CN101388683B - Code channel detection method in CDMA system - Google Patents

Abstract

The invention discloses a method for detecting code channels in a cdma system. The method comprises: firstly, estimating communication channels of a training sequence which receives user data, secondly, determining corresponding training sequence deviation according to the communication channel estimating results, finding out corresponding spread-spectrum codes according to configuration information, thirdly, using each spread-spectrum code in the corresponding spread-spectrum codes of the training sequence deviation to de-extend received data sections which are corresponding to taps, obtaining symbol energy, fourthly, sequencing the symbol energy according to size, wherein same with the symbol energy value which is de-extended by a minimum spread-spectrum factor, a spread-spectrum code which is corresponding to the maximum spread-spectrum factor in all spread-spectrum factors is the spread-spectrum code which is used by users. The method uses corresponding relation between the training sequence deviation and a spread-spectrum code set and de-extension of received data to realize blind detection of occupied code channel information of the users, and the transfer of the code channel information which is used by the users through a signaling message is avoided, hollow port resources are saved for systems, and the realizing complexity is small, the use is convenient, and the burden of a base station controller is lightened.

Description

A kind of method that in code division multiple access system, code channel detects

Technical field

The present invention relates to communication technique field, relate in particular to the method that in a kind of TD SDMA (TD-SCDMA) system, code channel detects.

Background technology

TD-CDMA (time division CDMA, particularly TD-SCDMA) is based on the 3-G (Generation Three mobile communication system) of time division duplex (TDD), wherein relates to a lot of key technologies, as Fast Channel estimation technique, associated detection technique, intelligent antenna technology etc.Use associated detection technique in the TD-SCDMA system, can effectively suppress multiple access and disturb (MAI) and intersymbol interference (ISI), the code channel of CU in system will be known exactly in the basis that these key technologies are used, i.e. the channel code of use.

In the TD-SCDMA system, the multi-cell joint detection algorithm need to know that adjacent cell is in the spreading code that the mobile terminal of state of activation uses, the inhibition that just can disturb.Present way is that each adjacent cell is crossed signaling with the information exchange of the excited users in own residential quarter and passed to this cell base station, has increased signaling consumption.In high speed uplink packet access (HSUPA), because UE can change automatically up spreading factor and code channel at different time slots, like this since very difficult spreading factor and the spreading code that obtains the adjacent cell cellphone subscriber by the mode of signaling in base station of this residential quarter.

Summary of the invention

Technical problem to be solved by this invention is to provide a kind of method that in code division multiple access system, code channel detects.

For solving the problems of the technologies described above, the present invention is achieved by the following technical solutions:

A kind of method that in code division multiple access system, code channel detects comprises the following steps:

A: the training sequence that will receive user data carries out channel estimating;

B: according to channel estimation results, determine corresponding training sequence deviation (Midamble shift), and find corresponding spreading code class according to the configuration information of Midamble shift;

C: respectively receive data section corresponding to tap carried out despreading with each spreading code in spreading code class corresponding to Midamble shift, obtain symbol energy;

D: symbol energy is sorted according to its size, the spreading code that in all spreading factors the same with the symbol energy values of minimal frequency spreading factor despreading, maximum spreading code corresponding to that spreading factor uses for the user.

Wherein, described in step a, training sequence can be expressed as:

e＝Gh+n

Wherein, the G matrix is the ring shift right matrix that is made of a basic Midamble code, and h is user's channel impulse response, and n is white Gaussian noise;

Obtaining channel estimating is: $\hat{h}={\left(G^{*T}G\right)}^{-1}{G}^{*T}e$

Because the G matrix is the ring shift right matrix, so obtain

$\hat{h}=G^{-1}e.$

Wherein, configuration information according to Midamble shift described in step b finds corresponding spreading code class to be specially: under Midamble code default (default) method of salary distribution, midamble code shift and the configuration mode of channel code in the different K values situation have been provided according to agreement, find out the corresponding spreading code class of midamble code shift according to this corresponding relation, wherein K is the maximum number of available training sequence code in a time slot.

Wherein, in step c, receive data section corresponding to tap being carried out despreading is specially: the transmission data vector of user k is

d _k＝(d ₁，d ₂，…，d _N)

Wherein N represents the symbolic number of each data block in each code channel;

Through spreading code c _k=(c ₁, c ₂..., c _Q)

After spread spectrum, become and expand signal S _k=d _c=(S ₁, S ₂..., S _NQ)

Obtaining symbol energy is:

${\mathrm{SE}}_{\mathrm{Qi}}={\mathrm{Sc}}_r^{*}$

$=\underset{k=1}{\overset{\mathrm{NQ}}{\mathrm{\Σ}}}{S}_k{c}_r^{*}$

$=\underset{k=1}{\overset{\mathrm{NQ}}{\mathrm{\Σ}}}{d}_k{c}_s{c}_r^{*}$

$=\underset{k=1}{\overset{\mathrm{NQ}}{\mathrm{\Σ}}}{d}_k\underset{l=1}{\overset{16/\mathrm{Qi}}{\mathrm{\Σ}}}{c}_{s,l}{c}_{r,l}^{*}$

Wherein Q represents scrambler cycle, c _rThe spreading code that expression receives, c _sThe spreading code of expression emission, ^*The expression conjugation.

The present invention has following beneficial effect:

Adopt utilize minamble shift and relation corresponding to spreading code set and the despreading of receive data of the present invention, complete the scheme of the detection of the spreading code that user uplink is used, can realize the blind Detecting of CU code channel information.Avoided like this using code channel information by signaling transmission user, for system has saved valuable interface-free resources, this scheme implementation complexity is little, easy to use, has alleviated the burden of base station controller, the simple detection of completing code channel that the user uses.

Description of drawings

Fig. 1 is the default corresponding relation figure of up channel code and up midamble code;

Fig. 2 is the inventive method flow chart.

Embodiment

Core concept of the present invention: go out the code channel situation that adjacent cell UE uses by receiving the Midamble sequencal estimation that contains in user data.At first carry out adjacent cell channel and estimate, utilize the corresponding relation of up channel code and midamble intersymbol, and in conjunction with to the user data despreading, the mode that varies in size according to the symbol energy of despreading detects the method that the user uses code channel.If the whole network uses this corresponding relation, this cell base station not to need through Iub (logic interfacing between RNC and Node B) oral instructions to pass this information, just can realize the detection of spreading code of the up use of mobile terminal of adjacent cell activation.

The present invention is described in further detail below in conjunction with drawings and Examples:

Isolate training sequence and receive signal from the data that the base station receives, complete channel estimating and channel post-processing to each user of adjacent cell.The training sequence that receives can be expressed as

e＝Gh+n (1)

Wherein, the G matrix is the ring shift right matrix that is made of a basic Midamble code (training sequence that adjacent cell is used), and h is user's channel impulse response, and n is white Gaussian noise.

Complete the maximal possibility estimation of channel impulse response h with formula (2)

Namely

$\hat{h}={\left(G^{*T}G\right)}^{-1}{G}^{*T}e---\left(2\right)$

Because the G matrix is the ring shift right matrix, have

$\hat{h}=G^{-1}e---\left(3\right)$

Therefore, utilize the cyclic shift characteristic of matrix, above-mentioned channel estimating can adopt fast fourier transform/inverse-Fourier transform (FFT/IFFT) fast method realization fast, thereby has improved significantly arithmetic speed.

As shown in the formula (4):

$\hat{h}=\mathrm{ifft}(\mathrm{fft}\left(e\right)/\mathrm{fft}\left(m\right))---\left(4\right)$

According to channel estimating

Result, have the window of tap to be the Midamble shift of activation, can know the operating position of Midamble shift.

In the foundation Midamble code default method of salary distribution, provide midamble code shift and the configuration mode of channel code in the different K values situation in agreement 25.221, found out a corresponding channel code class of midamble code shift (also claiming the spreading code class) according to this corresponding relation.

Wherein, the K value is the maximum number of available Midamble code in a time slot.

The below is take the K=8 situation as example, and in other K value situations, the method that code channel detects is identical.Specifically introduce the process of estimating code channel according to the configuration information of midamble shift.Under the Midamble code default method of salary distribution, the corresponding relation of up channel code and up midamble code has the default corresponding relation (K=8) of following up channel code class and up midamble code as shown in Figure 1 like this, as table 1:

Table 1

1

2

3

4

5

6

7

8

c 1 (1)，c 2 (1)，c 4 (1)，c 8 (1)，c 16 (1)

c 8 (2)，c 16 (3)，

c 4 (2)，c 8 (3)c 16 (5)

c 8 (4)，c 16 (7)，

c 2 (2)，c 4 (3)，c 8 (5)，c 16 (9)

c 8 (6)，c 16 (11)，

c 4 (4)，c 8 (7)，c 8 (13)

c 8 (8)，c 16 (15)

m (1)

m (2)

m (3)

m (4)

m (5)

m (6)

m (7)

m (8)

Channel code is directly to define with the code tree that generates ovsf code, the every one-level in code tree all correspondence a spreading factor Q.TD-SCDMA system requirements: if in code tree, a certain branch of certain one-level is used, must guarantee this branch left side so until all codes of root node all are not used, and on the right of this branch, the code of all subtrees can not be used again.According to such characteristics of system, in table 1 in each up channel code class, only have a spreading factor to be used in the above, the different spreading codes of the same spread factor can be used simultaneously.Therefore, in spreading code class corresponding to same midamble shift, only have a spreading code to be used by the user of current time slots.

Because spreading code in the TD-SCDMA system is elongated, therefore to carry out preliminary treatment to each user's spreading code.With spreading code expand to 16 long.If Q=2 repeats spreading code 8 times; If Q=4 repeats spreading code 4 times, the like.Spreading code expansion cross-reference, as shown in table 2:

Table 2

Spreading factor	Spreading code c before expansion i	Spreading code c after expansion i′
			Q＝1	(c 1)	(c 1，c 1，c 1，c 1，c 1，c 1，c 1，c 1，c 1，c 1，c 1，c 1，c 1，c 1，c 1，c 1)
Q＝2	(c 1，c 2)	(c 1，c 2，c 1，c 2，c 1，c 2，c 1，c 2，c 1，c 2，c 1，c 2，c 1，c 2，c 1，c 2)
			Q＝4	(c 1，c 2，c 3，c 4)	(c 1，c 2，c 3，c 4，c 1，c 2，c 3，c 4，c 1，c 2，c 3，c 4，c 1，c 2，c 3，c 4)
Q＝8	(c 1，c 2，c 3，c 4，c 5，c 6，c 7，c 8)	(c 1，c 2，c 3，c 4，c 5，c 6，c 7，c 8，c 1，c 2，c 3，c 4，c 5，c 6，c 7，c 8)
			Q＝16	(c 1，c 2，c 3，c 4，c 5，c 6，c 7，c 8，c 9，c 10，c 11，c 12，c 13，c 14，c 15，c 16)	(c 1，c 2，c 3，c 4，c 5，c 6，c 7，c 8，c 9，c 10，c 11，c 12，c 13，c 14，c 15，c 16)

If know the spreading code class that midamble shift is corresponding, then respectively tap corresponding data section is carried out despreading with each spreading code in same spreading code class, calculate possible symbol energy.The symbol energy that obtains according to despreading, then determine that the spreading code judgement that maximum is corresponding is spreading code corresponding to this tap.

Transmitting terminal, the transmission data vector of user k is

d _k＝(d ₁，d ₂，…，d _N) (5)

Wherein N represents the symbolic number of each data block in each code channel.

Through spreading code c _k=(c ₁, c ₂..., c _Q) (6)

After spread spectrum, become and expand signal

S _k＝d _c＝(S ₁，S ₂，...，S _NQ) (7)

With all spreading codes in spreading code class corresponding to midamble shift, the receive data section that tap is corresponding is carried out despreading, obtain symbol energy, use SE _Qi(symbol energy) expression, i represents the value of Q.

Carry out to received signal despreading

${\mathrm{SE}}_{\mathrm{Qi}}={\mathrm{Sc}}_r^{*}$

$=\underset{k=1}{\overset{\mathrm{NQ}}{\mathrm{\Σ}}}{S}_k{c}_r^{*}$

$=\underset{k=1}{\overset{\mathrm{NQ}}{\mathrm{\Σ}}}{d}_k{c}_s{c}_r^{*}$

$=\underset{k=1}{\overset{\mathrm{NQ}}{\mathrm{\Σ}}}{d}_k\underset{l=1}{\overset{16/\mathrm{Qi}}{\mathrm{\Σ}}}{c}_{s,l}{c}_{r,l}^{*}---\left(8\right)$

Wherein Q represents scrambler cycle, c _rThe spreading code that expression receives, c _sThe spreading code of expression emission, ^*The expression conjugation.The symbol lengths of despreading can be less than NQ, and data length is larger, and reliability is higher.

By table 1, table 2 as can be known, the code word in spreading code class corresponding to each midamble shift has such feature, expansion can generate the spreading code of little spreading factor and the large identical code word of spreading factor through simply copying again.

In conjunction with following formula, so the SE that the large spreading factor that receiving terminal is launched with little spreading factor demodulation transmitting terminal obtains is just arranged _{Little Q}The SE that obtains with the spreading factor demodulation of using with actual transmission _{Large Q}The same, otherwise but quite different, because the statistical property of symbol of emission is that 0.5 probability is { 0, send at random in the 1} set, two adjacent symbols can be different, therefore when transmitting terminal uses little spreading factor emission, when coming demodulation with large spreading factor, symbol energy has loss, SE _{Large Q}Less than SE _{Little Q}

Take spreading code class corresponding to first midamble shift as example:

Suppose that the spreading factor that the user uses is 8, so just have SE ₁₆＜SE ₈=SE ₄=SE ₂=SE ₁

Suppose that the spreading factor that the user uses is 4, so just have SE ₁₆≤ SE ₈＜SE ₄=SE ₂=SE ₁

Suppose that the spreading factor that the user uses is 16, so just have SE ₁₆=SE ₈=SE ₄=SE ₂=SE ₁

Thus, in despreading, to SE _QiSort, and the symbol energy values of minimal frequency spreading factor despreading, SE ₁In value all spreading factors equally, that maximum spreading factor is exactly the spreading factor that user's transmitting terminal uses, and the spreading code that uses this spreading factor in the midamble of correspondence shift spreading code class is exactly the channel code that the user uses.

When in system, network was set up, the concrete steps of code assignment scheme were consulted shown in Figure 2:

Step 201: the training sequence of receive data is proposed to do channel estimating;

The maximal possibility estimation of channel impulse response, $\hat{h}=G^{-1}e;$

Step 202: according to channel estimating

Result, determine the corresponding Midamble shift of theactivewindow.There is the window of tap to be the Midamble shift of activation, can arrives the operating position of Midamble shift; Under the Midamble code default method of salary distribution, provided midamble code shift and the configuration mode of channel code in the different K values situation according to agreement, find out a corresponding channel code class of midamble code shift according to this corresponding relation;

Step 203: respectively tap corresponding data section is carried out despreading with each spreading code in spreading code class corresponding to same midamble shift, calculate possible symbol energy, be designated as SE _Qi

Step 204: according to size, to SE _QiSort, and the symbol energy values of minimal frequency spreading factor despreading, SE ₁In all the same spreading factors, that maximum spreading factor is exactly the spreading factor that user's transmitting terminal uses, and the spreading code that uses this spreading factor in the midamble of correspondence shift spreading code class is exactly the channel code that the user uses; Return to step 202, until complete the detection that user under all Midamble shift uses channel code.

Adopt utilize minamble shift and relation corresponding to spreading code set and the despreading of receive data of the present invention, complete the scheme of the detection of the spreading code that user uplink is used, can realize the blind Detecting of CU code channel information.Avoided like this using code channel information by signaling transmission user, for system has saved valuable interface-free resources.

The present invention can also be used for all cdma systems of moving communicating field, particularly TDD-CDMA system and all SCDMA systems.

The above is only preferred embodiment of the present invention, not in order to limiting the present invention, all any modifications of doing within the spirit and principles in the present invention, is equal to and replaces and improvement etc., within all should being included in protection scope of the present invention.

Claims (2)

1. the method that in a code division multiple access system, code channel detects is characterized in that, comprises the following steps:

A: the training sequence that will receive user data carries out channel estimating;

B: according to channel estimation results, determine corresponding training sequence deviation, and find corresponding spreading code class according to the configuration information of training sequence deviation;

C: in the spreading code class that training sequence deviation is corresponding, each spreading code carries out the extended length preliminary treatment, respectively receive data section corresponding to tap is carried out despreading with each spreading code after expansion, obtains symbol energy;

D: symbol energy is sorted according to its size, the spreading code that in all spreading factors the same with the symbol energy values of minimal frequency spreading factor despreading, maximum spreading code corresponding to that spreading factor uses for the user;

Described in step a, training sequence is expressed as:

e＝Gh+n

Wherein, the G matrix is the ring shift right matrix that is made of a basic training sequences code, and h is user's channel impulse response, and n is white Gaussian noise;

Obtaining channel estimating is: $\hat{h}={\left(G^{*T}G\right)}^{-1}{G}^{*T}e$

Because the G matrix is the ring shift right matrix, so obtain

$\hat{h}=G^{-1}e;$

According to channel estimation results, determine that corresponding training sequence deviation is specially in step b: according to channel estimation results, have the window of tap to be the training sequence deviation of activation;

In step c, receive data section corresponding to tap being carried out despreading is specially: the transmission data vector of user k is

d _k＝(d ₁，d ₂，…，d _N)

Wherein N represents the symbolic number of each data block in each code channel;

Through spreading code c _k=(c ₁, c ₂..., c _Q)

After spread spectrum, become and expand signal S _k=dc=(S ₁, S ₂..., S _NQ)

Obtaining symbol energy is:

${\mathrm{SE}}_{\mathrm{Qi}}={\mathrm{Sc}}_r^{*}$

$=\underset{k=1}{\overset{\mathrm{NQ}}{\mathrm{\Σ}}}{S}_k{c}_r^{*}$

$=\underset{k=1}{\overset{\mathrm{NQ}}{\mathrm{\Σ}}}{d}_k{c}_s{c}_r^{*}$

$=\underset{k=1}{\overset{\mathrm{NQ}}{\mathrm{\Σ}}}{d}_k\underset{l=1}{\overset{16/\mathrm{Qi}}{\mathrm{\Σ}}}{c}_{s,l}{c}_{r,l}^{*}$

Wherein Q represents scrambler cycle, c _rThe spreading code that expression receives, c _sThe spreading code of expression emission, * represents conjugation.

2. the method that in code division multiple access system as claimed in claim 1, code channel detects, it is characterized in that, configuration information according to training sequence deviation described in step b finds corresponding spreading code class to be specially: under the training sequence code default distributing mode, provide training sequence deviation and the configuration mode of channel code in the different K values situation according to agreement, find out the corresponding spreading code class of training sequence deviation according to this corresponding relation, wherein K is the maximum number of available training sequence code in a time slot.

Images