CN105207753A - Block error rate measurement method, block error rate measurement system and power control system - Google Patents

Abstract

The application provides a block error rate measurement method, a block error rate measurement system and a power control system. Every bit of soft information is acquired by processing the received data signals according to the working mode of a current power control system baseband; then a PDF estimation value is obtained by utilizing the statistical characteristic of a soft information probability density function based on a kernel function; and then a block error rate estimation value is obtained according to every bit of corresponding soft information and the PDF estimation value through calculation. The preset threshold in hierarchical statistics in the prior art is not required, and block error rate estimation accuracy and signal-to-noise (interference) ratio level are decoupled so that block error rate can be directly used as a power control regulation indicator and accuracy can be enhanced. Meanwhile, loopback measurement of block error rate is not required so that the requirement of a self-adaption system for timeliness can be met.

Description

A kind of error rate measurement method, error rate measurement system and power control system

Technical field

The present invention relates to power control techniques field, particularly relate to a kind of error rate measurement method, error rate measurement system and power control system.

Background technology

The error rate (BLockErrorRate, BLER) measuring technique is mainly used in the outer shroud control of close-loop power control, quality of service (QualityofService can be directly reflected because BLER measures, QoS), so make described close-loop power control directly can carry out power control based on error rate measurement, avoid the uncertainty by signal to noise ratio reflection quality of service, but greatly cause real-time poor because of its sample requirement amount.

A kind of estimating method of the error rate based on cyclic redundancy check (CRC) code (CyclicRedundancyCode, CRC) is there is in prior art.The method is applicable to use in the system of CRC, has certain real-time online.But, it estimates that accuracy depends on the threshold value preset in hierarchical statistics, namely accuracy and signal to noise ratio rank (error rate) close coupling, still need great amount of samples in good signal to noise situations, causes its accuracy and real-time to need certain choice.

Summary of the invention

In view of this, the invention provides a kind of error rate measurement method, error rate measurement system and power control system, to solve the problem that in prior art, accuracy and real-time can not ensure simultaneously.

To achieve these goals, the technical scheme that provides of the embodiment of the present invention is as follows:

A kind of error rate measurement method, comprising:

The mode of operation of the data-signal received according to current power control system base band is processed, obtains the Soft Inform ation that each bit is corresponding;

According to current power control system modulation system and planisphere characteristic, initialization is carried out to PDF (ProbabilisticDensityFunction, the Soft Inform ation probability density function) estimated parameter based on kernel function;

According to described PDF estimated parameter and the very big algorithm of statistics, the Soft Inform ation corresponding to each bit described carries out PDF blind estimate, calculates PDF estimated value;

The Soft Inform ation corresponding according to each bit described and described PDF estimated value calculate bit error rate estimation value.

Preferably, described according to described current described power control system modulation system and planisphere characteristic, initialization is carried out to the PDF estimated parameter based on kernel function and comprises:

The Soft Inform ation corresponding to each bit described according to current described power control system planisphere characteristic is classified, and adds up the categorical measure of Soft Inform ation corresponding to each bit described and element number of all categories;

Calculate described of all categories in the prior probability of Soft Inform ation;

Calculate the smoothing parameter that the PDF based on kernel function estimates.

Preferably, described in calculate the smoothing parameter that the PDF based on kernel function estimates, comprising:

When the passage receiving described data-signal is Gaussian channel, then basis calculate the smoothing parameter that the PDF based on kernel function estimates;

Wherein, σ _jfor Gaussian function (N (μ, σ ²)) in a variable, N _jfor the element number that the Soft Inform ation in described PDF estimated parameter is of all categories, j is the quantity of described Soft Inform ation classification.

Preferably, described in calculate the smoothing parameter that the PDF based on kernel function estimates, comprising:

When the passage receiving described data-signal is Unknown Channel, then basis ${\mathrm{SP}}_{optimal}={\left(\frac{A\left(K\right)}{N\×B{\left(K\right)}^2\×C\left({\hat{f}}_X\right)}\right)}^{1/5}={\left(\frac{\∫{\left(K\left(t\right)\right)}^{2}dt}{N{\left(\∫t^{2}K\left(t\right)dt\right)}^2\∫{\left({\hat{f}}_X^{\′\′}\left(t\right)\right)}^{2}dt}\right)}^{1/5}$ Calculate the smoothing parameter that the PDF based on kernel function estimates;

Wherein, A and B is respectively constant, and N is the amount of bits of the described data-signal received, and K (.) is described kernel function, f _xfor function.

Preferably, described according to described PDF estimated parameter and the very big algorithm of statistics, the Soft Inform ation corresponding to each bit described carries out PDF blind estimate, calculates PDF estimated value and comprises:

S301, judge whether current iteration number of times t is greater than described maximum iteration time T;

If described current iteration number of times t is less than or equal to described maximum iteration time T;

S302, calculate the posterior probability of the Soft Inform ation in current Soft Inform ation classification;

Calculate posterior probability described in S303, basis, calculate and upgrade described PDF estimated parameter;

S304, according to the PDF estimated parameter after described renewal, calculate and upgrade described PDF estimated value;

Make t=t+1, repeat S301-S304, until current iteration number of times t is greater than described maximum iteration time T;

Wherein, the posterior probability calculating of a Soft Inform ation and the equal corresponding a kind of Soft Inform ation classification of calculating of a PDF estimated parameter.

Preferably, described according to described PDF estimated parameter and the very big algorithm of statistics, the Soft Inform ation corresponding to each bit described carries out PDF blind estimate, calculates PDF estimated value, comprising:

According to ${\mathrm{PDF}}_{X_i}={\hat{f}}_X\left(x\right)=\frac{1}{N\×SP}\underset{X_i}{\Σ}K\left(\frac{x-X_i}{SP}\right)$ Calculate PDF estimated value;

Wherein, X _ifor the Soft Inform ation (1≤i≤N of each bit described, N is positive integer), x is the independent variable of the Soft Inform ation of each bit described, SP is the smoothing parameter in described PDF estimated parameter, K (.) is described kernel function, and N is the element number of this kind of Soft Inform ation classification in described PDF estimated parameter.

Preferably, the Soft Inform ation that described in described basis, each bit is corresponding and described PDF estimated value calculate bit error rate estimation value, comprising:

Interval according to the error distribution on current described power control system planisphere, divide and defining integration interval;

Described bit error rate estimation value is calculated according to described integrating range and described PDF estimated value.

A kind of error rate measurement system, is applied to the power control system in radio digital communication system, comprises:

Decoding unit, for processing the mode of operation of the data-signal received according to current described power control system base band, obtains the Soft Inform ation of each bit;

PDF parameter generating unit, for according to current described power control system modulation system and planisphere characteristic, carries out initialization to the PDF estimated parameter based on kernel function;

PDF estimation unit, for according to described PDF estimated parameter and statistics very big algorithm, the Soft Inform ation corresponding to each bit described carries out PDF blind estimate, calculates PDF estimated value;

Bit error rate estimation unit, for calculating bit error rate estimation value according to Soft Inform ation corresponding to each bit described and described PDF estimated value.

A kind of power control system, be applied to radio digital communication system, described power control system comprises:

Decoding unit, for processing the mode of operation of the data-signal received according to current described power control system base band, obtains the Soft Inform ation of each bit;

PDF parameter generating unit, for according to current described power control system modulation system and planisphere characteristic, carries out initialization to the PDF estimated parameter based on kernel function;

PDF estimation unit, for according to described PDF estimated parameter and statistics very big algorithm, the Soft Inform ation corresponding to each bit described carries out PDF blind estimate, calculates PDF estimated value;

Bit error rate estimation unit, for calculating bit error rate estimation value according to Soft Inform ation corresponding to each bit described and described PDF estimated value;

Target comparing unit, for comparing described bit error rate estimation value and error rate target gate, generates comparative result;

Angle detecting unit, for detecting the direction that current power controls;

Algorithm setting unit, for regulating algorithm to arrange according to the external circule power control step parameter of current described power control system;

Adjustment unit, for according to described comparative result, described current power control direction and described algorithm arrange, generate and output power command bit.

The application provides a kind of error rate measurement method, by processing the mode of operation of the data-signal received according to current described power control system base band, obtains the Soft Inform ation of each bit; Recycle the statistical property based on the Soft Inform ation probability density function of kernel function, calculate PDF estimated value; Then described bit error rate estimation value is calculated according to Soft Inform ation corresponding to each bit described and described PDF estimated value, without the need to the threshold value preset in hierarchical statistics in prior art, make the accuracy of bit error rate estimation and noise (doing) than rank (error rate) decoupling zero, thus realize directly using the error rate as power control regulating index, improve accuracy; Meanwhile, the error rate is measured without the need to loopback, meets the requirement of Adaptable System for real-time.

Accompanying drawing explanation

In order to be illustrated more clearly in the embodiment of the present invention or technical scheme of the prior art, be briefly described to the accompanying drawing used required in embodiment or description of the prior art below, apparently, accompanying drawing in the following describes is only embodiments of the invention, for those of ordinary skill in the art, under the prerequisite not paying creative work, other accompanying drawing can also be obtained according to the accompanying drawing provided.

A kind of error rate measurement method flow diagram that Fig. 1 provides for the embodiment of the present application;

Another error rate measurement method flow diagram that Fig. 2 provides for another embodiment of the application;

Another error rate measurement method flow diagram that Fig. 3 provides for another embodiment of the application;

Another error rate measurement method flow diagram that Fig. 4 provides for another embodiment of the application;

A kind of Soft Inform ation sample probability density schematic diagram that Fig. 5 provides for another embodiment of the application;

A kind of error rate measurement system configuration schematic diagram that Fig. 6 provides for another embodiment of the application;

A kind of power control system structural representation that Fig. 7 provides for another embodiment of the application;

Another power control system structural representation that Fig. 8 provides for another embodiment of the application.

Embodiment

Below in conjunction with the accompanying drawing in the embodiment of the present invention, be clearly and completely described the technical scheme in the embodiment of the present invention, obviously, described embodiment is only the present invention's part embodiment, instead of whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art, not making the every other embodiment obtained under creative work prerequisite, belong to the scope of protection of the invention.

The invention provides a kind of error rate measurement method, be applied to the power control system in radio digital communication system, poor to solve real-time in prior art, accuracy is lower, or needs the problem of reception data being carried out to recompile.

Concrete, as shown in Figure 1, described error rate measurement method comprises:

S101, the mode of operation of the data-signal received according to current power control system base band to be processed, obtain the Soft Inform ation that each bit is corresponding;

The process that the described data-signal to receiving carries out according to the mode of operation of current described power control system base band is corresponding with described mode of operation.

Concrete, the process that the described data-signal to receiving carries out according to the mode of operation of current described power control system base band comprises: synchronous, demodulation code operations.

In concrete practical application, described process can also comprise the operations such as deinterleaving, decorrelation, despreading, herein and be not specifically limited.

S102, according to current power control system modulation system and planisphere characteristic, initialization is carried out to the PDF estimated parameter based on kernel function (KernelFunction);

Preferably, to the formula that the PDF based on kernel function estimates be:

${\mathrm{PDF}}_{X_i}={\hat{f}}_X\left(x\right)=\frac{1}{N\×SP}\underset{X_i}{\Σ}K\left(\frac{x-X_i}{SP}\right)$

Described formula is to the PDF estimation formulas based on kernel function for a kind of Soft Inform ation classification; Wherein, X _ifor the Soft Inform ation (1≤i≤N) of each bit described, x is the independent variable of the Soft Inform ation of each bit described, SP is the smoothing parameter in described PDF estimated parameter, K (.) can be described kernel function, and N is the element number of this kind of Soft Inform ation classification in described PDF estimated parameter.

What deserves to be explained is, described kernel function can be Epanechnikov, Biweight, Gaussian or other kernel functions, herein and be not specifically limited, depending on the applied environment that it is concrete.

Initialized described PDF estimated parameter is needed to comprise: C _j, α _j, N _jand SP _j.Wherein, C _jfor the Soft Inform ation classification in described PDF estimated parameter, N _jfor the Soft Inform ation in described PDF estimated parameter is classified element number of all categories, α _jfor the Soft Inform ation prior probability in described PDF estimated parameter, SP _jfor the smoothing parameter of each Soft Inform ation classification in described PDF estimated parameter, wherein j is the quantity of described Soft Inform ation classification.

In concrete practical application, the quantity j of described Soft Inform ation classification depends on the constellation point number of ideally planisphere; N _j, α _jand SP _jcomposition Density Estimator parameter set Ω _j, Ω _j={ N _j, α _j, SP _j}

S103, according to described PDF estimated parameter with add up very big algorithm (Expectation-MaximizationAlgorithm), the Soft Inform ation corresponding to each bit described carries out PDF blind estimate, calculates PDF estimated value;

In concrete practical application, according to above-mentioned formula, continuous iteration and renewal are carried out to described PDF estimated parameter and described PDF estimated value, until be met the described PDF estimated value of needs.

S104, calculate bit error rate estimation value according to Soft Inform ation corresponding to each bit described and described PDF estimated value.

The Soft Inform ation that described in obtaining according to said process, each bit is corresponding and described PDF estimated value, calculate bit error rate estimation value, can carry out power control for the power control system in described radio digital communication system.

The described error rate measurement method that the application provides, by processing the mode of operation of the data-signal received according to current described power control system base band, obtains the Soft Inform ation of each bit; Recycle the statistical property based on the Soft Inform ation probability density function of kernel function, calculate PDF estimated value; Then described bit error rate estimation value is calculated according to Soft Inform ation corresponding to each bit described and described PDF estimated value, without the need to the threshold value preset in hierarchical statistics in prior art, make the accuracy of bit error rate estimation and noise (doing) than rank (error rate) decoupling zero, thus realize directly using the error rate as power control regulating index, improve accuracy; Meanwhile, the error rate is measured without the need to loopback, meets the requirement of Adaptable System for real-time.

What deserves to be explained is, for the measurement of the error rate, also there are two kinds of methods in prior art, a kind of is utilize MC (Monte-Carlo that is classical or that expand, Monte Carlo) method, another kind is the estimated value utilizing the relation of signal to noise ratio and the error rate to obtain the error rate.

Wherein, classical MC method statistic receives the number of errors of bit information, and calculates the estimated value that it obtains the error rate with the ratio sending bit (actual receive bit).The method is in order to ensure accuracy, need the information bit sample of some, particularly when signal to noise ratio is higher, sample requirement amount is very big, thus cause the real-time of error rate measurement to reduce, although the MC method expanded can utilize statistical confidence principle to make compromise in certainty of measurement and Measuring Time, communication receiver needs the transmission bit information obtaining some all the time, and namely the method cannot realize the bit error rate estimation of real-time online.

Another utilizes the relation of signal to noise ratio and the error rate to obtain the method for the estimated value of the error rate, first received signal to noise ratio is measured, then obtain the estimated value of the error rate according to the mapping relations of signal to noise ratio and the error rate under given coding/decoding, modulating/demodulating mode and known channel condition.The method real-time is higher, but obtains channel condition due to the given transmitting-receiving mode of needs, and accuracy and applicability are all poor.

Therefore existing estimating method of the error rate all cannot ensure applicability, accuracy and real-time simultaneously, also just cannot be used for the improvement that power controls.

The described error rate measurement method that the application provides, the measurement of the error rate is realized by above-mentioned steps, not only ensure that accuracy and real-time, simultaneously, for the data-signal received without the need to knowing that it sends Bit data, only utilizing the corresponding Soft Inform ation of described bit can realize follow-up calculating, obtaining described bit error rate estimation value, without the need to the complexity that recompile in prior art causes, the applicability of described error rate measurement, accuracy and real-time have also namely been ensured simultaneously.And reduce the existing complexity simultaneously using open loop, closed loop, inner ring, external circule power control system.

Preferably, as shown in Figure 2, step S102 comprises:

S201, to classify according to the Soft Inform ation that current described power control system planisphere characteristic is corresponding to each bit described, the categorical measure of Soft Inform ation corresponding to each bit described and element number of all categories are added up;

In concrete practical application, the categorical measure of the Soft Inform ation that each bit described is corresponding depends on the constellation point number of ideally planisphere.With BPSK (BinaryPhaseShiftKeying, binary phase shift keying) for example, can according to described Soft Inform ation X _ito be positive and negatively initialized as with

$C_1^{\left(0\right)}=\left\{X_i\right|X_i\&GreaterEqual;0\},C_2^{\left(0\right)}=\{X_i|X_i<0\};$

Element number then of all categories is:

$N_1^{\left(0\right)}=Card\left\{C_1^{\left(0\right)}\right\},C_2^{\left(0\right)}=Card\left\{C_2^{\left(0\right)}\right\}.$

S202, calculate described of all categories in the prior probability of Soft Inform ation;

For BPSK, classification with in the prior probability of Soft Inform ation be:

${\mathrm{\&alpha;}}_1^{\left(0\right)}=\frac{N_1^{\left(0\right)}}{N},{\mathrm{\&alpha;}}_2^{\left(0\right)}=\frac{N_2^{\left(0\right)}}{N}.$

S203, calculate the smoothing parameter that the PDF based on kernel function estimates.

For real system, function f _xthe unknown, its Soft Inform ation classification C (f _x) cannot obtain, in concrete practical application, can f be supposed _xfor Gaussian function (N (μ, σ ²)), using the initial value as calculating postfitted orbit parameter SP.For BPSK, Soft Inform ation X _ipDF be:

${\mathrm{PDF}}_{X_i,C_1}={\hat{f}}_{X,N_1}\left(x\right)=\frac{1}{N_1\&times;{\mathrm{SP}}_1}\underset{X_i\&Element;C_1}{\&Sigma;}K\left(\frac{x-X_i}{{\mathrm{SP}}_1}\right)$

${\mathrm{PDF}}_{X_i,C_2}={\hat{f}}_{X,N_2}\left(x\right)=\frac{1}{N_2\&times;{\mathrm{SP}}_2}\underset{X_i\&Element;C_2}{\&Sigma;}K\left(\frac{x-X_i}{{\mathrm{SP}}_2}\right);$

Preferably, described in calculate the smoothing parameter that the PDF based on kernel function estimates, comprising:

When the passage receiving described data-signal is Gaussian channel, then basis calculate the smoothing parameter that the PDF based on kernel function estimates;

Wherein, σ _jfor Gaussian function (N (μ, σ ²)) in a variable, N _jfor the element number that the Soft Inform ation in described PDF estimated parameter is of all categories, j is the quantity of described Soft Inform ation classification.

For normal state Gaussian type kernel function, different classes of with corresponding smoothing parameter SP is initialized as respectively:

${\mathrm{SP}}_1^{\left(0\right)}={\left(\frac{4}{3N_1}\right)}^{1/5}{\mathrm{\&sigma;}}_1$

${\mathrm{SP}}_2^{\left(0\right)}={\left(\frac{4}{3N_2}\right)}^{1/5}{\mathrm{\&sigma;}}_2.$

In concrete practical application, the formula in above-mentioned steps is not necessarily defined in BPSK, can also adopt as the case may be.

Preferably, as shown in Figure 3, step S103 comprises:

S301, judge whether current iteration number of times t is greater than described maximum iteration time T;

Concrete, maximum iteration time T depends on practical communication system and channel conditions, when using chnnel coding (as Turbo code) or signal to noise ratio higher (if the corresponding error rate is at 10-5), maximum iteration time T should be set to high value (as 20).In concrete application, the setting that maximum iteration time T can carry out in advance according to concrete applied environment.

If described current iteration number of times t is less than or equal to described maximum iteration time T, S302, calculates the posterior probability of the Soft Inform ation in current Soft Inform ation classification;

Judge whether current iteration number of times t is greater than described maximum iteration time T, if current iteration number of times t equals described maximum iteration time T, illustrate and can meet system requirements after completing this iteration.

The described PDF estimated parameter obtained in last iteration is employed in computational process:

${\mathrm{\&beta;}}_j^{\left(t\right)}=\frac{{\mathrm{\&alpha;}}_j^{(t-1)}{\hat{f}}_{X,N_j^{(t-1)}}\left(X_i\right)}{\underset{j}{\&Sigma;}{\mathrm{\&alpha;}}_j^{(t-1)}{\hat{f}}_{X,N_j^{\left(t-1\right)}}\left(X_i\right)};$

For BPSK, different classes of with in the posterior probability of Soft Inform ation be respectively:

${\mathrm{\&beta;}}_1^{\left(t\right)}=\frac{{\mathrm{\&alpha;}}_1^{(t-1)}{\hat{f}}_{X,N_1^{(t-1)}}\left(X_i\right)}{{\mathrm{\&alpha;}}_1^{(t-1)}{\hat{f}}_{X,N_1^{\left(t-1\right)}}\left(X_i\right)+{\mathrm{\&alpha;}}_2^{(t-1)}{\hat{f}}_{X,N_2^{(t-1)}}\left(X_i\right)}$

${\mathrm{\&beta;}}_2^{\left(t\right)}=1-\frac{{\mathrm{\&alpha;}}_1^{(t-1)}{\hat{f}}_{X,N_1^{(t-1)}}\left(X_i\right)}{{\mathrm{\&alpha;}}_1^{(t-1)}{\hat{f}}_{X,N_1^{\left(t-1\right)}}\left(X_i\right)+{\mathrm{\&alpha;}}_2^{(t-1)}{\hat{f}}_{X,N_2^{(t-1)}}\left(X_i\right)}$

If current iteration number of times t reaches maximum iteration time T, jump to step S104.

Calculate posterior probability described in S303, basis, calculate and upgrade described PDF estimated parameter;

Prior probability α in described PDF estimated parameter _jfor:

${\mathrm{\&alpha;}}_j^{\left(t\right)}=\frac{{\mathrm{\&Sigma;}}_{i=1}^N{\mathrm{\&beta;}}_{i,j}^{\left(t\right)}}{N};$

For BPSK, different classes of with in the prior probability of Soft Inform ation be respectively:

${\mathrm{\&alpha;}}_1^{\left(t\right)}=\frac{1}{N}\underset{i=1}{\overset{N}{\&Sigma;}}{\mathrm{\&beta;}}_1^{\left(t\right)};$

${\mathrm{\&alpha;}}_2^{\left(t\right)}=1-{\mathrm{\&alpha;}}_1^{\left(t\right)}$

Wherein, Soft Inform ation classification C _jwith the element number N that Soft Inform ation is of all categories _jupgrade according to the planisphere characteristic of current described power control system, smoothing parameter SP _jits optimum value is calculated according to MISE minimum principle.

Under current iteration, the element number N1 of two classifications and N2 needs according to upgraded Soft Inform ation classification C ₁and C ₂obtain, Soft Inform ation classification C ₁and C ₂be calculated as follows, wherein for equally distributed variable between [0,1]:

$\begin{array}{c}{C}_1^{\left(t\right)}=\left\{X_i:{\mathrm{\&beta;}}_1^{\left(t\right)}\&GreaterEqual;U_i^{\left(t\right)}\right\}\\ C_2^{\left(t\right)}={\stackrel{\&OverBar;}{C}}_1^{\left(t\right)}\end{array};$

Preferably, described in calculate the smoothing parameter that the PDF based on kernel function estimates, comprising:

When the passage receiving described data-signal is Unknown Channel, then basis ${\mathrm{SP}}_{optimal}={\left(\frac{A\left(K\right)}{N\&times;B{\left(K\right)}^2\&times;C\left({\hat{f}}_X\right)}\right)}^{1/5}={\left(\frac{\&Integral;{\left(K\left(t\right)\right)}^{2}dt}{N{\left(\&Integral;t^{2}K\left(t\right)dt\right)}^2\&Integral;{\left({\hat{f}}_X^{\&prime;\&prime;}\left(t\right)\right)}^{2}dt}\right)}^{1/5}$ Calculate the smoothing parameter that the PDF based on kernel function estimates;

Wherein, A and B is respectively constant, and N is the amount of bits of the described data-signal received, and K (.) is described kernel function, f _xfor function.

For real system, channel situation is unknown, then can upgrade according to above formula the optimum value obtaining current iteration smoothing parameter SP.

S304, according to the PDF estimated parameter after described renewal, calculate and upgrade described PDF estimated value;

Make t=t+1, repeat S301-S304, until current iteration number of times t is greater than described maximum iteration time T;

Wherein, the posterior probability calculating of a Soft Inform ation and the equal corresponding a kind of Soft Inform ation classification of calculating of a PDF estimated parameter.

Preferably, can basis ${\mathrm{PDF}}_{X_i}={\hat{f}}_X\left(x\right)=\frac{1}{N\&times;SP}\underset{X_i}{\&Sigma;}K\left(\frac{x-X_i}{SP}\right)$ Calculate described PDF estimated value;

Wherein, X _ifor the Soft Inform ation (1≤i≤N of each bit described, N is positive integer), x is the independent variable of the Soft Inform ation of each bit described, SP is the smoothing parameter in described PDF estimated parameter, K (.) is described kernel function, and N is the element number of this kind of Soft Inform ation classification in described PDF estimated parameter.

In concrete practical application, corresponding different Soft Inform ation classification carries out the posterior probability calculating of Soft Inform ation and the calculating of PDF estimated parameter respectively, finally obtains the described PDF estimated value of Soft Inform ation corresponding to each bit described.

Preferably, as shown in Figure 4, step S104 comprises:

S1041, interval according to the error distribution on current described power control system planisphere, to divide and defining integration is interval;

In concrete practical application, there is error distribution interval in Soft Inform ation on planisphere, with the mistake interval division of Soft Inform ation on planisphere and defining integration is interval, for the calculating of described bit error rate estimation value.Modulate mutually for more much higher, need to divide mistake according to planisphere interval, then the integrating range of above formula is set by mistake interval.

S1042, calculate described bit error rate estimation value according to described integrating range and described PDF estimated value.

Behind defining integration interval, calculate the described error rate according to described integrating range and described PDF estimated value with the mistake interval division of Soft Inform ation on planisphere.

Fig. 5 is for carry out error rate calculation displaying for BPSK system.Ideally, + 1 and-1 is should be according to the Soft Inform ation sample (X) that the data-signal received calculates, be described for+1: during hard decision, due to+1 sample value received when the received also >0 sent, all bits all correctly receive, and there is not error code.But under certain wireless channel conditions, the probability density of Soft Inform ation sample (X) presents certain probability distribution (fX, N1 and fX, N2).Due to the effect of wireless channel, transmission+1 when receiving, corresponding sample value may <0, be now judged to be-1.Using 0 as decision threshold, then consider the probability distribution of sample value, the hatched area in the middle of known Fig. 5 is depicted as error rate region, is ber value.

Therefore for BPSK, the error rate is pressed following formula and is estimated:

$BER={\mathrm{\&alpha;}}_1^{\left(T\right)}{\&Integral;}_{-\mathrm{\&infin;}}^0{\hat{f}}_{X,N_1^{\left(T\right)}}\left(x\right)dx+{\mathrm{\&alpha;}}_2^{\left(T\right)}{\&Integral;}_0^{+\mathrm{\&infin;}}{\hat{f}}_{X,N_2^{\left(T\right)}}\left(x\right)dx.$

Wherein α ₁, α ₂for the prior probability of sample value, as got 1000 sampled points, wherein 300 corresponding transmission bits are the sample value of+1, and 700 corresponding transmission bits are the sample value of-1, then α ₁=0.3, α ₂=0.7.

Another embodiment of the present invention additionally provides a kind of error rate measurement system, as shown in Figure 6, is applied to radio digital communication system, and described power control system comprises:

Decoding unit 101, PDF parameter generating unit 102, PDF estimation unit 103 and bit error rate estimation unit 104.

Concrete operation principle is:

According to the mode of operation of current power control system base band, decoding unit 101 carries out that process is synchronous to the data-signal received, demodulation, decode operation, obtain the Soft Inform ation (X that each bit is corresponding _i, 1≤i≤N);

PDF parameter generating unit 102, according to current power control system modulation system and planisphere characteristic, carries out initialization to the Soft Inform ation probability density function PDF estimated parameter based on kernel function (KernelFunction);

PDF estimation unit 103 is according to described PDF estimated parameter and the very big algorithm of statistics, and the Soft Inform ation corresponding to each bit described carries out PDF blind estimate, calculates PDF estimated value;

Bit error rate estimation unit 104 calculates bit error rate estimation value according to Soft Inform ation corresponding to each bit described and described PDF estimated value.

Concrete operation principle is same as the previously described embodiments, repeats no more herein.

Another embodiment of the present invention additionally provides a kind of power control system, as shown in Figure 7, is applied to radio digital communication system, and described power control system comprises:

Decoding unit 101, PDF parameter generating unit 102, PDF estimation unit 103, bit error rate estimation unit 104, target comparing unit 105, angle detecting unit 106, algorithm setting unit 107 and adjustment unit 108.

Concrete operation principle is:

According to the mode of operation of current power control system base band, decoding unit 101 carries out that process is synchronous to the data-signal received, demodulation, decode operation, obtain the Soft Inform ation (X that each bit is corresponding _i, 1≤i≤N);

PDF parameter generating unit 102, according to current power control system modulation system and planisphere characteristic, carries out initialization to the Soft Inform ation probability density function PDF estimated parameter based on kernel function (KernelFunction);

PDF estimation unit 103 is according to described PDF estimated parameter and the very big algorithm of statistics, and the Soft Inform ation corresponding to each bit described carries out PDF blind estimate, calculates PDF estimated value;

Bit error rate estimation unit 104 calculates bit error rate estimation value according to Soft Inform ation corresponding to each bit described and described PDF estimated value;

Target comparing unit 105 compares described bit error rate estimation value and error rate target gate, generates comparative result;

Angle detecting unit 106 detects the direction that current power controls;

Algorithm setting unit 107 regulates algorithm to arrange according to the external circule power control step parameter of current described power control system;

Adjustment unit 108 is arranged according to the direction of described comparative result, the control of described current power and described algorithm, generates and output power command bit.

After described power control instruction bit is received by the opposing party, through demodulation coding and power control bit extraction, send to gain link PA to carry out gain-adjusted, realize the adjustment that power controls.

What deserves to be explained is, for the Poewr control method of the power control system in radio digital communication system in prior art, all use signal to noise ratio to carry out power control, the error rate measurement method described in above-described embodiment at present, then pass through:

Described bit error rate estimation value and error rate target gate are compared, generates comparative result;

Detect the direction that current power controls;

Algorithm is regulated to arrange according to the external circule power control step parameter of current described power control system;

The direction controlled according to described comparative result, described current power and described algorithm are arranged, and generate and output power command bit;

After above-mentioned steps, the architecturally existing power control system of reusable (as shown in Figure 8), after reception, decoding, bit error rate estimation, the outer shroud being demarcated adjustment by snr measurement and signal to noise ratio is controlled, more relatively after, realize power control bit and generate, send to the opposing party after demodulation coding and power control bit extraction, send to gain link PA to carry out gain-adjusted, complete inner ring and control, realize the adjustment that power controls; Signal to noise ratio is only replaced with the parameter index that described bit error rate estimation value controls as power by the present embodiment in control procedure.

The described power control system that the application provides, not only the same with above-described embodiment, utilize the statistical property based on the Soft Inform ation probability density function of kernel function, calculate described bit error rate estimation value, make the accuracy of bit error rate estimation and noise (doing) than rank (error rate) decoupling zero, thus realize directly using the error rate as power control regulating index, improve accuracy, use described bit error rate estimation value as power contorl parameters, make power control can more directly reflect system business quality; And for the data-signal received without the need to knowing that it sends Bit data, without the need to the complexity that recompile in prior art causes.And the error rate is measured without the need to loopback, meet the requirement of Adaptable System for real-time.Solve the existing problem high in conjunction with the power control techniques complexity of open loop, closed loop as shown in Figure 5 simultaneously, and solve with this problem that existing method cannot meet accuracy and the requirement of real-time two aspect simultaneously, greatly can improve the speed of open sea wharf, even control the low accuracy fast inner loop power based on signal to noise ratio/signal interference ratio and control based on the high accuracy slower outer loop power of the error rate power control system be integrated in the lump based on the error rate, for the cdma system that single user standard receives, when signal to noise ratio is 10dB, only need 700 samples can obtain variance compared with theoretical value and be less than the bit error rate estimation of 10-6, thus the real-time online power realizing high accuracy controls.

Simultaneously, utilize the printenv characteristic of Density Estimator method (Kernelestimation), the power control system described in the present embodiment is made to be applicable to Any Digit communication system, as cellular communication systems such as CDMA, WCDMA and other 3G, 4G network systems, or sensor network, mobile ad hoc network, car networking etc.

In the present invention, each embodiment adopts the mode of going forward one by one to describe, and what each embodiment stressed is the difference with other embodiments, between each embodiment identical similar portion mutually see.For device disclosed in embodiment, because it corresponds to the method disclosed in Example, so description is fairly simple, relevant part illustrates see method part.

Below be only the preferred embodiment of the present invention, those skilled in the art understood or realizes the present invention.To be apparent to one skilled in the art to the multiple amendment of these embodiments, General Principle as defined herein can without departing from the spirit or scope of the present invention, realize in other embodiments.Therefore, the present invention can not be restricted to these embodiments shown in this article, but will meet the widest scope consistent with principle disclosed herein and features of novelty.

Claims (9)

1. an error rate measurement method, is characterized in that, comprising:

The mode of operation of the data-signal received according to current power control system base band is processed, obtains the Soft Inform ation that each bit is corresponding;

According to current power control system modulation system and planisphere characteristic, initialization is carried out to the Soft Inform ation probability density function PDF estimated parameter based on kernel function;

According to described PDF estimated parameter and the very big algorithm of statistics, the Soft Inform ation corresponding to each bit described carries out PDF blind estimate, calculates PDF estimated value;

The Soft Inform ation corresponding according to each bit described and described PDF estimated value calculate bit error rate estimation value.

2. error rate measurement method according to claim 1, is characterized in that, described according to described current described power control system modulation system and planisphere characteristic, carries out initialization comprise the PDF estimated parameter based on kernel function:

The Soft Inform ation corresponding to each bit described according to current described power control system planisphere characteristic is classified, and adds up the categorical measure of Soft Inform ation corresponding to each bit described and element number of all categories;

Calculate described of all categories in the prior probability of Soft Inform ation;

Calculate the smoothing parameter that the PDF based on kernel function estimates.

3. error rate measurement method according to claim 2, is characterized in that, described in calculate the smoothing parameter that the PDF based on kernel function estimates, comprising:

When the passage receiving described data-signal is Gaussian channel, then basis calculate the smoothing parameter that the PDF based on kernel function estimates;

Wherein, σ _jfor Gaussian function (N (μ, σ ²)) in a variable, N _jfor the element number that the Soft Inform ation in described PDF estimated parameter is of all categories, j is the quantity of described Soft Inform ation classification.

4. error rate measurement method according to claim 2, is characterized in that, described in calculate the smoothing parameter that the PDF based on kernel function estimates, comprising:

When the passage receiving described data-signal is Unknown Channel, then basis ${\mathrm{SP}}_{optimal}={\left(\frac{A\left(K\right)}{N\&times;B{\left(K\right)}^2\&times;C\left({\hat{f}}_X\right)}\right)}^{1/5}={\left(\frac{\&Integral;{\left(K\left(t\right)\right)}^{2}dt}{N{\left(\&Integral;t^{2}K\left(t\right)dt\right)}^2\&Integral;{\left({\hat{f}}_X^{\&prime;\&prime;}\left(t\right)\right)}^{2}dt}\right)}^{1/5}$ Calculate the smoothing parameter that the PDF based on kernel function estimates;

Wherein, A and B is respectively constant, and N is the amount of bits of the described data-signal received, and K (.) is described kernel function, f _xfor function.

5. error rate measurement method according to claim 1, is characterized in that, described according to described PDF estimated parameter and the very big algorithm of statistics, the Soft Inform ation corresponding to each bit described carries out PDF blind estimate, calculates PDF estimated value and comprises:

S301, judge whether current iteration number of times t is greater than described maximum iteration time T;

If described current iteration number of times t is less than or equal to described maximum iteration time T;

S302, calculate the posterior probability of the Soft Inform ation in current Soft Inform ation classification;

Calculate posterior probability described in S303, basis, calculate and upgrade described PDF estimated parameter;

S304, according to the PDF estimated parameter after described renewal, calculate and upgrade described PDF estimated value;

Make t=t+1, repeat S301-S304, until current iteration number of times t is greater than described maximum iteration time T;

Wherein, the posterior probability calculating of a Soft Inform ation and the equal corresponding a kind of Soft Inform ation classification of calculating of a PDF estimated parameter.

6. error rate measurement method according to claim 1, is characterized in that, described according to described PDF estimated parameter and the very big algorithm of statistics, the Soft Inform ation corresponding to each bit described carries out PDF blind estimate, calculates PDF estimated value, comprising:

According to ${\mathrm{PDF}}_{X_i}={\hat{f}}_X\left(x\right)=\frac{1}{N\&times;SP}\underset{X_i}{\&Sigma;}K\left(\frac{x-X_i}{SP}\right)$ Calculate PDF estimated value;

Wherein, X _ifor the Soft Inform ation (1≤i≤N of each bit described, N is positive integer), x is the independent variable of the Soft Inform ation of each bit described, SP is the smoothing parameter in described PDF estimated parameter, K (.) is described kernel function, and N is the element number of this kind of Soft Inform ation classification in described PDF estimated parameter.

7. error rate measurement method according to claim 1, is characterized in that, the Soft Inform ation that described in described basis, each bit is corresponding and described PDF estimated value calculate bit error rate estimation value, comprising:

Interval according to the error distribution on current described power control system planisphere, divide and defining integration interval;

Described bit error rate estimation value is calculated according to described integrating range and described PDF estimated value.

8. an error rate measurement system, is characterized in that, is applied to the power control system in radio digital communication system, comprises:

Decoding unit, for processing the mode of operation of the data-signal received according to current described power control system base band, obtains the Soft Inform ation of each bit;

PDF parameter generating unit, for according to current described power control system modulation system and planisphere characteristic, carries out initialization to the PDF estimated parameter based on kernel function;

PDF estimation unit, for according to described PDF estimated parameter and statistics very big algorithm, the Soft Inform ation corresponding to each bit described carries out PDF blind estimate, calculates PDF estimated value;

Bit error rate estimation unit, for calculating bit error rate estimation value according to Soft Inform ation corresponding to each bit described and described PDF estimated value.

9. a power control system, is characterized in that, is applied to radio digital communication system, and described power control system comprises:

Decoding unit, for processing the mode of operation of the data-signal received according to current described power control system base band, obtains the Soft Inform ation of each bit;

PDF parameter generating unit, for according to current described power control system modulation system and planisphere characteristic, carries out initialization to the PDF estimated parameter based on kernel function;

PDF estimation unit, for according to described PDF estimated parameter and statistics very big algorithm, the Soft Inform ation corresponding to each bit described carries out PDF blind estimate, calculates PDF estimated value;

Bit error rate estimation unit, for calculating bit error rate estimation value according to Soft Inform ation corresponding to each bit described and described PDF estimated value;

Target comparing unit, for comparing described bit error rate estimation value and error rate target gate, generates comparative result;

Angle detecting unit, for detecting the direction that current power controls;

Algorithm setting unit, for regulating algorithm to arrange according to the external circule power control step parameter of current described power control system;

Adjustment unit, for according to described comparative result, described current power control direction and described algorithm arrange, generate and output power command bit.