CN102149103A - Network optimizing system and method - Google Patents

Abstract

The invention discloses a network optimizing system and a network optimizing method. The system comprises a signaling acquisition device, a database server, a data processing module and an optimizing application module, wherein the signaling acquisition device is used for acquiring signaling data from a mobile communication network and storing the acquired signaling data; the database server is used for storing acquired network optimizing analysis auxiliary data and storing the processed signaling data; the data processing module is used for gathering network optimizing analysis basic data and the network optimizing analysis auxiliary data and accomplishing a multilayer database framework for data processing and application; and an optimizing worker calls an analysis result in a multilayer data framework in a data analysis module through the application module so as to realize work in the network optimizing process. By the network optimizing system and the network optimizing method, comprehensive data can be acquired, network problems can be truly and comprehensively reflected, and network performance can be further optimized comprehensively.

Description

Network optimization system and method

Technical field

The present invention relates to the mobile communication network optimization technical field, specifically a kind of mobile communication network optimization system and method.

Background technology

Mobile communication network optimization is normally defined mobile operator finish the communication network development after, network optimization personnel find by the multiple technologies means and solve the variety of issue that exists in the network, reach the purpose that improves mobile communication network device performance, network examination quality, mobile subscriber's satisfaction.This work be the maintenance work of network conventional operation distillation, be the network O﹠M quality constantly promote indispensable routine work.

The mobile communication network optimization of present stage is mainly pinpointed the problems by means such as the test (DT) of driving, speech quality test (CQT), network performance statistical report, customer complaints.

Network optimization staff drives to test (DT) by the test road of planning in advance, and use test software is analyzed the wireless issue of discovery and solution institute test zone to sampled data.In addition, key area, city (as: government bodies, large enterprise, hospital, station, tourist attractions etc.) are carried out speech quality test (CQT), guarantee that these regional voice, data service are normal.At network level, the optimization personnel need report from network operation maintenance centre (OMCR) collection network performance statistics, carry out statistical analysis at not up to standard in the performance assessment criteria item (for example: wireless system call completing rate, cutting off rate, paging success rate, handover success rate etc.) or the bigger project of fluctuation, investigation problem reason is also solved.In addition, radio network optimization need be paid a return visit the complaint problem points of public users, and the problem that the user is reflected is handled.

Present stage, there was following shortcoming in the mobile communication network optimization technology:

1) means of pinpointing the problems are single, can't grasp network condition comprehensively

The current network optimization is only pinpointed the problems by means such as the test (DT) of driving, speech quality test (CQT), network performance statistical report, customer complaints.

Drive to test the wireless network performance situation that (DT) only can reflect the road radiation areas of testing, then can't find the network problem of its existence to the zone of not testing.

Speech quality test (CQT) same only at as: the key area that groups such as government bodies, large enterprise, hospital, station, tourist attractions patrol and examine is tested, more then can't its network coverage situation of detail knowledge such as residential block, high density building, high-rise overlay area, small towns, factories and miness.

More than two kinds of methods of finding network problems be to catch small probability anomalous event in the network by the work of magnanimity, capture rate is low, because the reproduction possibility of the in esse problem of network is little.So can not realize the perfect network optimization.

The network performance statistical indicator is to the homogenizing index of certain time period, certain Local Area Network service ability, then can't present in detail for accident, difficult incident and recessive problem etc.

Customer complaint generally is that network problem reaches capacity user's TL just can to take place, and the recessive network problem that those are not complained may be difficult to be found; Secondly there is very big randomness in the problem of customer complaint, is difficult to the recurrent network problem during on-the-spot test; The 3rd customer complaint and customer group self tolerance degree and structure of personnel have relation, and some customer group may not can report network problem with the form of complaining, and directly personal feeling is married again carrier network, influences the popular public sentiment evaluation of operator.

2) network optimization purpose is single, is purpose with the index lifting generally, can't take into account the actual service performance lifting of user's impression and network

The problem of finding in the analysis of network performance statistical report general only promote and drive at the network performance assessment criteria test (DT), speech quality test (CQT), the customer complaint is analyzed, can't find and predict the network potential problems, lifting for network integrity service performance, the problems such as optimization of network element performance also can't fully be paid attention to, and in fact these actions will be related to operator's income and the user experiences conscientiously.

3) optimize work quality and rely on optimization personnel's technical capability and optimization experience, be difficult to guarantee continual and steady

The network optimization is had relatively high expectations to optimizing personnel's technical capability and optimization experience.Different optimization engineers is different for same case study direction, processing means and optimization effect.Therefore when reality solved network problem, the difference of technical capability and optimization experience may cause its optimization effect completely different.

And network optimization work generally is according to project invitation for bid, carrying out at times, and therefore when occurring optimizing personnel amendment or network configuration variation, network service quality can't ensure.

4) accurate fixer network problem

Traditional statistical analysis mode, be simply network event to be carried out statistical summaries, its positioning accuracy minimum is cell-level index (cell), a general sub-district in the urban district covering diameter about 1.5 kilometers, the fringe area is bigger, there are the problem that positioning accuracy is not enough and incident can't be filtered in rural area even reach tens kilometers, this kind statistical.

5) personnel cost, equipment cost are too high

The structure of personnel of network optimization team is example (network size of base station number about 1000) with a medium-sized city at present, its test/test analysis personnel amount is generally 20, BSS engineer and is generally 8, optimizes 1 of project manager, other supporting personnel, as test 4 of drivers, 4 of tower workers, add up to the minimum 40-50 people that needs.Optimizing equipment relates generally to network test instrument, test analysis AM/BAM software, interference analysis instrument, frequency sweep equipment, testing vehicle, BSS performance evaluation software, signaling instrument, signalling analysis software or the like.Equipment cost is as follows: the quotation Zai $400 of the legal testing software of the Tems of Ericsson, more than 000; The quotation of Tektronix interference analysis instrument YBT250 exists, more than 000; Signaling instrument K1205 Zai $600 is more than 000.

Therefore personnel cost, equipment cost are too high becomes the drawback that general networks is optimized.If reduce personnel and equipment cost merely, must cause the decline of the optimization means and the ability of optimization, and then influence network optimization effect.

Summary of the invention

In order to solve the above-mentioned problems in the prior art, the invention provides a kind of network optimization system, the data of network optimization system of the present invention are comprehensive, can be truly reaction network problem comprehensively, and then network performance optimized comprehensively, change the working method that legacy network is optimized.

In order to solve the problems of the technologies described above, the present invention has adopted following technical scheme:

Network optimization system comprises signaling acquisition device, database server, data processing module and optimization application module, wherein

Signaling acquisition device is gathered signaling data as network optimization analysis foundation data from mobile communication network, and the signaling data of storage of collected;

Database server, auxiliary data is analyzed in the network optimization that comprises wireless network basic data, user data and DT data of storage of collected; And the data that are used for network optimization inquiry, add up and transfer that generate after the storage process data processing module processing signaling data;

Data processing module, gather the network optimization analysis foundation data and the network optimization and analyze auxiliary data, finish data processing and analysis, and will handle and multi-layered database framework that analysis result is stored to database server to realize the required data association of the network optimization, problem points alignment by union and problem data inquiry;

Optimizing application module, is the man-machine interface of network optimization system, is used for calling processing and analysis result in the multi-layer data framework of database server.

Further, described data acquisition unit comprises signal collecting unit (PCD) and signal collecting server; Mobile communication network comprises 2G network and 3G network, the signal collecting unit is from the A interface (interface between base station controller (BSC)-mobile switching centre (MSC)) of 2G network, Abis interface (interface between base station (BTS)-base station controller (BSC)) and Gb Interface (interface between base station controller (BSC)-serving GPRS support node (SGSN)) carry out the collection of signaling data or from the Iu_b interface (interface between radio network controller (RNC)-base station (NodeB)) of 3G network, Iu_CS interface (interface between radio network controller (RNC)-switched circuit territory) and Iu_PS interface (interface between radio network controller (RNC)-swap data territory) carry out the collection of signaling data, and the signaling data of gathering is converted into the magnanimity disk array (RAID) that is stored to the signal collecting server behind the IP form.

Advance-go on foot, described signal collecting unit is made up of the signal collecting integrated circuit board, when signal collecting integrated circuit board when being a plurality of, by the network switch signaling data of a plurality of signal collecting integrated circuit boards outputs is converged the magnanimity disk array that convergence transfers to the signal collecting server and store.

Further, wherein

Described multi-layered database framework comprises one-level database and second databases;

Described data processing module, generate the location factor according to the DT data, signaling data is handled and according to location factor physics and chemistry label additionally, above-mentioned signaling process result is stored to the one-level database, and the one-level database comprises the MR measured database that is mutually related, problem signaling process database and signaling optimization basis staqtistical data base; , generate corresponding analysis result and be stored to second databases one-level database difference finishing analysis according to network optimization content.

Further, described data processing module generates the location factor according to the DT data and additionally the step of physics and chemistry label is as follows to data signaling:

S1, the generation location factor:

A. at first, with the geographic area that system will cover, the size according to formulating is divided into grid, gives each grid numbering;

B. filter out the grid that the DT data cover;

C. based on the DT data, calculate above each grid each level range, the probability of appearance at each sub-district;

D. with one group of probability statistics value, as the location factor of a grid, preservation is got up;

E. next handle the grid that is not covered again by the DT data;

F. travel through the grid that all do not locate the factor;

G. calculate the quantity of each grid of not locating factor grid that the location factor is arranged on every side;

H. from around the maximum grid of not locating the factor of quantity of grid that the location factor is arranged begin to handle;

I. from around have the grid of the location factor to extract the location factor, go out one group of location factor according to mean value computation earlier, traversal is located the sub-district in the factor again, find the grid at these places, sub-district successively, calculate the azimuthal relation of current grid and place, sub-district grid and sub-district, extrapolating current grid should increase still at the probability of occurrence of sub-district and reduce, and the amount that changes;

J. result calculated is preserved as the location factor of current grid;

K. repeating step f to j is till all grids all have the location factor;

S2, filter out the conversation each time in the signaling data, the measurement report (MR) in the conversation is extracted, comprehensive statistics goes out each sub-district, in certain level range, and the probability of appearance;

S3, travel through the location factor of all grids, calculate the involved sub-district of itself and current talking and the matching degree of level range,, calculate probable value according to matching degree;

S4, find out 5 the highest grids of probability, calculate the center of gravity of these 5 grids, be the position of this conversation.

Another object of the present invention is for providing a kind of network optimized approach, network optimized approach of the present invention can be truly reaction network problem comprehensively, and then network performance optimized comprehensively.

The technical scheme that realizes this purpose is as follows:

Network optimized approach comprises the steps:

A. data acquisition is gathered signaling data and storage from mobile communication network, as network optimization analysis foundation data; Gather wireless network basic data, user data, geographical basis information data and DT data, analyze auxiliary data as the network optimization;

B. auxiliary data is analyzed in the aggregation step A network optimization analysis foundation data and the network optimization of gathering, and finishes data processing and analysis, and will handle and analysis result is stored to the multi-layered database framework; Realize the required data association of the network optimization, problem points alignment by union and problem data inquiry by the multi-layered database framework;

C. from the multi-layered database framework of step B, call corresponding data according to the content of the network optimization.

Further, mobile communication network comprises 2G network and 3G network, carry out the collection of signaling data or carry out the collection of signaling data from A interface, Abis interface and the Gb Interface of 2G network by the signal collecting unit, and store after the signaling data of gathering is converted into the IP form from Iu_b interface, Iu_CS interface and the Iu_PS interface of 3G network.

Further, described signal collecting unit is made up of the signal collecting integrated circuit board, when signal collecting integrated circuit board when being a plurality of, the magnanimity disk array that a plurality of signal collecting integrated circuit board outgoing signaling convergences convergences transfer to the signal collecting server is stored by the network switch.

Further, data processing among the step B comprises according to the DT data generation location factor with to signaling data to be handled, and the physics and chemistry label additionally of the signaling data after handling according to the location factor pair, thereby be configured for the one-level database on the basis of network optimization analysis, the one-level database comprises the MR measured database that is mutually related, problem signaling process database and signaling optimization basis staqtistical data base; And according to network optimization content to one-level database finishing analysis respectively, generate corresponding analysis result, constitute second databases.

Further, generate the location factor according to the DT data and additionally the step of physics and chemistry label is as follows to data signaling:

S1, the generation location factor:

A. at first, with the geographic area that system will cover, the size according to formulating is divided into grid, gives each grid numbering;

B. filter out the grid that the DT data cover;

C. based on the DT data, calculate above each grid each level range, the probability of appearance at each sub-district;

D. with one group of probability statistics value, as the location factor of a grid, preservation is got up;

E. next handle the grid that is not covered again by the DT data;

F. travel through the grid that all do not locate the factor;

G. calculate the quantity of each grid of not locating factor grid that the location factor is arranged on every side;

H. from around the maximum grid of not locating the factor of quantity of grid that the location factor is arranged begin to handle;

I. from around have the grid of the location factor to extract the location factor, go out one group of location factor according to mean value computation earlier, traversal is located the sub-district in the factor again, find the grid at these places, sub-district successively, calculate the azimuthal relation of current grid and place, sub-district grid and sub-district, extrapolating current grid should increase still at the probability of occurrence of sub-district and reduce, and the amount that changes;

J. result calculated is preserved as the location factor of current grid;

K. repeating step f to j is till all grids all have the location factor;

S2, filter out the conversation each time in the signaling data, the measurement report (MR) in the conversation is extracted, comprehensive statistics goes out each sub-district, in certain level range, and the probability of appearance;

S3, travel through the location factor of all grids, calculate the involved sub-district of itself and current talking and the matching degree of level range,, calculate probable value according to matching degree;

S4, find out 5 the highest grids of probability, calculate the center of gravity of these 5 grids, be the position of this conversation.

Compared with prior art, beneficial effect of the present invention is:

1) the present invention is by gathering signaling data at 2G network (A interface, Abis interface and Gb Interface) or 3G network (Iu_b interface, Iu_CS interface and Iu_PS interface), as network optimization analysis foundation data, overcome the single drawback of original means of pinpointing the problems with signaling data.

Signaling is the basis of the normal operation of network, any once foundation of conversation all is the mutual result of network signal, corresponding with it any primary network anomalous event all can be on signaling marks to some extent all, therefore all network performance indexes all are to be statistic unit with the signaling, we can say and have only the ruuning situation of the most comprehensive true reflection network of signaling.Therefore by A mouth and the Abis mouth signaling that continues, the collection of measurement report (MR) and to the association analysis of signaling data has overcome the single drawback of the means of pinpointing the problems;

2) the present invention pays close attention to the lifting of index, user's perception and network performance simultaneously, and does detailed partition at application side.

On the action basis that legacy network is optimized, consider analysis optimization to user's impression and network performance, utilize that signaling data is comprehensive, the accurate advantage of location algorithm, make original index optimization work, user's perception promote the level network performance and promote and obtained in-depth and unified.

The optimization applying portion of network optimization system of the present invention and method is detailed comprehensively, comprises: network index optimization, VIP user/VIP area guarantee, wireless parameter intelligent checking, switch analysis automatically, signaling optimization, frequency planning assistant analysis, zone (LAC district/RAC district) optimization, customer complaint assistant analysis, road/modules such as rds data analysis.Above-mentioned module all can be presented by GIS ground physics and chemistry information platform.

3) the invention provides powerful auxiliary optimizational function, offer the optimization personnel and optimize the experience reference, guarantee that optimum level is continual and steady.

Existing network is adjusted experience and optimized case is enough to provide powerful technical support.The present invention can provide the solution of reference to the network problem of finding.All can produce prompting automatically for the problem points of finding in the signalling analysis, and present, be alarmed in the mode of tabulation simultaneously by generalized information system.Reason for problem produces the invention provides auxiliary reference information, to make things convenient in the optimization personnel course of work in addition reference.After the network parameter modification was finished, the present invention can back up the related data of revising, and makes things convenient for the optimization personnel to be optimized compliance test result and basic data library management behind parameter modification, thereby guaranteed that optimization work continues efficiently to carry out.

4) accurate positioning of the present invention.

The present invention has created the new propagation model formula of a cover by the research to the theoretical propagation model of hata, and by real-time importing test data the propagation model factor is corrected.While is finally confirmed the grid (GRID) that each calling takes place to carry out main Serving cell, many sectors, adjacent sub-district alignment by union with the measurement report of once calling out (MR), is embodied as the problem calling and accurately locatees.By on-site verification, this algorithm positioning accuracy is in 40 meters, and locating accuracy reaches more than 98%, is higher than the positioning accuracy and the accuracy rate of other computational methods at present far away.

5) effectively reduce personnel cost, equipment cost.

The present invention at utmost saves personnel cost and equipment cost under the prerequisite that guarantees the optimization quality.Because the present invention carries out preliminary treatment with the signaling data that collects automatically, after the preliminary treatment with problem data screening and filtering in addition, network optimization personnel can login the function items that the intelligent optimization platform is provided with personalized interface and is paid close attention to, needed data and result will be presented in face of the user with the interface of ground physics and chemistry or formization, have saved the complicated processes that data extract in traditional optimizing process, backstage are analyzed, the result checks.Therefore wireless analysis project's teacher workload will reduce significantly.Simultaneously, because the concrete condition of network event can obtain by signaling data each time, therefore saved the work of substantive test personnel magnanimity test.After using the intelligent optimization platform, the work that the tester does only is to go to upgrade the location propagation model factor by test, and arranges barrier at optimizing the problem points of finding, has really accomplished to shoot the arrow at the target, and has improved personnel ' s efficiency.

Description of drawings

Fig. 1 is the structured flowchart of network optimization system of the present invention;

Fig. 2 is the structure principle chart of signaling acquisition device part in the 2G network of network optimization system of the present invention;

Fig. 3 is the structure principle chart of signaling acquisition device part in 3G network of network optimization system of the present invention;

Fig. 4 is the procedure chart of the data processing of network optimization system of the present invention and method.

Embodiment

Below in conjunction with the drawings and specific embodiments the present invention is described in further detail, but not as a limitation of the invention.

Network optimization system of the present invention and method can be used for 2G network or the 3G network in the mobile communication network, and its principle is identical.Just in different networks, gather the title difference of the interface of signaling data.

Network optimization system of the present invention and method all need be used following data, are illustrated earlier at this:

The signaling data storehouse:

The signaling data storehouse of 2G network comprises:

Abis message makes data, is all signaling messages that BTS-BSC transmits in the network normal voice services/data service communication process;

The A message data are all signaling messages that BSC-MSC transmits in the network normal voice service communication process;

Gb message makes data, is all signaling messages that BSC-SGSN transmits in the network normal data service communication process.

The signaling data storehouse of 3G network comprises:

Iu_b message makes data, is all signaling messages that Node_B-RNC transmits in the network normal voice services/data service communication process;

Iu_CS message makes data, is all signaling messages that transmit in RNC-switched circuit territory in the network normal voice service communication process;

Iu_PS message makes data, is all signaling messages that transmit in RNC-swap data territory in the network normal voice service communication process.

The basic data that all optimizational functions of network optimization system of the present invention and method realize.Signaling acquisition device is realized the collection work to above data.

Wireless network basic database (BSS/RNC basic database):

It in the 2G network BSS basic database, it in the 3G network RNC basic database, be all base station Back ground Informations of communication network (as latitude and longitude of base station, data such as antenna electrical characteristics, antenna azimuths), wireless network parameter, the set of data such as network timer and geography information.This part data is the auxiliary reference data of network optimization analysis or the data that need adjustment.This partial data upgrades from operation maintenance center (OMCR) in real time, and offers the supplementary of data processing module as data processing.

Customer data base:

Mainly be user's data set such as VIP user, emphasis road and railway, be used for guarantee this user.To VIP user's MSISDN the geography information of IMSI and emphasis road carry out self-definedly, this partial data will directly offer the supplementary of data processing module as data processing.

The DT database:

The set of DT data, it is the basic data that the propagation model factor is corrected in the location algorithm, to be stored in the data importing database module of the present invention after test is finished by the testing engineering personnel, this partial data will directly offer the supplementary of data processing module as data processing.

Fig. 1 is the structured flowchart of network optimization system of the present invention; Fig. 2 is the structure principle chart of signaling acquisition device part in the 2G network of network optimization system of the present invention; Fig. 4 is the procedure chart of the data processing of network optimization system of the present invention.As Fig. 1, Fig. 2 and shown in Figure 4, network optimization system comprises signaling acquisition device, database server, data processing module and optimization application module, wherein

Signaling acquisition device is used for gathering signaling data from mobile communication network.In the 2G network, Abis mouth (interface between base station (BTS)-base station controller (BSC)), A mouth (interface between base station controller (BSC)-mobile switching centre (MSC)) and Gb Interface (interface between base station controller (BSC)-serving GPRS support node (SGSN)) that signaling acquisition device is used for from the 2G network are gathered signaling data respectively, and with the signaling data storage of gathering.Fig. 3 is the structure principle chart of signaling acquisition device part in 3G network of network optimization system of the present invention.As shown in Figure 3, in 3G network, signaling acquisition device is that Iu_b interface (interface between radio network controller (RNC)-base station (NodeB)), Iu_CS (interface between radio network controller (RNC)-switched circuit territory) and the Iu_PS interface (interface between radio network controller (RNC)-swap data territory) from 3G network is gathered signaling respectively, and with the signaling data storage of gathering.

Signaling acquisition device mainly comprises signal collecting unit (PCD) and signal collecting server, also may need the network switch.The signal collecting unit is made of the signal collecting integrated circuit board, the signal collecting integrated circuit board is finished the collection of signaling data by corresponding interface in the network, and the signaling message of different bottom host-host protocols (E1/T1, STM-1, STM-4) restrained and converge, and unify to be converted to the magnanimity disk array (RAID) that the signal collecting server was exported and be stored in to the IP message form, be that the IP message form is then directly gathered as transformat.Because the amount of the signaling data of gathering is very huge, so need the signal collecting server to have the magnanimity disk array.And at data processing module the signaling data in the signal collecting server is handled accordingly, and, can be deleted the signaling data of handling in the signal server, realize the circulation storage after result's storage.Can also insert 4 pairs of STM-1 optical interfaces or 2 pairs of STM-1 optical interfaces in the time of 64 pairs of E1/T1 Interface of signal collecting integrated circuit board support, its capacity is 1024 64k signalings.Under the complicated network structure, signaling quantity situation how, the signal collecting unit is made up of polylith signal collecting integrated circuit board usually, so that satisfy the collection needs of data.Under the situation of polylith signal collecting integrated circuit board, data acquisition server carries out data collection for convenience just needs the network switch.The network switch is under the situation of polylith signal collecting integrated circuit board, for convenience the magnanimity disk array (RAID) of signal collecting server carries out data collection, the message content on many gigabit netting twines of each integrated circuit board output is converged converge to a gigabit netting twine and export.Certainly the magnanimity disk array (RAID) at the signal collecting database disposes under the situation of polylith network interface card, also can not use the network switch.

The electric parameter of the signal collecting integrated circuit board that this enforcement is adopted such as following table 1.

Table 1

Database server, storage is analyzed BSS/RNC basic data, user data and the DT data of auxiliary data as the network optimization, provides data source with the signaling data of signaling acquisition device collection for the network optimization.The related data that auxiliary data can increase needs according to the needs of the network optimization is analyzed in the network optimization, as VIP user data, customer complaint data and key area (as railway) data etc.

Data processing module will gather collected data, data be handled accordingly and will be analyzed, and store with the form of multi-layered database framework; By rational multi-layered database framework is the platform that application layer is built data query, called, and realizes at required basic operations of the network optimization such as multi-interface data association, problem points alignment by union, problem data inquiries.Data source mainly comprises data message, geographical basis data message (GIS data message), VIP user data and the DT data of signaling data, BSS/RNC basic database.Wherein signaling data is a network optimization system The main analysis content of the present invention; The BSS/RNC basic database is used to optimize the data source of aligning reference and verifying as wireless parameter; Ground physics and chemistry information be used for the GIS network positions and ground physics and chemistry show, also be the reference of User Defined geography information simultaneously; The data basis that the VIP user data ensures for VIP user; The basis that the DT data are calculated for the location factor.The ground physics and chemistry information of the main storage network of GIS information database.Data processing module will carry out preanalysis to signaling data, because signaling data belongs to mass data, in preanalysis process to signaling data, mainly useful data and information are concluded and extracted, and a large amount of hashes are given up after finishing statistics, so that realize the storage that circulates of signaling data to continuous collection; The calculated factor of the up-to-date location of network is preserved in factor storehouse, location.

Database server is used to store data source and the data that the back generates are handled and analyzed to data processing module.Wherein data processing module is handled and is analyzed in the one-level database and second databases of storage in database server of back generation.

The storage that data processing module carries out initial data producing after the preanalysis is in the one-level database.The one-level database mainly comprises MR measured database, problem signaling process database and signaling optimization basis staqtistical data base.

One-level database number is the basis of realizing that the network optimization is analyzed, and also is the network data that really needs storage, also is simultaneously the data source that second databases realizes inquiry.Mainly comprise MR measured database, problem signaling process database and signaling optimization basis staqtistical data base in the one-level database, wherein the MR measurement data is to be unit with the sub-district, measurement result after measurement report averaged directly reflects level, quality, the MR locating information of network; The problem signaling process is network optimization system to be set in-problem signaling process preserve; Signaling optimization basis staqtistical data base is for optimizing the basic data that application module is used.More than three databases interrelated, the integral body that formation problem series connection can be inquired about;

Second databases is the data content that produces for the optimization application module, and generally the form with pointer realizes the inquiry of one-level database and reviewing of historical data.The application of second databases will promote the operating efficiency of optimizing application module.

For convenience in the realization of the network optimization, we have carried out arrangement with the one-level database and have formed second databases, that is: to the statistics and the Query Information of one-level database, utilize second databases easily to realize to the inquiry of one-level database and reviewing of historical data, data content in the second databases is not fixed, and can data be increased and decreased according to user's demand.

Data processing module possesses powerful buffer memory and data-handling capacity; The factor data storehouse, location, the network ground physics and chemistry system (GIS) that bring in constant renewal in according to the DT data carry out the subregion modeling to network, in the signaling process process, calculate the locating information of signaling process, add ground physics and chemistry label in the data that the one-level database is preserved, be locating information, with the physics and chemistry location, ground of realizing using.

The location factor production method and to signaling data additionally the process of physics and chemistry label can describe in detail as follows:

S1, the data of gathering based on DT are by gathering a large amount of DT data, as the modeling basic data of entire city;

S2, the generation location factor (ls):

A. at first, with the geographic area that system will cover, the size according to formulating is divided into grid on generalized information system, gives each grid numbering;

B. filter out the grid that the DT data cover;

C. based on the DT data, calculate above each grid each level range, the probability of appearance at each sub-district;

D. with one group of probability statistics value, as the location factor of a grid, preservation is got up;

E. next handle the grid that is not covered again by the DT data;

F. travel through the grid that all do not locate the factor;

G. calculate the quantity of each grid of not locating factor grid that the location factor is arranged on every side;

H. from around the maximum grid of not locating the factor of quantity of grid that the location factor is arranged begin to handle;

I. from around have the grid of the location factor to extract the location factor, go out one group of location factor according to mean value computation earlier, traversal is located the sub-district in the factor again, find the grid at these places, sub-district successively, calculate the azimuthal relation of current grid and place, sub-district grid and sub-district, extrapolating current grid should increase still at the probability of occurrence of sub-district and reduce, and the amount that changes;

J. result calculated is preserved as the location factor of current grid;

K. repeating step f to j is till all grids all have the location factor;

S3, filter out the conversation each time in the signaling data, the measurement report (MR) in the conversation is extracted, comprehensive statistics goes out each sub-district, in certain level range, and the probability of appearance;

S4, travel through the location factor of all grids, calculate the involved sub-district of itself and current talking and the matching degree of level range,, calculate probable value according to matching degree;

S5, find out 5 the highest grids of probability, calculate the center of gravity of these 5 grids, be the position of this conversation.

Wherein step S1 to S2 is the process that generates the location factor according to the DT data, and step S3-S5 is the process to the network positions of signaling data.Along with the collection of DT data among the step S1 is upgraded, the location factor among the step S2 is upgraded thereupon.

Optimizing application module is man machine operation interface, realizes the network optimization with ground physics and chemistry figure, tabulation, method such as self-defined.Optimize the staff by optimizing application module except that finishing the optimization routine content, also taken into account user's perception and performance optimization two aspects.The power by location algorithm and software signalling analysis particularly, can carry out intelligently filters to problem points, and on the GIS graphical interfaces, carry out positioning problems, broken away from the big and unfavorable drawback of effect of pinpointing the problems, deal with problems of routine optimization workload, therefore in the application layer design process, lay special stress on is optimized main features such as content is comprehensive, signalling analysis is intelligent, computational speed is fast, locate mode is accurate, the user is easy to use.Below will concrete function piece that optimize application module be described.Having the database corresponding with following functional module in the second databases calls for functional module.

◆ the network index optimal module

Mainly solve traditional optimization problem, congested as TCH, insert problems such as success rate, cutting off rate be low at random.The network intelligence Optimization Platform (MUEIS) provide automatic retrieval to the index exception item, to the accurate location on the problem points realization map, and the detailed description of problem incident signaling process is provided and optimizes reference.After optimizing the personnel prioritization scheme, can realize reviewing of contrast of index front and back and historical data.

Except that physics and chemistry location, problem ground, sub-district covering analyzing function also is a bright spot in this part.Urban district in the wireless environment complexity, the actual coverage in sub-district is great for basis optimization work meaning, and network intelligence Optimization Platform (MUEIS) can rely on the shared peripheral cell traffic of network grid (GRID) ratio that the zone of the actual covering in this sub-district is confirmed!

◆ VIP user area assurance module

Network intelligence Optimization Platform (MUEIS) provides VIP user, VIP area definition interfaces, the optimization personnel can import Optimization Platform with relevant information, MUEIS will monitor in real time the calling each time in definition user, area in the process of signalling analysis and classification gathers, reservation problem evidence makes things convenient for the optimization personnel to carry out the reprocessing of problem points.These are that traditional optimization can't be accomplished.

◆ wireless parameter is verified module

The accuracy of wireless network basic database, reasonability are most important for the normal operation of network.Network intelligence Optimization Platform (MUEIS) is analyzed the wireless network basic database that imports switching in the wireless parameter, power control, the isoparametric reasonability of network timer is checked by ground, GIS base station physics and chemistry information, the analysis of cell call signaling process, MR, and generate check result automatically and optimize suggestion, guarantee the network parameter complete and accurate.

◆ the handover optimization analysis module

Switching problem is one of important process in optimizing, but because this workload is huge, and the 2G/3G network is also deposited and caused switching problem complicated more at present in addition, therefore will switch to analyze separately as a functional block to show that become must.This module is mainly sorted out handover event, as: Intra-cell switching, Inter-cell switching, double frequency switching, the switching of 2G/3G network, soft handover or the like, and, improve optimization efficiency to carrying out intellectual analysis and provide optimizing suggestion such as problems such as handoff failure, unidirectional switching, table tennis switchings.

◆ the signaling optimization module

Signaling optimization is mainly handled case study, the special topic processing that some conventional meanses can't be finished, as strong functions items such as connecting time analysis, SMS analysis, malicious call SMS interception, the unusual release cause analysis of A mouth/Iu, the analysis of RSL release cause, the real-time analysis of traffic focus, the budget of PCH/AGCH capacity are provided.The deep-seated problem of network can be found and solve to the application of signaling optimization module, and then excavate the network potentiality, promotes service quality.

◆ frequency planning assistant analysis module

The GSM frequency resource is very limited, therefore distributes very difficulty of frequency in the actual area frequency planning.On the one hand, it is huge that the Frequency Distribution workload is carried out in craft, inefficiency; On the other hand, the general purpose of design of existing frequency planning software is pre-planning, i.e. planning before the networking can't be taken into account the variation that existing network is planned regional wireless environment, therefore plans that effect is generally also undesirable; Network intelligence Optimization Platform (MUEIS) medium frequency planning assistant analysis functional block is carried out analysis-by-synthesis according to MR measurement report content and GIS geography information, cell configuration frequency hopping mode, on the basis that its radio characteristics, geographic properties are taken all factors into consideration, make frequency planning scheme, therefore plan respond well.In addition, the quenching frequency planning module can also be calculated the reasonability that is provided with of frequency hopping group, MAIO, HSN according to existing network frequency operating position.

◆ LAC, RAC optimal module

This work also is the emphasis that optimization routine is paid close attention to, but optimization means was more single in the past.The assistant analysis function that network intelligence Optimization Platform (MUEIS) provides LAC, the performance evaluation of RAC border and LAC, RAC to divide is to help the optimization personnel to carry out base station LAC, the ownership adjustment of RAC district and to optimize.

◆ optimal module is assisted in customer complaint

The a large amount of optimization personnel of the needs of customer complaint in the past carry test instrumentation and deal with problems at the complaint scene.Because it is sporadic and not reproducible that network problem exists, therefore the actual effect and the effect of dealing with problems is not very desirable.Utilize network intelligence Optimization Platform (MUEIS) platform to need not to complain the on-the-spot situations such as 24 hours level of problem points, quality, traffic of just can understanding, even can transfer and complain the user signaling flow process to analyze, the efficient that therefore solves customer complaint obviously promotes.Owing to can back up, just can realize that each problem points reviews to the historical data amendment record.

◆ emphasis road railway support module

The network test of emphasis road railway, the common method of move in being, UNICOM examining each branch company, therefore prefectures and cities operator to emphasis road railway support work attach great importance to, in routine optimization, drop into a large amount of optimization personnel, testing equipment and testing vehicle and go to finish this work.Can be by network intelligence Optimization Platform (MUEIS) system by the self-defined road railway information of optimization personnel, and level, quality, the call case of these road railways monitored in real time, anomalous event, the abnormal call that relates to the road railway filters to gather and makes things convenient for the optimization personnel to handle in real time, simultaneously the network parameter that relates to the road railway finished synchronous verification.To sum up saved the cost input of road railway support work greatly.

Above-mentioned application module can increase according to the actual needs of the network optimization or reduce just as exemplary illustrated.Not as restriction to network optimization system of the present invention.

Network optimized approach comprises the steps:

A. data acquisition needs to gather signaling data respectively as network optimization analysis foundation data from Abis mouth (interface between base station (BTS)-base station controller (BSC)), A mouth (interface between base station controller (BSC)-mobile switching centre (MSC)), Gb Interface (interface between base station controller (BSC)-serving GPRS support node (SGSN)) from 2G network (GSM) network; At 3G network (WCDMA, TD-SCDMA) need in the network to gather signaling data respectively, as network optimization analysis foundation data from Iu_b interface (interface between radio network controller (RNC)-base station (NodeB)), Iu_CS (interface between radio network controller (RNC)-switched circuit territory), Iu_PS interface (interface between radio network controller (RNC)-swap data territory); Gather BSS/RNC basic data, user data, geographical basis information data (GIS data) and DT data, analyze auxiliary data as the network optimization;

B. auxiliary data is analyzed in the network optimization analysis foundation data of aggregation step A collection and the network optimization, the multi-layered database framework of finish the storage of data processing and later stage, using; Build data query, calling platform by the multi-layered database framework, realize the required data association of the network optimization, problem points alignment by union and problem data inquiry basic operation;

C. from the multi-layered database framework of step B, call corresponding data according to the content of the network optimization, realize every work of the network optimization.

Wherein in the steps A: from the Abis mouth (interface between base station (BTS)-base station controller (BSC)) of 2G network, the Iu_b interface (interface between radio network controller (RNC)-base station (NodeB)) of A mouth (interface between base station controller (BSC)-mobile switching centre (MSC)) and Gb Interface (interface between base station controller (BSC)-serving GPRS support node (SGSN)) or 3G network, Iu_CS (interface between radio network controller (RNC)-switched circuit territory) and Iu_PS interface (interface between radio network controller (RNC)-swap data territory) carry out the collection of signaling data, the collection of signaling data is to be undertaken by signaling acquisition device, signaling acquisition device comprises signal collecting unit (PCD) and signal collecting server, the signal collecting unit is with original transformat (E1/T1, ATM-1 ATM-4) is converted into IP form and be stored to the signal collecting server.The signal collecting unit is made up of the signal collecting integrated circuit board, when signal collecting integrated circuit board when being a plurality of, generally need the network switch that the signaling data on many netting twines of a plurality of signal collecting integrated circuit boards outputs is converged and converge to the magnanimity disk array that a netting twine transfers to the signal collecting server and store.But but when the signal collecting server is equipped with the polylith network interface card also heterogeneous networks switch.

Wherein, data processing among the step B comprises according to the DT data generation location factor, with signaling data is handled and according to location factor physics and chemistry label additionally, generation is used for the one-level database of the network optimization, and the one-level database comprises the MR measured database that is mutually related, problem signaling process database and signaling optimization basis staqtistical data base; And according to network optimization content to one-level database finishing analysis respectively, generate corresponding analysis result, constitute second databases.

Further, the location factor production method and to signaling data additionally the process of physics and chemistry label can describe in detail as follows:

S1, the data of gathering based on DT are by gathering a large amount of DT data, as the modeling basic data of entire city;

S2, the generation location factor (ls):

A. at first, with the geographic area that system will cover, the size according to formulating is divided into grid on generalized information system, gives each grid numbering;

B. filter out the grid that the DT data cover;

C. based on the DT data, calculate above each grid each level range, the probability of appearance at each sub-district;

D. with one group of probability statistics value, as the location factor of a grid, preservation is got up;

E. next handle the grid that is not covered again by the DT data;

F. travel through the grid that all do not locate the factor;

G. calculate the quantity of each grid of not locating factor grid that the location factor is arranged on every side;

H. from around the maximum grid of not locating the factor of quantity of grid that the location factor is arranged begin to handle;

I. from around have the grid of the location factor to extract the location factor, go out one group of location factor according to mean value computation earlier, traversal is located the sub-district in the factor again, find the grid at these places, sub-district successively, calculate the azimuthal relation of current grid and place, sub-district grid and sub-district, extrapolating current grid should increase still at the probability of occurrence of sub-district and reduce, and the amount that changes;

J. result calculated is preserved as the location factor of current grid;

K. repeating step f to j is till all grids all have the location factor;

S3, filter out the conversation each time in the signaling data, the measurement report (MR) in the conversation is extracted, comprehensive statistics goes out each sub-district, in certain level range, and the probability of appearance;

S4, travel through the location factor of all grids, calculate the involved sub-district of itself and current talking and the matching degree of level range,, calculate probable value according to matching degree;

S5, find out 5 the highest grids of probability, calculate the center of gravity of these 5 grids, be the position of this conversation.

Wherein step S1 to S2 is the process that generates the location factor according to the DT data, and step S3-S5 is the process to the network positions of signaling data.Along with the collection of DT data among the step S1 is upgraded, the location factor among the step S2 is upgraded thereupon.The technical scheme part correlation content of network optimization system of the present invention and method complements one another, uses for reference, so duplicate contents repeats no more.

Above embodiment is an exemplary embodiment of the present invention only, is not used in restriction the present invention, and protection scope of the present invention is defined by the claims.Those skilled in the art can make various modifications or be equal to replacement the present invention in essence of the present invention and protection range, this modification or be equal to replacement and also should be considered as dropping in protection scope of the present invention.

Claims (10)

1. network optimization system is characterized in that, comprises signaling acquisition device, database server, data processing module and optimization application module, wherein

Signaling acquisition device is gathered signaling data as network optimization analysis foundation data from mobile communication network, and the signaling data of storage of collected;

Database server, auxiliary data is analyzed in the network optimization that comprises wireless network basic data, user data and DT data of storage of collected; And the data that are used for network optimization inquiry, add up and transfer that generate after the storage process data processing module processing signaling data;

Data processing module, gather the network optimization analysis foundation data and the network optimization and analyze auxiliary data, finish data processing and analysis, and will handle and multi-layered database framework that analysis result is stored to database server to realize the required data association of the network optimization, problem points alignment by union and problem data inquiry;

Optimizing application module, is the man-machine interface of network optimization system, is used for calling processing and analysis result in the multi-layer data framework of database server.

2. network optimization system according to claim 1 is characterized in that, described data acquisition unit comprises signal collecting unit and signal collecting server; Mobile communication network comprises 2G network and 3G network, the signal collecting unit carries out the collection of signaling data or carries out the collection of signaling data from Iu_b interface, Iu_CS interface and the Iu_PS interface of 3G network from A interface, Abis interface and the Gb Interface of 2G network, and the signaling data of gathering is converted into the magnanimity disk array that is stored to the signal collecting server behind the IP form.

3. network optimization system according to claim 2, it is characterized in that, described signal collecting unit is made up of the signal collecting integrated circuit board, when signal collecting integrated circuit board when being a plurality of, by the network switch signaling data of a plurality of signal collecting integrated circuit boards outputs is converged the magnanimity disk array that convergence transfers to the signal collecting server and store.

4. network optimization system according to claim 1 is characterized in that, wherein

Described multi-layered database framework comprises one-level database and second databases;

Described data processing module, generate the location factor according to the DT data, signaling data is handled and according to location factor physics and chemistry label additionally, above-mentioned signaling process result is stored to the one-level database, and the one-level database comprises the MR measured database that is mutually related, problem signaling process database and signaling optimization basis staqtistical data base; , generate corresponding analysis result and be stored to second databases one-level database difference finishing analysis according to network optimization content.

5. network optimization system according to claim 4 is characterized in that, described data processing module generates the location factor according to the DT data and additionally the step of physics and chemistry label is as follows to data signaling:

S1, the generation location factor:

A. at first, with the geographic area that system will cover, the size according to formulating is divided into grid, gives each grid numbering;

B. filter out the grid that the DT data cover;

C. based on the DT data, calculate above each grid each level range, the probability of appearance at each sub-district;

D. with one group of probability statistics value, as the location factor of a grid, preservation is got up;

E. next handle the grid that is not covered again by the DT data;

F. travel through the grid that all do not locate the factor;

G. calculate the quantity of each grid of not locating factor grid that the location factor is arranged on every side;

H. from around the maximum grid of not locating the factor of quantity of grid that the location factor is arranged begin to handle;

I. from around have the grid of the location factor to extract the location factor, go out one group of location factor according to mean value computation earlier, traversal is located the sub-district in the factor again, find the grid at these places, sub-district successively, calculate the azimuthal relation of current grid and place, sub-district grid and sub-district, extrapolating current grid should increase still at the probability of occurrence of sub-district and reduce, and the amount that changes;

J. result calculated is preserved as the location factor of current grid;

K. repeating step f to j is till all grids all have the location factor;

S2, filter out the conversation each time in the signaling data, the measurement report in the conversation is extracted, comprehensive statistics goes out each sub-district, in certain level range, and the probability of appearance;

S3, travel through the location factor of all grids, calculate the involved sub-district of itself and current talking and the matching degree of level range,, calculate probable value according to matching degree;

S4, find out 5 the highest grids of probability, calculate the center of gravity of these 5 grids, be the position of this conversation.

6. network optimized approach is characterized in that, comprises the steps:

A. data acquisition is gathered signaling data and storage from mobile communication network, as network optimization analysis foundation data; Gather wireless network basic data, user data, geographical basis information data and DT data, analyze auxiliary data as the network optimization;

B. auxiliary data is analyzed in the aggregation step A network optimization analysis foundation data and the network optimization of gathering, and finishes data processing and analysis, and will handle and analysis result is stored to the multi-layered database framework; Realize the required data association of the network optimization, problem points alignment by union and problem data inquiry by the multi-layered database framework;

C. from the multi-layered database framework of step B, call corresponding data according to the content of the network optimization.

7. network optimized approach according to claim 6, it is characterized in that, mobile communication network comprises 2G network and 3G network, carry out the collection of signaling data or carry out the collection of signaling data from A interface, Abis interface and the Gb Interface of 2G network by the signal collecting unit, and store after the signaling data of gathering is converted into the IP form from Iu_b interface, Iu_CS interface and the Iu_PS interface of 3G network.

8. network optimized approach according to claim 7, it is characterized in that, described signal collecting unit is made up of the signal collecting integrated circuit board, when signal collecting integrated circuit board when being a plurality of, the magnanimity disk array that a plurality of signal collecting integrated circuit board outgoing signaling convergences convergences transfer to the signal collecting server is stored by the network switch.

9. network optimized approach according to claim 6, it is characterized in that, data processing among the step B comprises according to the DT data generation location factor with to signaling data to be handled, and the physics and chemistry label additionally of the signaling data after handling according to the location factor pair, thereby be configured for the one-level database on the basis of network optimization analysis, the one-level database comprises the MR measured database that is mutually related, problem signaling process database and signaling optimization basis staqtistical data base; And according to network optimization content to one-level database finishing analysis respectively, generate corresponding analysis result, constitute second databases.

10. network optimized approach according to claim 9 is characterized in that, generates the location factor and additionally the step of physics and chemistry label is as follows to data signaling according to the DT data:

S1, the generation location factor:

A. at first, with the geographic area that system will cover, the size according to formulating is divided into grid, gives each grid numbering;

B. filter out the grid that the DT data cover;

C. based on the DT data, calculate above each grid each level range, the probability of appearance at each sub-district;

D. with one group of probability statistics value, as the location factor of a grid, preservation is got up;

E. next handle the grid that is not covered again by the DT data;

F. travel through the grid that all do not locate the factor;

G. calculate the quantity of each grid of not locating factor grid that the location factor is arranged on every side;

H. from around the maximum grid of not locating the factor of quantity of grid that the location factor is arranged begin to handle;

I. from around have the grid of the location factor to extract the location factor, go out one group of location factor according to mean value computation earlier, traversal is located the sub-district in the factor again, find the grid at these places, sub-district successively, calculate the azimuthal relation of current grid and place, sub-district grid and sub-district, extrapolating current grid should increase still at the probability of occurrence of sub-district and reduce, and the amount that changes;

J. result calculated is preserved as the location factor of current grid;

K. repeating step f to j is till all grids all have the location factor;

S2, filter out the conversation each time in the signaling data, the measurement report (MR) in the conversation is extracted, comprehensive statistics goes out each sub-district, in certain level range, and the probability of appearance;

S3, travel through the location factor of all grids, calculate the involved sub-district of itself and current talking and the matching degree of level range,, calculate probable value according to matching degree;

S4, find out 5 the highest grids of probability, calculate the center of gravity of these 5 grids, be the position of this conversation.

Images