CN106997666B - Method for obtaining traffic flow speed by switching positions of mobile phone signaling data - Google Patents
Method for obtaining traffic flow speed by switching positions of mobile phone signaling data Download PDFInfo
- Publication number
- CN106997666B CN106997666B CN201710114066.7A CN201710114066A CN106997666B CN 106997666 B CN106997666 B CN 106997666B CN 201710114066 A CN201710114066 A CN 201710114066A CN 106997666 B CN106997666 B CN 106997666B
- Authority
- CN
- China
- Prior art keywords
- base station
- sequence
- switching
- road section
- traffic flow
- 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.)
- Expired - Fee Related
Links
Images
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
Abstract
The invention relates to the technical field of road traffic running speed in-transit monitoring, in particular to a method for acquiring traffic flow speed by utilizing mobile phone signaling data position switching, which comprises the following steps: s1: taking a Thiessen polygon formed by a base station as a coverage area of the base station; s2: constructing a forward switching sequence and a reverse switching sequence of a target road section, and acquiring a road section distance set and a reverse distance set under the coverage of a single base station by using a GIS platform; s3: extracting a switching track of the user base station based on the label and the switching time of the base station; s4: carrying out track matching degree verification on the user track sequence and the standard switching sequence; s5: acquiring the speed of the successfully matched user; s6: steps S3 to S5 are performed for all users, and the traffic flow speed of the target link is found. By comparing the microwave detector with the traffic flow speed detection example, the method for acquiring the traffic flow speed based on the position switching can effectively acquire the road section traffic flow speed.
Description
Technical Field
The invention relates to the technical field of road traffic running speed in-transit monitoring, in particular to a method for acquiring traffic flow speed by utilizing mobile phone signaling data position switching.
Background
With the continuous development of urban traffic and the increasing travel demand of people, the method brings convenience to road traffic management and simultaneously puts higher requirements on the acquisition of road traffic speed and flow. The traditional monitoring methods mainly comprise GPS floating car detection, microwave detection, video detection, coil detection and the like, but various detection methods have certain defects: the GPS floating car has high detection precision, but because a taxi provided with a GPS mainly moves in a central area of a city, a coverage area is influenced; the microwave detector is greatly influenced by road conditions and needs a certain installation space; the video detection is greatly influenced by weather and light; the detection accuracy of the coil detection may be reduced when the traffic flow is excessively large. The mobile phone as a detector provides a brand-new visual angle mode for traffic information acquisition and provides a new application space for traffic information acquisition.
According to data display published by Mint of China industry and information technology (MIIT), by 2016 (5 months), mobile phone mobile communication users in China reach 12.96 hundred million, the popularity rate of mobile phones reaches 94%, and the mobile communication users can be used as wireless traffic information acquisition sensors, so that a facility foundation and a data foundation are provided for traffic information acquisition. The mobile communication network can update the user position under the coverage area of the mobile communication network regularly, and the method has the difficulty that the road sections need to be calibrated through a GIS platform, the users are matched to the corresponding road sections through different map matching algorithms, and then the speed is estimated through a traffic flow speed estimation method. The method comprises the steps of calibrating a target road section switching sequence, providing a method for solving the longest public subsequence for map matching between a user track and the target road section switching sequence, and estimating the traffic flow speed by using the ratio of the distance of a switching point to the travel time.
Disclosure of Invention
In order to solve the problem of traffic jam and realize real-time monitoring of traffic flow speed, the method uses a Thiessen polygon to calibrate the coverage area of a base station, further extracts a standard switching sequence of a target road section through a GIS system, introduces a Needleman-Wunsch algorithm in biology to perform road matching on a user track and the target road section, determines a matching effect by using a matching speed function, and estimates the speed by using a traffic flow speed estimation method. A new idea is provided for solving the traffic problem of the mobile phone data. The specific method of the invention is as follows:
a method for obtaining traffic flow speed by switching positions of mobile phone signaling data comprises the following steps:
s1: taking a Thiessen polygon formed by a base station as a coverage area of the base station;
s2: constructing a forward switching sequence and a reverse switching sequence of a target road section, and acquiring a road section distance set and a reverse distance set under the coverage of a single base station by using a GIS platform; the forward switching sequence is a mark number covering the base station from the starting point to the end point of the target road section, and the reverse switching sequence is a mark number covering the base station from the end point to the starting point of the target road section;
s3: extracting a switching track of the user base station based on the label and the switching time of the base station;
s4: carrying out track matching degree verification on the user track sequence and the standard switching sequence;
s5: acquiring the speed of the successfully matched user;
s6: steps S3 to S5 are performed for all users, and the traffic flow speed of the target link is found.
The track matching degree verification specifically comprises the following steps: and solving the longest public subsequence of the user track sequence and the standard switching sequence and the similarity of the longest public subsequence, comparing the similarity with a similarity threshold, and if the similarity passes the threshold, successfully matching, otherwise, failing.
The calculation method of the longest public subsequence specifically comprises the following steps: and matching the user track sequence with the standard switching sequence, constructing a length matrix of the longest public subsequence according to a recursion relation, updating matrix elements, and obtaining the longest public subsequence according to a backtracking path.
The calculation method of the similarity specifically comprises the following steps: similarity is the length of the longest common subsequence/length of the standard sequence.
The user base station switching track is specifically a base station switching sequence generated by the position updating of a mobile communication network user in a communication network.
The standard switching sequence is a base station switching set of a target road section, and the method for extracting the standard switching sequence of the target road section specifically comprises the following steps:
s41: converting the Thiessen polygon surface layer established by the base station into a surface layer boundary line layer by means of a GIS platform;
s42: intercepting a target road section from the road section to be detected, and establishing a new target road line layer;
s43: taking a traffic operation between the surface layer boundary line layer and the target road line layer, and extracting an intersection point to obtain a key node layer of the target road section, wherein the key node is a switching point which is switched in the driving process of the target road section;
s44: and sequentially extracting the key nodes according to the road direction to form a standard switching sequence.
The track matching degree is specifically a datamation measurement value of the similarity between the user switching track and the target road section standard switching track.
The matching includes disorder matching and directional matching.
The invention has the beneficial effects that: the invention takes the switching data of the mobile communication users as the basis, combines the GIS system to extract and match the single user, then carries out speed estimation on the successfully matched user, and finally obtains the speed of the traffic flow, and the whole flow chart is shown in figure 3. The precision of subsequent calculations sets a threshold δ for trajectory similarity, where user trajectories with Sim > δ are retained. In addition, the invention is based on the traffic flow speed estimation under the coverage of the mobile communication network base station, the coverage of the base station is determined, and after the road section is selected, the coverage point of the base station can be extracted and the road length can be calculated through a GIS system.
Drawings
FIG. 1 is a Thiessen polygon of the coverage area of a base station in Beijing;
FIG. 2 is a schematic diagram of trace matching;
FIG. 3 is a flow chart of traffic flow velocity estimation;
FIG. 4 is a schematic diagram of base station handover in a road segment;
FIG. 5 is a study road segment selection graph;
FIG. 6 is a re-numbered drawing of a Xiwuhuan Xiangquan bridge-Xingshikou bridge segment base station;
figure 7 road segment 1 forward speed contrast map;
figure 8 reverse speed comparison graph for segment 1;
FIG. 9 is a top-clear bridge-home bridge segment base station re-numbered diagram;
figure 10 road segment 2 forward speed comparison graph;
fig. 11 reverse speed comparison graph for link 2.
Detailed Description
The embodiments are described in detail below with reference to the accompanying drawings.
In order to solve the problem of traffic jam and realize real-time monitoring of traffic flow speed, the method uses a Thiessen polygon to calibrate the coverage area of a base station, further extracts a standard switching sequence of a target road section through a GIS system, introduces a Needleman-Wunsch algorithm in biology to perform road matching on a user track and the target road section, determines a matching effect by using a matching speed function, and estimates the speed by using a traffic flow speed estimation method. A new idea is provided for solving the traffic problem of the mobile phone data. The specific method of the invention is as follows:
(1) determining base station coverage
The mobile communication network is also commonly referred to as a cellular communication network, and ideally, the coverage area of a mobile network base station is an inscribed regular hexagon taking the mobile network base station as a circle center. As base stations are continuously added, and the coverage density of the base stations in the hot spot area is higher in the practical application process, the regular hexagon cannot cover the whole area without overlapping. The invention takes the Thiessen polygon formed by the base station as the coverage area of the base station.
A Thiessen Polygon (Thiessen Polygon) is a continuous Polygon formed by perpendicular bisectors of two adjacent straight lines. Such a polygon has the following features:
① each discrete point belongs to only one Thiessen polygon;
② the distance from the discrete point to other points in the Thiessen polygon is shortest;
③ the distances from a point on the side of the Thiessen polygon to two adjacent discrete points are equal.
When a point falls within a Thiessen polygon, the point is closest to the center point. The spatial relationship is exactly the same as the COO (cell of origin) positioning rule, so that the method has rationality in processing the coverage area of the base station by applying the Thiessen polygon.
In the process of drawing the Thiessen polygon of the coverage area of the base station, the invention uses GIS software to process the base station, firstly reads the data of the base station, and draws the base station layer through the longitude and latitude coordinates of the base station. Based on the base station point layer, the Thiessen polygons are manufactured by using a GIS tool, the coverage range of each polygon approximately represents the coverage range of a base station network signal, and the Thiessen polygons drawn by the mobile communication base station in Beijing are shown in figure 1.
(2) Target road section switching sequence calibration
When a mobile user moves, base station switching is generated, and the user position is uploaded when the position updating condition is met, so that the motion trail of the user is obtained after the position of the user is updated and extracted within a period of time. In order to determine which road the user specifically moves on, matching needs to be performed subsequently, and the base station switching set of the target road segment is called a standard switching sequence. The normal handover sequence is further divided into a forward normal handover sequence L (1, n) { x1,x2,…,xnAnd the reverse standard switching sequence L (n,1) { xn,xn-1,…,x1And the reverse standard switching sequence is all the base station numbers which are possibly switched in the process of driving the target road section to the starting point of the target road section along the road along the end point of the road.
In order to obtain a road section standard switching sequence, a large number of road tests are mainly carried out on a target road section, a 'road test vehicle signal' obtained by a base station is analyzed, and then a high-probability switching sequence is determined according to theoretical methods such as probability theory and the like so as to achieve a better switching sequence.
The first step is as follows: converting the surface layer of the Thiessen polygon established by the base station into a surface layer boundary line layer by means of a GIS platform;
the second step is that: intercepting the research road section, intercepting a target road section and establishing a new target road line layer;
the third step: taking a traffic operation between the surface layer boundary line layer and the target road line layer, and extracting an intersection point to obtain a target road section key node layer, wherein the key nodes can be considered as switching points switched in the driving process of the target road section;
the fourth step: and sequentially extracting the key nodes according to the road direction to form a standard switching sequence.
(3) Trajectory matching algorithm
The track matching algorithm is one of map matching, and the invention converts a more complex map matching algorithm into a simpler sequence comparison algorithm in data mining on the basis of user switching tracks. The heart of a sequence alignment algorithm is to compare the differences between two given sequences (text). The trajectory matching is performed by finding the longest common subsequence of the two sequences, wherein the specific algorithm used by the present invention is the Needleman-Wunsch algorithm.
The basic idea of the Needleman-Wunsch algorithm is to find the longest identical subsequence that exists in both sequences when matching the two sequences. The longest common subsequence does not require the elements to appear consecutively, but requires that the order of appearance be consistent, e.g., the sequence X ═ { P ═ P1,P2,P3,P4}，Y＝{P1,P3,P2Then their most common subsequence is { P }1,P3}. As a basic algorithm for sequence alignment, the Needleman-Wunsch algorithm belongs to dynamically searching the longest common subsequence in a global scope, and LCS (X, Y) is used for representing the longest common subsequence of the sequence X and the sequence Y in the invention. When lenLCS (X, Y) indicates the length of sequence X, lenLCS (X, Y) means Len (LCS (X, Y)), and when lenLCS (X, Y) is 0, it indicates that there is no common portion between the two sequences. The longest common subsequence has the following structure:
let sequence X ═ X1,x2,…,xmY and the sequence Y ═ Y1,y2,…,ynOne longest common subsequence is Z, which is LCS (X, Y) Z1,z2,…,zk}, the following properties:
properties 1: if xm＝ynThen z isk＝xm＝ynAnd Z isk-1Is Xm-1And Yn-1The longest common subsequence of (a);
properties 2: if xm≠ynAnd z isk≠xmZ is Xm-1And the longest common subsequence of Y;
properties 3: if xm≠ynAnd z isk≠ynZ is X and Yn-1The longest common subsequence.
Wherein, Xm-1＝{x1,x2,…,xm-1}，Yn-1＝{y1,y2,…,yn-1}，Zk-1＝{z1,z2,…,zk-1}。
From the structural properties of the longest common subsequence problem, the requirement sequence X ═ X1,x2,…,xmY and the sequence Y ═ Y1,y2,…,ynThe longest common subsequence LCS (X, Y) of (C) has mainly two cases:
first, when xm＝ynWhen LCS (X, Y) { LCS (X)m-1,Ym-1),xm}。
Second, when xm≠ynWhen LCS (X, Y) { LCS (X)i,Yj)|max(lenLCS(Xm-1,Y),lenLCS(X,Yn-1) I ═ m-1 or j ═ n-1 }.
From this recursive relationship, it can be seen that case 2 has sub-problem overlap, when calculating Xm-1And Y and X and Yn-1When calculating the longest common subsequence of (2), all X are calculatedm-1And Yn-1The longest common subsequence. Before finding the longest common subsequence of the sequence, a recursive relationship of the optimal values of the sub-problems needs to be established. When there are two sequences X to be matched { X ═ X1,x2,……,xi}，Y＝{y1,y2,......,yjCalculating the lengths of X and Y as len (i, j) or lenLCS (X, j) defined by lenLCS (i, j) i, j, and Yi,Yj). When i ═ 0 or j ═ 0, XiAnd YjIs a null sequence, lenLCS (i, j) ═ 0. Thus, the recursive relationship is established as follows:
in the process of solving the longest common subsequence, the Needleman-Wunsch algorithm compares each element in the sequence to be matched on the basis of the recursive relationship, and expresses the sequence matching length by establishing a scoring matrix based on the lenLCS. Now assume that there are two trajectory sequences T1、T2，T1＝{P1,P2,P4,P5,P6,P7,P9}，T2＝{P1,P3,P5,P6,P7And matching the two track sequences, namely obtaining an LCS (T1, T2), firstly constructing a longest common subsequence length lenLCS matrix according to a recursive relationship, then updating matrix elements, and finally obtaining the longest common subsequence LCS by backtracking.
Step 1: a lenLCS matrix is constructed.
First step, create and initialize lenLCS matrix, T1, T2 sequence element numbers 7 and 5, respectively, so construct a matrix of 6 × 8, as in the table below, with the extra rows and columns listed as the initial LCS number.
| 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 |
The second step is that: the first row of the matrix is calculated using the core formula.
| 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 |
The third step: calculating other rows of the matrix by using a core formula to obtain the final lenLCS (T)1,T2)。
From the lenLCS matrix, the lenLCS (T) can be derived1,T2)＝lenLCS(T2,T1) lenLCS (5,7) ═ 4. T is obtained through a Needleman-Wunsch algorithm matrix1And T2Has a length of 4. In order to obtain the longest common subsequence LCS, the above matrix needs to be traced back to find the corresponding matching sequence.
Step 2: the matrix is traced back to find the longest common subsequence.
The first step is as follows: the lower right corner element of the location lenLCS matrix.
The second step is that: backtracking the cells to the top left corner of the matrix. And backtracking the matrix unit according to the backtracking rule of the lower edge.
If T1(i)＝T2(j) If yes, backtracking to the upper left cell of the current cell;
if T1(i)≠T2(j) If the cells with the same maximum value exist, the priority is sequentially selected according to the sequence of the upper edge, the left edge and the upper left edge;
if the current cell is in the first row of the matrix, backtracking to the cell on the left side;
if the current cell is in the first column of the matrix, backtracking to the top cell.
The third step: and writing out the longest common subsequence according to the backtracking path.
If the cell at the upper left corner is traced back, the cell element belongs to the longest common subsequence element, the whole matching path is searched, and the sequence LCS (T) is matched1,T2)＝{P1,P5,P6,P7Finish.
(3) Track matching algorithm implementation
And modeling and calculating the length of the longest common subsequence according to the recursive relation of the algorithm. The algorithm takes two sequences X ═ X1,x2,…,xmY ═ Y1,y2,…,ynAs input, two arrays Z [0 … m,0 … n ] are output]And L [1 … m,1 … n ]]Wherein Z [ i, j ]]In which is stored XiAnd YjOf the longest common subsequence, L [ i, j ]]Then record the current Z [ i, j ]]The relation to the last sub-problem for the subsequent construction of the longest common subsequence. Thus, the length of the longest common subsequence of X and Y is stored as Z m, n]In (1). The calculation of the lenLCS pseudo-code is as follows:
by this procedure, the length of the longest common subsequence of sequence X and sequence Y can be obtained. Following the idea of using the Needleman-Wunsch algorithm, a search is performed through the L array, starting from L [ m, n ], in the direction indicated by the arrow, to find the longest common subsequence.
When L [ i, j ]]"↖" indicates LCS (X)i,Yj)＝{LCS(Xi-1,Yj-1),xiThat is to say xi＝yj；
When L [ i, j ]]═ @ ", denotes LCS (X)i,Yj)＝LCS(Xi-1,Yj)；
When L [ i, j ]]"←", representing LCS (X)i,Yj)＝LCS(Xi,Yj-1)。
This method also searches for LCS according to the reverse order, and in order to obtain LCS, only the extraction output that satisfies the first condition is needed, and the pseudo code of the extraction process is as follows:
the longest common subsequence LCS can be found by this procedure. The calculation is performed for the example in the previous section, and the results are as follows:
starting from L (5,7), searching in turn according to the arrow direction, recording the element corresponding to each '↖', and finally obtaining LCS (T)1,T2)＝{P1,P5,P6,P7Results shown are consistent with the analysis.
(3) Calculation of degree of matching
The sequence matching degree in the invention refers to the data measurement of the Similarity of the user switching track compared with the standard switching track of the target road section, namely the Similarity (Similarity) of two sequences is calculated. Now, assuming two information to be matched a, B, the similarity between the two information is represented by Sim (a, B), generally, in the case that both a and B are the same, the similarity function Sim should satisfy the following two properties:
the property one is as follows: sim (A, B) is more than or equal to 0 and less than or equal to 1;
property II: sim (a, B) ═ Sim (B, a).
However, in the present invention, it is required that the user switching trajectory matches with the target road segment standard switching sequence, and the multiple user sequences match with a single (or two) standard sequences, and in the matching process, the matching positions of the two sequences are different. In fig. 2(a), the standard switching sequence of the target road section and the driving direction of the user trajectory are not distinguished, that is, a sort of "disorder" of the overall trajectory is equivalent to matching the multi-user to the single standard sequence. In fig. 2(b), the standard sequences have directivity while distinguishing the user's traveling direction, so that the two standard sequences are matched by multiple users.
Therefore, based on the special sequence feature of the invention with the standard sequence, a similarity calculation method based on the longest common subsequence is defined. Suppose that the target link standard switching sequence is Trs＝{Ps1,Ps2,PS3,Ps4,Ps5User switching sequence Tru＝{Pu1,Pu2,Pu3,Pu4,Pu5,Pu6The longest common subsequence of both sequences is LCS (Tr)s,Tru) Length of the longest common subsequence is measured by lenLCS (Tr)s,Tru) Indicating that if the user's driving direction is distinguished, it is necessary to distinguish TrsAfter inversion, matching with the user, such as (c) and (d) in FIG. 2, the similarity Sim (Tr) between thems,Tru) Can be expressed as:
len (Tr) in the formulas) -Standard sequence TrsLength of (d).
Although the similarity function satisfies property one of a general similarity function and does not satisfy property two, the similarity function has enough condition support for the contents of subsequent researches of the invention and belongs to a sequence similarity function based on a standard sequence.
The invention takes the switching data of the mobile communication users as the basis, combines the GIS system to extract and match the single user, then carries out speed estimation on the successfully matched user, and finally obtains the speed of the traffic flow, and the whole flow chart is shown in figure 3. The precision of subsequent calculations sets a threshold δ for trajectory similarity, where user trajectories with Sim > δ are retained. In addition, the invention is based on the traffic flow speed estimation under the coverage of the mobile communication network base station, the coverage of the base station is determined, and after the road section is selected, the coverage point of the base station can be extracted and the road length can be calculated through a GIS system, as shown in figure 4.
The invention is further illustrated by the following figures and examples. The method selects five rings in Beijing as research objects, selects two representative road sections, and the research road sections are respectively a Xiwuhuan Xiangquan bridge-apricot stone bridge section, a Beijing Wuhuan Shangqing bridge-Yuanhuangyuan bridge section as shown in figure 5. The exit and entrance of the west pentacyclic Xiangquan bridge-apricot stone bridge section are fewer, the time of road congestion is shorter, the exit and entrance of the north pentacyclic Shangqing bridge-Yuanguang bridge section are more, and the road traffic condition is greatly influenced by the peak of going to and out of work at ordinary times.
Example 1: western pentacyclic spring bridge-apricot stone bridge
The total length of the western pentacyclic Xiangquan bridge-apricot bridge section is about 4.3 kilometers, the total length spans 12 base stations, CELLIDs of the base stations are sequentially extracted from south to north and are re-numbered, and the numbers are sequentially calibrated by 001-012, as shown in FIG. 6. According to the method for acquiring the standard switching sequence of the target road section, the switching key points of the road section are extracted, and meanwhile, the direction from south to north is positive, and the direction from south to north is reverse, so that the standard switching sequence of the positive direction and the negative direction of the Xiangquan bridge-apricot stone bridge road section can be determined.
Calibrating the road section 1 according to a target road section switching sequence calibration algorithm to obtain a forward switching sequence Trsf-001,002, …,012, reverse switching sequence TrsbAnd after the target link switching sequence is obtained, calculating the traffic flow speed of the link in different time periods according to the traffic flow speed estimation process {012,011, …,001 }. The time period selected by the invention is 6: 00-20: 00, and the user position is updated by switching the base station, so that the invention adopts the mode of comparing the user and the road section in the time period every 10 minutes1 are matched and then the corresponding speed is estimated. In order to better verify the feasibility of the algorithm, the invention extracts the traffic flow speed measured by the microwave sensors on the road section 1 at the same time interval, the sampling interval of the microwave sensors is 2 minutes, although the sampling interval is inconsistent with the interval of the algorithm of the invention, the sampling interval can also be used for verifying the algorithm through the integral trend, and the data of the microwave detectors are compared with the speed estimated by the algorithm, as shown in fig. 7 and 8.
Fig. 7 shows the forward traffic flow speed of the section 1 in the research period, and fig. 8 shows the reverse traffic flow speed of the section 1 in the research period, where blue is the speed of the algorithm of the present invention, and red is the data of the microwave detection speed. Because the road section 1 is positioned in the position of the Beijing Western Wuhuan, the number of entrances and exits is small, the traffic flow is relatively stable at ordinary times and is slightly influenced by the peak in the morning and at night, and the average speed of the forward road section is basically kept above 65 km/h. By observing the reverse traffic flow speed of the road section 1, the speed is reduced to some extent before and after 14 pm, but the speed has larger fluctuation, meanwhile, an obvious singular point appears between 12:00 and 13:00 in the algorithm, and analysis and research show that fewer users successfully matched in the time period and lower speeds of some users exist, so that the integral average speed is reduced, but the speed obtained by the algorithm is basically consistent with the microwave speed in terms of the integral trend.
Example 2: north pentacyclic Shangqing bridge-Lai Guang Yingqiang bridge
The total length of the whole length of the north five-ring supernatant bridge-come-wide operation bridge section is about 8.8 kilometers, 25 base stations are spanned, CELLIDs of the base stations are sequentially extracted from west to east and are re-numbered, and numbers are sequentially calibrated from 101 to 125, as shown in fig. 9. According to the method for acquiring the standard switching sequence of the target road section, the switching key points of the road section are extracted, and meanwhile, the direction from west to east is a forward direction, and the reverse direction is a reverse direction, so that the standard switching sequence of the forward and reverse directions of the supernatant bridge-come and wide-run bridge road section can be determined.
Calibrating the road section 1 according to the calibration algorithm of the target road section switching sequenceObtaining a forward switching sequence Trsf-101,002, …,125, reverse switching sequence TrsbAnd after the target link switching sequence is obtained, calculating the traffic flow speed of the link in different time periods according to the traffic flow speed estimation process {125,124, …,001 }. The method is the same as the road section 1, the selected time period is 6: 00-20: 00, matching is carried out on the user and the road section 2 in the time period every 10 minutes, and then the corresponding speed is calculated. Similarly, the present invention extracts the traffic flow speed measured by the microwave sensor at the same time interval on road section 2, the sampling interval of the microwave sensor is 2 minutes, and the data of the microwave detector is compared with the speed estimated by the algorithm as shown in fig. 10 and 11.
Fig. 10 shows the forward traffic flow speed of the section 2 during the research period, and fig. 11 shows the reverse traffic flow speed of the section 2 during the research period, where blue is the speed of the algorithm of the present invention, and red is the data of the microwave detection speed. As the road section 2 is positioned at the Beijing North five-ring position, the number of access and highway connection points is large, the fluctuation of traffic flow in the whole day is large, the influence of the peak in the morning and at night is large, the average speed of the forward road section continuously drops in the early peak period, the lowest point appears at about 9:30, about 45km/h is formed, the influence is realized in the late peak period, but the speed is always maintained at 55 km/h. The algorithm of the invention still keeps good consistency with the microwave speed under the condition of larger fluctuation. When the traffic flow speed of the road section 2 in the reverse direction is observed, although the average speed is always over 60km/h in the early peak period, the fluctuation of the traffic flow speed is large under the influence of the late peak, the fluctuation of the algorithm of the invention is larger in the period, and the speed obtained by the algorithm of the invention is basically consistent with the microwave speed in terms of the whole trend.
The first part of the algorithm of the invention is qualitatively analyzed, and the algorithm is quantitatively analyzed in order to more accurately prove the feasibility of the algorithm. Because the time interval of the estimation of the bidirectional switching speed of the base station used by the invention is 10min, and the sampling interval of the microwave speed measurement is 2min, in order to better compare with the microwave speed measurement, the invention carries out spline interpolation on the bidirectional switching speed of the base station to achieve the same data interval, and utilizes the absolute error e, the error percentage PE and the mean square error MSE to test the algorithm.
e＝|vh-vm|
In the formula vh-the base station double handover calculation speed;
vm-the microwave detector detects the speed;
n-experiment number.
The speed of the data of the two target road sections 06: 00-20: 00 is estimated, the obtained result is subjected to error analysis, and the final result is shown in table 1.
TABLE 1 traffic flow velocity comparison results
From the statistical results in table 1, we can see that the traffic flow speed estimation method based on the bidirectional switching of the mobile communication network base station has higher accuracy. The absolute errors of the road section 1 and the road section 2 are all below 5.5km/h, the error percentage of the forward direction of the road section 1 and the error percentage of the road section 2 are about 5%, the mean square error is below 5, and the reverse error percentage of the road section 1 is slightly higher than that of the other three. The sources of the overall error mainly include calculation error and measurement error. The calculation error is mainly that a certain error exists between the distance between the base station switching key points and the speed converted by the switching time difference. Since the distribution of the traffic flow in space is not uniform and has a certain spatial characteristic, the speed simply calculated by the ratio has an error to some extent. The measuring speed mainly comprises two parts, wherein one part is that certain measuring errors exist between the distance measured by the switching node and the actual distance through GIS software, and the other part is that errors also exist in the speed of the road section detected by the microwave detector.
Through the comparative analysis between the road section 1 reversal direction and other three parts, the reason that the precision is slightly low is that in the track matching process, users passing through the matching algorithm are slightly fewer, the speed difference with the microwave detector is larger, and therefore the error is higher. The road section 2 is a road section with more inlets and outlets, has larger traffic flow than the road section 1, has more matched users, and therefore, the overall calculated speed error is low. However, the errors in both cases are within an acceptable range, so the traffic flow speed estimation method based on base station bidirectional switching provided by the invention is effective.
The invention relates to a method for acquiring traffic flow speed by using a mobile phone as a mobile sensor and by using mobile communication switching data of the mobile phone, belonging to the technical field of traffic transportation. Map matching is carried out by using a method of solving the longest public subsequence by using a user track and a target road section switching sequence, and the traffic flow speed is obtained by using the ratio of the distance of a switching point to the travel time. The method for determining the coverage area of the base station uses Thiessen polygons for calibration, further extracts a target road section standard switching sequence through a GIS platform, determines the road length, matches a mobile phone user to a corresponding road section by using different map matching algorithms, introduces a Needleman-Wunsch algorithm in biology to construct a lenLCS matrix, backtracks the matrix to find a corresponding matching sequence, determines the sequence matching degree, namely similarity datamation measurement of comparing a user switching track with the target road section standard switching track, and further obtains the speed through a traffic flow speed obtaining method. By comparing the microwave detector with the traffic flow speed detection example, the method for acquiring the traffic flow speed based on the position switching can effectively acquire the road section traffic flow speed.
The present invention is not limited to the above embodiments, and any changes or substitutions that can be easily made by those skilled in the art within the technical scope of the present invention are also within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (5)
1. A method for obtaining traffic flow speed by utilizing mobile phone signaling data position switching is characterized by comprising the following steps:
s1: taking a Thiessen polygon formed by a base station as a coverage area of the base station;
s2: constructing a forward switching sequence and a reverse switching sequence of a target road section, and acquiring a road section distance set and a reverse distance set under the coverage of a single base station by using a GIS platform; the forward switching sequence is a mark number covering the base station from the starting point to the end point of the target road section, and the reverse switching sequence is a mark number covering the base station from the end point to the starting point of the target road section;
s3: extracting a switching track of the user base station based on the label and the switching time of the base station;
s4: carrying out track matching degree verification on the user track sequence and the standard switching sequence;
s5: acquiring the speed of the successfully matched user;
s6: performing steps S3 to S5 for all users, finding a traffic flow speed of the target link;
the track matching degree verification specifically comprises the following steps: solving the longest public subsequence of the user track sequence and the standard switching sequence and the similarity of the longest public subsequence, comparing the similarity with a similarity threshold, and if the similarity passes the threshold, successfully matching, otherwise, failing;
the calculation method of the longest public subsequence specifically comprises the following steps: matching the user track sequence with the standard switching sequence, constructing a length matrix of the longest public subsequence according to a recursion relation, updating matrix elements, and obtaining the longest public subsequence according to a backtracking path;
the calculation method of the similarity specifically comprises the following steps: similarity is the length of the longest common subsequence/length of the standard sequence.
2. The method according to claim 1, wherein the subscriber base station handover trajectory is a base station handover sequence of a mobile communication network subscriber in the communication network due to location update.
3. The method according to claim 1, wherein the standard handover sequence is a handover set of base stations of a target road segment, and the method for extracting the standard handover sequence of the target road segment specifically comprises:
s41: converting the Thiessen polygon surface layer established by the base station into a surface layer boundary line layer by means of a GIS platform;
s42: intercepting a target road section from the road section to be detected, and establishing a new target road line layer;
s43: taking a traffic operation between the surface layer boundary line layer and the target road line layer, and extracting an intersection point to obtain a key node layer of the target road section, wherein the key node is a switching point which is switched in the driving process of the target road section;
s44: and sequentially extracting the key nodes according to the road direction to form a standard switching sequence.
4. The method according to claim 1, wherein the track matching degree is a digitalized metric value of similarity between the user switching track and the target road section standard switching track.
5. The method of claim 1 or 4, wherein the matching comprises disorder matching and directional 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 CN106997666A (en) | 2017-08-01 |
| CN106997666B true 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) |
Families Citing this family (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 |
| 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 |
| 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 |
| CN110047277B (en) * | 2019-03-28 | 2021-05-18 | 华中科技大学 | Urban road traffic jam ranking 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 |
| 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 |
| 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 |
| CN113709006B (en) * | 2021-10-29 | 2022-02-08 | 上海闪马智能科技有限公司 | Flow determination method and device, storage medium and electronic device |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2003199154A (en) * | 2001-12-21 | 2003-07-11 | Fujitsu Ten Ltd | Communication means switching method |
| CN102722984B (en) * | 2012-06-19 | 2015-01-28 | 张家港市鸿嘉数字科技有限公司 | Real-time road condition monitoring method |
| 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 (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
Non-Patent Citations (2)
| Title |
|---|
| "基于位置数据的交通流速度分析应用";高瑞;《中国优秀硕士论文学位论文全文数据库 信息科技辑》;20120715;正文全文 * |
| "基于手机信令技术的区域交通出行特征研究";毛晓汶 等;《中国优秀硕士论文全文数据库 工程科技Ⅱ辑》;20150415;正文全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106997666A (en) | 2017-08-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106997666B (en) | Method for obtaining traffic flow speed by switching positions of mobile phone signaling data | |
| Mohamed et al. | Accurate real-time map matching for challenging environments | |
| Yuan et al. | An interactive-voting based map matching algorithm | |
| CN105809292B (en) | Bus IC card passenger getting off car website projectional technique | |
| Gong et al. | Data selection in machine learning for identifying trip purposes and travel modes from longitudinal GPS data collection lasting for seasons | |
| CN101866545B (en) | Method for acquiring road network matching track of mobile object | |
| CN113724489B (en) | Traffic jam tracing method based on multi-source data | |
| CN106454722A (en) | Dynamic differential positioning method based on map matching for mobile communication terminals | |
| CN106597363A (en) | Pedestrian location method in indoor WLAN environment | |
| CN106574975A (en) | Trajectory matching using peripheral signal | |
| CN104330089A (en) | Map matching method by use of historical GPS data | |
| CN104504099A (en) | Position-trajectory-based travel state splitting method | |
| Dumbliauskas et al. | Application of Google-based data for travel time analysis: Kaunas city case study | |
| US11049390B2 (en) | Method, apparatus, and system for combining discontinuous road closures detected in a road network | |
| CN105206057A (en) | Detection method and system based on floating car resident trip hot spot regions | |
| CN111189459A (en) | Method and device for matching positioning information with road | |
| CN104200104A (en) | Fine granularity air pollutant concentration area estimation method based on spatial characteristics | |
| CN104990551A (en) | Indoor and outdoor co-navigation method and system based on trajectory extraction | |
| CN108592903B (en) | Vehicle geomagnetic matching positioning method based on road network geomagnetic reference library | |
| CN111024098A (en) | Motor vehicle path fitting algorithm based on low-sampling data | |
| US20180307729A1 (en) | Method and apparatus for detecting a quality weather provider, weather station, or weather report | |
| CN111954160A (en) | Method for converting two-dimensional mobile phone signaling data into three-dimensional space trajectory data | |
| CN112556717A (en) | Travel mode screening method and travel route recommending method and device | |
| CN113077089B (en) | Method and device for evaluating influence of multiple factors on air quality | |
| Liu et al. | Determination of routing velocity with GPS floating car data and webGIS-based instantaneous traffic information dissemination |
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 |
Granted publication date: 20200703 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |