CN104754730B - The method and device of position prediction - Google Patents
The method and device of position prediction Download PDFInfo
- Publication number
- CN104754730B CN104754730B CN201310739194.2A CN201310739194A CN104754730B CN 104754730 B CN104754730 B CN 104754730B CN 201310739194 A CN201310739194 A CN 201310739194A CN 104754730 B CN104754730 B CN 104754730B
- Authority
- CN
- China
- Prior art keywords
- mobile data
- user
- sequence
- mobile
- discrete
- 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
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W64/00—Locating users or terminals or network equipment for network management purposes, e.g. mobility management
- H04W64/003—Locating users or terminals or network equipment for network management purposes, e.g. mobility management locating network equipment
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/02—Services making use of location information
- H04W4/029—Location-based management or tracking services
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W64/00—Locating users or terminals or network equipment for network management purposes, e.g. mobility management
- H04W64/006—Locating users or terminals or network equipment for network management purposes, e.g. mobility management with additional information processing, e.g. for direction or speed determination
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
The embodiment of the invention discloses a kind of method and devices of position prediction, are related to data processing field, can solve the problem of that there are the motion tracks that Accurate Prediction user's future is unable to when disturbing factor.The method of the present invention includes:Obtain user's mobile data;User's mobile data is pre-processed, General Mobile sequence is built;Short-term forecast and long-term forecast are carried out respectively according to the General Mobile sequence;The result of the short-term forecast is superimposed upon in the result of the long-term forecast, obtains the result of moving projection.The present invention is suitable for position prediction.
Description
Technical field
The present invention relates to data processing field more particularly to a kind of method and devices of position prediction.
Background technology
In mobile Internet business, location based service rapidly develops, and many convenience are brought to people’s lives.Example
Such as, the traffic information in front of the path of user's traveling is provided, helps user to carry out traffic path planning in advance, avoids traffic
It blocks;Travelling route recommendation etc. is provided for tourist.These application scenarios need to solve there are one common problem, that is, it is accurate to need
Predict the motion track in user's future.For the movement pattern problem in user's future, difference is proposed in the prior art
Predicting Technique:Such as sequence prediction may be carried out in which cell movement to user using Markov Chain method;Or
Correlation rule is established using frequent item set mining technology, and then predicts next cell etc. of user's movement.
At least there are the following problems in the prior art:The above-mentioned Predicting Technique provided, in the moving rail in prediction user's future
When mark, all assume that the user's mobile data got is accurate.However, due to weather and weather conditions variation, city ring
Signal switching etc., signal strength when can all user be caused to move in the irregular terrain profiles such as building, tunnel and wireless network in border
Change, signal interruption or the problems such as be switched to other base stations, so these unpredictable disturbing factors make user move
Contain noise in data, be not precise information, and then cause when predicting the motion track in user's future, the precision of prediction is very
It is low, it is unable to the motion track in Accurate Prediction user's future.
Invention content
The embodiment of the present invention provides a kind of method and device of position prediction, can solve there are when disturbing factor, no
The problem of motion track in energy Accurate Prediction user's future.
In order to achieve the above objectives, the embodiment of the present invention adopts the following technical scheme that:
In a first aspect, the embodiment of the present invention provides a kind of method of position prediction, including:
Obtain user's mobile data;
User's mobile data is pre-processed, General Mobile sequence is built;
Short-term forecast and long-term forecast are carried out respectively according to the General Mobile sequence;
The result of the short-term forecast is superimposed upon in the result of the long-term forecast, obtains the result of moving projection.
With reference to first aspect, in the first possible realization method of first aspect, user's mobile data includes:
The mark of cell residing for user, user reach the time point of the cell and collect the number of user's mobile data.
The possible realization method of with reference to first aspect the first, in second of possible realization method of first aspect
In, pretreatment user's mobile data, building General Mobile sequence includes:
User's mobile data described in discretization obtains discrete mobile data;
The value that General Mobile parameter is arranged is the first preset value;
According to the discrete mobile data and original discrete mobile data, the General Mobile sequence is built, it is described original
The acquisition time point of discrete mobile data is before the acquisition time point of the discrete mobile data, and close to the discrete movement
The number of the acquisition time point of data, original discretization data is first preset value.
Second of possible realization method with reference to first aspect, in the third possible realization method of first aspect
In, described according to the discrete mobile data and original discrete mobile data, building the General Mobile sequence includes:
According to the discrete mobile data and original discrete mobile data, in the pre-set interval of collection frequence, son is determined
The beginning and end of sequence, the subsequence are the discrete mobile data of cell ID frequent switching residing for user, described frequent
The cell ID number of switching is not more than first preset value, and the pre-set interval length is the second preset value, described
The initial value of pre-set interval is the initial collection frequence of original discrete mobile data;
The subsequence is indicated with identical universal location, constitutes the General Mobile sequence.
The third possible realization method with reference to first aspect, in the 4th kind of possible realization method of first aspect
In, it is described short-term forecast to be carried out according to the General Mobile sequence respectively and long-term forecast includes:
Subsequence storage is gathered to index;
Definition starting index;
Calculate the offset of the relatively described starting index of index of the position to be predicted;
According to the offset, the corresponding subsequence of index of the position to be predicted is determined;
According to the subsequence, prediction sets are calculated;
According to quantile parameter, the quantile of the prediction sets is calculated, obtains short-term forecast result.
Second of possible realization method with reference to first aspect, in the 5th kind of possible realization method of first aspect
In, it is described short-term forecast to be carried out according to the General Mobile sequence respectively and long-term forecast further includes:
According to movement law, preset time is divided into disjoint group;
Determine out-of-date mobile sequence;
According to the General Mobile sequence and the out-of-date mobile sequence, quantity and the discretization position of universal location are calculated
Cumulative number;
Determine the return group and same day index of location index to be predicted;
Calculate separately the number of the cumulative number of the discretization position of position to be predicted divided by the universal location of position to be predicted
The result of amount;
Determine the result not less than the result that the cell ID of threshold value is long-term forecast.
The 5th kind of possible realization method with reference to first aspect, in the 6th kind of possible realization method of first aspect
In, it is described according to the General Mobile sequence and the out-of-date mobile sequence, calculate quantity and the discretization position of universal location
Cumulative number include:
For the discretization position of the General Mobile sequence, the cumulative number adds 1;
For the discretization position of the out-of-date mobile sequence, the cumulative number subtracts 1;
For the universal location of the General Mobile sequence, the quantity of the universal location adds 1;
For the universal location of the out-of-date mobile sequence, the quantity of the universal location subtracts 1.
Second aspect, the embodiment of the present invention provide a kind of device of position prediction, including:
Acquiring unit, for obtaining user's mobile data;
Construction unit builds General Mobile sequence for pre-processing user's mobile data;
Predicting unit, for carrying out short-term forecast and long-term forecast respectively according to the General Mobile sequence;
Superpositing unit, the result for the result of the short-term forecast to be superimposed upon to the long-term forecast, obtains movement
The result of prediction.
In conjunction with second aspect, in the first possible realization method of second aspect, user's mobile data includes:
The mark of cell residing for user, user reach the time point of the cell and collect the number of user's mobile data.
In conjunction with the first possible realization method of second aspect, in second of possible realization method of second aspect
In, the construction unit includes:
Discrete subelement obtains discrete mobile data for user's mobile data described in discretization;
Subelement is set, and the value for General Mobile parameter to be arranged is the first preset value;
Subelement is built, for according to the discrete mobile data and original discrete mobile data, building the general shifting
Dynamic sequence, the acquisition time point of original discrete mobile data before the acquisition time point of the discrete mobile data, and
Close to the acquisition time point of the discrete mobile data, the number of original discretization data is first preset value.
In conjunction with second of possible realization method of second aspect, in the third possible realization method of second aspect
In, the structure subelement is specifically used for:
According to the discrete mobile data and original discrete mobile data, in the pre-set interval of collection frequence, son is determined
The beginning and end of sequence, the subsequence are the discrete mobile data of cell ID frequent switching residing for user, described frequent
The cell ID number of switching is not more than first preset value, and the pre-set interval length is the second preset value, described
The initial value of pre-set interval is the initial collection frequence of original discrete mobile data;
The subsequence is indicated with identical universal location, constitutes the General Mobile sequence.
In conjunction with the third possible realization method of second aspect, in the 4th kind of possible realization method of second aspect
In, the predicting unit includes:
Storing sub-units, for gathering subsequence storage to index;
Subelement is defined, for defining starting index;
First computation subunit, the offset of the relatively described starting index of index for calculating the position to be predicted;
First determination subelement, for according to the offset, determining the corresponding sub- sequence of the index of the position to be predicted
Row;
First computation subunit is additionally operable to, according to the subsequence, calculate prediction sets;
First computation subunit is additionally operable to calculate the quantile of the prediction sets according to quantile parameter, obtains
Short-term forecast result.
In conjunction with second of possible realization method of second aspect, in the 5th kind of possible realization method of second aspect
In, subelement is divided, for according to movement law, preset time to be divided into disjoint group;
Second determination subelement, for determining out-of-date mobile sequence;
Second computation subunit, for according to the General Mobile sequence and the out-of-date mobile sequence, calculating general position
The cumulative number of the quantity and discretization position set;
Second determination subelement is additionally operable to determine the return group of location index to be predicted and same day index;
Second computation subunit be additionally operable to calculate separately the cumulative number of the discretization position of position to be predicted divided by
The result of the quantity of the universal location of position to be predicted;
Second determination subelement is additionally operable to determine that the cell ID that the result is not less than threshold value is long-term forecast
Result.
In conjunction with the 5th kind of possible realization method of second aspect, in the 6th kind of possible realization method of second aspect
In, second computation subunit is specifically used for:
For the discretization position of the General Mobile sequence, the cumulative number adds 1;
For the discretization position of the out-of-date mobile sequence, the cumulative number subtracts 1;
For the universal location of the General Mobile sequence, the quantity of the universal location adds 1;
For the universal location of the out-of-date mobile sequence, the quantity of the universal location subtracts 1.
In the prior art, when carrying out position prediction, contain noise in the user's mobile data got, be not perfect number
According to, cause predict user's future motion track when, the precision of prediction is very low.Compared with prior art, the embodiment of the present invention
A kind of method and device of the position prediction provided, server obtains user's mobile data in the present invention, and is pre-processed, root
The result of Data preprocess builds General Mobile sequence, then carries out short-term forecast and long-term pre- respectively according to General Mobile sequence
It surveys, the result of short-term forecast is superimposed upon in the result of long-term forecast, obtains the result of moving projection.Due to obtaining user's shifting
After dynamic data and pretreatment, reduces the influence of interference, and predicted using the prediction technique of interference robust, improve prediction
Precision, be capable of the motion track in Accurate Prediction user's future.
Description of the drawings
It to describe the technical solutions in the embodiments of the present invention more clearly, below will be to needed in the embodiment
Attached drawing is briefly described, it should be apparent that, drawings in the following description are only some embodiments of the invention, for ability
For the those of ordinary skill of domain, without creative efforts, it can also be obtained according to these attached drawings other attached
Figure.
Fig. 1 is the method flow diagram that one embodiment of the invention provides;
Fig. 2 is the method flow diagram that further embodiment of this invention provides;
Fig. 3 is the user trajectory schematic diagram that further embodiment of this invention provides;
Fig. 4, Fig. 5 are the apparatus structure schematic diagram that further embodiment of this invention provides;
Fig. 6 is the apparatus structure schematic diagram that further embodiment of this invention provides.
Specific implementation mode
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation describes, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts all other
Embodiment shall fall within the protection scope of the present invention.
One embodiment of the invention provides a kind of method of position prediction, position prediction is used for, as shown in Figure 1, the method
Including:
101, server obtains user's mobile data.
Wherein, server obtains user's mobile data from the equipment for wirelessly collecting data, and user moves number
The time point of cell is reached according to the mark including cell residing for user, user and obtains the number of user's mobile data.
As long as it should be noted that having the moving projection of the executable user trajectory of the equipment of computing function, not only
It is server, common PC(Personal Computer, PC)Machine, other electronic equipments etc. can complete this function, this
Place is without limiting.
102, server pre-processes user's mobile data, builds General Mobile sequence.
Wherein, server obtains discrete mobile data first by user's mobile data discretization, and General Mobile ginseng is arranged
Several values is the first preset value, then according to discrete mobile data and original discrete mobile data, builds General Mobile sequence.It is former
There is the acquisition time point of discrete mobile data before the acquisition time point of discrete mobile data, and close to discrete mobile data
The number of acquisition time point, original discretization data is the first preset value.
It will be obtained after user's mobile data discretization of acquisition it should be noted that discrete mobile data is server.
According to discrete mobile data and original discrete mobile data, build General Mobile sequence includes server:According to discrete mobile number
According to original discrete mobile data the beginning and end of subsequence is determined in the pre-set interval of collection frequence;Subsequence is used
Identical universal location indicates, constitutes General Mobile sequence.Subsequence is the discrete shifting of cell ID frequent switching residing for user
The cell ID number of dynamic data, frequent switching is not more than first preset value, and pre-set interval length is the second preset value, in advance
If the initial value in section is the initial collection frequence of original discrete mobile data.
103, server carries out short-term forecast and long-term forecast respectively according to General Mobile sequence.
Wherein, short-term forecast process is:Server stores subsequence to after index set, definition starting index, and counts
Calculate the offset of the opposite starting index of index of position to be predicted;According to offset, determine that the index of position to be predicted is corresponding
Then subsequence calculates prediction sets;According to quantile parameter, the quantile of prediction sets is calculated, obtains short-term forecast result.
Further, long-term prediction procedure is:Preset time is divided into disjoint by server according to movement law
Group;Determine out-of-date mobile sequence;According to General Mobile sequence and out-of-date mobile sequence, the quantity and discretization of universal location are calculated
The cumulative number of position;Determine the return group and same day index of location index to be predicted;Calculate separately the discrete of position to be predicted
Change the result of the quantity of the cumulative number of position divided by the universal location of position to be predicted;Definitive result is small not less than threshold value
Area is identified as the result of long-term forecast.
For the discretization position of General Mobile sequence, cumulative number adds 1;For the discretization position of out-of-date mobile sequence
It sets, cumulative number subtracts 1;For the universal location of General Mobile sequence, the quantity of universal location adds 1;For out-of-date mobile sequence
Universal location, the quantity of universal location subtracts 1.
104, the result of short-term forecast is superimposed upon in the result of long-term forecast by server, obtains the result of moving projection.
In the prior art, when carrying out position prediction, contain noise in the user's mobile data got, be not perfect number
According to, cause predict user's future motion track when, the precision of prediction is very low.Compared with prior art, the embodiment of the present invention
Middle server obtains user's mobile data, and is pre-processed, and builds General Mobile sequence according to pretreated result, then root
Short-term forecast and long-term forecast are carried out respectively according to General Mobile sequence, and the result of short-term forecast is superimposed upon to the result of long-term forecast
On, obtain the result of moving projection.Due to after obtaining user's mobile data and pre-processing, reducing the influence of interference, and adopt
It is predicted with the prediction technique of interference robust, improves the precision of prediction, be capable of the motion track in Accurate Prediction user's future.
Further embodiment of this invention provides a kind of method of position prediction, is used for position prediction, wherein server is from passing through
User's mobile data is obtained in the equipment of wireless mode collection data;Then sliding-model control is carried out to user's mobile data, obtained
Go out General Mobile sequence;And then short-term forecast and long-term forecast are carried out according to General Mobile sequence pair position to be predicted;It will be short-term
Final prediction result is obtained after prediction result and the superposition of long-term forecast result.Server is the equipment for having computing function, such as PC
The electronic equipments such as machine.As shown in Fig. 2, the method includes:
201, wireless network demarcates cell ID.
Wherein, the mobile data that obtain user demarcates cell ID, to distinguish user institute to the cell that user occurs
Position.C={ c is set-1,c0,c1,c2,…,cCellsCount, c1,c2,…cCellsCountFor the cell mark of wireless network calibration
Know, cell ID is that cell imparts unique ID(Identity, identification number), cell number CellsCount is that setting is small
Area mark number, C for be possible to cell ID set.c-1And c0It is the position manually set, c-1Indicate mobile station
It closes, c0It indicates that signal can not be obtained from base station for some reason, makes user location that can not demarcate.The embodiment of the present invention with
It is arranged for ten cell IDs, i.e. CellsCount is equal to 10.
202, wireless device collects user's mobile data.
It should be noted that the collection of user's mobile data is mainly carried out by the wireless device with wireless mode
, for example, mobile phone, base station, GPS(Global Positioning System, global positioning system)、Wi-Fi(wireless
Fidelity, Wireless Fidelity)Etc. be used equally for collect user's mobile data.
Wherein, in the wireless network, when carrying out mobile handoff between each cell as user, the mobile data of user can quilt
It records.
For example, the data format of motion track can be used<ci,timei>, i=1,2 ..., TotalSwitchesCount's
Format, wherein switching total quantity TotalSwitchesCount indicates the total quantity switched by the end of current time user location,
Being turned on and off including mobile device, the interruption of mobile device signal or mobile device rediscover signal etc., ciIndicate i-th
User is moved to cell c by the cell after secondary handover event occurs where user, i.e. wireless networki, timeiIndicate user's movement
The signal of equipment is switched to cell ciTime.
In the embodiment of the present invention, the primary data of the mobile data of user is<c-1,time0>, time0Indicate a Zhou Zhong
The time of daystart, for example, 00:00AM(Ante Meridiem, the morning), value is less than or equal to time1.It indicates to survey with d
The time interval of data is measured, then Conventional Time lattice act on time-domain, are expressed as
For floor operation.Since the time daily measures, thenExpression observes use daily
The quantity of family mobile data.
It should be noted that in the value for determining d, therefore, to assure that DailyObsCount is integer.With
For DailyObsCount=1000, then d need to be taken as 0.001 day, i.e., measured user's mobile data every 0.001 day,
It is exactly 0.001*24*60 minutes=1.44 minutes.It is thus determined that d is 1.44 minutes, user's movement was measured every 1.44 minutes
Data measure 1000 times, it is ensured that DailyObsCount is integer for one day.
203, server obtains user's mobile data from wireless device.
As long as it should be noted that having the moving projection of the executable user trajectory of the equipment of computing function, not only
It is server, ordinary PC, other electronic equipments etc. can complete this function, herein without limiting.
Wherein, after measuring user's mobile data, n indicate discrete interval quantity, accumulation n be spaced quantity after again to
Family mobile data is handled, and n is the integer not less than 0.Therefore, user's mobile data is handled after accumulating n*d days.
It should be noted that carrying out data processing if often measuring user's mobile data, workload can be very big, institute
After user's mobile data n times measurement can be had collected, then carry out data processing.One-shot measurement spends d days, collects n times, flower
Time be exactly n*d.Wherein n is optional, and collecting how many a user's mobile datas every time can artificially determine.
204, server discretization user mobile data obtains discrete mobile data.
Wherein, user's mobile data according to when layout carry out discretization, the method for discretization is:Between each discretization
In, the cell residing for user, which takes, to be set to the user in the discretization interval and stays that longest cell of total time, to
Go out discretization mobile data ct-n+1,ct-n+2,…,ct。
For example, as shown in figure 3,1 is the trajectory diagram obtained according to user's mobile data, by when layout a for, can from 1
See, at this in layout, user's mobile data is first located at cell 1, is switched to cell 8 later, then switches back to cell again
1, it has been switched to cell 7 again followed by.At this in layout, it is easy to it is to be located at cell 1 to see user's most of time all,
Therefore after discretization, this when layout in think that user is constantly in cell 1, discretization results are passed through as shown in 2 in Fig. 3
After discretization, when each in layout, user is only positioned at a cell.
205, server determines General Mobile parameter.
Wherein, General Mobile parameter is indicated with k.In step 205, show that data length is the discretization mobile data of n
ct-n+1,ct-n+2,…,ct, in original discretization mobile data, determine k observation time close to ct-n+1,ct-n+2,…,
ctOriginal discretization mobile data ct-n-k+1,ct-n-k+2,…,ct-n, in conjunction with ct-n+1,ct-n+2,…,ct, constitute discrete mobile sequence
Arrange ct-n-k+1,ct-n-k+2,…,ct-n,ct-n+1,…,ct, to calculate General Mobile sequence.
206, server calculates General Mobile sequence.
It should be noted that General Mobile sequence in k original discrete mobile datas and step 205 based on obtaining
What what n discretization mobile data calculated went out, the General Mobile sequence obtained, the data in General Mobile sequence are known as general position
It sets, length is still n.
Wherein, it is based on discrete mobile sequence ct-n-k+1,ct-n-k+2,…,ct-n,ct-n+1,…,ct, user is found out no more than k
Then the sequence of frequent switching between a cell determines the beginning and end of frequent switching sequence, obtains frequent switching sequence
Section [starting point, terminal].The method for determining the beginning and end of frequent switching sequence is, for section [t-n-k+1 ..., t-
K] in each r, (1) finds out minimum value r1, it is allowed to meet r-k≤r1≤ r and(2) maximum value r is found out2, it is allowed to meet
r≤r2≤ r+k and(3) for eachSo thatIn conjunction with original General Mobile sequenceIt constitutes
General Mobile sequenceI.e. current General Mobile sequenceHere ∪ indicates relationship simultaneously.
For the data of the discrete mobile sequence of not frequent switching, corresponding universal location is directly determined according to data content,
The General Mobile sequence of final output is ut-n-k+1,ut-n-k+2,…,ut-k。
For example, the embodiment of the present invention, by taking k=3 as an example, user moves in 10 cells of calibration, according to the user of collection
After mobile data discretization, show that discretization sequence is c1,c2,c3,c4,c5,c2,c1,c2,c1,c4,c5,c6,c7,c8,c6,c7,
c8,c6,c7,c9,c10, that is, c1=c1,c2=c2,…,c6=c2,c7=c1,…,c20=c9,c21=c10.Due to k=3, it is therefore desirable to examine
The situation for considering user's frequent switching between no more than k cell, that is, need consideration user single subdistrict or two cells it
Between between frequent switching or three cells frequent switching situation.User's mobile data elder generation position it can be seen from discretization sequence
In cell 1, after have respectively entered cell 2,3,4,5, since these cells do not duplicate, can determine that its is corresponding logical
It is u with position1~u5, i.e. u1=u1=c1,…,u5=u5=c5.Then, user's mobile data is introduced into cell 2, after be switched to it is small
Then area 1 is switched to cell 2 again, last to switch back to cell 1 again, i.e., section [6,9] corresponding user's mobile data is at two
Minizone frequent switching, [6,9] corresponding discrete mobile sequence is frequent switching sequence, so r1It is 6, r2It is 9,ForThen the corresponding universal location of user's mobile data in the section [6,9] of frequent switching sequence is u6=u7=u8=u9=
{c6,c7,c8,c9}={c2,c1,c2,c1, then merger same section, obtains u6=u7=u8=u9=u1,2={c1,c2}.Frequent switching
The method that the method that the section [12,19] of sequence calculates universal location calculates universal location with frequent switching section [6,9] is identical,
Corresponding universal location is u12=…=u19=u6,7,8={c6,c7,c8, so the universal sequence of final output is:u1,u2,u3,u4,
u5,u1,2,u1,2,u1,2,u1,2,u4,u5,u6,7,8,u6,7,8,u6,7,8,u6,7,8,u6,7,8,u6,7,8,u6,7,8,u6,7,8,u9,u10, wherein
ui,j,…,k={ci,cj,…,ck}。
207, server update index set.
It should be noted that index set S () is calculated according to General Mobile sequence, for user location
Short-term forecast, this step be according to General Mobile sequence update index set, to ensure the accuracy of short-term forecast.
For example, the General Mobile sequence u determined in step 2071,u2,u3,u4,u5,u1,2,u1,2,u1,2,u1,2,u4,u5,
u6,7,8,u6,7,8,u6,7,8,u6,7,8,u6,7,8,u6,7,8,u6,7,8,u6,7,8,u9,u10In, u5≠u6=u7=u8=u9≠u10, then this sub- sequence
The section of row is (6,9), wherein tstart=6,tend=9, it is saved into index set S (), final indexed set is combined into
It should be noted that all universal location u, indexed set credit union disposes expired part, indicates to protect with w
Remain for the maximum parameter of the historical data of short-term forecast, i.e., disposes t in gathering from indexendThe part subset of≤t-w.
208, server determines starting index.
Wherein, with tbeginIt indicates determining starting index, the index of position to be predicted, t is indicated with tbeginIt is artificially defined
One starting index, it is therefore an objective to t is mapped to tbeginIn that day of beginning.
209, server calculates the mapping position of position to be predicted according to starting index.
Wherein, with Shift (t, tbegin) indicating the mapping position of position to be predicted, DailyObsCount is to be seen in one day
The number of user data is surveyed, then the calculation formula of the mapping position of position to be predicted is:
It is calculated by formula, t is mapped to by indexing tbeginCorresponding index position in that day of definition, in formulaFor under
Floor operation accords with, and % accords with for modulo operation.
For example, in the embodiment of the present invention by taking DailyObsCount is 1000 as an example, i.e., the number of user data is observed daily
It it is 1000 times, with tbeginFor 1002, expression is start observation for second day second, thenEqual to 1*1000=1000.T%DailyObsCount is to take the remainder, i.e.,
Offset is sought, by taking t is 1 as an example, then the value of offset is 1, that is, Shift (t, tbegin) be 1001, i.e., t is mapped to second
It first position.If t is 2001, the value of offset is still 1, i.e. Shift (t, tbegin) still it is 1001, still by t
It is mapped to second day first position.
210, server calculates prediction sets according to mapping position.
For example, with WithFor, then set D is there are one elementD has multiple elements for meeting condition, and 2 be one of element.
211, server obtains short-term forecast result by the quantile of calculating prediction sets.
Wherein, after obtaining prediction sets D, according to position parameter QuantileLevel is divided, the quantile of prediction sets is calculated
Quantile (D, QuantileLevel), you can prediction obtains universal location ut.QuantileLevel be artificially be arranged can
It adjusts parameter, indicates the element for being located at that position QuantileLevel*100% in prediction sets D, i.e. percentage in prediction sets D
Element its (QuantileLevel*100) less than that is derived as short-term forecast result.
For example, by taking QuantileLevel is equal to 0.5 as an example, quantile, expression is asked to need to obtain in D prediction sets D
An element, it is desirable that have 50% element in set D, value is less than that.It is with prediction sets D(1,2,3,4,5)For, meter
50% small that element of element ratio own value is calculated, it is 3 to obtain a result.QuantileLevel values are different, the result obtained
It is different.If QuantileLevel is equal to 1, then it represents that need to obtain an element in D, it is desirable that other elements 100% are all than it
Small, the result obtained is 5.
212, server division group.
Wherein, according to User Activity rule, ordinary day is divided into disjoint group of G={ g1,g2,…,gm, m is to divide
The number of group, it is contemplated that movable periodicity, these disjoint group will independently be handled.
For example, with one week for the period, by one week Mon(Monday, Monday), Tue(Tuesday, Tuesday), Wed
(Wednesday, Wednesday), Thu(Thursday, Thursday), Fri(Friday, Friday), Sat(Saturday, week
Six), Sun(Sunday, Sunday), divide:{ Mon, Tue, Wed, Thu, Fri } is one group, and { Sat, Sun } is another group;Or
Person was divided into { Mon, Tue, Wed, Thu, Fri } by one week, { Sat }, three groups of { Sun };Or it is divided into { Mon }, { Tue },
{ Wed }, { Thu }, { Fri }, { Sat }, seven groups of { Sun };Or { Mon, Tue, Wed, Thu, Fri, Sat, Sun } is classified as one group
Etc..
213, server determines the group and same day index of the return of position to be predicted.
Wherein, the index t for corresponding to position to be predicted indicates corresponding with the ordinary day group that g (t) indicates to return with i (t)
In the same day index in certain day of index t, i.e. i (t)=t%DailyObsCount.
214, server determines out-of-date mobile sequence.
Wherein, out-of-date mobile sequence is the mobile sequence of observation time at most.Prediction needs to use historical data, and more long
Remote data are not more worth, so needing to carry out data update, remove out-of-date data.Showing that length is the general shifting of n
After dynamic sequence, to determine that length is n, the out-of-date mobile sequence of observation time at most indicates to be preserved for long-term forecast with w
The maximum parameter of historical data, then the section of out-of-date mobile sequence is [t-k-n-w+1, t-k-n-w+2 ..., t-k-w].
215, server calculates quantity and the discretization position of universal location according to General Mobile sequence and out-of-date mobile sequence
The cumulative number set.
For example, the General Mobile sequence u obtained in step 2071,u2,u3,u4,u5,u1,2,u1,2,u1,2,u1,2,u4,u5,u6 ,7,8,u6,7,8,u6,7,8,u6,7,8,u6,7,8,u6,7,8,u6,7,8,u6,7,8,u9,u10, for each uτ(τ=1,2 ..., 21), i.e., from
u1Start to u21, often meet cell cr, willIt executes plus 1 operates, r=1,2 ..., 10.For out-of-date mobile sequence, often touch
To cell cr, willOperation that execution subtracts 1.Calculating it is similar.
216, server calculates the result of long-term forecast.
Optionally, the result of long-term forecast can be according to cjThe frequency of generation determines, for example, in a manner of descending, it is right
All cjIt is ranked up according to the frequency of generation, r cell before then selecting, makes the sum frequency number of generation more than some setting
Threshold value.
217, the result of short-term forecast is superimposed upon in the result of long-term forecast by server, obtains the result of moving projection.
Wherein, withIndicate short-term forecast as a result, withIndicate long-term forecast
As a result, subscript s and l indicate that short-term forecast and long-term forecast, h indicate the step number of prediction, h respectivelysIndicate the thresholding of short-term forecast
Value, that is, predict when step number is not more than the value to be short-term forecast, is then long-term forecast when being more than the value.By folded in long-term forecast
Add short-term forecast, then the final prediction result Forecast obtained is
In the prior art, when carrying out position prediction, contain noise in the user's mobile data got, be not perfect number
According to, cause predict user's future motion track when, the precision of prediction is very low.Compared with prior art, the embodiment of the present invention
Middle server obtains user's mobile data, and carries out sliding-model control, and General Mobile sequence is built according to the result of sliding-model control
Then row update index set according to General Mobile sequence, carry out short-term forecast, carried out according to General Mobile sequence long-term pre-
It surveys, the result of short-term forecast is superimposed upon in the result of long-term forecast, obtains the result of moving projection.Due to obtaining user's shifting
After dynamic data and sliding-model control, reduce the influence of interference, and using the prediction technique of interference robust carry out short-term forecast and
Long-term forecast improves the precision of prediction, is capable of the motion track in Accurate Prediction user's future.
Further embodiment of this invention provides a kind of device 30 of position prediction, as shown in figure 4, described device 30 includes:
Acquiring unit 31, for obtaining user's mobile data;
Construction unit 32 builds General Mobile sequence for pre-processing user's mobile data;
Predicting unit 33, for carrying out short-term forecast and long-term forecast respectively according to the General Mobile sequence;
Superpositing unit 34, the result for the result of the short-term forecast to be superimposed upon to the long-term forecast obtain shifting
The result of dynamic prediction.
Wherein, user's mobile data includes:The mark of cell residing for user, user reach the time point of the cell
With the number for collecting user's mobile data.
Further, as shown in figure 5, the construction unit 32 includes:
Discrete subelement 321 obtains discrete mobile data for user's mobile data described in discretization;
Subelement 322 is set, and the value for General Mobile parameter to be arranged is the first preset value;
Subelement 323 is built, for according to the discrete mobile data and original discrete mobile data, structure to be described general
Mobile sequence, the acquisition time point of original discrete mobile data before the acquisition time point of the discrete mobile data,
And close to the acquisition time point of the discrete mobile data, the number of original discretization data is first preset value.
Further, the structure subelement 323 is specifically used for:
According to the discrete mobile data and original discrete mobile data, in the pre-set interval of collection frequence, son is determined
The beginning and end of sequence, the subsequence are the discrete mobile data of cell ID frequent switching residing for user, described frequent
The cell ID number of switching is not more than first preset value, and the pre-set interval length is the second preset value, described
The initial value of pre-set interval is the initial collection frequence of original discrete mobile data;
The subsequence is indicated with identical universal location, constitutes the General Mobile sequence.
Further, as shown in figure 5, the predicting unit 33 includes:
Storing sub-units 331, for gathering subsequence storage to index;
Subelement 332 is defined, for defining starting index;
First computation subunit 333, offset of the index relative to the starting index for calculating the position to be predicted
Amount;
First determination subelement 334, for according to the offset, determining the corresponding son of index of the position to be predicted
Sequence;
First computation subunit 333 is additionally operable to, according to the subsequence, calculate prediction sets;
First computation subunit 333 is additionally operable to calculate the quantile of the prediction sets according to quantile parameter, obtains
Go out short-term forecast result.
Further, as shown in figure 5, the predicting unit 33 includes:
Subelement 335 is divided, for according to movement law, preset time to be divided into disjoint group;
Second determination subelement 336, for determining out-of-date mobile sequence;
Second computation subunit 337, for according to the General Mobile sequence and the out-of-date mobile sequence, calculating general
The cumulative number of the quantity and discretization position of position;
Second determination subelement 336 is additionally operable to determine the return group of location index to be predicted and same day index;
The cumulative number that second computation subunit 337 is additionally operable to calculate separately the discretization position of position to be predicted removes
With the result of the quantity of the universal location of position to be predicted;
Second determination subelement 336 is additionally operable to determine that the cell ID that the result is not less than threshold value is long-term pre-
The result of survey.
Further, second computation subunit 337 is specifically used for:
For the discretization position of the General Mobile sequence, the cumulative number adds 1;
For the discretization position of the out-of-date mobile sequence, the cumulative number subtracts 1;
For the universal location of the General Mobile sequence, the quantity of the universal location adds 1;
For the universal location of the out-of-date mobile sequence, the quantity of the universal location subtracts 1.
In the prior art, when carrying out position prediction, contain noise in the user's mobile data got, be not perfect number
According to, cause predict user's future motion track when, the precision of prediction is very low.Compared with prior art, the embodiment of the present invention
Middle device 30 obtains user's mobile data, and is pre-processed, and builds General Mobile sequence according to pretreated result, then root
Short-term forecast and long-term forecast are carried out respectively according to General Mobile sequence, and the result of short-term forecast is superimposed upon to the result of long-term forecast
On, obtain the result of moving projection.Due to after obtaining user's mobile data and pre-processing, reducing the influence of interference, and adopt
It is predicted with the prediction technique of interference robust, improves the precision of prediction, be capable of the motion track in Accurate Prediction user's future.
Further embodiment of this invention provides a kind of device 40 of routing diffusion, as shown in fig. 6, described device 40 includes:
Processor 41, for obtaining user's mobile data;And for pre-processing user's mobile data, structure is logical
Use mobile sequence;And for carrying out short-term forecast and long-term forecast respectively according to the General Mobile sequence;And it is used for
The result of the short-term forecast is superimposed upon in the result of the long-term forecast, obtains the result of moving projection.
Wherein, user's mobile data includes:The mark of cell residing for user, user reach the time point of the cell
With the number for collecting user's mobile data.
Further, the processor 41 is additionally operable to user's mobile data described in discretization, obtains discrete mobile data;With
And the value for General Mobile parameter to be arranged is the first preset value;And for according to the discrete mobile data and it is original from
Mobile data is dissipated, builds the General Mobile sequence, the acquisition time point of original discrete mobile data is in the discrete shifting
Before the acquisition time point of dynamic data, and close to the acquisition time point of the discrete mobile data, original discretization data
Number be first preset value.
Further, the processor 41 is additionally operable to according to the discrete mobile data and original discrete mobile data,
In the pre-set interval of collection frequence, determine that the beginning and end of subsequence, the subsequence are that cell ID residing for user is frequent
The cell ID number of the discrete mobile data of switching, the frequent switching is not more than first preset value, described pre-
If siding-to-siding block length is the second preset value, the initial value of the pre-set interval is original discrete mobile data initial collection time
Number;And for indicating the subsequence with identical universal location, constitute the General Mobile sequence.
Further, the processor 41 is additionally operable to gather subsequence storage to index;And for defining
Begin index;And the offset of the relatively described starting index of index for calculating the position to be predicted;And it is used for root
According to the offset, the corresponding subsequence of index of the position to be predicted is determined;And for according to the subsequence, meter
Calculate prediction sets;And for according to quantile parameter, calculating the quantile of the prediction sets, obtaining short-term forecast result.
Further, the processor 41 is additionally operable to according to movement law, and preset time is divided into disjoint group;With
And for determining out-of-date mobile sequence;And for according to the General Mobile sequence and the out-of-date mobile sequence, calculating
The cumulative number of the quantity and discretization position of universal location;And return group for determining location index to be predicted and work as
Day indexes;And the discretization position for calculating separately position to be predicted cumulative number divided by position to be predicted it is general
The result of the quantity of position;And for determining the result not less than the result that the cell ID of threshold value is long-term forecast.
Further, the processor 41 is additionally operable to the discretization position for the General Mobile sequence, described accumulative
Number adds 1;And for the discretization position for the out-of-date mobile sequence, the cumulative number subtracts 1;And for pair
In the universal location of the General Mobile sequence, the quantity of the universal location adds 1;And for for the out-of-date movement
The quantity of the universal location of sequence, the universal location subtracts 1.
In the prior art, when carrying out position prediction, contain noise in the user's mobile data got, be not perfect number
According to, cause predict user's future motion track when, the precision of prediction is very low.Compared with prior art, the embodiment of the present invention
Middle device 40 obtains user's mobile data, and is pre-processed, and builds General Mobile sequence according to pretreated result, then root
Short-term forecast and long-term forecast are carried out respectively according to General Mobile sequence, and the result of short-term forecast is superimposed upon to the result of long-term forecast
On, obtain the result of moving projection.Due to after obtaining user's mobile data and pre-processing, reducing the influence of interference, and adopt
It is predicted with the prediction technique of interference robust, improves the precision of prediction, be capable of the motion track in Accurate Prediction user's future.
The embodiment of the method for above-mentioned offer may be implemented in a kind of device of position prediction provided in an embodiment of the present invention, specifically
Function realizes the explanation referred in embodiment of the method, and details are not described herein.A kind of position prediction provided in an embodiment of the present invention
Method and device can be adapted for position prediction, but be not limited only to this.
Each embodiment in this specification is described in a progressive manner, identical similar portion between each embodiment
Point just to refer each other, and each embodiment focuses on the differences from other embodiments.Especially for equipment reality
For applying example, since it is substantially similar to the method embodiment, so describing fairly simple, related place is referring to embodiment of the method
Part explanation.
One of ordinary skill in the art will appreciate that realizing all or part of flow in above-described embodiment method, being can be with
Relevant hardware is instructed to complete by computer program, the program can be stored in a computer read/write memory medium
In, the program is when being executed, it may include such as the flow of the embodiment of above-mentioned each method.Wherein, the storage medium can be magnetic
Dish, CD, read-only memory(Read-Only Memory, ROM)Or random access memory(Random Access
Memory, RAM)Deng.
The above description is merely a specific embodiment, but scope of protection of the present invention is not limited thereto, any
Those familiar with the art in the technical scope disclosed by the present invention, all answer by the change or replacement that can be readily occurred in
It is included within the scope of the present invention.Therefore, protection scope of the present invention should be subject to the protection scope in claims.
Claims (10)
1. a kind of method of position prediction, which is characterized in that including:
Obtain user's mobile data;
User's mobile data is pre-processed, General Mobile sequence is built;
Short-term forecast and long-term forecast are carried out respectively according to the General Mobile sequence;
The result of the short-term forecast is superimposed upon in the result of the long-term forecast, obtains the result of moving projection;
It is described the short-term forecast to be carried out according to the General Mobile sequence respectively and the long-term forecast includes:
Subsequence storage is gathered to index;Definition starting index;Calculate the relatively described starting index of index of position to be predicted
Offset;According to the offset, the corresponding subsequence of index of the position to be predicted is determined;According to the subsequence,
Calculate prediction sets;According to quantile parameter, the quantile of the prediction sets is calculated, obtains short-term forecast result;
According to movement law, preset time is divided into disjoint group;Determine out-of-date mobile sequence;According to the General Mobile
Sequence and the out-of-date mobile sequence, calculate the cumulative number of the quantity and discretization position of universal location;It is waited for described in determination pre-
Survey the return group and same day index of location index;Calculate separately the discretization position of the position to be predicted cumulative number divided by
The result of the quantity of the universal location of the position to be predicted;Determine that the cell ID that the result is not less than threshold value is long-term
The result of prediction.
2. according to the method described in claim 1, it is characterized in that, user's mobile data includes:Cell residing for user
Mark, user reach the time point of the cell and collect the number of user's mobile data.
3. according to the method described in claim 2, it is characterized in that, pretreatment user's mobile data, structure are general
Mobile sequence includes:
User's mobile data described in discretization obtains discrete mobile data;
The value that General Mobile parameter is arranged is the first preset value;
According to the discrete mobile data and original discrete mobile data, the General Mobile sequence is built, it is described original discrete
The acquisition time point of mobile data is before the acquisition time point of the discrete mobile data, and close to the discrete mobile data
Acquisition time point, the numbers of original discretization data is first preset value.
4. according to the method described in claim 3, it is characterized in that, described according to the discrete mobile data and original discrete shifting
Dynamic data, building the General Mobile sequence includes:
According to the discrete mobile data and original discrete mobile data subsequence is determined in the pre-set interval of collection frequence
Beginning and end, the subsequence be user residing for cell ID frequent switching discrete mobile data, the frequent switching
The cell ID number be not more than first preset value, the pre-set interval length is the second preset value, described default
The initial value in section is the initial collection frequence of original discrete mobile data;
The subsequence is indicated with identical universal location, constitutes the General Mobile sequence.
5. according to the method described in claim 1, it is characterized in that, it is described according to the General Mobile sequence and it is described cross time shift
Dynamic sequence, the cumulative number of the quantity and discretization position that calculate universal location include:
For the discretization position of the General Mobile sequence, the cumulative number adds 1;
For the discretization position of the out-of-date mobile sequence, the cumulative number subtracts 1;
For the universal location of the General Mobile sequence, the quantity of the universal location adds 1;
For the universal location of the out-of-date mobile sequence, the quantity of the universal location subtracts 1.
6. a kind of device of position prediction, which is characterized in that including:
Acquiring unit, for obtaining user's mobile data;
Construction unit builds General Mobile sequence for pre-processing user's mobile data;
Predicting unit, for carrying out short-term forecast and long-term forecast respectively according to the General Mobile sequence;
Superpositing unit, the result for the result of the short-term forecast to be superimposed upon to the long-term forecast, obtains moving projection
Result;
The predicting unit includes:Storing sub-units, for gathering subsequence storage to index;
Subelement is defined, for defining starting index;First computation subunit, for calculating the index of position to be predicted with respect to institute
State the offset of starting index;First determination subelement, for according to the offset, determining the index of the position to be predicted
The corresponding subsequence;First computation subunit is additionally operable to, according to the subsequence, calculate prediction sets;Described first
Computation subunit is additionally operable to calculate the quantile of the prediction sets according to quantile parameter, obtains the short-term forecast result;
The predicting unit further includes:Subelement is divided, for according to movement law, preset time being divided into disjoint
Group;Second determination subelement, for determining out-of-date mobile sequence;Second computation subunit, for according to the General Mobile sequence
Row and the out-of-date mobile sequence, calculate the cumulative number of the quantity and discretization position of universal location;Described second determines son
Unit is additionally operable to determine the return group of the location index to be predicted and same day index;Second computation subunit is additionally operable to point
The quantity of the cumulative number of the discretization position of the position to be predicted divided by the universal location of the position to be predicted is not calculated
Result;Second determination subelement is additionally operable to determine that the cell ID that the result is not less than threshold value is described long-term pre-
The result of survey.
7. device according to claim 6, which is characterized in that user's mobile data includes:Cell residing for user
Mark, user reach the time point of the cell and collect the number of user's mobile data.
8. device according to claim 7, which is characterized in that the construction unit includes:
Discrete subelement obtains discrete mobile data for user's mobile data described in discretization;
Subelement is set, and the value for General Mobile parameter to be arranged is the first preset value;
Subelement is built, for according to the discrete mobile data and original discrete mobile data, building the General Mobile sequence
Row, the acquisition time point of original discrete mobile data before the acquisition time point of the discrete mobile data, and close to
The number of the acquisition time point of the discrete mobile data, original discretization data is first preset value.
9. device according to claim 8, which is characterized in that the structure subelement is specifically used for:
According to the discrete mobile data and original discrete mobile data subsequence is determined in the pre-set interval of collection frequence
Beginning and end, the subsequence be user residing for cell ID frequent switching discrete mobile data, the frequent switching
The cell ID number be not more than first preset value, the pre-set interval length is the second preset value, described default
The initial value in section is the initial collection frequence of original discrete mobile data;
The subsequence is indicated with identical universal location, constitutes the General Mobile sequence.
10. device according to claim 6, which is characterized in that second computation subunit is specifically used for:
For the discretization position of the General Mobile sequence, the cumulative number adds 1;
For the discretization position of the out-of-date mobile sequence, the cumulative number subtracts 1;
For the universal location of the General Mobile sequence, the quantity of the universal location adds 1;
For the universal location of the out-of-date mobile sequence, the quantity of the universal location subtracts 1.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310739194.2A CN104754730B (en) | 2013-12-27 | 2013-12-27 | The method and device of position prediction |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310739194.2A CN104754730B (en) | 2013-12-27 | 2013-12-27 | The method and device of position prediction |
Publications (2)
Publication Number | Publication Date |
---|---|
CN104754730A CN104754730A (en) | 2015-07-01 |
CN104754730B true CN104754730B (en) | 2018-07-31 |
Family
ID=53593665
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201310739194.2A Active CN104754730B (en) | 2013-12-27 | 2013-12-27 | The method and device of position prediction |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104754730B (en) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109309956B (en) * | 2017-07-28 | 2020-11-24 | 中国移动通信集团山西有限公司 | Position locating method, device, equipment and computer readable storage medium |
CN108009134A (en) * | 2017-12-26 | 2018-05-08 | 武汉大学 | A kind of Information of Ancient Human Activity method for digging |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102194316A (en) * | 2011-03-23 | 2011-09-21 | 中兴通讯股份有限公司 | Method and system for acquiring road condition information in real time |
CN102509170A (en) * | 2011-10-10 | 2012-06-20 | 浙江鸿程计算机系统有限公司 | Location prediction system and method based on historical track data mining |
CN102918898A (en) * | 2010-06-03 | 2013-02-06 | 索尼爱立信移动通讯有限公司 | Method and apparatus for location prediction |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP4818803B2 (en) * | 2006-05-01 | 2011-11-16 | 株式会社エヌ・ティ・ティ・ドコモ | Radio communication method and radio communication apparatus based on variable TTI length control |
-
2013
- 2013-12-27 CN CN201310739194.2A patent/CN104754730B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102918898A (en) * | 2010-06-03 | 2013-02-06 | 索尼爱立信移动通讯有限公司 | Method and apparatus for location prediction |
CN102194316A (en) * | 2011-03-23 | 2011-09-21 | 中兴通讯股份有限公司 | Method and system for acquiring road condition information in real time |
CN102509170A (en) * | 2011-10-10 | 2012-06-20 | 浙江鸿程计算机系统有限公司 | Location prediction system and method based on historical track data mining |
Also Published As
Publication number | Publication date |
---|---|
CN104754730A (en) | 2015-07-01 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US20220221552A1 (en) | Providing transit information | |
US10332019B2 (en) | Ranking nearby destinations based on visit likelihoods and predicting future visits to places from location history | |
Gong et al. | Inferring trip purposes and uncovering travel patterns from taxi trajectory data | |
US9285231B2 (en) | Providing transit information | |
Ahas et al. | Everyday space–time geographies: using mobile phone-based sensor data to monitor urban activity in Harbin, Paris, and Tallinn | |
Oliveira et al. | On the regularity of human mobility | |
Scellato et al. | Nextplace: a spatio-temporal prediction framework for pervasive systems | |
CN102460534A (en) | Predicting expected road traffic conditions based on historical and current data | |
Yazici et al. | Analysis of travel time reliability in New York city based on day-of-week and time-of-day periods | |
Zhang et al. | Accuracy analysis of freeway traffic speed estimation based on the integration of cellular probe system and loop detectors | |
KR101642487B1 (en) | Method for predicting user's future location and, apparatus and method for providing contents using the same method | |
Demiryurek et al. | Towards k-nearest neighbor search in time-dependent spatial network databases | |
CN106028444A (en) | Method and device for predicting location of mobile terminal | |
Mazumdar et al. | An approach to compute user similarity for GPS applications | |
CN104754730B (en) | The method and device of position prediction | |
US10419298B2 (en) | Monitoring device and memory medium for monitoring communication between terminal devices in a group | |
Deeshma et al. | Travel time modeling for bus transport system in Bangalore city | |
Khan et al. | Ensuring energy efficient coverage for participatory sensing in urban streets | |
Kesharwani et al. | Empowering bus transportation system using wireless sensor networks | |
Jariyasunant | Improving traveler information and collecting behavior data with smartphones | |
JP2016048529A (en) | Apparatus, program and method capable of determining stay of user carrying mobile terminal | |
Zhao et al. | This place is swarming: using a mobile social app to study human traffic in cities | |
CN109297492A (en) | A kind of determination method and device of the parked point of motion track | |
Anastasiou et al. | Time-Dependent Reachability Analysis: A Data-Driven Approach | |
Lüders et al. | Macroscopic Patterns in Sparse Location Data: Identifying Mobility Prototypes |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
TR01 | Transfer of patent right |
Effective date of registration: 20210423 Address after: Unit 3401, unit a, building 6, Shenye Zhongcheng, No. 8089, Hongli West Road, Donghai community, Xiangmihu street, Futian District, Shenzhen, Guangdong 518040 Patentee after: Honor Device Co.,Ltd. Address before: 518129 Bantian HUAWEI headquarters office building, Longgang District, Guangdong, Shenzhen Patentee before: HUAWEI TECHNOLOGIES Co.,Ltd. |
|
TR01 | Transfer of patent right |