CN103326743B - Method and device for acquiring merging weight - Google Patents

Abstract

The invention provides a method and device for acquiring merging weight. The method comprises the steps of selecting a time quantum used for calculating channel estimated values of a channel, calculating the channel estimated values of the channel in the time quantum, carrying out interpolation calculation on the channel estimated values, and obtaining merging weight of a receiver according to the channel estimated values obtained after the interpolation calculation. According to the technical scheme, the technical problems that in the prior art, merging weight updating speed is slow, and channel changes can not be accurately reflected are solved, merging weight updating speed is improved, the granularity of merging weight is made to be smaller, and therefore a fading channel capable of resisting quick time varying and a certain extent of residual frequency offset can be achieved, and the error rate of the receiver is lowered.

Description

Merge acquisition methods and the device of weight

Technical field

The present invention relates to the communications field, in particular to a kind of acquisition methods merging weight and device.

Background technology

In code division multiple access system, Rake (Rake) receiver is widely used.Rake receiver is to the multipath searching Carry out descrambling and de-spreading, then maximum-ratio combing is carried out with certain weight.Common merging weight is the channel of a certain bar multipath The conjugation of valuation.With the development of advanced receiver techniques, merge the energy footpath that multipath is no longer limited to search, also necessarily The noise path of quantity.Now, merging weight is no longer channel guess value, but obtained based on certain algorithm, such as lowest mean square Error (MMSE) or senior Rake receiver (Arake).Arake receiver is the extension of Rake receiver, the multipath that it merges Number is referred to as noise path more than the bar number of true multipath, the part having more.Certainly, merge weight to remain based on channel guess value 's.Either Rake receiver or senior receiver, the calculating merging weight is all most important, decides final error code Rate.

In order to antinoise, channel guess value is typically all average result in a period of time, average after channel guess value meeting It is applied in this time period.Channel response when channel is quick time-varying, or when having larger residual frequency departure, in average time section It is also time-varying.It is inappropriate it is impossible to accurately reflect channel variation that one constant channel guess value is applied to this time period. During channel quick time-varying, the bit error rate can improve, for example vehicle-mounted mobile platform.Then cause planisphere when there is larger residual frequency departure Rotation.

For the problems referred to above in correlation technique, effective solution is not yet proposed at present.

Content of the invention

Slowly ask it is impossible to accurately reflect the technology such as channel variation for merging weight renewal speed present in correlation technique Topic, the present invention provides a kind of acquisition methods merging weight and device, at least to solve above-mentioned technical problem.

According to an aspect of the invention, it is provided a kind of acquisition methods merging weight, including；Choose for calculating letter The time period of the channel guess value in road；Calculate the channel guess value in the described time period for the described channel；Described channel guess value is inserted Value computing, and the merging weight of receiver is obtained according to the channel guess value carrying out obtaining after interpolation arithmetic.

Above-mentioned receiver is Rake Rake receiver；Channel guess value according to carrying out obtaining after interpolation arithmetic obtains receiver Merging weight, including：Conjugate operation is carried out to the channel guess value carrying out obtaining after interpolation arithmetic, obtains described Rake receiver Merging weight.

Above-mentioned receiver is senior Rake Arake receiver；Before interpolation arithmetic is carried out to described channel guess value, above-mentioned Method also includes：Obtain the initial merging weight of described Arake receiver；

Channel guess value according to carrying out obtaining after interpolation arithmetic obtains the merging weight of receiver, including：With described initial Merging weight is initial value, to carry out the input as sef-adapting filter of the channel guess value that obtains after interpolation arithmetic, to described just The merging weight that begins is adjusted, and obtains the merging weight of the described Arake receiver after interpolation arithmetic.

The merging weight of the described Arake receiver after following iterative process obtains interpolation arithmetic, including：

w(0)

E (n)=d (n)-w^H(n)u(n)

W (n+1)=w (n)+k (n) e^*N (), wherein, w (0) is described initial merging weight, after w (n) is nth iteration Merging weight, w^HN () represents the conjugate transpose of w (n), u (n) be carry out interpolation arithmetic after the channel guess value that obtains, d (n) is The expected value of described sef-adapting filter output, e (n) is that the difference between described sef-adapting filter output and expected value is missed Difference, e (n) represents the conjugation of e (n), and k (n) is weight renewal vector.

In described iterative process, methods described also includes：Described error amount during current iteration is more than previous During the error amount of secondary iteration, regenerate described weight renewal vector so that iterative process tends to restraining.

Described iterative process, when meeting one below condition, jumps out described iterative process：The iteration of described iterative process Number of times reaches predetermined threshold value；Described error amount meets the predetermined condition jumping out iterative process.

Above-mentioned channel linearly changed within the described time period chosen.

Interpolation arithmetic is carried out to described channel guess value, including：Using linear spline interpolation algorithm, described channel guess value is entered Row interpolation computing.

According to another aspect of the present invention, there is provided a kind of acquisition device merging weight, in receiver, bag Include；Choose module, for choosing the time period of the channel guess value for calculating channel；Computing module, for calculating described channel Channel guess value in the described time period；Interpolating module, for carrying out interpolation arithmetic to described channel guess value；Acquisition module, is used for Channel guess value according to carrying out obtaining after interpolation arithmetic obtains the merging weight of receiver

Above-mentioned acquisition module, is additionally operable to carry out conjugate operation to the channel guess value carrying out obtaining after interpolation arithmetic, obtains auspicious The merging weight of gram Rake receiver.

Above-mentioned acquisition module includes：First acquisition unit, before carrying out interpolation arithmetic to described channel guess value, obtains The initial merging weight of senior Rake Arake receiver；Second acquisition unit, for described initial merging weight be initial Value, to carry out the input as sef-adapting filter of the channel guess value that obtains after interpolation arithmetic, is carried out to described initial merging weight Adjustment, obtains the merging weight of the described Arake receiver after interpolation arithmetic.

By the present invention, obtain, using based on the channel guess value after interpolation, the technical scheme merging weight, solve correlation In technology, the merging weight renewal speed existing, slowly it is impossible to accurately reflect the technical problems such as channel variation, improves merging weight Renewal speed, and make merging weight granularity less, thus reached can resist the fading channel of quick time-varying with certain The residual frequency departure of degree, reduces the bit error rate of receiver.

Brief description

Accompanying drawing described herein is used for providing a further understanding of the present invention, constitutes the part of the application, this Bright schematic description and description is used for explaining the present invention, does not constitute inappropriate limitation of the present invention.In the accompanying drawings：

Fig. 1 is the flow chart of the acquisition methods of the merging weight according to the embodiment of the present invention；

Fig. 2 is the schematic diagram that according to the preferred embodiment of the present invention, quick time-varying channel is divided with average area；

Fig. 3 is the schematic diagram obtaining average channel valuation in average area according to the preferred embodiment of the present invention；

Fig. 4 is the principle schematic of the adaptive-filtering according to the preferred embodiment of the present invention；

Fig. 5 is the schematic diagram entering row interpolation according to the preferred embodiment of the present invention using linear spline interpolation algorithm；

Fig. 6 be based on the optimization of embodiment illustrated in fig. 4 after adaptive-filtering principle schematic；

Fig. 7 is the structured flowchart of the acquisition device of the merging weight according to the embodiment of the present invention；

Fig. 8 is the structured flowchart of the acquisition module based on embodiment illustrated in fig. 7；

Fig. 9 is the structural representation of the device of the maximum-ratio combing weight according to the preferred embodiment of the present invention；

Figure 10 is the principle schematic realizing adaptive-filtering according to the preferred embodiment of the present invention using LMS.

Specific embodiment

To describe the present invention in detail below with reference to accompanying drawing and in conjunction with the embodiments.It should be noted that not conflicting In the case of, the embodiment in the application and the feature in embodiment can be mutually combined.

Fig. 1 is the flow chart of the acquisition methods of the merging weight according to the embodiment of the present invention.As shown in figure 1, the method bag Include：

Step S102, chooses the time period of the channel guess value for calculating channel；

Step S104, calculates the channel guess value in the described time period for the described channel；

Step S106, carries out interpolation arithmetic, and is estimated according to the channel carrying out obtaining after interpolation arithmetic to described channel guess value Value obtains the merging weight of receiver.

No matter channel be whether quick time-varying or have larger residual frequency departure, all can be obtained using above-mentioned processing procedure The merging weight of receiver.Especially in channel for quick time-varying or when having larger residual frequency departure, can solve due to merging weight Renewal speed cannot accurately reflect the technical problem of channel variation slowly, makes merging weight (such as maximum-ratio combing weight) update speed Faster, granularity is less for degree, can resist the fading channel of quick time-varying and a certain degree of residual frequency departure, reduce the mistake of receiver Code check.

In step S102 and step S104, can be realized using following processing procedure：When channel is time-varying, such as Fig. 2 Shown, choose certain average area (i.e. time period) so that channel variation is approximately linear in this interval.After now average Channel guess value reflection be average slot center channel response.Based on approximately linear it is assumed that inserting to channel guess value Value (entering row interpolation for example with linear spline interpolation algorithm) is reducing the granularity of channel guess value.

In step s 106, row interpolation can be entered to channel guess value using following processing procedure：

As shown in Figure 3 it is assumed that the average widow time of channel guess value is T, then average result reflection is in window The channel response of heart position, makes window center position manipulative indexing n1, n2, n3, and corresponding channel guess value is w_n1, w_n2, w_n3.Adopt Enter row interpolation with linear spline algorithm, make location of interpolation index as n.

If the corresponding window of n1, n2 is first and last average window respectively, then for n ＜ n1 or n ＞ n3 For, interpolation belongs to extrapolation, and for n1≤n≤n3, interpolation belongs to interpolation.Extrapolation belongs to prediction, and therefore extrapolation method must Known trend must be met, otherwise error can be than larger.Known trend is：Frequency deviation or decline can cause the rotation of channel guess value Turn, but turn not too large in average window inward turning, can be with a curve Lai approximate.

In step s 106, merge the acquisition scheme of weight to Rake receiver (or referred to as common Rake receiver) and height It is different for level Rake receiver.Specifically,

When receiver is for Rake Rake receiver, above-mentioned merging weight can be obtained by following processing procedure：To entering The channel guess value obtaining after row interpolation computing carries out conjugate operation, obtains the merging weight of described Rake receiver.And, due to Channel guess value is (approximate) linear change, and therefore, the merging weight of Rake receiver also meets linear relationship, so, for Channel guess value after interpolation may be directly applied to maximum-ratio combing.

When receiver is for senior Rake Arake receiver, above-mentioned merging weight can be obtained by following processing procedure： Before interpolation arithmetic is carried out to described channel guess value, obtain the initial merging weight of described Arake receiver；With described initial Merging weight is initial value, to carry out the input as sef-adapting filter of the channel guess value that obtains after interpolation arithmetic, to described just The merging weight that begins is adjusted, and obtains the merging weight of the described Arake receiver after interpolation arithmetic.

Wherein it is possible to the merging weight of the described Arake receiver after following iterative process obtains interpolation arithmetic, bag Include：

w(0)

E (n)=d (n)-w^H(n)u(n)

W (n+1)=w (n)+k (n) e^*(n)

Wherein, w (0) is described initial merging weight, and w (n) is the merging weight after nth iteration, w^HN () represents w (n) Conjugate transpose, u (n) be carry out interpolation arithmetic after the channel guess value that obtains, d (n) is the phase of described sef-adapting filter output Prestige value, e (n) is the difference between described sef-adapting filter output and expected value is error, and e (n) represents the conjugation of e (n), k N () is weight renewal vector.Different adaptive algorithms has the different forms of expression.

In general, as shown in figure 4, the input of sef-adapting filter is the training sequence receiving, because training sequence Expected value be known.Through the slip of the training sequence of certain length, the coefficient of transversal filter can tend towards stability.Upper State in iterative process, the input of sef-adapting filter is the channel guess value after interpolation, be exactly channel impulse response in fact, that is, train Sequence is equivalent to a unit impulse function.After channel fading, the signal u receiving (obtains after carrying out interpolation arithmetic Channel guess value) be no longer impulse function, but be desirable to be reduced to unit impulse function after horizontal filtering.So, such as W if fruit power normalizing^HThe output of u just should be equal to 1, power not in the case of normalizing the imaginary part of at least output valve be zero.Institute So that the expected value of wave filter output is w^HThe real part of u, error amount is：

E (n)=- i*imag (w^Hu)

Wherein imag represents imaginary part, i.e. imaginary part in plural number." i " is imaginary unit, and it is a square root of " -1 ".

In above-mentioned iterative process, because weight renewal vector determines the convergence process of above-mentioned iterative process, in order to protect Card convergence rate, accordingly, it would be desirable to consider the generation of weight renewal vector, the generation of weight renewal vector needs to consider error amount Convergence process.The generation of initial weight renewal vector is to maximize convergence rate as principle, once finding certain iterative process In (or during current iteration) error amount be more than front an iteration error amount, do not restrain, then need to regenerate weight Renewal vector is so that iterative process tends to restraining.

In order to control computation complexity, require when the output result (channel guess value after interpolation) of sef-adapting filter meets When need to terminate iterative process, above-mentioned iterative process, when meeting one below condition, can jump out described iterative process：Described The iterationses of iterative process reach predetermined threshold value；Described error amount meets the predetermined condition jumping out iterative process.

In step s 102, in order that merging weight also assume linear change, within the above-mentioned time period chosen, channel is special Property can linearly change, linear change herein can be in certain linear change allowing in deviation range however it is not limited to Absolute linear change.

Above-mentioned channel guess value is carried out with interpolation arithmetic and can adopt multiple interpolation algorithms, one in the present invention is preferable to carry out In mode, it is preferred to use linear spline interpolation algorithm carries out interpolation arithmetic to described channel guess value.

As shown in figure 5, linear spline interpolation is from unlike linear fit, the former is sampled point to be connected into line segment obtain Interpolated point, stretches out line segment and obtains extrapolation point.This interpolation algorithm is：

$w_n=\frac{w_{n2}-w_{n1}}{n2-n1}(n-n1)+w_{n1},n\∈\left\{\begin{array}{cccc}1& 2& \·\·\·& n2\end{array}\right\}$

$w_n=\frac{w_{n3}-w_{n2}}{n3-n2}(n-n2)+w_{n2},n\∈\left\{\begin{array}{cc}n2+1& \·\·\·\end{array}\right\}$

In order to more fully understand the adaptive-filtering process (iterative process) in above-described embodiment, this mistake is described briefly below The principle of journey.

As shown in fig. 6, the adaptive filtration theory after optimizing is as follows, in this prioritization scheme, increased adaptive iteration Escape mechanism and weight renewal vector generation mechanism.Specific as follows：

Step (1)：Channel guess value is entered with row interpolation, the channel guess value after each group of interpolation corresponds to an adaptive-filtering Device；

Step (2)：Calculate initial weight renewal vector using the channel guess value after interpolation；

Step (3)：The channel guess value u in all footpaths in some moment is laterally filtered, filter factor is i.e. current to be closed And weight w_now, i.e. sum (u.*conj (w_now))；Wherein sum is summation, and conj is conjugate operation, and " .* " transports for dot product Calculate, that is, two vectorial corresponding elements are multiplied.

Step (4)：The imaginary part taking the result of horizontal filtering is error, meets e (n)=- i*imag (w^Hu)；

Step (5)：If meeting maximum iteration time or jumping out condition (i.e. error meets condition), jump out iteration；

Step (6)：Jump out and jump out condition if meeting during iteration, then final weighted value is last iteration Weight, otherwise for the weighted value of last iteration；

Step (7)：If continuing iteration, and then judging whether current iteration dissipates, if dissipated, regenerating power Weight renewal vector, ignores current iteration；Otherwise update weighted value；

Step (8)：The current renewal merging weighted value, meets below equation：

w(0)

E (n)=d (n)-w^H(n)u(n)

W (n+1)=w (n)+k (n) e^*(n)；

Step (9)：If continuing iteration, repeat step (3)～step (8).

Additionally provide a kind of acquisition device merging weight in the present embodiment, this device is located in receiver, for reality Existing above-described embodiment and preferred implementation, had carried out repeating no more of explanation, be related to module in this device below Illustrate.As used below, the software of predetermined function and/or the combination of hardware can be realized in term " module ".Although Device described by following examples preferably to be realized with software, but hardware, or the realization of the combination of software and hardware It is also may and to be contemplated.Fig. 7 is the structured flowchart of the acquisition device of the merging weight according to the embodiment of the present invention.As Fig. 7 Shown, this device includes：

Choose module 70, connect to computing module, for choosing the time period of the channel guess value for calculating channel；

Computing module 72, connects to interpolating module 74, for calculating the channel guess value in the described time period for the described channel；

Interpolating module 74, connects to acquisition module 76, for carrying out interpolation arithmetic to described channel guess value；

Acquisition module 76, for obtaining the merging weight of receiver according to the channel guess value carrying out obtaining after interpolation arithmetic.

As described in said method embodiment, merge the acquisition scheme of weight (or referred to as common to Rake receiver Rake receiver) and senior Rake receiver for be different.For Rake receiver, can be realized using below scheme： Acquisition module 76, is additionally operable to carry out conjugate operation to the channel guess value carrying out obtaining after interpolation arithmetic, obtains Rake receiver Merge weight.

In order that merging weight also assume linear change, within the above-mentioned time period choosing module 70 selection, the characteristic of channel Can linearly change, linear change herein can be however it is not limited to absolutely in certain linear change allowing in deviation range To linear change.

For senior Rake (Arake) receiver, can be realized using below scheme：As shown in figure 8, above-mentioned acquisition module 76 include：

First acquisition unit 760, before carrying out interpolation arithmetic to described channel guess value, obtains senior Rake Arake The initial merging weight of receiver；

Second acquisition unit 762, for described initial merging weight as initial value, with carry out obtaining after interpolation arithmetic Channel guess value is the input of sef-adapting filter, described initial merging weight is adjusted, obtains described after interpolation arithmetic The merging weight of Arake receiver

In order to more fully understand above-described embodiment below in conjunction with preferred embodiment and relevant drawings detailed description.Need explanation , in the examples below, relate generally to WCDMA (Wideband Code Division Multiple Access, referred to as WCDMA) channel estimation and equalization in band receiver of base technology, with Rake receiver is especially relevant.Below Embodiment describes a kind of interpolation scheme of maximum-ratio combing weight in detail.

As shown in figure 9, the device of maximum-ratio combing weight is included with lower module：

Control channel descrambling and de-spreading module 90：It is used for control channel is done with descrambling and de-spreading process, one solution of each multipath Disturb despreading resource.

Data channel descrambling and de-spreading module 92：It is used for data channel is done with descrambling and de-spreading process, one solution of each multipath Disturb despreading resource, including energy footpath and noise path.

Channel estimation module 94：For estimating the channel response of each multipath.

Senior receiver merges weight computation module 96：Be used for calculating energy footpath and noise path in Arake merges power Weight, is then used for calculating all tap coefficients of transversal filter in MMSE receiver.Here refer to all and Rake receiver Different merging weight calculation, the merging weight calculating is not equal to the conjugation of channel guess value.

Maximum-ratio combing module 98：Maximum-ratio combing is carried out to the sign level data needing the multipath merging.

Demodulation coding module 100：Maximum-ratio combing result is demodulated, decoding, obtains the information bit of transmitting terminal.

The linear interpolation module 102 of channel guess value：Row interpolation is entered to the channel estimation after average, obtains the less letter of granularity Road valuation.

Adaptive-filtering interpolating module 104：The result of weight calculation is merged as initial value using senior receiver, with interpolation Channel guess value afterwards is the input of sef-adapting filter, is adjusted to merging weight by adaptive iteration, after obtaining interpolation Merging weight.

Control, the descrambling and de-spreading module of data channel, channel estimation merges weight computation module, maximum with senior receiver Ratio merges module, the demodulation coding module composition necessary module of whole channel estimation and equalization, and the interpolation of channel guess value and Adaptive-filtering interpolation is then the nucleus module merging Power Interpolation, for improving systematic function, opposing quick time-varying channel with Residual frequency departure.

It should be noted that the linear interpolation module 102 of the channel guess value in the present embodiment is equivalent in above-described embodiment Selection module 70, computing module 72 and interpolating module 74；On adaptive-filtering interpolating module 104 in the present embodiment is equivalent to State the acquisition module 76 in embodiment.

The handling process of maximum-ratio combing weight to be described below based on the principle in embodiment illustrated in fig. 9：

The first step：All multipaths are controlled with the descrambling and de-spreading of channel, data channel, obtain sign level data.

Second step：The symbol data of control channel is processed, obtains the channel guess value in every footpath, and in average window Inside carry out average.

3rd step：The channel statistics such as the channel guess value being obtained using second step calculate the merging power of senior receiver Weight.

4th step：The channel guess value that second step is obtained carries out linear spline interpolation, obtains the Rake receiver after interpolation Merge weight, meet below equation：

$w_n=\frac{w_{n2}-w_{n1}}{n2-n1}(n-n1)+w_{n1},n\∈\left\{\begin{array}{cccc}1& 2& \·\·\·& n2\end{array}\right\}$

$w_n=\frac{w_{n3}-w_{n2}}{n3-n2}(n-n2)+w_{n2},n\∈\left\{\begin{array}{cc}n2+1& \·\·\·\end{array}\right\}$

5th step：Channel guess value after the interpolation being obtained with the 4th step is input, the senior receiver being obtained with the 3rd step Merging weight be initial value, carry out adaptive-filtering interpolation, obtain the senior receiver after interpolation and merge weight.

6th step：Maximum-ratio combing is carried out to the symbol data needing the data channel merging multipath, merges weight and be derived from 4th step or the 5th step.

7th step：Data after merging is demodulated and decodes, obtain information bit.

5th step is further broken into following steps, as shown in Figure 10, adaptive-filtering adopts LMS to realize, and LMS is adaptive The iterative formula that should filter is：

w(0)

E (n)=d (n)-w^H(n)u(n)

W (n+1)=w (n)+μ u (n) e^*(n)

μ is iteration step length, and μ u (n) constitutes weight renewal vector k (n).

Step 1：Channel guess value is entered row interpolation, senior receiver merging weight is copied to the granularity after interpolation, each Channel guess value after group interpolation constitutes a LMS adaptive-filtering.

Step 2：Calculate initial step length using the channel guess value u after interpolation, meet

$\mathrm{\μ}=\frac{2}{\mathrm{NS}}=\frac{2}{N*\mathrm{mean}\left\{{\left|u\right|}^2\right\}*3}$

Wherein S is the maximum of the power spectral density of input u, and N is filter order, and mean represents and takes average, uses mean {|u|²* 3 approximate S.

Step 3：The channel guess value u in all footpaths in some moment is laterally filtered, filter factor is i.e. current to be merged Weight

Step 4：The imaginary part taking the result of horizontal filtering is error, meets e (n)=- i*imag (w^Hu)

Step 5：If meeting maximum iteration time or jumping out condition, jump out LMS iteration, the condition of jumping out is

Wherein th is threshold value, and e is error amount.Wherein real is the real part in plural number, and imag is imaginary part.

Step 6：Jump out and jump out condition if meeting during iteration, then final weighted value is the power of last iteration Weight, otherwise for the weighted value of last iteration

Step 7：If continuing iteration, and then judge whether current iteration dissipates, that is, meet

| e (n) | ＞ | e (n-1) |

If dissipated, step-length removes 2, regenerates weight renewal vector；Otherwise update a front weighted value once to change with front For error, even a front weight is equal to present weight, a front iteration error is equal to error current.

Step 8：The current renewal merging weighted value, meets below equation：

w(0)

E (n)=d (n)-w^H(n)u(n)

W (n+1)=w (n)+μ u (n) e^*(n).

Step 9：If continuing iteration, repeat step 3～step 8.

In another embodiment, additionally provide a kind of software, this software is used for executing above-described embodiment and preferably real Apply the technical scheme described in mode.

In another embodiment, additionally provide a kind of storage medium, be stored with this storage medium above-mentioned software, should Storage medium includes but is not limited to：CD, floppy disk, hard disk, scratch pad memory etc..

Obviously, those skilled in the art should be understood that each module of the above-mentioned present invention or each step can be with general Computing device realizing, they can concentrate on single computing device, or be distributed in multiple computing devices and formed Network on, alternatively, they can be realized with the executable program code of computing device, it is thus possible to they are stored To be executed by computing device in the storage device, and in some cases, can be with different from shown in order execution herein The step going out or describing, or they are fabricated to respectively each integrated circuit modules, or by the multiple modules in them or Step is fabricated to single integrated circuit module to realize.So, the present invention is not restricted to any specific hardware and software combination.

The foregoing is only the preferred embodiments of the present invention, be not limited to the present invention, for the skill of this area For art personnel, the present invention can have various modifications and variations.All within the spirit and principles in the present invention, made any repair Change, equivalent, improvement etc., should be included within the scope of the present invention.

Claims (11)

1. a kind of acquisition methods merging weight are it is characterised in that include；

Choose the time period of the channel guess value for calculating channel；

Calculate the channel guess value in the described time period for the described channel；

Described channel guess value is carried out with interpolation arithmetic, and receiver is obtained according to the channel guess value carrying out obtaining after interpolation arithmetic Merge weight；

Wherein, described receiver is senior receiver, and the channel guess value according to carrying out obtaining after interpolation arithmetic obtains receiver Merge weight, including：With the initial weight that merges as initial value, filtered with the channel guess value carrying out obtaining after interpolation arithmetic for self adaptation The input of ripple device, is adjusted to described initial merging weight, obtains the merging power of the described senior receiver after interpolation arithmetic Weight.

2. method according to claim 1 is it is characterised in that described receiver is Rake Rake receiver；According to carry out The channel guess value obtaining after interpolation arithmetic obtains the merging weight of receiver, including：

Conjugate operation is carried out to the channel guess value carrying out obtaining after interpolation arithmetic, obtains the merging weight of described Rake receiver.

3. method according to claim 1 is it is characterised in that described receiver is senior Rake Arake receiver；

Before carrying out interpolation arithmetic to described channel guess value, methods described also includes：Obtain the initial of described Arake receiver Merge weight.

4. method according to claim 3 it is characterised in that by following iterative process obtain interpolation arithmetic after described The merging weight of Arake receiver, including：

w(0)

E (n)=d (n)-w^H(n)u(n)

W (n+1)=w (n)+k (n) e^*(n)

Wherein, w (0) is described initial merging weight, and w (n) is the merging weight after nth iteration, w^HN () represents being total to of w (n) Yoke transposition, u (n) be carry out interpolation arithmetic after the channel guess value that obtains, d (n) is the expected value of described sef-adapting filter output, E (n) is the difference between described sef-adapting filter output and expected value is error, e^*N () represents the conjugation of e (n), k (n) is Weight renewal vector.

5. method according to claim 4 is it is characterised in that in described iterative process, methods described also includes：

When described error amount during current iteration is more than the error amount of front an iteration, regenerates described weight and update Vector is so that iterative process tends to restraining.

6. method according to claim 4, it is characterised in that described iterative process is when meeting one below condition, is jumped Go out described iterative process：

The iterationses of described iterative process reach predetermined threshold value；

Described error amount meets the predetermined condition jumping out iterative process.

7. method according to claim 1 is it is characterised in that described channel linearly became within the described time period chosen Change.

8. the method according to any one of claim 1 to 7 is it is characterised in that carry out interpolation arithmetic to described channel guess value, Including：

Interpolation arithmetic is carried out to described channel guess value using linear spline interpolation algorithm.

9. a kind of acquisition device merging weight, it is characterised in that including in receiver；

Choose module, for choosing the time period of the channel guess value for calculating channel；

Computing module, for calculating the channel guess value in the described time period for the described channel；

Interpolating module, for carrying out interpolation arithmetic to described channel guess value；

Acquisition module, for obtaining the merging weight of receiver according to the channel guess value carrying out obtaining after interpolation arithmetic；Wherein, institute Stating receiver is senior receiver, and the channel guess value according to carrying out obtaining after interpolation arithmetic obtains the merging weight of receiver, bag Include：With the initial weight that merges as initial value, to carry out the input as sef-adapting filter of the channel guess value that obtains after interpolation arithmetic, Described initial merging weight is adjusted, obtains the merging weight of the described senior receiver after interpolation arithmetic.

10. device according to claim 9 is it is characterised in that described acquisition module, is additionally operable to carrying out after interpolation arithmetic The channel guess value obtaining carries out conjugate operation, obtains the merging weight of Rake Rake receiver.

11. devices according to claim 9 are it is characterised in that described acquisition module includes：

First acquisition unit, before carrying out interpolation arithmetic to described channel guess value, obtains senior Rake Arake receiver Initial merging weight；

Second acquisition unit, for described initial merging weight as initial value, being estimated with the channel carrying out obtaining after interpolation arithmetic It is worth the input for sef-adapting filter, described initial merging weight is adjusted, obtains the described Arake after interpolation arithmetic and connect The merging weight of receipts machine.