CN103906106A - Automatic neighbor relation generating and optimizing method and system based on weight model - Google Patents

Abstract

The invention discloses an automatic neighbor relation generating and optimizing method and system based on a weight model. The method includes the steps that UE reports a measurement report; an eNB inquires whether an NCL is configured in a cell or not to an OSS according to a physical cell identity in the measurement report, and if not, SINR scanning is performed on the eNB which the cell belongs to, and the SINRieNBscan of a series of cells is obtained; the SINRTHreshold is obtained according to the SINR threshold selection strategy, whether the cell i meets the condition that the SINRieNBscan is larger than SINRThreshold or not is judged, and an initial NCL is formed; the weight values Wk of all the cells in the initial NCL are calculated according to the weight model; the weight values are compared, the initial NCL is reordered, and a new NCL is obtained. According to the automatic neighbor relation generating and optimizing method and system based on the weight model, when cell selection is performed, the types of neighbor cells and the cell load conditions can be fully taken into consideration, completeness and effectiveness of the NCL are ensured, the network capacity is improved, and the target of reducing the operation and maintenance cost is finally achieved.

Description

A kind of Automatic Neighboring Relation based on weight model produces optimization method and system

Technical field

The invention belongs to mobile communication technology field, particularly a kind of for LTE(Long Term Evolution, Long Term Evolution) Automatic Neighboring Relation based on weight model in system produces optimization method and system.

Background technology

LTE (Long Term Evolution, Long Term Evolution) project is the evolution of 3G, it is a transition between 3G and 4G technology, the global standards of 3.9G, it improves and has strengthened the aerial access technology of 3G, adopt orthogonal frequency division multiplex OFDM and the multiple-input and multiple-output MIMO sole criterion as its wireless network evolution, main purpose is the performance in order to have improved Cell Edge User, improves cell capacity and reduces system delay.

Along with the deployment of LTE, the quantity of network parameter and structure become more and more, become increasingly complex, next generation mobile networks requires highly intelligent, and network should have good reconstruct, scalability, self-organization, in order to meet the communication requirement of varying environment, different user.Therefore 3GPP LTE has proposed a kind of new O&M strategy, i.e. self-organizing network SON.LTE operator can obviously reduce O&M cost by SON mechanism, thereby further promotes the competitive advantage of LTE.SON refers to possess self-configuring (Self-Configuration), self-optimizing (Self-Optimization) and certainly recovers the wireless network of (Self-Healing).

Different from neighboring BS relationship allocation and optimization method in traditional wireless network is in LTE network, to have introduced SON ANR(Automatic Neighbor Relation, Automatic Neighboring Relation) function, Automatic Neighboring Relation ANR function mainly comprises in LTE/frequently in and between system/frequently between automatic configuration and the self-optimizing of adjacent area.The object of ANR function is to allow operator free from the work of manual configuration adjacent area.ANR function allow OSS (Operation support system, OSS) to Neighboring Relation Table increase, the operations such as modification, deletion.If Neighboring Relation Table changes, can notify OSS.The object of application ANR function is exactly accurate as far as possible, complete configuring adjacent cell relation list NCL.Because if need to be switched to the community not having in NCL list, just there will be switching delay, there will be severe the phenomenon of call drop; On the other hand, not the community of neighboring BS relationship if comprised in NCL list, will waste network management overhead.

Therefore, the technical problem that those skilled in the art are devoted to solve is, propose a kind of Automatic Neighboring Relation and produce optimization method, can be in the time carrying out community selection, take into full account neighbor cell type and cell load, guarantee the completeness and efficiency of NCL, to improve network capacity, the final target that reduces O&M cost (Operating Expense, OPEX) that realizes.

Summary of the invention

For the problems referred to above, the present invention proposes a kind of Automatic Neighboring Relation based on weight model and produces optimization method and system, can be in the time carrying out community selection, take into full account neighbor cell type and cell load, guarantee the completeness and efficiency of NCL, to improve network capacity, the final target that reduces O&M cost that realizes.

For achieving the above object, the present invention takes following technical scheme, comprising:

(1) mobile subscriber terminal UE reports a measurement report about community.Report comprises all adjacent areas of physical-layer cell identifier Cell ID, the UE of community, the received power of each adjacent area reference signal, the load information of each adjacent area.

(2) whether base station eNB, according to the Physical Cell Identifier in measurement report, is furnished with neighbor cell list NCL to this community of OSS system queries, if do not had, the eNB under this community scans the signal to noise ratio SINR that obtains a series of communities ⁱ _eNBscan; If be furnished with NCL, this NCL is called to storage.

Select different signal-noise ratio threshold SINR according to the concrete condition of community _threshold, SINR thresholding is selected different, and systematic function is had to considerable influence.

Preferably, SINR thresholding selection strategy comprises: determine to arrange different SINR thresholdings according to different newly-increased cell types.If newly-increased eNB community is macrocell, the coverage of newly-increased eNB is greater than microcell type, and also can increase with the ultimate range of neighbor cell.In this case, SINR thresholding should be set to S1, if newly-increased eNB community is Microcell, because picocell coverage area is less, so be only that being positioned near newly-increased eNB community is neighbor cell.In this case, SINR thresholding should be set to S2, S1<S2.

Preferably, SINR thresholding selection strategy also comprises: according to the type of neighbor cell, different decisions arrange different SINR thresholdings.Because macrocell coverage area is larger, so the ratio Microcell of through-put power setting is higher and along with its signal attenuation of increase of distance is not very serious; On the other hand, the through-put power of Microcell is lower and signal attenuation is more serious.This is because the not equal reason of antenna height causes.Although therefore identical from the SINR intensity of dissimilar eNB from cell edge detect, the eNB place of Xin community detects to such an extent that SINR intensity is also different.So the situation when SINR value of the adjacent macro cell measuring in place, newly-increased community is Microcell higher than neighbor cell, based on above reason, in the time that neighbor cell is macrocell, SINR thresholding is set to S3; In the time that neighbor cell is Microcell, SINR thresholding is set to S4, S3>S4.

(3) obtain SINR according to SINR thresholding selection strategy _threshold, the signal to noise ratio of the community that judgement scanning obtains meets SINR ⁱ _eNBscan> SINR _threshold, Ze Ba community joins initial NCL, and the rank of community in list is i, is designated as cell i, i=1,2 ... N, N represents the length of initial NCL; If do not meet judgment condition, re-start SINR scanning.

(4) calculate the weighted value W of each community in initial NCL according to weight model _k, the signal to noise ratio SINR of cell i ⁱ _eNBscancan be abbreviated as Si, W _kprior probability factor of influence G (P _i) and the weighted value of pilot signal function f (Si),

k=1,2 ... N.

Preferably, G (P _i)=kP _i+ b;

G (P _i) represent prior probability factor of influence, wherein k and b can be represented by pilot signal strength and switching probability information, P _irepresent that UE is switched to the switching probability of adjacent area j from cell i.

(5) weighted value is compared; NCL list is resequenced by weighted value size, obtain new NCL, store.

Accordingly, the invention also discloses a kind of neighboring BS relationship based on weight and produce system, comprising:

Measure inquiry module, for receiving the measurement data of reporting of user, comprise all adjacent areas of physical-layer cell identifier Cell ID, the UE of community, received power, the load information of each adjacent area etc. of each adjacent area reference signal, and whether be furnished with neighbor cell list NCL to this community of OSS system queries;

Scanning judging module, for scanning the signal to noise ratio SINR that obtains a series of communities reception signals ⁱ _eNBscan, compare judgement with signal-noise ratio threshold: SINR ⁱ _eNBscan> SINR _threshold, obtain initial NCL;

Calculate poll order module, for calculating the weights W of initial NCLZhong Ge community _k, and polling ratio is, and weighted value is sorted;

Memory module, for storing the new NCL producing through processing.

Compared with prior art, the present invention, owing to taking above technical scheme, has the following advantages:

First, weight model can make when SINR thresholding arrange too small time, the community that in NCL is not neighbouring relations is deleted, reduce the redundancy rate of NCL, like this can the spending of minimizing system, reduce system operation maintenance cost.

Secondly, this model can make when SINR thresholding arrange too high time, the community that is originally neighbouring relations is not but added to NCLZhong community and joins in NCL, can reduce like this UE cutting off rate and raising handover success rate.

Finally, adopt the mode of calculating weight poll sequence can guarantee the completeness and efficiency of obtained neighbor cell list NCL.

Accompanying drawing explanation

Fig. 1 is that the Automatic Neighboring Relation based on weight model produces optimization method flow chart;

Fig. 2 is the affect schematic diagrames of different SINR thresholdings on adjacent area scope;

Fig. 3 selects SINR thresholding schematic diagram according to newly-increased cell type;

Fig. 4 is according to neighbor cell type selecting SINR thresholding schematic diagram;

Fig. 5 is that the Automatic Neighboring Relation based on weight model produces system block diagram;

Switching probability schematic diagram between Tu6Shi tri-communities.

Embodiment

For making the object, technical solutions and advantages of the present invention clearer, below in conjunction with accompanying drawing, the present invention is described in further detail.

The present invention proposes a kind of Automatic Neighboring Relation based on weight model and produces optimization method, can, in the time carrying out community selection, take into full account neighbor cell type and cell load, guarantee the completeness and efficiency of NCL, to improve network capacity, the final target that reduces O&M cost that realizes.

With reference to Fig. 1, provide a kind of automatic adjacent section based on weight model described in the embodiment of the present invention and produced optimization method flow chart.

Step 1:UE reports a measurement report about community;

In LTE system, in subscriber equipment moving process, measure while reporting thresholding reaching, mobile device is to base station reporting measurement reports, and report comprises all adjacent areas of physical-layer cell identifier Cell ID, the UE of community, received power, the load information of each adjacent area etc. of each adjacent area reference signal.

Whether step 2:eNB, according to the Physical Cell Identifier in measurement report, is furnished with neighbor cell list NCL to this community of OSS system queries, if do not had, performs step 3; If be furnished with NCL, this NCL is called to storage;

Step 3: the eNB under this community carries out SINR and scans the signal to noise ratio SINR that obtains a series of communities ⁱ _eNBscan;

With reference to Fig. 2, SINR thresholding is selected different, and systematic function is had to considerable influence.The adjacent area of Wei Xin community, figure small area 1～6 (Cell1～6) (New cell), 7～11(Cell7～11, community) be non-adjacent area.Along with selecting different SINR thresholdings, the scope of neighbor cell is also different.If it is too high that SINR thresholding arranges, and is only to have the community of high SINR can join in neighbor cell list NCL, so likely can lose some handover candidate cells (as the Cell6 in Fig. 2); On the contrary, if the setting of SINR thresholding is too low, join in NCL with regard to having many potential handover candidate cells, NCL will comprise the communities (as Cell7～11 in Fig. 2) that do not belong in a large number neighbouring relations like this, can increase like this expense of grid.

With reference to Fig. 3, SINR thresholding selection strategy comprises: determine to arrange different SINR thresholdings according to different newly-increased cell types.LTE network provides different cell types, so the cell type of newly-increased eNB can affect the setting of SINR thresholding.If newly-increased eNB community is macrocell, the coverage of newly-increased eNB is greater than microcell type, and also can increase with the ultimate range of neighbor cell, and in this case, SINR thresholding should be set to S1; If newly-increased eNB community is Microcell, because picocell coverage area is less, so be only that being positioned near newly-increased eNB community is neighbor cell, in this case, SINR thresholding should be set to S2, S1<S2.

With reference to Fig. 4, SINR thresholding selection strategy also comprises: according to the type of neighbor cell, different decisions arrange different SINR thresholdings.Because macrocell coverage area is larger, so the ratio Microcell of through-put power setting is higher and along with its signal attenuation of increase of distance is not very serious; On the other hand, the through-put power of Microcell is lower and signal attenuation is more serious.This is because the not equal reason of antenna height causes.Although therefore identical from the SINR intensity of dissimilar eNB from cell edge detect, the eNB place of Xin community detects to such an extent that SINR intensity is also different.So the situation when SINR value of the adjacent macro cell measuring in place, newly-increased community is Microcell higher than neighbor cell, based on above reason, in the time that neighbor cell is macrocell, SINR thresholding is set to S3; In the time that neighbor cell is Microcell, SINR thresholding is set to S4, S3>S4.

Consider above two kinds of situations, select suitable SINR thresholding SINR _threshold.

Step 4: obtain SINR according to SINR thresholding selection strategy _threshold, the signal to noise ratio of the community that judgement scanning obtains meets SINR ⁱ _eNBscan> SINR _threshold, Ze Ba community joins initial NCL, and the rank of community in list is i, is designated as cell i, i=1,2 ... N, N represents the length of initial NCL; If do not meet judgment condition, return to step 3 and re-start SINR scanning;

Step 6: utilize weight model to calculate the weight of each community in initial NCL;

The probability that moves to different neighbor cells according to UE in pilot signal strength and real system carrys out the weights W of calculation plot _k, the signal to noise ratio SINR of cell i ⁱ _eNBscancan be abbreviated as Si, W _kprior probability factor of influence G (P _i) and the weighted value of pilot signal function f (Si),

k=1,2 ... N.

Calculate the weighted value W of each community in initial NCL according to weight model _k,

Wi＝f(Si) （1）

f(Si)＝Si+S _base （2）

Preferably, S _basea very important parameter in this algorithm model, while getting empirical value 15dB, Optimal performance the best of this model.

In addition, consider the situation of real system, in LTE network, in eNB, preserve UE and be switched to from cell i the switching probability P of adjacent area j _i, add prior probability factor of influence G (P _i), the weight of cell i can be rewritten as:

Wi＝G(P _i)f(Si) （3）

Wi＝G(P _i)(Si+S _base) （4）

G(P _i)＝kP _i+b （5）

Preferably, G (P _i) represent prior probability factor of influence, wherein k and b can be represented by pilot signal strength and switching probability information, P _irepresent that UE is switched to the switching probability of adjacent area j from cell i.

In LTE network, suppose that the movement of subscriber equipment meets markoff process (Markov process), switching is all random switching, and the probability matrix from cell i to community j is:

With reference to Fig. 6, with Tu Zhong tri-community Cell A, Cell B, Cell C is example, the probability that Cell A is switched to Cell A is a, the probability that Cell A is switched to Cell B is b, the probability that Cell A is switched to Cell C is c, the probability that Cell B is switched to Cell A is d, and the probability that Cell B is switched to Cell B is e, and the probability that Cell B is switched to Cell C is f, the probability that Cell C is switched to Cell A is g, the probability that Cell C is switched to Cell B is h, and the probability that Cell C is switched to Cell C is i, switching probability matrix

$P=\left(\begin{array}{ccc}a& b& c\\ d& e& f\\ g& h& i\end{array}\right)---\left(7\right)$

P _i=[P] [P _i-1]=[P _i-1] [P] (8)

By above calculating, weighted value is weighted and is obtained:

Step 7: the weighted value obtaining is calculated in each community in initial NCL and compare, judgement W _i> W _j, the sequence of cell i in advance; Otherwise the sequence of Ze Ba community j in advance;

Step 8,9: N community polling ratio in initial NCL sorted;

Preferably, each community in initial NCL is through the processing of above step, by the descending rearrangement of weight.

Step 10: the NCL newly obtaining is stored.

In sum, when the application sorts in definite NCL small area, neighbor cell type and cell load have been considered, and then according to weighted value, rank is weighted to processing, guarantee the completeness and efficiency of NCL, to improve network capacity, the final target that reduces O&M cost that realizes.

With reference to Fig. 5, a kind of Automatic Neighboring Relation based on weight model having provided described in the embodiment of the present application produces system block diagram.

The invention also discloses a kind of neighboring BS relationship based on weight and produce system, comprising:

Measure inquiry module, for receiving the measurement data of reporting of user, comprise all adjacent areas of physical-layer cell identifier Cell ID, the UE of community, received power, the load information of each adjacent area etc. of each adjacent area reference signal, and whether be furnished with neighbor cell list NCL to this community of OSS system queries;

Scanning judging module, for scanning the SINR that obtains a series of communities ⁱ _eNBscan, judgement SINR ⁱ _eNBscan> SINR _threshold, obtain initial NCL;

Calculate poll order module, for calculating the weights W of initial NCLZhong Ge community _k, and polling ratio sorts;

Memory module, for storing the new NCL producing through processing.

A kind of automatic adjacent section based on weight model above the application being provided produces optimization method and system is described in detail, applied principle and the execution mode of specific embodiment to the application herein and set forth, the explanation of above embodiment is just for helping to understand the application's method and core concept thereof.

Claims (6)

1. the Automatic Neighboring Relation based on weight model produces an optimization method, it is characterized in that comprising following steps:

(1) in LTE system, user terminal UE reports the measurement report about community;

(2) base station eNB is according to the Physical Cell Identifier in measurement report, inquire about this community to OSS OSS and whether be furnished with neighbor cell list NCL, if carrying out signal to noise ratio SINR scanning, the eNB no, under this community obtains the signal to noise ratio SINR of the reception signal of a series of communities ⁱ _eNBscan;

(3) obtain signal-noise ratio threshold SINR according to SINR thresholding selection strategy _threshold, the signal to noise ratio of the community that judgement scanning obtains is greater than signal-noise ratio threshold value, meets SINR ⁱ _eNBscan> SINR _threshold, Ze Ba community joins initial NCL;

(4) calculate the weighted value W of each community in initial NCL according to weight model _k;

(5) weighted value is compared, initial NCL list is resequenced, obtain new NCL.

2. a kind of Automatic Neighboring Relation based on weight model according to claim 1 produces optimization method, it is characterized in that in step (1): in LTE system, in subscriber equipment moving process, measure while reporting thresholding reaching, mobile device is to base station reporting measurement reports, and report comprises all adjacent areas of physical-layer cell identifier Cell ID, the UE of community, the received power of each adjacent area reference signal, the load information of each adjacent area.

3. a kind of Automatic Neighboring Relation based on weight model according to claim 1 produces optimization method, it is characterized in that described SINR thresholding selection strategy comprises: determine to arrange different SINR thresholdings according to different newly-increased cell types; If newly-increased eNB community is macrocell, SINR thresholding is set to S1; If newly-increased eNB community is Microcell, SINR thresholding is set to S2; S1<S2.

4. a kind of Automatic Neighboring Relation based on weight model according to claim 3 produces optimization method, it is characterized in that described SINR thresholding selection strategy also comprises: different decisions arrange different SINR thresholdings according to the type of neighbor cell, in the time that neighbor cell is macrocell, SINR thresholding is set to S3; In the time that neighbor cell is Microcell, SINR thresholding is set to S4, S3>S4.

5. a kind of Automatic Neighboring Relation based on weight model according to claim 1 produces optimization method, it is characterized in that described weight model is

wherein, G (P _i)=kP _i+ b; G (P _i) represent prior probability factor of influence, wherein k and b are represented by pilot signal strength and switching probability information, P _irepresent that UE is switched to the switching probability of adjacent area j from cell i.

6. the Automatic Neighboring Relation based on weight model produces an optimization system, it is characterized in that, comprising:

Measure inquiry module, for receiving the measurement data of reporting of user, comprise all adjacent areas of physical-layer cell identifier Cell ID, the UE of community, received power, the load information of each adjacent area etc. of each adjacent area reference signal, and whether be furnished with neighbor cell list NCL to this community of OSS system queries;

Scanning judging module, for scanning the signal to noise ratio SINR that obtains a series of communities reception signals ⁱ _eNBscan, compare judgement with signal-noise ratio threshold: SINR ⁱ _eNBscan> SINR _threshold, obtain initial NCL;

Calculate poll order module, for calculating the weights W of initial NCLZhong Ge community _k, and polling ratio is, and weighted value is sorted;

Memory module, for storing the new NCL producing through processing.

Images