CN100411309C - Burst detecting equipment and method for radio receiver - Google Patents

Abstract

The present invention discloses a burst detection device and a detection method thereof. The burst detection device comprises a square module set, a flexible measurement mean square root module set, and a quality determination module set; the burst detection method orderly comprises the following steps: calculating the reciprocal of the quadratic sum of received signals, deducing the flexible measurement root mean square value of the received signals, and comparing the product of the reciprocal and the root mean square value (RMS, root mean square) with a given critical value to generate a Boolean value, wherein the quality of the received signals is determined by the Boolean value; if the product is smaller than the critical value, then the received signals are regarded as noise, else the received signals are regarded as signals.

Description

Be applied to the pulse train arrangement for detecting and the method for radio receiver

Technical field

The invention relates to radio receiver, particularly about being applied in pulse train arrangement for detecting and the method on the radio receiver.

Background technology

In an actual communication environment, the whole bag of tricks that receives radio wave must be told the signal that will receive from noise.Very naturally,, above-mentioned the whole bag of tricks is to identify signal and noise completely, so noise can be thought by mistake the probability of signal is designated as P _Signal/noise, similarly, think signal by mistake probability of noise then be designated as P _Noise|signalThe various radiotechnics that is developed in this field all desire reduces the probability of above-mentioned two kinds of misidentifications.

A kind of method that is called pulse train (burst) detecting is in order to judge that the signal that is received is real signal or noise.In some communication system, the false judgment that above-mentioned pulse train method for detecting has been done will cause delay, interruption, the data loss of communication, even channel loses connection.For example in the mechanical stage standard in the middle of the global system for mobile telecommunications service that is widely known by the people (GSM, the Global System for Mobile communication) system, stipulated that tranmitting frequency must keep certain accuracy.Usually for the holding frequency accuracy, all adopted automatic frequency control.In addition, in the standard of numbering 11.10,, must in this temporary transient period, keep the frequency and the time-code locking (timing) of wireless radio transmission if mechanical stage can't correctly be received desired signal in a period of time.When the signal that is received is that above-mentioned pulse train method for detecting is judged as noise, then the result who is estimated by above-mentioned automatic frequency controlled function will be rejected, and keep needed frequency accuracy with this.If also be used by the estimated frequency resultant of coming out of these noises, its consequence may cause the mistake connection of communication.

One of above-mentioned pulse train method for detecting is in order to calculate relevant (correlation) algorithm of a correlation for a kind of.For carrying out this relevant algorithm, the signal bursts domestic demand comprises one section known in advance midamble code (training code), and must finish synchronized action earlier.Because of noise uncorrelated with above-mentioned known midamble code, so when the signal strength signal intensity of receiving is identical, should be at the correlation that noise produced much smaller than correlation at real signal.In view of the above, the peak value of the correlation that draws after the calculating (peak) after signal strength signal intensity normalization (normalize), can do one relatively with a critical value (threshold), and whether the signal that receives with decision serves as true.In the system of global system for mobile telecommunications service, the result that the correlation of a subcritical value will cause above-mentioned automatic frequency controlled function to be estimated is abandoned, otherwise then this result who estimates will be accepted.

See through the mode of adjusting above-mentioned critical value, in keeping P _Noise|signalIn the time of a quite low probability, also can reduce P effectively _Signal/noiseTo an acceptable degree.Yet the method has two weak points:

The first, midamble code length is 26, might occur pseudo-peak (false peak) when relevant, therefore, the noise erroneous judgement may be signal.

The second, when multi-path (multi-path) transmits, if two or the approaching path of more intensity.Signal power can be dispersed in two (or more) paths, removes one of them peak value of normalization (normalize) with the signal strength signal intensity of two paths, and the value that obtains may be lower than critical value, thereby the signal erroneous judgement is noise.

In sum, need develop the probability of a kind of apparatus and method of radio reception with while reducing noise and the mutual misidentification of signal.

Summary of the invention

In above-mentioned background of invention, the present invention discloses a kind of pulse train (burst) arrangement for detecting that is applied to radio reception.In an embodiment, above-mentioned pulse train arrangement for detecting comprises one square of module, soft measuring (soft-metric) root mean square (RMS, root mean square) module, with quality decision module.Wherein, the soft therewith root mean square module of measuring of above-mentioned square module is all accepted digital input signals, this digital input signals is to be quantized to produce with the conversion back through analog/digital converter by radio signal, and above-mentioned radio signal is to comprise a homophase (in-phase) input signal and a quadrature (quadrature) input signal.One output valve of above-mentioned square of module is the inverse for an above-mentioned input signal square.Simultaneously, an above-mentioned soft output valve of measuring the root mean square module is to be the root mean square according to the soft measurement value that this input signal produced.Thereafter, above-mentioned quality decision module is the product and a given critical value of this two output valve of comparison; If the product of this two output valve, judges then that the signal that is received is a noise less than this critical value; Otherwise, then be regarded as real signal.

In another embodiment of the present invention, one by the pulse train arrangement for detecting that disclosed comprise a homophase module, a quadrature module, soft measure etc. changed module, a root mean square module, with quality decision module.Wherein, above-mentioned soft change module such as measure is accepted digital input signals, this digital input signals is to be quantized to produce with the conversion back through analog/digital converter by radio signal, and above-mentioned radio signal is to comprise a homophase (in-phase) input signal and a quadrature (quadrature) input signal.One output valve of above-mentioned homophase module is to be an above-mentioned in-phase input signals for 1 quadratic sum.Similarly, an output valve of above-mentioned quadrature module is to be an above-mentioned positive input signal quadratic sum.Simultaneously, an above-mentioned soft sequence output valve of the change module such as measuring is to be the soft measurement value according to this input signal produced.Then, the above-mentioned soft sequence output valve of the change module such as measuring is to produce a root-mean-square value via above-mentioned root mean square module.Thereafter, above-mentioned quality decision module is that the inverse with the summation of this two quadratic sum multiply by this root-mean-square value, and compares with a given critical value; If above-mentioned product, judges then that the signal that is received is a noise less than this critical value; Otherwise, then be regarded as real signal.

The present invention also discloses a pulse train method for detecting, and wherein the method comprises the following step: the inverse that at first calculates received signal square; Then, derive the root-mean-square value of a soft measurement value according to received signal; At last, the product of above-mentioned inverse and root-mean-square value and a given critical value are compared; If above-mentioned product, judges then that the signal that is received is a noise less than this critical value; Otherwise, then be regarded as real signal.

A purpose of the present invention is in the environment of single-pathway and multi-path radio propagation, reduces and thinks signal by mistake probability of noise.

Another object of the present invention is to keep when thinking signal the low probability of noise by mistake, improve the probability of noise being thought by mistake signal.

Description of drawings

The present invention is a kind of pulse train arrangement for detecting and method that is applied to radio reception in this direction of inquiring into.In order to understand the present invention up hill and dale, detailed step and composition thereof will be proposed in following description.Apparently, execution of the present invention is not defined in the specific details that the operator had the knack of of radio receiver.On the other hand, well-known composition or step are not described in the details, with the restriction of avoiding causing the present invention unnecessary.Preferred embodiment meeting of the present invention is described in detail as follows, yet except these detailed descriptions, the present invention can also be implemented among other the embodiment widely, and scope of the present invention do not limited, its with after claim be as the criterion, wherein:

Fig. 1 is the block schematic diagram according to the pulse train arrangement for detecting that one embodiment of the invention provided;

Fig. 2 is the block schematic diagram for a pulse train arrangement for detecting that is provided according to another embodiment of the present invention; And

It is to be the schematic flow sheet according to the pulse train method for detecting that one embodiment of the invention provided for Fig. 3.

Embodiment

With reference to shown in Figure 1, it is to be the block schematic diagram according to the pulse train arrangement for detecting 100 that one embodiment of the invention provided.Then narrate the content of Fig. 1 embodiment.Above-mentioned pulse train arrangement for detecting comprises one square of module, 110, one soft measuring (soft-metric) root mean square (RMS, root mean square) module 120, with quality decision module 130.Wherein, the soft therewith root mean square module 120 of measuring of above-mentioned square module 110 is all accepted a digital input signals 140, this input signal 140 is to be quantized to produce with the conversion back through analog/digital converter by radio signal, and above-mentioned radio signal is to comprise a homophase (in-phase) input signal 142 and quadrature (quadrature) input signal 144.Be convenient to the event of explanation, this in-phase input signals for 1 142 is designated as r _I, and this positive input signal 144 is designated as r _Q

In present embodiment, this square module 110 is in order to calculate an inverse of institute's receiving inputted signal 140 quadratic sums.Is input with above-mentioned in-phase input signals for 1 142 with positive input signal 144, and this square module 110 is to derive an output valve according to following first party formula:

$m_1=\frac{1}{\underset{i}{\mathrm{\Σ}}{\left|r_I\left(i\right)\right|}^2+\underset{i}{\mathrm{\Σ}}{\left|r_Q\left(i\right)\right|}^2}$ The first party formula

In addition, the above-mentioned soft root mean square module 120 of measuring is to carry out a soft calculating of measuring an algorithm and a root-mean-square value.Real being to use that the present invention is suitable for, but the soft algorithm of measuring that is not limited to, be the one piece of paper that comes from international electronic motor engineering association (IEEE), promptly by Ke He (WolfgangKoch) and Ba Er (A1fred Baier) proposed " because of time-variance scatter cause coded identification mutually the best of the enciphered data of mutual interference detect with suboptimum " (OPTIMUM AND SUB-OPTIMUMDETECTION OF CODED DATA DISTRIBUTED BY TIME-VARYING INTERSYMBOLINTERFERENCE).This piece paper provides the first-classization algorithm, and under the situation that ideal is interlocked, these change algorithms is to can be channel decoder the best soft signal of measuring is provided.In addition, this paper also provides complicated simple parameter of traditional Viterbi (Viterbi) eqalizing cricuit, several in view of the above soft signals of measuring that suboptimum is provided.After deriving soft measurement value, the above-mentioned soft root mean square module 120 of measuring then calculates a root-mean-square value according to following second party formula; The also soft for this reason output valve of measuring root mean square module 120 of this root-mean-square value:

$m_2=\sqrt{\frac{\underset{i}{\mathrm{\Σ}}{(\mathrm{soft}-\mathrm{metric}\left(i\right))}^2}{N}}$ The second party formula

Wherein be for corresponding to the number of the enciphered data in the pulse train.

As shown in Figure 1, divided by the output valve m of above-mentioned square of module 110 ₁With the above-mentioned soft output valve m that measures root mean square module 120 ₂Outside input, this quality decision module 130 also receives a prior determined critical value 150.This quality decision module 130 is with this two output valve m ₁With m ₂The multiply each other product of gained compares the boolean results 160 that the back produces this pulse train arrangement for detecting 100 to this critical value 150.Quality decision module 130 performed functions are can following third party's formula represented thus:

$\mathrm{quality}-\mathrm{outcome}=m_1\⊗m_2?\mathrm{threshold}$ Third party's formula

If the product of this two output valve, judges then that the signal that is received is a noise less than this critical value 150; Otherwise, then be regarded as real signal.

The numerical values recited of above-mentioned soft measurement value is represented the confidence level of this judgement, so can be in order to the detecting noise.When receiving noise, the confidence level of this soft measurement value representative will be quite low, and the confidence level of the soft measurement value that serious positive signal produced is low at least.Utilize soft this characteristic of measuring algorithm, can reduce the possibility of noise being thought by mistake signal; Simultaneously, also exempted the situation that the signal strength signal intensity that obtains with mulitpath is removed a peak value in normalization (normalize) correlation, therefore reduced the possibility of signal being thought by mistake noise, in other words, the soft algorithm of measuring can reduce P _Signal|noiseAnd P _Noise|signalLook from another viewpoint, propagate down in multi-path and think signal by mistake the problem of noise still exist.When making an uproar news when quite low, promptly the signal that receives of toilet be very, yet to measure the result that algorithm draws be very trustless by soft.In view of the above, when when measuring low confidence level result that algorithm draws and being lower than a normal critical value, also abandoning the result that the automatic frequency controlled function is estimated according to soft.

In another preferable example of this embodiment of the present invention, when being applied to the system of global system for mobile telecommunications service, the parameter in above-mentioned first and second equation can be designated.The pulse train content of the communication channel of global system for mobile telecommunications service is transmitting than inferior form according to first table.

Bit number zone length area contents

0-2 3 tail bits

3-60 58 encrypts bit (e0-e57)

61-86 26 training sequence bits

87-144 58 encrypts bit (e58-e115)

145-147 3 tail bits

Bit during 148-156 8 monitors

First table: the pulse train content of global system for mobile telecommunications service

In view of the above, it is as follows to derive the cubic formula that can be applicable to the global system for mobile telecommunications service by the first party formula:

$m_1=\frac{1}{\underset{i=0}{\overset{147}{\mathrm{\Σ}}}{\left|r_I\left(i\right)\right|}^2+\underset{i=0}{\overset{147}{\mathrm{\Σ}}}{\left|r_Q\left(i\right)\right|}^2}$ Cubic formula

In the cubic journey, the needed bit number of signal calculated intensity, can from 148 bits, choose all as required or wherein a part.

Comprise 116 in the pulse train for the global system for mobile telecommunications service shown in first table and encrypt bit; Be as the flag of controlling channel signals, so two bits are not participate in the soft calculating process of measuring algorithm wherein because of the 57 and the 58 bit.In view of the above, the 5th equation of being derived by the second party formula is to calculate above-mentioned 114 soft root-mean-square values of measuring of encrypting bit.

$m_2=\sqrt{\frac{\underset{i=0}{\overset{113}{\mathrm{\Σ}}}{(\mathrm{soft}-\mathrm{metric}\left(i\right))}^2}{114}}$ The 5th equation

If the product of this two output valve, judges then that the signal that is received is a noise less than this critical value 150, in view of the above, the frequency discreet value that the automatic frequency controlled function is produced will be rejected; Otherwise if the product of this two output valve is greater than this critical value 150, then above-mentioned frequency discreet value will be accepted.

With reference to shown in Figure 2, it is the block schematic diagram for a pulse train arrangement for detecting 200 that is provided according to another embodiment of the present invention.This pulse train arrangement for detecting 200 comprises soft measure etc. of a homophase module 210, a quadrature module 220, and changes module 230, a root mean square module 240, with quality decision module 250.Wherein, the input signal of being accepted 260 comprises an in-phase input signals for 1 262 and a positive input signal 264.The event of being convenient to illustrate, this in-phase input signals for 1 262 also is designated as r _I, and this positive input signal 264 is designated as r _Q

In present embodiment, this homophase module 210 is in order to calculate the quadratic sum of the in-phase input signals for 1 262 that is received.With above-mentioned in-phase input signals for 1 262 is input, and this homophase module 210 is to derive an output valve m according to following the 6th equation ₃:

$m_3=\underset{i}{\mathrm{\Σ}}{\left|r_I\left(i\right)\right|}^2$ The 6th equation

Similarly, this quadrature module 220 is in order to calculate the quadratic sum of the positive input signal 264 that is received.With above-mentioned positive input signal 264 is input, and this quadrature module 220 is to derive an output valve m according to following the 7th equation ₄:

$m_4=\underset{i}{\mathrm{\Σ}}{\left|r_Q\left(i\right)\right|}^2$ The 7th equation

In present embodiment, this is soft measure etc. change module 230 be carry out above-mentioned by change algorithms such as Ke He and Ba Er proposed.With above-mentioned input signal 260, that is in-phase input signals for 1 262 is input with positive input signal 264, and this is soft measures that to wait change module 230 be to derive an above-mentioned soft sequence output valve m who the change algorithm such as measures according to this input signal 260 ₅Then, in receiving this sequence output valve m ₅After, can produce a root-mean-square value m according to following all directions formula ₆, this root-mean-square value m ₆It also is the output valve of above-mentioned root mean square module 240.

$m_6=\sqrt{\frac{\underset{i}{\mathrm{\Σ}}{\left(m_5\right)}^2}{N}}$ The all directions formula

Wherein be for corresponding to the number of the enciphered data in the pulse train.

As shown in Figure 2, divided by above-mentioned output valve m ₃, m ₄, with m ₆Beyond input, this quality decision module 250 also receives a prior determined critical value 270.This quality decision module 250 is with this two output valve m ₃With m ₄With derivative and m ₆The multiply each other product of gained compares the boolean results 280 that the back produces this pulse train arrangement for detecting 200 to this critical value 270.Quality decision module 250 performed functions are can following the 9th equation represented thus:

$\mathrm{quality}-\mathrm{outcome}=\frac{m_6}{m_3+m_4}?\mathrm{threshold}$ The 9th equation

If this product, judges then that the signal that is received is a noise less than this critical value 270; Otherwise, then be regarded as real signal.

In another preferable example of this embodiment of the present invention, above-mentioned the 6th, the 7th when being applied to the system of global system for mobile telecommunications service, the parameter in the formula can be designated from all directions with the.The pulse train content of the communication channel of global system for mobile telecommunications service is transmitting than inferior form according to first table.In view of the above, it is as follows to derive the tenth equation that can be applicable to the global system for mobile telecommunications service by the 6th equation:

$m_3=\underset{i=0}{\overset{147}{\mathrm{\Σ}}}{\left|r_I\left(i\right)\right|}^2$ The tenth equation

Similarly, it is as follows to derive the 11 equation that can be applicable to the global system for mobile telecommunications service by the 7th equation:

$m_4=\underset{i=0}{\overset{147}{\mathrm{\Σ}}}{\left|r_Q\left(i\right)\right|}^2$ The 11 equation

Comprise 116 in the pulse train for the global system for mobile telecommunications service shown in first table and encrypt bit (the 0th to the 115 bit); Wherein because of the 57 and the 58 bit is as the flag of controlling channel signals, so two ratios are not participate in the soft calculating process of measuring algorithm.In view of the above, the 12 equation of being derived by all directions formula is to calculate above-mentioned 114 soft root-mean-square values of measuring of encrypting bits.

$m_6=\sqrt{\frac{\underset{i=0}{\overset{113}{\mathrm{\Σ}}}{\left(m_5\right)}^2}{114}}$ The 12 equation

If this product, judges then that the signal that is received is a noise less than this critical value 270, in view of the above, the frequency discreet value that the automatic frequency controlled function that defines in the global system for mobile telecommunications service regulation is produced will be rejected; Otherwise if the product of this two output valve is greater than this critical value 270, then above-mentioned frequency discreet value will be accepted.

As shown in Figure 3, it is to be the schematic flow sheet according to the pulse train method for detecting that one embodiment of the invention provided.At first, carry out step 310, calculate the inverse of received signal square; Then, carry out step 320, derive the root-mean-square value of a soft measurement value according to received signal; At last, in step 330, the product of the root-mean-square value of the inverse of above-mentioned steps 310 gained and above-mentioned steps 320 gained and a given critical value are compared, and then produce a boolean results judged result of quality decision for this reason.If this product, judges then that the signal that is received is a noise less than this critical value, otherwise, then be regarded as real signal.

Should notice that above-mentioned steps 310 and step 320 are to carry out simultaneously, otherwise or carry out with order mutually; The present invention is all applicable to above-mentioned application.

In another preferable example of this embodiment of the present invention, when being applied to the system of global system for mobile telecommunications service, in the step 310 received signal be for first the table shown last 140 8 bits received signals be 148 bits, but during signal calculated intensity, can choose the intensity that whole or outstanding partial bit is wherein come characterization signal as required; Wherein, soft in the step 320 to measure algorithm be that to encrypt bits (the 0th to the 56 bit and the 59 to the 115 bit) with 114 be input value.

Apparently, according to the description among the top embodiment, the present invention has many corrections and difference.Therefore need be understood in the scope of its additional claim item, except above-mentioned detailed description, the present invention can also implement in other embodiment widely.Above-mentioned is preferred embodiment of the present invention only, is not in order to limit claim of the present invention; Every other do not break away from the equivalence of being finished under the disclosed spirit and changes or modification, all should be included in the following claim.

Claims (11)

1. pulse train arrangement for detecting that is applied to radio receiver is characterized in that this pulse train arrangement for detecting comprises:

One square of module, in order to calculate an input signal square an inverse;

The one soft root mean square module of measuring is in order to a root mean square of a soft metric calculation value of this input signal of deriving; And

One quality decision module, wherein above-mentioned quality decision module calculate this inverse with should be root mean square long-pending, and will amass and a given critical value compares, to produce boolean's comparative result.

2. a pulse train arrangement for detecting that is applied to radio receiver as claimed in claim 1 is characterized in that wherein above-mentioned input signal comprises the input signal and an input signals in quadrature of a homophase.

3. a pulse train arrangement for detecting that is applied to radio receiver as claimed in claim 1 is characterized in that, wherein above-mentioned input signal is to quantize the conversion digital format signal that radio signal produced through an analog/digital converter.

4. a pulse train arrangement for detecting that is applied to radio receiver as claimed in claim 1 is characterized in that, wherein above-mentioned input signal is contained in the 140 interior 8 bits of a pulse train of global system for mobile telecommunications service.

5. a pulse train arrangement for detecting that is applied to radio receiver as claimed in claim 1 is characterized in that, wherein above-mentioned soft measure calculation is to carry out according to the encryption bit in the pulse train that is contained in the global system for mobile telecommunications service.

6. a pulse train arrangement for detecting that is applied to radio receiver as claimed in claim 5, it is characterized in that wherein above-mentioned soft measure calculation is to encrypt bit according to 114 in the pulse train that is contained in the global system for mobile telecommunications service to carry out.

7. pulse train arrangement for detecting that is applied to radio receiver is characterized in that this pulse train arrangement for detecting comprises:

One homophase module is in order to the quadratic sum of the input signal that calculates a homophase;

One quadrature module is in order to calculate the quadratic sum of an input signals in quadrature;

One soft measures etc. changed module, and wherein above-mentioned soft measuring waits the change module to derive a soft sequence output valve of change such as measuring according to an input signal;

One root mean square module, wherein above-mentioned root mean square module calculates a root-mean-square value according to this soft sequence output valve of change such as measuring; And

One quality decision module, wherein above-mentioned quality decision module calculate these two quadratic sum sums one reciprocal with should be root mean square long-pending, and will amass and a given critical value compares, to produce boolean's comparative result.

8. pulse train method for detecting that is applied to radio receiver is characterized in that this pulse train method for detecting comprises:

Calculate an input signal square an inverse;

The derive root mean square of a soft metric calculation value of this input signal; And

Calculate this inverse with should be root mean square long-pending, and will amass and a given critical value compares, to produce boolean's comparative result.

9. a pulse train method for detecting that is applied to radio receiver as claimed in claim 8 is characterized in that wherein above-mentioned input signal comprises the input signal and an input signals in quadrature of a homophase.

10. a pulse train method for detecting that is applied to radio receiver as claimed in claim 8 is characterized in that, wherein above-mentioned input signal is to quantize the conversion digital format signal that radio signal produced through an analog/digital converter.

11. a pulse train method for detecting that is applied to radio receiver as claimed in claim 8 is characterized in that, wherein above-mentioned soft measure calculation is to carry out according to the encryption bit in the pulse train that is contained in the global system for mobile telecommunications service.

Images