CN106781479A - A kind of method for obtaining highway running status in real time based on mobile phone signaling data - Google Patents

Abstract

The present invention relates to a kind of method for obtaining highway running status in real time based on mobile phone signaling data, the method delimit highw ay m onitoring region first, and carry out rasterizing treatment to monitored area；Raster series in grid system are mapped to according to what the grid system that establishes obtained highway；Base station information in monitored area is then obtained, positional information and grid system according to base station obtain grid where each base station, and obtain the road grid that base station maps are arrived according to base station range；Set up base station information table；Then the signaling data produced in monitored area in the monitoring time period is obtained, the track sets that edit obtains each user is carried out to signaling data；User trajectory and its historical track according to the monitoring time period judge whether it travels within the monitoring time period on expressway；Obtain monitoring the time period interior user for being travelled on expressway and producing new track sets data accordingly；The signaling data updated followed by these users calculates road grid real time execution speed.The present invention is without extra hardware device, and its implementation is simple, and the degree of accuracy is higher.

Description

A kind of method for obtaining highway running status in real time based on mobile phone signaling data

Technical field

The invention belongs to highway real-time road monitoring technical field, it is related to one kind to be obtained in real time based on mobile phone signaling data The method for taking highway running status.

Background technology

With being continuously increased for China's vehicle fleet size, highway vehicle flowrate constantly increases, and causes expressway traffic accident frequently Hair, congestion occurrence frequency more and more higher.The collection of current traffic state information is main to use fixed sensor.As tachymeter, Vehicle checker, shooting are first-class.And the expense of the cost, installation and maintenance of these hardware facilities is of a relatively high, and its monitoring range has Limit, it is impossible to cover whole piece highway.Therefore its exigence one kind can monitor whole highway network, and lower-cost Method.

Mobile phone signaling be it is a kind of result from mobile communications network and can reflect that cellphone subscriber occurs this communication behavior where Geographical position and the data of time.Such data contain cellphone subscriber's travel behaviour feature, can be not only used for analysis and use The trip mode and trip preference at family, moreover it is possible to by analyzing user's sample of colony, obtain the real-time running state of highway.

The content of the invention

In view of this, highway operation is obtained based on mobile phone signaling data in real time it is an object of the invention to provide a kind of The method of state, the method analyzes the signaling data of Real-time Collection using correlation technique, recognizes the user travelled on expressway； And the signaling data analytical calculation produced in monitoring cycle using user obtains express highway section average running speed, according to The road-section average speed of service obtains highway real-time road.

To reach above-mentioned purpose, the present invention provides following technical scheme：

A kind of method for obtaining highway running status in real time based on mobile phone signaling data, the method includes following step Suddenly：

Step 1：According to highway actual path, rectangle monitored area delimited；The square grid of L is using length and width Rasterizing treatment is carried out to monitored area；With the monitored area lower left corner as starting point, with geographical horizontal line as transverse axis, vertical line is the longitudinal axis Divide grid；Grid is expressed as B [x, y], and wherein x is line number where this grid, and y is columns where this grid；If highway A certain travel direction is forward direction, if the grid containing express highway section is road grid, thus obtains highway forward direction road Road raster series L_HB={ B [h₁,w₁],B[h₂,w₂],B[h₃,w₃],…,B[h_i,w_i],…,B[h_n,w_n], wherein B [h_i,w_i] It is a road grid, h_iThe line number where this grid, w_iThe columns where this grid；

Step 2：The base station data in monitored area is obtained, and the grating map determined according to step 1, set up base station information Table；Base station information table (is base comprising base station location area numbering LAC, base station cell numbering CellID, base station geographic position information Stand latitude and longitude coordinates, referred to as LOC), affiliated grid (referred to as CB), whether cover highway (referred to as IOW), corresponding Road grid (referred to as CWB)；

Step 3：If T moment user profile collection is combined into S_U, wherein user profile comprising user's unique mark UID and this Track sets of the user before the T moment；All signaling datas that monitored area produces within T to the T+t monitoring time periods are obtained, Wherein signaling data is numbered comprising user's unique mark UID, signaling data own base station position area numbering LAC, base station cell CellID and time stamp T imeStamp fields；Then by signaling data classify by user and to the data of each user temporally Stamp TimeStamp is ranked up, and filters pingpang handoff, and co-located data are carried out in continuous time section by user Merge, so as to build track sets of the user within the monitoring time period；User profile set S_U is then updated according to UID, if S_ There is this user in U, then update the track sets of this user；If not existing this user in S_U, to adding this user in S_U Information, thus obtains the user profile set S_U1 that track data updates within the monitoring time period；

Step 4：It is highway user to be located at the user travelled on expressway, obtains T moment highway user set S_HU (wherein S_HU is contained in S_U)；Track sets to each user in S_U1 carry out effectively trip section segmentation, and analyze it most The latter section of track sets of section of effectively going on a journey, judge whether it travels on expressway, and updating maintenance highway is used accordingly Family set S_HU and generation calculate each road grid real-time speed of highway needed for positive track sets set S_PE and negative To track sets set S_NE；

Step 5：Each track sets in S_PE and S_NE are extracted, the positive negative sense that analytical calculation obtains each road grid is fast in real time Degree；Wherein road grid real-time speed is the express highway section real-time speed that road grid is included, wherein positive negative direction road Road grid real-time speed computational methods are consistent.

Further, in step 2, the affiliated grid CB in base station is determined according to base station LOC first, model is then covered according to base station Enclose, determine the grid set of base station covering；If containing road grid in the grid set of base station covering, by this base station IOW category Property value is set to iw, and this IOW property value otherwise is set into ow；Wherein iw represents that this base station range contains highway, ow tables Show that this base station can not cover highway；If base station IOW property values are iw, road grid and its place that calculation base station is covered The distance of grid CB, it is the grid CWB in base station maps to road to take the minimum road grid of distance；If base station IOW property values It is ow, then its CWB value is null；If grid where certain base station is B [a, b], the road grid set of its covering is S_CHB= {B[c₁,d₁],B[c₂,d₂],B[c₃,d₃],…,B[c_i,d_i],…,B[c_m,d_m]}；Calculate each grid and base station place in S_CHB Grid B [a, b] apart from S_D={ D_c1d1,D_c2d2,D_c3d3,…,D_cidi,…,D_cmdm(wherein D_cidiIt is B [c_i,d_i] and B [a, b] The distance between, D_cidi=((c_i-a)²+(d_i-b)²)^(1/2)), the minimum grid of S_D intermediate values is chosen for base station maps arrive road Grid.

Further, in step 4, specifically include：

Step 4.1：The track sets of each user in S_U1 are extracted, and carries out effectively trip section segmentation；Used in extraction S_U1 Family final stage is effectively gone on a journey section, if this effectively trip section signaling number more than certain threshold value if jump to step 4.2, otherwise judge Whether there is this user in S_HU, if in the presence of this user, it is being monitored into the track in the time period according to user's travel direction Sequence is added in corresponding track sets set；

Step 4.2：Obtain effectively trip section track sets L_P={ P₁,P₂,P₃,…,P_i,…,P_u(wherein P_iRepresent rail I-th signaling data in mark sequence, in chronological sequence order is arranged between each signaling), the base station information table according to step 2 is obtained The corresponding road grid CWB in Pi base stations, thus obtains the corresponding road raster series L_PCWB={ CWB of track sets₁,CWB₂, CWB₃,…,CWB_i,…,CWB_u(wherein CWB_iIt is P_iCorresponding CWB, CWB_iMay be null)；

Step 4.3：Calculate the accounting of the CWB of non-null in L_PCWB sequences；Value is null in obtaining L_PCWB sequences CWB number N1, and the CWB of the non-null of value number N2；Calculate the accounting R1=N2/ (N1+ of the CWB of non-null N2)；Step 4.4 is gone to if R1 is more than certain threshold value, this user non-freeway user is otherwise assert, step 4.5 is jumped to；

Step 4.4：Value is the road grid of the corresponding non-NULLs of CWB composition track sets P of non-null in extraction L_PCWB Lattice sequence L_PCWB2={ CWB_g1,CWB_g2,CWB_g3,…,CWB_gi,…,CWB_gq(wherein L_PCWB2 is contained in L_PCWB, g_i∈ [1,u])；Extract CWB_giSequence number O in Expressway Road raster series L_HB_gi, then corresponding sequence number sequences of L_PCWB2 It is L_O={ O_g1,O_g2,O_g3,…,O_gi,…,O_gq}；Calculate the average value Agi=(O of two neighboring element in L_O_gi+O_g(i+1))/ 2, then the sequence after L_O smoothing processings is L_OA={ A_g1,A_g2,A_g3,…,A_gi,…,A_g(q-1)}；If L_OA sequences exist bright Aobvious growth trend, then assert that the moving direction of this user is matched and it is highway user with highway；If increasing Gesture is for just, then it is assumed that this user's moving direction is highway positive direction, and its track sets within the monitoring time period is added It is added in S_PE；If growth trend is negative, then it is assumed that this user's moving direction is highway negative direction, and by it in monitoring Between track sets in section be added in S_NE；

Step 4.5：Update highway user set S_HU；Obtain T moment highway user collection and be combined into S_HU, if this User is that highway user and S_HU do not exist this user, then add this user in S_HU；If this user is judged as not high There is this user in fast highway user and S_HU, then this user is deleted from S_HU.

Further, in steps of 5, so that positive direction speed is asked for as an example, its step includes：

Step 5.1：It is that each road grid sets corresponding speed container in L_HB, for depositing mobile phone in each track sets Translational speed of the user between two tracing points；Speed is divided into by Z grade according to the path length between two tracing points, then often Z speed container is included in the speed container of individual road grid, each speed container deposits corresponding speed data；

Step 5.2：Translational speed of the user between two tracing points in each track sets in calculating S_PE, according to two tracks Distance between point carries out grade classification to the speed being calculated, and puts it into the corresponding speed appearance of corresponding road grid In device；

Step 5.3：The speed data in each speed container using clustering algorithm to each road grid in L_HB is gathered Alanysis, obtains the corresponding speed of each road grid difference distance grade；If road grid B [h_i,w_i] corresponding speed sets of containers It is combined into S_B [h_i,w_i] _ Con={ B [h_i,w_i]_C₁,B[h_i,w_i]_C₂,B[h_i,w_i]_C₃,…,B[h_i,w_i]_C_j,…,B[h_i, w_i]_C_z, calculate B [h using focusing solutions analysis_i,w_i]_C_jMiddling speed degrees of data obtains the corresponding speed of grade j for Con_ Speed_j, then road grid B [h_i,w_i] different brackets sets of speeds be S_Con_Speed={ Con_Speed₁,Con_Speed₂, Con_Speed₃,…,Con_Speed_j,…,Con_Speed_z}；Each speed container medium velocity data sample number is obtained, then road Grid B [h_i,w_i] different brackets speed sample number collection is combined into S_Con_NS={ NS₁,NS₂,NS₃,…,NS_j,…NS_z}；Then road Grid B [h_i,w_i] speed total sample number B [h_i,w_i] _ SpeedSwatch_Sum=NS₁+NS₂+NS₃+…+NS_j+…+NS_z；Utilize Formula below calculates road grid B [h_i,w_i] real-time speed B [h_i,w_i]_RTSpeed：

Step 5.4：Real time execution speed according to road grid determines the traffic behavior of express highway section therein.

Further, in the step 5.2, comprise the following steps that：

Step 5.2.1：Obtain a certain track sets L_EP={ R in S_PE₁,R₂,R₃,…,R_i,…,R_k(wherein R_iRepresent I-th tracing point in this track sets)；Each tracing point in order traversal L_EP, calculate cellphone subscriber itself and follow-up each point it Between translational speed, and the speed that will be calculated is included into the speed container of corresponding road grid；

Step 5.2.2：If tracing point R_i-1And R_iCorresponding road grid is respectively B [e_(i-1),f_(i-1)] and B [e_i,f_i], Obtain B [e_(i-1),f_(i-1)] and B [e_i,f_i] sequence number in L_HB, if the difference of two sequence numbers is all just with the moving direction of this user Or be all negative, then jump to 5.2.3；

Step 5.2.3：Calculate B [e_(i-1),f_(i-1)] and B [e_i,f_i] between distance Dis_(i-1)i；Obtain B in L_HB [e_(i-1),f_(i-1)] and B [e_i,f_i] between subsequence L_EHB={ B [s₁,l₁],B[s₂,l₂],…,B[s_i,l_i],…,B[s_v, l_v], then B [e_(i-1),f_(i-1)] and B [e_i,f_i] between distance Dis_(i-1)iCan be calculated with formula below；

Step 5.2.4：If tracing point R_i-1And R_iCorresponding time stamp T imeStamp_(i-1)And TimeStamp_i, then user In R_i-1And R_iBetween translational speed Speed_(i-1)i=Dis_(i-1)i/(TimeStamp_i-TimeStamp_(i-1))；

Step 5.2.5：According to the distance Dis that step 5.2.3 is obtained_(i-1)iBy Speed_(i-1)iIt is put into B [e_(i-1),f_(i-1)] With B [e_i,f_i] between subsequence L_EHB in the corresponding speed container of each road grid in；If Speed_(i-1)iIt is corresponding Grade is Level_Speed_(i-1)i, rate sequence is L_Level={ Level₁,Level₂,Level₃,…,Level_j,…, Level_z(wherein Level_jRepresent j-th speed class)；Then Level_Speed can be determined by below equation_(i-1)i；If being Dis_(i-1)iMore than or equal to DisNum_jAnd less than DisNum_j+1, then Level_Speed_(i-1)iIt is Level_j；Wherein DisNum_j+1Greatly In DisNum_jIf, Dis_(i-1)iIn not interval any one, then this speed data is given up；

The beneficial effects of the present invention are：Signaling data by analyzing Real-time Collection of the invention, recognizes on expressway The user of traveling；And the signaling data analytical calculation produced in monitoring cycle using user is obtained express highway section and averagely transported Scanning frequency degree, highway real-time road is obtained according to the road-section average speed of service；The present invention without extra hardware device, and its Implementation method is simple, and the degree of accuracy is higher.

Brief description of the drawings

In order that the purpose of the present invention, technical scheme and beneficial effect are clearer, the present invention provides drawings described below and carries out Explanation：

A kind of method for obtaining highway running status in real time based on mobile phone signaling data that Fig. 1 is provided for the present invention Flow.

Specific embodiment

Below in conjunction with accompanying drawing, the preferred embodiments of the present invention are described in detail.

A kind of method for obtaining highway running status in real time based on mobile phone signaling data that Fig. 1 is provided for the present invention Flow, as illustrated, the method specifically includes following steps：

Step 1：According to highway actual path, rectangle monitored area delimited.The square grid of L is using length and width Rasterizing treatment is carried out to monitored area.With the monitored area lower left corner as starting point, with geographical horizontal line as transverse axis, vertical line is the longitudinal axis Divide grid.Grid is expressed as B [x, y], and wherein x is line number where this grid, and y is columns where this grid.If highway A certain travel direction is forward direction, if the grid containing express highway section is road grid, thus obtains highway forward direction road Road raster series L_HB={ B [h₁,w₁],B[h₂,w₂],B[h₃,w₃],…,B[h_i,w_i],…,B[h_n,w_n], wherein B [h_i,w_i] It is a road grid, h_iThe line number where this grid, w_iThe columns where this grid.

Step 2：The base station data in monitored area is obtained, and the grating map determined according to step 1, set up base station information Table.Base station information table (is base comprising base station location area numbering LAC, base station cell numbering CellID, base station geographic position information Stand latitude and longitude coordinates, referred to as LOC), affiliated grid (referred to as CB), whether cover highway (referred to as IOW), corresponding Road grid (referred to as CWB).The affiliated grid CB in base station is determined according to base station LOC first, then according to base station range, really Determine the grid set of base station covering.If containing road grid in the grid set of base station covering, this base station IOW property values are set It is iw, this IOW property value is otherwise set to ow.Wherein iw represents that this base station range contains highway, and ow represents this base Standing can not cover highway.If base station IOW property values are iw, the road grid and grid CB where it of calculation base station covering Distance, it is the grid CWB in base station maps to road to take the minimum road grid of distance.If base station IOW property values are ow, Its CWB value is null.If grid where certain base station is B [a, b], the road grid set of its covering is S_CHB={ B [c₁, d₁],B[c₂,d₂],B[c₃,d₃],…,B[c_i,d_i],…,B[c_m,d_m]}.Calculate each grid and grid B where base station in S_CHB [a, b] apart from S_D={ D_c1d1,D_c2d2,D_c3d3,…,D_cidi,…,D_cmdm(wherein D_cidiIt is B [c_i,d_i] Yu B [a, b] between Distance, D_cidi=((c_i-a)²+(d_i-b)²)^(1/2)), it is the grid of base station maps to road to choose the minimum grid of S_D intermediate values Lattice.

Step 3：If T moment user profile collection is combined into S_U, wherein user profile comprising user's unique mark UID and this Track sets of the user before the T moment.All signaling datas that monitored area produces within T to the T+t monitoring time periods are obtained, Wherein signaling data is numbered comprising user's unique mark UID, signaling data own base station position area numbering LAC, base station cell CellID and time stamp T imeStamp fields.Then by signaling data classify by user and to the data of each user temporally Stamp TimeStamp is ranked up, and filters pingpang handoff, and co-located data are carried out in continuous time section by user Merge, so as to build track sets of the user within the monitoring time period.User profile set S_U is then updated according to UID, if S_ There is this user in U, then update the track sets of this user.If not existing this user in S_U, to adding this user in S_U Information.Thus the user profile set S_U1 that track data updates within the monitoring time period is obtained.

Step 4：It is highway user to be located at the user travelled on expressway, obtains T moment highway user set S_HU (wherein S_HU is contained in S_U).Track sets to each user in S_U1 carry out effectively trip section segmentation, and analyze it most The latter section of track sets of section of effectively going on a journey, judge whether it travels on expressway, and updating maintenance highway is used accordingly Family set S_HU and generation calculate each road grid real-time speed of highway needed for positive track sets set S_PE and negative To track sets set S_NE.Comprise the following steps that：

Step 4.1：The track sets of each user in S_U1 are extracted, and carries out effectively trip section segmentation.Used in extraction S_U1 Family final stage is effectively gone on a journey section, if this effectively trip section signaling number more than certain threshold value if jump to step 4.2, otherwise judge Whether there is this user in S_HU, if in the presence of this user, it is being monitored into the track in the time period according to user's travel direction Sequence is added in corresponding track sets set.

Step 4.2：Obtain effectively trip section track sets L_P={ P₁,P₂,P₃,…,P_i,…,P_u(wherein P_iRepresent rail I-th signaling data in mark sequence, in chronological sequence order is arranged between each signaling), the base station information table according to step 2 is obtained The corresponding road grid CWB in Pi base stations, thus obtains the corresponding road raster series L of track sets_{_}PCWB={ CWB₁,CWB₂, CWB₃,…,CWB_i,…,CWB_u(wherein CWB_iIt is P_iCorresponding CWB, CWB_iMay be null).

Step 4.3：Calculate the accounting of the CWB of non-null in L_PCWB sequences.Value is null in obtaining L_PCWB sequences CWB number N1, and the CWB of the non-null of value number N2.Calculate the accounting R1=N2/ (N1+ of the CWB of non-null N2).Step 4.4 is gone to if R1 is more than certain threshold value, this user non-freeway user is otherwise assert, step 4.5 is jumped to.

Step 4.4：Value is the road grid of the corresponding non-NULLs of CWB composition track sets P of non-null in extraction L_PCWB Lattice sequence L_PCWB2={ CWB_g1,CWB_g2,CWB_g3,…,CWB_gi,…,CWB_gq(wherein L_PCWB2 is contained in L_PCWB, g_i∈ [1,u]).Extract CWB_giSequence number O in Expressway Road raster series L_HB_gi, then corresponding sequence number sequences of L_PCWB2 It is L_O={ O_g1,O_g2,O_g3,…,O_gi,…,O_gq}.Calculate the average value Agi=(O of two neighboring element in L_O_gi+O_g(i+1))/ 2, then the sequence after L_O smoothing processings is L_OA={ A_g1,A_g2,A_g3,…,A_gi,…,A_g(q-1)}.If L_OA sequences exist bright Aobvious growth trend, then assert that the moving direction of this user is matched and it is highway user with highway.If increasing Gesture is for just, then it is assumed that this user's moving direction is highway positive direction, and its track sets within the monitoring time period is added It is added in S_PE.If growth trend is negative, then it is assumed that this user's moving direction is highway negative direction, and by it in monitoring Between track sets in section be added in S_NE.

Step 4.5：Update highway user set S_HU.Obtain T moment highway user collection and be combined into S_HU, if this User is that highway user and S_HU do not exist this user, then add this user in S_HU.If this user is judged as not high There is this user in fast highway user and S_HU, then this user is deleted from S_HU.

Step 5：Each track sets in S_PE and S_NE are extracted, the positive negative sense that analytical calculation obtains each road grid is fast in real time Degree.Wherein road grid real-time speed is the express highway section real-time speed that road grid is included.Wherein positive negative direction road Road grid real-time speed computational methods are consistent.Below by taking the calculating of positive direction road grid real-time speed as an example, its specific steps is such as Under：

Step 5.1：It is that each road grid sets corresponding speed container in L_HB, for depositing mobile phone in each track sets Translational speed of the user between two tracing points.And speed is divided into by Z grade according to the path length between two tracing points again, Z speed container is then included in the speed container of each road grid, each speed container deposits corresponding speed data.

Step 5.2：Translational speed of the user between two tracing points in each track sets in calculating S_PE, according to two tracks Distance between point carries out grade classification to the speed being calculated, and puts it into the corresponding speed appearance of corresponding road grid In device.Comprise the following steps that：

Step 5.2.1：Obtain a certain track sets L_EP={ R in S_PE₁,R₂,R₃,…,R_i,…,R_k(wherein R_iRepresent I-th tracing point in this track sets).Each tracing point in order traversal L_EP, calculate cellphone subscriber itself and follow-up each point it Between translational speed, and the speed that will be calculated is included into the speed container of corresponding road grid.

Step 5.2.2：If tracing point R_i-1And R_iCorresponding road grid is respectively B [e_(i-1),f_(i-1)] and B [e_i,f_i], Obtain B [e_(i-1),f_(i-1)] and B [e_i,f_i] sequence number in L_HB, if the difference of two sequence numbers is all just with the moving direction of this user Or be all negative, then jump to 5.2.3.

Step 5.2.3：Calculate B [e_(i-1),f_(i-1)] and B [e_i,f_i] between distance Dis_(i-1)i.Obtain B in L_HB [e_(i-1),f_(i-1)] and B [e_i,f_i] between subsequence L_EHB={ B [s₁,l₁],B[s₂,l₂],…,B[s_i,l_i],…,B[s_v, l_v], then B [e_(i-1),f_(i-1)] and B [e_i,f_i] between distance Dis_(i-1)iCan be calculated with formula below.

Step 5.2.4：If tracing point R_i-1And R_iCorresponding time stamp T imeStamp_(i-1)And TimeStamp_i, then user In R_i-1And R_iBetween translational speed Speed_(i-1)i=Dis_(i-1)i/(TimeStamp_i-TimeStamp_(i-1))。

Step 5.2.5：According to the distance Dis that step 5.2.3 is obtained_(i-1)iBy Speed_(i-1)iIt is put into B [e_(i-1),f_(i-1)] With B [e_i,f_i] between subsequence L_EHB in the corresponding speed container of each road grid in.If Speed_(i-1)iIt is corresponding Grade is Level_Speed_(i-1)i, rate sequence is L_Level={ Level₁,Level₂,Level₃,…,Level_j,…, Level_z(wherein Level_jRepresent j-th speed class).Then Level_Speed can be determined by below equation_(i-1)i.If being Dis_(i-1)iMore than or equal to DisNum_jAnd less than DisNum_j+1, then Level_Speed_(i-1)iIt is Level_j.Wherein DisNum_j+1Greatly In DisNum_jIf, Dis_(i-1)iIn not interval any one, then this speed data is given up.

Step 5.3：The speed data in each speed container using clustering algorithm to each road grid in L_HB is gathered Alanysis, obtains the corresponding speed of each road grid difference distance grade.If road grid B [h_i,w_i] corresponding speed sets of containers It is combined into S_B [h_i,w_i] _ Con={ B [h_i,w_i]_C₁,B[h_i,w_i]_C₂,B[h_i,w_i]_C₃,…,B[h_i,w_i]_C_j,…,B[h_i, w_i]_C_z, calculate B [h using focusing solutions analysis_i,w_i]_C_jMiddling speed degrees of data obtains the corresponding speed of grade j for Con_ Speed_j, then road grid B [h_i,w_i] different brackets sets of speeds is：

S_Con_Speed={ Con_Speed₁,Con_Speed₂,Con_Speed₃,…,Con_Speed_j,…,Con_ Speed_z}.Each speed container medium velocity data sample number is obtained, then road grid B [h_i,w_i] different brackets speed sample manifold It is combined into S_Con_NS={ NS₁,NS₂,NS₃,…,NS_j,…NS_z}.Then road grid B [h_i,w_i] speed total sample number B [h_i,w_i]_ SpeedSwatch_Sum=NS₁+NS₂+NS₃+…+NS_j+…+NS_z.Road grid B [h are calculated using formula below_i,w_i] in real time Speed B [h_i,w_i]_RTSpeed。

Step 5.4：Real time execution speed according to road grid determines the traffic behavior of express highway section therein.

Finally illustrate, preferred embodiment above is merely illustrative of the technical solution of the present invention and unrestricted, although logical Cross above preferred embodiment to be described in detail the present invention, it is to be understood by those skilled in the art that can be Various changes are made to it in form and in details, without departing from claims of the present invention limited range.

Claims (5)

1. a kind of method for obtaining highway running status in real time based on mobile phone signaling data, it is characterised in that：The method bag Include following steps：

Step 1：According to highway actual path, rectangle monitored area delimited；The square grid of L is to prison using length and width Surveying region carries out rasterizing treatment；With the monitored area lower left corner as starting point, with geographical horizontal line as transverse axis, vertical line is divided for the longitudinal axis Grid；Grid is expressed as B [x, y], and wherein x is line number where this grid, and y is columns where this grid；If highway is a certain Travel direction is forward direction, if the grid containing express highway section is road grid, thus obtains highway forward direction road grid Lattice sequence L_HB={ B [h₁,w₁],B[h₂,w₂],B[h₃,w₃],…,B[h_i,w_i],…,B[h_n,w_n], wherein B [h_i,w_i] it is one Individual road grid, h_iThe line number where this grid, w_iThe columns where this grid；

Step 2：The base station data in monitored area is obtained, and the grating map determined according to step 1, set up base station information table； Base station information table (is base station comprising base station location area numbering LAC, base station cell numbering CellID, base station geographic position information Latitude and longitude coordinates, referred to as LOC), affiliated grid (referred to as CB), whether cover highway (referred to as IOW), corresponding road Road grid (referred to as CWB)；

Step 3：If T moment user profile collection is combined into S_U, wherein user profile includes user's unique mark UID and this user Track sets before the T moment；All signaling datas that monitored area produces within T to the T+t monitoring time periods are obtained, wherein Signaling data includes user's unique mark UID, signaling data own base station position area numbering LAC, base station cell numbering CellID With time stamp T imeStamp fields；Then the data by signaling data by user's classification and to each user are temporally stabbed TimeStamp is ranked up, and filters pingpang handoff, and co-located data are closed in continuous time section by user And, so as to build track sets of the user within the monitoring time period；User profile set S_U is then updated according to UID, if S_U It is middle to there is this user, then update the track sets of this user；If not existing this user in S_U, believe to this user is added in S_U Breath, thus obtains the user profile set S_U1 that track data updates within the monitoring time period；

Step 4：It is highway user to be located at the user travelled on expressway, obtains T moment highway user set S_HU (wherein S_HU is contained in S_U)；The track sets of each user in S_U1 are carried out with effectively trip section segmentation, and analyzes its last The track sets of Duan Youxiao trip sections, judge whether it travels on expressway, and updating maintenance highway user collection accordingly Positive track sets set S_PE and negative sense rail needed for closing S_HU and generation calculating each road grid real-time speed of highway Mark arrangement set S_NE；

Step 5：Each track sets in S_PE and S_NE are extracted, analytical calculation obtains the positive negative sense real-time speed of each road grid； Wherein road grid real-time speed is the express highway section real-time speed that road grid is included, wherein positive negative direction road grid Lattice real-time speed computational methods are consistent.

2. a kind of method for obtaining highway running status in real time based on mobile phone signaling data according to claim 1, It is characterized in that：In step 2, the affiliated grid CB in base station is determined according to base station LOC first, then according to base station range, Determine the grid set of base station covering；If containing road grid in the grid set of base station covering, by this base station IOW property values Iw is set to, this IOW property value is otherwise set to ow；Wherein iw represents that this base station range contains highway, and ow represents this Base station can not cover highway；If base station IOW property values are iw, the road grid and grid where it of calculation base station covering The distance of CB, it is the grid CWB in base station maps to road to take the minimum road grid of distance；If base station IOW property values are ow, Then its CWB value is null；If grid where certain base station is B [a, b], the road grid set of its covering is S_CHB={ B [c₁,d₁],B[c₂,d₂],B[c₃,d₃],…,B[c_i,d_i],…,B[c_m,d_m]}；Calculate each grid and grid where base station in S_CHB Lattice B [a, b] apart from S_D={ D_c1d1,D_c2d2,D_c3d3,…,D_cidi,…,D_cmdm(wherein D_cidiIt is B [c_i,d_i] with B [a, b] it Between distance, D_cidi=((c_i-a)²+(d_i-b)²)^(1/2)), it is the grid of base station maps to road to choose the minimum grid of S_D intermediate values Lattice.

3. a kind of method for obtaining highway running status in real time based on mobile phone signaling data according to claim 2, It is characterized in that：In step 4, specifically include：

Step 4.1：The track sets of each user in S_U1 are extracted, and carries out effectively trip section segmentation；Extract S_U1 in user most Latter section of section of effectively going on a journey, effectively step 4.2 is jumped to if this if trip section signaling number is more than certain threshold value, otherwise judges S_HU In whether there is this user, if in the presence of this user, according to user's travel direction by its track sets within the monitoring time period It is added in corresponding track sets set；

Step 4.2：Obtain effectively trip section track sets L_P={ P₁,P₂,P₃,…,P_i,…,P_u(wherein P_iRepresent track sequence I-th signaling data in row, in chronological sequence order is arranged between each signaling), the base station information table according to step 2 obtains Pi bases Stand corresponding road grid CWB, thus obtains the corresponding road raster series L of track sets_{_}PCWB={ CWB₁,CWB₂, CWB₃,…,CWB_i,…,CWB_u(wherein CWB_iIt is P_iCorresponding CWB, CWB_iMay be null)；

Step 4.3：Calculate the accounting of the CWB of non-null in L_PCWB sequences；Value is the CWB of null in obtaining L_PCWB sequences Number N1, and the CWB of the non-null of value number N2；Calculate the accounting R1=N2/ (N1+N2) of the CWB of non-null；If R1 Step 4.4 is then gone to more than certain threshold value, this user non-freeway user is otherwise assert, step 4.5 is jumped to；

Step 4.4：Value is the road grid sequence of the corresponding non-NULLs of CWB composition track sets P of non-null in extraction L_PCWB Row L_PCWB2={ CWB_g1,CWB_g2,CWB_g3,…,CWB_gi,…,CWB_gq(wherein L_PCWB2 is contained in L_PCWB, g_i∈[1, u])；Extract CWB_giSequence number O in Expressway Road raster series L_HB_gi, then the corresponding sequence number sequences of L_PCWB2 are L_ O={ O_g1,O_g2,O_g3,…,O_gi,…,O_gq}；Calculate the average value Agi=(O of two neighboring element in L_O_gi+O_g(i+1))/2, then Sequence after L_O smoothing processings is L_OA={ A_g1,A_g2,A_g3,…,A_gi,…,A_g(q-1)}；If L_{_}OA sequences exist obvious Growth trend, then assert that the moving direction of this user is matched and it is highway user with highway；If growth trend is Just, then it is assumed that this user's moving direction is highway positive direction, and its track sets within the monitoring time period is added to In S_PE；If growth trend is negative, then it is assumed that this user's moving direction is highway negative direction, and by it in the monitoring time period Interior track sets are added in S_NE；

Step 4.5：Update highway user set S_HU；Obtain T moment highway user collection and be combined into S_HU, if this user For highway user and S_HU do not exist this user, then this user is added in S_HU；If this user is judged as that non-high-speed is public There is this user in road user and S_HU, then this user is deleted from S_HU.

4. a kind of method for obtaining highway running status in real time based on mobile phone signaling data according to claim 3, It is characterized in that：In steps of 5, so that positive direction speed is asked for as an example, its step includes：

Step 5.1：It is that each road grid sets corresponding speed container in L_HB, for depositing cellphone subscriber in each track sets Translational speed between two tracing points；Speed is divided into by Z grade according to the path length between two tracing points, then each road Z speed container is included in the speed container of road grid, each speed container deposits corresponding speed data；

Step 5.2：Calculate in S_PE translational speed of the user between two tracing points in each track sets, according to two tracing points it Between distance grade classification is carried out to the speed being calculated, and put it into the corresponding speed container of corresponding road grid In；

Step 5.3：The speed data in each speed container using clustering algorithm to each road grid in L_HB carries out cluster point Analysis, obtains the corresponding speed of each road grid difference distance grade；If road grid B [h_i,w_i] corresponding speed container set is S_B[h_i,w_i] _ Con={ B [h_i,w_i]_C₁,B[h_i,w_i]_C₂,B[h_i,w_i]_C₃,…,B[h_i,w_i]_C_j,…,B[h_i,w_i]_ C_z, calculate B [h using focusing solutions analysis_i,w_i]_C_jMiddling speed degrees of data obtains the corresponding speed of grade j for Con_Speed_j, then Road grid B [h_i,w_i] different brackets sets of speeds be S_Con_Speed={ Con_Speed₁,Con_Speed₂,Con_ Speed₃,…,Con_Speed_j,…,Con_Speed_z}；Each speed container medium velocity data sample number is obtained, then road grid B [h_i,w_i] different brackets speed sample number collection is combined into S_Con_NS={ NS₁,NS₂,NS₃,…,NS_j,…NS_z}；Then road grid B [h_i,w_i] speed total sample number B [h_i,w_i] _ SpeedSwatch_Sum=NS₁+NS₂+NS₃+…+NS_j+…+NS_z；Using following Formula calculates road grid B [h_i,w_i] real-time speed B [h_i,w_i]_RTSpeed：

Step 5.4：Real time execution speed according to road grid determines the traffic behavior of express highway section therein.

5. a kind of method for obtaining highway running status in real time based on mobile phone signaling data according to claim 4, It is characterized in that：In the step 5.2, comprise the following steps that：

Step 5.2.1：Obtain a certain track sets L_EP={ R in S_PE₁,R₂,R₃,…,R_i,…,R_k(wherein R_iRepresent this rail I-th tracing point in mark sequence)；Each tracing point in order traversal L_EP, calculating cellphone subscriber is at it between follow-up each point Translational speed, and the speed that will be calculated is included into the speed container of corresponding road grid；

Step 5.2.2：If tracing point R_i-1And R_iCorresponding road grid is respectively B [e_(i-1),f_(i-1)] and B [e_i,f_i], obtain B [e_(i-1),f_(i-1)] and B [e_i,f_i] sequence number in L_HB, if the moving direction of the difference of two sequence numbers and this user is all just or together It is negative, then jumps to 5.2.3；

Step 5.2.3：Calculate B [e_(i-1),f_(i-1)] and B [e_i,f_i] between distance Dis_(i-1)i；Obtain B [e in L_HB_(i-1), f_(i-1)] and B [e_i,f_i] between subsequence L_EHB={ B [s₁,l₁],B[s₂,l₂],…,B[s_i,l_i],…,B[s_v,l_v], then B[e_(i-1),f_(i-1)] and B [e_i,f_i] between distance Dis_(i-1)iCan be calculated with formula below；

${\mathrm{Dis}}_{(i-1)i}=\underset{i=2}{\overset{v}{\Σ}}\sqrt{{(s_i-s_{i-1})}^2+{(l_i-l_{i-1})}^2}\×L$

Step 5.2.4：If tracing point R_i-1And R_iCorresponding time stamp T imeStamp_(i-1)And TimeStamp_i, then user is in R_i-1 And R_iBetween translational speed Speed_(i-1)i=Dis_(i-1)i/(TimeStamp_i-TimeStamp_(i-1))；

Step 5.2.5：According to the distance Dis that step 5.2.3 is obtained_(i-1)iBy Speed_(i-1)iIt is put into B [e_(i-1),f_(i-1)] and B [e_i,f_i] between subsequence L_EHB in the corresponding speed container of each road grid in；If Speed_(i-1)iCorresponding grade It is Level_Speed_(i-1)i, rate sequence is L_Level={ Level₁,Level₂,Level₃,…,Level_j,…,Level_z} (wherein Level_jRepresent j-th speed class)；Then Level_Speed can be determined by below equation_(i-1)i；If being Dis_(i-1)i More than or equal to DisNum_jAnd less than DisNum_j+1, then Level_Speed_(i-1)iIt is Level_j；Wherein DisNum_j+1It is more than DisNum_jIf, Dis_(i-1)iIn not interval any one, then this speed data is given up；

$Level\_{\mathrm{Speed}}_{(i-1)i}=\left\{\begin{array}{cc}{\mathrm{Level}}_1& {\mathrm{Dis}}_{(i-1)i}\∈\[{\mathrm{DisNum}}_1,{\mathrm{DisNum}}_2)\\ {\mathrm{Level}}_2& {\mathrm{Dis}}_{(i-1)i}\∈\[{\mathrm{DisNum}}_2,{\mathrm{DisNum}}_3)\\ \mathrm{......}& \mathrm{......}\\ {\mathrm{Level}}_j& {\mathrm{Dis}}_{(i-1)i}\∈\[{\mathrm{DisNum}}_j,{\mathrm{DisNum}}_{j+1})\\ \mathrm{......}& \mathrm{......}\\ {\mathrm{Level}}_s& {\mathrm{Dis}}_{(i-1)i}\∈\[{\mathrm{DisNum}}_s,{\mathrm{DisNum}}_{s+1}\]\end{array}\right..$