching sequence, belong to
Single (or the two) standard sequence of multi-user's sequences match, in the matching process, the matching status of two sequences is differed.
In Fig. 2 (a), target road section standard switching sequence and user trajectory travel direction are not differentiated between, that is, overall track one kind
" randomness ", then be equivalent to multi-user and matched to single standard sequence.In Fig. 2 (b), standard sequence has directionality, together
When user's travel direction is made a distinction, be thus that multi-user matches to two standard sequences.
Therefore, this special sequence feature of standard sequence is had based on the present invention, definition is a kind of to be based on most long public sub- sequence
The similarity calculating method of row.It is assumed that target road section standard switching sequence is Trs={ Ps1,Ps2,PS3,Ps4,Ps5, user cuts
Change sequence Tru={ Pu1,Pu2,Pu3,Pu4,Pu5,Pu6, the longest common subsequence of two sequences is LCS (Trs,Tru), most long public affairs
Length lenLCS (the Tr of common subsequences,Tru) represent, if distinguishing user's travel direction, need to TrsInverted
Matched afterwards with user, (c) and (d), both similarity Sim (Tr in such as Fig. 2s,Tru) be represented by：
Len (Tr in formulas) --- standard sequence TrsLength.
Although the similarity function meets the property one of general similarity function, and property two is not met, for
The content of follow-up study of the present invention has enough condition supports, belongs to based on the sequence similarity function under standard sequence.
The present invention based on mobile communication subscriber switch data, with reference to generalized information system, unique user is extracted,
Match somebody with somebody, velocity estimation then is carried out to the user that the match is successful, traffic flow speed is finally tried to achieve, whole flow process figure is as shown in Figure 3.
The precision subsequently calculated has carried out threshold values δ settings, wherein Sim to track similarity>δ user trajectory is retained.In addition, this
Invention is based on the traffic flow speed estimation under mobile communication network base station coverage, by being carried out to base station range
Determine, choose behind section, base station range point can extract by GIS systems and be calculated with link length, such as
Shown in Fig. 4.
The present invention is further illustrated with reference to the accompanying drawings and examples.The present invention have chosen Beijing five rings as research
Object, have chosen two sections of representative sections of comparison, research section section as shown in Figure 5 be respectively the fragrant spring bridge in West 5th Ring Road-
Apricot stone mouthful bridge section, North 5th Ring Road supernatant bridge-extensively seek bridge section.Wherein the fragrant spring bridge in West 5th Ring Road-apricot stone mouthful bridge section gateway is less,
The crowded duration of road is shorter, and North 5th Ring Road supernatant bridge-come is wide to seek that bridge section gateway is more, and usually road situation is by above and below
Class's peak influence is larger.
Embodiment 1：West 5th Ring Road perfume spring bridge-apricot stone mouthful bridge
West 5th Ring Road perfume spring bridge-apricot stone mouthful bridge section total length total length is about 4.3 kilometers, altogether across 12 base stations, to these
The CELLID of base station extractions and again label successively from south to north, number is sequentially demarcated by 001~012, as shown in Figure 6.According to
Target road section standard switching sequence acquisition methods, the switching key point to the section is extracted, and is provided simultaneously, from south to north
Direction be forward direction, opposite direction be reverse, it is possible thereby to determine the standard of the both forward and reverse directions of fragrant spring bridge-apricot stone mouth bridge section
Switching sequence.
According to target road section switching sequence calibration algorithm, section 1 is demarcated, positive switching sequence Tr is obtainedsf=
{ 001,002 ..., 012 }, reverse switching sequence Trsb={ 012,011 ..., 001 }, after obtaining target road section switching sequence,
Flow is estimated according to traffic flow speed, the traffic flow speed of section different time sections is calculated.The time that the present invention chooses
Section is 6:00~20:00, because user position update is that, by switching base station, therefore the present invention is taken every 10 minutes to this
User is matched with section 1 in one period, then estimates corresponding speed.In order to more preferably verify the feasibility of this algorithm, this
Invention is extracted to section 1 with the traffic flow speed that the microwave remote sensor of period is measured, and the sampling interval of microwave remote sensor is
2 minutes, although the interval with inventive algorithm is inconsistent, but checking to this algorithm can be used for by overall trend, it is micro-
The speed that ripple detector data is estimated with this algorithm is contrasted, as illustrated in figs. 7 and 8.
Shown in Fig. 7 is to study the positive traffic flow speed in period section 1, and Fig. 8 is the reverse friendship in research period section 1
Through flow velocity, wherein blueness is inventive algorithm speed, red is microwave detection speed data, by contrast as can be seen that originally
The velocity perturbation of invention algorithm estimation is big compared with the velocity perturbation that microwave is measured, but overall trend is basically identical.Due to road
Section 1 is in Beijing West 5th Ring Road position, and gateway is less, and usually wagon flow is stablized relatively, is influenceed smaller by early evening peak, positive
Road average-speed substantially remains in more than 65km/h.The traffic flow speed reverse by observing section 1, in the afternoon before 14 points
Speed has declined afterwards, but speed has larger fluctuation, while inventive algorithm is 12:00 and 13:Occur one between 00
Individual obvious singular point, by analysis and research, find the successful user of period fits it is less, and exist certain user's speed compared with
It is low, therefore the average speed of entirety has been dragged down, but for overall trend, the speed that inventive algorithm is tried to achieve and microwave speed
Degree is basically identical.
Embodiment 2：North 5th Ring Road supernatant bridge-extensively seek bridge
North 5th Ring Road supernatant bridge-next wide bridge section total length total length of seeking is about 8.8 kilometers, altogether across 25 base stations, to these
The CELLID of base station extractions and again label successively from west to east, number is sequentially demarcated by 101~125, as shown in Figure 9.According to
Target road section standard switching sequence acquisition methods, the switching key point to the section is extracted, and is provided simultaneously, from west to east
Direction be forward direction, opposite direction be reverse, it is possible thereby to determine the standard of the both forward and reverse directions of supernatant bridge-extensively to seek bridge section
Switching sequence.
According to target road section switching sequence calibration algorithm, section 1 is demarcated, positive switching sequence Tr is obtainedsf=
{ 101,002 ..., 125 }, reverse switching sequence Trsb={ 125,124 ..., 001 }, after obtaining target road section switching sequence,
Flow is estimated according to traffic flow speed, the traffic flow speed of section different time sections is calculated.It is identical with section 1, this hair
The period of bright selection is 6:00~20:00, take and user in this period is matched with section 2 every 10 minutes,
Then corresponding speed is calculated.Similarly, the present invention is carried out to section 2 with the traffic flow speed that the microwave remote sensor of period is measured
Extract, the sampling interval of microwave remote sensor is 2 minutes, Figure 10 is with shown in Figure 11 being that microwave detector data are estimated with this algorithm
The speed of calculation is contrasted.
Shown in Figure 10 is to study the positive traffic flow speed in period section 2, and Figure 11 is that research period section 2 is reverse
Traffic flow speed, wherein blue is inventive algorithm speed, red is microwave detection speed data, be can be seen that by contrast
The velocity perturbation of inventive algorithm estimation is big compared with the velocity perturbation that microwave is measured, but overall trend is basically identical.Due to
Section 2 is in Beijing North 5th Ring Road position, and gateway and highway tie point are more, and whole day vehicle flow fluctuating is larger, by morning
The influence of evening peak is big, positive Road average-speed continuous decrease during morning peak, 9:30 or so there is minimum point, greatly
About 45km/h, though having an impact during evening peak, speed maintains 55km/h all the time.Inventive algorithm is same in fluctuation
In the case that situation is larger, good uniformity has been remained in that with microwave speed.The reverse traffic flow speed in section 2 is observed,
Although average speed is influenceed all the time in more than 60km/h by evening peak during morning peak, traffic flow speed rise and fall compared with
Greatly, the algorithm of the same present invention is in the period fluctuation compared with usually bigger, and equally for overall trend, inventive algorithm is asked
The speed and microwave speed obtained is basically identical.
Forward part inventive algorithm has done qualitative analysis, in order to more accurately prove the feasibility of this algorithm, to this calculation
Method carries out quantitative analysis.Because the time interval that the two-way switch speed in base station that the present invention is used is estimated is 10min, and microwave
The sampling interval tested the speed is 2min, in order to preferably be contrasted with Microwave Velocity, and the present invention enters to the two-way switch speed in base station
Row spline interpolation is to reach same data break, using absolute error e, percentage error PE and mean square error MSE come to calculating
Method is tested.
E=| vh-vm|
V in formulah--- the double switching calculating speeds in base station；
vm--- microwave detector detection speed；
N --- experiment numbers.
By to two target road sections 06:00~20:00 data carry out velocity estimation, and the result of acquisition carries out error point
Analysis, final result is shown in Table 1.
The traffic flow speed comparing result of table 1
From the statistical result of table 1, it will be seen that the traffic flow velocity based on the two-way switching of mobile communication network base station
Spending evaluation method has higher precision.The absolute error in section 1 and section 2 is on below 5.5km/h, the forward direction of section 1 and road
The percentage error of section 2 is 5% or so, and mean square error is below 5, and the reverse percentage error in section 1 is slightly higher compared with other threes.
Global error source mainly includes calculation error and measurement error.Calculation error is mainly used between base station switching key point
The speed of distance and switching time differential conversion there is certain error.Because the distribution of traffic flow spatially is not uniform point
Cloth, with certain spatial character, the speed calculated simply by ratio can have error to a certain extent.
Measuring speed mainly includes two parts, and a part is that the distance measured by GIS software switching node exists with actual range
Certain measurement error, another part is that there is also error for the section speed that detects of microwave detector.
By the comparative analysis to section 1 reversely between other threes, the reason for its precision is slightly lower is in path matching
During, slightly few by the user of matching algorithm, the speed difference with microwave detector is larger, so that it is higher to result in error.
Section 2 is multiple port section, and big compared with the vehicle flowrate of section 1, by matching, user is more, therefore the overall velocity error calculated is low.
But the error of two kinds of situations is within tolerance interval, therefore, the traffic proposed by the present invention based on the two-way switching in base station
It is effective to flow velocity estimation.
The present invention carries out traffic flow speed using mobile phone as " movable sensor " using handset mobile communications switch data
The method of acquisition, belongs to technical field of traffic transportation.Using user trajectory and target road section switching sequence using asking most long public
The method of subsequence carries out map match, and traffic flow speed is obtained with the ratio of journey time using switching point distance.This hair
The determination mode of bright base station range is demarcated using Thiessen polygon, and then by GIS platform to target road section standard
Switching sequence is extracted, and determines link length, and cellphone subscriber is matched into corresponding road using different map-matching algorithms
Section, the Needleman-Wunsch algorithms being introduced into biology construct lenLCS matrixes, and carrying out backtracking to matrix finds accordingly
Sequence is matched, the similarity data that sequences match degree i.e. user's switching track is compared with target road section standard switching track is determined
Change measurement, and then speed acquisition is carried out by traffic flow speed acquisition methods.By detecting traffic flow speed with microwave detector
Comparison example, demonstrating the traffic flow speed acquisition methods proposed by the invention switched based on position can effectively be obtained
Road traffic delay speed.
This embodiment is only the present invention preferably embodiment, but protection scope of the present invention is not limited thereto,
Any one skilled in the art the invention discloses technical scope in, the change or replacement that can be readily occurred in,
It should all be included within the scope of the present invention.Therefore, protection scope of the present invention should be with scope of the claims
It is defined.
Claims (8)
1. a kind of method that utilization mobile phone signaling data position switching obtains traffic flow speed, it is characterised in that including following step
Suddenly：
S1：Coverage of the Thiessen polygon formed using base station as the base station；
S2：The positive switching sequence of target road section and reverse switching sequence are built, is obtained using GIS platform under single base station range
Section distance set and reverse distance set；Wherein, the positive switching sequence is the mark of target road section origin-to-destination coverage base station
Number, the reverse switching sequence is label of the target road section terminal to starting point coverage base station；
S3：Label and switching time based on the base station extract user base station switching track；
S4：Path matching degree checking is carried out to user trajectory sequence and standard switching sequence；
S5：Speed acquisition is carried out to the user that the match is successful；
S6：Step S3 to S5 is performed to all users, the traffic flow speed of target road section is asked for.
2. method according to claim 1, it is characterised in that the path matching degree checking is specially：Ask for the user
The longest common subsequence and its similarity of track sets and the standard switching sequence, by above-mentioned similarity and similarity threshold
Value is compared, and by threshold value, then the match is successful, otherwise fails.
3. method according to claim 2, it is characterised in that the computational methods of the longest common subsequence are specially：It is right
The user trajectory sequence is matched with the standard switching sequence, and the length of longest common subsequence is built according to recurrence relation
Matrix is spent, and updates matrix element, longest common subsequence is tried to achieve according to backtracking path.
4. method according to claim 2, it is characterised in that the computational methods of the similarity are specially：Similarity=most
The length of length/standard sequence of long common subsequence.
5. method according to claim 1, it is characterised in that the user base station switching track is specially mobile communications network
The base station switching sequence that user is produced due to location updating in a communication network.
6. method according to claim 1, it is characterised in that the standard switching sequence for target road section base station switching collection
Close, be specially to the method that the standard switching sequence of target road section is extracted：
S41：The Thiessen polygon surface layer set up by GIS platform to base station is changed, and is converted into surface layer boundary line layer；
S42：Target road section is intercepted out by section to be measured, and sets up new target track route layer；
S43：Surface layer boundary line layer is taken with target track route layer and ships calculation, intersection point is extracted and obtains target road section key node layer,
The key node is the switching point switched during target road section traveling；
S44：The key node is extracted to form standard switching sequence successively according to road direction.
7. method according to claim 1, it is characterised in that the path matching degree is specially user's switching track and target
The digitization metric of the similitude of section standard switching track.
8. according to the methods described of claim 1 or 7, it is characterised in that the matching includes randomness matching and directionality matching.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710114066.7A CN106997666B (en) | 2017-02-28 | 2017-02-28 | Method for obtaining traffic flow speed by switching positions of mobile phone signaling data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710114066.7A CN106997666B (en) | 2017-02-28 | 2017-02-28 | Method for obtaining traffic flow speed by switching positions of mobile phone signaling data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106997666A true CN106997666A (en) | 2017-08-01 |
| CN106997666B CN106997666B (en) | 2020-07-03 |
Family
ID=59431081
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710114066.7A Expired - Fee Related CN106997666B (en) | 2017-02-28 | 2017-02-28 | Method for obtaining traffic flow speed by switching positions of mobile phone signaling data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106997666B (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107798182A (en) * | 2017-10-19 | 2018-03-13 | 江苏金陵科技集团有限公司 | A kind of quick calculating of base station radiation scope and update method |
| CN108322891A (en) * | 2018-01-11 | 2018-07-24 | 江苏欣网视讯软件技术有限公司 | Region congestion recognition methods based on user mobile phone signaling data |
| CN109872529A (en) * | 2017-12-01 | 2019-06-11 | 北京万集科技股份有限公司 | Traffic flow information collection method and device based on LTE-V communication |
| CN110047277A (en) * | 2019-03-28 | 2019-07-23 | 华中科技大学 | Road traffic congestion arrangement method and system based on signaling data |
| CN110708664A (en) * | 2019-10-11 | 2020-01-17 | 同帅科技（天津）有限公司 | Traffic flow sensing method and device, computer storage medium and electronic equipment |
| CN111402575A (en) * | 2018-12-17 | 2020-07-10 | 中国移动通信集团北京有限公司 | Method and device for evaluating traffic running state |
| CN113205700A (en) * | 2021-03-26 | 2021-08-03 | 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 |
| CN113379296A (en) * | 2021-06-28 | 2021-09-10 | 平安信托有限责任公司 | Report index normalization method and device, electronic equipment and readable storage medium |
| CN113709006A (en) * | 2021-10-29 | 2021-11-26 | 上海闪马智能科技有限公司 | Flow determination method and device, storage medium and electronic device |
| CN113778986A (en) * | 2021-08-20 | 2021-12-10 | 江西省军民融合研究院 | Vehicle supervision analysis method based on Beidou space-time data reconstruction and data mining |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2003199154A (en) * | 2001-12-21 | 2003-07-11 | Fujitsu Ten Ltd | Communication means switching method |
| CN102722984A (en) * | 2012-06-19 | 2012-10-10 | 张家港市鸿嘉数字科技有限公司 | Real-time road condition monitoring method |
| CN103440772A (en) * | 2013-08-28 | 2013-12-11 | 北京工业大学 | Method for calculating moving speed of user by means of mobile phone location data |
| CN105389996A (en) * | 2015-12-17 | 2016-03-09 | 北京亚信蓝涛科技有限公司 | Traffic operation condition characteristic parameter extraction method based on big data |
-
2017
- 2017-02-28 CN CN201710114066.7A patent/CN106997666B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2003199154A (en) * | 2001-12-21 | 2003-07-11 | Fujitsu Ten Ltd | Communication means switching method |
| CN102722984A (en) * | 2012-06-19 | 2012-10-10 | 张家港市鸿嘉数字科技有限公司 | Real-time road condition monitoring method |
| CN103440772A (en) * | 2013-08-28 | 2013-12-11 | 北京工业大学 | Method for calculating moving speed of user by means of mobile phone location data |
| CN105389996A (en) * | 2015-12-17 | 2016-03-09 | 北京亚信蓝涛科技有限公司 | Traffic operation condition characteristic parameter extraction method based on big data |
Non-Patent Citations (2)
| Title |
|---|
| 毛晓汶 等: ""基于手机信令技术的区域交通出行特征研究"", 《中国优秀硕士论文全文数据库 工程科技Ⅱ辑》 * |
| 高瑞: ""基于位置数据的交通流速度分析应用"", 《中国优秀硕士论文学位论文全文数据库 信息科技辑》 * |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107798182A (en) * | 2017-10-19 | 2018-03-13 | 江苏金陵科技集团有限公司 | A kind of quick calculating of base station radiation scope and update method |
| CN109872529A (en) * | 2017-12-01 | 2019-06-11 | 北京万集科技股份有限公司 | Traffic flow information collection method and device based on LTE-V communication |
| CN109872529B (en) * | 2017-12-01 | 2021-01-15 | 北京万集科技股份有限公司 | Traffic flow information acquisition method and device based on LTE-V communication |
| CN108322891B (en) * | 2018-01-11 | 2020-08-28 | 江苏欣网视讯软件技术有限公司 | Traffic area congestion identification method based on user mobile phone signaling |
| CN108322891A (en) * | 2018-01-11 | 2018-07-24 | 江苏欣网视讯软件技术有限公司 | Region congestion recognition methods based on user mobile phone signaling data |
| CN111402575A (en) * | 2018-12-17 | 2020-07-10 | 中国移动通信集团北京有限公司 | Method and device for evaluating traffic running state |
| CN110047277A (en) * | 2019-03-28 | 2019-07-23 | 华中科技大学 | Road traffic congestion arrangement method and system based on signaling data |
| CN110708664B (en) * | 2019-10-11 | 2020-11-06 | 同帅科技（天津）有限公司 | Traffic flow sensing method and device, computer storage medium and electronic equipment |
| CN110708664A (en) * | 2019-10-11 | 2020-01-17 | 同帅科技（天津）有限公司 | Traffic flow sensing method and device, computer storage medium and electronic equipment |
| CN113205700A (en) * | 2021-03-26 | 2021-08-03 | 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 |
| 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 |
| CN113379296A (en) * | 2021-06-28 | 2021-09-10 | 平安信托有限责任公司 | Report index normalization method and device, electronic equipment and readable storage medium |
| CN113778986A (en) * | 2021-08-20 | 2021-12-10 | 江西省军民融合研究院 | Vehicle supervision analysis method based on Beidou space-time data reconstruction and data mining |
| CN113778986B (en) * | 2021-08-20 | 2023-08-29 | 江西省军民融合研究院 | Carrier supervision analysis method based on Beidou space-time data reconstruction and data mining |
| CN113709006A (en) * | 2021-10-29 | 2021-11-26 | 上海闪马智能科技有限公司 | Flow determination method and device, storage medium and electronic device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106997666B (en) | 2020-07-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106997666A (en) | A kind of method that utilization mobile phone signaling data position switching obtains traffic flow speed | |
| CN108955693B (en) | Road network matching method and system | |
| CN105493109B (en) | Inferred using the air quality of multiple data sources | |
| CN106912015B (en) | Personnel trip chain identification method based on mobile network data | |
| Yuan et al. | An interactive-voting based map matching algorithm | |
| CN110213724B (en) | Pseudo base station motion trajectory identification method | |
| US10578451B2 (en) | Method and system for utilizing a trip history to predict a destination | |
| CN102521973B (en) | A kind of mobile phone switches the road matching method of location | |
| Gong et al. | Data selection in machine learning for identifying trip purposes and travel modes from longitudinal GPS data collection lasting for seasons | |
| CN108109423B (en) | Underground parking lot intelligent navigation method and system based on WiFi indoor positioning | |
| CN107516417B (en) | A kind of real-time highway flow estimation method for excavating spatial and temporal association | |
| JP5815718B2 (en) | Method for determining and verifying navigation priority settings using probe data | |
| CN109688532B (en) | Method and device for dividing city functional area | |
| CN103929719B (en) | The optimization method and optimization device of location information | |
| CN105809962A (en) | Traffic trip mode splitting method based on mobile phone data | |
| Jiang et al. | The impact of the transportation network companies on the taxi industry: Evidence from Beijing’s GPS taxi trajectory data | |
| CN101925939A (en) | Method and system for estimating road traffic | |
| CN106710208A (en) | Traffic state acquisition method and device | |
| CN110472999B (en) | Passenger flow mode analysis method and device based on subway and shared bicycle data | |
| CN110598917B (en) | Destination prediction method, system and storage medium based on path track | |
| CN110880238B (en) | Road congestion monitoring method based on mobile phone communication big data | |
| CN112556717B (en) | Travel mode screening method and travel route recommending method and device | |
| CN109348404B (en) | Method for extracting individual travel road track in big data environment | |
| CN104282142A (en) | Bus station arrangement method based on taxi GPS data | |
| Li et al. | Estimating crowd flow and crowd density from cellular data for mass rapid transit |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20200703 |