CN105409306B - Mobile terminal locations prediction technique and device - Google Patents
Mobile terminal locations prediction technique and device Download PDFInfo
- Publication number
- CN105409306B CN105409306B CN201480009660.4A CN201480009660A CN105409306B CN 105409306 B CN105409306 B CN 105409306B CN 201480009660 A CN201480009660 A CN 201480009660A CN 105409306 B CN105409306 B CN 105409306B
- Authority
- CN
- China
- Prior art keywords
- track
- mobile terminal
- fragment group
- historical track
- trajectory
- 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
Abstract
A kind of mobile terminal locations prediction technique and device, next position of destination mobile terminal is predicted according to the historical track information of multiple mobile terminals, it obtains first and the historical track end of multiple users is clustered to obtain track fragment group, wherein, the historical track section of user divides to obtain from the entire historical track information of user.Then, target trajectory fragment group similar with the current track segment of target user is obtained from the fragment group of track, finally, predicting according to the next position of the orbit segment in target trajectory fragment group to target user.Historical track is divided into multiple historical track sections, each period, there is independent orbit segment therefore if sometime the track of mobile terminal is changed, the situation of change of track can be captured quickly, to improve the accuracy rate of prediction result.
Description
Technical field
The present invention relates to the prediction techniques and device of field of communication technology more particularly to mobile terminal locations.
Background technique
Base station can determine the current geographic position of the mobile terminal (user) in its coverage area, and base station records user's
Geographical location obtains the historical track information of user.Next position of user can be predicted by the historical track information of user
It sets.
Due to having certain relevance between some user's factum and the behavior of the user friend, it is existing
A kind of user location prediction technique having, the position of user is predicted using the trace information of the similar users with similar behavior
It sets, specifically, the entire historical track of user is analyzed as a whole, excavates the part of user behavior rule,
Establish suitable personal behavior model.And need to predict a certain moment of user location in future, using personal behavior model, and
Historical track information before a certain moment is matched, and obtains user in the location information at a certain moment.
Since user trajectory is continually changing, current similar user may also be no longer similar in future, causes to use
The position prediction result accuracy rate at family is low.In addition, as a whole by the entire historical track of user, the change to user behavior
To change insensitive, i.e., the nearest Behavioral change of user influences little in the probability distribution of entire behavior pattern, it is difficult to embody,
The prediction result accuracy rate for further resulting in user location reduces.
Summary of the invention
A kind of mobile terminal locations prediction technique and device are provided in the embodiment of the present invention, it is in the prior art to solve
The low problem of position prediction result accuracy rate.
In order to solve the above-mentioned technical problem, the embodiment of the invention discloses following technical solutions:
In a first aspect, the present invention provides a kind of mobile terminal locations prediction technique, comprising:
Track fragment group after obtaining the historical track section cluster to multiple mobile terminals, the historical track section is according to shifting
The historical track information of dynamic terminal divides to obtain；From the track fragment group, the current track with destination mobile terminal is obtained
The similar target trajectory fragment group of section；According to the target trajectory fragment group, determine the destination mobile terminal described current
Next position after orbit segment.
With reference to first aspect, in the first possible implementation of the first aspect, the method also includes: according to going through
Similarity between history orbit segment clusters the historical track section using clustering algorithm, obtains track fragment group.
The possible implementation of with reference to first aspect the first, in second of possible implementation of first aspect
In, according to the similarity between historical track section, the historical track section is clustered using clustering algorithm, comprising: respectively
Obtain the trajectory model of each historical track section；The similarity between the trajectory model is calculated, is obtained and the track
Similarity between the corresponding historical track section of mode；According to the similarity between the historical track section, similarity is met
Mono- track fragment group of historical track Duan Juwei of preset threshold.
The possible implementation of second with reference to first aspect, in the third possible implementation of first aspect
In, the trajectory model of each historical track section is obtained respectively, comprising: sequentially in time from each historical track section
Middle extraction obtains multiple sub-trajectories；Support is selected to be not less than the sub- rail of minimum support threshold value from the multiple sub-trajectory
Mark obtains the trajectory model of the historical track section.
The possible implementation of second with reference to first aspect, in the 4th kind of possible implementation of first aspect
In, calculate the similarity between the trajectory model, comprising: determine the public sub- sequence of longest between two trajectory models
Column；In conjunction with the position continuity parameter of trajectory model, the longest common subsequence is calculated respectively in two trajectory models
In shared ratio；Calculate the longest common subsequence respectively in described two trajectory models proportion average value,
Obtain the similarity between described two trajectory models.
With reference to first aspect or the first any one into the 4th kind of possible implementation, the of first aspect
In five kinds of possible implementations, according to the target trajectory fragment group, determine that the destination mobile terminal works as front rail described
Next position after mark section, comprising:
According to the corresponding historical track section of destination mobile terminal in the target trajectory fragment group, exist to destination mobile terminal
Next position of the current track segment is predicted, the first predictive information is obtained, and first predictive information includes prediction
Position and corresponding first probability value；
According to the corresponding historical track section of mobile terminals other in the target trajectory fragment group, exist to destination mobile terminal
Next position after current track segment is predicted, the second predictive information is obtained, and second predictive information includes prediction
Position and corresponding second probability value；
Merge first predictive information and the second predictive information, determines that the maximum predicted position of probability value is mobile for target
Next position of terminal.
The 5th kind of possible implementation with reference to first aspect, in the 6th kind of possible implementation of first aspect
In, merge first predictive information and the second predictive information, determines that the maximum corresponding predicted position of probability value is mobile for target
Next position of terminal, comprising:
Obtain corresponding first weight of first predictive information and corresponding second power of second predictive information
Weight；
By in first predictive information the first probability value and first multiplied by weight, obtain the first Weight prediction letter
Breath；
By in second predictive information the second probability value and second multiplied by weight, obtain the second Weight prediction letter
Breath；
Corresponding predicted position in the first Weight prediction information and the second Weight prediction information is corresponding
Probability value is overlapped merging, obtains merging predictive information；
Using the maximum predicted position of probability value in the merging predictive information as the described next of destination mobile terminal
Position.
Second aspect, present invention implementation also provide a kind of mobile terminal locations prediction meanss, comprising:
Receiving unit, it is described to go through for receiving the track fragment group after the historical track section to multiple mobile terminals clusters
History orbit segment divides to obtain according to the historical track information of mobile terminal；
Processing unit, for obtaining similar with the current track segment of destination mobile terminal from the track fragment group
Target trajectory fragment group；And according to the target trajectory fragment group, determine the destination mobile terminal in the current track
Next position after section.
In conjunction with second aspect, in the first possible implementation of the second aspect, the processing unit is also used to point
The trajectory model for not obtaining each historical track section, calculates the similarity between the trajectory model, obtains and the rail
Similarity between the corresponding historical track section of mark mode；According to the similarity between the historical track section, similarity is expired
Mono- track fragment group of historical track Duan Juwei of sufficient preset threshold.
In conjunction with the first possible implementation of second aspect, in second of possible implementation of second aspect
In, the processing unit, specifically for determining the longest common subsequence between two trajectory models, in conjunction with trajectory model
Position continuity parameter, calculate longest common subsequence ratio shared in two trajectory models respectively, meter
Calculate the longest common subsequence respectively in described two trajectory models proportion average value, obtain described two tracks
Similarity between mode.
In conjunction with any one of the first of second aspect or second aspect into second of possible implementation,
In the third possible implementation of two aspects, the processing unit is specifically used for:
According to the corresponding historical track section of destination mobile terminal in the target trajectory fragment group, exist to destination mobile terminal
Next position of the current track segment is predicted, the first predictive information is obtained, and first predictive information includes prediction
Position and corresponding first probability value；
According to the corresponding historical track section of mobile terminals other in the target trajectory fragment group, exist to destination mobile terminal
Next position after current track segment is predicted, the second predictive information is obtained, and second predictive information includes prediction
Position and corresponding second probability value；
Merge first predictive information and the second predictive information, determines that the maximum predicted position of probability value is mobile for target
Next position of terminal.
In conjunction with the third possible implementation of second aspect, in the 4th kind of possible implementation of second aspect
In, the processing unit is specifically used for:
Obtain corresponding first weight of first predictive information and corresponding second power of second predictive information
Weight；
By in first predictive information the first probability value and first multiplied by weight, obtain the first Weight prediction letter
Breath；
By in second predictive information the second probability value and second multiplied by weight, obtain the second Weight prediction letter
Breath；
Corresponding predicted position in the first Weight prediction information and the second Weight prediction information is corresponding
Probability value is overlapped merging, obtains merging predictive information；
Using the maximum predicted position of probability value in the merging predictive information as the described next of destination mobile terminal
Position.
By above technical scheme as it can be seen that mobile terminal locations prediction technique provided in an embodiment of the present invention and device, first
Acquisition clusters to obtain track fragment group to the historical track end of multiple users, wherein the historical track section of user is from the whole of user
A historical track information divides to obtain.Then, mesh similar with the current track segment of target user is obtained from the fragment group of track
Track fragment group is marked, finally, carrying out according to the next position of the orbit segment in target trajectory fragment group to target user
Prediction.
Historical track is divided into multiple historical track sections by the method, and each historical track section is historical track information
Therefore a part if sometime the track of mobile terminal is changed, can capture the situation of change of track quickly,
To improve the accuracy rate of prediction result.Moreover, traditional similar users (mobile terminal) are converted similar rail by the method
Mark section does not compare the corresponding historical track information of mobile terminal and believes as a whole with the historical track of other mobile terminals
Breath, judges the similitude between mobile terminal.But the corresponding historical track information of mobile terminal is divided into multiple history rails
Mark section judges the similitude between orbit segment, in such manner, it is possible to historical track that is subtleer, more accurately capturing mobile terminal
Similitude can further increase the accuracy rate of prediction result.
Detailed description of the invention
The accompanying drawings constituting a part of this application is used to provide further understanding of the present invention, and of the invention shows
Examples and descriptions thereof are used to explain the present invention for meaning property, does not constitute improper limitations of the present invention.In the accompanying drawings:
Fig. 1 shows a kind of flow diagram of mobile terminal locations prediction technique of the embodiment of the present invention；
Fig. 2 shows the method flow schematic diagrams of the step S120 in embodiment illustrated in fig. 1；
Fig. 3 shows the method flow schematic diagram of the step S140 in embodiment illustrated in fig. 1；
Fig. 4 shows a kind of sub-trajectory schematic diagram of the embodiment of the present invention；
Fig. 5 shows the method flow schematic diagram of the step S130 in embodiment illustrated in fig. 1；
Fig. 6 shows the method flow schematic diagram of the step S133 in embodiment illustrated in fig. 5；
Fig. 7 shows a kind of block diagram of mobile terminal locations prediction meanss of the embodiment of the present invention.
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, for those of ordinary skill in the art
Speech, without any creative labor, is also possible to obtain other drawings based on these drawings.
Specific embodiment
To achieve the purpose of the present invention, the present invention provides mobile terminal locations prediction technique and devices, mobile terminal
Entire historical track be divided into multiple historical track sections, and historical track section is clustered, similar historical track section is poly-
For a track fragment group；From the track fragment group, the moment to be predicted affiliated orbit segment with destination mobile terminal is obtained
Similar target trajectory fragment group determines destination mobile terminal in the prediction bits at moment to be predicted according to target trajectory fragment group
It sets.Historical track is divided into multiple historical track sections according to time response, each period has independent orbit segment, therefore,
If sometime the track of mobile terminal is changed, the situation of change of track can be captured quickly, to improve prediction
As a result accuracy rate.
It is core of the invention thought above, in order to make those skilled in the art more fully understand the present invention program, below
The attached drawing in the embodiment of the present invention will be combined, technical solution in the embodiment of the present invention is purged, is fully described by, and shows
So, the embodiment of the description is only a part of the embodiment of the present invention, instead of all the embodiments.Based on the reality in the present invention
Example is applied, every other embodiment obtained by those of ordinary skill in the art without making creative efforts all belongs to
In the scope of protection of the invention.
Referring to Figure 1, a kind of flow diagram of mobile terminal locations prediction technique of the embodiment of the present invention, the party are shown
Method is applied in wireless network.As shown in Figure 1, the method may include following steps:
Step S110, the track fragment group after obtaining the historical track section cluster to multiple mobile terminals, the history rail
Mark section divides to obtain according to the historical track information of mobile terminal.
It divides the historical track information of multiple mobile terminals to obtain historical track section.It is assumed that mobile terminal (that is, user)
In moving process, in tiMoment accesses active wireless network, wherein access position is li, i=1,2,3 ..., n, and in tn
Moment has left active wireless network, then defines S={ (t1, tn): < l1, l2..., ln> be the user a historical track
Section, historical track information are made of several orderly position sequences.The process that user moves in a network can generate multiple history
Orbit segment, all these historical tracks constitute the historical track information D={ S of the mobile terminal1, S2..., SN}.From user's
Each orbit segment therein is extracted in entire historical track, obtains the historical track section of the user.
After the historical track section for obtaining user, it can use clustering algorithm and the historical track section gathered in advance
Similarity is met mono- track fragment group of historical track Duan Juwei of default similarity, i.e., by similar historical track section by class
Gather for a track fragment group.If using cosine similarity measure historical track section between similitude, when historical track section it
Between cosine similarity be greater than preset threshold when, show that the similitude between corresponding historical track section is very big, can be divided into
In one group.
User location prediction need not be carried out every time to the process of historical track section cluster to be carried out, it can be in advance to database
In existing historical track section clustered, then, predicted using position of the cluster result to user.
Step S120 obtains target similar with the current track segment of destination mobile terminal from the track fragment group
Track fragment group.
In one embodiment of the invention, as shown in Fig. 2, step S120 may include step S121~S125:
Step S121 obtains the current track segment of destination mobile terminal；The current track segment is that destination mobile terminal exists
Orbit segment locating for current time.
Step S122, judges whether the similarity between the current track segment and track fragment group meets preset threshold.
If meeting preset threshold (for example, the similarity between current track segment and track fragment group is greater than default threshold
Value), in step S123, determine that the track fragment group is target trajectory fragment group.Then, then step S124 is executed.
If being unsatisfactory for preset threshold, in step S124, judge whether that whole track fragment groups have all judged.
If whole tracks fragment group has all judged, in step S125, target trajectory fragment group is determined.It will be with current track
The track fragment group that similarity between section meets preset threshold is determined as target trajectory fragment group.
If returning to step S122 there is also the track fragment group not judged, continue to judge next track stage group
Whether group is target trajectory fragment group.
Step S130 determines the destination mobile terminal after current track segment according to the target trajectory fragment group
Next position.
To next position of prediction destination mobile terminal (i.e. target user) after current track segment, target is utilized
The historical track section of target user itself in the fragment group of track and the historical track section of other users are respectively to described next
A position is predicted, obtains the prediction result comprising predicted position and corresponding probability, obtained prediction result is merged,
In, using the maximum position of probability value as next position of target user.
Signified user location prediction of the invention need to only predict next position of user current location, reach without considering
At the time of next position.Time only is used to do the sequence of historical position and time window division is used, that is, guarantees that user goes through
History trace information is ordered into sequence, and historical track section can be effectively divided according to the time.
Mobile terminal locations prediction technique provided in this embodiment, predicts mesh according to the historical track information of multiple users
Next position of user is marked, obtains first and the historical track end of multiple users is clustered to obtain track fragment group, wherein
The historical track section of user divides to obtain from the entire historical track information of user.Then, acquisition and mesh from the fragment group of track
The similar target trajectory fragment group of current track segment for marking user, finally, according to the orbit segment in target trajectory fragment group to mesh
It predicts next position of mark user.
Historical track information is divided into multiple historical track sections by the method, and each historical track section is historical track letter
Therefore a part of breath if sometime the track of user is changed, can capture quickly rail from historical track section
The situation of change of mark, to improve the accuracy rate of prediction result.Moreover, the method convert traditional similar users to it is similar
Orbit segment, i.e., be not by the corresponding historical track information of user as a whole with the historical track information ratio of other users
Compared with, judge the similitude between user, but the corresponding historical track information of user is divided into multiple historical track sections, judge
Similitude between orbit segment.In such manner, it is possible to the similitude between historical track that is subtleer, more accurately capturing user, energy
Enough further increase the accuracy rate of position prediction result.
Optionally, embodiment shown in FIG. 1 before step S110 can with the following steps are included:
Step S140 carries out the historical track section using clustering algorithm according to the similarity between historical track section
Cluster obtains track fragment group.
For example, when the cosine similarity between historical track section A and historical track section B is greater than preset threshold, then by A and B
It clusters to a track fragment group.
It can use classical clustering algorithm (for example, K-means algorithm), according to the similitude pair between historical track section
The corresponding historical track section of multiple users clusters, and obtains track fragment group.Historical track section in each track fragment group
All be it is similar, corresponding historical track section may be assigned to different track fragment groups to each user in different time period
In, and the historical track section of different user may be assigned in the same track fragment group.
It optionally, can be to newly-increased historical track section and existing when increasing new historical track section in database
Track fragment group is clustered using incremental clustering algorithm, obtains new cluster result.
Fig. 3 is referred to, the method flow schematic diagram of step S140 in Fig. 1 is shown, in the present embodiment, is obtaining history
Before similarity between orbit segment, trajectory model is extracted for each historical track section, then, based on the similar of trajectory model
Degree, clusters different historical track sections, obtains the similar track fragment group of trajectory model.
As shown in figure 3, the method may include steps:
Step S141 obtains the trajectory model of each historical track section respectively.
In one embodiment of the invention, 11)~12 step S141 may comprise steps of):
Step 11) is extracted to obtain multiple sub-trajectories from each historical track section sequentially in time.
If there is a track α, it includes position sequence, there are whole position sequences of track α in another track β, i.e. α is
The subset of β, such situation track α are referred to as the sub-trajectory of track β.
For example, length is track α=< α of m1, α2, α3..., αm>track sets β=<the β for being n with length1, β2,
β3..., βn>, 1≤k of integer if it exists1< k2< ... < km≤ n, so that α1=βk1, α2=βk2... αm=βkm, then claim track
α is the sub-trajectory of track β, i.e.,α is the subset of β.
Optionally, the entire historical track information of user is divided into several time windows, includes one in each time window
The user's history track of section time, i.e., each time window correspond to a sub-trajectory of the historical track section.It is assumed that
Time window length is T, then, is moved forward by Δ T of step-length, the corresponding sub-trajectory of time window 1 are as follows: t1:a, b, c, d, e, f；
The corresponding sub-trajectory of time window 2 are as follows: t2:e, f, g, h, i；The corresponding sub-trajectory of time window 3 are as follows: t3:h, i, j, k, l.
Step 12) selects support to be not less than the sub-trajectory of minimum support threshold value, obtains from the multiple sub-trajectory
The trajectory model of the historical track section.
Firstly, introducing the concept of support and trajectory model.
(1) support: the support of track α is the percentage shared in historical track section of the sub-trajectory comprising α.
(2) trajectory model: if support of some sub-trajectory in historical track section is more than minimum support threshold value, institute
State the trajectory model that sub-trajectory is known as the historical track section.
For example, table 1 is five sub-trajectories of historical track section α, if minimum support threshold value is 0.4.
Wherein, include in sub-trajectory 1,2,4<e,f>, therefore, sub-trajectory<e,f>support sup (<e,f>)=3/5
=0.6 > 0.4, therefore,<e, f>be historical track section α a trajectory model.Similarly, comprising <b, c in sub-trajectory 3,4,5
>, sub-trajectory<b,c>support sup (<b,c>)=3/5=0.6 > 0.4, therefore,<b,c>it is also the one of historical track section α
A trajectory model.That is the trajectory model of historical track section α is<e,f>with<b,c>.
Table 1
Step S142 calculates the similarity between the trajectory model, obtains history rail corresponding with the trajectory model
Similarity between mark section.
Calculate the similarity between trajectory model in one embodiment of the invention, it is first determined two trajectory models it
Between longest common subsequence then calculate longest common subsequence ratio shared in two trajectory models respectively,
The average value for calculating two ratios again obtains the similarity of two trajectory models.
In another embodiment of the present invention, step S142 may include step 21)~23):
Step 21) determines the longest common subsequence of two trajectory models；
Step 22) calculates longest common subsequence respectively described in two in conjunction with the position continuity parameter of trajectory model
Shared ratio in trajectory model.
The continuity of position in two tracks is different, then may indicate two kinds of entirely different behaviors.For example, a rail
Mark is " office -> nursery -> dining room ", it may be possible to a family party；And if track is " office -> dining room ", it can
It can be that one action is had a dinner party.It can be seen that the continuity between two positions is critically important factor, therefore, track is being calculated
The process of the similarity of mode consider position continuity parameter be completely it is necessary to.
Give two trajectory models α and β, it is assumed that the longest common subsequence of α and β is θ, then θ institute in trajectory model α
Accounting example R (α, θ) is as shown in formula 1:
Wherein, | α | indicate the quantity of position in trajectory model α, | θ | indicate the quantity of position in longest common subsequence θ,
hjIndicate the continuity parameter of two positions.
M (α in formula 1i, θi) as shown in formula 2:
In formula 1Indicate the continuity between two positions, δiIt is as shown in formula 3:
U, v in formula 3 respectively indicate αu、αvPosition in trajectory model, | u-v | between indicating between two positions
Every.For example, trajectory model are as follows:<a, b, c, d>, if αuFor b, αvFor d, then u=2, v=4, then | u-v |=2, indicate position b and d
Between interval.
For example, α: a- > b- of trajectory model > c- > d- > e, β: a- > c- of trajectory model > e- > f- > g.The public son of the longest of α and β
Sequence θ is { a- > c- > e }.
According to formula 1- formula 3, θ proportion R (α, θ)=(1+e in α is calculated-1/5·1+e-1/5·1)/5；Meter
Calculation obtains θ proportion R (β, θ)=(1+1+1)/5 in β, it can be seen that, the numerical value of R (α, θ) and R (β, θ) is not identical.
From the position sequence in α and β it is found that for α, before a place of arrival c of place, centre also be have passed through
Place b, and for β, then it is from the direct place of arrival c of place a, centre is without passing through other places.
Step 23), calculating the longest common subsequence, the proportion in described two trajectory models is averaged respectively
Value, obtains the similarity between described two trajectory models.
According to longest common subsequence θ ratio R (α, θ) and R (β, θ) shared in the trajectory model, two are calculated
Similarity sim (α, β) between trajectory model, sim (α, β) are the average value of θ proportion in α and β, as shown in formula 4:
Similarity is met the historical track section of preset threshold according to the similarity between historical track section by step S143
Gather for a track fragment group.
The similarity between the trajectory model of two historical track sections is calculated according to above-mentioned similarity calculating method, is made
For the similarity between two historical track sections, according to the similarity between historical track section, using clustering algorithm to whole
Historical track section is clustered, mono- track fragment group of historical track Duan Juwei with similar trajectory model.Each track
Historical track section in fragment group be all it is similar, historical track section of each user in different moments may be divided into difference
Track fragment group in, and the historical track section of different user may be divided into the same track fragment group.
The similarity calculation mode of historical track section provided in this embodiment is extracted from the historical track section of mobile terminal
Trajectory model calculates the similarity of trajectory model as the similarity between corresponding historical track section, the track mould extracted
The data volume of formula is far smaller than the data volume of historical track section, and therefore, the similarity calculated between trajectory model can subtract significantly
Subtotal calculates data volume, saves the resource inside the processing unit of occupancy, can reduce the requirement to processing unit performance.Moreover,
The present embodiment considers the continuity of position in trajectory model, that is, considers user's when calculating the similarity between trajectory model
The similarity of behavior, the trajectory model being calculated in this way is more acurrate, and the user behavior similitude of similar historical track section is more
Greatly.
Fig. 5 is referred to, the method flow diagram of step S130 shown in Fig. 1 is shown, includes in the target trajectory fragment group
The corresponding historical track section of destination mobile terminal and the corresponding historical track section of other mobile terminals.
As shown in figure 5, step S130 may comprise steps of:
S131, it is mobile to target according to the corresponding historical track section of destination mobile terminal in the target trajectory fragment group
Next position of the terminal after current track segment is predicted, the first predictive information is obtained.The first predictive information packet
Include predicted position and corresponding first probability value.
S132, it is mobile to target according to the corresponding historical track section of mobile terminals other in the target trajectory fragment group
Next position of terminal is predicted, the second predictive information is obtained.Second predictive information include predicted position and
Corresponding second probability value.
S133 merges first predictive information and the second predictive information, determines that the maximum predicted position of probability value is mesh
Mark next position of mobile terminal.
In one embodiment of the invention, merge the first predictive information and the second predictive information, it can be using such as lower section
The corresponding probability value of predicted position identical in the first predictive information and the second predictive information is added up, is obtained described pre- likes:
Total probability value that location is set.
For example, the first predictive information isSecond predictive information is Wherein,WithFor same position, then the predictive information after merging is From this
The corresponding prediction result of maximum probability value is found out in predictive information, next position as destination mobile terminal.
In another embodiment of the present invention, it is predicted using corresponding first weight of first predictive information and second
Second weight of information merges first predictive information and second predictive information.As shown in fig. 6, step S133 can be with
The following steps are included:
Step S1331, obtains corresponding first weight of the first predictive information and second predictive information is corresponding
The second weight.
First weight of the first predictive information is ξ, and the second weight of the second predictive information is (1- ξ), wherein ξ be greater than 0 and
Less than 1, in order to make the corresponding prediction result of historical track section of the more partial destination mobile terminal of prediction result itself after merging,
Therefore, 0.5 < ξ < 1 can be set.
First probability value of first predictive information and first multiplied by weight are obtained the first power by step S1332
Weight predictive information.
It is assumed that the first predictive information obtained according to the historical track section of the destination mobile terminal in target trajectory fragment group
ForThen the first Weight prediction information is
Second probability value of second predictive information and second multiplied by weight are obtained the second power by step S1333
Weight predictive information.
It is assumed that obtaining the second predictive information according to the historical track section of other mobile terminals in target trajectory fragment group Then the second Weight prediction information is
Wherein, the execution sequence of step S1332 and step S1333 can be interchanged, for example, step S1333 can be first carried out,
Step S1332 is executed again, and the present invention is not intended to limit this.
Step S1334, by the identical prediction in the first Weight prediction information and the second Weight prediction information
The corresponding probability value in position is overlapped merging, obtains merging predictive information.
Assuming that in the first predictive informationIn the second predictive informationFor same position, i.e.,Then the first power
Weight predictive information and the second Weight prediction information are merged into
Step S1335, using the maximum predicted position of probability value in the merging predictive information as destination mobile terminal
Next position.
Compare the size for merging the probability value in predictive information, the maximum predicted position of probability value is under destination mobile terminal
The prediction result of one position.
The method of next position of the prediction mobile terminal provided in this embodiment after current track segment, according to target
The historical track of destination mobile terminal obtains one group of predictive information in the fragment group of track, and, according in target trajectory fragment group
The historical track sections of other terminals obtain another group of predictive information, in conjunction with predictive information weight will in two groups of predictive information it is pre-
Location is set corresponding probability and is merged, and probability value is then compared, the maximum predicted position, that is, destination mobile terminal of probability value
The prediction result of next position.In conjunction with weight, can make destination mobile terminal predictive information and other mobile terminals it is pre-
Measurement information merges according to a certain percentage, weight is set according to the importance of predictive information, to improve the standard of prediction result
True rate.
Corresponding to above-mentioned embodiment of the method, present invention also provides Installation practices.
Fig. 7 is referred to, a kind of block diagram of mobile terminal locations prediction meanss of the embodiment of the present disclosure, the device application are shown
In wireless network, as shown in fig. 7, the apparatus may include: receiving unit 110 and processing unit 120.
Receiving unit 110 is configured as receiving the track fragment group after the historical track section cluster to multiple mobile terminals,
The historical track section divides to obtain according to the historical track information of mobile terminal.
The receiving unit 110 can be the structure that network element etc. has certain transfer function.
Processing unit 120 is configured as from the track fragment group, obtains the current track segment with destination mobile terminal
Similar target trajectory fragment group；And according to the target trajectory fragment group, determine that the destination mobile terminal is worked as described
Next position after preceding orbit segment.
In one embodiment of the invention, processing unit 120 obtains and destination mobile terminal from the fragment group of track
When the similar target trajectory fragment group of current track segment, specifically it is configurable to: obtains the current track segment of destination mobile terminal,
Whether within a preset range to judge similarity between the current track segment and track fragment group, if within a preset range,
Determine that the track fragment group is target trajectory fragment group.
In another embodiment of the present invention, processing unit 120 is described in determine according to the target trajectory fragment group
Destination mobile terminal is specifically configurable at next position after the current track segment:
According to the corresponding historical track section of destination mobile terminal in the target trajectory fragment group, exist to destination mobile terminal
Next position of the current track segment is predicted, the first predictive information is obtained, and first predictive information includes prediction
Position and corresponding first probability value；
According to the corresponding historical track section of mobile terminals other in the target trajectory fragment group, exist to destination mobile terminal
Next position after current track segment is predicted, the second predictive information is obtained, and second predictive information includes prediction
Position and corresponding second probability value；
Merge first predictive information and the second predictive information, determines that the maximum predicted position of probability value is mobile for target
Next position of terminal.
The processing unit 120 is merging first predictive information and the second predictive information, determines that probability value is maximum
When predicted position is next position of destination mobile terminal, specifically it is configurable to:
Obtain corresponding first weight of first predictive information and corresponding second power of second predictive information
Weight；
By in first predictive information the first probability value and first multiplied by weight, obtain the first Weight prediction letter
Breath；
By in second predictive information the second probability value and second multiplied by weight, obtain the second Weight prediction letter
Breath；
Corresponding predicted position in the first Weight prediction information and the second Weight prediction information is corresponding
Probability value is overlapped merging, obtains merging predictive information；
Using the maximum predicted position of probability value in the merging predictive information as the described next of destination mobile terminal
Position.
The processing unit 120 can be general processor (CPU), digital signal processor (DSP), specific integrated circuit
(ASIC), ready-made programmable gate array (FPGA) either other programmable logic device, discrete gate or transistor logic,
Discrete hardware components.It may be implemented or execute disclosed each method, step and the logic diagram in the embodiment of the present invention.It is general
Processor can be microprocessor or any conventional processor etc..
Mobile terminal locations prediction meanss provided in this embodiment receive the history rail to multiple users by receiving unit
The track fragment group that mark section is clustered, wherein the historical track section of user is drawn from the entire historical track information of user
Get.Then, target trajectory fragment group similar with the current track segment of target user is obtained from the fragment group of track, most
Afterwards, it is predicted according to the next position of the orbit segment in target trajectory fragment group to target user.By historical track
Information is divided into multiple historical track sections, and each historical track section is a part of historical track information, therefore, if sometime
The track of user is changed, and the situation of change of track can be captured quickly from historical track section, to improve prediction
As a result accuracy rate.Moreover, the corresponding historical track information of user is divided into multiple historical track sections, judge between orbit segment
Similitude, in this way can it is subtleer, more accurately capture user historical track between similitude, can further increase
The accuracy rate of position prediction result.
In another embodiment of the present invention, processing unit 120 is additionally configured to according to the phase between historical track section
Like degree, the historical track section is clustered using clustering algorithm, track fragment group is obtained, is supplied to the receiving unit
110。
In one embodiment of the invention, processing unit 120 is clustering historical track section using clustering algorithm
When, it is specifically configurable to: obtaining the trajectory model of each historical track section respectively；It calculates between the trajectory model
Similarity obtains the similarity between historical track section corresponding with the trajectory model；According between the historical track section
Similarity, similarity is met to mono- track fragment group of historical track Duan Juwei of preset threshold.In each track fragment group
Historical track section be all it is similar, corresponding historical track section may be assigned to difference to each user in different time period
Track fragment group in, and the historical track section of different user may be assigned in the same track fragment group.
In one embodiment of the invention, processing unit 120 is in the track for obtaining each historical track section respectively
It when mode, is specifically configurable to: being extracted from each historical track section obtain multiple sub-trajectories sequentially in time, from
It selects support to be not less than the sub-trajectory of minimum support threshold value in the multiple sub-trajectory, obtains the rail of the historical track section
Mark mode.
In one embodiment of the invention, processing unit 120 is calculating the corresponding trajectory model of two historical track sections
Between similarity when, be specifically configurable to: the longest common subsequence between two trajectory models determined, in conjunction with rail
The position continuity parameter of mark mode calculates longest common subsequence ratio shared in two trajectory models respectively
Example, then calculate the longest common subsequence respectively in described two trajectory models proportion average value, obtain described
Similarity between two trajectory models, using the similarity between trajectory model as similar between corresponding historical track section
Degree.
The predicted position device of mobile terminal provided in this embodiment extracts track from the historical track section of mobile terminal
Mode, calculates the similarity of trajectory model as the similarity between corresponding historical track section, the trajectory model extracted
Data volume is far smaller than the data volume of historical track section, and therefore, the similarity calculated between trajectory model can greatly reduce meter
Data volume is calculated, the resource inside the processing unit of occupancy is saved, can reduce the requirement to processing unit performance.Moreover, this reality
Example is applied when calculating the similarity between historical track section, the continuity of position in trajectory model is considered, that is, considers the row of user
For the similarity for the trajectory model being calculated in this way is more acurrate, and the user behavior similitude of similar historical track section is bigger.
All the embodiments in this specification are described in a progressive manner, same and similar portion between each embodiment
Dividing may refer to each other, and each embodiment focuses on the differences from other embodiments.Especially for device or
For system embodiment, since it is substantially similar to the method embodiment, so describing fairly simple, related place is referring to method
The part of embodiment illustrates.Apparatus and system embodiment described above is only schematical, wherein the conduct
The unit of separate part description may or may not be physically separated, component shown as a unit can be or
Person may not be physical unit, it can and it is in one place, or may be distributed over multiple network units.It can root
According to actual need that some or all of the modules therein is selected to achieve the purpose of the solution of this embodiment.Ordinary skill
Personnel can understand and implement without creative efforts.
The present invention can describe in the general context of computer-executable instructions executed by a computer, such as program
Module.Generally, program module includes routines performing specific tasks or implementing specific abstract data types, programs, objects, group
Part, data structure etc..The present invention can also be practiced in a distributed computing environment, in these distributed computing environments, by
Task is executed by the connected remote processing devices of communication network.In a distributed computing environment, program module can be with
In the local and remote computer storage media including storage equipment.
It should be noted that, in this document, the relational terms of such as " first " and " second " or the like are used merely to one
A entity or operation with another entity or operate distinguish, without necessarily requiring or implying these entities or operation it
Between there are any actual relationship or orders.Moreover, the terms "include", "comprise" or its any other variant are intended to
Cover non-exclusive inclusion, so that the process, method, article or equipment for including a series of elements not only includes those
Element, but also including other elements that are not explicitly listed, or further include for this process, method, article or setting
Standby intrinsic element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that
There is also other identical elements in the process, method, article or apparatus that includes the element.
The above is only a specific embodiment of the invention, it is noted that for the ordinary skill people of the art
For member, various improvements and modifications may be made without departing from the principle of the present invention, these improvements and modifications are also answered
It is considered as protection scope of the present invention.
Claims (14)
1. a kind of mobile terminal locations prediction technique characterized by comprising
Track fragment group after obtaining the historical track section cluster to multiple mobile terminals, the historical track section is according to mobile whole
The historical track information at end divides to obtain；
From the track fragment group, target trajectory fragment group similar with the current track segment of destination mobile terminal is obtained；
According to the target trajectory fragment group, next position of the destination mobile terminal after the current track segment is determined
It sets；
Wherein, according to the target trajectory fragment group, under determining the destination mobile terminal after the current track segment
One position, comprising:
According to the corresponding historical track section of destination mobile terminal in the target trajectory fragment group, to destination mobile terminal described
Next position of current track segment is predicted, obtains the first predictive information, first predictive information includes predicted position
With corresponding first probability value；
According to the corresponding historical track section of mobile terminals other in the target trajectory fragment group, to destination mobile terminal current
Next position after orbit segment is predicted, obtains the second predictive information, second predictive information includes predicted position
With corresponding second probability value；
Merge first predictive information and the second predictive information, determines that the maximum predicted position of probability value is destination mobile terminal
Next position.
2. the method according to claim 1, wherein the method also includes:
According to the similarity between historical track section, the historical track section is clustered using clustering algorithm, obtains track
Fragment group.
3. according to the method described in claim 2, it is characterized in that, utilizing cluster according to the similarity between historical track section
Algorithm clusters the historical track section, comprising:
The trajectory model of each historical track section is obtained respectively；
The similarity between the trajectory model is calculated, is obtained similar between historical track section corresponding with the trajectory model
Degree；
According to the similarity between the historical track section, similarity is met to mono- rail of historical track Duan Juwei of preset threshold
Mark fragment group.
4. according to the method described in claim 3, it is characterized in that, obtaining the track mould of each historical track section respectively
Formula, comprising:
It is extracted from each historical track section sequentially in time and obtains multiple sub-trajectories；
It selects support to be not less than the sub-trajectory of minimum support threshold value from the multiple sub-trajectory, obtains the historical track
The trajectory model of section.
5. according to the method described in claim 3, it is characterized in that, calculating the similarity between the trajectory model, comprising:
Determine the longest common subsequence between two trajectory models；
In conjunction with the position continuity parameter of trajectory model, the longest common subsequence is calculated respectively in two trajectory models
In shared ratio；
Calculate the longest common subsequence respectively in described two trajectory models proportion average value, obtain described two
Similarity between a trajectory model.
6. method according to claim 1-5, which is characterized in that from the track fragment group, acquisition and mesh
Mark the similar target trajectory fragment group of current track segment of mobile terminal, comprising:
Obtain the current track segment of destination mobile terminal；
Whether within a preset range similarity between the current track segment and track fragment group is judged, if in preset range
It is interior, determine that the track fragment group is target trajectory fragment group.
7. the method according to claim 1, wherein merge first predictive information and the second predictive information,
Determine that the maximum corresponding predicted position of probability value is next position of destination mobile terminal, comprising:
Obtain corresponding first weight of first predictive information and corresponding second weight of second predictive information；
By in first predictive information the first probability value and first multiplied by weight, obtain the first Weight prediction information；
By in second predictive information the second probability value and second multiplied by weight, obtain the second Weight prediction information；
By the corresponding probability of corresponding predicted position in the first Weight prediction information and the second Weight prediction information
Value is overlapped merging, obtains merging predictive information；
Using the maximum predicted position of probability value in the merging predictive information as next position of destination mobile terminal.
8. a kind of mobile terminal locations prediction meanss characterized by comprising
Receiving unit, for receiving the track fragment group after the historical track section to multiple mobile terminals clusters, the history rail
Mark section divides to obtain according to the historical track information of mobile terminal；
Processing unit, for obtaining target similar with the current track segment of destination mobile terminal from the track fragment group
Track fragment group；And according to the target trajectory fragment group, determine the destination mobile terminal the current track segment it
Next position afterwards；
The processing unit is specifically used for:
According to the corresponding historical track section of destination mobile terminal in the target trajectory fragment group, to destination mobile terminal described
Next position of current track segment is predicted, obtains the first predictive information, first predictive information includes predicted position
With corresponding first probability value；
According to the corresponding historical track section of mobile terminals other in the target trajectory fragment group, to destination mobile terminal current
Next position after orbit segment is predicted, obtains the second predictive information, second predictive information includes predicted position
With corresponding second probability value；
Merge first predictive information and the second predictive information, determines that the maximum predicted position of probability value is destination mobile terminal
Next position.
9. device according to claim 8, it is characterised in that:
The processing unit is also used to, according to the similarity between historical track section, using clustering algorithm to the historical track
Duan Jinhang cluster, obtains track fragment group.
10. device according to claim 9, it is characterised in that:
The processing unit calculates the track mould specifically for obtaining the trajectory model of each historical track section respectively
Similarity between formula obtains the similarity between historical track section corresponding with the trajectory model；According to the history rail
Similarity is met mono- track fragment group of historical track Duan Juwei of preset threshold by the similarity between mark section.
11. device according to claim 10, it is characterised in that:
The processing unit obtains multiple sub- rails specifically for extracting from each historical track section sequentially in time
Mark selects support to be not less than the sub-trajectory of minimum support threshold value, obtains the historical track from the multiple sub-trajectory
The trajectory model of section.
12. device according to claim 10, it is characterised in that:
The processing unit, specifically for determining the longest common subsequence between two trajectory models, in conjunction with track mould
The position continuity parameter of formula calculates longest common subsequence ratio shared in two trajectory models respectively,
Calculate the longest common subsequence respectively in described two trajectory models proportion average value, obtain described two rails
Similarity between mark mode.
13. according to the described in any item devices of claim 8-12, it is characterised in that:
The processing unit judges the current track segment and rail specifically for obtaining the current track segment of destination mobile terminal
Similarity between mark fragment group whether within a preset range, if within a preset range, determining that the track fragment group is mesh
Mark track fragment group.
14. device according to claim 8, which is characterized in that the processing unit is specifically used for:
Obtain corresponding first weight of first predictive information and corresponding second weight of second predictive information；
By in first predictive information the first probability value and first multiplied by weight, obtain the first Weight prediction information；
By in second predictive information the second probability value and second multiplied by weight, obtain the second Weight prediction information；
By the corresponding probability of corresponding predicted position in the first Weight prediction information and the second Weight prediction information
Value is overlapped merging, obtains merging predictive information；
Using the maximum predicted position of probability value in the merging predictive information as next position of destination mobile terminal.
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| PCT/CN2014/079655 WO2015188324A1 (en) | 2014-06-11 | 2014-06-11 | Method and apparatus for predicting location of mobile terminal |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105409306A CN105409306A (en) | 2016-03-16 |
| CN105409306B true CN105409306B (en) | 2019-01-18 |
Family
ID=54832704
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201480009660.4A Active CN105409306B (en) | 2014-06-11 | 2014-06-11 | Mobile terminal locations prediction technique and device |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN105409306B (en) |
| WO (1) | WO2015188324A1 (en) |
Families Citing this family (23)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107655490B (en) * | 2017-08-29 | 2020-03-17 | 重庆邮电大学 | Hot spot path discovery method based on mobile user track segmentation and hottest search |
| CN109768869B (en) * | 2017-11-06 | 2022-05-31 | 中国移动通信有限公司研究院 | Service prediction method, system and computer storage medium |
| CN108289279A (en) * | 2018-01-30 | 2018-07-17 | 浙江省公众信息产业有限公司 | Processing method, device and the computer readable storage medium of location information |
| CN109727141B (en) * | 2018-03-02 | 2021-06-04 | 中国平安人寿保险股份有限公司 | Information processing method, device and equipment based on vehicle moving card |
| FR3079309B1 (en) * | 2018-03-21 | 2020-07-17 | Sigfox | METHOD AND SYSTEM FOR GEOLOCATION OF TERMINALS EVOLVING IN A GROUP |
| CN108566618B (en) * | 2018-04-04 | 2020-07-28 | 广州杰赛科技股份有限公司 | Method, device, equipment and storage medium for acquiring user residence law |
| CN110928914A (en) * | 2018-08-30 | 2020-03-27 | 百度在线网络技术（北京）有限公司 | Method and apparatus for outputting information |
| CN109041217B (en) * | 2018-09-21 | 2020-01-10 | 北京邮电大学 | Hierarchical mobility prediction method in heterogeneous network |
| CN109041218B (en) * | 2018-09-25 | 2020-08-11 | 广东小天才科技有限公司 | Method for predicting user position and intelligent hardware |
| CN109388757B (en) * | 2018-10-10 | 2021-11-02 | 广州力挚网络科技有限公司 | Hot track extraction method and device |
| CN111723123B (en) * | 2019-03-20 | 2023-07-14 | 杭州海康威视数字技术股份有限公司 | Track prediction method, track prediction device, electronic equipment and storage medium |
| CN110209560B (en) * | 2019-05-09 | 2023-05-12 | 北京百度网讯科技有限公司 | Data anomaly detection method and detection device |
| CN112116805B (en) * | 2019-06-21 | 2022-04-05 | 杭州海康威视系统技术有限公司 | Method and device for determining specified driving route |
| CN110825833B (en) * | 2019-11-11 | 2022-05-17 | 杭州数澜科技有限公司 | Method for predicting user moving track point |
| CN111339449A (en) * | 2020-03-24 | 2020-06-26 | 青岛大学 | User motion trajectory prediction method, device, equipment and storage medium |
| CN111753214A (en) * | 2020-06-24 | 2020-10-09 | 平安科技（深圳）有限公司 | Data pushing method and system based on behavior track and computer equipment |
| CN111929715A (en) * | 2020-06-28 | 2020-11-13 | 杭州云起智慧校园科技有限公司 | Positioning method, device and equipment for school badge and storage medium |
| CN112667763B (en) * | 2020-12-29 | 2022-09-13 | 电子科技大学 | Track prediction method based on self-adaptive timestamp and multi-scale feature extraction |
| CN112613673A (en) * | 2020-12-29 | 2021-04-06 | 北京梧桐车联科技有限责任公司 | Travel track determination method and device and computer readable storage medium |
| CN112714398B (en) * | 2021-01-26 | 2024-03-29 | 上海明略人工智能(集团)有限公司 | Method, device and equipment for correcting positioning coordinate drift of indoor positioning system |
| CN113784362A (en) * | 2021-07-29 | 2021-12-10 | 三维通信股份有限公司 | Aerial base station deployment method, aerial base station deployment device, electronic device and storage medium |
| CN114756367B (en) * | 2022-04-15 | 2024-02-23 | 中国电信股份有限公司 | Service migration method, device, medium and electronic equipment |
| CN116558513B (en) * | 2023-07-06 | 2023-10-03 | 中国电信股份有限公司 | Indoor terminal positioning method, device, equipment and medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102175253A (en) * | 2010-12-28 | 2011-09-07 | 清华大学 | Multi-hypothesis map matching method based on vehicle state transition |
| WO2011150971A1 (en) * | 2010-06-03 | 2011-12-08 | Sony Ericsson Mobile Communications Ab | Method and apparatus for location prediction |
| CN102984799A (en) * | 2012-11-27 | 2013-03-20 | 中国人民解放军信息工程大学 | Method and system for judging user positions |
| CN103747523A (en) * | 2014-01-14 | 2014-04-23 | 上海河广信息科技有限公司 | User position predicating system and method based on wireless network |
-
2014
- 2014-06-11 WO PCT/CN2014/079655 patent/WO2015188324A1/en active Application Filing
- 2014-06-11 CN CN201480009660.4A patent/CN105409306B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2011150971A1 (en) * | 2010-06-03 | 2011-12-08 | Sony Ericsson Mobile Communications Ab | Method and apparatus for location prediction |
| CN102175253A (en) * | 2010-12-28 | 2011-09-07 | 清华大学 | Multi-hypothesis map matching method based on vehicle state transition |
| CN102984799A (en) * | 2012-11-27 | 2013-03-20 | 中国人民解放军信息工程大学 | Method and system for judging user positions |
| CN103747523A (en) * | 2014-01-14 | 2014-04-23 | 上海河广信息科技有限公司 | User position predicating system and method based on wireless network |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2015188324A1 (en) | 2015-12-17 |
| CN105409306A (en) | 2016-03-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105409306B (en) | Mobile terminal locations prediction technique and device | |
| Fan et al. | Citymomentum: an online approach for crowd behavior prediction at a citywide level | |
| CN104123398B (en) | A kind of information-pushing method and device | |
| Zhou et al. | Understanding crowd behaviors in a social event by passive wifi sensing and data mining | |
| Gidófalvi et al. | When and where next: Individual mobility prediction | |
| CN107018493B (en) | Geographic position prediction method based on continuous time sequence Markov model | |
| Chen et al. | Constructing and comparing user mobility profiles for location-based services | |
| CN109688532B (en) | Method and device for dividing city functional area | |
| Chen et al. | Predicting next locations with object clustering and trajectory clustering | |
| CN106792523B (en) | A kind of anomaly detection method based on extensive WiFi activity trajectory | |
| Zhang et al. | Accuracy analysis of freeway traffic speed estimation based on the integration of cellular probe system and loop detectors | |
| Ye et al. | B-Loc: Scalable floor localization using barometer on smartphone | |
| CN105120433A (en) | WLAN indoor positioning method based on continuous sampling and fuzzy clustering | |
| CN103929719B (en) | The optimization method and optimization device of location information | |
| Li et al. | Robust inferences of travel paths from GPS trajectories | |
| CN109326087B (en) | Urban waterlogging early warning method and device based on drainage pipe network monitoring | |
| CN108882172A (en) | A kind of prediction technique of the indoor moving track data based on HMM model | |
| CN110213710A (en) | A kind of high-performance indoor orientation method, indoor locating system based on random forest | |
| Umair et al. | Discovering personal places from location traces | |
| Furey et al. | HABITS: a Bayesian filter approach to indoor tracking and location | |
| de Graaff et al. | Automated semantic trajectory annotation with indoor point-of-interest visits in urban areas | |
| Yu et al. | Floor positioning method indoors with smartphone’s barometer | |
| Chen | Mining individual behavior pattern based on significant locations and spatial trajectories | |
| CN107133689B (en) | Position marking method | |
| Chen et al. | InferLoc: calibration free based location inference for temporal and spatial fine-granularity magnitude |
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 |