CN103906078A

CN103906078A - Relation matrix establishment system and method and PCI intelligent optimization system and method

Info

Publication number: CN103906078A
Application number: CN201410177128.5A
Authority: CN
Inventors: 阮民
Original assignee: 阮民
Current assignee: Chengdu Edwards Communication Technology Co., Ltd.
Priority date: 2014-04-29
Filing date: 2014-04-29
Publication date: 2014-07-02
Anticipated expiration: 2034-04-29
Also published as: CN103906078B

Abstract

The invention relates to intelligent analysis and intelligent optimization of an LTE mobile communication system. The problem that the coverage relation, the switching relation and the interference relation among cells cannot be accurately quantified and analyzed in an existing LTE system is solved, the problem that the degree of intelligence and the degree of automation are low in an PCI optimization method of the existing LTE system is also solved, and a relation matrix establishment system and method and a PCI intelligent optimization system and method are provided. According to the technical scheme, the relation matrix establishment system is formed by the way that an MR data processing module is connected with an LTE system body, and a CIM cell relation matrix module is connected with the MR data processing module; the PCI intelligent optimization system is formed by the way that a PCI intelligent optimization module is connected with the CIM cell relation matrix module in the cell relation matrix establishment system. The relation matrix establishment system and method and the PCI intelligent optimization system and method have the advantages of being capable of not only quantifying the inference relation among the cells, but also planning and optimizing PCI setting of the cells, suitable for LTE system analysis and wireless optimization and also suitable for 4G service providers to improve operation efficiency.

Description

Relational matrix constructing system and method and PCI intelligent optimization system and method

Technical field

The present invention relates to the network optimization of LTE mobile communication system, be particularly related to the measurement report (MR in LTE mobile communication system, Measurement Reports) application and planning and the optimization of cell physical layer identification code (PCI, Physical Cell Identifier).

Background technology

In existing LTE (Long Term Evolution) mobile communication system, two kinds of coverings, switching and interference relationships analytical methods below general employing:

1, the covering based on DT drive test data, switching and interference relationships analytical method:

In LTE (Long Term Evolution) system, based on the prediction and calculation of radio transmission model, utilize the emulation of relevant planning software and Planning Tool, operator can obtain adjacent area topological relation (NCR) design of wireless system conventionally.Due to prediction and emulation, usually there is any discrepancy with actual landform and propagation characteristic, operator has to by means of drive test (DT, Drive Test) mode judges actual covering intersection and impulsive, and then again plan and optimize NCR neighboring BS relationship, and optimize on this basis the setting of PCI, such as avoiding neighbor cell to occur that mould 3, mould 6 and mould 30 disturb.

Because LTE data service occurs in indoor mostly, and following trend is the HNB (Femotcell in most of indoor heating system, the resident family that relies on interior of building of indoor access of terminal, Home eNodeB) etc., traditional drive test mode almost cannot contain so numerous indoor access signal, rely on drive test to find out the concrete situation that dense urban covering, switching and mould 3 disturb and become very difficult, and cost is very high, be difficult to realize.

2, the covering based on OSS statistics, switching and interference relationships analytical method:

In LTE system, cover intersection and mould 3 and disturb the call drop and the handoff failure that usually cause terminal, even cause the increase of degradation and the data transmission delay of terminal transmission speed, utilize OSS statistics to do after labor the KPI such as handoff failure (Key Performance Indication) index, usually can find that certain community and the covering between peripheral cell intersect, the facts such as edge covering, thereby for analyzing, evading PCI conflict (PCI Collision) and PCI obscures (PCI Confusion) and reduces mould 3 interference data analysis basis is provided.

Operator is after normally the KPI of Mou community worsens, and just notices to intersect that coverings, edge cover, user is intensive, switch the problems such as setting mould 3 unreasonable and PCI disturbs of being with between this community and peripheral cell.

Owing to lacking the complete original signaling information of bottom or wireless environment data, OSS statistics cannot completely present covering, switching and mould 3 interference relationships between any community and the whole associated cell of periphery, and this is the major defect that covers, switches and disturb based on OSS analysis of statistical data.

Above-mentioned DT drive test analytical method and OSS statistical analysis technique, all the time there are three more significant shortcomings, cause covering, switching and the interference analysis of eUTRAN (Evolved Universal Terrestrial Radio Access Network) systematization to carry out.

One, covering relation, handoff relation and the interference relationships based between DT Drive Test Data Analysis eNodeB, sample rate is too low, sampling is difficult and sampling human cost is high, and cannot gather and critical-path analysis for groups of building, extensive area and the whole network expanding data.

Two, covering relation, handoff relation and the interference relationships based between OSS analysis of statistical data eNodeB, it is usually a kind of analysis afterwards, after the deterioration of KPI extreme or customer complaint, just can pinpoint the problems and attempt to deal with problems, daily KPI monitoring is difficult to notice slight KPI problem, is also difficult to notice hiding user awareness KQI (the Key Quality Indicators) problem in slight KPI problem behind.

Three, covering relation, handoff relation and the interference relationships based between OSS analysis of statistical data eNodeB, normally after KPI extremely worsens, just can pinpoint the problems and attempt to deal with problems, the bottom detailed information that now causes KPI to worsen has been difficult to recall, cannot obtain covering, mould 3 disturb and switch between spatial relation figure, have a strong impact on the effect that KPI analyzes and optimizes.

In LTE mobile communication system, terminal UE can be to system reporting measurement reports (MR, Measurement Report).The crucial metrical information comprising in MR data has: terminal produces timestamp, cell-of-origin CGI and PCI, cell-of-origin RSRP and SRQ, adjacent cell CGI and PCI, adjacent cell RSRP and the RSRQ of this measurement report;

SON in LTE mobile communication system (Self-Organizing Network) function.Operator improves neighboring BS relationship (NCR, Neighbor Cell Relationship) by SON conventionally.Perfect neighboring BS relationship contributes to the real covering of MR data reaction and switches present situation, and in prior art, not system or the method for a matrix of cell relations intuitively that can directly generate.

PCI planning and optimization.The effect of physical-layer cell ID (PCI, Physical Cell Identifier) is in cell search process, facilitates terminal, distinguishes the wireless signal of different districts.In the wireless signal of multiple communities that terminal receives simultaneously, can not there is identical PCI, otherwise can form interference.In LTE/LTE-A system, distributable PCI number has 504, is divided into 168 groups, every group of 3 marks.Number of cells in general LTE system, considerably beyond 504, makes same PCI must in different districts, reuse (Reuse), but must the enough distances in interval.PCI planning and the general objective of optimizing are to be the PCI that each cell allocation is suitable by rational instrument and algorithm, the interference bringing to lower identical PCI as far as possible.

When physical planning and optimization PCI, need follow successively following five principles:

A, avoid PCI conflict (PCI collision-free)-be coated with overlapping neighbor cell can not have identical PCI, otherwise terminal is difficult to demodulation pilot frequency channel;

Can not there be identical PCI B, two adjacent areas avoiding PCI to obscure (PCI confusion-free)-Yi Ge community, otherwise terminal switch time can obscure target adjacent section;

C, avoid mould 3 to disturb (Mod3interference-free)-avoid organizing the PCI that interior ID is identical to be distributed on relative or adjacent community as far as possible.The be divided by value (mould 3 of PCI is worth) of remainder number of PCI value and 3 is exactly ID in group.That is be that the mould 3 of PCI is worth two identical PCI, avoids being distributed on adjacent or relative Liang Ge community as far as possible, otherwise can cause the interference of master sync signal PSS (Primary Synchronization Signal);

D, to avoid mould 6 to disturb the be divided by value of remainder number of (Mod6interference-free)-PCI value and 6 be exactly that the mould 6 of PCI is worth.The mould 6 of PCI is worth two identical PCI, avoids being distributed on adjacent or relative Liang Ge community as far as possible, otherwise can cause the interference of downlink reference signal RS (Reference Signal);

E, to avoid mould 30 to disturb (Mod30interference-free) .PCI value and 30 values of being divided by remainder number be exactly that the mould 30 of PCI is worth.The mould 30 of PCI is worth two identical PCI, avoids being distributed on adjacent or relative Liang Ge community as far as possible, otherwise can cause the interference of uplink reference signals RS (Reference Signal).

In prior art, the optimization of PCI comprises following two kinds of methods:

1. the PCI planing method based on emulation of coverage capability technology

In LTE system Construction process, operator conventionally need to be by LTE planning software, import the basic parameters such as base station height, aerial angle, transmitting power, predicting telephone traffic and propagation model, the theoretic coverage in the each community of emulation on geographical graphic, thereby covering relation, handoff relation and the interference relationships of simulation minizone.

Operator, by means of above-mentioned emulation of coverage capability result, further plans NCR neighboring BS relationship, and is the configuration of all plot planning PCI on this basis, to avoiding relatively or PCI conflict appears in neighbor cell, PCI obscures and mould 3 disturbs.

2. the PCI optimization method based on DT drive test technology

After treating that LTE system is opened, operator need to organize drive test (DT conventionally regularly, Drive Test), arrange engineer's use test terminal and testing software, streets and lanes in city are tested repeatedly, to finding that certain community and the covering between peripheral cell intersect, the mould 3 of edge covering, QoS problem, PCI the problem such as disturbs, and then take Optimized Measures, cover such as adjusting, revise the means such as indivedual communities PCI settings, evade PCI conflict (PCI Collision), PCI obscures (PCI Confusion) and reduces mould 3 and disturb.

Above-mentioned conventional method, emulation of coverage capability planning PCI configuration, and DT drive test analysis optimization PCI configuration, have following significant shortcoming all the time, cause the interference of LTE system evade and interference optimization cannot systematization, intelligent enforcement.

One, plan PCI configuration by LTE planning software, because theoretical propagation model and actual radio characteristics have very large deviation, emulation of coverage capability result can not reflect real scene, and the PCI that therefore plans gained is configured in unavoidably still to there will be in actual LTE system and covers the overlapping mould bringing 3 and disturb.

Its two, plan PCI configuration by LTE planning software, be a kind of paper planning, cannot be by real user traffic data and user awareness data, include PCI planning in.Therefore the PCI of planning is configured in and in actual LTE system, cannot obtains optimum effect, also cannot obtain optimum user awareness and experience.

Its three, optimize PCI configuration by DT drive test mode, because drive test can only carry out on major trunk roads, be difficult to enter indoor, sample rate is too low, can only find local problem, is difficult to building and residential quarter to LTE dense traffic, tests and optimizes.

Its four, optimize PCI configuration by DT drive test mode, because test zone is limited, find that after PCI problem, the PCI temporarily carrying out optimizes and revises, and often attends to one thing and lose sight of another, cannot be for whole LTE system, implement comprehensive PCI and plan as a whole to optimize.

Summary of the invention

Technical problem to be solved by this invention, first is that covering relation, handoff relation and the interference relationships of minizone cannot accurately quantize the shortcoming presenting, and a kind of relational matrix constructing system and method are provided in order to solve in present LTE system; Second is intellectuality and the not high shortcoming of automaticity in order to solve PCI optimization method in present LTE system, and a kind of PCI intelligent optimization system and method are provided.

The present invention solves described first technical problem, the technical scheme adopting is, relational matrix constructing system, comprise LTE system ontology, it is characterized in that, also comprise MR data processing module and CIM cell relations matrix (CIM, Cell Interacting Matrix) module, described MR data processing module is connected with LTE system ontology, and CIM cell relations matrix module is connected with MR data processing module;

Described MR data processing module, obtains MR data for the Operation and Maintenance server from LTE system ontology, and extracts the crucial metrical information of MR data, is transferred to CIM cell relations matrix module;

Described CIM cell relations matrix module, according to the crucial metrical information of the MR data that receive in certain hour section, calculate according to computation rule, obtain any cell-of-origin number of times accumulated value that conflicts with any PCI between adjacent cell, PCI obscures number of times accumulated value, the accumulated value of RSRP absolute difference, between cell-of-origin and adjacent cell, there is the accumulated value of the frequency of neighboring BS relationship, the accumulated value of the frequency that cell-of-origin and its adjacent cell occur as other adjacent areas, community simultaneously, between cell-of-origin and adjacent cell, PCI mould 3 disturbs number of times accumulated value, between cell-of-origin and adjacent cell, mould 6 disturbs number of times accumulated value, between cell-of-origin and adjacent cell, mould 30 disturbs the accumulated value of number of times, and the network configuration information that calculates gained and current LTE system is merged, generation can reflect the CIM relational matrix table of any two minizone relation value in LTE system.

Concrete, the crucial metrical information of described MR data comprises timestamp, cell-of-origin CGI, cell-of-origin PCI, cell-of-origin RSRP and the cell-of-origin RSRQ of MR data, also comprises adjacent cell CGI, adjacent cell PCI, adjacent cell RSRP and adjacent cell RSRQ.

Relational matrix construction method, is characterized in that, comprises the following steps:

Step 1, system obtain the MR data of terminal to report;

Step 2, from the MR data that obtain, extract crucial metrical information;

Step 3, system is according to the crucial metrical information in certain hour section, calculate any cell-of-origin number of times accumulated value that conflicts with any PCI between adjacent cell according to computation rule, PCI obscures number of times accumulated value, the accumulated value of RSRP absolute difference, between cell-of-origin and adjacent cell, there is the accumulated value of the frequency of neighboring BS relationship, the accumulated value of the frequency that cell-of-origin and its adjacent cell occur as other adjacent areas, community simultaneously, between cell-of-origin and adjacent cell, PCI mould 3 disturbs number of times accumulated value, between cell-of-origin and adjacent cell, mould 6 disturbs mould 30 between number of times accumulated value and cell-of-origin and adjacent cell to disturb number of times accumulated value, and the network configuration information that calculates gained and current LTE system is merged, generation can reflect the CIM relational matrix table of any two minizone relation value in LTE system.

Concrete, in step 1, the MR data of described acquisition terminal to report, are the MR data acquisition of the Operation and Maintenance server realization by LTE system ontology.

Further, in step 2, described crucial metrical information comprises MR timestamp, cell-of-origin CGI, cell-of-origin PCI, cell-of-origin RSRP and cell-of-origin RSRQ, also comprises adjacent cell CGI, adjacent cell PCI, adjacent cell RSRP and adjacent cell RSRQ.

Further, step 3 comprises following concrete steps:

Step 301, obtain the crucial metrical information in certain hour section;

Step 302, calculate any cell-of-origin number of times accumulated value that conflicts with any PCI between adjacent cell, be designated as C1value;

Step 303, the PCI calculating between any cell-of-origin and any adjacent cell obscure number of times accumulated value, are designated as C2value;

Step 304, calculate any cell-of-origin and any accumulated value of the RSRP absolute difference between adjacent cell, be designated as R value;

Step 305, calculate any cell-of-origin and between adjacent cell, occur arbitrarily the accumulated value of the frequency of neighboring BS relationship, be designated as F value;

Step 306, calculate any cell-of-origin and its any adjacent cell as the accumulated value of the frequency of other community neighboring BS relationship, be designated as N value;

Step 307, calculate any cell-of-origin and arbitrarily between adjacent cell, occur that mould 3 disturbs the accumulated value of number of times, is designated as M3value;

Step 308, calculate any cell-of-origin and arbitrarily between adjacent cell, occur that mould 6 disturbs the accumulated value of number of times, is designated as M6value;

Step 309, calculate any cell-of-origin and arbitrarily between adjacent cell, occur that mould 30 disturbs the aggregate-value of number of times, is designated as M30value;

C1value, C2value, R value, F value, N value, M3value, M6value and M30value that step 310, basis obtain, merge with the network configuration information of current LTE system, according to cell-of-origin dimension and adjacent cell dimension mode one to one, synthesize a two-dimentional relation matrix table, obtain reflecting the CIM relational matrix table of any two minizone relation value in LTE system.

Concrete, in step 302, conflict with any PCI between the adjacent cell method of number of times accumulated value of any cell-of-origin of described calculating is: when terminal accesses LTE system and reports MR data in certain cell-of-origin, add up one by one the whole MR data in this cell-of-origin, when if find, the PCI of its adjacent cell is identical with the PCI of this cell-of-origin, be judged to be PCI conflict, in this cell-of-origin to matrix position that should adjacent cell by C1value cumulative 1;

In step 303, the method that any cell-of-origin of described calculating and the arbitrarily PCI between adjacent cell obscure number of times accumulated value is: when terminal accesses LTE system and reports MR data in certain cell-of-origin, add up one by one the whole MR data in this cell-of-origin, if find, there are two adjacent cells certain cell-of-origin, and the PCI of these two adjacent cells is when identical, be judged to be these two adjacent cell PCI and obscure, at the mutual corresponding matrix position of these two adjacent cells, C2value is added up to 1.

Further, in step 304, any cell-of-origin of described calculating and the arbitrarily method of the RSRP difference accumulated value between adjacent cell are: when terminal accesses LTE system and reports MR data in certain cell-of-origin, add up one by one the whole MR data in this cell-of-origin, calculate the absolute difference of the RSRP of this cell-of-origin and the RSRP of its adjacent cell, then in this cell-of-origin to matrix position that should adjacent cell by cumulative R value this absolute difference.

Concrete, in step 305, between any cell-of-origin of described calculating and any adjacent cell, the method for the accumulated value of occurrence frequency is: when terminal accesses LTE system and reports MR data in certain cell-of-origin, add up one by one whole MR data, if find Wei Qi adjacent area, another community, in this cell-of-origin to matrix position that should adjacent cell by F value cumulative 1.

Further, in step 306, any cell-of-origin of described calculating and its any adjacent cell as the method for the accumulated value of the frequency of other community neighboring BS relationship are: when accessing LTE system in terminal Mou community and reporting MR data, add up one by one whole MR data, if find, this cell-of-origin and its adjacent cell, simultaneously as the adjacent area of aforementioned certain community, add up 1 at the matrix position of corresponding its adjacent cell in this cell-of-origin by N value.

Concrete, in step 307, the method that occurs the number of times that mould 3 disturbs between any cell-of-origin of described calculating and arbitrarily adjacent cell is: when terminal accesses LTE system and reports MR data in certain cell-of-origin, add up one by one the whole MR data in this cell-of-origin, if PCI mould 3 values of finding its adjacent cell and the PCI mould 3 of this cell-of-origin are worth when identical, be judged to be mould 3 and disturb, in this cell-of-origin to matrix position that should adjacent cell by M3value cumulative 1;

In step 308, the method that occurs the number of times that mould 6 disturbs between any cell-of-origin of described calculating and arbitrarily adjacent cell is: when terminal accesses LTE system and reports MR data in certain cell-of-origin, add up one by one the whole MR data in this cell-of-origin, if PCI mould 6 values of finding its adjacent cell and the PCI mould 6 of this cell-of-origin are worth when identical, be judged to be mould 6 and disturb, in this cell-of-origin to matrix position that should adjacent cell by M6value cumulative 1;

In step 309, the method that occurs the number of times that mould 30 disturbs between any cell-of-origin of described calculating and arbitrarily adjacent cell is: when terminal accesses LTE system and reports MR data in certain cell-of-origin, add up one by one the whole MR data in this cell-of-origin, if PCI mould 30 values of finding its adjacent cell and the PCI mould 30 of this cell-of-origin are worth when identical, be judged to be mould 30 and disturb, in this cell-of-origin to matrix position that should adjacent cell by M30value cumulative 1.

The present invention solves described second technical problem, and the technical scheme of employing is, PCI intelligent optimization system, it is characterized in that, comprise above-mentioned cell relations matrix constructing system, also comprise PCI intelligent optimization module, described PCI intelligent optimization module is connected with CIM cell relations matrix module;

Described CIM cell relations matrix module, is transferred to PCI intelligent optimization module by the CIM relational matrix table that can reflect any two minizone relation value in LTE system of its generation;

Described PCI intelligent optimization module, for according to the CIM relational matrix table receiving, by computation rule, obtains the PCI allocation optimum of each community.

Concrete, according to the CIM relational matrix table receiving, calculate the C1value total degree in current C IM relational matrix table, C2value total degree, M3value total degree, M6value total degree and M30value total degree, then, from the PCI candidate list of each community, be the optimization settings that PCI is selected in each community repeatedly, after every suboptimization, all need to upgrade CIM relational matrix table according to CIM computation rule, finally make C1value total degree, C2value total degree, M3value total degree, these five numerical value minimums of M6value total degree and M30value total degree, the PCI candidate list of described each community, its initial table is that PCI span is the list of digital 0-503.

PCI intelligent optimization method, is characterized in that, comprises the following steps:

Step 10, system acquisition are to CIM relational matrix table;

C1value total degree, C2value total degree, M3value total degree, M6value total degree and M30value total degree in step 20, system-computed current C IM relational matrix table, be designated as respectively Σ C1, Σ C2, Σ M3, Σ M6 and Σ M30;

Step 30, execution PCI Optimizing Flow, optimize the PCI settings of community, upgrade CIM relational matrix table according to update rule, recalculate current Σ C1 value, Σ C2 value, Σ M3 value, Σ M6 value and Σ M30 value, judge whether current Σ C1, Σ C2, Σ M3, Σ M6 and Σ M30 are optimized to respectively optimum, if optimize completely, exit PCI Optimizing Flow, if not repeating step 30.

Concrete, in step 30, described PCI Optimizing Flow refers to: optimize sub-process with Σ C1 optimization sub-process, Σ C2 optimization sub-process, Σ M3 optimization sub-process, Σ M6 optimization sub-process and Σ M30 the PCI settings of each community are optimized.

Further, in step 30, described update rule refers to: in the time that the PCI value of cell-of-origin and adjacent cell is identical, this cell-of-origin equals F value value to the C1value value of matrix position that should adjacent cell, otherwise this cell-of-origin equals 0 to the C1value value of matrix position that should adjacent cell;

In the time that the PCI value of cell-of-origin and adjacent cell is identical, this cell-of-origin equals N value value to the C2value value of matrix position that should adjacent cell, otherwise this cell-of-origin equals 0 to the C2value value of matrix position that should adjacent cell;

In the time that the PCI value of cell-of-origin and adjacent cell is identical divided by the remainder of 3 gained respectively, this cell-of-origin equals F value value to the M3value value of matrix position that should adjacent cell, otherwise this cell-of-origin equals 0 to the M3value value of matrix position that should adjacent cell;

In the time that the PCI value of cell-of-origin and adjacent cell is identical divided by the remainder of 6 gained respectively, this cell-of-origin equals F value value to the M6value value of matrix position that should adjacent cell, otherwise this cell-of-origin equals 0 to the M6value value of matrix position that should adjacent cell;

In the time that the PCI value of cell-of-origin and adjacent cell is identical divided by the remainder of 30 gained respectively, this cell-of-origin equals F value value to the M30value value of matrix position that should adjacent cell, otherwise this cell-of-origin equals 0 to the M30value value of matrix position that should adjacent cell.

Concrete, described Σ C1 optimizes sub-process and comprises following concrete steps:

A1, C1value value in CIM relational matrix table is arranged from large to small, select cell-of-origin corresponding to maximum C1value value as Σ C1 optimization aim community;

B1, judge whether the C1value value of this Target cell is 0, if Σ C1 optimizes sub-process and finishes, if not enter next step;

C1, obtain the PCI candidate list of this Target cell, judge in the PCI candidate list of this Target cell, whether PCI quantity is 0, if keep this community current PC I constant, Target cell is replaced by cell-of-origin corresponding to next maximum C1value value, get back to step B1; If without next C1value value, Σ C1 optimization sub-process finishes; If now in PCI candidate list, PCI quantity is not 0 selects a PCI value from PCI candidate list, give this Target cell, enter next step;

D1, upgrade C1value value, C2value value, M3value value, M6value value and the M30value value in CIM relational matrix table according to update rule, whether the Σ C1 value of calculating in CIM relational matrix table is now less than the Σ C1 value in CIM relational matrix table last time, if judge that this Target cell PCI is now effective, get back to steps A 1, if not judge that this Target cell PCI is now invalid, PCI candidate list by this PCI value before this Target cell, delete, enter step e 1;

E1, judge in the PCI candidate list of this Target cell, whether PCI quantity is 0, if the PCI value of this Target cell is returned to the PCI value before this suboptimization and the renewal of this CIM relational matrix is return, Target cell is replaced by cell-of-origin corresponding to next maximum C1value value, get back to step B1, if without next C1value value, Σ C1 optimization sub-process finishes; If now in PCI candidate list, PCI quantity is not 0 selects a PCI value from PCI candidate list, give this Target cell, get back to step D1.

Further, described Σ C2 optimization sub-process comprises following concrete steps:

A2, system judgement enter next step when Σ C1 optimizes after sub-process finishes;

B2, C2value value in CIM relational matrix table is arranged from large to small, select cell-of-origin corresponding to maximum C2value value as Σ C2 optimization aim community;

C2, judge whether the C2value value of this Target cell is 0, if Σ C2 optimizes sub-process and finishes, if not enter next step;

D2, obtain the PCI candidate list of this Target cell, judge in the PCI candidate list of this Target cell, whether PCI quantity is 0, if keep this community current PC I constant, Target cell is replaced by cell-of-origin corresponding to next maximum C2value value, get back to step C2, if without next C2value value, Σ C2 optimization sub-process finishes; If now in PCI candidate list, PCI quantity is not 0 selects a PCI value from PCI candidate list, give this Target cell, enter next step;

E2, upgrade C1value value, C2value value, M3value value, M6value value and the M30value value in CIM relational matrix table according to update rule, whether the Σ C1 value of calculating in CIM relational matrix table is now greater than the Σ C1 value in CIM relational matrix table last time, if the Σ C1 value being greater than in CIM relational matrix table last time judges that this Target cell PCI is now invalid, this PCI value is deleted from PCI candidate list before, enter step G2, if be less than or equal to the Σ C1 value in CIM relational matrix table last time, enter step F 2;

Whether F2, the Σ C2 value of calculating in CIM relational matrix table are now less than the Σ C2 value in CIM relational matrix table last time, if judge that this Target cell PCI is now effective, get back to step B2, if not judge that this Target cell PCI is now invalid, this PCI value is deleted from PCI candidate list before, enter step G2;

G2, judge in the PCI candidate list of this Target cell, whether PCI quantity is 0, if the PCI value of this Target cell is returned to the PCI value before this suboptimization and the renewal of this CIM relational matrix is return, Target cell is replaced by cell-of-origin corresponding to next maximum C2value value, get back to step C2, if without next C2value value, Σ C2 optimization sub-process finishes; If now in PCI candidate list, PCI quantity is not 0 selects a PCI value from PCI candidate list, give this Target cell, get back to step e 2.

Concrete, described Σ M3 optimizes sub-process and comprises following concrete steps:

A3, system judgement enter next step when Σ C2 optimizes after sub-process finishes;

B3, M3value value in CIM relational matrix table is arranged from large to small, select cell-of-origin corresponding to maximum M3value value as Σ M3 optimization aim community;

C3, judge whether the M3value value of this Target cell is 0, if Σ M3 optimizes sub-process and finishes, if not enter next step;

D3, obtain the PCI candidate list of this Target cell, judge in the PCI candidate list of this Target cell, whether PCI quantity is 0, if keep this community current PC I constant, Target cell is replaced by corresponding cell-of-origin or the adjacent cell of next maximum M3value value, get back to step C3, if without next M3value value, Σ M3 optimization sub-process finishes; If now in PCI candidate list, PCI quantity is not 0 selects a PCI value from PCI candidate list, give this Target cell, enter next step;

E3, upgrade the C1value value in CIM relational matrix table according to update rule, C2value value, M3value value, M6value value and M30value value, calculate whether Σ C1 value in CIM relational matrix table is now greater than the Σ C1 value in CIM relational matrix table last time and whether the Σ C2 value calculated in CIM relational matrix table is now greater than the Σ C2 value in CIM relational matrix table last time, if Σ C1 value and Σ C2 value that Σ C1 value now and Σ C2 value are all less than or equal in CIM relational matrix table last time enter step F 3, otherwise judge that this Target cell PCI is now invalid, this PCI value is deleted from PCI candidate list before, enter step G3,

Whether F3, the Σ M3 value of calculating in CIM relational matrix table are now less than the Σ M3 value in CIM relational matrix table last time, if judge that this Target cell PCI is now effective, get back to step B3, if not judge that this Target cell PCI is now invalid, this PCI value is deleted from PCI candidate list before, enter step G3;

G3, judge in the PCI candidate list of this Target cell, whether PCI quantity is 0, if the PCI value of this Target cell is returned to the PCI value before this suboptimization and the renewal of this CIM relational matrix is return, Target cell is replaced by cell-of-origin corresponding to next maximum M3value value, get back to step C3, if without next M3value value, Σ M3 optimization sub-process finishes; If now in PCI candidate list, PCI quantity is not 0 selects a PCI value from PCI candidate list, give this Target cell, get back to step e 3.

Further, described Σ M6 optimization sub-process comprises following concrete steps:

A4, system judgement enter next step when Σ M3 optimizes after sub-process finishes;

B4, M6value value in CIM relational matrix table is arranged from large to small, select cell-of-origin corresponding to maximum M6value value as Σ M6 optimization aim community;

C4, judge whether the M6value value of this Target cell is 0, if Σ M6 optimizes sub-process and finishes, if not enter next step;

D4, obtain the PCI candidate list of this Target cell, judge in the PCI candidate list of this Target cell, whether PCI quantity is 0, if keep this community current PC I constant, Target cell is replaced by cell-of-origin corresponding to next maximum M6value value, get back to step C4, if without next M6value value, Σ M6 optimization sub-process finishes; If now in PCI candidate list, PCI quantity is not 0 selects a PCI value from PCI candidate list, give this Target cell, enter next step;

E4, upgrade C1value value, C2value value, M3value value, M6value value and the M30value value in CIM relational matrix table according to update rule, whether the Σ C1 value of calculating in CIM relational matrix table is now greater than the Σ C1 value in CIM relational matrix table last time, and whether the Σ C2 value of calculating in CIM relational matrix table be now greater than the Σ C2 value in CIM relational matrix table last time, and whether the Σ M3 value of calculating in CIM relational matrix table is now greater than the Σ M3 value in CIM relational matrix table last time; If Σ C1 value now, Σ C2 value and Σ M3 value are all less than or equal to Σ C1 value, Σ C2 value and Σ M3 value in CIM relational matrix table last time, enter step F 4, otherwise judge that this Target cell PCI is now invalid, this PCI value is deleted from PCI candidate list before, enter step G4;

Whether F4, the Σ M6 value of calculating in CIM relational matrix table are now less than the Σ M6 value in CIM relational matrix table last time, if judge that this Target cell PCI is now effective, get back to step B4, if not judge that this Target cell PCI is now invalid, this PCI value is deleted from PCI candidate list before, enter step G4;

G4, judge in the PCI candidate list of this Target cell, whether PCI quantity is 0, if the PCI value of this Target cell is returned to the PCI value before this suboptimization and the renewal of this CIM relational matrix is return, Target cell is replaced by cell-of-origin corresponding to next maximum M6value value, get back to step C4, if without next M6value value, Σ M6 optimization sub-process finishes, if now in PCI candidate list, PCI quantity is not 0 selects a PCI value from PCI candidate list, give this Target cell, get back to step e 4.

Concrete, described Σ M30 optimizes sub-process and comprises following concrete steps:

A5, system judgement enter next step when Σ M6 optimizes after sub-process finishes;

B5, M30value value in CIM relational matrix table is arranged from large to small, select cell-of-origin corresponding to maximum M30value value as Σ M30 optimization aim community;

C5, judge whether the M30value value of this Target cell is 0, if Σ M30 optimizes sub-process and finishes, if not enter next step;

D5, obtain the PCI candidate list of this Target cell, judge in the PCI candidate list of this Target cell, whether PCI quantity is 0, if keep this community current PC I constant, Target cell is replaced by cell-of-origin corresponding to next maximum M30value value, gets back to step C5, if without next M30value value, Σ M30 optimization sub-process finishes, if now in PCI candidate list, PCI quantity is not 0 selects a PCI value from PCI candidate list, give this Target cell, enter next step;

E5, upgrade the C1value value in CIM relational matrix table according to update rule, C2value value, M3value value, M6value value and M30value value, whether the Σ C1 value of calculating in CIM relational matrix table is now greater than the Σ C1 value in CIM relational matrix table last time, and whether the Σ C2 value of calculating in CIM relational matrix table is now greater than the Σ C2 value in CIM relational matrix table last time, and whether the Σ M3 value of calculating in CIM relational matrix table is now greater than the Σ M3 value in CIM relational matrix table last time, and whether the Σ M6 value of calculating in CIM relational matrix table is now greater than the Σ M6 value in CIM relational matrix table last time, if Σ C1 value now, Σ C2 value, Σ M3 value and Σ M6 value are all less than or equal to the Σ C1 value in CIM relational matrix table last time, Σ C2 value, Σ M3 value and Σ M6 value, enter step F 5, otherwise judge that this Target cell PCI is now invalid, this PCI value is deleted from PCI candidate list before, enter step G5,

Whether F5, the Σ M30 value of calculating in CIM relational matrix table are now less than the Σ M30 value in CIM relational matrix table last time, if judge that this Target cell PCI is now effective, get back to step B5, if not judge that this Target cell PCI is now invalid, this PCI value is deleted from PCI candidate list before, enter step G5;

G5, judge in the PCI candidate list of this Target cell, whether PCI quantity is 0, if the PCI value of this Target cell is returned to the PCI value before this suboptimization and the renewal of this CIM relational matrix is return, Target cell is replaced by cell-of-origin corresponding to next maximum M30value value, get back to step C5, if without next M30value value, Σ M30 optimization sub-process finishes, if now in PCI candidate list, PCI quantity is not 0 selects a PCI value from PCI candidate list, give this Target cell, get back to step e 5.

The invention has the beneficial effects as follows, by above-mentioned relation matrix constructing system and method, the MR data that receive according to system, calculating gets the CIM relational matrix table that can reflect any two minizone relation value in LTE system, its can quantitative analysis minizone covering, switching and interference relationships, thereby help 4G operator to improve efficiency and the effect of adjacent section planning in LTE system, frequency planning, PCI planning; By above-mentioned PCI intelligent optimization system and method, with automation and intelligent mode, computing repeatedly by flow process obtains PCI configuration scheme, thereby greatly reduce the interference total amount of LTE wireless system, significantly improve network quality (QoS, Quality of Service) and user awareness (QoE, Quality of Experience).Meanwhile, automation or intelligentized PCI plan optimization scheme, be conducive to 4G operator and cut operating costs, improve efficiency and the effect of the network optimization.

Brief description of the drawings

Fig. 1 is the system block diagram of cell relations matrix constructing system of the present invention.

Fig. 2 is the system block diagram of PCI intelligent optimization system of the present invention.

Embodiment

Below in conjunction with drawings and Examples, describe technical scheme of the present invention in detail.

Relational matrix constructing system of the present invention, its system block diagram is as Fig. 1, formed by LTE system ontology, MR data processing module and CIM cell relations matrix module, this MR data processing module is connected with LTE system ontology, and CIM cell relations matrix module is connected with MR data processing module, wherein, MR data processing module, for obtaining MR data from the Operation and Maintenance server of LTE system ontology, and extract the crucial metrical information of MR data, be transferred to CIM cell relations matrix module, CIM cell relations matrix module, according to the crucial metrical information of the MR data that receive in certain hour section, calculate according to computation rule, obtain any cell-of-origin number of times accumulated value that conflicts with any PCI between adjacent cell, PCI obscures number of times accumulated value, the accumulated value of RSRP absolute difference, between cell-of-origin and adjacent cell, there is the accumulated value of the frequency of neighboring BS relationship, the accumulated value of the frequency that cell-of-origin and its adjacent cell occur as other adjacent areas, community simultaneously, between cell-of-origin and adjacent cell, PCI mould 3 disturbs number of times accumulated value, between cell-of-origin and adjacent cell, mould 6 disturbs number of times accumulated value, between cell-of-origin and adjacent cell, mould 30 disturbs the accumulated value of number of times, and the network configuration information that calculates gained and current LTE system is merged, generation can reflect the CIM relational matrix table of any two minizone relation value in LTE system.Relational matrix construction method of the present invention is: first, system obtains the MR data of terminal to report, then from the MR data that obtain, extract crucial metrical information, final system is according to the crucial metrical information in certain hour section, calculate any cell-of-origin number of times accumulated value that conflicts with any PCI between adjacent cell according to computation rule, PCI obscures number of times accumulated value, the accumulated value of RSRP absolute difference, between cell-of-origin and adjacent cell, there is the accumulated value of the frequency of neighboring BS relationship, the accumulated value of the frequency that cell-of-origin and its adjacent cell occur as other adjacent areas, community simultaneously, between cell-of-origin and adjacent cell, PCI mould 3 disturbs number of times accumulated value, between cell-of-origin and adjacent cell, mould 6 disturbs mould 30 between number of times accumulated value and cell-of-origin and adjacent cell to disturb number of times accumulated value, and the network configuration information that calculates gained and current LTE system is merged, generation can reflect the CIM relational matrix table of any two minizone relation value in LTE system.

PCI intelligent optimization system of the present invention, its system block diagram is as Fig. 2, formed by above-mentioned cell relations matrix constructing system and PCI intelligent optimization module, this PCI intelligent optimization module is connected with the CIM cell relations matrix module in above-mentioned cell relations matrix constructing system, wherein, CIM cell relations matrix module, is transferred to PCI intelligent optimization module by the CIM relational matrix table that can reflect any two minizone relation value in LTE system of its generation; PCI intelligent optimization module, for according to the CIM relational matrix table receiving, by computation rule, obtains the PCI allocation optimum of each community.PCI intelligent optimization method of the present invention is: first, system acquisition is to CIM relational matrix table, then calculate the C1value total degree in current C IM relational matrix table, C2value total degree, M3value total degree, M6value total degree and M30value total degree, be designated as respectively Σ C1, Σ C2, Σ M3, Σ M6 and Σ M30, carry out again PCI Optimizing Flow, optimize the PCI settings of community, upgrade CIM relational matrix table according to update rule, recalculate current Σ C1 value, Σ C2 value, Σ M3 value, Σ M6 value and Σ M30 value, judge current Σ C1, Σ C2, Σ M3, whether Σ M6 and Σ M30 are optimized to respectively optimum, if optimize complete, exit PCI Optimizing Flow, if not repeat this step of PCI Optimizing Flow.

Embodiment 1

The system block diagram of this routine relational matrix constructing system as shown in Figure 1.

The system that this routine relational matrix builds, is made up of LTE system ontology, MR data processing module and CIM cell relations matrix module, and this MR data processing module is connected with LTE system ontology, and CIM cell relations matrix module is connected with MR data processing module.

Wherein, MR data processing module is used for obtaining MR data from the Operation and Maintenance server of LTE system ontology, and extracts the crucial metrical information of MR data, is transferred to CIM cell relations matrix module.

In LTE system, LTE system ontology, refer to LTE or the LTE-A4G mobile communication system of operator's construction and operation, its feature is, once any terminal access LTE system, to automatically produce MR data, these MR data, are stored in the Operation and Maintenance server of LTE system ontology equipment vendor, operator and third party entity, can, regularly from this Operation and Maintenance server, obtain MR data.

Obtain after MR data decoding, therefrom at least can extract the crucial metrical information of following MR data:

A. the timestamp that these MR data produce;

B. handover source community key message:

Source?cell?CGI

Source?cell?PCI

Source?cell?RSRP

Source?cell?RSRQ

C switches adjacent cell key message:

Neighbor?cell?CGI

Neighbor?cell?PCI

Neighbor?cell?RSRP

Neighbor?cell?RSRQ

Visible, timestamp, cell-of-origin CGI, cell-of-origin PCI, cell-of-origin RSRP and cell-of-origin RSRQ that it comprises MR, also comprise adjacent cell CGI, adjacent cell PCI, adjacent cell RSRP and adjacent cell RSRQ.

If attempt to obtain more information from MR data, need to revise MeasuConfig parameter, allow the more information of terminal to report, such as the nearby region measurement result of inter-RAT.

The present invention only considers to switch relevant MR data with LTE internal system, do not consider and intersystem handover (as the switching between LTE and CDMA2000), because the present invention seeks to based on MR data construct relational matrix, final purpose is to promote PCI intelligent optimization.PCI conflict is LTE system inner question, and irrelevant between system.

Terminal to report MR data have three kinds of reasons, and the one, handover event (A1/A2/A3/A4/A5) triggers, and the 2nd, periodically report, the 3rd, Event triggered periodic report, i.e. same trigger event, repeatedly reports according to certain hour interval.

The present invention is indifferent to eUTRAN system and how configures, and being indifferent to MR data is that what reason reports, and is only concerned about that the Back ground Information comprising in the middle of each MR collecting can successfully obtain and record warehouse-in.

CIM cell relations matrix module is used for according to the crucial metrical information of the MR data that receive in certain hour section, calculate according to computation rule, obtain any cell-of-origin number of times accumulated value that conflicts with any PCI between adjacent cell, PCI obscures number of times accumulated value, the accumulated value of RSRP absolute difference, between cell-of-origin and adjacent cell, there is the accumulated value of the frequency of neighboring BS relationship, the accumulated value of the frequency that cell-of-origin and its adjacent cell occur as other adjacent areas, community simultaneously, between cell-of-origin and adjacent cell, PCI mould 3 disturbs number of times accumulated value, between cell-of-origin and adjacent cell, mould 6 disturbs number of times accumulated value, between cell-of-origin and adjacent cell, mould 30 disturbs the accumulated value of number of times, and the network configuration information that calculates gained and current LTE system is merged, generation can reflect the CIM relational matrix table of any two minizone relation value in LTE system.CIM (Cell Interacting Matrix) cell relations matrix module is core of the present invention.

In the time that system is used, its concrete relational matrix construction method comprises the steps:

Step 1, system obtain the MR data of terminal to report.

In this step, obtaining the method for the MR data of terminal to report, can be the MR data acquisition of the Operation and Maintenance server realization by LTE system ontology.

Step 2, from the MR data that obtain, extract crucial metrical information.

In this step, crucial metrical information comprises MR timestamp, cell-of-origin CGI, cell-of-origin PCI, cell-of-origin RSRP and cell-of-origin RSRQ, also comprises adjacent cell CGI, adjacent cell PCI and adjacent cell RSRP.

In this step, it comprises following concrete steps:

Step 301, obtain the crucial metrical information in certain hour section.

In this step, for the crucial metrical information obtaining, by the logical construction according to example shown in table 1, synthetic two-dimentional relation matrix, experiences with the real wireless environment of complete reaction, network interferences reality and user.

Table 1CIM cell relations matrix logic signal table

Wherein, Source represents the cell-of-origin identification information obtaining from MR data, has two kinds of RMs of CGI and PCI.Specific cell, its CGI and PCI have one-to-one relationship.CGI is without multiplexing, and PCI sum only has 504, and having to, it is multiplexing to consider.

Neighbor represents the adjacent cell identification information obtaining from MR data, has two kinds of RMs of CGI and PCI.Specific cell, its CGI and PCI have one-to-one relationship.CGI is without multiplexing, and PCI sum only has 504, and having to, it is multiplexing to consider.

Between the C value representative cell-of-origin that statistics is obtained from MR data and adjacent cell, conflict total degree (C1) and PCI of PCI obscures total degree (C2);

The aggregate-value of RSRP absolute difference between the R value representative cell-of-origin that statistics is obtained from MR data and adjacent cell;

Between the F value representative cell-of-origin that statistics is obtained from MR data and adjacent cell, there is the frequency value of neighboring BS relationship;

The frequency value that the N value representative cell-of-origin that statistics is obtained from MR data and adjacent cell occur as other adjacent areas, community simultaneously;

Between the M representative cell-of-origin that statistics is obtained from MR and adjacent cell, there is the cumulative number that Mod disturbs, comprise that the so-called mould 3 of number of times (M30.) that number of times (M3) that mould 3 disturbs, number of times (M6) that mould 6 disturbs and mould 30 disturb calculates, refer to the remainder of PCI value divided by 3 rear gained, mould 6 calculates and refers to that PCI value is divided by 6 rear gained remainders, and mould 30 calculates and refers to that PCI value is divided by 30 rear gained remainders.It is identical that so-called mould 3 disturbs the mould 3 of the PCI value that refers to adjacent two communities to be worth, and in like manner, mould 6 disturbs or mould 30 disturbs the mould 6 of the PCI value that also refers to adjacent two communities to be worth or mould 30 is worth identical.

Step 302, calculate any cell-of-origin number of times accumulated value that conflicts with any PCI between adjacent cell, be designated as C1value.Its computational methods are: when terminal accesses LTE system and reports MR data in certain cell-of-origin, add up one by one the whole MR data in this cell-of-origin, when if find, the PCI of its adjacent cell is identical with the PCI of this cell-of-origin, be judged to be PCI conflict, in this cell-of-origin to matrix position that should adjacent cell by C1value cumulative 1.

Step 303, the PCI calculating between any cell-of-origin and any adjacent cell obscure number of times accumulated value, are designated as C2value.Its computational methods are: when terminal accesses LTE system and reports MR data in certain cell-of-origin, add up one by one the whole MR data in this cell-of-origin, if find, there are two adjacent cells certain cell-of-origin, and the PCI of these two adjacent cells is when identical, be judged to be these two adjacent cell PCI and obscure, at the mutual corresponding matrix position of these two adjacent cells, C2value is added up to 1.

Step 304, calculate any cell-of-origin and any RSRP absolute difference accumulated value between adjacent cell, be designated as R value.Its computational methods are: when terminal accesses LTE system and reports MR data in certain cell-of-origin, add up one by one the whole MR data in this cell-of-origin, calculate the absolute difference of the RSRP of this cell-of-origin and the RSRP of its adjacent cell, then in this cell-of-origin to matrix position that should adjacent cell by cumulative R value this absolute difference.

Step 305, calculate any cell-of-origin and any accumulated value of occurrence frequency between adjacent cell, be designated as F value.Its computational methods are: when terminal accesses LTE system and reports MR data in certain cell-of-origin, add up one by one whole MR data, if find Wei Qi adjacent area, another community, in this cell-of-origin to matrix position that should adjacent cell by F value cumulative 1.

Step 306, calculate any cell-of-origin and its any adjacent cell as the accumulated value of the frequency of other community neighboring BS relationship, be designated as N value.Its computational methods are: when accessing LTE system in terminal Mou community and reporting MR data, add up one by one whole MR data, if find this cell-of-origin and its adjacent cell while adjacent area as aforementioned certain community, at the matrix position of corresponding its adjacent cell in this cell-of-origin, N value is added up to 1, in like manner, also in this adjacent cell as cell-of-origin, and this cell-of-origin as the matrix position of adjacent cell by N value cumulative 1.

Step 307, calculate any cell-of-origin and arbitrarily between adjacent cell, occur that mould 3 disturbs the accumulated value of number of times, is designated as M3value.Its computational methods are: when terminal accesses LTE system and reports MR data in certain cell-of-origin, add up one by one the whole MR data in this cell-of-origin, if PCI mould 3 values of finding its adjacent cell and the PCI mould 3 of this cell-of-origin are worth when identical, be judged to be mould 3 and disturb, in this cell-of-origin to matrix position that should adjacent cell by M3value cumulative 1.

Step 308, calculate any cell-of-origin and arbitrarily between adjacent cell, occur that mould 6 disturbs the accumulated value of number of times, is designated as M6value.Its computational methods are: when terminal accesses LTE system and reports MR data in certain cell-of-origin, add up one by one the whole MR data in this cell-of-origin, if PCI mould 6 values of finding its adjacent cell and the PCI mould 6 of this cell-of-origin are worth when identical, be judged to be mould 6 and disturb, in this cell-of-origin to matrix position that should adjacent cell by M6value cumulative 1.

Step 309, calculate any cell-of-origin and arbitrarily between adjacent cell, occur that mould 30 disturbs the aggregate-value of number of times, is designated as M30value.Its computational methods are: when terminal accesses LTE system and reports MR data in certain cell-of-origin, add up one by one the whole MR data in this cell-of-origin, if PCI mould 30 values of finding its adjacent cell and the PCI mould 30 of this cell-of-origin are worth when identical, be judged to be mould 30 and disturb, in this cell-of-origin to matrix position that should adjacent cell by M30value cumulative 1.

C1value, C2value, R value, F value, M3value, M6value and M30value that step 310, basis obtain, merge with the network configuration information of current LTE system, according to cell-of-origin dimension and adjacent cell dimension mode one to one, synthesize a two-dimentional relation matrix table, obtain reflecting the CIM relational matrix table of any two minizone relation value in LTE system.

Be exemplified below:

100: system obtains MR data by Operation and Maintenance server or other modes of LTE system ontology.

200: from the MR data that obtain, extract crucial metrical information, the information input databases one by one such as the timestamp, cell-of-origin CIG/PCI, adjacent cell CGI/PCI, cell-of-origin RSRP/RSRQ, adjacent area RSRP/RSRQ that the crucial metrical information by MR data: MR produces.

301: C1value between calculation plot.C1 represents PCI conflict (PCI Collision) cumulative number.When UE connecting system while reporting MR, if find that CGI8/PCI8 and CGI4/PCI4 have neighboring BS relationship and PCI8=PCI4, so C1Value is added up to 1 at CGI8/PCI8->CGI4/PCI4 matrix position in CGI8/PCI8 community.In like manner, when UE connecting system while reporting MR, if find that CGI4/PCI4 and CGI8/PCI8 have neighboring BS relationship and PCI4=PCI8, add up 1 at CGI4/PCI4->CGI8/PCI8 matrix position by C1value in CGI4/PCI4 community.

302: C2value between calculation plot.C2 represents that PCI obscures (PCI Confusion) cumulative number.No matter CGI4/PCI4 and CGI8/PCI8 have or not handoff relation to have or not covering relation, as long as the Zhe Liang community adjacent area existence such as CGI2/PCI2 as other communities respectively, from the MR of cell-of-origin CGI2/PCI2, find that CGI4/PCI4 and CGI8/PCI8 are all its Target cells, and PCI4=PCI8, so C2value is added up to 1 at CGI4/PCI4->CGI8/PCI8 matrix position, simultaneously need to be at CGI8/PCI8->CGI4/PCI4 matrix position by C2value cumulative 1

303: P value between calculation plot.P represents RSRP difference (Received Signal Reference Power) aggregate-value.When UE connecting system while reporting MR in CGI8/PCI8 community, calculate the absolute difference of the RSRP of cell-of-origin CGI8/PCI8 and the RSRP of its adjacent area CGI4/PCI4, then at CGI8/PCI8->CGI4/PCI4 matrix position by P Value this absolute difference that adds up.In like manner, when UE connecting system while reporting MR in CGI4/PCI4 community, calculate the RSRP of cell-of-origin CGI4/PCI4 and the RSRP absolute difference of Target cell CGI8/PCI8, then at CGI4/PCI4->CGI8/PCI8 matrix position by P value this absolute difference that adds up.

304: F value between calculation plot.F represents the accumulated value of cell-of-origin and adjacent cell occurrence frequency (Frequence).When UE connecting system while reporting MR in CGI8/PCI8 community, if find that CGI4/PCI4 is its adjacent area, so at CGI8/PCI8->CGI4/PCI4 matrix position by cumulative 1 time of F Value.In like manner, when UE connecting system while reporting MR in CGI4/PCI4 community, if find that CGI8/PCI8 is its adjacent area, so at CGI4/PCI4->CGI8/PCI8 matrix position by cumulative 1 time of F value.

305: N value between calculation plot.N represents the frequency that cell-of-origin and adjacent cell occur as this situation of adjacent cell of other communities simultaneously.No matter CGI4/PCI4 and CGI8/PCI8 have or not handoff relation to have or not covering relation, as long as the adjacent area existence such as CGI2/PCI2 as other communities simultaneously of Zhe Liang community, finds from the MR data of cell-of-origin CGI2/PCI2 that CGI4/PCI4 and CGI8/PCI8 are its adjacent cells simultaneously.So at CGI4/PCI4->CGI8/PCI8 matrix position, N value is added up to 1, simultaneously need to be at CGI8/PCI8->CGI4/PCI4 matrix position by N value cumulative 1.N value has added up the frequency that cell-of-origin and adjacent cell occur as the adjacent cell of other communities simultaneously, with cell-of-origin and adjacent cell PCI settings irrelevant.

306: M3value between calculation plot.M3 represents and between cell-of-origin and adjacent cell, occurs the number of times that mould 3 disturbs.When UE connecting system while reporting MR in CGI8/PCI8 community, if find that CGI4/PCI4 is its adjacent area, and it is identical that mould 3 values of PCI8 and the mould 3 of PCI4 are worth, then at CGI8/PCI8->CGI4/PCI4 matrix position by cumulative 1 time of M3Value.In like manner, when UE connecting system while reporting MR in CGI4/PCI4 community, if find that CGI8/PCI8 is its adjacent area, and mould 3 values of PCI4 and the mould 3 of PCI8 be worth identical, so M3value is added up 1 time at CGI4/PCI4->CGI8/PCI8 matrix position.

307: M6value between calculation plot.M6 represents and between cell-of-origin and adjacent cell, occurs the number of times that mould 6 disturbs.When UE connecting system while reporting MR in CGI8/PCI8 community, if find that CGI4/PCI4 is its adjacent area, and it is identical that mould 6 values of PCI8 and the mould 6 of PCI4 are worth, then at CGI8/PCI8->CGI4/PCI4 matrix position by cumulative 1 time of M6Value.In like manner, when UE connecting system while reporting MR in CGI4/PCI4 community, if find that CGI8/PCI8 is its adjacent area, and mould 6 values of PCI4 and the mould 6 of PCI8 be worth identical, so M6value is added up 1 time at CGI4/PCI4->CGI8/PCI8 matrix position.

308: M30value between calculation plot.M30 represents and between cell-of-origin and adjacent cell, occurs the number of times that mould 30 disturbs.When UE connecting system while reporting MR in CGI8/PCI8 community, if find that CGI4/PCI4 is its adjacent area, and it is identical that mould 30 values of PCI8 and the mould 30 of PCI4 are worth, then at CGI8/PCI8->CGI4/PCI4 matrix position by cumulative 1 time of M30Value.In like manner, when UE connecting system while reporting MR in CGI4/PCI4 community, if find that CGI8/PCI8 is its adjacent area, and mould 30 values of PCI4 and the mould 30 of PCI8 be worth identical, so M30value is added up 1 time at CGI4/PCI4->CGI8/PCI8 matrix position.

400: synthetic CIM minizone relational matrix.By above-mentioned calculating gained, according to cell-of-origin and the adjacent cell corresponding synthetic XY relational matrix table of mode one by one, Y direction is listed all cell-of-origins, X-direction is listed the whole adjacent cells relevant with cell-of-origin, so can inquire the relation value C1/C2/R/F/N/M3/M6/M30 value of arbitrary cell-of-origin and the whole adjacent cells of periphery from this table.So far, the calculation process of CIM cell relations matrix completes.

CIM relational matrix will present as table 2 pattern in tables of data:

Table 2CIM relational matrix table

The feature of CIM cell relations matrix is: F value is relevant with adjacent cell occurrence frequency with N value and cell-of-origin, with the PCI configuration-independent of cell-of-origin and adjacent cell.C value is relevant, also relevant with F value and N value with PCI configuration with M value, under relationship is shown in:

If cell-of-origin is identical with the PCI value of adjacent cell, C1value=F value, otherwise C1value=0;

If cell-of-origin is identical with the PCI value of adjacent cell, C2value=N value, otherwise C2value=0;

If cell-of-origin is identical divided by the remainder of 3 gained respectively with the PCI value of adjacent cell, M3value=F value, otherwise M3value=0;

If cell-of-origin is identical divided by the remainder of 6 gained respectively with the PCI value of adjacent cell, M6value=F value, otherwise M6value=0;

If cell-of-origin is identical divided by the remainder of 30 gained respectively with the PCI value of adjacent cell, M30value=F value, otherwise M30value=0.

Embodiment 2

The system block diagram of this routine PCI intelligent optimization system as shown in Figure 2.

The PCI intelligent optimization system that this is routine, is made up of the relational matrix constructing system in embodiment 1 and PCI intelligent optimization module, and this PCI intelligent optimization module is connected with the CIM cell relations matrix module in above-mentioned relational matrix constructing system.

Wherein, CIM cell relations matrix module, is transferred to PCI intelligent optimization module by the CIM relational matrix table that can reflect any two minizone relation value in LTE system of its generation.

PCI intelligent optimization module, for according to the CIM relational matrix table receiving, by computation rule, obtains the PCI allocation optimum of each community.

This computation rule can be specially: according to the CIM relational matrix table receiving, calculate the C1value total degree in current C IM relational matrix table, C2value total degree, M3value total degree, M6value total degree and M30value total degree, then, from the PCI candidate list of each community, be the optimization settings that PCI is selected in each community repeatedly, after every suboptimization, all need to upgrade CIM relational matrix table according to CIM update rule, finally make C1value total degree, C2value total degree, M3value total degree, these five numerical value minimums of M6value total degree and M30value total degree, here, the PCI candidate list of each community, its initial table is that PCI span is the list of digital 0-503.

When use, its PCI intelligent optimization method specifically comprises the steps:

Step 10, system acquisition are to CIM relational matrix table;

C1value total degree, C2value total degree, M3value total degree, M6value total degree and M30value total degree in step 20, system-computed current C IM relational matrix table, be designated as respectively Σ C1, Σ C2, Σ M3, Σ M6 and Σ M30.

Step 30, execution PCI Optimizing Flow, optimize the PCI settings of community, upgrade CIM relational matrix table according to update rule, recalculate current Σ C1 value, Σ C2 value, Σ M3 value, Σ M6 value and Σ M30 value, judge whether current Σ C1, Σ C2, Σ M3, Σ M6 and Σ M30 are optimized to respectively optimum, if optimize completely, exit PCI Optimizing Flow, if not repeat this step.

In this step, PCI Optimizing Flow refers to: the order of optimizing sub-process, Σ C2 optimization sub-process, Σ M3 optimization sub-process, Σ M6 optimization sub-process and Σ M30 optimization sub-process with Σ C1 is optimized the PCI settings of each community respectively.Update rule refers to: in the time that the PCI value of cell-of-origin and adjacent cell is identical, this cell-of-origin equals F value value to the C1value value of matrix position that should adjacent cell, otherwise this cell-of-origin equals 0 to the C1value value of matrix position that should adjacent cell; In the time that the PCI value of cell-of-origin and adjacent cell is identical, this cell-of-origin equals Nvalue value to the C2value value of matrix position that should adjacent cell, otherwise this cell-of-origin equals 0 to the C2value value of matrix position that should adjacent cell; In the time that the PCI value of cell-of-origin and adjacent cell is identical divided by the remainder of 3 gained respectively, this cell-of-origin equals F value value to the M3value value of matrix position that should adjacent cell, otherwise this cell-of-origin equals 0 to the M3value value of matrix position that should adjacent cell; In the time that the PCI value of cell-of-origin and adjacent cell is identical divided by the remainder of 6 gained respectively, this cell-of-origin equals F value value to the M6value value of matrix position that should adjacent cell, otherwise this cell-of-origin equals 0 to the M6value value of matrix position that should adjacent cell; In the time that the PCI value of cell-of-origin and adjacent cell is identical divided by the remainder of 30 gained respectively, this cell-of-origin equals F value value to the M30value value of matrix position that should adjacent cell, otherwise this cell-of-origin equals 0 to the M30value value of matrix position that should adjacent cell.

Wherein, Σ C1 optimization sub-process comprises following concrete steps:

Σ C2 optimizes sub-process and comprises following concrete steps:

Σ M3 optimizes sub-process and comprises following concrete steps:

Σ M6 optimizes sub-process and comprises following concrete steps:

Σ M30 optimizes sub-process and comprises following concrete steps:

After Σ C1 optimization sub-process, Σ C2 optimization sub-process, Σ M3 optimization sub-process, Σ M6 optimization sub-process and Σ M30 optimization sub-process all finish, this PCI Optimizing Flow finishes, and now obtains the PCI settings of Ge community, is PCI and has optimized value.

Claims

1. relational matrix constructing system, comprise LTE system ontology, it is characterized in that, also comprise MR data processing module and CIM cell relations matrix module, described MR data processing module is connected with LTE system ontology, and CIM cell relations matrix module is connected with MR data processing module;

2. relational matrix construction method, is characterized in that, comprises the following steps:

Step 1, system obtain the MR data of terminal to report;

Step 2, from the MR data that obtain, extract crucial metrical information;

3. relational matrix construction method according to claim 2, is characterized in that, step 3 comprises following concrete steps:

Step 301, obtain the crucial metrical information in certain hour section;

4. relational matrix construction method according to claim 3, it is characterized in that, in step 302, conflict with any PCI between the adjacent cell method of number of times accumulated value of any cell-of-origin of described calculating is: when terminal accesses LTE system and reports MR data in certain cell-of-origin, add up one by one the whole MR data in this cell-of-origin, when the PCI of its adjacent cell is identical with the PCI of this cell-of-origin if find, be judged to be PCI conflict, in this cell-of-origin to matrix position that should adjacent cell by C1value cumulative 1;

In step 303, the method that any cell-of-origin of described calculating and the arbitrarily PCI between adjacent cell obscure number of times accumulated value is: when terminal accesses LTE system and reports MR data in certain cell-of-origin, add up one by one the whole MR data in this cell-of-origin, if find, there are two adjacent cells certain cell-of-origin, and the PCI of these two adjacent cells is when identical, be judged to be these two adjacent cell PCI and obscure, at the mutual corresponding matrix position of these two adjacent cells, C2value is added up to 1;

In step 307, the method that occurs the number of times that mould 3 disturbs between any cell-of-origin of described calculating and arbitrarily adjacent cell is: when terminal accesses LTE system and reports MR data in certain cell-of-origin, add up one by one the whole MR data in this cell-of-origin, if PCI mould 3 values of finding its adjacent cell and the PCI mould 3 of this cell-of-origin are worth when identical, be judged to be mould 3 and disturb, in this cell-of-origin to matrix position that should adjacent cell by M3value cumulative 1;

5. relational matrix construction method according to claim 3, it is characterized in that, in step 304, any cell-of-origin of described calculating and the arbitrarily method of the RSRP difference accumulated value between adjacent cell are: when terminal accesses LTE system and reports MR data in certain cell-of-origin, add up one by one the whole MR data in this cell-of-origin, calculate the absolute difference of the RSRP of this cell-of-origin and the RSRP of its adjacent cell, then in this cell-of-origin to matrix position that should adjacent cell by cumulative R value this absolute difference;

In step 305, between any cell-of-origin of described calculating and any adjacent cell, the method for the accumulated value of occurrence frequency is: when terminal accesses LTE system and reports MR data in certain cell-of-origin, add up one by one whole MR data, if find Wei Qi adjacent area, another community, in this cell-of-origin to matrix position that should adjacent cell by F value cumulative 1.

6. according to relational matrix construction method described in claim 3 or 4 or 5, it is characterized in that, in step 306, any cell-of-origin of described calculating and its any adjacent cell as the method for the accumulated value of the frequency of other community neighboring BS relationship are: when accessing LTE system in terminal Mou community and reporting MR data, add up one by one whole MR data, if find, this cell-of-origin and its adjacent cell, simultaneously as the adjacent area of aforementioned certain community, add up 1 at the matrix position of corresponding its adjacent cell in this cell-of-origin by N value.

7.PCI intelligent optimization system, is characterized in that, comprises relational matrix constructing system claimed in claim 1, also comprises PCI intelligent optimization module, and described PCI intelligent optimization module is connected with CIM cell relations matrix module;

8. PCI intelligent optimization system according to claim 7, it is characterized in that, described computation rule is: according to the CIM relational matrix table receiving, calculate the C1value total degree in current C IM relational matrix table, C2value total degree, M3value total degree, M6value total degree and M30value total degree, then, from the PCI candidate list of each community, be the optimization settings that PCI is selected in each community repeatedly, after every suboptimization, all need to upgrade CIM relational matrix table according to CIM update rule, finally make C1value total degree, C2value total degree, M3value total degree, these five numerical value minimums of M6value total degree and M30value total degree, the PCI candidate list of described each community, its initial table is that PCI span is the list of digital 0-503.

9.PCI intelligent optimization method, is characterized in that, comprises the following steps:

Step 10, system acquisition are to CIM relational matrix table;

10. PCI intelligent optimization method according to claim 9, it is characterized in that, in step 30, described PCI Optimizing Flow refers to: optimize sub-process, Σ C2 optimization sub-process, Σ M3 optimization sub-process, Σ M6 optimization sub-process and Σ M30 optimization sub-process with Σ C1 the PCI settings of each community are optimized.

11. PCI intelligent optimization methods according to claim 10, it is characterized in that, in step 30, described update rule refers to: in the time that the PCI value of cell-of-origin and adjacent cell is identical, this cell-of-origin equals F value value to the C1value value of matrix position that should adjacent cell, otherwise this cell-of-origin equals 0 to the C1value value of matrix position that should adjacent cell;

12. according to PCI intelligent optimization method described in claim 11, it is characterized in that, described Σ C1 optimizes sub-process and comprises following concrete steps:

E1, judge in the PCI candidate list of this Target cell, whether PCI quantity is 0, if the PCI value of this Target cell is returned to the PCI value before this suboptimization and the renewal of this CIM relational matrix table is return, Target cell is replaced by cell-of-origin corresponding to next maximum C1value value, get back to step B1, if without next C1value value, Σ C1 optimization sub-process finishes; If now in PCI candidate list, PCI quantity is not 0 selects a PCI value from PCI candidate list, give this Target cell, get back to step D1.

13. according to PCI intelligent optimization method described in claim 11, it is characterized in that, described Σ C2 optimizes sub-process and comprises following concrete steps:

G2, judge in the PCI candidate list of this Target cell, whether PCI quantity is 0, if the PCI value of this Target cell is returned to the PCI value before this suboptimization and the renewal of this CIM relational matrix table is return, Target cell is replaced by cell-of-origin corresponding to next maximum C2value value, get back to step C2, if without next C2value value, Σ C2 optimization sub-process finishes; If now in PCI candidate list, PCI quantity is not 0 selects a PCI value from PCI candidate list, give this Target cell, get back to step e 2.

14. according to PCI intelligent optimization method described in claim 11, it is characterized in that, described Σ M3 optimizes sub-process and comprises following concrete steps:

G3, judge in the PCI candidate list of this Target cell, whether PCI quantity is 0, if the PCI value of this Target cell is returned to the PCI value before this suboptimization and the renewal of this CIM relational matrix table is return, Target cell is replaced by cell-of-origin corresponding to next maximum M3value value, get back to step C3, if without next M3value value, Σ M3 optimization sub-process finishes; If now in PCI candidate list, PCI quantity is not 0 selects a PCI value from PCI candidate list, give this Target cell, get back to step e 3.

15. according to PCI intelligent optimization method described in claim 11, it is characterized in that, described Σ M6 optimizes sub-process and comprises following concrete steps:

G4, judge in the PCI candidate list of this Target cell, whether PCI quantity is 0, if the PCI value of this Target cell is returned to the PCI value before this suboptimization and the renewal of this CIM relational matrix table is return, Target cell is replaced by cell-of-origin corresponding to next maximum M6value value, get back to step C4, if without next M6value value, Σ M6 optimization sub-process finishes, if now in PCI candidate list, PCI quantity is not 0 selects a PCI value from PCI candidate list, give this Target cell, get back to step e 4.

16. according to PCI intelligent optimization method described in claim 11 or 12 or 13 or 14 or 15, it is characterized in that, described Σ M30 optimizes sub-process and comprises following concrete steps:

G5, judge in the PCI candidate list of this Target cell, whether PCI quantity is 0, if the PCI value of this Target cell is returned to the PCI value before this suboptimization and the renewal of this CIM relational matrix table is return, Target cell is replaced by cell-of-origin corresponding to next maximum M30value value, get back to step C5, if without next M30value value, Σ M30 optimization sub-process finishes, if now in PCI candidate list, PCI quantity is not 0 selects a PCI value from PCI candidate list, give this Target cell, get back to step e 5.