CN104217593B - A kind of method for obtaining road condition information in real time towards mobile phone travelling speed - Google Patents
A kind of method for obtaining road condition information in real time towards mobile phone travelling speed Download PDFInfo
- Publication number
- CN104217593B CN104217593B CN201410428715.7A CN201410428715A CN104217593B CN 104217593 B CN104217593 B CN 104217593B CN 201410428715 A CN201410428715 A CN 201410428715A CN 104217593 B CN104217593 B CN 104217593B
- Authority
- CN
- China
- Prior art keywords
- road
- mobile phone
- travelling speed
- traffic state
- road traffic
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Landscapes
- Traffic Control Systems (AREA)
Abstract
The invention discloses a kind of method for obtaining road condition information in real time towards mobile phone travelling speed, belong to intelligent transportation field, the method comprises: historical traffic data excavates, and road traffic state is inferred, the calculating of road travel speed. The present invention is the mobile phone travelling speed calculating towards mobile phone signaling data, and these translational speeds are carried out fusion calculation and then obtained traffic information. Because the principle of obtaining of mobile phone travelling speed is network wireless location technology, therefore this takes full advantage of existing mobile communication facility and Internet resources, can realize and cover system-wide net, round-the-clock real-time traffic information of road collection with little investment. So can meet towards the traffic information fusion calculation of mobile phone travelling speed the demand that wide regional real-time road calculates.
Description
Technical field
The present invention relates to intelligent transportation field, particularly a kind of method for obtaining road condition information in real time towards mobile phone travelling speed.
Background technology
In recent years, information technology is constantly applied to traffic and transport field, and transport information resource is constantly abundant, traffic-information serviceMeans little by little perfect. By providing transport information to the public, can before travel make good planning, go out for travelerIn row, evade the section that blocks up, thereby reach the object of effectively utilizing road, minimizing to block up.
Real-time traffic information collection is the basis of advanced transportation information service systems efficiently. Traffic information collection mode is divided into fixingFormula gathers and movable type gathers two kinds. Traditional fixed acquisition mode because it is with high costs, need manual maintenance, easy to wear etc.Feature makes it be difficult to realize the traffic information collection that covers system-wide net. In recent years, along with wireless sensor network and location technology (asGPS) fast development, numerous radio hand-held equipment and mobile units with positioning function are popularized in a large number, portable collection sideFormula has obtained development fast. The transport information that the existing traffic information service system based on Floating Car provides is covered at downtown roadsLid rate and accuracy rate have all reached quite high level. But because existing floating car data source derives from taxi and public transport mostlyCar, makes still to have obvious shortage of data phenomenon at surrounding city (suburb, highway). Locate based on wireless mobileThe traffic data collection technology of technology is because its investment is little, and data volume is large, gathers the features such as wide coverage, can effectively make upThe deficiency of existing acquisition mode, is subject to the attention of transport body just day by day. At present, using wireless mobile locator data as data sourceCarrying out traffic information calculating has become an important development direction of transport information treatment technology.
In sum, the transport information treatment technology of realizing based on wireless mobile location is of great practical significance. ExistingThe transport information treatment technology scheme based on wireless mobile location be divided into two large classes: adopt high-precision fixed bit data to calculate road conditions and believeBreath and employing are switched locator data based on base station and are calculated traffic information. Adopting high-precision fixed bit data to calculate traffic information technology becomesRipe, accuracy rate is high, but because high-precision fixed bit data cost is higher and relate to privacy of user problem, is difficult to large under existence conditionsScale gathers, and cannot really bring into play radio data feature. And adopt based on base station switch positioning principle calculate road conditions modes byLow in its positioning precision, directly cause the traffic information that calculates inaccurate, even mistake.
Transport information treatment technology based on mobile phone mobile signaling protocol, its principle is to switch and calculate traffic information based on base station, it facesLot of challenges. As mobile phone carrier wide material sources, both can be from the other crowd on foot of road, also can come bus and carPassenger can, from the driver who drives bicycle, be also that on road, the cellphone subscriber in motor vehicle is only available but only have carrierThe sample calculating in road conditions, because carrier is difficult to distinguish, has caused the traffic information calculating can not reflect faithfully the friendship of roadLogical situation; And for example mobile phone mobile signaling protocol data have the features such as randomness is strong, positioning precision is low, and at adjacent different timeThe cellphone subscriber's sample size fluctuation that plays phone (triggering communication event) in section is larger, thereby the calculating of road travel speed is madeBecome deviation (shake); It is inaccurate that these situations all make traffic information calculate.
Summary of the invention
In view of this, the object of the present invention is to provide a kind of method for obtaining road condition information in real time towards mobile phone travelling speed, withReach and improve the precision that road conditions are calculated, the object of realize accurately, the real-time traffic information with wide coverage calculating, andWhole object is to improve the service quality of intelligent transportation system.
Embodiments of the invention provide a kind of method for obtaining road condition information in real time towards mobile phone travelling speed, and traffic information mergesBe that the traffic information that on every road, all cell phones produce is done to further fusion treatment, generate the reality taking road as unitTime dynamic information. Comprise:
Step 1, historical traffic data excavate;
Step 2, road traffic state are inferred;
Step 3, road travel speed are calculated.
The present invention applies existing mobile phone travelling speed data, and then its fusion calculation is obtained to traffic information. Due to towards mobile phoneThe traffic information fusion calculation method of translational speed takes full advantage of existing mobile communication facility and Internet resources, can be with seldomInvestment realize and cover system-wide net, round-the-clock real-time traffic information of road collection, therefore towards the road conditions letter of mobile phone travelling speedBreath fusion calculation method can meet the demand that wide regional real-time road calculates.
Brief description of the drawings
Fig. 1 is the principle schematic that mobile phone travelling speed obtains; Fig. 1 (a) is previous base station coverage area, Fig. 1 (b)For the real region of switching for there is cellular base station of juncture area that adjacent two base stations cover, i.e. switch area, Fig. 1 (c) is nextIndividual base station coverage area;
The flow chart of the method for obtaining road condition information in real time towards mobile phone travelling speed that Fig. 2 provides for the embodiment of the present invention;
The historical traffic data excavation flow chart that Fig. 3 provides for the embodiment of the present invention;
The road travel speed calculation flow chart that Fig. 4 provides for the embodiment of the present invention.
Detailed description of the invention
For making the object, technical solutions and advantages of the present invention clearer, below in conjunction with accompanying drawing, the present invention is done further in detailDescribe.
The present embodiment is applied existing mobile phone travelling speed data, and then its fusion calculation is obtained to traffic information. Due to towards handThe method for obtaining road condition information in real time of machine translational speed takes full advantage of existing mobile communication facility and Internet resources, can be with veryFew investment realizes and covers system-wide net, round-the-clock real-time traffic information of road collection, therefore towards the road conditions of mobile phone travelling speedInformation fusion calculation method can meet the demand that wide regional real-time road calculates. The embodiment of the present invention is according to mobile phone travelling speedThe feature of data, proposes historical traffic data method for digging, road traffic state estimation method, and road travel speed calculation method,Thereby obtain traffic information.
The mobile phone travelling speed data of the present embodiment application are the mobile phone mobile signaling protocol numbers that enlivens cellphone subscriber providing in operatorAccording to basis on calculate, any active ues refers to trigger mobile communication event (call, note sending and receiving, normal position renewal etc.)User. The data format of mobile phone travelling speed data is as shown in the table.
Table 1
| Field name | Describe | Type | Maximum length |
| TimeSeq | Time period sequence number | Integer character constant string | 3 |
| User ID | User's mobile phone cipher number | Integer character constant string | 11 |
| Link ID | Road chain number | Integer character constant string | 10 |
| Speed | Velocity amplitude | Floating type | 10 |
Mobile phone, in mobile process, can keep continual with base station and communicate by letter, when the signal strength signal intensity of mobile phone Current Serving BTS subtractsA little less than, the signal strength signal intensity of neighbor base station exceedes current base station, and mobile phone signal can be switched to neighbor base station, to obtain better signal.In the process of switching, operator can retain the relevant record that switches, and this provides required signaling for calculating mobile phone travelling speedData. The position that mobile phone switching occurs is called switching point (HandoverPoint), and the section that recurs twice mobile phone switching isSwitch section, switching point and switching section continuous on road have formed road handover network (HandoverNetwork) jointly.The a series of switching point of mobile phone, determines the driving path under mobile phone by map match, completes after route matching, carries out mobile phoneThe calculating of speed.
Based on mobile phone switch carry out the calculating of mobile phone speed schematic diagram as shown in Figure 1, the junctional area that in figure, adjacent two base stations coverReal region of switching for there is cellular base station, territory, switch area is as (b), when mobile phone is led to by previous base station coverage area (a)When crossing switch area and entering next base station coverage area (b), can record a switching time and switching point, same, work as warpWhile crossing next switch area, can record next switching time and switching point. So just produce a time difference, according to switchingDistance between district just can calculate the speed that mobile phone moves between two switching points. In application the method, need manyInferior test is determined the switching point of mobile phone and is switched section, calculates the road section length between adjacent two switching points. Utilize mobile phone to switchPoint gathers the method for transport information, need to, through repeatedly measuring, determine handover network, the convenient further location to mobile phone.
Fig. 2 is the flow chart of the traffic information fusion calculation method towards mobile phone travelling speed that provides of the embodiment of the present invention, the partyMethod comprises:
Step 201, historical traffic data excavate. Because traffic data has periodically and similitude to a certain extent, as earlyThe tidal regime of evening peak. Therefore be necessary historical traffic data to carry out law mining, excavate its Changing Pattern, and finalIts rule is applied in the road traffic state prediction that current (real-time) road traffic state is differentiated and future is possible.In the time that historical traffic data is excavated, need to complete successively extraction characteristic value; Set up characteristic vector; Structure training tuple; Carry outThe steps such as the training of tuple-set (as Fig. 3).
Step 2011, extraction characteristic value. The method of extracting characteristic value specifically refers to, extracts with road and hand over from historical traffic dataThe logical relevant value of state, the value of mobile phone user number on the value of time period sequence number and road. Characteristic value is as data mining mouldThe input of type, is the abstractdesription to data, therefore characteristic value choose extremely importantly, can it reflect number to be sorted exactlyAccording to feature will directly affect final classifying quality. The characteristic of division that the present embodiment is chosen comprises: time period sequence number feature and roadThe feature of mobile phone user number on road.
Because road traffic state value is always different along with the difference of time period, so can be used as traffic behavior, time period sequence number sentencesOther feature.
Time period in whole day can represent with following formula:
[(t/12):(t％12)×5,(t/12):(t％12)×5+5](1)
Wherein t represents time period sequence number, because the sampling interval of mobile phone signaling data is 5 minutes, therefore, one day totalThe individual time period. Time period sequence number 0 represent the time period [0:00,0:05), the time period, sequence number t represented the time period[(t/12): (t%12) × 5, (t/12): (t%12) × 5+5]. The concept of time period sequence number t has been simplified retouching the data rise timeState. The time period sequence number feature of such one day can be with gathering I={0,1,2 ... n, n=1,2,3 ... 288} represents.
Because road traffic state value is usually subject to the impact of mobile phone user number fluctuation on road, mobile phone user numberGreat Yi causes traffic congestion, and little the making of mobile phone user number has a good transport and communication network, so mobile phone user number can be done on roadFor a feature of road traffic state differentiation.
Step 2012, set up characteristic vector. The method of setting up characteristic vector specifically refers to, according to the characteristic value extracting above,Set up characteristic vector, the one-component using each characteristic value as characteristic vector. Described characteristic vector is for decision treeThe input of training pattern.
Step 2013, structure training tuple. The method of structure training tuple specifically refers to: according to the characteristic vector of above foundation,Structure training tuple, an attribute using each component of characteristic vector as training tuple, finally also will add a class markSign attribute, class tag attributes refers to road traffic state (unimpeded, slow, block up), derives from the demarcation of floating car technology.
The training of step 2014, execution tuple-set. The training of carrying out tuple-set specifically refers to, inputs tuple-set to decision-makingIn tree training pattern, carry out training and operation. Apply a large amount of historical traffic datas and carry out the training of decision tree forecast model, can obtainOptimum forecast model. Decision tree (DecisionTree) is an important method in Classification Algorithms in Data Mining. It is oneIndividual forecast model, what its represented is a kind of mapping relations between object properties and object value. Its operation principle is known eachOn the basis of the situation of kind probability of happening, the desired value of asking for net present value (NPV) by forming decision tree is more than or equal to zero probability, is straightSee a kind of diagram method of using probability analysis. After input training tuple-set in decision tree training pattern, decision tree trainingModel, through training study process, will produce forecast model, and this forecast model is to pick in follow-up road travel speed is calculatedExcept disturbing bunch collection and select optimum bunch guiding decision support is provided, be i.e. the validity of data identification. The validity identification of dataBe a part for data fusion, data fusion refers to the translational speed of mobile phone is merged to the road travel speed becoming taking road as unitDegree. Effective identification of data is the committed steps in data fusion, differentiates and could participate in road for effective mobile phone travelling speed dataThe fusion calculation of condition information.
Step 202, road traffic state are inferred. Read the decision tree forecast model training, can be by from mobile phone travelling speed numberAccording to the characteristic vector of middle extraction be divided into block up, slow and unimpeded three classes. With reference to decision tree training pattern, historical tuple-set is trainedThe characteristic parameter going out, thus judge that the classification results of current (real-time) characteristic vector to be sorted is road traffic state.
The step that step 203, road travel speed are calculated is followed successively by and adopts KMeans clustering algorithm to be polymerized to mobile phone travelling speed3 bunches; From 3 bunches, select optimum bunch; Utilize Kalman filtering algorithm to carry out school to the central value of current optimum bunchJust (as Fig. 4).
Step 2031, employing KMeans clustering algorithm are polymerized to 3 bunches to mobile phone travelling speed. Adopt KMeans cluster to calculateMethod is polymerized to 3 bunches to the mobile phone travelling speed in current (real-time) same time period.
Step 2032, from 3 bunches, select optimum bunch. Utilize road traffic state estimation method to infer and current roadRoad traffic behavior, then in 3 bunches from cluster, select to meet most current road traffic condition bunch, optimum bunch, itSelection rule as follows:
If current road traffic state is unimpeded, central value maximum bunch be optimum;
If current road traffic state is slowly, what speed number was maximum bunch is optimum;
If current road traffic state is to block up, central value minimum bunch be optimum;
Step 2033, utilize Kalman filtering algorithm to proofread and correct the central value of current optimum bunch. Utilize Kalman filterRipple algorithm to current optimum bunch central value carry out filtering processing, when filtering is processed with reference to the travelling speed of front 3 adjacent time intervalsValue is proofreaied and correct the central value when prevariety, proofread and correct complete after, the result after output calibration, i.e. current road travel velocity amplitude.
In a word, the foregoing is only preferred embodiment of the present invention, be not intended to limit protection scope of the present invention.
Claims (3)
1. towards a method for obtaining road condition information in real time for mobile phone travelling speed, it is characterized in that, comprising:
Step 1, historical traffic data excavate;
Step 2, road traffic state are inferred;
Step 3, road travel speed are calculated;
Wherein, described historical traffic data method for digging specifically comprises:
Extract characteristic value, set up characteristic vector, structure training tuple, carries out the training of tuple-set;
Wherein, the method for described extraction characteristic value specifically comprises:
From all fields of historical traffic data, extract the field relevant to road traffic state, the field of time period sequence number and roadThe field of mobile phone user number on road, extraction time section sequence number characteristic value and road on the feature of mobile phone user numberValue;
Wherein, the described method of setting up characteristic vector specifically comprises:
According to the characteristic value extracting above, set up characteristic vector, the one-component using each characteristic value as characteristic vector;
Wherein, the method for described structure training tuple specifically comprises:
According to the characteristic vector of above foundation, structure training tuple, using each component of characteristic vector as one of training tupleIndividual attribute, finally also will add a class tag attributes, and class tag attributes refers to road traffic state, specifically utilizes 0 representative smoothLogical, 1 representative slowly, 2 representatives block up, and derives from the road traffic state that floating car technology (FCD) obtains;
Wherein, the training of described execution tuple-set specifically comprises:
Input tuple-set, in decision tree training pattern, is carried out training study task.
2. the method for obtaining road condition information in real time towards mobile phone travelling speed according to claim 1, is characterized in that,Described road traffic state estimation method specifically comprises:
The characteristic parameter that utilizes decision tree training pattern to train historical tuple-set, thus judge that current is real-time treat pointThe classification results of category feature vector, i.e. road traffic state identification.
3. the method for obtaining road condition information in real time towards mobile phone travelling speed according to claim 1, is characterized in that,The step that described road travel speed is calculated specifically comprises:
A, adopt KMeans clustering algorithm to current be that mobile phone travelling speed in real-time same time period is polymerized to 3 bunches,Bunch this is because all speed of a motor vehicle on road generally present high, medium and low three kinds of speed colonies;
B, utilize road traffic state estimation method to infer current be real-time road traffic state, then from cluster3 bunches in select to meet most current road traffic condition bunch, optimum bunch, its selection rule is as follows:
If current road traffic state is unimpeded, central value maximum bunch be optimum;
If current road traffic state is slowly, what speed number was maximum bunch is optimum;
If current road traffic state is to block up, central value minimum bunch be optimum;
C, utilize Kalman filtering algorithm to current optimum bunch central value carry out filtering processing, with reference to first 3 when adjacentSection travelling speed value proofread and correct the central value when prevariety, proofread and correct complete after, proofread and correct result export, export current roadRoad travelling speed value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410428715.7A CN104217593B (en) | 2014-08-27 | 2014-08-27 | A kind of method for obtaining road condition information in real time towards mobile phone travelling speed |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410428715.7A CN104217593B (en) | 2014-08-27 | 2014-08-27 | A kind of method for obtaining road condition information in real time towards mobile phone travelling speed |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104217593A CN104217593A (en) | 2014-12-17 |
| CN104217593B true CN104217593B (en) | 2016-05-25 |
Family
ID=52099027
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410428715.7A Active CN104217593B (en) | 2014-08-27 | 2014-08-27 | A kind of method for obtaining road condition information in real time towards mobile phone travelling speed |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104217593B (en) |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104636611A (en) * | 2015-02-03 | 2015-05-20 | 上海奕行信息科技有限公司 | Urban road/ road segment vehicle speed evaluation method |
| CN105679021B (en) * | 2016-02-02 | 2018-11-06 | 招商局重庆交通科研设计院有限公司 | Journey time fusion forecasting and querying method based on traffic big data |
| CN105702041A (en) * | 2016-04-21 | 2016-06-22 | 东南大学 | Highway multisource data fusion state estimation system based on neural network and method thereof |
| CN105869402B (en) * | 2016-05-12 | 2019-04-19 | 中山大学 | Express highway section speed modification method based on polymorphic type floating car data |
| CN106548645B (en) * | 2016-11-03 | 2019-07-12 | 济南博图信息技术有限公司 | Vehicle route optimization method and system based on deep learning |
| CN106448168B (en) * | 2016-11-24 | 2019-02-19 | 中山大学 | Traffic event automatic detection method based on tendency index and fluctuation index |
| CN106981197B (en) * | 2017-05-23 | 2019-10-22 | 招商局重庆交通科研设计院有限公司 | Multistage road grid traffic method for control speed |
| CN111190984B (en) * | 2019-12-30 | 2024-03-08 | 北京世纪高通科技有限公司 | Method and device for extracting job place and computer readable storage medium |
| CN113344239B (en) * | 2021-04-25 | 2022-11-29 | 山东师范大学 | Traffic congestion condition prediction method and system based on two-stage spectral clustering |
| CN114333324A (en) * | 2022-01-06 | 2022-04-12 | 厦门市美亚柏科信息股份有限公司 | Real-time traffic state acquisition method and terminal |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101706996A (en) * | 2009-11-12 | 2010-05-12 | 北京交通大学 | Method for identifying traffic status of express way based on information fusion |
| CN102737510A (en) * | 2012-07-03 | 2012-10-17 | 浙江大学 | Real-time traffic condition acquisition method based on mobile intelligent terminal |
| CN103116990A (en) * | 2013-01-18 | 2013-05-22 | 杨飞 | Traffic speed vehicle-mounted acquisition system and method based on mobile phone switch position |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20000032614A (en) * | 1998-11-16 | 2000-06-15 | 이흥수 | Method and apparatus for collecting traffic information and method for processing speed data of traffic information |
-
2014
- 2014-08-27 CN CN201410428715.7A patent/CN104217593B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101706996A (en) * | 2009-11-12 | 2010-05-12 | 北京交通大学 | Method for identifying traffic status of express way based on information fusion |
| CN102737510A (en) * | 2012-07-03 | 2012-10-17 | 浙江大学 | Real-time traffic condition acquisition method based on mobile intelligent terminal |
| CN103116990A (en) * | 2013-01-18 | 2013-05-22 | 杨飞 | Traffic speed vehicle-mounted acquisition system and method based on mobile phone switch position |
Non-Patent Citations (1)
| Title |
|---|
| 基于手机位置信息的实时路况信息服务;史风林等;《计算机应用》;20080531(第5期);正文第48-49页、图1 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104217593A (en) | 2014-12-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104217593B (en) | A kind of method for obtaining road condition information in real time towards mobile phone travelling speed | |
| Iqbal et al. | Development of origin–destination matrices using mobile phone call data | |
| US9830817B2 (en) | Bus station optimization evaluation method and system | |
| CN106878951B (en) | User trajectory analysis method and system | |
| CN106912015B (en) | Personnel trip chain identification method based on mobile network data | |
| CN102521973B (en) | A kind of mobile phone switches the road matching method of location | |
| CN106781479B (en) | A method of highway operating status is obtained based on mobile phone signaling data in real time | |
| CN102799897B (en) | Computer recognition method of GPS (Global Positioning System) positioning-based transportation mode combined travelling | |
| CN106530716A (en) | Method for calculating highway section average speed based on mobile phone signaling data | |
| CN108320501A (en) | Public bus network recognition methods based on user mobile phone signaling | |
| CN108171993B (en) | Highway vehicle speed calculation method based on mobile phone signaling big data | |
| CN106101999B (en) | A kind of recognition methods of user trajectory and device | |
| CN106197458A (en) | A kind of cellphone subscriber's trip mode recognition methods based on mobile phone signaling data and navigation route data | |
| CN105142106A (en) | Traveler home-work location identification and trip chain depicting method based on mobile phone signaling data | |
| CN104765808B (en) | The method for digging and system of one kind of groups track | |
| CN105809962A (en) | Traffic trip mode splitting method based on mobile phone data | |
| CN109769201A (en) | A kind of smart city management platform for realizing user's precise positioning | |
| CN105101092A (en) | Mobile phone user travel mode recognition method based on C4.5 decision tree | |
| CN109688532A (en) | A kind of method and device dividing city function region | |
| CN109272032A (en) | Trip mode recognition methods, device, computer equipment and storage medium | |
| CN105682097A (en) | Pseudo base station recognizing and positioning method and device | |
| CN105070057A (en) | Real-time road traffic monitoring method | |
| CN104636611A (en) | Urban road/ road segment vehicle speed evaluation method | |
| CN109348404B (en) | Method for extracting individual travel road track in big data environment | |
| CN110460962A (en) | Rail traffic user identification method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |