CN102883329B - The distribution method of subdistrict frequency point and device - Google Patents

Abstract

The present invention provides distribution method and the device of a kind of subdistrict frequency point, relates to moving communicating field, does not account for, during for solving distribution frequency in prior art, the factor that community interferes so that network small area to be allocated interferes relatively larger technical problem and invents.The distribution method of described subdistrict frequency point, including: from the community of network to be planned, select the first community；According to the frequency of other communities distributing frequency except described first community in described community and signal intensity, calculate the degree of disturbance of described network to be planned when described first cell configuration is each frequency respectively；From each frequency described, the first frequency of described first community corresponding when selecting the degree of disturbance of described network to be planned minimum, as the frequency of described first community.The present invention can reduce interfering between the community of frequency to be allocated.

Description

The distribution method of subdistrict frequency point and device

Technical field

The present invention relates to moving communicating field, particularly relate to distribution method and the device of a kind of subdistrict frequency point.

Background technology

Due to the feature of TD-SCDMA (TimeDivision-SynchronousCodeDivisionMultipleAccess, TD SDMA) network formats, current TD-SCDMA adopts the networking mode of N frequency.N frequency point networking is to distribute N number of carrier frequency in a sector, determines that a frequency is as main carrier frequency in these carrier frequency, and remaining frequency is assistant carrier frequency.Only sending DwPTS (descending guiding time slot) and broadcast message in sector on main carrier frequency, multiple frequencies share a broadcast.Visible, TD-SCDMA dominant frequency point is the networking mode adopting alien frequencies, so needing to carry out the planning of rational dominant frequency point, bigger frequency interference otherwise can be brought to cause system performance degradation.

In prior art, the automatic planning algorithm of frequency includes heuritic approach (greedy algorithm, local search algorithm etc.), intelligent algorithm (including genetic algorithm, simulated annealing etc.).In existing algorithm, algorithm itself is utilized to find relative optimal solution, it does not have to consider the factor that community interferes so that network small area to be allocated interferes relatively larger.

Summary of the invention

The technical problem to be solved in the present invention is to provide distribution method and the device of a kind of subdistrict frequency point, it is possible to reduce interfering of network small area to be allocated.

For solving above-mentioned technical problem, embodiments of the invention provide technical scheme as follows:

On the one hand, it is provided that the distribution method of a kind of subdistrict frequency point, including:

The first community is selected from the community of network to be planned；

According to the frequency of other communities distributing frequency except described first community in described community and signal intensity, calculate the degree of disturbance of described network to be planned when described first cell configuration is each frequency respectively；

From each frequency described, the first frequency of described first community corresponding when selecting the degree of disturbance of described network to be planned minimum, as the frequency of described first community.

Described first community be the unallocated frequency of described network to be planned community in the highest community of risk factor, the risk factor of described first community represents that described first community and other each communities except described first community in described network to be planned occur the frequency of the probability of interference, described first frequency and described other communities having distributed frequency to differ.

Described first community be described network to be planned community in the highest community of risk factor；The risk factor of described first community represents that described first community, with other each communities except described first community in described network to be planned, the probability of interference occurs；

Described from each frequency described, the first frequency of described first community corresponding when selecting the degree of disturbance of described network to be planned minimum, the step as the frequency of described first community includes:

When described first community is assigned with frequency, using described first frequency to update the former frequency of described first community, the frequency as other communities being updated frequency in the new frequency of described first community, described first frequency and described network to be planned differs.

The degree of disturbance of described network to be planned calculates according to below equation:

$\underset{i=1}{\overset{N}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}E(j-i)*X(j-i),$

Wherein, i and j is community sequence number, i ∈ [1, N], j ∈ [1, N], i ≠ j；N is the number of cells in described network to be planned；

E (j-i) is the community j correlation intensity to the first cell i；

When the frequency of community j and the first cell i is identical, X (j-i)=1；When the frequency of community j and the first cell i differs, X (j-i)=0.

The risk factor of described first community is calculated by below equation:

$D_i=\underset{j=1}{\overset{N}{\mathrm{\Σ}}}E(j-i),$

Wherein, Di is the risk factor of the first cell i；E (j-i) is the community j correlation intensity to the first cell i, represents the community j general impacts degree to the first cell i；

J and i is community sequence number, i ∈ [1, N], j ∈ [1, N], i ≠ j；N is the quantity of all communities in network to be planned.

The correlation intensity E (j-i) of the first cell i is calculated by described community j by below equation:

Wherein, f (j-i) is at point (x, the signal of community, the position j y) correlation intensity to the signal of the first cell i；(x y) is positioned at the prime coverage area territory of the first cell i to point；As Pi-Pj < lim, f (j-i)=1；As Pi-Pj >=lim, f (j-i)=0；Pj is that community j is at point (x, signal intensity y)；Pi is that the first cell i is at point (x, signal intensity y)；Lim is interference threshold；

Or, the correlation intensity E (j-i) of the first cell i is calculated by described community j by below equation:

Wherein, R is the quantity of the grid divided according to predefined size of the community j in described network to be planned；R is the sequence number of the grid of community j；Q is the quantity of the grid that the first cell i in described network to be planned divides according to predefined size；Q is the sequence number of the grid of the first cell i；The position that f (q-r) is grid q, the signal of the grid r correlation intensity to the signal of grid q；During Pq-Pr < lim, f (q-r)=1；During Pq-Pr >=lim, f (q-r)=0；Pr is the community j signal intensity at grid r place；Pq is first cell i signal intensity at grid r place；Lim is interference threshold value.

The difference of the signal intensity of described grid r and first cell i minimum signal strength in its prime coverage area territory is more than or equal to described interference threshold value.

On the other hand, it is provided that the distributor of a kind of subdistrict frequency point, including:

Select unit, from the community of network to be planned, select the first community；

Network interferences degree computing unit, according to the frequency of other communities distributing frequency except described first community in described community and signal intensity, calculates the degree of disturbance of described network to be planned when described first cell configuration is each frequency respectively；

Selecting unit, from each frequency described, the first frequency of described first community corresponding when selecting the degree of disturbance of described network to be planned minimum, as the frequency of described first community.

Described first community be the unallocated frequency of described network to be planned community in the highest community of risk factor, the risk factor of described first community represents that described first community and other each communities except described first community in described network to be planned occur the frequency of the probability of interference, described first frequency and described other communities having distributed frequency to differ.

Described first community be described network to be planned community in the highest community of risk factor；The risk factor of described first community represents that described first community, with other each communities except described first community in described network to be planned, the probability of interference occurs；

Described device also includes:

Updating block, when described first community is assigned with frequency, using described first frequency to update the former frequency of described first community, the frequency as other communities being updated frequency in the new frequency of described first community, described first frequency and described network to be planned differs.

Described network interferences degree computing unit calculates the degree of disturbance of described network to be planned according to below equation:

$\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}\underset{j=1}{\overset{N}{\mathrm{\&Sigma;}}}E(j-i)*X(j-i),$

Wherein, i and j is community sequence number, i ∈ [1, N], j ∈ [1, N], i ≠ j；N is the number of cells in described network to be planned；

E (j-i) is the community j correlation intensity to the first cell i；

When the frequency of community j and the first cell i is identical, X (j-i)=1；When the frequency of community j and the first cell i differs, X (j-i)=0.

The distributor of described subdistrict frequency point, also includes:

Community risk factor computing unit, for being calculated the risk factor of described first community by below equation:

$D_i=\underset{j=1}{\overset{N}{\mathrm{\&Sigma;}}}E(j-i),$

Wherein, Di is the risk factor of the first cell i；E (j-i) is the community j correlation intensity to the first cell i, represents the community j general impacts degree to the first cell i；

J and i is community sequence number, i ∈ [1, N], j ∈ [1, N], i ≠ j；N is the quantity of all communities in network to be planned.

The distributor of described subdistrict frequency point, also includes:

Correlation intensity computing unit, for calculating the described community j correlation intensity E (j-i) to the first cell i by below equation:

Wherein, f (j-i) is at point (x, the signal of community, the position j y) correlation intensity to the signal of the first cell i；(x y) is positioned at the prime coverage area territory of the first cell i to point；As Pi-Pj < lim, f (j-i)=1；As Pi-Pj >=lim, f (j-i)=0；Pj is that community j is at point (x, signal intensity y)；Pi is that the first cell i is at point (x, signal intensity y)；Lim is interference threshold；

Or, for calculating the described community j correlation intensity E (j-i) to the first cell i by below equation:

Wherein, R is the quantity of the grid divided according to predefined size of the community j in described network to be planned；R is the sequence number of the grid of community j；Q is the quantity of the grid that the first cell i in described network to be planned divides according to predefined size；Q is the sequence number of the grid of the first cell i；The position that f (q-r) is grid q, the signal of the grid r correlation intensity to the signal of grid q；During Pq-Pr < lim, f (q-r)=1；During Pq-Pr >=lim, f (q-r)=0；Pr is the community j signal intensity at grid r place；Pq is first cell i signal intensity at grid r place；Lim is interference threshold value.

Embodiments of the invention have the advantages that

In such scheme, when distributing frequency to community, from the community of network to be planned, select the first community；According to the frequency of other communities distributing frequency except described first community in described community and signal intensity, calculate the degree of disturbance of described network to be planned when described first cell configuration is each frequency respectively；From each frequency described, the first frequency of described first community corresponding when selecting the degree of disturbance of described network to be planned minimum, as the frequency of described first community.Due to when distributing frequency to community, consider the first cell configuration degree of disturbance of described network to be planned when being each frequency, and the first frequency of described first community corresponding when selecting the degree of disturbance of network to be planned minimum, frequency as described first community, therefore, the frequency point allocation scheme of community can reduce interfering of network small area to be allocated.

Accompanying drawing explanation

Fig. 1 is the schematic flow sheet of the distribution method of a kind of subdistrict frequency point of the present invention；

Fig. 2 is in the application scenarios of the distribution method of a kind of subdistrict frequency point of the present invention, and initial frequency generates the schematic flow sheet of scheme；

Fig. 3 is in the application scenarios of the distribution method of a kind of subdistrict frequency point of the present invention, optimizes frequency and generates the schematic flow sheet of scheme；

Fig. 4 is the structural representation of the distributor of a kind of subdistrict frequency point of the present invention.

Detailed description of the invention

For making embodiments of the invention solve the technical problem that, technical scheme and advantage clearly, be described in detail below in conjunction with the accompanying drawings and the specific embodiments.

As it is shown in figure 1, be the distribution method of a kind of subdistrict frequency point of the present invention, including:

Step 11, selects the first community from the community of network to be planned；

Step 12, according to the frequency of other communities distributing frequency except described first community in described community and signal intensity, calculates the degree of disturbance of described network to be planned when described first cell configuration is each frequency respectively；

Step 13, from each frequency described, the first frequency of described first community corresponding when selecting the degree of disturbance of described network to be planned minimum, as the frequency of described first community.

When being applied to the scene of original allocation scheme generating frequency for the community of network to be planned, described first community be the unallocated frequency of described network to be planned community in the highest community of risk factor, the risk factor of described first community represents that described first community and other each communities except described first community in described network to be planned occur the frequency of the probability of interference, described first frequency and described other communities having distributed frequency to differ.Below to be applied to greedy algorithm, embodiments of the invention are described.Greedy algorithm is to utilize certain Experience norms, is quickly generated a frequency allocation plan.This algorithm generates for the initial solution of following computing intelligence.Concrete steps include:

Step 1, finds in whole community, the community that risk factor is maximum；

Step 2, distributes any one frequency to the community that described risk factor is maximum；

Step 3, finds in the community of unallocated frequency, the first community that risk factor is maximum；

Step 4, distributes a frequency to the first community so that object function (namely the degree of disturbance of network to be planned) is minimum；Particularly as follows: according to the frequency of other communities distributing frequency except described first community in described community and signal intensity, calculate the degree of disturbance of described network to be planned when described first cell configuration is each frequency respectively；From each frequency described, the first frequency of described first community corresponding when selecting the degree of disturbance of described network to be planned minimum, as the frequency of described first community.

Step 5, if also having the community of unallocated frequency, then jumps to step 4.

When being applied to the scene of allocative decision of the optimizing cells frequency for network to be planned, described first community be described network to be planned community in the highest community of risk factor；The risk factor of described first community represents that described first community, with other each communities except described first community in described network to be planned, the probability of interference occurs；Described from each frequency described, first frequency of described first community corresponding when selecting the degree of disturbance of described network to be planned minimum, step as the frequency of described first community includes: when described first community is assigned with frequency, described first frequency is used to update the former frequency of described first community, frequency as other communities being updated frequency in the new frequency of described first community, described first frequency and described network to be planned differs.Below to be applied to local search algorithm, embodiments of the invention are described.Local search algorithm is on existing frequency allocation plan basis, being constantly attempting to the former allocative decision of improvement, until can not improve.Concrete steps include:

Step 1, imports original frequency programme；

Step 2, in whole communities, chooses the first community that risk factor is maximum；

Step 3, trial is that all of frequency is distributed in the first community, and selects the frequency making target function value minimum；Particularly as follows: according to the frequency of other communities distributing frequency except described first community in described community and signal intensity, calculate the degree of disturbance of described network to be planned when described first cell configuration is each frequency respectively；From each frequency described, the first frequency of described first community corresponding when selecting the degree of disturbance of described network to be planned minimum, as the frequency of described first community.

Step 4, finds in the community without trial distribution frequency, the second community that risk factor is maximum；

Step 5, trial is that all of frequency is distributed in the second community, and selects the frequency making target function value minimum；

Step 6, if also having the community of unallocated frequency, then jumps to step 5.

Algorithm constantly scans community, attempts choosing a better frequency to each community.If it find that choose better frequency can to certain community, target function value can be made to reduce, then accept this and revise.If algorithm scans all of community, all can not change the frequency that they distribute, then algorithm terminates.

The Experience norms that above-mentioned greedy algorithm and local search algorithm use is: the cell preference distribution that risk factor is maximum, and the frequency that the interference that to select to be subject to dividing frequency community is minimum.

Wherein, the degree of disturbance of described network to be planned calculates according to below equation:

$\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}\underset{j=1}{\overset{N}{\mathrm{\&Sigma;}}}E(j-i)*X(j-i),$

Wherein, i and j is community sequence number, i ∈ [1, N], j ∈ [1, N], i ≠ j；N is the number of cells in described network to be planned；

E (j-i) is the community j correlation intensity to the first cell i；

When the frequency of community j and the first cell i is identical, X (j-i)=1；When the frequency of community j and the first cell i differs, X (j-i)=0.

Described risk factor is calculated by below equation:

$D_i=\underset{j=1}{\overset{N}{\mathrm{\&Sigma;}}}E(j-i),$

Wherein, Di is the risk factor of cell i；

J and i is community sequence number, i ∈ [1, N], j ∈ [1, N], i ≠ j；N is the quantity of all communities in network to be planned；

E (j-i) is the community j correlation intensity to the first cell i, represents the community j general impacts degree to the first cell i.

The correlation intensity E (j-i) of the first cell i is calculated by described community j by below equation:

Wherein, f (j-i) is at point (x, the signal of community, the position j y) correlation intensity to the signal of cell i；(x y) is positioned at the prime coverage area territory of cell i to point；

As Pi-Pj < lim, f (j-i)=1；As Pi-Pj >=lim, f (j-i)=0；Pj is that community j is at point (x, signal intensity y)；Pi is that cell i is at point (x, signal intensity y)；Lim is interference threshold.

Optionally, the correlation intensity E (j-i) of cell i is calculated by described community j by below equation,

Wherein, R is the quantity of the grid divided according to predefined size of the community j in described network to be planned；R is the sequence number of the grid of community j；

Q is the quantity of the grid that the cell i in described network to be planned divides according to predefined size；Q is the sequence number of the grid of cell i；

The position that f (q-r) is grid q, the signal of the grid r correlation intensity to the signal of grid q；

During Pq-Pr < lim, f (q-r)=1；During Pq-Pr >=lim, f (q-r)=0；Pr is the community j signal intensity at grid r place；Pq is the cell i signal intensity at grid r place；Lim is interference threshold.

Described grid r is the grid of the signal intensity ratio cell i little at least lim of the minimum signal strength in its prime coverage area territory.

As in figure 2 it is shown, be another embodiment of the distribution method of subdistrict frequency point of the present invention, including:

Step 21, obtains all communities in network to be planned；

Step 22, selects the first community that risk factor is the highest from described all communities, and the risk factor of described first community is the sum to the correlation intensity of described first community of other communities except described first community in described network to be planned；

Step 23, to described first community distribution one first frequency；

Step 24, from described all communities in the residue community except described first community, searches the second community that risk factor is the highest；

Step 25, the first frequency according to the signal intensity of the first community and the first community, calculate the degree of disturbance of described network to be planned when described second cell configuration is each frequency respectively；

Step 26, from each frequency described, the second frequency of described second community corresponding when selecting the degree of disturbance of described network to be planned minimum, as the frequency of described second community, described second frequency is different from described first frequency；

Step 27, according to the method distributing the second frequency to described second community, the community given in described residue community except described second community successively distribution frequency.This step 27 particularly as follows:

Step 271, from described all communities in the residue community except described first community, the second community, searches the 3rd community that risk factor is the highest；

Step 272, the second frequency according to the first frequency of the signal intensity of the first community and the second community and the first community and the second community, calculate the degree of disturbance of described network to be planned when described 3rd cell configuration is each frequency respectively；

Step 273, from each frequency described, the 3rd frequency of described 3rd community corresponding when selecting the degree of disturbance of described network to be planned minimum, as the frequency of described 3rd community, described 3rd frequency is different from described first frequency and described second frequency.

According to the method described above, distribute frequency to each community, generate the preliminary allocative decision of subdistrict frequency point.Hereinafter the tentative programme of subdistrict frequency point is optimized.

As it is shown on figure 3, the distribution method of described subdistrict frequency point, also include:

Step 28, searches the first community that risk factor is the highest from described all communities；

Step 29, according to the frequency of other communities except described first community in described community and signal intensity, calculates the degree of disturbance of described network to be planned when described first cell configuration is each frequency respectively；In described community, the frequency of other communities except described first community can obtain from above-mentioned preliminary allocative decision.

Step 210, from each frequency described, the 4th frequency of described first community corresponding when selecting the degree of disturbance of described network to be planned minimum；

Step 211, uses described 4th frequency to update described first frequency；

Step 212, from described all communities in the residue community except described first community, searches the second community that risk factor is the highest；

Step 213, according to the frequency of other communities except described first community, the second community in described community and signal intensity, calculates the degree of disturbance of described network to be planned when described second community is each frequency described respectively；In described community, the frequency of other communities except described first community can obtain from above-mentioned preliminary allocative decision, and the frequency of described first community is the 4th above-mentioned frequency.

Step 214, from each frequency described, the 5th frequency of described second community corresponding when selecting the degree of disturbance of described network to be planned minimum, described 5th frequency is not equal to described 4th frequency；

Step 215, uses described 5th frequency to update second frequency (the former frequency of the second community) of described second community, as the new frequency of the second community.

Step 216, according to the method to the second frequency described in described second cell update, gives described residue community cell update frequency except described second community successively.Step 216 includes:

Step 2161, from described all communities in the residue community except described first community, the second community, searches the 3rd community that risk factor is the highest；

Step 2162, according to the frequency of other communities except described first community, the second community, the 3rd community in described community and signal intensity, calculates the degree of disturbance of described network to be planned when described 3rd community is each frequency described respectively；In described community, the frequency of other communities except described first community, the second community can obtain from above-mentioned preliminary allocative decision, and the frequency of described first community is the 4th above-mentioned frequency, and the frequency of described second community is the 5th above-mentioned frequency,

Step 2162, from each frequency described, the 6th frequency of described 3rd community corresponding when selecting the degree of disturbance of described network to be planned minimum, described 6th frequency is not equal to described 4th frequency, the 5th frequency；

Step 2163, uses described 6th frequency to update the 3rd frequency (the former frequency of the 3rd community) of described 3rd community, as the new frequency of the 3rd community.

According to the method described above, update the frequency of each community, generate the renewal allocative decision of subdistrict frequency point.

As it is shown on figure 3, be the distributor of a kind of subdistrict frequency point of the present invention, including:

First selects unit 31, selects the first community from the community of network to be planned；

Network interferences degree computing unit 32, according to the frequency of other communities distributing frequency except described first community in described community and signal intensity, calculates the degree of disturbance of described network to be planned when described first cell configuration is each frequency respectively；

Second selects unit 33, and from each frequency described, the first frequency of described first community corresponding when selecting the degree of disturbance of described network to be planned minimum, as the frequency of described first community.

Described first community be the unallocated frequency of network to be planned community in the highest community of risk factor, the risk factor of described first community is the sum to the correlation intensity of described first community of other communities except described first community in described network to be planned, and the frequency of described first frequency and described other communities having distributed frequency differs.

Described first community be described network to be planned community in the highest community of risk factor；The risk factor of described first community is the sum to the correlation intensity of described first community of other communities except described first community in described network to be planned；

Described device also includes: updating block 34, when described first community is assigned with frequency, described first frequency is used to update the former frequency of described first community, frequency as other communities being updated frequency in the new frequency of described first community, described first frequency and described network to be planned differs.

Described network interferences degree computing unit 32 calculates the degree of disturbance of described network to be planned according to below equation:

$\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}\underset{j=1}{\overset{N}{\mathrm{\&Sigma;}}}E(j-i)*X(j-i),$

Wherein, i and j is community sequence number, i ∈ [1, N], j ∈ [1, N], i ≠ j；N is the number of cells in described network to be planned；

E (j-i) is the community j correlation intensity to the first cell i；

When the frequency of community j and the first cell i is identical, X (j-i)=1；When the frequency of community j and the first cell i differs, X (j-i)=0.

The distributor of described subdistrict frequency point, also includes:

Community risk factor computing unit 35, for being calculated the risk factor of described first community by below equation:

$D_i=\underset{j=1}{\overset{N}{\mathrm{\&Sigma;}}}E(j-i),$

Wherein, Di is the risk factor of the first cell i；E (j-i) is the community j correlation intensity to the first cell i, represents the community j general impacts degree to the first cell i；

J and i is community sequence number, i ∈ [1, N], j ∈ [1, N], i ≠ j；N is the quantity of all communities in network to be planned.

The distributor of described subdistrict frequency point, also includes:

Correlation intensity computing unit 36, for calculating the described community j correlation intensity E (j-i) to the first cell i by below equation:

Wherein, f (j-i) is at point (x, the signal of community, the position j y) correlation intensity to the signal of the first cell i；(x y) is positioned at the prime coverage area territory of the first cell i to point；As Pi-Pj < lim, f (j-i)=1；As Pi-Pj >=lim, f (j-i)=0；Pj is that community j is at point (x, signal intensity y)；Pi is that the first cell i is at point (x, signal intensity y)；Lim is interference threshold；

Or, for calculating the described community j correlation intensity E (j-i) to the first cell i by below equation:

$E(j-i)=\underset{r=1}{\overset{R}{\mathrm{\&Sigma;}}}\underset{q=1}{\overset{Q}{\mathrm{\&Sigma;}}}f(q-r);$ Wherein, R is the quantity of the grid divided according to predefined size of the community j in described network to be planned；R is the sequence number of the grid of community j；Q is the quantity of the grid that the first cell i in described network to be planned divides according to predefined size；Q is the sequence number of the grid of the first cell i；The position that f (q-r) is grid q, the signal of the grid r correlation intensity to the signal of grid q；During Pq-Pr < lim, f (q-r)=1；During Pq-Pr >=lim, f (q-r)=0；Pr is the community j signal intensity at grid r place；Pq is first cell i signal intensity at grid r place；Lim is interference threshold value.

The application scenarios of the distribution method of of the present invention subdistrict frequency point is described below.

TD Wireless network frequency planing method of the present invention, mainly includes the content of three aspects: TD network planning characteristic data model (i.e. TD network interferences model), model objective function and the automatic planning algorithm of dominant frequency point.The whole net interference size situation of model objective function reflection TD network to be planned.TD network planning characteristic data model (i.e. TD network interferences model) measures " influence each other ability " between network internal base station cell to be planned.The present invention stands in the height that Automatic Frequency Planning method is overall, and model objective function " mates " as much as possible with TD network planning characteristic data model (i.e. TD network interferences model), decreases unnecessary computing, improves the overall efficiency of method.

(1) TD network planning characteristic data model (TD network interferences model)

All communities in planning region are numbered, if number of cells is N in frequency planning region, then have cell number value ∈ [1, N].There is cell i (i ∈ [1, N]) and community j (j ∈ [1, N]), cell i and community j are neighbor cells, solve all of E (j-i), it is possible to characterize the community correlation intensity model in frequency planning region with this matrix.E (j-i) is community correlation intensity, represents the community j general impacts degree to cell i.

The prime coverage area territory of cell i is Si, for certain point in the Si of region, the influence degree of cell i signal, at the signal intensity respectively Pi (dBm) of this point and Pj (dBm), is e (j-i) at this community j signal by cell i and community j.

For P-CCPCH, the C/I thresholding of TD terminal receiver is 0.1dB, assume that interference margins is mar, in other words, certain point in cell i prime coverage area territory, if cell i and community j adopt same frequency, once Pi-Pj is less than (0.1+mar) dB, the signal of cell i will be interfered by the signal of community j.If interference threshold is lim, and lim=(0.1+mar) dB.

Order: e (j-i)=f (j-i)

Wherein:

F (j-i)=1, as Pi-Pj < lim；

F (j-i)=0, as Pi-Pj >=lim；

Then the correlation intensity of cell i can be expressed as by community j:

It is to say, owing to the impact of community A is act on whole region Si by community B, thus community j to the general impacts degree E (j-i) of cell i equal to the e (j-i) integration on the Si of region.E (j-i) reflects in the prime coverage area territory Si of cell i, the size in the region of Pi-Pj < lim (dB).

In actual calculating process, it is possible to simplify the calculating of community correlation intensity:

Frequency planning region is being given gridding, after adopting discrete point to be calculated, owing to each grid area is the same, it is possible to characterize E (j-i) with meshes number rather than area.It is to say, in order to reduce amount of calculation, simultaneously take account of computer and carry out the mechanism of Practical Calculation, it is possible to this continuity point of integration is calculated process and is converted into discrete point calculating process.Whole frequency planning region being given gridding, namely divides the area into continuous print square net one by one, the set comprehensive of square length of side d considers the precision of amount of calculation size and result of calculation, usually, it is possible to take d=10 to 50 meter.

Theoretically, along with distance increases, cell signal strength only can decay to infinitely small, but will not disappear.But in the Practical Calculation community j process to cell i correlation intensity, if in the prime coverage area territory Si of cell i, the signal intensity ratio cell i of community j minimum signal in its prime coverage area territory also wants the grid of little lim (dB), just need not participate in calculating.That is, it is necessary to the difference of the signal intensity of the grid r of consideration and first cell i minimum signal strength in its prime coverage area territory is more than or equal to described interference threshold value.

(2) model objective function

First definition symbol is as shown in the table:

Model objective function symbol definition

Model objective function is set as follows:

$\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}\underset{j=1}{\overset{N}{\mathrm{\&Sigma;}}}E(j-i)*X(j-i),$

TD network interferences model can measure the ability that influences each other between network internal base station cell to be planned, model objective function is designed as interference model data here that obtained be multiplied with X (j-i) and sue for peace, the interference level of whole network to be planned can not only be reflected intuitively, namely target function value is more big, disturbed condition is more serious, vice versa, and make use of data with existing as much as possible, decreases amount of calculation.

(3) the automatic planning algorithm of dominant frequency point (model solution algorithm)

In the present invention, it is possible to adopt algorithm formation algorithm combination two kinds different to solve the dominant frequency point planning problem of TD network.First by greedy algorithm to be quickly generated an original frequency allocative decision；Then local search approach is adopted to finely tune above obtaining a result, to obtain final scheme.

Owing to greedy algorithm and local search algorithm belong to heuritic approach, it is necessary to utilize certain Experience norms, according to community correlation intensity model matrix, a concept will used in algorithm below to be introduced: community risk factor.

Cell i risk factor is: in network to be planned, other all communities correlation intensity sum to cell i.

If the risk factor of cell i is Di, then have:

$D_i=\underset{j=1}{\overset{N}{\mathrm{\&Sigma;}}}E(j-i);$ Wherein, i ∈ [1, N], j ∈ [1, N], i ≠ j.

One community of the risk factor attribute characterization of community size to result superiority-inferiority power of influence when being assigned with frequency.One community risk factor is more big, it is possible to the community that interference occurs with it is more many, and its frequency allocation plan is more big on the good and bad impact of result.Otherwise it is then contrary.In other words, in order to obtain the optimal solution of problem, it is first determined the frequency distribution of high-risk community so that it is contribute to the optimization of result as far as possible, the frequency distribution of relatively low risk factor community is then determined on this basis again.

Automatic Frequency Planning method of the present invention mainly includes the content of three aspects: TD network planning characteristic data model (i.e. TD network interferences model, referred to as interference model), model objective function and the automatic planning algorithm of dominant frequency point (i.e. model solution algorithm).

The present invention proposes community correlation intensity model, a kind of new wireless network interference model method for building up is proposed, solve in prior art can only qualitative cannot quantitative analysis or cannot comprehensively reflect the defect of influence factor's (such as, it is impossible to reflect the impact of the geographic factors such as topography and geomorphology objectively)." the community correlation intensity model " of the present invention reflects the potential impact ability of minizone, it is possible to achieve the quantitative analysis under composite factor, modeling.Base station relevant parameter in network to be planned and topography and geomorphology have been carried out quantitative sign for the impact of frequency planning, can reflecting real network situation, the rational for following model object function and the automatic planning algorithm of dominant frequency point is laid a good foundation.

It addition, the present invention proposes model objective function, it is the concrete application form of wireless network interference model, it is possible to take into account whole structure and the efficiency of Automatic Frequency Planning method.Utilizing " community correlation intensity model " to build model objective function, can not only reflect the interference level of whole network to be planned intuitively, namely target function value is more big, disturbed condition is more serious, vice versa, and make use of data with existing as much as possible, decreases amount of calculation.

It addition, the present invention proposes model solution algorithm (i.e. the combinational algorithm of greedy algorithm and local search algorithm), maximize and use a priori criteria, efficiently controlled amount of calculation.The combinational algorithm adopting greedy algorithm and local search algorithm completes model solution, it is generally adopted single algorithm with existing scheme to compare, feature for algorithms of different itself is combined utilization, and enhance a priori criteria directive significance for algorithm during the course, improve the computational efficiency of algorithm.Adopt, additionally, overcome, the uncertainty causing operation effect and operation time during intelligent algorithm, enhance the practicality of algorithm.Overcome the following defect of the model solution algorithm of employing at present: utilize single algorithm itself to look for relative optimal solution, lack the guidance of Wireless network frequency planning technology priori, lacking direction property of algorithm, cause that efficiency of algorithm is low.

One of ordinary skill in the art will appreciate that, realize all or part of step in above-described embodiment method to can be by the hardware that program carrys out instruction relevant and complete, described program can be stored in a computer read/write memory medium, this program is upon execution, step including such as above-mentioned embodiment of the method, described storage medium, as: magnetic disc, CD, read-only store-memory body (Read-OnlyMemory, or random store-memory body (RandomAccessMemory, RAM) etc. ROM).

In each embodiment of the method for the present invention; the sequence number of described each step can not be used for limiting the sequencing of each step; for those of ordinary skill in the art, under the premise not paying creative work, the priority of each step is changed also within protection scope of the present invention.

The above is the preferred embodiment of the present invention; it should be pointed out that, for those skilled in the art, under the premise without departing from principle of the present invention; can also making some improvements and modifications, these improvements and modifications also should be regarded as protection scope of the present invention.

Claims (5)

1. the distribution method of a subdistrict frequency point, it is characterised in that including:

The first community is selected from the community of network to be planned；

According to the frequency of other communities distributing frequency except described first community in described community and signal intensity, calculate the degree of disturbance of described network to be planned when described first cell configuration is each frequency respectively；

From each frequency described, the first frequency of described first community corresponding when selecting the degree of disturbance of described network to be planned minimum, as the frequency of described first community；

Described first community be the unallocated frequency of described network to be planned community in the highest community of risk factor, the risk factor of described first community represents that described first community and other each communities except described first community in described network to be planned occur the frequency of the probability of interference, described first frequency and described other communities having distributed frequency to differ；Or

Described first community be described network to be planned community in the highest community of risk factor；The risk factor of described first community represents that described first community, with other each communities except described first community in described network to be planned, the probability of interference occurs；

Described from each frequency described, the first frequency of described first community corresponding when selecting the degree of disturbance of described network to be planned minimum, the step as the frequency of described first community includes:

When described first community is assigned with frequency, using described first frequency to update the former frequency of described first community, the frequency as other communities being updated frequency in the new frequency of described first community, described first frequency and described network to be planned differs；

The risk factor of described first community is calculated by below equation:

$D_i=\underset{j=1}{\overset{N}{\&Sigma;}}E(j-i),$

Wherein, Di is the risk factor of the first cell i；E (j-i) is the community j correlation intensity to the first cell i, represents the community j general impacts degree to the first cell i；

J and i is community sequence number, i ∈ [1, N], j ∈ [1, N], i ≠ j；N is the quantity of all communities in network to be planned；

The degree of disturbance of described network to be planned calculates according to below equation:

$\underset{i=1}{\overset{N}{\&Sigma;}}\underset{j=1}{\overset{N}{\&Sigma;}}E(j-i)*X(j-i),$

Wherein, i and j is community sequence number, i ∈ [1, N], j ∈ [1, N], i ≠ j；N is the number of cells in described network to be planned；

E (j-i) is the community j correlation intensity to the first cell i；

When the frequency of community j and the first cell i is identical, X (j-i)=1；When the frequency of community j and the first cell i differs, X (j-i)=0.

2. the distribution method of subdistrict frequency point according to claim 1, it is characterised in that

The correlation intensity E (j-i) of the first cell i is calculated by described community j by below equation:

E (j-i)=∮ f (j-i) dxdy,

Wherein, f (j-i) is at point (x, the signal of community, the position j y) correlation intensity to the signal of the first cell i；(x y) is positioned at the prime coverage area territory of the first cell i to point；When Pi-Pj < during lim, f (j-i)=1；As Pi-Pj >=lim time, f (j-i)=0；Pj is that community j is at point (x, signal intensity y)；Pi is that the first cell i is at point (x, signal intensity y)；Lim is interference threshold；

Or, the correlation intensity E (j-i) of the first cell i is calculated by described community j by below equation:

$E(j-i)=\underset{r=1}{\overset{R}{\&Sigma;}}\underset{q=1}{\overset{Q}{\&Sigma;}}f(q-r);$

Wherein, R is the quantity of the grid divided according to predefined size of the community j in described network to be planned；R is the sequence number of the grid of community j；Q is the quantity of the grid that the first cell i in described network to be planned divides according to predefined size；Q is the sequence number of the grid of the first cell i；The position that f (q-r) is grid q, the signal of the grid r correlation intensity to the signal of grid q；Pq-Pr < during lim, f (q-r)=1；During Pq-Pr >=lim, f (q-r)=0；Pr is the community j signal intensity at grid r place；Pq is first cell i signal intensity at grid r place；Lim is interference threshold value.

3. the distribution method of subdistrict frequency point according to claim 2, it is characterised in that the difference of the signal intensity of described grid r and first cell i minimum signal strength in its prime coverage area territory is more than or equal to described interference threshold value.

4. the distributor of a subdistrict frequency point, it is characterised in that including:

First selects unit, selects the first community from the community of network to be planned；

Network interferences degree computing unit, according to the frequency of other communities distributing frequency except described first community in described community and signal intensity, calculates the degree of disturbance of described network to be planned when described first cell configuration is each frequency respectively；

Second selects unit, and from each frequency described, the first frequency of described first community corresponding when selecting the degree of disturbance of described network to be planned minimum, as the frequency of described first community；

Described first community be the unallocated frequency of described network to be planned community in the highest community of risk factor, the risk factor of described first community represents that described first community and other each communities except described first community in described network to be planned occur the frequency of the probability of interference, described first frequency and described other communities having distributed frequency to differ；Or

Described first community be described network to be planned community in the highest community of risk factor；The risk factor of described first community represents that described first community, with other each communities except described first community in described network to be planned, the probability of interference occurs；

Described device also includes:

Updating block, when described first community is assigned with frequency, using described first frequency to update the former frequency of described first community, the frequency as other communities being updated frequency in the new frequency of described first community, described first frequency and described network to be planned differs；

Described device also includes:

Community risk factor computing unit, for being calculated the risk factor of described first community by below equation:

$D_i=\underset{j=1}{\overset{N}{\&Sigma;}}E(j-i),$

Wherein, Di is the risk factor of the first cell i；E (j-i) is the community j correlation intensity to the first cell i, represents the community j general impacts degree to the first cell i；

J and i is community sequence number, i ∈ [1, N], j ∈ [1, N], i ≠ j；N is the quantity of all communities in network to be planned；

Described network interferences degree computing unit calculates the degree of disturbance of described network to be planned according to below equation:

$\underset{i=1}{\overset{N}{\&Sigma;}}\underset{j=1}{\overset{N}{\&Sigma;}}E(j-i)*X(j-i),$

Wherein, i and j is community sequence number, i ∈ [1, N], j ∈ [1, N], i ≠ j；N is the number of cells in described network to be planned；

E (j-i) is the community j correlation intensity to the first cell i；

When the frequency of community j and the first cell i is identical, X (j-i)=1；When the frequency of community j and the first cell i differs, X (j-i)=0.

5. the distributor of subdistrict frequency point according to claim 4, it is characterised in that also include:

Correlation intensity computing unit, for calculating the described community j correlation intensity E (j-i) to the first cell i by below equation:

E (j-i)=∮ f (j-i) dxdy,

Wherein, f (j-i) is at point (x, the signal of community, the position j y) correlation intensity to the signal of the first cell i；(x y) is positioned at the prime coverage area territory of the first cell i to point；When Pi-Pj < during lim, f (j-i)=1；As Pi-Pj >=lim time, f (j-i)=0；Pj is that community j is at point (x, signal intensity y)；Pi is that the first cell i is at point (x, signal intensity y)；Lim is interference threshold；

Or, for calculating the described community j correlation intensity E (j-i) to the first cell i by below equation:

$E(j-i)=\underset{r=1}{\overset{R}{\&Sigma;}}\underset{q=1}{\overset{Q}{\&Sigma;}}f(q-r);$

Wherein, R is the quantity of the grid divided according to predefined size of the community j in described network to be planned；R is the sequence number of the grid of community j；Q is the quantity of the grid that the first cell i in described network to be planned divides according to predefined size；Q is the sequence number of the grid of the first cell i；The position that f (q-r) is grid q, the signal of the grid r correlation intensity to the signal of grid q；Pq-Pr < during lim, f (q-r)=1；During Pq-Pr >=lim, f (q-r)=0；Pr is the community j signal intensity at grid r place；Pq is first cell i signal intensity at grid r place；Lim is interference threshold value.