CN107750051A - The optimization method and its device of radio transmission model - Google Patents
The optimization method and its device of radio transmission model Download PDFInfo
- Publication number
- CN107750051A CN107750051A CN201710912056.8A CN201710912056A CN107750051A CN 107750051 A CN107750051 A CN 107750051A CN 201710912056 A CN201710912056 A CN 201710912056A CN 107750051 A CN107750051 A CN 107750051A
- Authority
- CN
- China
- Prior art keywords
- data
- block
- radio transmission
- value
- model
- 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.)
- Granted
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W24/00—Supervisory, monitoring or testing arrangements
- H04W24/02—Arrangements for optimising operational condition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W24/00—Supervisory, monitoring or testing arrangements
- H04W24/06—Testing, supervising or monitoring using simulated traffic
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W28/00—Network traffic management; Network resource management
- H04W28/02—Traffic management, e.g. flow control or congestion control
- H04W28/06—Optimizing the usage of the radio link, e.g. header compression, information sizing, discarding information
Abstract
The embodiment of the invention discloses a kind of optimization method of radio transmission model and its device, wherein, method includes:According to GIS map data acquisition plane map;Region division and convergence processing are carried out to plane map to obtain multiple blocks;The corresponding relation of MR data and user terminal location is established to obtain target data, and base station user and user terminal location are belonged into block;If the data volume in block reaches preset data threshold value, regression fit is carried out to each block to obtain multiple block radio transmission models based on wireless universal propagation model, using big data technology and target data;Multiple block radio transmission models are carried out to test checking and iteration optimization.Implement the embodiment of the present invention, participated in without artificial, fitting precision is high, improves model optimization effect, and high efficiency and time conservation.
Description
Technical field
The present invention relates to mobile communication technology field, and in particular to the optimization method and its dress of a kind of radio transmission model
Put.
Background technology
Radio transmission model is to mobile communication network planning and optimization, including the configuration of base station selection, base station work ginseng and use
Family precise positioning etc., value are particularly significant.At present it is known that various typical scenes radio transmission model up to ten it is several, often
Kind model has the clear and definite theoretical scope of application (scene), and its parameters/coefficients, which has, clearly to be defined.But due to actually using field
Scape typical scene theoretical absolutely not, the polymorphic intertexture of actual scene, complicated and diversified influence be present, these theoretical models are in practical application
When, it is extremely difficult to theoretic Expected Results.Base station needs that are unreasonable or handing over dimension are planned in the site for causing mobile base station by this
Secondary, three suboptimization even more suboptimization, or the positioning of the user based on wireless base station are inaccurate.
Further, typical radio propagation model has the characteristics that simple, time saving.But wirelessly passed due to influenceing electromagnetic wave
Factor during broadcasting is more, even if being applicable the model of a certain typical scene, other scene individual differences are also very big, especially exist
During actual environment complexity, any single scene typical radio passes model and is all difficult to directly be applicable, therefore, typical radio propagation model
Error it is larger.Based on this, in actually applicable typical radio propagation model, typically all it is optimized.Specifically, it is existing
There is the optimization method to be usually:A large amount of artificial actual measurement sampled datas (such as height above sea level, electromagnetic interference, weather interference) are first passed through,
Typical radio propagation model is optimized by above-mentioned sampled data again.The optimization method is wasted time and energy, and due to sampled point
Limited, model optimization effect is also limited.
The content of the invention
The purpose of the embodiment of the present invention is the optimization method and its device for providing a kind of radio transmission model, to improve mould
Type effect of optimization, and high efficiency and time conservation.
To achieve the above object, in a first aspect, the embodiments of the invention provide a kind of optimization method of radio transmission model,
Comprise the following steps:
According to GIS map data acquisition plane map;
Region division and convergence processing are carried out to the plane map to obtain multiple blocks;
The corresponding relation of MR data and user terminal location is established to obtain target data, and by base station location and the use
Family terminal location belongs into the block;
If the data volume in the block reaches preset data threshold value, based on wireless universal propagation model, big number is utilized
Regression fit is carried out to obtain multiple block radio transmission models to each block according to technology and the target data;
Multiple block radio transmission models are carried out to test checking and iteration optimization.
As an alternative embodiment, region division and convergence processing are carried out to plane map to obtain multiple blocks
Specifically include:
Region division is carried out to the plane map according to Essential Geomorphic Features;
Convergence processing is carried out to the plane map after region division according to related geomorphic feature, and by homogeneous region root
It is divided into multiple blocks according to distribution characteristics.
As an alternative embodiment, the corresponding relation for establishing MR data and the user terminal location specifically wraps
Include:
Obtain MR data, signaling data and base station location;
The MR data are closed by base coded, MME marks, UE marks and sampling time with the signaling data
Join to obtain user IMEI;
User's OTT data are obtained, parse the user OTT data to obtain user terminal location;
By the user IMEI and the user terminal location to establish pair of MR data and the user terminal location
It should be related to.
As an alternative embodiment, each block is carried out using big data technology and the target data
Regression fit is specifically included with obtaining block radio transmission model:
For any block, according to the initial prediction of any sampled point in the target data calculating cycle T and
Actual value;
According to the predicted value and the deviation of any sampled point of calculated with actual values;
The total deviation value of all sampled points in each block is calculated according to the deviation;
Each parameter in wireless universal propagation model is modified to obtain target prediction value according to the total deviation value;
Loop iteration is carried out according to the target prediction value and actual value to calculate to obtain the wireless universal propagation model
In target component;
The target component is substituted into the wireless universal propagation model to obtain block radio transmission model.
As an alternative embodiment, the block radio transmission model is carried out to test checking and iteration optimization tool
Body includes:
Obtain multiple new sampled points;
Any new sampled point is predicted by the block radio transmission model to obtain any described newly adopt
The predicted value and actual value of sampling point;
The predicted value of any new sampled point and the deviation of actual value are calculated, and calculates multiple new sampled points
The mean error of deviation;
If the mean error is more than default error threshold, optimization is iterated to the block radio transmission model.
Second aspect, the embodiment of the present invention additionally provide a kind of optimization device of radio transmission model, including:
Acquisition module, for according to GIS map data acquisition plane map;
First processing module, for carrying out region division and convergence processing to the plane map to obtain multiple blocks;
Module is established, for establishing the corresponding relation of MR data and the user terminal location to obtain target data;
Belong to module, for base station location and the user terminal location to be belonged into the block;
Second processing module, if reaching preset data threshold value for the data volume in the block, based on wireless universal
Propagation model, using big data technology and the target data regression fit is carried out to obtain multiple blocks to each block
Radio transmission model;
3rd processing module, for carrying out testing checking and iteration optimization to multiple block radio transmission models.
As an alternative embodiment, the first processing module is specifically used for:
Region division is carried out to the plane map according to Essential Geomorphic Features;
Convergence processing is carried out to the plane map after region division according to related geomorphic feature, and by homogeneous region root
It is divided into multiple blocks according to distribution characteristics.
As an alternative embodiment, the module of establishing is specifically used for:
Obtain MR data, signaling data and base station location;
The MR data are closed by base coded, MME marks, UE marks and sampling time with the signaling data
Join to obtain user IMEI;
User's OTT data are obtained, parse the user OTT data to obtain user terminal location;
The user IMEI and the user OTT data are associated to establish MR data and the user terminal location
Corresponding relation.
As an alternative embodiment, the Second processing module is specifically used for:
For any block, according to the initial prediction of any sampled point in the target data calculating cycle T and
Actual value;
According to the predicted value and the deviation of any sampled point of calculated with actual values;
The total deviation value of all sampled points in each block is calculated according to the deviation;
Each parameter in wireless universal propagation model is modified to obtain target prediction value according to the total deviation value;
Loop iteration is carried out according to the target prediction value and actual value to calculate to obtain the wireless universal propagation model
In target component;
The target component is substituted into the wireless universal propagation model to obtain block radio transmission model.
As an alternative embodiment, the 3rd processing module is specifically used for:
Obtain multiple new sampled points;
Any new sampled point is predicted by the block radio transmission model to obtain any described newly adopt
The predicted value and actual value of sampling point;
The predicted value of any new sampled point and the deviation of actual value are calculated, and calculates multiple new sampled points
The mean error of deviation;
If the mean error is more than default error threshold, optimization is iterated to the block radio transmission model.
The optimization method and its device for the radio transmission model that the embodiment of the present invention is provided, first according to GIS map data
Obtain multiple blocks, resettle the corresponding relation of MR data and user terminal location, afterwards based on wireless universal propagation model, adopt
Regression fit is carried out to obtain block radio transmission model to block with big data technology, finally block radio transmission model entered
Row test checking and iteration optimization, sample size is sufficiently large, and without manually participating in, fitting precision is high, improves model optimization effect
Fruit, and high efficiency and time conservation.
Brief description of the drawings
, below will be to specific in order to illustrate more clearly of the specific embodiment of the invention or technical scheme of the prior art
The required accompanying drawing used is briefly described in embodiment or description of the prior art.In all of the figs, similar element
Or part is typically identified by similar reference.In accompanying drawing, each element or part might not be drawn according to the ratio of reality.
Fig. 1 is the schematic flow diagram of the optimization method for the radio transmission model that first embodiment of the invention provides;
Fig. 2 is the sub-process figure of step S103 in Fig. 1;
Fig. 3 is the sub-process figure of step S104 in Fig. 1;
Fig. 4 is the structural representation of the optimization device for the radio transmission model that first embodiment of the invention provides.
Embodiment
The embodiment of technical solution of the present invention is described in detail below in conjunction with accompanying drawing.Following examples are only used for
Clearly illustrate technical scheme, therefore be intended only as example, and the protection of the present invention can not be limited with this
Scope.
It should be noted that unless otherwise indicated, technical term or scientific terminology used in this application should be this hair
The ordinary meaning that bright one of ordinary skill in the art are understood.
Fig. 1 is refer to, is the flow signal of the optimization method for the radio transmission model that first embodiment of the invention is provided
Figure, as illustrated, this method may include steps of:
S101, according to GIS map data acquisition plane map.
By taking Chongqing as an example, according to GIS-Geographic Information System (Geographic Information System, GIS) map number
According to the plane map that can get Chongqing.
S102, region division and convergence processing are carried out to plane map to obtain multiple blocks.
Specifically, region division first is carried out to plane map according to Essential Geomorphic Features.Wherein, Essential Geomorphic Features include
Plain, mountain region, river, high building area (more than 6 floor), in the type such as low building area (6 floor and following) and road.Further according to correlation
Plane map after region division is carried out convergence processing by geomorphic feature, and homogeneous region is divided into according to distribution characteristics multiple convex
Polygon block, further, also each block is encoded.Wherein, high building area low building area with may be defined as phase
Close geomorphic feature.
For example, region division is first carried out to Chongqing plane map with forming region figure according to Essential Geomorphic Features, should
Administrative division map includes A areas, B areas, C areas, D areas and E areas etc..Administrative division map is subjected to convergence processing to obtain further according to related geomorphic feature
Homogeneous region, such as A areas belong to homogeneous region 1, B areas and D areas with C areas and belong to homogeneous region 2, and by homogeneous region according to distribution
Feature is divided into multiple convex polygon blocks, such as homogeneous region 1 and homogeneous region 2 are formed into convex polygon block 1.It is appreciated that
Ground, according to the above method, multiple convex polygon blocks can be formed on plane map, and each block is encoded, for example, weight
Celebrating block is encoded to 023.
S103, establish the corresponding relation of MR data and user terminal location to obtain target data, and by base station location and
User terminal location belongs into block.
Further, as shown in Fig. 2 step S103 is specifically included:
S1031, obtain MR data, signaling data and base station location;
Specifically, MR (Measurement Report, measurement report) data are obtained from base station, from mobile management entity
(Mobility Management Entity, MME) place obtains signaling data, and base station engineering parameter is obtained from database, should
Base station engineering parameter includes base station location;
S1032, MR data are associated by base coded, MME marks, UE marks and sampling time and signaling data
To obtain user IMEI;
Specifically, MR data are passed through into base coded (eci), MME marks, UE (user equipment, user equipment)
Mark and sampling time are associated to obtain user IMEI (International Mobile Equipment with signaling data
Identity, International Mobile Equipment Identity code);
S1033, user's OTT data are obtained, parse user OTT data to obtain user terminal location;
OTT data are Over The Top, refer to and think that user provides various application services by internet, map APP is it
A kind of middle application service.When user is using map APP etc., gateway can obtain user's OTT data, and user OTT data are carried out
Parse to obtain user terminal location;
S1034, by user IMEI user terminal locations to establish the corresponding relation of MR data and user terminal location.
It should be noted that after establishing the corresponding relation of MR data and user terminal location, target data can be obtained, should
Target data is used for following model parameter fitting, and detailed process will be described in detail in subsequent content, and will not be repeated here.Further
Ground, establish the corresponding relation of MR data and user terminal location, also achieve by base station location and user terminal location belong to
In the block of foregoing division.
S104, if the data volume in block reaches preset data threshold value, based on wireless universal propagation model, utilize big number
Regression fit is carried out to each block to obtain multiple block radio transmission models according to technology and the regression fit parameter.
S1041, for either block, according to the initial prediction of any sampled point in regression fit parameter calculating cycle T
And actual value;
S1042, according to predicted value and the deviation of any sampled point of calculated with actual values;
S1043, the total deviation value of all sampled points in each block is calculated according to deviation;
S1044, each parameter in wireless universal propagation model is modified to obtain target prediction according to total deviation value
Value;
S1045, loop iteration is carried out according to target prediction value and actual value and calculated to obtain in wireless universal propagation model
Target component;
S1046, target component is substituted into wireless universal propagation model to obtain block radio transmission model.
For preferably illustrative step S1041 to S1046, now it is explained as follows:
With use of the user to mobile network, the data volume in each block is continuously increased.For single block, when
When its data volume reaches predetermined threshold value (100,000), wireless universal propagation model can be based on, intended using big data technology and recurrence
Close parameter and regression fit is carried out to each block to obtain multiple block radio transmission models.
It should be noted that for block, due to density of stream of people, the accumulative speed of some block datas is fast,
Some block datas add up speed it is slow, generally the block in Chongqing entirely reach 100,000 data take around one about when
Between, herein refer to when most slow time, be exactly the minimum block of density of stream of people, to accumulate about one months that 100,000 records need
Time., can be by the affiliated administrative region of block by block classification, because administrative region to a certain extent may be used in practical operation
To characterize density of stream of people, such as main city zone is just more than the people in outer suburbs, so main city zone data are more, the setting of threshold value can it is big
A bit (such as 100,000), outer suburbs data are few, a little bit smaller (such as 50,000) that threshold value is just set.
On the other hand, because administrative region area is excessive, internal environment is unduly complex, and is unfavorable for being fitted, so one
Individual administrative region usually contains multiple GIS blocks.
Wherein, wireless universal propagation model is as follows:
Lmodel=K1+K2*log(d)+K3*log(HTexff)+K4*DiffractionLoss+K5*log(d)*
log(HTexff)+K6*HRxeff+Kclutter*f(clutter)
There is involved parameter in the model:K1, constant, unit dB;K2, constant;D, the distance of receiver to emitter
(the relative position distance of user and base station), unit is rice;K3, constant;HTexff, transmitting antenna height equivlent (antenna for base station etc.
Effect height), unit is rice;K4, constant;Diffraction loss, line penetrate consume, unit dB;K5, constant;K6, constant;
HRxeff, reception antenna height equivlent (terminal antenna height equivlent), unit is rice;Kclutter, constant factor;F (clutter),
Landform average loss, it is relevant with landforms.Above-mentioned parameter in model can be taking human as manually setting following initial value:K1:160.93;
K2:44.90;K3:-13.82;K4:0.20;K5:-6.55;K6:0.00;Kcluter:1.00.
For any one block, there is following anticipation function:
Lmodel=F (d, HTexff, Diffraction Loss, HRxeff, f (clutter))
(1) for any one block, in each calculating cycle T, unequal number amount n (n can be obtained>=105) position
The corresponding relation with comment loss is put, relevant parameter is substituted into above-mentioned anticipation function, the initial of each sampled point i can be obtained
Predicted value Lpre=F (di, Htxeffi,Diffraction Lossi,HRexffi, f (clutter) i) and actual value Ltrue;Wherein,
Relevant parameter refers to d, HTexff、Diffraction loss、HRxeffIt is and resulting in f (clutter) etc., i.e. step S103
Target data.The relevant parameter is resulting after MR data associate with user terminal location.
(2) the deviation d of any sampled point is calculatedL=Ltrue–Lpre+λ;λ is the regular terms for preventing over-fitting.
(3) for sampled point all in each block, total deviation value ∑ d is calculated using step (1) and the method for (2)L
=∑ (wt*dL)/∑wt, wherein, according to the time gap current time (model training time) of each sampled point, assign each
Sampled point difference weight wi。
It should be noted that a period of time T (T typically takes 3 months) is preset, for 0.5 times of T of time between T
Sampled point weights are 1,0.5 times of T sampled point are less than for the time, weights reduce and are incremented by with the time, for time T and 2 times of T
Between sampled point, weights increase over time and successively decrease, sufficient in available sampling point quantity for 2 times of more than T sampled point
In the case of, it is not involved in calculating.
(4) the total deviation value is the function H (K on parameters in wireless universal propagation model1, K2, K3, K4, K5, K6,
Kclutter), function H is for each parameter partial derivative.
(5) Ki '=Ki+dH ' ki*dki are modified by local derviation for each parameter.
(6) for the parameters of last time amendment, new predicted value (i.e. target prediction value) can be obtained.
Step (1) to (6) is repeated, optimal parameter K can be obtained1、K2、K3、K4、K5、K6And Kclutter, can be with
The radio transmission formula of the block is obtained, i.e., carrying out loop iteration according to target prediction value and actual value calculates to obtain general nothing
Target component in line propagation model, target component is substituted into wireless universal propagation model to obtain block radio transmission model.
It should be noted that when repeating step (1) to (6), following two end conditions be present:Cycle-index is more than
Preset times N;Average deviation ∑ dL/ n is less than preset value, and n is sampled point number.
It should also be noted that, the process described by step S104, can regard a model training as.First time model
During training, the parameter K in wireless universal propagation model1、K2、K3、K4、K5、K6And KcluterInitial value be manually set, and
In model training next time, the initial value of above-mentioned parameter is then the result after last model training.
S105, multiple block radio transmission models are carried out to test checking and iteration optimization.
Step S105 includes test checking and two processes of iteration optimization.Wherein, it is as follows to test verification process:
Due to system automation gathered data, using the time sampling point obtained after last time model training, that is, obtain multiple
New sampled point.Then, any new sampled point is predicted by block radio transmission model formula, to obtain any new sampling
The predicted value and actual value of point, and the predicted value of any new sampled point and the deviation of actual value are calculated, further, calculate
The mean error of the deviation of multiple new sampled points.According to the mean error and new sampled point number, it can represent that block is wireless
The accuracy of propagation model.
Iterative optimization procedure is as follows:
When the mean error of test checking is more than default error threshold, restarts above-mentioned model training program, model is entered
Row re -training.
Implement the optimization method for the radio transmission model that the embodiment of the present invention is provided, first obtained according to GIS map data
Multiple blocks, the corresponding relation of MR data and user terminal location is resettled, afterwards based on wireless universal propagation model, using big
Data technique carries out regression fit to block to obtain block radio transmission model, and finally block radio transmission model is surveyed
Experiment card and iteration optimization, sample size is sufficiently large, and without manually participating in, fitting precision is high, improves model optimization effect, and
High efficiency and time conservation.
Fig. 4 is refer to again, is the structural representation of the optimization device for the radio transmission model that first embodiment of the invention provides
Figure, as illustrated, the optimization device includes:
Acquisition module 10, for according to GIS map data acquisition plane map;
First processing module 11, for carrying out region division and convergence processing to plane map to obtain multiple blocks;
Module 12 is established, for establishing the corresponding relation of MR data and user terminal location to obtain target data;
Belong to module 13, for base station location and user terminal location to be belonged into block;
Second processing module 14, if reaching predetermined threshold value for the data volume in block, based on wireless universal propagating mode
Type, using big data technology and target data regression fit is carried out to each block to obtain multiple block radio transmission models;
3rd processing module 15, for carrying out testing checking and iteration optimization to multiple block radio transmission models.
Further, first processing module 11 is specifically used for:
Region division is carried out to plane map according to Essential Geomorphic Features;
According to related geomorphic feature to after region division plane map carry out convergence processing, and by homogeneous region according to divide
Cloth feature is divided into multiple blocks.
Further, module 12 is established to be specifically used for:
Obtain MR data, signaling data and base station location;
MR data are associated to obtain by base coded, MME marks, UE marks and sampling time and signaling data
User IMEI;
User's OTT data are obtained, parse the user OTT data to obtain user terminal location;
The corresponding relation of MR data and user terminal location is established by user IMEI and user terminal location.
Further, Second processing module 14 is specifically used for:
For either block, according to the initial prediction and actual value of any sampled point in target data calculating cycle T;
According to predicted value and the deviation of any sampled point of calculated with actual values;
The total deviation value of all sampled points in each block is calculated according to deviation;
Each parameter in wireless universal propagation model is modified to obtain target prediction value according to total deviation value;
Loop iteration is carried out according to target prediction value and actual value to calculate to obtain the target in wireless universal propagation model
Parameter;
Target component is substituted into wireless universal propagation model to obtain block radio transmission model.
Further, the 3rd processing module 15 is specifically used for:
Obtain multiple new sampled points;
Any new sampled point is predicted by block radio transmission model to obtain the predicted value of any new sampled point
And actual value;
The predicted value of any new sampled point and the deviation of actual value are calculated, and calculate the deviation of multiple new sampled points
Mean error;
If mean error is more than predetermined threshold value, optimization is iterated to block radio transmission model.
Implement the optimization device for the radio transmission model that the embodiment of the present invention is provided, first obtained according to GIS map data
Multiple blocks, the corresponding relation of MR data and user terminal location is resettled, afterwards based on wireless universal propagation model, using big
Data technique carries out regression fit to block to obtain block radio transmission model, and finally block radio transmission model is surveyed
Experiment card and iteration optimization, sample size is sufficiently large, and without manually participating in, fitting precision is high, improves model optimization effect, and
High efficiency and time conservation.
It should be noted that the specific workflow of the optimization device of radio transmission model refer to method in the present embodiment
Partial description, will not be repeated here.
Various embodiments above is merely illustrative of the technical solution of the present invention, rather than its limitations;Although with reference to foregoing each reality
Example is applied the present invention is described in detail, it will be understood by those within the art that:It still can be to foregoing each
Technical scheme described in embodiment is modified, and either carries out equivalent substitution to which part or all technical characteristic;And
These modifications are replaced, and the essence of appropriate technical solution is departed from the scope of various embodiments of the present invention technical scheme, its
It all should cover among the claim of the present invention and the scope of specification.
Claims (10)
1. a kind of optimization method of radio transmission model, it is characterised in that comprise the following steps:
According to GIS map data acquisition plane map;
Region division and convergence processing are carried out to the plane map to obtain multiple blocks;
The corresponding relation of MR data and user terminal location is established to obtain target data, and base station location and the user is whole
End position belongs into the block;
If the data volume in the block reaches preset data threshold value, based on wireless universal propagation model, big data skill is utilized
Art and the target data carry out regression fit to obtain multiple block radio transmission models to each block;
Multiple block radio transmission models are carried out to test checking and iteration optimization.
2. the optimization method of radio transmission model as claimed in claim 1, it is characterised in that region is carried out to plane map and drawn
Divide and convergence processing is specifically included with obtaining multiple blocks:
Region division is carried out to the plane map according to Essential Geomorphic Features;
According to related geomorphic feature to after region division the plane map carry out convergence processing, and by homogeneous region according to divide
Cloth feature is divided into multiple blocks.
3. the optimization method of radio transmission model as claimed in claim 1, it is characterised in that establish MR data and user terminal
The corresponding relation of position specifically includes:
Obtain MR data, signaling data and base station location;
By the MR data by base coded, MME marks, UE marks and sampling time and the signaling data be associated with
Obtain user IMEI;
User's OTT data are obtained, parse the user OTT data to obtain user terminal location;
By the user IMEI and the user terminal location, to establish, MR data are corresponding with the user terminal location to close
System.
4. the optimization method of the radio transmission model as described in any one of claims 1 to 3, it is characterised in that utilize big data
Technology and the target data carry out regression fit to each block and specifically included with obtaining block radio transmission model:
For any block, according to the initial prediction and reality of any sampled point in the target data calculating cycle T
Value;
According to the predicted value and the deviation of any sampled point of calculated with actual values;
The total deviation value of all sampled points in each block is calculated according to the deviation;
Each parameter in wireless universal propagation model is modified to obtain target prediction value according to the total deviation value;
Loop iteration is carried out according to the target prediction value and actual value to calculate to obtain in the wireless universal propagation model
Target component;
The target component is substituted into the wireless universal propagation model to obtain block radio transmission model.
5. the optimization method of radio transmission model as claimed in claim 4, it is characterised in that to the block radio transmission mould
Type carries out testing checking and iteration optimization specifically includes:
Obtain multiple new sampled points;
Any new sampled point is predicted by the block radio transmission model to obtain any new sampled point
Predicted value and actual value;
The predicted value of any new sampled point and the deviation of actual value are calculated, and calculates the deviation of multiple new sampled points
The mean error of value;
If the mean error is more than default error threshold, optimization is iterated to the block radio transmission model.
A kind of 6. optimization device of radio transmission model, it is characterised in that including:
Acquisition module, for according to GIS map data acquisition plane map;
First processing module, for carrying out region division and convergence processing to the plane map to obtain multiple blocks;
Module is established, for establishing the corresponding relation of MR data and the user terminal location to obtain target data;
Belong to module, for base station location and the user terminal location to be belonged into the block;
Second processing module, if reaching preset data threshold value for the data volume in the block, propagated based on wireless universal
Model, using big data technology and the target data to carry out regression fit to each block wireless to obtain multiple blocks
Propagation model;
3rd processing module, for carrying out testing checking and iteration optimization to multiple block radio transmission models.
7. the optimization device of radio transmission model as claimed in claim 6, it is characterised in that the first processing module is specific
For:
Region division is carried out to the plane map according to Essential Geomorphic Features;
According to related geomorphic feature to after region division the plane map carry out convergence processing, and by homogeneous region according to divide
Cloth feature is divided into multiple blocks.
8. the optimization device of radio transmission model as claimed in claim 6, it is characterised in that the module of establishing specifically is used
In:
Obtain MR data, signaling data and base station location;
By the MR data by base coded, MME marks, UE marks and sampling time and the signaling data be associated with
Obtain user IMEI;
User's OTT data are obtained, parse the user OTT data to obtain user terminal location;
The corresponding relation of MR data and the user terminal location is established by the user IMEI and the user terminal location.
9. the optimization device of the radio transmission model as described in claim any one of 6-8, it is characterised in that the second processing
Module is specifically used for:
For any block, according to the initial prediction and reality of any sampled point in the target data calculating cycle T
Value;
According to the predicted value and the deviation of any sampled point of calculated with actual values;
The total deviation value of all sampled points in each block is calculated according to the deviation;
Each parameter in wireless universal propagation model is modified to obtain target prediction value according to the total deviation value;
Loop iteration is carried out according to the target prediction value and actual value to calculate to obtain in the wireless universal propagation model
Target component;
The target component is substituted into the wireless universal propagation model to obtain block radio transmission model.
10. the optimization device of radio transmission model as claimed in claim 9, it is characterised in that the 3rd processing module tool
Body is used for:
Obtain multiple new sampled points;
Any new sampled point is predicted by the block radio transmission model to obtain any new sampled point
Predicted value and actual value;
The predicted value of any new sampled point and the deviation of actual value are calculated, and calculates the deviation of multiple new sampled points
The mean error of value;
If the mean error is more than default error threshold, optimization is iterated to the block radio transmission model.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710912056.8A CN107750051B (en) | 2017-09-29 | 2017-09-29 | Optimization method and device of wireless propagation model |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710912056.8A CN107750051B (en) | 2017-09-29 | 2017-09-29 | Optimization method and device of wireless propagation model |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107750051A true CN107750051A (en) | 2018-03-02 |
CN107750051B CN107750051B (en) | 2020-06-23 |
Family
ID=61255924
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710912056.8A Active CN107750051B (en) | 2017-09-29 | 2017-09-29 | Optimization method and device of wireless propagation model |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107750051B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110708665A (en) * | 2019-10-12 | 2020-01-17 | 中国联合网络通信集团有限公司 | Method and device for determining home area of access network equipment |
CN111817805A (en) * | 2020-07-08 | 2020-10-23 | 中国人民解放军国防科技大学 | Method, device and medium for adjusting channel propagation model parameters |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102065432A (en) * | 2009-11-13 | 2011-05-18 | 中国移动通信集团黑龙江有限公司 | Transmission model-based network coverage correcting method and system |
CN202617358U (en) * | 2012-04-12 | 2012-12-19 | 成都福兰特电子技术有限公司 | System for optimizing wireless network based on TD-LTE (time division long term evolution) |
US20160249193A1 (en) * | 2014-11-24 | 2016-08-25 | Qualcomm Incorporated | Location by reference for an over-the-top emergency call |
CN106231621A (en) * | 2016-07-29 | 2016-12-14 | 武汉大学 | A kind of many scene adaptives optimization method of propagation model in FDD LTE system |
CN106851570A (en) * | 2017-01-19 | 2017-06-13 | 中国移动通信集团江苏有限公司 | Based on the method and device that MR is positioned to mobile terminal |
CN107205237A (en) * | 2016-03-18 | 2017-09-26 | 中国移动通信集团贵州有限公司 | The optimization method and device of a kind of network structure |
-
2017
- 2017-09-29 CN CN201710912056.8A patent/CN107750051B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102065432A (en) * | 2009-11-13 | 2011-05-18 | 中国移动通信集团黑龙江有限公司 | Transmission model-based network coverage correcting method and system |
CN202617358U (en) * | 2012-04-12 | 2012-12-19 | 成都福兰特电子技术有限公司 | System for optimizing wireless network based on TD-LTE (time division long term evolution) |
US20160249193A1 (en) * | 2014-11-24 | 2016-08-25 | Qualcomm Incorporated | Location by reference for an over-the-top emergency call |
CN107205237A (en) * | 2016-03-18 | 2017-09-26 | 中国移动通信集团贵州有限公司 | The optimization method and device of a kind of network structure |
CN106231621A (en) * | 2016-07-29 | 2016-12-14 | 武汉大学 | A kind of many scene adaptives optimization method of propagation model in FDD LTE system |
CN106851570A (en) * | 2017-01-19 | 2017-06-13 | 中国移动通信集团江苏有限公司 | Based on the method and device that MR is positioned to mobile terminal |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110708665A (en) * | 2019-10-12 | 2020-01-17 | 中国联合网络通信集团有限公司 | Method and device for determining home area of access network equipment |
CN111817805A (en) * | 2020-07-08 | 2020-10-23 | 中国人民解放军国防科技大学 | Method, device and medium for adjusting channel propagation model parameters |
CN111817805B (en) * | 2020-07-08 | 2022-07-19 | 中国人民解放军国防科技大学 | Method, device and medium for adjusting channel propagation model parameters |
Also Published As
Publication number | Publication date |
---|---|
CN107750051B (en) | 2020-06-23 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Rhee et al. | On the levy-walk nature of human mobility | |
CN107333232B (en) | Terminal positioning method and network equipment | |
CN101267643B (en) | Method for improving network plan simulation precision | |
Mollel et al. | Comparison of empirical propagation path loss models for mobile communication | |
CN110418354A (en) | It is a kind of that propagation model wireless network planning method is exempted from based on machine learning | |
CN108306699A (en) | A kind of antenna feeder optimization method estimated based on gain | |
CN105163344B (en) | A kind of interior localization method interfered of TD-LTE system | |
CN107846688B (en) | Wireless network site planning method and device based on multiple operators | |
CN108540987B (en) | LTE network coverage state evaluation method and device | |
CN109526019A (en) | Wireless private network base station range propagation model revision method based on measured data | |
CN105101228B (en) | The pre- addressing algorithm of the modified multi-operator of quadruple | |
CN108259097B (en) | Method and device for correcting wireless propagation model based on MR data | |
CN107750051A (en) | The optimization method and its device of radio transmission model | |
Robinson | Hunting for oxes with sheaves | |
CN109165455B (en) | Hydrological dependent structure modeling method based on mutual information and vine copula | |
CN110326323A (en) | A kind of method and apparatus obtaining emission probability, transition probability and sequence positioning | |
CN109982349A (en) | Configuration method, configuration device, equipment and the medium of the different system adjacent cell of LTE | |
CN114828026A (en) | Base station planning method, device, equipment, storage medium and program product | |
CN104349326A (en) | Network planning method | |
Mazzoleni et al. | Can assimilation of crowdsourced streamflow observations in hydrological modelling improve flood prediction? | |
CN107682864A (en) | A kind of base station construction method assessed based on coverage rate | |
KR101694521B1 (en) | Apparatus and method for generating radio fingerprint map | |
CN108495262B (en) | Sparse representation and matching positioning method for indoor space ubiquitous positioning signal fingerprint database | |
CN107846689A (en) | The method that cordless communication network frequency planning is realized based on generalized information system | |
Li et al. | Research on TD-LTE wireless communication network propagation model optimization based on visual simulation and GIS |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |