CN105142106A - Traveler home-work location identification and trip chain depicting method based on mobile phone signaling data - Google Patents

Abstract

The invention discloses a traveler home-work location identification and trip chain depicting method based on mobile phone signaling data, and belongs to the field of analysis of transportation planning data. The method comprises the steps of: triggering an information database by communication network records provided by a mobile network operator; after cleaning, integrating and calculating by using a position conversion algorithm, converting the data of the operator into a reference dataset; aiming at each user, converting the dataset into a mobile time-space trajectory, identifying stopover points, high frequency points and long time points in the mobile time-space trajectory, and on the basis, identifying home-work locations, activity target locations and insignificant stopover points; and finally depicting a closed trip chain of the user by being combined with the mobile time-space trajectory. The traveler home-work location identification and trip chain depicting method based on mobile phone signaling data has the advantages of being conductive to reducing trip survey costs of residents, obtaining more precision trip characteristics of crowds with different identities, and thereby allowing the traffic demand forecasting to be more accurate; and 2, being conductive to improving scientificity of planning layout construction, such as transport corridors, public transit networks and traffic facilities, in urban traffic planning.

Description

Traveler duty residence based on mobile phone signaling data identifies and Trip chain depicting method

Technical field

The invention belongs to traffic programme data analysis field, be specifically related to a kind of traveler duty residence based on mobile phone signaling data and identify and Trip chain depicting method.

Background technology

The duty residence of traveler and Trip chain obtain traffic trip data such as OD (" O " derives from English ORIGIN, point out the departure place of row, " D " derives from English DESTINATION, points out the destination of row), basis such as the significant data of Resident Trip Characteristics, traveler travel behaviour feature etc.Recommend for a long time according to mechanisms such as Institute of Traffic Engineers of the U.S. (ITE) and Urban Traffic Planning academic board of China Urban Planning Society, China Highway Institute and assert, the method obtaining traffic trip data is investigation of filling in a form, and mainly contains the methods such as door-to-door survey, telephone poll, roadside or public place inquiry.These methods not only need to drop into a large amount of human and material resources, and sampling rate is lower, and control time is short, and data precision is not enough.In view of mobile phone popularity rate constantly increases, the feature such as the easy Real-time Obtaining of mobile phone signaling data, cost are low, broad covered area, sampling period are flexible, can be used as and depict passerby's Trip chain, identifies the new data source of passerby's duty residence.But technical method design is unreasonable may bring and differentiate the large and data disaster of error.The present invention portrays based on a kind of traveler Trip chain adapting to large data processing of mobile phone signaling data exploitation the technical method identified with duty residence just.

Prior art one related to the present invention

The technical scheme of prior art one

Patent of invention: in mobile phone communication data, user travels frequently the method for digging of OD

Technical scheme: be comprise cell-phone number, call base station location and the data format of on by mobile phone communication data processing; In the statistics setting period, each cell-phone number is at the talk times of difference call base station location; Merge that overlapping in coverage to close on call base station be a base station area, add up the talk times of each cell-phone number at base station area; According to residence and the place of working of determining the user of described cell-phone number at the talk times of base station area of each cell-phone number, the base station area that talk times is high is residence or the place of working of user.

The shortcoming of prior art one

In data source, rely on call bill data (CDR) and CELL-ID (locate mode based on GSM) method to carry out customer location judgement, precision is not good, and the temporal discreteness of position data is large;

On method of discrimination, only consider user and rest on situation in a certain honeycomb for a long time, not based on the stop place of the time of staying, frequency of occurrences synthetic determination user, therefore not enough to the judgement accuracy of user stop place.

Prior art two related to the present invention

The technical scheme of prior art two

Patent of invention: based on the acquisition methods of the resident travel characteristic parameter of mobile phone location data

Technical scheme: collect mobile phone location data and filter, then matching traffic zone, by mobile phone location data after the coupling in a day according to time sequence, and continuous print data in same traffic zone are merged into one, obtains mobile phone location preprocessed data; According to the number merged and affect duration and judge dwell point; The trip route reduced between two dwell point, obtains trip distance, trip speed; Thus obtain the trip record sheet of all users; Based on the mobile phone location data after coupling, statistics obtains residence and work earth's surface; To go on a journey record sheet and residence and place of working result table Conjoint Analysis, to obtain user's trip characteristics parameter.

The shortcoming of prior art two

On method of discrimination, use the method for shortest path coupling to infer the Trip chain of user in minizone, lack data supporting, cannot embody actual trip, accuracy is poor; Meanwhile, in the judgement of dwell point, the point being greater than threshold value all time of staying, all as research object, does not reject insignificant stop, causes failing to remove for the insignificant trip of traffic programme, causes the calculated distortion for trip characteristics parameter.

In technical goal, based on the path after coupling, calculate the trip characteristics parameters such as distance, speed, with physical presence error, and characteristic parameter is considered too early, add data processing content and data volume, reduce the accuracy of judgement.

Prior art three related to the present invention

The technical scheme of prior art three

Patent of invention: the personnel based on the mobile phone location data of sparse sampling reside place recognition methods

Technical scheme: by target cities gridding, set up the mapping relations of mobile phone location data and grid, the probability of occurrence of counting user and the frequency of occurrences, respectively space-time cluster is carried out to each cellphone subscriber's frequency of occurrences and probability of occurrence, occur to be incorporated in Time and place the network that feature is similar, and the resident place of carrying out each cellphone subscriber identifies.

The shortcoming of prior art three

1, in data source, adopt the data of sparse sampling as basis of characterization, do not utilize mobile phone real time data reflect the trip information that passerby is complete and excavate;

2, on method of discrimination, only consider from the frequency of occurrences and probability, do not consider the stay time of user, fail effectively to react the temporal characteristics of resident point " resident ".

Summary of the invention

The present invention is directed to prior art data source to obtain not accurately and method of discrimination fails the abundant information of reflection and propose a kind of traveler duty residence based on mobile phone signaling data and identify and Trip chain depicting method.

For overcoming the above problems, technical method of the present invention is to provide a kind of traveler duty residence based on mobile phone signaling data and identifies and Trip chain depicting method, comprises the following steps:

Step 1: the communication network record trigger message database provided by Virtual network operator, by data encryption, integrated with conversion, is converted to data set; For each user, complete the cleaning of data set, sequence after namely each data in a few days being extracted one by one, and eliminate " drift " and " pingpang handoff ";

Step 2: the basic principle divided according to traffic zone, is divided into 2 and above traffic zone, and matches with urban function region and mobile cellular position by survey region;

Each data transformations is in a few days the Mobile Space-time track on the same day, according to distance range threshold value Th by step 3: for each user, according to timestamp one by one _ranand time of staying threshold value Th _tidentify the dwell point of each user every day;

Step 4: for each user, according to cumulative frequency threshold value Th in a day _{pfre_o}and be more than or equal to cumulative frequency threshold value Th on the two _{pfre_N}, for one day and be more than or equal to two, high frequency points and be more than or equal to high frequency points in two days in identifying; According to cumulative time threshold value Th in a day _{tacc_o}and be more than or equal to cumulative time threshold value Th on the two _{tacc_N}, long time point and be more than or equal to long time point in two days in identifying;

Step 5: according to high frequency, the long time point recognition result of each user, consider the various combination of high frequency points, long time point, set up the criterion of identification of duty residence, movable destination, meaningless dwell point, identify the duty residence of user, movable destination and meaningless dwell point;

Step 6: according to the recognition result of each customer objective point, in conjunction with the Mobile Space-time track of this user every day, portrays one day and is more than or equal to Trip chain complete in two days.

As preferably:

In conjunction with the recognition result of duty residence, movable destination, meaningless logical dwell point, in conjunction with nth user's m days Mobile Space-time tracks, the Trip chain that can obtain this user same day is as follows:

Wherein, tb _qrepresentative of consumer starts the time stopped at this point, to _qrepresentative of consumer stops the time stopping at this point, tk _qrepresent the character of this point, 1 is residence, and 2 is place of working, and 3 is movable destination, and 4 is meaningless dwell point, and 5 is by point, when for by some time, to and tb is equal, Δ t _{q-1 → q}representative of consumer to move to the time of q some cost from q-1, when q-1 put and q put be by some time, Δ t _{q-1 → q}be 0;

Then for nth user's m days Trip chain, if there is a kind subchain tk _pand tk _qbe not all 4 or 5, then can think that this subchain represents a travel behaviour of traveler, the set of all subchains in whole day, namely features nth user's all trips of m days.

The present invention can increase a dimension for the resident trip acquisition of information work based on mobile phone signaling data in sum, and can realize following effect:

1. be conducive to reducing resident trip survey cost, obtain the different identity crowd trip characteristics that precision is higher, make Traffic Demand Forecasting more accurate;

2. be conducive to promoting the science that the programming and distribution such as transport channel, public transport network, means of transportation in Urban Traffic Planning are built;

3. be conducive in urban planning, for duty residence balance, Public Establishment Location Selection, all kinds of outlet plannings etc. provide scientific basis.

Accompanying drawing explanation

Fig. 1 is main-process stream schematic diagram of the present invention;

Fig. 2 is the Trip chain schematic diagram that the present embodiment is portrayed.

Embodiment

For making object of the present invention, technical scheme and advantage clearly understand, to develop simultaneously embodiment referring to accompanying drawing, the present invention is described in further details.

That the present invention utilizes is that Mobile Network Operator provides, meet the mobile phone signaling data of state's laws about individual privacy, and these data have the features such as obtain manner is simple, procurement cost is low, sample cycle is flexible, quantity is large; Effectively can depict the Trip chain of passerby; More adequately can identify the duty residence of passerby; Can as the basis of the work such as the establishment of the table of traffic trip OD, the identification of traveler occupational identity.

Mobile phone signaling data in the present invention refers to, among mobile communication process, answer the call when occurring to beat, send short messages, switching on and shutting down, across LAC region, refresh at regular intervals time, by mobile phone call bill data (CDR) and the mobile phone traffic data (TDR) of mobile phone operators record.

Embodiment:

Step 1: the collection of data and preliminary treatment

The collection of step 1.1 data

The present invention adopt based on CDR the mobile phone signaling data of TDR data, through carrier data cleaning, integrated calculate with position transfer algorithm after, be converted to the data set of reference of the present invention, data set field comprises mobile subscriber's coding, timestamp, customer location longitude, customer location latitude, age of user, user's sex, SIM ownership place, and the form of timestamp is yyyymmddhhmmss.

The extraction of step 1.2 data and sequence

It is considered herein that the corresponding user of mobile phone and mobile subscriber's coding, according to mobile subscriber's coding and timestamp, extract all data of same user every day, and temporally stamp order ascending order arrangement.Carry out aid illustration by way of example, the data (part) in 20 days Mays in 2015 of user A000001 are as shown in table 1.

Table 1

Mobile subscriber encodes	Timestamp	Latitude	Longitude	Age	Sex	SIM ownership place
							A000001	20150520010456	30.0993	120.4826	25	Man	Shaoxing
A000001	20150520010052	30.0995	120.4840	25	Man	Shaoxing
							A000001	20150520012314	30.0991	120.4808	25	Man	Shaoxing
A000001	20150520011911	30.1013	120.4838	25	Man	Shaoxing
							A000001	20150520013931	30.0997	120.4824	25	Man	Shaoxing

The removal that step 1.3 is drifted about

From the 2nd row of each user's data every day, calculate one by one every bar timestamp correspondence position with upper some time difference (time_div) and enter instantaneous velocity (in_speed), with the difference that the upper some time difference is this data timestamp and a upper data timestamp, entering instantaneous velocity is the corresponding latitude and longitude coordinates of these data and the distance between the corresponding latitude and longitude coordinates of upper bar data, and with the business of upper some time difference, put upper some time difference of the 1st row data and entered instantaneous velocity and be 0.Travel through every a line of each user's data every day, remove and be less than and upper some time difference threshold value (TH with the upper some time difference _{diff_time}, desirable 10 ~ 60s, depending on the adjustment of data sampling interval), and enter instantaneous velocity and be greater than and enter instantaneous velocity threshold value (TH _{in_speed}, desirable 50 ~ 80km/h, viewed area mobile cellular density fluctuates) row.Carry out aid illustration by way of example, the drift of the data (part, has concealed age, sex and SIM card ownership place, and Biao Zhong unit omits) in 20 days Mays in 2015 of user A000001 removes result, TH _{diff_time}get 60s, TH _{in_speed}get km/h, as shown in table 2.

Table 2

The removal of step 1.4 pingpang handoff

From the 2nd row of each user's data every day, until inverse the 2nd behavior of each user's data every day only, meter current data is that p is capable, the distance counted between the capable correspondence position of p-1 and the capable correspondence position of p is AB, distance between the capable correspondence position of p and the capable correspondence position of p+1 is BC, distance between the capable correspondence position of p-1 and the capable correspondence position of p+1 is AC, and AC is less than soldier pang decision threshold A (TH _{pp_A}, desirable 100 ~ 200m, fluctuates depending on survey region mobile cellular density), and the value of (AB+BC)/AC is greater than soldier pang decision threshold B (TH _{pp_B}, desirable 1.5 ~ 2.0), then delete the capable data of m.Carry out aid illustration by way of example, the pingpang handoff of the data (part) in 20 days Mays in 2015 of user A000001 removes result, TH _{pp_A}get 200m, TH _{pp_B}get 2.0, as shown in table 3.

Table 3

Step 2: traffic zone divides mates with urban function region and mobile cellular

Step 2.1 traffic zone divides

According to the basic principle that traffic zone divides, survey region is divided into several traffic zones;

Step 2.2 traffic zone is mated with urban function region and mobile cellular

Based on traffic zone division result, in ArcGIS, build traffic zone layer; Based on the latitude and longitude coordinates of base station in survey region, use the service area of Voronoi diagram root base station, build mobile cellular layer; Based on the criteria for classifying of urban land in table 2 and corresponding function subregion, build functional region of city layer.Above 3 layer are imported in same ArcGIS database, as the basic database of research.

Step 3: Mobile Space-time track portray the identification with dwell point

Define 1. dwell points

Certain comprising at data set is investigated in a few days, at time of staying threshold value Th _durin, locus is constant, or change is at distance range threshold value Th _raninterior point, is dwell point (StayPoint, referred to as PS).

The expression of step 3.1 Mobile Space-time track

For nth user m days through pretreated data, be that data of p are expressed as (t by wherein timestamp _p, x _p, y _p, s _p), wherein, t _pfor timestamp, x _pfor longitude, y _pfor latitude, s _pfor current state, if consistent with the longitude and latitude of last bar data or in identification error threshold value (TH _err), then put 1, i.e. resting state, otherwise be set to 0, be i.e. mobile status, s ₁be set to 1.The total data on this user same day can be converted to the Mobile Space-time track of, be expressed as:

${\mathrm{traj}}_n^m\{(t_1,x_1,y_1,s_1),...,(t_k,x_k,y_k,s_k)\}---\left(1\right)$

The merging of step 3.2 Mobile Space-time track

For nth user's m days Mobile Space-time tracks, from the 2nd row, to last whereabouts, if s _kand s _k-1be 1, then put and stop intermission to _k-1for t _k, and delete row k, on the contrary put stop initial time tb _kand stop intermission to _kbe t _k, the mobile trip track after finally must merging:

${\mathrm{trajC}}_n^m\left\{\left({\mathrm{tb}}_1,{\mathrm{to}}_1,x_1,y_1,s_1\right),\mathrm{...},\left({\mathrm{tb}}_p,{\mathrm{to}}_p,x_p,y_p,s_p\right)\right\},p\≤k---\left(2\right)$

The primary election of step 3.3 dwell point

Determine Th _durand Th _ran, Th _durdesirable 15 ~ 30mins, Th _ranvalue depending on mobile phone Mobile data sample local cellular network density, 50 ~ 100m can be removed; According to dwell point definition, to the mobile trip track of all users every day, for each record, and to _kwith tb _kdifference be td _k, i.e. stop durations, if td _k>Th _dur, then think that kth bar record forms a dwell point, note tb _kwith to _kintermediate time for stop in time tm _kand stop place d ^pS(x ^pS, y ^pS), form the n-th traveler m days dwell point set ask for the dwell point set of all users every day successively.

Step 3.4 dwell point merges

Determine to stop threshold value Th interval time _diff, desirable 1 ~ 3mins, specifically looks the adjustment of data sampling interval.For each dwell point of each user, suspense PS ₁, corresponding to ₁with tb ₁if, Th around its stop place _ranin, there is other dwell points, suspense is PS ₂, corresponding to ₂with tb ₂if meet to ₂with tb ₁time difference <Th _diff, then PS is merged ₁and PS ₂, the tb of merging gets tb ₁, toto ₂get, stop durations and stop in time also revise accordingly, d ^psget each d needing the PS merged ^pscenter of gravity, corresponding to upgrade for dwell point (part) result of determination of the data on May 20th, 2015 of user A000001, Th _diffget 1mins, Th _ranget 200m, Th _durget 15mins, as shown in table 4.

Table 4

Step 4: the identification of high frequency points, long time point

Define high frequency points in 2. 1 days

Certain comprising at data set is investigated in a few days, and odd-numbered day occurrence number is d _{fre_o}be more than or equal to cumulative frequency threshold value Th in a day _{pfre_o}point, be called high frequency points in a day (HighFrequencyPointinOneDay, referred to as PHO).

Definition more than 3. in a few days high frequency points

In the investigation number of days D that data set comprises, can be d as the number of days of high frequency points in a day _{fre_N}, many days accumulative frequency of occurrences p _{fre_N}(d _{fre_N}/ D) be greater than many days cumulative frequency threshold value Th _{pfre_N}point, be called many in a few days high frequency points (HighFrequencyPointinNDays, referred to as PHN).

Define long time point in 4. 1 days

Certain comprising at data set is investigated in a few days, odd-numbered day accumulated dwelling time t _{acc_o}be more than or equal to cumulative time threshold value Th in a day _{tacc_o}point, be called long time point in a day (LongDurationPointinOneDay, referred to as PLO).

Definition more than 5. in a few days long time point

In the investigation number of days D that data set comprises, many days accumulated dwelling time t _{acc_N}(each day t _{acc_o}sum) be greater than many days cumulative times threshold value Th _{tacc_N}, and many days accumulative frequency of occurrences p _{fre_N}be greater than many days cumulative frequency threshold value Th _{pfre_N}point, be called many days long time points (LongDurationPointinNDays, referred to as PLN).

Asking for of step 4.1 odd-numbered day data

For nth user, according to its dwell point set in m days for among each dwell point, be set to current dwell point, in, find all longitudes and latitudes consistent or distance between two points (TH in identification error threshold value with it _err, desirable 30 ~ 50m, specifically fluctuates depending on survey region mobile cellular density) dwell point; If there is such dwell point, included in alternate location set, the quantity of alternate location set mid point is added the odd-numbered day occurrence number d that 1 is designated as current dwell point _{fre_o}, and by the time of staying summation of dwell points all in alternate location set, add the time of staying of current dwell point, be designated as the odd-numbered day accumulated dwelling time t of current dwell point _{acc_o}, all dwell points in the current dwell point of middle deletion and alternate location, put sky by alternative set, are placed in by current dwell point in; If there is not such dwell point, then direct the current dwell point of middle deletion, and current dwell point is placed in in, until till sky.For each user, all complete above operation.

Asking for of step 4.2 many days data

In probation D days, for nth user, owned merge into a set for each interior dwell point, is set to current dwell point, the all longitudes and latitudes of interior searching are consistent or distance between two points (TH in identification error threshold value with it _err) dwell point, if there is such dwell point, included in alternate location set, find out in alternate location set, the date that the time of staying is corresponding and current dwell point are not the dwell point of same day, then its quantity adds 1, and what be designated as current dwell point occurs number of days d for many days _{fre_N}, and calculate many days accumulative frequency of occurrences p _{fre_N}(d _{fre_N}/ D), and add up all dwell points in alternative set, and the odd-numbered day accumulated dwelling time of current dwell point, obtain the many days time of staying t of current dwell point _{acc_N}, dwell points all in the current dwell point of middle deletion and alternative set, sky is put in alternative set; If there is not such dwell point, then direct the current dwell point of middle deletion, until for empty position, for each user, all complete above operation.

The judgement of step 4.3 high frequency points

According to the result that odd-numbered day and many days data are asked for, in conjunction with the basic definition of high frequency points, for each user, identify high frequency points and many in a few days high frequency points in its all one day.For the high frequency points result of determination on May 19th ~ 21,2015 of user A000001, Th _{tacc_o}get 3, Th _{pfre_o}get 2, as shown in table 5, in the 4th, 5,6 row, numerical portion is d on the same day _{fre_o}for high frequency points on the same day judges conclusion in bracket.

Table 5

The judgement of the long time point of step 4.4

According to the result that odd-numbered day and many days data are asked for, in conjunction with the basic definition of long time point, for each user, identify long time point and many in a few days long time points in its all one day.For the long time point result of determination on May 19th ~ 21,2015 of user A000001, Th _{tacc_o}get 30mins, Th _{tacc_N}get 90mins, as shown in table 6, in the 4th, 5,6 row, numerical portion is t on the same day _{acc_o}in bracket be the same day long time point judge conclusion.

Table 6

Step 5: the identification of duty residence, movable destination, meaningless dwell point

The identification of step 5.1 duty residence

For all users, in probation D days, traveling through all current states in its Mobile Space-time track is the point of 1, place of working and residence is identified by the principle of table 7, wherein, working hour generally can be 7:00 to 17:00, and the period of living is other periods in one day, and working hour and period of living can make corresponding change according to the volume of traffic time varying characteristic in city, user place.

Table 7

The identification of the movable destination of step 5.2

For all users, in probation D days, traveling through all current states in its Mobile Space-time track is the point of 1, and is not residence, place of working, identifies general trip purpose place by the principle of table 8.

Table 8

The identification of the meaningless dwell point of step 5.3

For all users, in probation D days, traveling through all current states in its Mobile Space-time track is the point of 1, and is not the dwell point of residence, place of working, movable point of destination, is identified as meaningless dwell point.

For the result of determination on May 19th ~ 21,2015 of user A000001, as shown in table 9, in the 4th, 5,6 row, numerical portion is t on the same day _{acc_o}in bracket be the same day long time point judge conclusion.

Table 9

Step 6: portraying of Trip chain

Thus in conjunction with the recognition result of duty residence, movable destination, meaningless logical dwell point, in conjunction with nth user's m days Mobile Space-time tracks, the Trip chain that can obtain this user same day is as follows:

Wherein, tb _qrepresentative of consumer starts the time stopped at this point, to _qrepresentative of consumer stops the time stopping at this point, tk _qrepresent the character of this point, 1 is residence, and 2 is place of working, and 3 is movable destination, and 4 is meaningless dwell point, and 5 is by point, when for by some time, to and tb is equal, Δ t _{q-1 → q}representative of consumer to move to the time of q some cost from q-1, when q-1 put and q put be by some time, Δ t _{q-1 → q}be 0.

Then for nth user's m days Trip chain, if there is a kind subchain tk _pand tk _qbe not all 4 or 5, then can think that this subchain represents a travel behaviour of traveler, the set of all subchains in whole day, namely features nth user's all trips of m days.

For the Trip chain on May 20th, 2015 of user A000001, take longitude as x-axis, latitude is y-axis, and the time is z-axis, portrays its Trip chain, as shown in Figure 2.

Those of ordinary skill in the art will appreciate that, embodiment described here is to help reader understanding's implementation method of the present invention, should be understood to that protection scope of the present invention is not limited to so special statement and embodiment.Those of ordinary skill in the art can make various other various concrete distortion and combination of not departing from essence of the present invention according to these technology enlightenment disclosed by the invention, and these distortion and combination are still in protection scope of the present invention.

Claims (2)

1. the traveler duty residence based on mobile phone signaling data identifies and a Trip chain depicting method, it is characterized in that, comprises the following steps:

Step 1: the communication network record trigger message database provided by Virtual network operator, by data encryption, integrated with conversion, is converted to data set; For each user, complete the cleaning of data set, sequence after namely each data in a few days being extracted one by one, and eliminate " drift " and " pingpang handoff ";

Step 2: the basic principle divided according to traffic zone, is divided into 2 and above traffic zone, and matches with urban function region and mobile cellular position by survey region;

Each data transformations is in a few days the Mobile Space-time track on the same day, according to distance range threshold value Th by step 3: for each user, according to timestamp one by one _ranand time of staying threshold value Th _tidentify the dwell point of each user every day;

Step 4: for each user, according to cumulative frequency threshold value Th in a day _{pfre_o}and be more than or equal to cumulative frequency threshold value Th on the two _{pfre_N}, for one day and be more than or equal to two, high frequency points and be more than or equal to high frequency points in two days in identifying; According to cumulative time threshold value Th in a day _{tacc_o}and be more than or equal to cumulative time threshold value Th on the two _{tacc_N}, long time point and be more than or equal to long time point in two days in identifying;

Step 5: according to high frequency, the long time point recognition result of each user, consider the various combination of high frequency points, long time point, set up the criterion of identification of duty residence, movable destination, meaningless dwell point, identify the duty residence of user, movable destination and meaningless dwell point;

Step 6: according to the recognition result of each customer objective point, in conjunction with the Mobile Space-time track of this user every day, portrays one day and is more than or equal to Trip chain complete in two days.

2. a kind of traveler duty residence based on mobile phone signaling data according to claim 1 identifies and Trip chain depicting method, and it is characterized in that, the concrete grammar of step 6 is:

In conjunction with the recognition result of duty residence, movable destination, meaningless logical dwell point, in conjunction with nth user's m days Mobile Space-time tracks, the Trip chain that can obtain this user same day is as follows:

Wherein, tb _qrepresentative of consumer starts the time stopped at this point, to _qrepresentative of consumer stops the time stopping at this point, tk _qrepresent the character of this point, 1 is residence, and 2 is place of working, and 3 is movable destination, and 4 is meaningless dwell point, and 5 is by point, when for by some time, to and tb is equal, Δ t _{q-1 → q}representative of consumer to move to the time of q some cost from q-1, when q-1 put and q put be by some time, Δ t _{q-1 → q}be 0;

Then for nth user's m days Trip chain, if there is a kind subchain tk _pand tk _qbe not all 4 or 5, then can think that this subchain represents a travel behaviour of traveler, the set of all subchains in whole day, namely features nth user's all trips of m days.