CN1492607A - Speed matching and parameter optimizing method of wide band CDMA cut-ni system - Google Patents

Abstract

A method of rate matched parameters optimization for a broad band CDMA access system is to get the maximum rate matched parameter combination for each channel according to code back BER limitation of every channel, the said parameter combination is got by derivation of a set of equations made up of channel rate matched gain equations. The method can get rate matched factors of the transmission channels in quantity so that the compound channel CCtr CH performance can meet the different performance requirements of transmission channels joining in complexion at the same time, unbalanced situation will not appear, when this set of rate matched factors is applied for rate matched factors is applied for rate match.

Description

Speed matching and parameter optimizing method of wide band CDMA cut-ni system

Technical field:

The present invention relates to a kind of at Wideband Code Division Multiple Access (WCDMA) access (Wideband Code DivisionMultiple Access, WCDMA) method of optimal parameters, more particularly: be the method for distributing transmission channel RM in the WCDMA system (rate-matched, Rate Matching) parameter rationally.

Background technology:

In the uplink and downlink flow process of WCDMA system physical layer base band flow process, the step of rate-matched is arranged all.In the different TTI in same transmission channel (Transmission Time Interval, Transport TimeInterval), the size of transmission code block and the number of code block all may change.If directly with each bar transmission channel compound be a physical channel, can not guarantee all the time that the CCtrCH (sign indicating number multiplexing transmission information channel Coded Composite Transport Channel) of compound formation can be mapped on the physical channel according to the frame format of stipulating in the agreement; That is to say: the bit number that can comprise in a frame on the physical channel can be more inconsistent than the number that compound channel requires it to send in a frame.Therefore must control the compound process of channel.

Repeat or the dynamic size of adjusting code stream of bit stream punching by bit, make the CCtrCH channel that obtains at last can correctly be mapped to physical channel.In the repetition and punch operation of reality, operated bit can be scattered in the whole transmission channel as far as possible, to guarantee the minimum that influences to decoding.The process of descending and upstream rate coupling is in the computation rate matching operation, promptly destroys or repeats, and the formula of bit number is consistent, and the computational methods of regulation are shown below among WCDMA agreement 3GPP TS 25.212 V4.1.0 (2001-01):

Wherein the value of i from 1 to I

Δ N _i=Z _i-Z _I-1-N ₁, wherein, i=1...I

Symbol description in the formula is as follows:

I is the transmission channel that comprises among the CCtrCH;

N _DataFor according to the bit number that should comprise by each radio frames of physical transfer requirement for restriction compound channel CCtrCH;

N _mIt is m bar transmission channel bit number in the corresponding 10ms radio frames before rate-matched;

RM _mIt is the parameters of rate matching of m bar transmission channel;

Z _IBe an intermediate object program;

Δ N _I, repeats the i bar transmission channel that calculates the bit number of (if value just is) or punching (if value is for negative) for needing the through-rate coupling.

From above-mentioned formula as can be seen: the rate matching attribute actual influence to the bit rate that how CCtrCH required be assigned on each transmission channel.Rate matching attribute parameter how to set different transmission channels is not described in the agreement of WCDMA.If can not just can not obtain the compound channel of optimum performance according to the balanced speed of distributing of transmission channel performance requirement to different transmission channels.

Summary of the invention:

Main purpose of the present invention is to provide a kind of method of optimizing the transmission channel parameters of rate matching, take all factors into consideration the different performance requirement of each transmission channel, utilize rational parameters of rate matching to carry out rate adaptation operating, the balanced speed of distributing is to different transmission channels, thus acquisition best complex channel performance; And can calculate parameters of rate matching value separately quantitatively according to the performance requirement of each compound bar transmission channel.

The object of the present invention is achieved like this:

Calculate the error rate restriction of each compound in base band flow process bar transmission channel.If channel quality requires to provide with the form of error rate restriction, can directly use it.If but the channel quality requirement is the form of Block Error Rate or frame error rate restriction, just go out corresponding error rate restriction according to the following derivation of equation.

$\mathrm{BER}=-\frac{\ln (1-\mathrm{ERROR}\_\mathrm{Ratio})}{\mathrm{size}}$

Wherein,

BER (bit error rate, Bit Error Ratio) is error rate restriction to be asked;

ERROR_Ratio is given Block Error Rate restriction or frame error rate restriction;

Size is the size of a radio frames of code block size or transmission channel correspondence.

If multiple possible code block size is arranged in the format set of this transmission channel, then should calculate according to the probability that these different-format collection occur in this transmission channel, obtain the radio frames size of an expectation code block size or expectation.

Error correction coding gain according to different transmission channels is required to obtain the error rate requirement of the later transmission channel of process coding by the error rate of transmission channel.

In the present invention, the error rate before the transmission channel coding requires to be the error rate before the coding, and the error rate of the transmission channel behind the coding requires to be expressed as the coding back error rate.Because coding gain, the coding back error rate is greater than the preceding error rate of coding.

According to the error rate behind the coding of each multiplexing transmission channel, find the solution the equation group that the rate-matched gain equation by each channel constitutes, obtain to satisfy the parameters of rate matching combination of the best of each multiplexing transmission channel performance.

The form of the rate-matched gain equation of channel is as follows: if channel is to obtain postiive gain by the bit repetitive operation, corresponding gain equation is:

$Q^{\′}=Q(1-\frac{1}{2}*\frac{\mathrm{\ΔN}}{N}),\mathrm{if\ΔN}>0$

If opposite, channel carries out the bit punch operation and obtains negative gain, and corresponding gain equation is:

$Q^{\&prime;}=Q(1+\frac{\mathrm{\&Delta;N}}{N})-\frac{1}{2}*\frac{\mathrm{\&Delta;N}}{N},\mathrm{if\&Delta;N}<0$

In the above-mentioned formula,

Q is the channel bit error rate after the rate adaptation operating;

Q ' is the channel bit error rate before the rate adaptation operating;

Δ N is the rate adaptation operating bit number of channel, its on the occasion of the time represent repetition bits operation, it represents that punching bit operates during for negative;

N is the sign indicating number section bit number before the transmission channel rate-matched.

The repetitive operation rate-matched gain equation form that operation obtains with punching bit is different.Can carry out the repetition bits operation owing to can't determine which transmission channel before parameters of rate matching designs, which carries out the punching bit operation, so the form of equation group is indefinite; Concrete method for solving is:

A) suppose that all channels all can carry out repetitive operation, make up the equation group that the rate-matched gain equation by the postiive gain form of all channels constitutes and find the solution.

B) group of solving an equation is to carry out the punching bit operation or the hypothesis of repetition bits operation if all Δ N that solve meet the equation group of structure to each channel, then begins to carry out d) step; Otherwise carry out c) step.

C) channel setting of correspondence the maximum bit error rate restriction makes up new equation group and execution in step b for carrying out punch operation (because its performance requirement is minimum));

D) release the parameters of rate matching of each channel by the Δ N value of each channel.

Use method of the present invention, can obtain the rate matching attribute of transmission channel quantitatively, when having guaranteed to adopt this group rate matching attribute to carry out rate-matched, can make the performance of compound channel CCtrCH satisfy the multiplexing different performance requirement of transmission channel of participation simultaneously, and unbalanced situation can not occur.That is to say: reduced each transmission channel its quality requirements during wireless transmission aloft.The present invention can reduce channel transmitting power, has in fact just improved the capacity of system.

Description of drawings:

Fig. 1 is the present invention 1/2 speed convolutional encoding error correction gain matched curve figure.

Fig. 2 is the present invention 1/3 speed convolutional encoding error correction gain matched curve figure.

Fig. 3 is Turbo code correction gain matched curve figure of the present invention.

Fig. 4 is an overview flow chart of the present invention.

Fig. 5 for the coding that calculates each transmission channel among the present invention after the flow chart of error rate restriction.

Fig. 6 finds the solution flow chart for the rate-matched gain equation of channel among the present invention.

Embodiment:

The present invention is described in further detail below in conjunction with accompanying drawing and specific embodiment:

Transmission channel parameters of rate matching method for designing is divided into following step in the system provided by the invention:

At first, calculate the error rate restriction that in the base band flow process, participates in each multiplexing bar transmission channel.

The error rate restriction of transmission channel has reflected the performance requirement of channel; But actual channel property parameter might be to provide by the form that Block Error Rate and frame error rate limit, in this step, and need be with they unified forms that be converted to error rate restriction.

Derive the below formula of this conversion:

The define symbol variable:

Err: for the given error rate of sign indicating number section form (code block or frame);

Ber: the error rate;

Size: the length of given sign indicating number section;

X: code stream length;

ErrNum: error code section counting.

The utilization statistical theory can draw: $\mathrm{Err}=\underset{x\&RightArrow;\&infin;}{\lim }\frac{\mathrm{ErrNum}}{x/\mathrm{size}}$

Definition num=x/size, it is that length is the sign indicating number hop count on the code stream of x.Length is that the errored bit number on the code stream of x is x * Ber.According to probability theory, when an error bit position was randomized into num sign indicating number section and goes up, wherein not to be assigned to the probability of this errored bit be (1-1/num) to certain yard section, and then (x * Ber) inferior is not divided to advance the probability of error bit to be to certain yard section continuously ${\left(\frac{\mathrm{num}-1}{\mathrm{num}}\right)}^{x\&CenterDot;\mathrm{Ber}},$ So most probable error code hop count order is:

$\mathrm{ErrNum}=\mathrm{num}*(1-{\left(\frac{\mathrm{num}-1}{\mathrm{num}}\right)}^{x\&CenterDot;\mathrm{Ber}})$

Thus, so can derive:

$\mathrm{Err}=\underset{x\&RightArrow;\&infin;}{\lim }\frac{\mathrm{ErrNum}}{\mathrm{num}}=\underset{x\&RightArrow;\&infin;}{\lim }(1-{\left(\frac{\mathrm{num}-1}{\mathrm{num}}\right)}^{x\&CenterDot;\mathrm{Ber}})=\underset{x\&RightArrow;\&infin;}{\lim }(1-{\left(\frac{x-\mathrm{size}}{x}\right)}^{x\&CenterDot;\mathrm{Ber}})=1-e^{-\mathrm{size}\&CenterDot;\mathrm{Ber}}$

It is as follows that thereby the mistake that obtains transforming sign indicating number section form is restricted to the formula of error rate limited form:

$\mathrm{Ber}=-\frac{\ln (1-\mathrm{Err})}{\mathrm{size}}$

If multiple possible sign indicating number section size is arranged in the format set of this transmission channel, therefore the probability calculation that should occur in this transmission channel according to these different size sign indicating number sections earlier obtains an expected code section size, brings above-mentioned formula again into and tries to achieve error rate restriction.

Then, according to error rate restriction behind the coding of each transmission channel of error correction coding gain calculating of different transmission channels.

The error correction coding gain of considering each transmission channel is to obtain their performance requirements before rate adaptation operating, and just the coding back error rate limits.

Regulation has three kinds of coded systems in WCDMA agreement 3GPP TS 25.212 V4.1.0 (2001-01), and they are respectively: the convolutional encoding of 1/2 speed, the convolutional encoding of 1/3 speed and Turbo coding.The gain of these coded systems has very big difference.The method that obtains the gain of certain transmission channel coding the most accurately is that the parameter designing simulated program according to this channel calculates statistics.But be difficult in practice realizing.A simple and practical method is transmission channel as a reference of design, serves as that l-G simulation test is carried out to three kinds of different coded systems stipulating in the agreement in the basis with this channel, obtains the coding gain result of the test.When the error rate limits after needs are asked the coding of certain transmission channel, only need to carry out interpolation and just can obtain according to the data in the form.Reference channel has following characteristics:

1) the supposition error bit is evenly distributed on the whole transmission channel code stream behind the coding.This also conforms to the situation of reality, holds because the flow process that interweaves in the physical layer can make the error code distribution disperse.

2) decoding algorithm is hard decision.

3) be to be that 400 random binary code stream has carried out 2000 coding and decoding tests to length to the convolutional encoding of 1/2 speed and 1/3 speed gain l-G simulation test among the present invention, the accuracy of the bit-error rate results of acquisition should be on the 1e-4 accuracy grade; The Turbo coding gain

L-G simulation test is to be that 1000 random bit stream has carried out 100,000 coding and decoding tests to length.

The accuracy of the bit-error rate results that obtains should be on the 1e-6 accuracy grade.

Gain data under three kinds of coded systems that emulation obtains is listed as follows:

Table 1

1/2 convolution		1/3 convolution		Turbo
						The coding back error rate	The error rate before the coding	The coding back error rate	The error rate before the coding	The coding back error rate	The error rate before the coding
0.0200	2.0875e-004	0.0200	0.0175e-003	0.07	5.54e-6
						0.0240	3.5000e-004	0.0240	0.0288e-003	0.073	6e-6
0.0280	4.5125e-004	0.0280	0.0487e-003	0.075	6.73e-6
						0.0320	7.6500e-004	0.0320	0.0588e-003	0.08	1.28e-5
0.0360	8.9000e-004	0.0360	0.0838e-003	0.085	2.38e-5
						0.0400	1.550e-003	0.0400	0.1138e-003	0.09	5.5e-5
0.0440	2.4250e-003	0.0440	0.1800e-003	0.095	1.307e-4
						0.0480	2.7912e-003	0.0480	0.2275e-003	0.1	3.767e-4
0.0520	4.3338e-003	0.0520	0.1925e-003	0.1050	1.0703e- 003
						0.0560	5.7188e-003	0.0560	0.3212e-003	0.1100	2.9606e- 003
0.0600	8.3825e-003	0.0600	0.3838e-003	0.1150	7.6783e- 003
								0.0640	5.1500e-004
		0.0680	6.0000e-004
								0.0720	7.5625e-004
		0.0760	8.5500e-004
								0.0800	1.3113e-003
		0.0840	1.5612e-003
								0.0880	1.9338e-003
		0.0920	2.5525e-003

Referring to Fig. 1, Fig. 2, Fig. 3, the gain curve of three kinds of coded systems that the data fitting in its table 1 of serving as reasons obtains.

Further, the error rate behind the coding of each transmission channel of substitution is found the solution the equation group that the rate-matched gain equation by channel constitutes, thereby obtains the parameters of rate matching combination of each best channel.

For convenience of description, the symbolic variable that is defined as follows:

N _iBeing i bar transmission channel rate adaptation operating bit number, if positive number is just carried out repetitive operation, is that negative is then carried out punch operation;

N _iIt is i bar transmission channel bit number in the corresponding radio frames before rate-matched;

RM _iIt is the parameters of rate matching of i bar transmission channel;

N _DataBe the bit number in the last radio frames of CCtrCH;

I is for participating in multiplexing transmission channel number;

Q _iIt is error rate restriction behind the coding of i bar transmission channel;

Q is the error rate restriction of CCtrCH;

X _iBe the rate of change of i bar transmission channel rate adaptation operating, be Δ N _i/ N _I

Puncture is an intermediate variable, the channel bar number of the execution punch operation of current hypothesis.

The computational methods of each transmission channel operation bit number in the rate-matched link, can release by the formula among WCDMA agreement 3GPP TS 25.212 V4.1.0 (2001-01):

${\mathrm{\&Delta;N}}_i=\frac{N_i\&CenterDot;{\mathrm{RM}}_i\&CenterDot;N_{\mathrm{data}}}{\underset{j=1}{\overset{I}{\mathrm{\&Sigma;}}}{N}_j\&CenterDot;{\mathrm{RM}}_j}-N_i,i=\mathrm{1,2},...I$

In addition, the rate adaptation operating according to channel is that bit repeats or the bit punching should be carried out according to following rule:

If the bit repetitive operation, the statistics error rate after the rate-matched is Q; The statistics error rate before the rate-matched is Q '.In general Δ N _i＜N _i, that is to say not have to be repeated once above bit.The demodulation result of repetition bits merge make the probable bit error reduction half, so the gain equation that obtains under the following bit repetitive operation is:

$Q^{\&prime;}=Q(1-\frac{1}{2}*\frac{\mathrm{\&Delta;}{N}_i}{N_i}),\mathrm{if\&Delta;}{N}_i>0$

If the bit punch operation, the bit of being struck off can be 0 or 1 by arbitrarily mending the recipient, and the possibility of mistake is 1/2.So the gain equation under the punch operation is:

$Q^{\&prime;}=Q(1+\frac{\mathrm{\&Delta;}{N}_i}{N_i})-\frac{1}{2}*\frac{\mathrm{\&Delta;}{N}_i}{N_i},\mathrm{if\&Delta;}{N}_i<0$

Because can't judge which channel need carry out the bit punch operation before the design speed match parameter, which channel need carry out the bit repetitive operation, so can't determine the rate-matched gain equation form of each channel.Therefore program circuit is that the process of checking is then found the solution in an exploration.

Referring to Fig. 4, it is an overview flow chart of the present invention, comprises following step:

Step 1: calculate the error rate restriction that in the base band flow process, participates in each multiplexing bar transmission channel;

Step 2:, calculate error rate restriction behind the coding of each transmission channel according to the gain of the error correction coding of different transmission channels;

Step 3: according to the error rate behind the coding of each transmission channel, find the solution the equation group that the rate-matched gain equation by channel constitutes, obtain the parameters of rate matching combination of each best channel.

Referring to Fig. 5, its for the coding that calculates each transmission channel among the present invention after the flow chart of error rate restriction, comprising: at first, a reference transmission channel is set; Then,, there are three kinds of coded systems to carry out l-G simulation test, obtain coding gain test data table regulation among WCDMA agreement 3GPP TS 25.212 V4.1.0 (2001-01) based on this reference channel; At last, according to this coding gain test data table, interpolation is carried out in error rate restriction behind the coding of corresponding transmission channel.

Referring to Fig. 6, it is the flow chart of the rate-matched gain equation method for solving of channel among the present invention, comprising:

Step 1: suppose that all channels all can carry out repetitive operation, make up the equation group that the rate-matched gain equation by the postiive gain form of all channels constitutes and find the solution;

Step 2: the group of solving an equation is to carry out the punching bit operation or the hypothesis of repetition bits operation if all Δ N that solve meet the equation group of structure to each channel, then begins execution in step 4; Otherwise execution in step 3;

Step 3: the channel setting of correspondence the maximum bit error rate restriction makes up new equation group and execution in step 2 for the execution punch operation;

Step 4:, calculate the parameters of rate matching of each channel by the Δ N value of each channel according to the computing formula of each transmission channel rate adaptation operating bit number among WCDMA agreement 3GPP TS 25.212 V4.1.0 (2001-01).

Thereby method provided by the invention can accurately dispose each parameters of rate matching value that participates in multiplexing transmission channel and make that the error rate restriction of compound channel is maximum.

It should be noted that at last: above embodiment only in order to the explanation the present invention and and unrestricted technical scheme described in the invention; Therefore, although this specification has been described in detail the present invention with reference to each above-mentioned embodiment,, those of ordinary skill in the art should be appreciated that still and can make amendment or replacement to the present invention with being equal to; And all do not break away from the technical scheme and the improvement thereof of the spirit and scope of the present invention, and it all should be encompassed in the middle of the claim scope of the present invention.

Claims (4)

1, a kind of speed matching and parameter optimizing method of wide band CDMA cut-ni system is characterized in that: according to error rate restriction behind the coding of each transmission channel, obtain the parameters of rate matching combination of each best channel.

2, speed matching and parameter optimizing method of wide band CDMA cut-ni system according to claim 1, it is characterized in that: the combination of the parameters of rate matching of each channel of described the best, the equation group that constitutes by the rate-matched gain equation of finding the solution by channel obtains; The construction method of this equation group is:

When channel obtains postiive gain by the bit repetitive operation:

$Q^{\&prime;}=Q(1-\frac{1}{2}\&times;\frac{\mathrm{\&Delta;N}}{N}),\mathrm{\&Delta;N}>0$ Perhaps,

When channel carries out the negative gain of bit punch operation acquisition:

$Q^{\&prime;}=Q(1+\frac{\mathrm{\&Delta;N}}{N})-\frac{1}{2}\&times;\frac{\mathrm{\&Delta;N}}{N},\mathrm{\&Delta;N}<0$

Wherein:

Q is the channel bit error rate after the rate adaptation operating;

Q ' is the channel bit error rate before the rate adaptation operating;

Δ N is the rate adaptation operating bit number of channel, its on the occasion of the time represent repetition bits operation, it represents that punching bit operates during for negative;

N is the sign indicating number section bit number before the transmission channel rate-matched.

3, the method for transmission channel rate-matched in the WCDMA access system according to claim 1 is characterized in that: the structure and the method for solving of described channel speed coupling gain equation are:

Step 1: all channels of initial construction are all carried out the equation group of repetitive operation, promptly make up the equation group that the rate-matched gain equation by the postiive gain form of all channels constitutes and find the solution;

Step 2: the group of solving an equation, if be to carry out the punching bit operation to each channel when all Δ N that solve meet the equationof structure group, the still setting operated of repetition bits is then to begin execution in step 4; Otherwise execution in step 3;

Step 3: the channel setting of correspondence the maximum bit error rate restriction makes up new equation group and execution in step 2 for the execution punch operation according to new setting;

Step 4:, calculate the parameters of rate matching of each channel by the Δ N value of each channel according to the computing formula of each transmission channel rate adaptation operating bit number among WCDMA agreement 3GPP TS 25.2 12 V4.1.0 (2001-01).

4, speed matching and parameter optimizing method of wide band CDMA cut-ni system according to claim 1 is characterized in that: when channel quality required form with Block Error Rate or frame error rate restriction to provide, the following formula of corresponding error rate restricted passage calculated:

$\mathrm{BER}=-\frac{\ln (1-\mathrm{ERROR}\_\mathrm{Ratio})}{\mathrm{size}}$

Wherein,

BER (bit error rate, Bit Error Ratio) is error rate restriction to be asked;

ERROR_Ratio is given Block Error Rate restriction or frame error rate restriction;

Size is the size of a radio frames of code block size or transmission channel correspondence.

Images