CN1845541A - Blind modulation classification detecting method and apparatus and decoding method and apparatus - Google Patents

Abstract

The provided detection method and device for blind modulation mode to detect the wireless block burst with lots of bursts in receiver of wireless communication system comprise: a) calculating current determinant variable corresponding to modulation modes from at least two different modulation modes; b) to every mode, combining at least one former determinant variable in same block with current variable; and c) comparing the variables of before and after combining to decide the current as one of two modulation modes with the determinant variable as the maximal channel tap energy, noise energy, channel energy, SNR, or them combination. Besides, it also provides a new decode method and device. This invention can improve block error code.

Description

Blind modulation classification detection method and device and coding/decoding method and device

Technical field

The present invention relates to wireless communication field, particularly be used for blind modulation classification detection method and device and the coding/decoding method and the device of the receiver of wireless communication system.

Background technology

Such as EDGE (Enhanced Data GSM Environment, enhanced data rates global system for mobile communications evolution technology) or EGPRS (Enhanced General Packet Radio Service, the enhanced general grouping service wireless) in etc. the wireless communication system, at PDTCH (Packet DataTransmission Channel, packet data traffic channel) uses two kinds of modulation classification GMSK (Gaussian Minimum Shift Keying on the channel, Guassian Minimum Shift Keying GMSK) and 8PSK (8-PhaseShift Keying, 8 phase phase-shift keyings), but any modulation classification of employing is not prenoticed the receiver to correspondence on current PDTCH, so that save Radio Resource.In the receiver of wireless communication system, blind modulation classification detects for selecting follow-up equilibrium and PDTCH data (specifically, Radio Link control/media access control layer (RLC/MAC from the upper strata, Radio Link Control/MediumAccess Control) data block) decoding and USF (Uplink State Flag, uplink state flags) decoding is absolutely necessary.Like this, multiple blind modulation classification detection method has been proposed in the prior art, for example referring to below with reference to document: (1) non-patent literature (books): E.Seurre, P.Savelli, P.Pietri, " EDGE for Mobile Internet ", Artech House, 2003; (2) patent documentation: publication No. is US2004/0097207A1, and denomination of invention is " Receiver to Determine Modulation Type ".

Summary is got up, in these existing documents and materials, detect for the blind modulation classification between GMSK and these two kinds of modulation classifications of 8PSK, proposed modulation classification detection algorithm based on channel energy, noise energy or SNR (Signal toNoise Ratio, signal to noise ratio) criterion.Channel estimating is for these criterions normally necessary, and in these documents, emphasized to utilize based on LS (Least Square, least square) or LSSE (Least Sum of Squared Errors, least mean-square error and)) channel estimating improve channel estimating performance, this will separate burst synchronization and channel estimation task before the equilibrium treatment of receiver, and the disposal ability of receiver is produced further burden.Based on the general flow figure of the detection method of these criterions as shown in Figure 1.Below with reference to Fig. 1 it is specifically described.

At first, receive the code element sample (step S110) of burst.Then, at the GMSK or the 8PSK modulation classification of each supposition, calculate channel energy, noise energy or SNR between burst period respectively.

Specifically, at step S120a (or step S120b), GMSK (or 8PSK) modulation classification at supposition, with the code element sample despining of the burst that received and the corresponding predetermined phase pi/2 of GMSK modulation classification (or with the corresponding predetermined phase 3pi/8 of 8PSK modulation classification), perhaps will be stored in the code element sample rotation pi/2 (or 3pi/8) of the known training sequence in the receiver.

Next, at step S130a (or step S130b), because in wireless communication system such as EDGE or EGPRS etc., 26 central 16 training symbol samples in centre of training symbol sample altogether of known training sequence have very good correlation properties, therefore by moving maximum 5 code element samples forward and backward, 16 training symbol samples in centre of known training sequence are despun with process, the burst code element sample that receives during the training sequence of burst moves relevant, perhaps will move through 16 the training symbol samples in centre of the known training sequence of rotation and the burst code element sample that during the training sequence that happens suddenly, receives relevant, thereby can obtain 11 correlations altogether.

Next, at step 140a (or step S140b), to delay composing the window size of selecting according to channel, from these 11 correlations, form a series of correlation windows, and search has the correlation window of ceiling capacity from this correlation window series, thereby realize the synchronous of burst, wherein first channel tap is synchronous with first tap of the correlation window with ceiling capacity.

After synchronously, if adopt the LS channel estimating, then this flow process is advanced along the path that indicates with " A " shown in Figure 1, otherwise this flow process is advanced along the path that indicates with " B " shown in Figure 1.That is to say that under the situation of channel energy criterion, if adopt the LS channel estimating, then this flow process advances to step 150a (or step S150b), wherein further carry out the channel estimating based on LS, this flow process advances to step S170 then.On the other hand, if do not adopt the LS channel estimating, then this flow process directly advances to step S170.Next, at step S170, in the channel energy that compares between GMSK that supposes and the 8PSK modulation classification between burst period, and carry out modulation classification according to comparative result and judge, thereby finish this flow process.Similarly, under the situation of noise energy or SNR criterion, if adopt the LS channel estimating, then this flow process advances to step 150a (or step S150b), wherein further carry out the channel estimating based on LS, this flow process advances to step 160a (or step S160b) then.On the other hand, if do not adopt the LS channel estimating, then this flow process directly advances to step 160a (or step S160b).Next, at step 160a (or step S160b), by the known training sequence in the receiver is passed through estimated channel, the received signal of reconstruct training period, then with actual reception to the signal subtraction of training period estimate the noise and the energy thereof of training period perhaps further to calculate SNR.Next, at step S180, at the noise energy or the SNR that compare between GMSK that supposes and the 8PSK modulation classification between burst period, and carry out modulation classification according to comparative result and judge, thereby finish this flow process.

From as can be seen last, all above-mentioned detection strategies are usually all based on burst synchronization and channel estimating, and the LS channel estimating has produced further burden to the disposal ability of receiver, and apparently, the computation complexity of noise criteria and SNR criterion is higher than the computation complexity of channel energy criterion.

As mentioned above, though the LS channel estimating may be more accurate than based on moving relevant channel estimating, the LS channel estimating has produced further burden to the disposal ability of receiver.In addition, though detecting the detection that is better than slightly on the performance based on noise criteria and channel energy criterion based on the detection of SNR criterion, it must carry out two division arithmetics, so its computation complexity is higher.For detection based on noise criteria, on the channel that the difference of stipulating in the wireless communication system such as EDGE or EGPRS etc. delays composing, it not necessarily gets better than the detected representation based on the channel energy criterion, that is to say, though consumed more disposal ability, do not guaranteed more performance.

In addition, above-mentioned modulation classification detection method is only carried out the modulation classification detection based on the burst of current reception.Can know that in the wireless communication system such as EDGE or EGPRS etc., a radio block is made up of 4 in successive frame bursts usually, and can determine that all 4 bursts have identical modulation classification.Therefore can consider that the modulation detection of utilizing this information to make next 3 bursts is more reliable than the 1st burst.

In addition, in the prior art, for data decode or USF decoding, normally only judge under all identical situation and just decode at the modulation classification of all 4 bursts of forming radio block, so data or USF BLER (Block Error Rate, bLock error rate) performance will be subject to the probability of false detection of the modulation classification of first burst, and the more reliable modulation classification of 3 bursts in back in the same wireless piece detects useless.Therefore can consider to be designed to utilize the more reliable modulation of next 3 bursts to judge dexterously data or USF decoding policy.

Summary of the invention

The present invention proposes in view of the said circumstances of prior art, and its purpose is detection of the blind modulation classification in the wireless communication system and decoding processing are improved, to improve data or the USF BLER performance under cochannel and adjacent-channel interference situation.

According to one embodiment of the invention, a kind of method that is used for detecting at the receiver of wireless communication system the modulation classification of the burst in the radio block of being made up of a plurality of bursts is provided, said method comprising the steps of:, calculate the maximum channel tap energy between described burst period a) respectively at each modulation classification at least two different modulating standards; And b) between the maximum channel tap energy of calculating, compare for described each modulation classification, and will be wherein the modulation classification of maximum channel tap energy maximum be judged to be the modulation classification of described burst.Should be based on the complexity of the novel detection method of maximum channel tap energy criteria less than traditional detection method based on noise criteria or SNR criterion, and only slightly more complicated in detection method (if not adopting the LS channel estimating) based on the channel energy criterion.But emulation demonstrates maximum channel tap energy criteria proposed by the invention and provides better detection performance compromise than channel energy criterion and noise criteria on the different channel types of delaying composing, and only is worse than the most complicated SNR criterion slightly.

According to another embodiment of the present invention, a kind of method that is used for detecting at the receiver of wireless communication system the modulation classification of the burst in the radio block of being made up of a plurality of bursts is provided, said method comprising the steps of: a), calculate the decision variable of current burst respectively at each modulation classification at least two different modulating standards; B), that the decision variable of the decision variable of at least one burst block of front in the same radio block and current burst is combined respectively at described each modulation classification; And c) between decision variable described each modulation classification, after the combination, compares, and according to comparative result, the modulation classification of current burst is judged to be in described at least two different modulating standards one.By this modulation classification detection method, the decision variable that happens suddenly previously in the same wireless piece such as maximum channel tap energy, noise energy, channel energy, signal to noise ratio or its combination in any etc. are combined in the current burst, thereby can carry out more reliable judgement current burst.

According to another embodiment of the present invention, a kind of method of the radio block of being made up of a plurality of bursts being decoded at the receiver of wireless communication system of being used for is provided, said method comprising the steps of: a) at each burst in described a plurality of bursts of described radio block, by the blind modulation classification detection method, from least two different modulating standards, detect the modulation classification of described each burst; B) according to the described testing result of described each burst, equilibrium is carried out in described each burst, and obtained soft output separately; C),, determine to have the modulation classification of higher reliability according to number and/or its relative position in described radio block of the burst that belongs to described each modulation classification at each modulation classification in described at least two different modulating standards; And d) based on the soft output of described each burst, described radio block is decoded, wherein for belonging to described burst in described a plurality of bursts with modulation classification of higher reliability, its soft output is contributed described decoding, and, its soft output is contributed to described decoding for not belonging to described burst in described a plurality of bursts with modulation classification of higher reliability.Compare with traditional coding/decoding method, the Block Error Rate of the coding/decoding method that this is novel will not be subject to the modulation classification probability of false detection of first burst, thereby can improve overall data or USF BLER performance, and, provide bigger surplus for data under cochannel and the adjacent-channel interference situation or USF BLER performance requirement.

According to another embodiment of the present invention, blind modulation classification checkout gear in a kind of receiver of wireless communication system is provided, be used for detecting the modulation classification of the burst of the radio block of forming by a plurality of bursts, comprise: maximum channel tap energy calculation unit, be used for respectively calculating the maximum channel tap energy between described burst period at each modulation classification of at least two different modulating standards; And modulation classification identifying unit, be used between the maximum channel tap energy of calculating for described each modulation classification, comparing by described maximum channel tap energy calculation unit, and will be wherein the modulation classification of maximum channel tap energy maximum be judged to be the modulation classification of described burst.

According to another embodiment of the present invention, a kind of receiver of wireless communication system is provided, it receives the burst in the radio block of being made up of a plurality of bursts, described receiver comprises: the decision variable computing unit, be used for respectively calculating the decision variable of described burst at each modulation classification of at least two different modulating standards; The decision variable assembled unit, be used for respectively at described each modulation classification, the decision variable of at least one burst block of front in the decision variable of the current burst that will be calculated by described decision variable computing unit and the same radio block of being calculated by described decision variable computing unit is combined; And modulation classification identifying unit, be used between decision variable, comparing by described each modulation classification after the described decision variable assembled unit combination, and, the modulation classification of current burst is judged to be in described at least two different modulating standards one according to comparative result.

According to another embodiment of the present invention, decoding device in a kind of receiver of wireless communication system is provided, be used for the radio block of being made up of a plurality of bursts is decoded, described decoding device comprises: the blind modulation classification detecting unit, be used for each burst at described a plurality of bursts of described radio block, by the blind modulation classification detection method, from least two different modulating standards, detect the modulation classification of described each burst; Balanced unit is used for the modulation classification according to described each burst that is detected by described blind modulation classification detecting unit, and equilibrium is carried out in described each burst, and obtains soft output separately; Higher reliability modulation classification determining unit, be used for each modulation classification at described at least two different modulating standards, according to number or its relative position in described radio block of the burst that belongs to described each modulation classification, determine to have the modulation classification of higher reliability; And decoding unit, be used for soft output based on described each burst that obtains by described balanced unit, described radio block is decoded, wherein for belonging to described burst in described a plurality of bursts with modulation classification of higher reliability, its soft output is contributed described decoding, and, its soft output is contributed to described decoding for not belonging to described burst in described a plurality of bursts with modulation classification of higher reliability.

Description of drawings

In conjunction with the drawings the preferred embodiments of the present invention are described in detail, above and other objects of the present invention, characteristic, advantage will become apparent, wherein:

Fig. 1 shows the general flow figure of blind modulation classification detection method of the prior art;

Fig. 2 show according to of the present invention, based on the exemplary process diagram of the blind modulation classification detection method of maximum channel tap energy criteria;

Fig. 3 show according to of the present invention, by making up the exemplary process diagram of the method that modulation classification that the decision variable that happens suddenly previously in the same radio block carries out current burst detects;

Fig. 4 shows the general flow figure according to coding/decoding method of the present invention;

Fig. 5 show according to of the present invention, with the block diagram of the corresponding blind modulation classification checkout gear of blind modulation classification detection method shown in Figure 2;

Fig. 6 show according to of the present invention, with the block diagram of the corresponding blind modulation classification checkout gear of blind modulation classification detection method shown in Figure 3; And

Fig. 7 show according to of the present invention, with the block diagram of the corresponding decoding device of coding/decoding method shown in Figure 4.

Embodiment

Each embodiment of the present invention describes in the context of EDGE or EGPRS system, and wherein Tu Fa modulation classification is GMSK or 8PSK.But it will be clear to one skilled in the art that each embodiment of the present invention can be used for any other wireless communication system, and the modulation classification of burst also can be any other modulation classification.

Below with reference to flow chart description shown in Figure 2 according to of the present invention, based on the blind modulation classification detection method of maximum channel tap energy criteria.

At first, at step S210, receive the code element sample of burst.Then, respectively at the GMSK or the 8PSK modulation classification of each supposition, parallel or calculate maximum channel tap energy between burst period (step S220a to S240a or step S220b to S240b) serially.

Specifically, at step S220a (or step S220b), GMSK (or 8PSK) modulation classification at supposition, with the code element sample despining of the burst that received and the corresponding predetermined phase pi/2 of GMSK modulation classification (or with the corresponding predetermined phase 3pi/8 of 8PSK modulation classification), perhaps will be stored in the code element sample rotation pi/2 (or 3pi/8) of the known training sequence in the receiver.

Next, at step S230a (or step S230b), move all or part of code element sample of known training sequence burst code element sample despun with process, that during the training sequence of burst, receive relevant, perhaps will move through all or part of code element sample of the known training sequence of rotation and the burst code element sample that during the training sequence that happens suddenly, receives relevant, thereby obtain one group of correlation.It should be noted that, because as previously mentioned in wireless communication system such as EDGE or EGPRS etc., in 26 training symbol samples altogether of known training sequence, middle 16 training symbol samples have very good correlation properties, therefore preferably adopt 16 training symbol samples in centre of known training sequence to move relevant, in relevant treatment, move maximum 5 code element samples thus respectively forward and backward, thereby can obtain 11 correlations altogether.

Next, at step S240a (or step S240b),, obtain the maximum channel tap energy between burst period according to resulting this group correlation.Under difference required, this processing can realize by different modes.For example, at computation complexity with calculate between the accuracy and more pay attention under the situation of computation complexity, can search for maximum related value in this group correlation as the maximum channel tap energy between burst period.In addition, between computation complexity and calculating accuracy, more pay attention to calculating under the situation of accuracy, this processing can realize in the following manner: to delay composing the window size of selecting according to channel, from this group correlation, form a series of correlation windows, and search has the correlation window of ceiling capacity from this correlation window series, search for the maximum related value in this correlation window then, as the maximum channel tap energy between burst period with ceiling capacity.In these two kinds of implementations that provide as example, the latter is in the above

Preferred implementation.

At last, at step S250, the maximum channel tap energy that between these two modulation classification GMSK that suppose and 8PSK, is relatively obtained, and select that bigger modulation classification of the central maximum channel tap energy of these two modulation classifications as the modulation classification of being judged.

From as can be seen last, with compare based on the modulation classification detection method of noise criteria or SNR criterion, this complexity that detects strategy is less, and especially, it has got rid of the extra process of LS channel estimation methods, if and do not consider the LS channel estimating, it only slightly more complicated in modulation classification based on the channel energy criterion.But emulation demonstrates it can produce the compromise of good detection performance between the channel type that the difference of EDGE/EGPRS defined delays composing.

According to a further aspect in the invention, proposed when receiver carries out the modulation classification detection to the burst except first burst of a radio block, the decision variable of the decision variable that happens suddenly previously and current burst is combined, so that more reliable judgement is carried out in current burst.Below with reference to exemplary process diagram shown in Figure 3 this modulation classification detection method is described.

At first, at step S310, receive the code element sample of current burst.Then, at step S320a or step S320b, GMSK or 8PSK modulation classification at supposition, calculate the decision variable of current burst, wherein this decision variable for example is maximum channel tap energy, noise energy, channel energy, signal to noise ratio or its combination in any between burst period, but is not limited to the above-mentioned decision variable of enumerating.Next, at step S330a or step S330b, GMSK or 8PSK modulation classification at supposition, the decision variable of the decision variable of at least one burst block of front in the same radio block and current burst is combined, wherein preferably, the decision variable of the decision variable of all burst blocks of front in the same radio block and current burst is combined.At last, at step S340, the decision variable between the GMSK of supposition and 8PSK modulation classification after the comparison combination, and judge that according to comparative result the modulation classification of current burst is GMSK or 8PSK modulation classification.

Form and 4 bursts in the radio block are expressed as burst by 4 bursts at a radio block below according to time sequencing ₀, burst ₁, burst ₂And burst ₃Situation under, specifically describe the preferred implementation of above-mentioned detection method respectively at maximum channel tap energy criteria, noise energy criterion, channel energy criterion or signal-to-noise ratio (SNR) Criterion.

Under detection case, if will be a burst in the radio block at GMSK and 8PSK modulation classification respectively based on maximum channel tap energy criteria _iThe maximum channel tap energy that (i=0,1,2,3) calculates is expressed as H respectively _{G, max}(i), i=0,1,2,3 and H _{P, max}(i), i=0,1,2,3, wherein i=0,1,2,3 corresponds respectively to burst _i(i=0,1,2,3), the then following modulation classification that carries out is judged:

If $\underset{i=0}{\overset{n}{\mathrm{\Σ}}}{H}_{g,\max }\left(i\right)\≥\underset{i=0}{\overset{n}{\mathrm{\Σ}}}{H}_{p,\max }\left(i\right),$ Then with burst _nThe modulation classification of (n=0,1,2,3) is judged to be GMSK,

Else if $\underset{i=0}{\overset{n}{\mathrm{\&Sigma;}}}{H}_{g,\max }\left(i\right)<\underset{i=0}{\overset{n}{\mathrm{\&Sigma;}}}{H}_{p,\max }\left(i\right),$ Then with burst _nThe modulation classification of (n=0,1,2,3) is judged to be 8PSK.

Under detection case, if will be a burst in the radio block at GMSK and 8PSK modulation classification respectively based on the channel energy criterion _iThe channel energy that (i=0,1,2,3) calculates is expressed as E respectively _g(i), i=0,1,2,3 and E _p(i), i=0,1,2,3, wherein i=0,1,2,3 corresponds respectively to burst _i(i=0,1,2,3), the then following modulation classification that carries out is judged:

If $\underset{i=0}{\overset{n}{\mathrm{\&Sigma;}}}{E}_g\left(i\right)\&GreaterEqual;\underset{i=0}{\overset{n}{\mathrm{\&Sigma;}}}{E}_p\left(i\right),$ Then with burst _nThe modulation classification of (n=0,1,2,3) is judged to be GMSK,

Else if $\underset{i=0}{\overset{n}{\mathrm{\&Sigma;}}}{E}_g\left(i\right)<\underset{i=0}{\overset{n}{\mathrm{\&Sigma;}}}{E}_p\left(i\right),$ Then with burst _nThe modulation classification of (n=0,1,2,3) is judged to be 8PSK.

Under detection case, if will be a burst in the radio block at GMSK and 8PSK modulation classification respectively based on the noise energy criterion _iThe noise energy that (i=0,1,2,3) calculates is expressed as noise respectively _g(i), i=0,1,2,3 and noise _p(i), i=0,1,2,3, wherein i=0,1,2,3 corresponds respectively to burst _i(i=0,1,2,3), the then following modulation classification that carries out is judged:

If $\underset{i=0}{\overset{n}{\mathrm{\&Sigma;}}}\mathrm{nois}{e}_g\left(i\right)<\underset{i=0}{\overset{n}{\mathrm{\&Sigma;}}}{\mathrm{noise}}_p\left(i\right),$ Then with burst _nThe modulation classification of (n=0,1,2,3) is judged to be GMSK,

Else if $\underset{i=0}{\overset{n}{\mathrm{\&Sigma;}}}{\mathrm{noise}}_g\left(i\right)\&GreaterEqual;\underset{i=0}{\overset{n}{\mathrm{\&Sigma;}}}{\mathrm{noise}}_p\left(i\right),$ Then with burst _nThe modulation classification of (n=0,1,2,3) is judged to be 8PSK.

Under the detection case based on signal-to-noise ratio (SNR) Criterion, by above-mentioned representation, the following modulation classification that carries out is judged:

If $\frac{\underset{i=0}{\overset{n}{\mathrm{\&Sigma;}}}{E}_g\left(i\right)}{\underset{i=0}{\overset{n}{\mathrm{\&Sigma;}}}{\mathrm{noise}}_g\left(i\right)}\&GreaterEqual;\frac{\underset{i=0}{\overset{n}{\mathrm{\&Sigma;}}}{E}_p\left(i\right)}{\underset{i=0}{\overset{n}{\mathrm{\&Sigma;}}}{\mathrm{noise}}_p\left(i\right)},$ Then with burst _nThe modulation classification of (n=0,1,2,3) is judged to be GMSK,

Else if $\frac{\underset{i=0}{\overset{n}{\mathrm{\&Sigma;}}}{E}_g\left(i\right)}{\underset{i=0}{\overset{n}{\mathrm{\&Sigma;}}}{\mathrm{noise}}_g\left(i\right)}<\frac{\underset{i=0}{\overset{n}{\mathrm{\&Sigma;}}}{E}_p\left(i\right)}{\underset{i=0}{\overset{n}{\mathrm{\&Sigma;}}}{\mathrm{noise}}_p\left(i\right)},$ Then with burst _nThe modulation classification of (n=0,1,2,3) is judged to be 8PSK.

Because therefore the utilisable energy that happens suddenly previously in the same radio block of combination, can infer that the modulation classification of back burst detects the modulation classification detection performance that performance is better than more preceding burst so that the modulation classification detection of current burst is more reliable.Like this, in data decode or USF decoding, in order to utilize the better detection performance of back burst, under the situation that a radio block is made up of 4 bursts, should utilize " last n; n 〉=2 or n, n 〉=3 " or " any n, n 〉=3 " strategy to improve conceptual data or USF BLER performance under cochannel and the adjacent-channel interference situation.Specifically, for " last n, n 〉=2 or n; n 〉=3 " strategy, if the last n in radio block, n 〉=2 or n, the modulation classification of n 〉=3 burst is judged identical, then make these bursts that the equilibrium treatment by corresponding types obtains soft output decoding processing is contributed; For " any n, n 〉=3 " strategy, if any n, the modulation classification of n 〉=3 burst is judged identical, then make these bursts that the equilibrium treatment by corresponding types obtains soft output decoding processing is contributed.Apparently, such decoding policy is not limited to above-mentioned specific design, but goes for any decoding processing in any wireless communication system, so that improve the overall performance after the decoding.Describe according to general solution code method of the present invention below with reference to the flow chart of Fig. 4.

At first, at step S410, receive the burst in the radio block.Then, at step S420, by the blind modulation classification detection method, from at least two different modulating standards, detect the modulation classification of burst, wherein this blind modulation classification detection method can be traditional based on the channel energy criterion, the detection method of noise energy criterion or signal-to-noise ratio (SNR) Criterion, also can be according to the detection method based on maximum channel tap energy criteria of the present invention, according to of the present invention with reference to figure 3 described detection methods, perhaps any other blind modulation classification detection method, and according to of the present invention are preferred blind modulation classification detection methods in this coding/decoding method with reference to figure 3 described detection methods.

Next, at step S430,, equilibrium is carried out in burst, and obtained its soft output according to the testing result of burst.Then, at step S440, all bursts in judging whether radio block treated.If judge all bursts in the radio block that still is untreated at step S440, then this flow process turns back to step S410, to continue to handle next burst.Otherwise this flow process advances to step S450.

Next, at step S450, number and/or its relative position in radio block according to the burst that belongs to each modulation classification, in these two different modulating standards, determine to have the modulation classification of higher reliability at least, the number of burst that wherein belongs to a modulation classification is many more, then the reliability of this modulation classification is high more, and/or belong to a modulation classification the relative position of burst in radio block lean on more after, then the reliability of this modulation classification is high more.

At last, at step S460, soft output based on each burst in the radio block, radio block is decoded, wherein for belonging to determined burst with modulation classification of higher reliability, its soft output is contributed described decoding, and, its soft output is contributed to described decoding for not belonging to determined burst with modulation classification of higher reliability.

Apparently, according to instruction herein, those skilled in the art can design different decoding policies in light of the circumstances, and is not limited to above-mentioned " last n, n 〉=2 or n, n 〉=3 " or " any n, n 〉=3 " strategy.

Below with reference to Fig. 5-7 describe according to of the present invention, with blind modulation classification detection method and corresponding blind modulation classification checkout gear of coding/decoding method and the decoding device shown in Fig. 2-4.

At first, with reference to the block diagram of figure 5 describe according to of the present invention, with the corresponding blind modulation classification checkout gear of blind modulation classification detection method shown in Figure 2.This blind modulation classification checkout gear is used for detecting the modulation classification of the burst of the radio block of being made up of a plurality of bursts, and comprise: maximum channel tap energy calculation unit 510, be used for respectively each modulation classification, calculate the maximum channel tap energy between burst period at least two different modulating standards; And modulation classification identifying unit 520, be used between the maximum channel tap energy of calculating for each modulation classification, comparing by maximum channel tap energy calculation unit 510, and will be wherein the modulation classification of maximum channel tap energy maximum be judged to be the modulation classification of burst.

As shown in the figure, in a preferred embodiment of the invention, maximum channel tap energy calculation unit 510 comprises: despining/rotary unit 511, be used for despining of code element sample and the corresponding phase place of each modulation classification with burst, perhaps the code element sample with the known training sequence in the receiver rotates and the corresponding phase place of each modulation classification; Mobile correlation unit 512, be used for all or part of code element sample of the known training sequence of receiver with move relevant by despining/rotary unit 511 burst code element samples despun, that during the training sequence of burst, receive, perhaps will move by all or part of code element sample of known training sequence despining/rotary unit 511 rotation, in the receiver and the burst code element sample that during the training sequence that happens suddenly, receives relevant, thereby obtain one group of correlation; And maximum channel tap energy acquisition unit 513, be used for obtaining maximum channel tap energy according to this group correlation that obtains by mobile correlation unit 512.

Next, with reference to the block diagram of figure 6 describe according to of the present invention, with the corresponding blind modulation classification checkout gear of blind modulation classification detection method shown in Figure 3.This blind modulation classification checkout gear is used for detecting the modulation classification of the burst of the radio block of being made up of a plurality of bursts, and comprise: decision variable computing unit 610, be used for respectively each modulation classification, calculate the decision variable of burst at least two different modulating standards; Decision variable assembled unit 620, be used for respectively at each modulation classification, the decision variable of at least one burst block of front in the decision variable of the current burst that will be calculated by decision variable computing unit 610 and the same radio block of being calculated by decision variable computing unit 610 is combined; And modulation classification identifying unit 630, be used between decision variable, comparing by each modulation classification after 620 combinations of decision variable assembled unit, and, the modulation classification of current burst is judged to be in these at least two different modulating standards one according to comparative result.

Next, with reference to the block diagram of figure 7 describe according to of the present invention, with the corresponding decoding device of coding/decoding method shown in Figure 4.This decoding device is used for the radio block of being made up of a plurality of bursts is decoded, and comprise: blind modulation classification detecting unit 710, be used for each burst at described a plurality of bursts of radio block, by the blind modulation classification detection method, detect the modulation classification of each burst from least two different modulating standards, wherein blind modulation classification detecting unit 710 can be blind modulation classification checkout gear shown in Figure 5, blind modulation classification checkout gear or any other blind modulation classification checkout gear shown in Figure 6; Balanced unit 720 is used for the modulation classification according to each burst that is detected by blind modulation classification detecting unit 710, and equilibrium is carried out in each burst, and obtains soft output separately; Higher reliability modulation classification determining unit 730, be used for each modulation classification at described at least two different modulating standards, according to number or its relative position in described radio block of the burst that belongs to each modulation classification, determine to have the modulation classification of higher reliability; And decoding unit 740, be used for soft output based on each burst that obtains by balanced unit 720, radio block is decoded, wherein for the burst that belongs to modulation classification in described a plurality of bursts with higher reliability, its soft output is contributed decoding processing, and, its soft output is contributed to decoding processing for the burst that does not belong to modulation classification in described a plurality of bursts with higher reliability.

Skilled in the art will recognize that, blind modulation classification checkout gear of the present invention and method and decoding device and method can be by softwares only, only the mode that combines with hardware of hardware and/or software realizes, therefore, its implementation is not construed as limiting technical scope of the present invention.

In addition, time sequencing described in method of the present invention is not limited to is to specifications carried out, also can be according to other time sequencing, or parallel or carry out independently, therefore, the execution sequence of the method for describing among the present invention is not construed as limiting technical scope of the present invention.

Although abovely the present invention is described in detail with reference to specific embodiment, those skilled in the art will appreciate that under the situation that does not break away from the spirit and scope of the present invention, can make various modifications, replacement and change to embodiments of the invention.

Claims (25)

1. method that is used for detecting at the receiver of wireless communication system the modulation classification of the burst in the radio block of being made up of a plurality of bursts said method comprising the steps of:

A) respectively at each modulation classification at least two different modulating standards, calculate the decision variable of current burst;

B), that the decision variable of the decision variable of at least one burst block of front in the same radio block and current burst is combined respectively at described each modulation classification; And

C) between decision variable described each modulation classification, after the combination, compare, and according to comparative result, the modulation classification of current burst is judged to be in described at least two different modulating standards one.

2. method according to claim 1, wherein said decision variable comprise maximum channel tap energy, noise energy, channel energy and signal to noise ratio one of them and the combination in any thereof between described burst period.

3. method according to claim 2 is that the decision variable that calculates described burst may further comprise the steps under the situation of the maximum channel tap energy between described burst period at described decision variable wherein:

A1) with code element sample despining in the described burst and the corresponding phase place of described each modulation classification, perhaps the code element sample with the known training sequence in the described receiver rotates and the corresponding phase place of described each modulation classification;

A2) all or part of code element sample with the known training sequence in the described receiver moves relevant with the code element sample of process described burst despun, that receive during the training sequence of described burst, perhaps will move through all or part of code element sample of known training sequence rotation, in the described receiver and the code element sample of the described burst that during the training sequence of described burst, receives relevant, thereby obtain one group of correlation; And

A3), obtain described maximum channel tap energy according to described one group of correlation.

4. method according to claim 3, wherein step a3) may further comprise the steps:

A3-1) to delay composing the window size of selecting according to channel, from described one group of correlation, form a series of correlation windows, and from described a series of correlation windows, search for correlation window with ceiling capacity; And

A3-2) the described interior maximum related value of correlation window with ceiling capacity of search is as described maximum channel tap energy.

5. method according to claim 1 also comprises following subsequent step:

D) according to the described modulation classification result of determination of each burst in described a plurality of bursts of described radio block, equilibrium is carried out in described each burst, and obtained soft output separately;

E),, determine to have the modulation classification of higher reliability according to number and/or its relative position in described radio block of the burst that belongs to described each modulation classification at each modulation classification in described at least two different modulating standards; And

F) based on the soft output of described each burst, described radio block is decoded, wherein for belonging to described burst in described a plurality of bursts with modulation classification of higher reliability, its soft output is contributed described decoding, and for not belonging to described burst with modulation classification of higher reliability in described a plurality of bursts, its soft output is contributed described decoding

Wherein said of being decoded as in data decode and the uplink state flags decoding.

6. according to each the described method in the claim 1 to 5, wherein said wireless communication system is in enhanced general wireless packet service system and the enhanced data rates global system for mobile communications evolution technology system, and described at least two different modulating standards comprise Guassian Minimum Shift Keying GMSK and 8 phase-shift keying modulation classifications mutually.

7. method that is used for detecting at the receiver of wireless communication system the modulation classification of the burst in the radio block of being made up of a plurality of bursts said method comprising the steps of:

A), calculate the maximum channel tap energy between described burst period respectively at each modulation classification at least two different modulating standards; And

B) between the maximum channel tap energy of calculating, compare for described each modulation classification, and will be wherein the modulation classification of maximum channel tap energy maximum be judged to be the modulation classification of described burst.

8. method according to claim 7, wherein step a) may further comprise the steps:

A1) with code element sample despining in the described burst and the corresponding phase place of described each modulation classification, perhaps the code element sample with the known training sequence in the described receiver rotates and the corresponding phase place of described each modulation classification;

A2) all or part of code element sample with the known training sequence in the described receiver moves relevant with the code element sample of process described burst despun, that receive during the training sequence of described burst, perhaps will move through all or part of code element sample of known training sequence rotation, in the described receiver and the code element sample of the described burst that during the training sequence of described burst, receives relevant, thereby obtain one group of correlation; And

A3), obtain described maximum channel tap energy according to described one group of correlation.

9. method according to claim 8, wherein step a3) may further comprise the steps:

A3-1) to delay composing the window size of selecting according to channel, from described one group of correlation, form a series of correlation windows, and from described a series of correlation windows, search for correlation window with ceiling capacity; And

A3-2) the described interior maximum related value of correlation window with ceiling capacity of search is as described maximum channel tap energy.

10. according to each the described method in the claim 7 to 9, wherein said wireless communication system is in enhanced general wireless packet service system and the enhanced data rates global system for mobile communications evolution technology system, and described at least two different modulating standards comprise Guassian Minimum Shift Keying GMSK and 8 phase-shift keying modulation classifications mutually.

11. one kind is used for the method for the radio block of being made up of a plurality of bursts being decoded at the receiver of wireless communication system, may further comprise the steps:

A),, from least two different modulating standards, detect the modulation classification of described each burst by the blind modulation classification detection method at each burst in described a plurality of bursts of described radio block;

B) according to the described testing result of described each burst, equilibrium is carried out in described each burst, and obtained soft output separately;

C),, determine to have the modulation classification of higher reliability according to number and/or its relative position in described radio block of the burst that belongs to described each modulation classification at each modulation classification in described at least two different modulating standards; And

D) based on the soft output of described each burst, described radio block is decoded, wherein for belonging to described burst in described a plurality of bursts with modulation classification of higher reliability, its soft output is contributed described decoding, and, its soft output is contributed to described decoding for not belonging to described burst in described a plurality of bursts with modulation classification of higher reliability.

12. method according to claim 11, the reliability of one of them modulation classification is high more to comprise one of them and combination in any thereof of following situation:

The number of burst that belongs to a described modulation classification is many more; And

After the relative position of burst in described radio block that belongs to a described modulation classification leans on more.

13. method according to claim 12 is that four bursts and described at least two different modulating standards are under the situation of two modulation classifications in described a plurality of bursts wherein, during the modulation classification of determining to have higher reliability may further comprise the steps one:

1) if the burst number that belongs to a modulation classification in described two modulation classifications more than or equal to 3, then is defined as described modulation classification with higher reliability with a described modulation classification;

2) if belonging to the burst of a modulation classification in described two modulation classifications is last n burst in the described radio block, wherein n is the integer more than or equal to 2, then a described modulation classification is defined as described modulation classification with higher reliability; And

3) if belonging to the burst of a modulation classification in described two modulation classifications is last n burst in the described radio block, wherein n is the integer more than or equal to 3, then a described modulation classification is defined as described modulation classification with higher reliability.

14. method according to claim 11, wherein said of being decoded as in data decode and the uplink state flags decoding.

15. according to each the described method in the claim 11 to 14, wherein said wireless communication system is in enhanced general wireless packet service system and the enhanced data rates global system for mobile communications evolution technology system, and described at least two different modulating standards comprise Guassian Minimum Shift Keying GMSK and 8 phase-shift keying modulation classifications mutually.

16. the receiver of a wireless communication system, it receives the burst in the radio block of being made up of a plurality of bursts, and described receiver comprises:

The decision variable computing unit is used for respectively calculating the decision variable of described burst at each modulation classification of at least two different modulating standards;

The decision variable assembled unit, be used for respectively at described each modulation classification, the decision variable of at least one burst block of front in the decision variable of the current burst that will be calculated by described decision variable computing unit and the same radio block of being calculated by described decision variable computing unit is combined; And

The modulation classification identifying unit, be used between decision variable, comparing by described each modulation classification after the described decision variable assembled unit combination, and, the modulation classification of current burst is judged to be in described at least two different modulating standards one according to comparative result.

17. receiver according to claim 16, wherein said decision variable comprise maximum channel tap energy, noise energy, channel energy and signal to noise ratio one of them and combination in any thereof between described burst period.

18. receiver according to claim 16 also comprises:

Balanced unit is used for respectively each burst at described a plurality of bursts of described radio block, and the modulation classification according to described each burst of being judged by described modulation classification identifying unit carries out equilibrium to described each burst, and obtains soft output separately;

Higher reliability modulation classification determining unit, be used for each modulation classification at described at least two different modulating standards, according to number or its relative position in described radio block of the burst that belongs to described each modulation classification, determine to have the modulation classification of higher reliability; And

Decoding unit, be used for soft output based on described each burst that obtains by described balanced unit, described radio block is decoded, wherein for belonging to described burst in described a plurality of bursts with modulation classification of higher reliability, its soft output is contributed described decoding, and, its soft output is contributed to described decoding for not belonging to described burst in described a plurality of bursts with modulation classification of higher reliability.

19. according to each the described receiver in the claim 16 to 18, wherein said wireless communication system is in enhanced general wireless packet service system and the enhanced data rates global system for mobile communications evolution technology system, and described at least two different modulating standards comprise Guassian Minimum Shift Keying GMSK and 8 phase-shift keying modulation classifications mutually.

20. the blind modulation classification checkout gear in the receiver of a wireless communication system is used for detecting the modulation classification of the burst of the radio block of being made up of a plurality of bursts, comprising:

Maximum channel tap energy calculation unit is used for respectively calculating the maximum channel tap energy between described burst period at each modulation classification of at least two different modulating standards; And

The modulation classification identifying unit, be used between the maximum channel tap energy of calculating for described each modulation classification, comparing by described maximum channel tap energy calculation unit, and will be wherein the modulation classification of maximum channel tap energy maximum be judged to be the modulation classification of described burst.

21. blind modulation classification checkout gear according to claim 20, wherein maximum channel tap energy calculation unit comprises:

Despining/rotary unit is used for despining of code element sample and the corresponding phase place of described each modulation classification with described burst, and perhaps the code element sample with the known training sequence in the described receiver rotates and the corresponding phase place of described each modulation classification;

Mobile correlation unit, be used for moving relevant with the code element sample of described burst despun by described despining/rotary unit, that during the training sequence of described burst, receive all or part of code element sample of the known training sequence of described receiver, perhaps will move by all or part of code element sample of known training sequence despining/rotary unit rotation, in the described receiver and the code element sample of the described burst that during the training sequence of described burst, receives relevant, thereby obtain one group of correlation; And

Maximum channel tap energy obtains the unit, is used for basis by described one group of correlation that described mobile correlation unit obtains, and obtains described maximum channel tap energy.

22. according to claim 20 or 21 described blind modulation classification checkout gears, wherein said wireless communication system is in enhanced general wireless packet service system and the enhanced data rates global system for mobile communications evolution technology system, and described at least two different modulating standards comprise Guassian Minimum Shift Keying GMSK and 8 phase-shift keying modulation classifications mutually.

23. the decoding device in the receiver of a wireless communication system is used for the radio block of being made up of a plurality of bursts is decoded, described decoding device comprises:

The blind modulation classification detecting unit is used for each burst at described a plurality of bursts of described radio block, by the blind modulation classification detection method, detects the modulation classification of described each burst from least two different modulating standards;

Balanced unit is used for the modulation classification according to described each burst that is detected by described blind modulation classification detecting unit, and equilibrium is carried out in described each burst, and obtains soft output separately;

Higher reliability modulation classification determining unit, be used for each modulation classification at described at least two different modulating standards, according to number or its relative position in described radio block of the burst that belongs to described each modulation classification, determine to have the modulation classification of higher reliability; And

Decoding unit, be used for soft output based on described each burst that obtains by described balanced unit, described radio block is decoded, wherein for belonging to described burst in described a plurality of bursts with modulation classification of higher reliability, its soft output is contributed described decoding, and, its soft output is contributed to described decoding for not belonging to described burst in described a plurality of bursts with modulation classification of higher reliability.

24. decoding device according to claim 23, wherein said decoding unit carry out one in data decode and the uplink state flags decoding.

25. according to claim 23 or 24 described decoding devices, wherein said wireless communication system is in enhanced general wireless packet service system and the enhanced data rates global system for mobile communications evolution technology system, and described at least two different modulating standards comprise Guassian Minimum Shift Keying GMSK and 8 phase-shift keying modulation classifications mutually.

Images