CN102546076A - Method and system for distributing scrambling codes of TD-SCDMA (Time Division-Synchronous Code Division Multiple Access) network - Google Patents

Abstract

The invention relates to the technical field of mobile communication, in particular to a method and system for distributing scrambling codes among cells of a TD-SCDMA (Time Division-Synchronous Code Division Multiple Access) network, for solving the problem of incapability of effectively reducing whole-network interference by scrambling code distribution, caused by only considering the difference of dependences among scrambling codes/ composite codes without considering the influence of an actual wireless signal propagation environment to the interference among cell pairs. The method comprises the steps of: computing a frame error rate among the cell pairs, computing a frame error rat matrix, distributing scrambling code groups and selecting scrambling codes. The device comprises a computing module of the frame error rate among the cell pairs, a frame error rate matrix computing module, a scrambling code group distributing module and a scrambling code selecting module. According to the invention, through considering an actual wireless propagation environment, an optimal scrambling code distributing scheme can be obtained to ensure that the whole-network total frame error rate is minimum; and the scrambling code groups are defined and scrambling code distributed are carried out by two steps of scrambling code group distribution and scrambling code selection by using properties of the scrambling code groups, thus the scrambling code distribution complexity is greatly reduced.

Description

A kind of scrambling code distribution method of TD-SCDMA network and distribution system thereof

Technical field

The present invention relates to the cellular mobile communication field, particularly the scrambling code distribution method of TD-SCDMA network and distribution system thereof.

Background technology

The TD-SCDMA network is a code division multiple access, and its capacity is interference-limited.Certainly the one of the main reasons of disturbing of TD-SCDMA network is that the scrambler code length is shorter.Because scrambler has only 16, scrambler between correlation poor, the compound key (by scrambler and walsh code is compound obtain) that is used for spread spectrum between correlation also poor, cause interference among adjacent cells strong.Therefore, scrambler is most important from interference, lifting service quality to reducing the TD-SCDMA network in the allocative decision of minizone.

128 scramblers of TD-SCDMA network belong to 32 code characters, and 4 scramblers of a code character and 1 descending synchronous code, 8 uplink synchronous codes and 4 Midamble sign indicating numbers are bound, and are as shown in table 1.Scrambler of cell allocation of standard code, the descending synchronous code of neighbor cell must be different, must belong to different code characters so distribute to the scrambler of neighbor cell.

The code character of table 1TD-SCDMA standard code

Existing some scrambling code distribution methods.One class methods are the suggestions according to the TD-SCDMA standard, after descending synchronous code is assigned in the sub-district, under this descending synchronous code, select one 4 of code character scramblers again.Another kind of distribution method is to be target with the whole network minimize interference, as constraints, or directly gives the cell allocation scrambler with the code character difference of neighbor cell, or gives cell allocation base scrambler (being equivalent to one group of scrambler); The foundation of distributing scrambler or scrambler group be scrambler to or the scrambler group to or compound key between the correlation size.For example, application number is to disclose in 200610057869.5 the one Chinese patent application to reduce the scrambling code distribution method that disturbs with frequently in a kind of CDMA mobile communication systems: choose signal interval [t in relative time delay ₁, t ₂]; Calculate [t ₁, t ₂] interior scrambler group maximum correlation; Calculate [t ₁, t ₂] interior compound key maximum correlation and probability distribution thereof, and obtain compound key maximum correlation average and variance; According to scrambler group maximum correlation and compound key maximum correlation average and variance, select the scrambler group to distribute.

Yet the intensity of presence of intercell interference is not only relevant with the scrambler that is assigned to, and relevant with the wireless propagation environment of minizone.Scrambler between correlation bigger, radio signal propagation path loss is more little, then presence of intercell interference is strong more.Therefore, only with the right correlation size of scrambler be not enough to reflect the sub-district between the true power of disturbing, only correlation is carried out scrambler and distributes and be difficult to really realize the whole network minimize interference according to scrambler.

Summary of the invention

The invention provides a kind of scrambling code distribution method and distribution system thereof of TD-SCDMA network, can't distribute scrambler to complicated wireless propagation environment to solve in the prior art, reduce the total problem of disturbing of the whole network.

A kind of scrambling code distribution method of TD-SCDMA network comprises the steps:

(1) in the dispensed zone sub-district of all sub-districts to a FER;

Belong to same scrambler family according to the scrambler of compound key set equality, 128 scramblers of TD-SCDMA network are divided into 12 scrambler families;

Sub-district based on the sub-district uses the arbitrary scrambler in the same scrambler family to obtain is identical to a FER, traversal calculate all sub-districts to use different scrambling codes family to the time FER;

All location point places on the overlay area, main plot are disturbed by neighboring area signal the FER cause and as the sub-district of this main plot to a FER;

Described location point place by neighboring area signal disturb the FER cause be this main plot, location point place and adjacent area 16 * 16 totally 256 code channels to the mean value of last FER;

(2) calculate the FER matrix;

Multiplied each other to a FER and cell weight in the sub-district of all sub-districts, the sub-district that obtains all sub-districts is to a weighting FER;

Described cell weight artificially is provided with according to the significance level of sub-district;

With all sub-districts a weighting FER is constructed a FER matrix;

(3) according to the FER matrix, to all cell allocation scrambler families in the range of distribution;

Scrambler family is assigned as twice planning problem of 0-1, according to the FER matrix that obtains in the step (2), finds the solution decision variable, obtains scrambler family allocative decision;

(4) according to the scrambler family allocative decision that obtains, scrambler is selected in all sub-districts in the range of distribution;

The scrambler of assigning to based on neighbor cell belongs to different code characters, from the scrambler family that the sub-district is assigned to, selects scrambler.

Scrambling code distribution method of the present invention, further, the FER that all location point places on the overlay area, main plot are disturbed by neighboring area signal described in the step (1) causes and as the sub-district of this main plot a FER is comprised step:

Based on the Back ground Information of range of distribution, uniform grid is carried out in the range of distribution divide;

Based on the Back ground Information of range of distribution, with the carrier/interface ratio and the relative time delay at electronic chart and all grid lattice point places of radio signal propagation path loss theoretical model calculating;

According to the carrier/interface ratio at all grid lattice point places with calculate the FER at all grid lattice point places relative time delay;

With the FER at grid lattice point place and the FER sum at interior all the location point places of approximate this grid of corresponding grid product of areas, as the FER of this grid;

FER sum with all location point places in approximate this sub-district of FER sum of all grids in the sub-district.

Described Back ground Information comprises that size, each base station location and the transmitting power of range of distribution area coverage and radio signal propagation path loss theoretical model calculate required out of Memory.

Scrambling code distribution method of the present invention further, in step (1) before, also comprises step: the measured data of gathering existing network; Described measured data comprise in the range of distribution main plot signal and its all neighboring area signals on some places the carrier/interface ratio measured value and relative time delay measured value; Described measured data does not require it is geographical equally distributed;

The FER that all location point places on the overlay area, main plot are disturbed by neighboring area signal described in the described step (1) causes and as the sub-district of this main plot a FER is comprised step:

Calculate the scope in relative time delay: the lower bound of scope was rounded up by the minimum value in relative time delay in the measured data and obtained relative time delay; The upper bound is rounded up by the maximum in relative time delay in the measured data that is not more than 12dB corresponding to carrier/interface ratio and obtains, again according to scope and interval width calculating in relative time delay interval number in relative time delay;

FER by interval quantization, is obtained the FER quantized value; Again according to the quantized interval boundary value of FER, traversal calculate all scrambler families in the relative time delay scope all relative time delays interval intermediate value places carrier/interface ratio interval border value;

Interval and relative time delay, the interval was added up measured data according to carrier/interface ratio, obtained the joint probability distribution function in carrier/interface ratio and relative time delay;

According to the joint probability distribution function and the FER quantized value in carrier/interface ratio and relative time delay, the traversal calculation plot to use different scrambling codes family to the time FER.

A kind of scrambler distribution system of TD-SCDMA network comprises:

The sub-district is to a FER computing module:

The Back ground Information and the grid lattice point interval that connect the input range of distribution; The Back ground Information of range of distribution comprises that size, each base station location and the transmitting power of regional area coverage and radio signal propagation path loss theoretical model calculate required out of Memory; Uniform grid is carried out in the range of distribution divides, according to base station transmitting power and propagation path loss theoretical model, obtain all grid lattice point place main plot signals and all other cell signals the carrier/interface ratio value and relative time delay value; Based on this carrier/interface ratio value and relative time delay value, traversal calculate all sub-districts to use different scrambling codes family to the time the sub-district to a FER as the sub-district to a FER set, export this sub-district to a FER set;

FER matrix computations module:

Receive described sub-district to a FER set; According to the significance level of sub-district the cell weight of all sub-districts is set, the sub-district of each sub-district cell weight to a FER and this sub-district is multiplied each other, the sub-district that obtains all sub-districts is to a weighting FER; Sub-district with described all sub-districts constitutes the FER matrix to a weighting FER, exports this FER matrix;

Scrambler family distribution module:

Receive described FER matrix; Distribute 0-1 quadratic programming model with FER matrix construction scrambler family, and find the solution the decision variable of optimum scrambler family allocative decision; Export the decision variable of this optimum scrambler family allocative decision;

Scrambler is selected module:

Receive the decision variable of optimum scrambler family allocative decision; The scrambler of assigning to based on neighbor cell belongs to different code characters, from the scrambler family that the sub-district is assigned to, selects scrambler as the scrambler allocation result, exports this scrambler allocation result.

Scrambler distribution system of the present invention, further, one group of submodule replacement can also be used to a FER computing module in described sub-district:

Relative time delay interval number calculating sub module:

The measured data that input is gathered is calculated the scope in relative time delay; According to range computation in this interval number in relative time delay wherein, all section boundaries value in relative time delay in relative time delay; Export section boundaries value in these all relative time delays;

FER quantizes and carrier/interface ratio interval division submodule:

Receive section boundaries value in described all in relative time delay; FER is pressed interval quantization; Again according to the intermediate value in relative time delay in the boundary value of each FER quantized interval and each interval, calculate a pair of scrambler family in the boundary value of this FER quantized interval carrier/interface ratio value with intermediate value place in this in relative time delay in relative time delay; With this carrier/interface ratio value as carrier/interface ratio section boundaries value; It is right to travel through all scrambler families, obtain all scrambler families in the boundary value of all FER quantized intervals, relative time delay interval relative time delay the intermediate value place carrier/interface ratio section boundaries value; Export all carrier/interface ratio section boundaries values; ,

The joint probability distribution function calculating sub module in carrier/interface ratio and relative time delay:

Receive described all in relative time delay the section boundaries value with all carrier/interface ratio section boundaries values; Section boundaries value in all according to each scrambler family is right is added up measured data with all carrier/interface ratio section boundaries values in relative time delay, obtains the right carrier/interface ratio of this scrambler family and the joint probability distribution function in relative time delay; It is right to travel through all scrambler families, obtains the right carrier/interface ratio of all scrambler families and the joint probability distribution function in relative time delay; Export the right carrier/interface ratio of these all scrambler families and the joint probability distribution function in relative time delay;

The sub-district is to a FER calculating sub module:

Receive the right carrier/interface ratio of described all scrambler families and the joint probability distribution function in relative time delay; According to each sub-district to use different scrambling codes family to the time carrier/interface ratio and the joint probability distribution function in relative time delay and FER quantized value, calculate this sub-district to use different scrambling codes family to the time the sub-district to a FER; Traversal calculate all sub-districts to use different scrambling codes family to the time the sub-district to a FER as the sub-district to a FER set, export this sub-district to a FER set.

Beneficial effect of the present invention is:

One, scrambler has been considered the actual wireless communication environments in distributing; Mainly refer to signal relative time delay and these two parameters of signal carrier/interface ratio; Can accurately reflect actual network interferences intensity; And through scrambler family distribute make scrambler to correlation distribute and wireless propagation environment is complementary, effectively reduce the total FER of the whole network;

Two, scrambler is distributed be reduced to that scrambler family distributes and scrambler selected for two steps, greatly reduce computation complexity;

Three, be applicable to that the scrambler in the optimization of the TD-SCDMA network planning and existing network distributes.

Description of drawings

Fig. 1 is the schematic flow sheet of the scrambling code distribution method of the embodiment of the invention 1;

Fig. 2 is the schematic flow sheet of the scrambling code distribution method of the embodiment of the invention 2;

Fig. 3 is the signal carrier/interface ratio interval border value calculation process sketch map of the embodiment of the invention 2;

Fig. 4 is the structural representation of the scrambler distribution system of the embodiment of the invention 3;

Fig. 5 is the structural representation of the scrambler distribution system of the embodiment of the invention 4;

The simulating scenes sketch map that Fig. 6 distributes for scrambler of the present invention.

Embodiment

The inventive method is distributed scrambler abstract in following Mathematical Modeling, promptly seeks to make the minimum scrambler allocative decision of total FER of the whole network,

$\underset{\{s_p,s_q\}}{\min }I$

s.t.s _p，s _q∈S

$I=\underset{p=1}{\overset{M}{\mathrm{\Σ}}}{\mathrm{\β}}_p{I}_p---\left(1\right)$

$I_p=\underset{q=1,q\≠p}{\overset{M}{\mathrm{\Σ}}}{I}_{p,q}^{s_p,s_q}$

In the formula (1), I is total FER of the whole network; s _pAnd s _qBe respectively p sub-district and q the scrambler that use the sub-district; S is the scrambler set of TD-SCDMA network; M is the number of sub-district in the cellular network; β _pBe the weight that scrambler divides p sub-district of timing, represent the significance level of this sub-district, can artificially stipulate, the important more then weight in sub-district is big more; I _pIt is the FER of p the sub-district of causing of the interference by all other sub-districts;

The FER of p the sub-district that causes for the interference by q sub-district is the overlay area of p sub-district

In FER sum on each location point;

$I_{p,q}^{s_p,s_q}=={\mathrm{\Ω}}_p\underset{(\mathrm{\η},\mathrm{\τ})\∈{\mathrm{\Θ}}_{p,q}}{\∫\∫}{f}_{p,q}(\mathrm{\η},\mathrm{\τ})G_{s_p,s_q}(\mathrm{\η},\mathrm{\τ})\mathrm{d\ηd\τ}---\left(2b\right)$

In the formula (2a);

is place (x in p the sub-district; Y) the signal carrier/interface ratio of last this sub-district and q sub-district is relevant with the transmitting power of the radio signal propagation condition of two minizones, two base stations;

is place (x; Y) the signal relative time delay of last this sub-district and q sub-district is relevant with the radio signal propagation condition of two minizones;

is place (x, y) the last FER that is caused by the interference of q sub-district.

(2a) does Jacobi's transformation to formula; Make integration variable by position x and y be transformed to carrier/interface ratio η and relative time delay τ; P cell signal is to the function of the carrier/interface ratio η of q cell signal, relative time delay τ, shown in (2b) in then

is expressed as

.Here, Ω _pFor

Area; Θ _{P, q}It is function

The domain of definition (span of η and τ); f _{P, q}(η τ) is the joint probability density function of η and τ, characterizes the signal interference relationships of two minizones, and is relevant with the transmitting power of the radio signal propagation condition of two minizones, two base stations.

FER

is defined as all code channels (j=1 of q sub-district in the formula (2b); 2 ..., N) to all code channels (i=1 of p sub-district; 2; ..., the average FER that N) causes

Here, N is the number of a scrambler corresponding compound sign indicating number, N=16;

is the FER of the code channel i (with the code channel of i compound key spread spectrum) that caused by code channel j (with the code channel of j compound key spread spectrum); Be the function of the frame length L and the error rate

,

error rate

be that j the compound key of i compound key and q sub-district of p sub-district is at the function of correlation at τ place in relative time delay.Basic modulation system QPSK with TD-SCDMA under the Gaussian channel is an example, ${\mathrm{\ξ}}_{s_p,s^q}^{i,j}=2Q\left(\sqrt{{2\mathrm{\ρ}}_{s_p,s_q}^{i,j}}\right)[1-\frac{1}{2}Q\left(\sqrt{{2\mathrm{\ρ}}_{s_p,s_q}^{i,j}}\right)],$ The code channel (signal code channel) of wherein using i compound key spread spectrum to the despreading of the code channel (interference code channel) of j compound key spread spectrum after signal to noise ratio Here J the compound key of i compound key and q sub-district that is p sub-district is at the correlation at τ place in relative time delay.

The FER of p the sub-district that causes by the interference of q sub-district in sum,

Receive the influence of two factors: (1) characterizes the joint probability density function f of two minizone geographical features (radio signal propagation condition) _{P, q}(η, τ), (2) characterize compound key to correlation properties

The inventive method has been utilized the compound key aggregating characteristic of scrambler: in the same scrambler family, and set of the compound key of a scrambler and the compound key set equality of another scrambler arbitrarily.12 altogether in the TD-SCDMA scrambler family of the present invention's definition, the scrambler number is not identical entirely in the family.Scrambler family is different with code character in the table 1, and two scramblers that belong to same scrambler family maybe be in different code characters.Table 2 has provided the ID of the scrambler in each scrambler family, the ID of the affiliated code character of scrambler.

The corresponding relation of code character under table 2 scrambler family, scrambler, the scrambler

With scrambler s _pAffiliated scrambler family is designated as T _p, scrambler s _qAffiliated scrambler family is designated as T _qBy the compound key aggregating characteristic of above-mentioned scrambler, right

s _q, s ' _q∈ T _q, necessarily have,

$G_{s_p^{\′},s_q}(\mathrm{\η},\mathrm{\τ})=G_{s_p,s_q^{\′}}(\mathrm{\η},\mathrm{\τ})=G_{s_p^{\′},s_q^{\′}}(\mathrm{\η},\mathrm{\τ})=G_{s_p,s_q}(\mathrm{\η},\mathrm{\τ})=G_{T_p,T_q}(\mathrm{\η},\mathrm{\τ})---\left(3\right)$

In the formula (3), the new symbol of introducing

Be to be η, relative time delay to be the τ place by using T in the signal carrier/interface ratio _qIn the interference of q sub-district of arbitrary scrambler to using T _pIn p the FER that cause the sub-district of arbitrary scrambler.

With the T of scrambler family _pIn any scrambler distribute to p sub-district, the T of scrambler family _qIn any scrambler distribute to q sub-district, the FER of p the sub-district that is caused by the interference of q sub-district is designated as Then have with reference to formula (2a) with (2b),

$I_{p,q}^{T_p,T_q}={\mathrm{\Ω}}_p\underset{(\mathrm{\η},\mathrm{\τ})\∈{\mathrm{\Θ}}_{p,q}}{\∫\∫}{f}_{p,q}(\mathrm{\η},\mathrm{\τ})G_{T_p,T_q}(\mathrm{\η},\mathrm{\τ})\mathrm{d\ηd\τ}---\left(4b\right)$

$I_{p,q}^{T_p,T_q}=I_{p,q}^{s_p,s_q}---\left(4c\right)$

Formula (4c) obtains according to formula (3).

Therefore, 128 scramblers of formula (1) Mathematical Modeling description can be realized through two steps in the optimum allocation of minizone: the first step is the optimum allocations of 12 scrambler families in the minizone; Second step was in the scrambler family that the sub-district is assigned to, to select a scrambler, guaranteed that the scrambler of neighbor cell belongs to different code characters.Visible by table 2, comprise 4 scramblers among the F of scrambler family, belong to 4 code characters; The scrambler number that other scrambler family comprises is more, and affiliated code character number is also more.Painted according to the map trichromatism, the second step total energy is accomplished.

The realization of this two-step, optimum allocation has guaranteed that total FER of the whole network is minimum in the minizone in the scrambler family of its first step, the scrambler in second step is selected total FER that can not change the whole network.

The realization of this two-step can reduce the complexity that scrambler distributes greatly.

Introduce embodiment of the present invention in detail below in conjunction with accompanying drawing.

Embodiment 1:

The scrambling code distribution method of TD-SCDMA network of the present invention is applied to a kind of scrambling code distribution method of the range of distribution in network planning stage, as shown in Figure 1, comprise step:

The sub-district of all sub-districts is to a FER in the S101 dispensed zone;

Belong to same scrambler family according to the scrambler of compound key set equality, 128 scramblers of TD-SCDMA network are divided into 12 scrambler families;

Sub-district based on the sub-district uses the arbitrary scrambler in the same scrambler family to obtain is identical to a FER, traversal calculate all sub-districts to use different scrambling codes family to the time FER;

All location point places on the overlay area, main plot are disturbed by neighboring area signal the FER cause and as the sub-district of this main plot to a FER;

Described location point place by neighboring area signal disturb the FER cause be this main plot, location point place and adjacent area 16 * 16 totally 256 code channels to the mean value of last FER;

Further, S101 specifically comprises:

Based on the Back ground Information of range of distribution, uniform grid is carried out in the range of distribution divide;

Based on the Back ground Information of range of distribution, with the carrier/interface ratio and the relative time delay at electronic chart and all grid lattice point places of radio signal propagation path loss theoretical model calculating;

According to the carrier/interface ratio at all grid lattice point places with calculate the FER at all grid lattice point places relative time delay;

With the FER at grid lattice point place and the FER sum at interior all the location point places of approximate this grid of corresponding grid product of areas, as the FER of this grid;

FER sum with all location point places in approximate this sub-district of FER sum of all grids in the sub-district.

Further, S101 comprises in the practical implementation process:

At interval uniform grid being carried out in the range of distribution according to the Back ground Information of the range of distribution in network planning stage and grid lattice point divides.The Back ground Information of the range of distribution in network planning stage comprises that size, each base station location and the transmitting power of regional overlay area area, propagation path loss theoretical model calculate required out of Memory.

In the present embodiment, can grid lattice point position be designated as (x _b, y _b), the lateral separation Δ x=x of lattice point _B+1-x _b, longitudinal separation Δ y=y _B+1-y _b, the grid area is Δ x Δ y.Calculate each lattice point (x _b, y _b) carrier/interface ratio located

Relative time delay

And FER

T _p, T _q=1 ..., 12.Substitution formula (4a); Calculation plot is to a FER; FER sum with all location point places in approximate this grid of lattice point place FER

and grid product of areas; FER sum with all location point places in all lattice point place FERs in the sub-district and approximate this sub-district of grid product of areas sum; Be that the sub-district is to a FER

p，q＝1，...，M (5)

T _p，T _q＝1，...，12

S102 calculates the FER matrix;

Multiplied each other to a FER and cell weight in the sub-district of all sub-districts, the sub-district that obtains all sub-districts is to a weighting FER;

Described cell weight artificially is provided with according to the significance level of sub-district;

With all sub-districts a weighting FER is constructed a FER matrix.

S102 specifically comprises in the present embodiment:

When representing that with FER matrix Γ different scrambling codes family is used in the sub-district the FER that might occur,

Its submatrix Γ _{P, q}Be p sub-district, a q sub-district use different scrambling codes family to the time, the FER submatrix of p the sub-district of causing by q sub-district.By Γ _{P, q}Physical significance can know submatrix Γ when p=q _{P, q}Be empty matrix.

Capable the 1st column element of k of FER matrix Γ

Represent p the sub-district use scrambler T of family _p, q sub-district uses the scrambler T of family _qThe time, the weighting FER of p the sub-district that cause q sub-district, k, l=1,2 ..., 12M.

γ _{K, l}Also be submatrix Γ _{P, q}In a _pRow a _qThe element of row.a _p=k-12 (p-1) is the T of scrambler family _pFamily's sequence number, a _p=1,2 ..., 12.a _q=l-12 (q-1) is the T of scrambler family _qFamily's sequence number, a _q=1,2 ..., 12.

S103 is according to the FER matrix, to all cell allocation scrambler families in the range of distribution;

Scrambler family is assigned as twice planning problem of 0-1, according to the FER matrix that obtains in the step (2), finds the solution decision variable, obtains scrambler family allocative decision.

S103 specifically comprises at present embodiment:

Total FER of the whole network can be expressed as I=x ^TΓ x.Wherein, decision vector x=[x _1,1x _1,2... x _1,12x _2,1x _2,2... x _2,12... x _{M, 1}x _{M, 2}... x _{M, 12}] ^TRepresent each cell scrambling family operating position, x _{P, a}A scrambler family used, x in p sub-district of=1 expression _{P, a}A scrambler family do not used in p sub-district of=0 expression, p=1, and 2 ..., M, a=1,2 ..., 12.Because a scrambler can and can only be used in each sub-district, therefore $\underset{a=1}{\overset{12}{\mathrm{\Σ}}}{x}_{p,q}=1,p=\mathrm{1,2},...,M.$

Formula (1) scrambler apportion model can be rewritten as a 0-1 quadratic programming model, that is,

$\underset{x}{\min }{x}^T\mathrm{\Γx}$

s.t.x＝[x _1，1x _1，2...x _1，12x _2，1x _2，2...x _2，12...x _M，1x _M，2...x _M，12] ^T

$\underset{a=1}{\overset{12}{\mathrm{\Σ}}}{x}_{p,a}=1,p=\mathrm{1,2},...,M---\left(8\right)$

x _{P, a}=0 or 1

${\mathrm{\γ}}_{k,l}={\mathrm{\β}}_p{I}_{p,q}^{T_p,T_q}$

Find the solution the 0-1 quadratic programming, obtain optimizing decision variable x ^Opt, promptly optimum scrambler family allocative decision.

S104 selects scrambler according to the scrambler family allocative decision that obtains to all sub-districts in the range of distribution;

The scrambler of assigning to based on neighbor cell belongs to different code characters, from the scrambler family that the sub-district is assigned to, selects scrambler.

A scrambler is selected in each sub-district from the scrambler family that has assigned to, the principle of selecting is that the scrambler of neighbor cell belongs to different code characters.For example, can adopt the chain type apportion design: optional sub-district begins, and from the scrambler family that this sub-district is assigned to, selects any scrambler; Select scrambler for its adjacent area, guarantee that the scrambler that two adjacent areas are picked out belongs to different code characters; By that analogy, all pick out scrambler until all sub-districts.

Embodiment 2:

The scrambling code distribution method of TD-SCDMA network of the present invention is applied to a kind of scrambling code distribution method of the range of distribution of existing network optimizing phase, as shown in Figure 2, comprise step:

S201 gathers the measured data of existing network; Described measured data comprise in the range of distribution main plot signal and its all neighboring area signals on some places the carrier/interface ratio measured value and relative time delay measured value; Described measured data does not require it is geographical equally distributed.

The present embodiment method is applicable to that the scrambler of existing network optimizing phase distributes.The input data of the range of distribution of existing network optimizing phase collect on road with equipment usually, and promptly data acquisition place/track is limited, are difficult on equally spaced place, gather.

Present embodiment by carrier/interface ratio data η and relative time delay τ come direct calculation plot to a FER, and be similar to the joint probability density function f of replacement η and τ with the joint probability distribution function of η and τ _{P, q}(η, τ).Thereby, need not know the positional information of data, need not carry out f yet _{P, q}(η, accurate Calculation τ) or estimation.

Be similar to the joint probability density function f that replaces η and τ in this zone for joint probability distribution function with η and τ _{P, q}(η, τ), with carrier/interface ratio η and relative time delay τ be divided into U interval and V interval respectively by numerical value.Correspondingly, carrier/interface ratio and relative time delay subdomain

U=1,2 ..., U, v=1,2 ..., V; Carrier/interface ratio and relative time delay subdomain Θ ^{U, v}With

The common factor of the domain of definition

Be called feasible subdomain; Carrier/interface ratio and relative time delay subdomain Θ ^{U, v}On

Quantized value do

Then, can further obtain by formula (4b) model,

$I_{p,q}^{T_p,T_q}={\mathrm{\Ω}}_p\underset{(\mathrm{\η},\mathrm{\τ})\∈{\mathrm{\Θ}}_{p,q}}{\∫\∫}{f}_{p,q}(\mathrm{\η},\mathrm{\τ})G_{T_p,T_q}(\mathrm{\η},\mathrm{\τ})\mathrm{d\ηd\τ}$

$={\mathrm{\&Omega;}}_p\underset{(\mathrm{\&eta;},\mathrm{\&tau;})\&Element;{\mathrm{\&Theta;}}_{p,\mathrm{<figure-callout\; id="1"\; label="q"\; filenames="HDA0000116203070000021.png"\; state="\{\{state\}\}">q</figure-callout>}}^{\mathrm{1,1}}}{\&Integral;\&Integral;}{f}_{p,q}(\mathrm{\&eta;},\mathrm{\&tau;})G_{T_p,T_q}(\mathrm{\&eta;},\mathrm{\&tau;})\mathrm{d\&eta;d\&tau;}+{\mathrm{\&Omega;}}_p\underset{(\mathrm{\&eta;},\mathrm{\&tau;})\&Element;{\mathrm{\&Theta;}}_{p,\mathrm{<figure-callout\; id="1"\; label="q"\; filenames="HDA0000116203070000021.png"\; state="\{\{state\}\}">q</figure-callout>}}^{\mathrm{1,2}}}{\&Integral;\&Integral;}{f}_{p,q}(\mathrm{\&eta;},\mathrm{\&tau;})G_{T_p,T_q}(\mathrm{\&eta;},\mathrm{\&tau;})\mathrm{d\&eta;d\&tau;}+...+$

${\mathrm{\&Omega;}}_p\underset{(\mathrm{\&eta;},\mathrm{\&tau;})\&Element;{\mathrm{\&Theta;}}_{p,\mathrm{<figure-callout\; id="1"\; label="q"\; filenames="HDA0000116203070000021.png"\; state="\{\{state\}\}">q</figure-callout>}}^{1,V}}{\&Integral;\&Integral;}{f}_{p,q}(\mathrm{\&eta;},\mathrm{\&tau;})G_{T_p,T_q}(\mathrm{\&eta;},\mathrm{\&tau;})\mathrm{d\&eta;d\&tau;}+...+{\mathrm{\&Omega;}}_p\underset{(\mathrm{\&eta;},\mathrm{\&tau;})\&Element;{\mathrm{\&Theta;}}_{p,q}^{U,V}}{\&Integral;\&Integral;}{f}_{p,q}(\mathrm{\&eta;},\mathrm{\&tau;})G_{T_p,T_q}(\mathrm{\&eta;},\mathrm{\&tau;})\mathrm{d\&eta;d\&tau;}$

$\&ap;{\mathrm{\&Omega;}}_p\underset{(\mathrm{\&eta;},\mathrm{\&tau;})\&Element;{\mathrm{\&Theta;}}_{p,\mathrm{<figure-callout\; id="1"\; label="q"\; filenames="HDA0000116203070000021.png"\; state="\{\{state\}\}">q</figure-callout>}}^{\mathrm{1,1}}}{\&Integral;\&Integral;}{f}_{p,q}(\mathrm{\&eta;},\mathrm{\&tau;})G_{T_p,{\mathrm{<figure-callout\; id="1"\; label="T\; q"\; filenames="HDA0000116203070000021.png"\; state="\{\{state\}\}">\; <figure-callout\; id="1"\; label="T\; q"\; filenames="HDA0000116203070000021.png"\; state="\{\{state\}\}">T</figure-callout>\; </figure-callout>}}_q}^{\mathrm{1,1}}\mathrm{d\&eta;d\&tau;}+{\mathrm{\&Omega;}}_p\underset{(\mathrm{\&eta;},\mathrm{\&tau;})\&Element;{\mathrm{\&Theta;}}_{p,\mathrm{<figure-callout\; id="1"\; label="q"\; filenames="HDA0000116203070000021.png"\; state="\{\{state\}\}">q</figure-callout>}}^{\mathrm{1,2}}}{\&Integral;\&Integral;}{f}_{p,q}(\mathrm{\&eta;},\mathrm{\&tau;})G_{T_p,{\mathrm{<figure-callout\; id="1"\; label="T\; q"\; filenames="HDA0000116203070000021.png"\; state="\{\{state\}\}">T</figure-callout>}}_q}^{\mathrm{1,2}}\mathrm{d\&eta;d\&tau;}+...+$

${\mathrm{\&Omega;}}_p\underset{(\mathrm{\&eta;},\mathrm{\&tau;})\&Element;{\mathrm{\&Theta;}}_{p,\mathrm{<figure-callout\; id="1"\; label="q"\; filenames="HDA0000116203070000021.png"\; state="\{\{state\}\}">q</figure-callout>}}^{1,V}}{\&Integral;\&Integral;}{f}_{p,q}(\mathrm{\&eta;},\mathrm{\&tau;})G_{T_p,{\mathrm{<figure-callout\; id="1"\; label="T\; q"\; filenames="HDA0000116203070000021.png"\; state="\{\{state\}\}">T</figure-callout>}}_q}^{1,V}\mathrm{d\&eta;d\&tau;}+...+{\mathrm{\&Omega;}}_p\underset{(\mathrm{\&eta;},\mathrm{\&tau;})\&Element;{\mathrm{\&Theta;}}_{p,q}^{U,V}}{\&Integral;\&Integral;}{f}_{p,q}(\mathrm{\&eta;},\mathrm{\&tau;})G_{T_p,T_q}^{U,V}\mathrm{d\&eta;d\&tau;}$

$={\mathrm{\&Omega;}}_p\underset{u=1}{\overset{U}{\mathrm{\&Sigma;}}}\underset{v=1}{\overset{V}{\mathrm{\&Sigma;}}}\left[G_{T_p,T_q}^{u,v}(\mathrm{\&eta;},\mathrm{\&tau;})\underset{(\mathrm{\&eta;},\mathrm{\&tau;})\&Element;{\mathrm{\&Theta;}}_{p,q}^{u,v}}{\&Integral;\&Integral;}{f}_{p,q}(\mathrm{\&eta;},\mathrm{\&tau;})\mathrm{d\&eta;d\&tau;}\right]$

$={\mathrm{\&Omega;}}_p\underset{u=1}{\overset{U}{\mathrm{\&Sigma;}}}\underset{v=1}{\overset{V}{\mathrm{\&Sigma;}}}[G_{T_p,T_q}^{u,v}(\mathrm{\&eta;},\mathrm{\&tau;})\&CenterDot;P((\mathrm{\&eta;},\mathrm{\&tau;})\&Element;{\mathrm{\&Theta;}}_{p,q}^{u,v})]---\left(9\right)$

In the formula (9); The joint probability distribution function of η and τ

is in the coverage of p sub-district; P cell signal to the carrier/interface ratio η of q cell signal and relative time delay τ value fall into the probability of feasible subdomain , u=1,2; ...; U, v=1,2; ...., V.Use

Come the joint probability density function f of approximate η of replacement and τ _{P, q}(η, τ).

The measured data of the collection of present embodiment is the data that collect on road through drive test in the existing network optimizing project, promptly the carrier/interface ratio value η on the geographical position that road distributes and relative time delay measured value.

The sub-district of all sub-districts is to a FER in the S202 dispensed zone;

Belong to same scrambler family according to the scrambler of compound key set equality, 128 scramblers of TD-SCDMA network are divided into 12 scrambler families;

Sub-district based on the sub-district uses the arbitrary scrambler in the same scrambler family to obtain is identical to a FER, traversal calculate all sub-districts to use different scrambling codes family to the time FER;

All location point places on the overlay area, main plot are disturbed by neighboring area signal the FER cause and as the sub-district of this main plot to a FER;

Described location point place by neighboring area signal disturb the FER cause be this main plot, location point place and adjacent area 16 * 16 totally 256 code channels to the mean value of last FER.

Further, S202 specifically comprises:

Calculate the scope in relative time delay: the lower bound of scope was rounded up by the minimum value in relative time delay in the measured data and obtained relative time delay; The upper bound is rounded up by the maximum in relative time delay in the measured data that is not more than 12dB corresponding to carrier/interface ratio and obtains, again according to scope and interval width calculating in relative time delay interval number in relative time delay;

The measured data of present embodiment is the data that in the engineering existing network collected, and the interval width of formula (9) τ in middle relative time delay is taken as a chip width, promptly

Δτ _v＝τ _v-τ _v-1＝1，v＝1，2，...，V

Then v interval does

τ _vAnd τ _V-1Be respectively v interval up-and-down boundary value, integer

Be v interval interval intermediate value.

From measured data, obtain

by the minimum value in relative time delay in the measured data and round up and obtain.

The maximum in relative time delay rounds up and obtains in the measured data of from measured data, obtaining and being not more than 12dB by carrier/interface ratio.Only getting the measured data of relative time delay in , is because carrier/interface ratio can not exert an influence to the proper communication of this community user greater than the then adjacent area interference of 12dB.

Computation interval number V, $V=\frac{{\stackrel{\&OverBar;}{\mathrm{\&tau;}}}_V-{\stackrel{\&OverBar;}{\mathrm{\&tau;}}}_0}{\mathrm{\&Delta;\; \&tau;}}+1.$

Table 3 has provided measured data and interval division one example in relative time delay.

Interval intermediate value, affiliated interval and sequence number one example thereof

in table 3 relative time delay

FER by interval quantization, is obtained the FER quantized value; Again according to the quantized interval boundary value of FER, traversal calculate all scrambler families in the relative time delay scope all relative time delays interval intermediate value places carrier/interface ratio interval border value; Interval and relative time delay, the interval was added up measured data according to carrier/interface ratio, obtained the joint probability distribution function in carrier/interface ratio and relative time delay;

In the present embodiment, the interval of carrier/interface ratio is calculated by the quantized interval of FER and is obtained.By user's level of perceived (U) of operator regulation, the codomain of FER

is divided into U interval.For example, FER is divided into three quantized interval≤10 ^-2, (10 ^-2, 10 ^-1] and＞10 ^-1, correspond respectively to user's level of perceived " well ", " relatively poor " and " unacceptable ".

Obtain U interval of carrier/interface ratio from the quantized interval of FER.The lower bound η of carrier/interface ratio interval 1 wherein ₀=-∞, the upper bound η of the interval U of carrier/interface ratio _U=∞.Table 4 provides interval corresponding relation one example of user's level of perceived, FER quantized interval and carrier/interface ratio.

The interval example (U=3) of table 4 user level of perceived, FER quantized interval, FER quantized value and carrier/interface ratio

Carrier/interface ratio section boundaries value η ₁, η ₂..., η _U-1It is relative time delay The function right with scrambler family.Travel through 12 * 12 scrambler families to V relative time delay; Calculate the interval intermediate value of different delay

and locate the carrier/interface ratio interval border value of different scrambling codes family to correspondence, the flow process of calculating is as shown in Figure 3.Carrier/interface ratio interval border value according to table 3 and table 4 example calculation obtain is seen table 5.

According to the quantized interval of FER, calculate the quantized value of FER.Quantized value can be the intermediate value of quantized interval up-and-down boundary, also can be any numerical value in upper boundary values, lower boundary or the quantized interval.With table 4 is example, and the quantized value of FER can be taken as

$G_{T_p,T_q}^{u,v}=\left\{\begin{array}{c}0,G_{T_p,T_q}(\mathrm{\&eta;},\mathrm{\&tau;})\&le;{10}^{-2}\\ 0.05,G_{T_p,T_q}(\mathrm{\&eta;},\mathrm{\&tau;})\&Element;({10}^{-2},{10}^{-1}]\\ 0.1,G_{T_p,T_q}(\mathrm{\&eta;},\mathrm{\&tau;})>{10}^{-1}\end{array}\right.$

According to the joint probability distribution function and the FER quantized value in carrier/interface ratio and relative time delay, the traversal calculation plot to use different scrambling codes family to the time FER.

According to the interval division of relative time delay and carrier/interface ratio, the data computation that is obtained by actual measurement obtains the joint probability distribution function in carrier/interface ratio and relative time delay

$P((\mathrm{\&eta;},\mathrm{\&tau;})\&Element;{\mathrm{\&Theta;}}_{p,q}^{u,v})=\frac{Z((\mathrm{\&eta;},\mathrm{\&tau;})\&Element;{\mathrm{\&Theta;}}_{p,q}^{u,v})}{Z((\mathrm{\&eta;},\mathrm{\&tau;})\&Element;{\mathrm{\&Theta;}}_{p,q})}$

u＝1，2，...，U (10)

v＝1，2，...，V

In the formula (10),

Be the actual measurement carrier/interface ratio η and relative time delay τ fall into feasible subdomain

The data number, Z ((η, τ) ∈ Θ _{P, q}) be the data sum of p sub-district.

At last, with formula (10) substitution formula (9), calculate the sub-district to a FER, promptly

$I_{p,q}^{T_p,T_q}={\mathrm{\&Omega;}}_p\underset{u=1}{\overset{U}{\mathrm{\&Sigma;}}}\underset{v=1}{\overset{V}{\mathrm{\&Sigma;}}}[G_{T_p,T_q}^{u,v}\&CenterDot;P((\mathrm{\&eta;},\mathrm{\&tau;})\&Element;{\mathrm{\&Theta;}}_{p,q}^{u,v})]$

p，q＝1，...，M

T _p，T _q＝1，...，12

S203 calculates the FER matrix;

Multiplied each other to a FER and cell weight in the sub-district of all sub-districts, the sub-district that obtains all sub-districts is to a weighting FER;

Described cell weight artificially is provided with according to the significance level of sub-district;

With all sub-districts a weighting FER is constructed a FER matrix.

The practical implementation process of present embodiment S203 can be with reference to S102.

S204 is according to the FER matrix, to all cell allocation scrambler families in the range of distribution;

Scrambler family is assigned as twice planning problem of 0-1, according to the FER matrix that obtains in the step (2), finds the solution decision variable, obtains scrambler family allocative decision;

The practical implementation process of present embodiment S204 can be with reference to S103.

S205 selects scrambler according to the scrambler family allocative decision that obtains to all sub-districts in the range of distribution;

The scrambler of assigning to based on neighbor cell belongs to different code characters, from the scrambler family that the sub-district is assigned to, selects scrambler.

The practical implementation process of present embodiment S205 can be with reference to S104.

The signal carrier/interface ratio interval border value that table 5 obtains according to the example calculation of table 3 and table 4

Embodiment 3

Technical scheme provided by the invention also can adopt distribution system as shown in Figure 4 to realize.A kind of scrambler distribution system that can be applicable to the range of distribution of grid planning stage comprises:

The sub-district is to a FER computing module 401:

The Back ground Information and the grid lattice point interval that connect the input range of distribution; The Back ground Information of range of distribution comprises that size, each base station location and the transmitting power of regional area coverage and radio signal propagation path loss theoretical model calculate required out of Memory; Uniform grid is carried out in the range of distribution divides, according to base station transmitting power and propagation path loss theoretical model, obtain all grid lattice point place main plot signals and all other cell signals the carrier/interface ratio value and relative time delay value; Based on this carrier/interface ratio value and relative time delay value, traversal calculate all sub-districts to use different scrambling codes family to the time the sub-district to a FER as the sub-district to a FER set, export this sub-district to a FER set;

FER matrix computations module 402:

Receive described sub-district to a FER set; According to the significance level of sub-district the cell weight of all sub-districts is set, the sub-district of each sub-district cell weight to a FER and this sub-district is multiplied each other, the sub-district that obtains all sub-districts is to a weighting FER; Sub-district with described all sub-districts constitutes the FER matrix to a weighting FER, exports this FER matrix.

Scrambler family distribution module 403:

Receive described FER matrix; Distribute 0-1 quadratic programming model with FER matrix construction scrambler family, and find the solution the decision variable of optimum scrambler family allocative decision; Export the decision variable of this optimum scrambler family allocative decision.

Scrambler is selected module 404:

Receive the decision variable of optimum scrambler family allocative decision; The scrambler of assigning to based on neighbor cell belongs to different code characters, from the scrambler family that the sub-district is assigned to, selects scrambler as the scrambler allocation result, exports this scrambler allocation result.

Embodiment 4

Present embodiment provides the another kind of distribution system that scrambler distributes, and its input data are the existing network measured datas of from the range of distribution of existing network optimizing phase, gathering, and collect on road with equipment usually.

The scrambler distribution system of a kind of TD-SCDMA network as shown in Figure 5 can be applicable to the range of distribution of existing network optimizing phase, comprising:

The sub-district connects the input image data to a FER computing module 501, and calculation plot is to a FER; The sub-district is made up of one group of submodule a FER computing module 501, specifically comprises:

Relative time delay, interval number calculating sub module 511:

The measured data that input is gathered is calculated the scope in relative time delay; According to range computation in this interval number in relative time delay wherein, all section boundaries value in relative time delay in relative time delay; Export section boundaries value in these all relative time delays;

FER quantizes and carrier/interface ratio interval division submodule 512:

Receive section boundaries value in described all in relative time delay; FER is pressed interval quantization; Again according to the intermediate value in relative time delay in the boundary value of each FER quantized interval and each interval, calculate a pair of scrambler family in the boundary value of this FER quantized interval carrier/interface ratio value with intermediate value place in this in relative time delay in relative time delay; With this carrier/interface ratio value as carrier/interface ratio section boundaries value; It is right to travel through all scrambler families, obtain all scrambler families in the boundary value of all FER quantized intervals, relative time delay interval relative time delay the intermediate value place carrier/interface ratio section boundaries value; Export all carrier/interface ratio section boundaries values; ,

The joint probability distribution function calculating sub module 513 in carrier/interface ratio and relative time delay:

Receive described all in relative time delay the section boundaries value with all carrier/interface ratio section boundaries values; Section boundaries value in all according to each scrambler family is right is added up measured data with all carrier/interface ratio section boundaries values in relative time delay, obtains the right carrier/interface ratio of this scrambler family and the joint probability distribution function in relative time delay; It is right to travel through all scrambler families, obtains the right carrier/interface ratio of all scrambler families and the joint probability distribution function in relative time delay; Export the right carrier/interface ratio of these all scrambler families and the joint probability distribution function in relative time delay;

The sub-district is to a FER calculating sub module 514:

Receive the right carrier/interface ratio of described all scrambler families and the joint probability distribution function in relative time delay; According to each sub-district to use different scrambling codes family to the time carrier/interface ratio and the joint probability distribution function in relative time delay and FER quantized value, calculate this sub-district to use different scrambling codes family to the time the sub-district to a FER; Traversal calculate all sub-districts to use different scrambling codes family to the time the sub-district to a FER as the sub-district to a FER set, export this sub-district to a FER set.

FER matrix computations module 502:

Receive described sub-district to a FER set; According to the significance level of sub-district the cell weight of all sub-districts is set, the sub-district of each sub-district cell weight to a FER and this sub-district is multiplied each other, the sub-district that obtains all sub-districts is to a weighting FER; Sub-district with described all sub-districts constitutes the FER matrix to a weighting FER, exports this FER matrix.

Scrambler family distribution module 503:

Receive described FER matrix; Distribute 0-1 quadratic programming model with FER matrix construction scrambler family, and find the solution the decision variable of optimum scrambler family allocative decision; Export the decision variable of this optimum scrambler family allocative decision.

Scrambler is selected module 504:

Receive the decision variable of optimum scrambler family allocative decision; The scrambler of assigning to based on neighbor cell belongs to different code characters, from the scrambler family that the sub-district is assigned to, selects scrambler as the scrambler allocation result, exports this scrambler allocation result.

Beneficial effect

The beneficial effect of technical scheme according to the invention comprises:

(1) scrambler has been considered the actual wireless communication environments in distributing; Can accurately reflect actual network interferences intensity; And through scrambler family distribute make scrambler to correlation distribute and wireless propagation environment is complementary; The little sub-district of propagation path loss is right to the scrambler of giving a little less than the correlation, guarantees that the total FER of the whole network is minimum;

(2) scrambler is distributed be reduced to that scrambler family distributes and scrambler selected for two steps, greatly reduce computation complexity.It is that 12 scrambler families are distributed M minizone that first step scrambler family distributes, and the algorithm complex that traversal is found the solution is O (12M).Second step was distributed scrambler, and complexity is O (M) when adopting the chain type apportion design.The complexity of whole algorithm is O (12M)+O (M) ≈ O (12M), finds the solution complexity O (128M) far below the traversal that 128 scramblers are directly distributed M minizone;

(3) be applicable to the scrambler distribution that the network planning stage utilizes electronic chart and propagation path loss theoretical model to carry out, also be applicable to the scrambler distribution that the existing network optimizing phase utilizes the limited measured data of precision to carry out.

The beneficial effect of technical scheme according to the invention can also be verified through emulation.For example, system works is in the 2G frequency range, and propagation path loss adopts microcellulor PCS model.The sub-district is that omnidirectional covers, and omnidirectional antenna is adopted in the base station, with the power of 25dBm to full Cell Broadcast CB pilot signal.The simulating area size is 7000m*6200m, and 100 base stations that distributed in the zone are as shown in Figure 6.Whole network is divided uniform grid, and sizing grid is 10m*10m.The weight of each sub-district is identical, i.e. β _p=1, i=1,2 ..., M.Using genetic algorithm to carry out scrambler family to 100 sub-districts in this zone distributes.Genetic algorithm is tending towards convergence after iteration 1000 times, the total FER of the whole network (the fitness function value in the genetic algorithm) is reduced to 34.51 by 67.06.This explanation employing scrambler of the present invention family distributes can effectively reduce the total FER of the whole network.

Obviously, those skilled in the art can carry out various changes and distortion to the present invention and not break away from the spirit and scope of the present invention.Like this, if these modifications of the present invention and distortion belong within the scope of claim of the present invention and equivalent technologies thereof, then the present invention also is intended to comprise these changes and is out of shape interior.

Claims (6)

1. the scrambling code distribution method of a TD-SCDMA network is characterized in that, comprises the steps:

(1) in the dispensed zone sub-district of all sub-districts to a FER;

Belong to same scrambler family according to the scrambler of compound key set equality, 128 scramblers of TD-SCDMA network are divided into 12 scrambler families;

Sub-district based on the sub-district uses the arbitrary scrambler in the same scrambler family to obtain is identical to a FER, traversal calculate all sub-districts to use different scrambling codes family to the time FER;

All location point places on the overlay area, main plot are disturbed by neighboring area signal the FER cause and as the sub-district of this main plot to a FER;

Described location point place by neighboring area signal disturb the FER cause be this main plot, location point place and adjacent area 16 * 16 totally 256 code channels to the mean value of last FER;

(2) calculate the FER matrix;

Multiplied each other to a FER and cell weight in the sub-district of all sub-districts, the sub-district that obtains all sub-districts is to a weighting FER;

Described cell weight artificially is provided with according to the significance level of sub-district;

With all sub-districts a weighting FER is constructed a FER matrix;

(3) according to the FER matrix, to all cell allocation scrambler families in the range of distribution;

Scrambler family is assigned as twice planning problem of 0-1, according to the FER matrix that obtains in the step (2), finds the solution decision variable, obtains scrambler family allocative decision;

(4) according to the scrambler family allocative decision that obtains, scrambler is selected in all sub-districts in the range of distribution;

The scrambler of assigning to based on neighbor cell belongs to different code characters, from the scrambler family that the sub-district is assigned to, selects scrambler.

2. scrambling code distribution method according to claim 1 is characterized in that,

The FER that all location point places on the overlay area, main plot are disturbed by neighboring area signal described in the described step (1) causes and as the sub-district of this main plot a FER is comprised step:

Based on the Back ground Information of range of distribution, uniform grid is carried out in the range of distribution divide;

Based on the Back ground Information of range of distribution, with the carrier/interface ratio and the relative time delay at electronic chart and all grid lattice point places of radio signal propagation path loss theoretical model calculating;

According to the carrier/interface ratio at all grid lattice point places with calculate the FER at all grid lattice point places relative time delay;

With the FER at grid lattice point place and the FER sum at interior all the location point places of approximate this grid of corresponding grid product of areas, as the FER of this grid;

FER sum with all location point places in approximate this sub-district of FER sum of all grids in the sub-district.

3. scrambling code distribution method according to claim 2 is characterized in that, described Back ground Information comprises that size, each base station location and the transmitting power of range of distribution area coverage and radio signal propagation path loss theoretical model calculate required out of Memory.

4. scrambling code distribution method according to claim 1 is characterized in that,

In step (1) before, also comprise step: the measured data of gathering existing network; Described measured data comprise in the range of distribution main plot signal and its all neighboring area signals on some places the carrier/interface ratio measured value and relative time delay measured value; Described measured data does not require it is geographical equally distributed;

The FER that all location point places on the overlay area, main plot are disturbed by neighboring area signal described in the described step (1) causes and as the sub-district of this main plot a FER is comprised step:

Calculate the scope in relative time delay: the lower bound of scope was rounded up by the minimum value in relative time delay in the measured data and obtained relative time delay; The upper bound is rounded up by the maximum in relative time delay in the measured data that is not more than 12dB corresponding to carrier/interface ratio and obtains, again according to scope and interval width calculating in relative time delay interval number in relative time delay;

FER by interval quantization, is obtained the FER quantized value; Again according to the quantized interval boundary value of FER, traversal calculate all scrambler families in the relative time delay scope all relative time delays interval intermediate value places carrier/interface ratio interval border value;

Interval and relative time delay, the interval was added up measured data according to carrier/interface ratio, obtained the joint probability distribution function in carrier/interface ratio and relative time delay;

According to the joint probability distribution function and the FER quantized value in carrier/interface ratio and relative time delay, the traversal calculation plot to use different scrambling codes family to the time FER.

5. the scrambler distribution system of a TD-SCDMA network is characterized in that, described system comprises:

The sub-district is to a FER computing module:

The Back ground Information and the grid lattice point interval that connect the input range of distribution; The Back ground Information of range of distribution comprises that size, each base station location and the transmitting power of regional area coverage and radio signal propagation path loss theoretical model calculate required out of Memory; Uniform grid is carried out in the range of distribution divides, according to base station transmitting power and propagation path loss theoretical model, obtain all grid lattice point place main plot signals and all other cell signals the carrier/interface ratio value and relative time delay value; Based on this carrier/interface ratio value and relative time delay value, traversal calculate all sub-districts to use different scrambling codes family to the time the sub-district to a FER as the sub-district to a FER set, export this sub-district to a FER set;

FER matrix computations module:

Receive described sub-district to a FER set; According to the significance level of sub-district the cell weight of all sub-districts is set, the sub-district of each sub-district cell weight to a FER and this sub-district is multiplied each other, the sub-district that obtains all sub-districts is to a weighting FER; Sub-district with described all sub-districts constitutes the FER matrix to a weighting FER, exports this FER matrix;

Scrambler family distribution module:

Receive described FER matrix; Distribute 0-1 quadratic programming model with FER matrix construction scrambler family, and find the solution the decision variable of optimum scrambler family allocative decision; Export the decision variable of this optimum scrambler family allocative decision;

Scrambler is selected module:

Receive the decision variable of optimum scrambler family allocative decision; The scrambler of assigning to based on neighbor cell belongs to different code characters, from the scrambler family that the sub-district is assigned to, selects scrambler as the scrambler allocation result, exports this scrambler allocation result.

6. device according to claim 5 is characterized in that, one group of submodule replacement can also be used to a FER computing module in described sub-district:

Relative time delay interval number calculating sub module:

The measured data that input is gathered is calculated the scope in relative time delay; According to range computation in this interval number in relative time delay wherein, all section boundaries value in relative time delay in relative time delay; Export section boundaries value in these all relative time delays;

FER quantizes and carrier/interface ratio interval division submodule:

Receive section boundaries value in described all in relative time delay; FER is pressed interval quantization; Again according to the intermediate value in relative time delay in the boundary value of each FER quantized interval and each interval, calculate a pair of scrambler family in the boundary value of this FER quantized interval carrier/interface ratio value with intermediate value place in this in relative time delay in relative time delay; With this carrier/interface ratio value as carrier/interface ratio section boundaries value; It is right to travel through all scrambler families, obtain all scrambler families in the boundary value of all FER quantized intervals, relative time delay interval relative time delay the intermediate value place carrier/interface ratio section boundaries value; Export all carrier/interface ratio section boundaries values; ,

The joint probability distribution function calculating sub module in carrier/interface ratio and relative time delay:

Receive described all in relative time delay the section boundaries value with all carrier/interface ratio section boundaries values; Section boundaries value in all according to each scrambler family is right is added up measured data with all carrier/interface ratio section boundaries values in relative time delay, obtains the right carrier/interface ratio of this scrambler family and the joint probability distribution function in relative time delay; It is right to travel through all scrambler families, obtains the right carrier/interface ratio of all scrambler families and the joint probability distribution function in relative time delay; Export the right carrier/interface ratio of these all scrambler families and the joint probability distribution function in relative time delay;

The sub-district is to a FER calculating sub module:

Receive the right carrier/interface ratio of described all scrambler families and the joint probability distribution function in relative time delay; According to each sub-district to use different scrambling codes family to the time carrier/interface ratio and the joint probability distribution function in relative time delay and FER quantized value, calculate this sub-district to use different scrambling codes family to the time the sub-district to a FER; Traversal calculate all sub-districts to use different scrambling codes family to the time the sub-district to a FER as the sub-district to a FER set, export this sub-district to a FER set.

Images