CN101288256B - Method and device for determining indexes allocated to correction symbols - Google Patents

Abstract

The method involves generating coded code symbols using a generator matrix of block codes form a number of source symbols. The code symbols are transmitted from a transmitting device to a receiving device, where transmission error occurs in the received code symbols. Indexes of error correction symbols that are to be transmitted are determined by identifying a region of the code symbols by two parameters. The region can be requested in the form of the correction symbol by the receiving device from the transmitting device for reconstructing the source symbols in an error-free manner. An independent claim is also included for a device for executing a method of determining error correction symbols.; The invention relates to a method for determining indexes (SI) allocated to error correcting symbols (FKS), wherein encoded code symbols (CC) are generated by means of a generator matrix (G) of a block code (BC) from number (K) of source symbols and the encoded code symbols (CC) are transmitted from a transmitting device (SV) to a receiving device (EV), transmission errors (UEF) occur in the received code symbols (C), the indexes of the error correcting symbols (FKS) are determined by unambiguously identifying the area of the encoded code symbols (CC) by means of first and second parameters (Rmin, Rmax), which can be requested in the form of at least one error correcting symbol (FKS) by the receiving device (EV) from the transmitting device (SV) for reconstructing the source symbols (Q) in an error-free manner.

Description

Be used to confirm distribute to the method and apparatus of the index of correction symbols

The present invention relates to a kind of method as described in the preamble and a kind of equipment as described in the preamble according to claim 13 according to claim 1.

With packet, such as audio frequency or video data from transmitter when receiver transmits, owing to the wrong reception to this packet takes place transmission error.What especially in downloading service, possibly need is can carry out reconstruct to all packets at the receiver place error-freely.

In MBMS professional (MBMS-multimedia broadcast/group broadcast service (MultimediaBroadcast/Multicast Service)); Packet is transferred to a plurality of receivers from a transmitter, wherein said MBMS business such as in 3GPP (3GPP-third generation partner program) by standardization.At this, under the situation of different received machine, different packets possibly arrive respective receiver wrongly.

In order to guarantee error free reception, repeatedly transmits data packets makes that wrong reception is extremely impossible.But this execution mode is extremely inefficient, because receive irrelevantly with whether having misconnection, packet sends to all receivers with being repeated.

In addition, except packet, can also be transferred to receiver to the error protection of parity data form.Thereby can correct the wrong packet of maximum quantity, but can not therefore guarantee error free reception.

Replacedly or additionally, after the wrong reception to one or more packet, receiver can be set up point-to-point the connection with transmitter, be used for the wrong packet of mode request with error correction packets.In this case, can ask all packets of mistake specially.Inquire more packets, for the more bandwidth of MBMS service needed.In general, packet, parity code check packet and/or error correction packets can be made up of one or more symbol (Symbole).At this, packet comprises the source symbol, and parity code check packet comprises parity check symbol, and error correction packets comprises error correction symbol.

The present invention based on task be to explain and a kind ofly can select the method and apparatus of error correction symbol simply and effectively in order to error-free ground reconstructed source symbol.

This task solves through independent claims.

At this, can confirm according to the method for the invention a plurality of with divide into groups (Block) relevant error correction symbol, wherein need said error correction symbol be used for error-free ground reconstructed source symbol.At this, for the required bandwidth of request transmission error correction symbol is reduced, reason is with the quantity of required error correction symbol irrelevant, and only two parameters are transferred to transmitting apparatus from receiving equipment.Realize through reducing the transmission time thus, because transmission bandwidth is limited in RTTS.

Other improvement of the present invention provides in the dependent claims.

In preferred extension according to claim 4 and 5, can confirm to minimize the index of quantity and the error correction symbol of being distributed, can save transmission cost thus, because only the error correction symbol of lesser amt must be transmitted.

Preferably, according to claim 6, being applied to system and link nonsystematic block code according to the method for the invention.This realizes that through generating by the sign indicating number matrix of deriving with nonsystematic generator matrix of system wherein, said derivation can be implemented through simple copy operation.

In addition, in favourable flexible program according to the inventive method of claim 7, be applied to two systems with one after the nonsystematic block code that connects.At this, can produce encoder matrix with simple especially mode, wherein can under the situation of no complicated matrix manipulation, produce.

In addition, according to Claim 8 or in the reasonable extension scheme of the inventive method of 9, this method can be applied to the Raptor sign indicating number of nonsystematic or system.At this, can change into two systems to foreign key and internal code with a nonsystematic block code, can confirm encoder matrix simply thus.When using system Raptor sign indicating number, before the application system block code, must mend the correction matrix of nonsystematic block code, wherein when generating encoder matrix, do not consider correction matrix.

Preferably, according to claim 10 or 11, utilize the perhaps nonsystematic block code of system to use according to the method for the invention.Can generate encoder matrix with simple especially mode through duplicating generator matrix at this.

The value of preferably, distributing binary value or Galois territory for source symbol, coding code sign and code sign.Can be applied to the different numerical space thus according to the method for the invention and therefore be used for multiple different application.

The invention still further relates to and a kind ofly be used to implement and the equipment of the relevant with it described method of dependent claims according to the present invention.Therefore can in equipment, especially portable vehicular radio, carry out and implement according to the method for the invention.

Come the present invention and improvement thereof are further explained by accompanying drawing below.

Fig. 1 illustrates the embodiment of the inventive method with transmission and receiving equipment and wrong transmission channel;

Fig. 2 illustrates the flow chart of the inventive method;

Fig. 3 illustrates first flexible program of the inventive method with systematic block codes and nonsystematic block code;

Fig. 4 illustrates the generator matrix according to first flexible program;

Fig. 5 illustrates second flexible program of the inventive method with two systematic block codes and a nonsystematic block code;

Fig. 6 illustrates the generator matrix according to second flexible program;

Fig. 7 illustrates the application of apparatus of the present invention in portable set.

In Fig. 1 to Fig. 7, the element with identical function and working method is equipped with identical Reference numeral.

Through Fig. 1 and 2 to being further described according to the method for the invention.In Fig. 1, can see transmitting apparatus SV, for example MBMS transmitting apparatus (MBMS-multimedia broadcast/group broadcast service).This transmitting apparatus SV comes the source symbol Q of quantity K is encoded through the generator matrix G of block code BCN.Said source symbol Q representes such as voice that compressed or image file.For example the voice document that compressed has been carried out encoding and according to Joint Photographic Experts Group (JPEG-joint photographic experts group) image file that compressed being encoded according to AMR standard (AMR-AMR).Source symbol Q generally can describe the compressed or unpressed data of any possibility.Use vector or matrix literary style to be used for other enforcement.

For following embodiment, use the binary source symbol, such as

Q＝[1?1?0?1] ^T， (1)

Wherein, the first quantity K of source symbol is K=4.

Generator matrix G can reproduce as follows:

In generator matrix G, adopt natural number to describe line index.The mark that allows to discern uniquely the row of institute's mark generally can be used as line index.Except natural number, also can use such as letter or storage address.

Can from K=4 source symbol Q, form N coding code sign CC through said generator matrix G.Because said generator matrix G descriptive system block code BC, so K=4 coding code sign CC and K=4 source symbol Q are consistent, and remaining N-K coding code sign CC is corresponding with parity check symbol P.Reference numeral N is called as the second quantity N.Coding code sign CC is calculated as:

CC＝G×Q＝[1，1，0，1，0，0，1，…] ^T (3)

Now, the coding code sign CC of the 3rd quantity N ' is transmitted, and wherein, the 3rd quantity N ' is less than or equal to the second quantity N.Such as, the coding code sign of a N '=5 is transmitted:

CC＝[1，1，0，1，0] ^T (4)

When the transmitting element SE of code sign CC from transmitting apparatus SV that will encode was transferred to the receiving element EE of receiving equipment EV through transmission channel UE, transmission error UEF appearred.Receiving equipment EV can be constructed to the MBMS receiving platform.Especially through wireless transmission channel UE, such as through according to the mobile radio channel transmission of GSM standard (GSM-global system for mobile communications) or UMTS standard (UMTS-Universal Mobile Telecommunications System) work the time, transmission error UEF appears.In addition, also error possibly appear, such as under IP (IP-Internet Protocol, the LAN-local area network (LAN)) situation through LAN in Wired transmission path.At this, coding code sign CC will be received with the distortion mode, or coding code sign CC is deleted and can not arrive receiving element EE or arrives receiving element EE with the order of changing when transmission.In the present invention, coding code sign CC arrives receiving equipment EV as the code sign C of wrong reception.In Fig. 1, for example, distortion, deletion or transposing be labeled drawings attached mark " X " owing to containing the wrong code sign C that is received.In Fig. 2, being received among the step S21 of code sign C reproduced.

In the present embodiment, the code sign C that is received shows as follows, and wherein the coding code sign CC of the second quantity N all is transmitted, and only is a part wherein:

C＝[X，1，0，X，X] ^T (5)

Therefore, the code sign C with index 2 and 3 is received, or the code sign C with index 1,4 and 5 is not received.Code sign C with index 5 possibly deleted when transmission.Receiving equipment EV does not also know whether the code sign C with index 5 is sent out.Thereby the code sign that only has index 2 and 3 is known by receiving equipment.Index provides the position of code sign in the row vector of (aufspannenten) that launch through all code signs, such as, index 2 is second code signs with value " 1 ", index 5 illustrates the have value last code sign of " X ".In this embodiment, use the binary character that has mark " 0 " and " 1 " to be used for the coefficient of source symbol Q, coding code sign CC, coded identification C and matrix.

In another treatment step S22, confirm institute's largest index of the code sign C of reception to some extent.This is index " 3 " in the above-described embodiments.Next, give the first parameters R min apportioning cost, this value is greater than the index of the last index that receives.Rmin for example is changed to Rmin=4, wherein also can select bigger value.

In another treatment step S23, from the encoder matrix CM that is derived by generator matrix G, form first matrix M 1.Encoder matrix CM is affiliated generator matrix G system or under the nonsystematic block code BC situation.Within the scope of the present invention, notion nonsystematic block code also can be understood that systematic block codes.For example can use LDGM sign indicating number or LPDC sign indicating number (LDGM-low-density generated matrix, LDPC-low density parity check code) as block code.The encoder matrix CM of the block code that is used for successively being implemented explains implementing modification.Because in this embodiment, block code is systematic block codes BC, so encoder matrix CM is consistent with the generator matrix CM in (2).

For the i coded identification C of error free reception from encoder matrix CM through duplicating capable first matrix M 1 that produces of i with the mode (zeilenweise) of row.Therefore, the row that has of encoder matrix CM number is that 2 and 3 capable ZX is copied in first matrix M 1, that is the code sign C with index 2 and 3 is received error-freely.Thereby, obtain first matrix M 1 and be:

$M1=\left[\begin{array}{cccc}0& 1& 0& 0\\ 0& 0& 1& 0\end{array}\right]---\left(6\right)$

In this case, in step S24, each row has been labeled column index, and wherein, corresponding column index SI is corresponding to the row of generator matrix GN number.In order to distinguish coefficient and column index SI, corresponding line index SI is placed in the quotation marks.So row 1 are marked with index ' a ', row 2 are marked with index ' b ', and row 3 are marked with index ' c ', and the rest may be inferred.Thereby second matrix M 2 that is labeled is following:

In first matrix M 1, used letter to describe column index.Usually, the mark that allows to identify uniquely the row of institute's mark can be used as column index.Except letter, also can use natural number or storage address.

In another step S25, through elementary row conversion and/or row transposing first matrix M 1 is transformed to second matrix, wherein produce incoherent capable ZU RG as a result of, wherein RG is corresponding to the order of second matrix M 2.According to [1] the 61st page, operation is known below under the elementary row conversion:

(I) delegation and other repeatedly addition of row

(II) two row exchanges

(III) scalar lambda ≠ 0 multiply by in delegation

For confirming second matrix M 2, at first duplicate first matrix M 1, implement following job step then, wherein regard to the intermediate object program that each job step has provided second matrix M 2 down:

A) row with index ' a ' and row exchange with index ' b ':

B) row with index ' a ' and row exchange with index ' c ':

In this case, order RG (M2)=2 and therefore in second matrix M 2 RG irrelevant with row ZU.

In another step S26, the second parameters R max is according to formula

Rmax＝Rmin+m-1. (10)

Confirm.The 3rd necessary in formula (11) parameter m is confirmed according to following formula formula

m≤L-RG(M2). (11)

L representes the quantity of intermediate symbols, and is corresponding with the quantity K of source symbol Q in the present embodiment: L=4.Through being called intermediate symbols with the inner most block code symbol that produces code sign that multiplies each other.When sign indicating number linked, intermediate symbols and source symbol were inconsistent.The 3rd parameter m is any natural number in the result, this natural number be in 1 and the difference of the order RG of the quantity L of intermediate symbols and second matrix M 2 between.

In the present embodiment, according to formula 11 and 12, draw for Rmax:

Rmax≤Rmin+L-RG(M2)-1＝4+4-2-1＝5。

In step S27, the 4th parameters R is set as

R＝Rmin， (12)

Draw R=4 thus in the present embodiment.Next the mode with row forms the 3rd matrix M 3 in step 28, and said the 3rd matrix M 3 is from second matrix M 2 and from the encoder matrix of being derived by generator matrix G, produce.This carries out as follows, promptly at first all row of second matrix M 2 is copied in the 3rd matrix M 3, then, comprises R interior and comprise Rmax all row between interior and copy in the 3rd matrix at encoder matrix CM.At this, the row of encoder matrix CM duplicates and comprises: for the row that will duplicate is carried out and the identical row transposing of being carried out for formation second matrix M 2.Because R=Rmax=4 in the present embodiment, the index of reference ' 4 ' of encoder matrix is copied in the 3rd matrix M 3 with the row of ' 5 ' expression.

For the foregoing description, this shows, in the row of the index that indicates encoder matrix ' 4 ' and ' 5 ', carries out following conversion:

A) row with index ' a ' and row exchange with index ' b ';

B) row with index ' a ' and row exchange with index ' c ';

C) row with index ' a ' and row exchange with index ' d ';

So it is as follows to obtain the 3rd matrix:

In the step S29 of elementary row conversion and/or row transposing, the 3rd matrix M 3 is transformed to the 4th matrix M 4, wherein, generate incoherent capable ZH RH as a result of, wherein RH is corresponding to the order of the 4th matrix M 4.

For confirming the 4th matrix M 4, at first duplicate matrix M 3, carry out following operative steps then, the wherein following median that provides the 4th matrix M 2 for each job step:

A) exchange the 3rd row and the 4th row:

B) be added to fourth line to first row:

C) be added to fourth line to second row:

D) be added to fourth line to the third line:

In this case, order RH (M4)=3, and RH is capable irrelevant with ZH in the 4th matrix M 4.Whether check the 4th matrix M 4 has full rank in step S30.Owing to also do not have full rank according to (17) the 4th matrix M 4, so need proceed this algorithm.

Thereby, in step S30, be made as Rmax to the 4th parameters R:

R＝Rmax＝5 (18)

Count n according to the also definite Wucan of following formula:

n≤L-RH(M4). (19)

In addition according to the computes second parameters R max:

Rmax＝Rmax+n。

As said, L representes the quantity of intermediate symbols, and is corresponding with the quantity K of source symbol Q in the present embodiment: L=4.It is any natural number that Wucan is counted n (as the 3rd parameter m), be in 1 and the difference of the order RH of the quantity L of intermediate symbols and second matrix M 4 between.

Wucan is counted n so is n=1 in the present embodiment, makes to obtain Rmax=5+1=6 according to formula (19) for Rmax.

Next, repeating step S28 and S29 still, wherein replace second matrix M 2 with the 4th matrix M 4.

According to above explanation, in step S28, next the mode with row forms the 3rd matrix M 3,1, and said the 3rd matrix is from the 4th matrix M 4 and from the encoder matrix of being derived by generator matrix G, produce.Through the literary style after the 3rd matrix M 3 ', 1 ' representes first iteration (i iteration generally speaking) of step S28.Therefore, at first all row with the 4th matrix M 4 copy in the 3rd matrix M 3,1, wherein with the zero row that takies matrix M 4 are cleared up (Bereinigung) about its coefficient in advance.Then, encoder matrix CM comprised R interior and comprise Rmax all row between interior and copy in the 3rd matrix M 3,1.At this, duplicating that encoder matrix CM is capable comprises: the row that duplicate is carried out and the identical row exchange of being carried out for formation second matrix M 2.Because R=Rmax=6 in the present embodiment, only the row that is marked with index ' 6 ' of encoder matrix CM is copied in the 3rd matrix M 3,1.

For present embodiment, this shows, carries out following operation at the row of the index that indicates encoder matrix CM ' 6 ':

A) row with index ' a ' and row exchange with index ' b ';

B) row with index ' a ' and row exchange with index ' c ';

C) row with index ' a ' and row exchange with index ' d '.

So it is as follows to obtain the 3rd matrix M 3,1:

In the step S29 of elementary row conversion and/or row transposing, the 3rd matrix M 3,1 is transformed to the 4th matrix M 4,1, wherein, produce incoherent capable ZH RH as a result of, and RH is corresponding to the order of the 4th matrix M 4,1.

In order to confirm the 4th matrix M 4,1, at first duplicate matrix M 3,1, carry out following operative steps then, wherein regard to the intermediate object program that each job step provides the 4th matrix down:

A) be added to fourth line to second row;

B) be added to fourth line to the third line;

The order RH of the 4th matrix M 4,1 (M4,1) is RH (4,1)=4, thereby the 4th matrix M 4,1 has full rank.Thus, iterative cycles that can skips steps S30 and turn to step S31.

In step S31, the first parameters R min and the second parameters R max are transferred to transmitting apparatus SV through receiving equipment EV.Confirm the scope of the index of coding code sign clearly through Rmin and Rmax, by the scope of its error correction symbol FKS that can confirm to transmit.So error correction symbol FKS can be confirmed and is transferred to receiving equipment by transmitting apparatus SV.

In the present embodiment, Rmin=4, Rmax=6 makes and only asks three error correction symbol FKS.Generally, can inquire a plurality of error correction symbol.Owing to error correction symbol FKS, sentence error-free mode such as coding code sign CC at receiving equipment EV and exist, so can carry out the complete reconstruct of source symbol Q with index 4 and 6.For this reason, containing wrong code sign C is replaced by the error correction symbol FKS of error free reception.Reconstruct is not explained, because the reconstruct to the source symbol from code sign is known by block code here.

The part of iff error correction symbol is received with error-free mode; Can be applied to code sign that is received and the error correction symbol that is received so according to the method for the invention; Wherein, as a result of obtain to add these additional error correction symbol that need for complete reconstructed source symbol.

In a form of implementation, can count n to Wucan and be changed to 1 always, can carry out iterative step S28 and the S29 that accelerates in some cases thus.Yet thereby the error correction symbol that will transmit of minimum number also confirmed in the index that will confirm that can confirm minimum number in this case.

On the other hand, can confirm enough but the index range of non-minimum through the higher value that Wucan is counted n.

Except being applied to the perhaps nonsystematic block code of unique system, when using the grouping block of a plurality of series connection, also can use according to the method for the invention.In the following examples according to Fig. 3, at first symbol Q in source utilizes the first generator matrix G1 of systematic block codes BC1 to be encoded, and the symbol I that is encoded thus utilizes the nonsystematic generator matrix GN of nonsystematic block code BCN to be encoded.Output place at nonsystematic block code BCN provides coding code sign CC.

In Fig. 4, can find out the structure of encoder matrix CM, wherein it should be noted that this expansion scheme, use encoder matrix CM to be used for confirming error correction symbol FKS, and be not used for producing coding code sign CC according to the inventive method.Encoder matrix CM is through following generating step:

A) nonsystematic generator matrix GN is copied among the encoder matrix CM with the mode of row.Through after this step, encoder matrix CM is as follows:

CM＝GN (23)

B) the first generator matrix G1 is divided into first generation part (Generatorteil) GT1 and the second generative part GT2, and first generates part GT1 produces system symbol, and second generative part GT2 produces parity check symbol P, is the nonsystematic symbol.If the first generator matrix G1 for example presses mode and sets up

$G1=\left[\begin{array}{cccc}1& 0& 0& 0\\ 0& 1& 0& 0\\ 0& 0& 1& 0\\ 0& 0& 0& 1\\ 0& 1& 1& 1\\ 1& 0& 1& 1\\ 1& 1& 0& 1\end{array}\right]---\left(24\right)$

Then draw and first and second generate part GT1, GT2 is:

$\mathrm{GT}1=\left[\begin{array}{cccc}1& 0& 0& 0\\ 0& 1& 0& 0\\ 0& 0& 1& 0\\ 0& 0& 0& 1\end{array}\right]---\left(25\right)$ With

$\mathrm{GT}2=\left[\begin{array}{cccc}0& 1& 1& 1\\ 1& 0& 1& 1\\ 1& 1& 0& 1\end{array}\right],---\left(26\right)$

Wherein, first generate part corresponding to unit matrix.

C) coefficient of second generative part GT2 is copied among the encoder matrix CM.After treatment step, obtaining encoder matrix CM is:

$\mathrm{CM}=\left[\begin{array}{ccccc}& & & & \\ & & & \mathrm{GN}& \\ & & & & \\ 0& 1& 1& 1& \\ 1& 0& 1& 1& \\ 1& 1& 0& 1& \end{array}\right]---\left(27\right)$

D) be in the right side at second generative part GT2 and in encoder matrix CM, have unit matrix E1, wherein the order of unit matrix E1 is corresponding to the line number of second generative part GT2.Therefore, encoder matrix CM is as follows:

$\mathrm{CM}=\left[\begin{array}{ccccccc}& & & & & & \\ & & & \mathrm{GN}& & & \\ & & & & & & \\ 0& 1& 1& 1& 1& 0& 0\\ 1& 0& 1& 1& 0& 1& 0\\ 1& 1& 0& 1& 0& 0& 1\end{array}\right]=\left[\begin{array}{ccc}& \mathrm{GN}& \\ \mathrm{GT}2& & E1\end{array}\right]---\left(28\right)$

In this flexible program, intermediate symbols quantity L corresponding to the quantity of the symbol I that is encoded.

The step a) of this flexible program is to d) be called as step S32, and can in according to the step S22 of Fig. 2, additionally be implemented.After implementation step S32, produce a M1 through step S23.These row that comprise second generative part GT2 and cell matrix E1 that it should be noted that encoder matrix CM in this case additionally are copied in first matrix M 1.Other treatment step S24 according to Fig. 2 is performed the index that is used for definite error correction symbol FKS that will transmit then to S31.

In another flexible program, under the situation of using a plurality of systematic block codes BC1, BC2 and nonsystematic block code BCN thereafter, can use according to the method for the invention.In Fig. 5, from the symbol Q of source, produce the first intermediate symbols I1 by the second generator matrix G2 of the second systematic block codes BC2, from the first intermediate symbols I1, produce the second intermediate symbols I2 by the second generator matrix G2 of the second systematic block codes BC.Under the situation of the nonsystematic generator matrix GN that uses nonsystematic block code BCN, handle the second intermediate symbols I2, make to produce coding code sign CC.

For generating encoder matrix CM, in according to the step S33 of Fig. 2, carry out following treatment step:

A) for direct systematic block codes BC1 and nonsystematic block code BCN before nonsystematic block code BCN, according to the step S32 enforcement encoder matrix CM of Fig. 2.Encoder matrix CM in that this generated is:

$\mathrm{CM}=\left[\begin{array}{ccc}& \mathrm{GN}& \\ \mathrm{GT}2\left(\mathrm{GT}\right)& & E1\end{array}\right]---\left(29\right)$

In this case, second generative part GT2 (G1) corresponding to the parity check symbol of the generation second intermediate symbols I2 of the first generator matrix G1, be this scope of nonsystematic symbol.The line number of the second generative part GT2 (G1) of the order of the first unit matrix E1 and the first generator matrix G1 is consistent.

B) according to Fig. 6, the second systematic block codes BC2 before the first systematic block codes BC1 is added among the encoder matrix CM in step by step following:

-from the second generator matrix G2, extract to produce the first intermediate symbols I1 parity check symbol, be this second generative part GT2 (G1) of nonsystematic symbol;

-the said second generative part GT2 (G2) that extracts is copied to the end of encoder matrix CM with the mode of row;

-mend into the second unit matrix E2 on the right side of the second generative part GT2 (G2) that is duplicated of the second generator matrix G2, wherein the order of this second unit matrix E2 is corresponding with the line number of the second generative part GT2 (G2) that is duplicated of the second generator matrix G2;

-benefit into the second unit matrix M2 right side, take with zero based on the unappropriated coefficient of line length of nonsystematic generator matrix GN.This comes mark with Reference numeral N in Fig. 6.

Therefore, this expansion according to the method for the invention draws encoder matrix CM and is:

$\mathrm{CM}=\left[\begin{array}{ccc}& \mathrm{GN}& \\ \mathrm{GT}2\left(G1\right)& & E1\\ \mathrm{GT}2\left(G2\right)& E2& N\end{array}\right]---\left(30\right)$

Said additional step marks with treatment step S33 in Fig. 2 and before step S23, is implemented.

Next, produce the first matrix M I according to step S23.It should be noted that additionally that at this all row of encoder matrix CM are added among the first matrix M I, said row comprises the second generative part G2 (G1) of the first generator matrix G1 and the second generation part G2 (G1) of the second generator matrix G2.In this flexible program, the quantity sum of the quantity of the parity check symbol of the quantity of the quantity L of intermediate symbols through source symbol Q, the first systematic block codes BC1 output and the parity check symbol of second systematic block codes (BC2) output is explained.Carry out the index that is used for definite error correction symbol FKS that will transmit according to other treatment step S23 of Fig. 2 to S31 then.

Be applicable to the error correction symbol FKS that will ask that confirms minimum number system or under the situation of nonsystematic Raptor sign indicating number according to the method for the invention.At this, as shown in Figure 6, nonsystematic block code BC2 is corresponding to the internal code of Raptor sign indicating number, and the first and second block code BC1, BC2 are corresponding to the respective external sign indicating number of Raptor sign indicating number.Internal code is also known as LT sign indicating number (LT-Luby conversion).In addition, under the situation of nonsystematic Raptor sign indicating number, the correction matrix G0 of nonsystematic block code BC0 is mended before the second systematic block codes BC2, in order to the coded identification CC of generation system.Produce the Raptor sign indicating number of system thus, and coding code sign CC that can Request System is as error correction symbol.

Be able to explanation through binary character according to the method for the invention.Usually, can utilize the value in binary value or Galois territory, such as with Galois territory (2 ⁸) mode use according to the method for the invention.

Under the situation of using a plurality of matrixes, be able to explanation according to the method for the invention.Usually, the matrix that utilizes lesser amt with utilize step S22 to implement according to the method for the invention at least one other sequence of S31.

Can use equipment V to implement according to the method for the invention, wherein receiving element EE carries out and implementation step S21, and the first device ML1 carries out and implementation step S22 and carrying out alternatively and implementation step S32 and S33; The second device ML2 carries out and implementation step S23, and the 3rd device ML3 carries out and implementation step S24, and the 4th device ML4 carries out and implementation step S25; The 5th device ML5 carries out and implementation step S26; The 6th device ML6 carries out and implementation step S27, and the 7th device ML7 carries out and implementation step S28, and the 8th device ML8 carries out and implementation step S29; The 9th device ML9 carries out and implementation step S30, and the tenth device ML10 carries out and implementation step S31.In addition, this equipment can also have the 11 device ML11, utilizes said the 11 device can realize expansion according to the method for the invention.

Can in terminal equipment EG, be implemented by equipment V according to the method for the invention.In Fig. 7, this terminal equipment EG illustrates with the form of portable set.These portable sets for example according to the mobile wireless electric standard, especially according to GSM standard (GSM-global system for mobile communications), UMTS standard (UMTS-Universal Mobile Telecommunications System), DAB standard (DAB-digital audio broadcasting) or DVB standard (DVB-DVB) work.

Reference list:

Prof.C.Preston，″Ein?Skript?für?Lineare?Algebra?I?undII″，2003/04， http://dhcp24-159.mathematik.uni- bielefeld.de/～preston/files/teaching/linalg/skripten/ linalg.pdf

Claims (16)

1. be used to confirm to distribute to the method for the index (ZI) of error correction symbol (FKS); Wherein under the situation of the generator matrix that uses block code (BC) (G), be generation coding code sign (CC) the source symbol of K from quantity; And said coding code sign (CC) is transferred to receiving equipment (EV) from transmitting apparatus (SV); Transmission error (UEF) wherein in the code sign that is received (C), occurs, the index of the error correction symbol that wherein will transmit (FKS) is confirmed according to following step:

A) largest index (ZI) of definite all code signs (C) that received,

B) confirm the first parameters R min, the said first parameters R min has the value greater than the largest index (ZI) of all code signs (C) that received,

C) so form first matrix (M1) from the encoder matrix (CM) of being derived with the mode of going, make to by i the code sign (C) of error free reception by said generator matrix (G), with capable the copying in first matrix (M1) of i of said encoder matrix (CM),

D) use each row that comes mark first matrix (M1) with the row number corresponding column index (SI) of said encoder matrix (CM),

E) through elementary row conversion and/or row transposings said first matrix (M1) is transformed to second matrix (M2) with the row (ZU) that has nothing to do with RG, wherein RG is corresponding to the order of said second matrix (M2),

F) confirm the second parameters R max according to following formula

Rmax＝Rmin+m-1，

Wherein confirm the 3rd numeric parameter m according to following formula from the difference of the order RG of the quantity L of middle symbol and said second matrix (M2)

m≤L-RG(M2)，

G) with the mode that equals the said first parameters R min the 4th parameters R being set is R=Rmin,

H) so form the 3rd matrix (M3) from said second matrix (M2) with from the encoder matrix (CM) of being derived with the mode of going by said generator matrix (G); Make said encoder matrix (CM) comprised R interior and comprise Rmax all row between interior and copy in said the 3rd matrix (M3); Wherein said duplicating is included as the row that will duplicate and is embodied as and forms the row exchange that said second matrix (M2) is carried out

I) through elementary row conversion and/or row transposings said the 3rd matrix (M3) is transformed to the 4th matrix (M4) with the row (ZH) that has nothing to do with RH, wherein RH is corresponding to the order of said the 4th matrix (M4),

J) if said the 4th matrix (M4) does not have full rank (RH):

-said the 4th parameters R is set is R=Rmax,

-confirm that Wucan counts n, said Wucan count n from the difference of the order RH of the quantity L of middle symbol and said the 4th matrix (M4) according to following formula

n≤L-RH(M4)

Confirm,

-the said second parameters R max is set is Rmax=Rmax+n, and

-repeating step h) to i), wherein replace said second matrix (M2) and use said the 4th matrix (M4),

K) if said the 4th matrix (M4) has full rank (RH):

Through the said first and second parameter (Rmin; Rmax) scope of identification code code sign (CC) uniquely, wherein with the form of at least one error correction symbol (FKS) by receiving equipment (EV) from the said scope of transmitting apparatus (SV) request in order to error-free ground reconstructed source symbol (Q).

2. the method for claim 1 is characterized in that, said intermediate symbols is corresponding to said source symbol (Q), and the quantity of said intermediate symbols (L) is corresponding to the quantity (K) of said source symbol.

3. according to claim 1 or claim 2 method is characterized in that it is 1 that said Wucan number (n) is set.

4. the method for claim 1 is characterized in that, said the 3rd parameter (m) be 1 and the difference of the order (RG) of the quantity (L) of said intermediate symbols and said second matrix (M2) between arbitrary integer.

5. the method for claim 1 is characterized in that, said Wucan number (n) be 1 and the difference of the order (RH) of the quantity (L) of said intermediate symbols and said the 4th matrix (M4) between arbitrary integer.

6. the method for claim 1; It is characterized in that; Said source symbol (Q) at first utilizes first generator matrix (G1) of first systematic block codes (BC1) to be encoded; The symbol that is encoded thus (I) utilizes the nonsystematic generator matrix (GN) of nonsystematic block code (BCN) to be encoded, and wherein said encoder matrix (CM) produces through the following step:

A) mode of said nonsystematic generator matrix (GN) with row copied in the said encoder matrix (CM),

B) said first generator matrix (G1) is divided into first and generates part (GT1) and second generative part (GT2), said first generates part (GT1) produces said source symbol (Q), and said second generative part (GT2) produces parity check symbol (P),

C) coefficient with said second generative part (GT2) copies in the said encoder matrix (CM) with the mode of row in said nonsystematic generator matrix (GN),

D) in said encoder matrix (CM), be in the right side at said second generative part (GT2) and enclose first unit matrix (E1), wherein, the order of said unit matrix (E1) is corresponding to the line number of said second generative part (GT2),

Additionally, the row that comprises second generative part (GT2) and unit matrix (E1) of said encoder matrix (CM) mode with row is copied in said first matrix (M1), and the quantity of said intermediate symbols (L) is corresponding to the quantity of the symbol that is encoded (I).

7. method as claimed in claim 6; It is characterized in that; For having the first and second generator matrix (G1; A plurality of systematic block codes G2), that implemented before in the nonsystematic block code (BCN) with nonsystematic generator matrix (GN) (BC1, BC2), carry out following steps in order to generate said encoder matrix (CM):

A) for first generator matrix (G1) of first systematic block codes of directly being implemented before in said nonsystematic block code (BCN) (BC1) and for said nonsystematic generator matrix (GN), confirm said encoder matrix (CM) according to claim 6,

Second generator matrix (G2) of second systematic block codes (BC2) that b) will be implemented before at first systematic block codes (BC1) so is added in the said encoder matrix (CM), makes

-the output that from said second generator matrix (G2), is extracted in said second systematic block codes (BC2) produces this second generative part (GT2 (G2)) of parity check symbol;

-copy to the end of said encoder matrix (CM) to the said second generative part that is extracted (GT2 (G2)) with the mode of row,

-to mend into second unit matrix (E2) on the right side of the second generative part that is duplicated (GT2 (G2)), the order of wherein said second unit matrix (E2) is corresponding with the line number of the second generative part that is duplicated (GT2 (G2)),

-benefit into the right side of second unit matrix (E2) the long and unappropriated coefficient of row (N) based on nonsystematic generator matrix (GN) is taken with zero,

Additionally; The row of the second generative part (GT2 (G2)) of the second generative part that comprises first generator matrix (G1) (GT2 (G1)) of said encoder matrix (CM) and second generator matrix (G2) mode with row is copied in said first matrix (M1), and the quantity sum of the parity check symbol of the quantity of the parity check symbol of output place of the quantity of the quantity of said intermediate symbols (L) through said source symbol (Q), said first systematic block codes (BC1) and output place of said second systematic block codes (BC2) is specified.

8. method as claimed in claim 7 is characterized in that, uses nonsystematic Raptor sign indicating number, and (BC1 BC2) corresponds respectively to one of foreign key to wherein said first and second systematic block codes, and nonsystematic block code (BCN) is corresponding to the internal code of said Raptor sign indicating number.

9. method as claimed in claim 7 is characterized in that, using system Raptor sign indicating number, wherein,

(BC1 BC2) corresponds respectively to one of foreign key to-said first and second systematic block codes, and said nonsystematic block code (BCN) is corresponding to the internal code of said Raptor sign indicating number;

-for generating system coding symbol (CC), be added into said second systematic block codes (BC2) to the correction matrix of nonsystematic block code (BC0) (G0) before.

10. the method for claim 1 is characterized in that, uses the nonsystematic block code as block code (BC), generates said encoder matrix (CM) through duplicating said generator matrix (G).

11. the method for claim 1 is characterized in that, as block code (BC) using system block code, generates said encoder matrix (CM) through duplicating said generator matrix (G).

12. the method for claim 1 is characterized in that, the value of binary value or Galois territory (GF) is assigned to source symbol (Q), coding code sign (CC) and code sign (C).

13. carry out the equipment (V) of the method be used for confirming error correction symbol (FKS), wherein, under the situation of the generator matrix that uses block code (BC) (G), be to generate the code sign (CC) of encoding the source symbol of K from quantity; And said coding code sign (CC) is transferred to receiving equipment (EV) from transmitting apparatus (SV); Wherein, transmission error (UEF) in the code sign that is received (C), occurs, the index of the error correction symbol that wherein will transmit (FKS) is confirmed according to one of aforementioned claim; It is characterized in that

A) first device is used to confirm the largest index (ZI) of all code signs (C) of being received,

B) second device is used for confirming the first parameters R min that the said first parameters R min has the value greater than the largest index of all code signs (C) that received,

C) the 3rd device; Be used for forming first matrix (M1) from the encoder matrix (CM) of being derived by said generator matrix (G) with the mode of row; Wherein to by i the code sign (C) of error free reception, copy in said first matrix (M1) the i of said encoder matrix (CM) is capable

D) the 4th install, be used to come each row of said first matrix of mark (M1) with the row number corresponding column index (SI) of said encoder matrix (CM),

E) the 5th device is used for through elementary row conversion or row transposings said first matrix (M1) being transformed to second matrix (M2) with the row (ZU) that has nothing to do with RG, and wherein RG is corresponding to the order (RG) of said second matrix (M2),

F) the 6th device is used for confirming the second parameters R max according to following formula

Rmax＝Rmin+m-1，

Through said the 6th device, from the difference of the order RG of the quantity L of middle symbol and said second matrix (M2) according to following formula

m≤L-RG(M2)

Confirm the 3rd parameter m, and with the mode that equals the first parameters R min the 4th parameters R to be set be R=Rmin through said the 6th device,

G) the 7th device; Be used for forming the 3rd matrix (M3) with the mode of going from said second matrix (M2) with from the said encoder matrix (CM) of being derived by said generator matrix (G); Make said encoder matrix (CM) comprised R interior and comprise Rmax all row between interior and copy in said the 3rd matrix (M3); Wherein duplicate and comprise: form the row exchange that said second matrix (M2) is carried out for the row that will duplicate is implemented as

H) the 8th device is used for through elementary row conversion and/or row transposings said the 3rd matrix (M3) being transformed to the 4th matrix (M4) with the row (ZH) that has nothing to do with RH, and wherein RH is corresponding to the order of said the 4th matrix (M4),

I) the 9th device is used to check said the 4th matrix (M4) whether to have full rank (RH), and said the 9th device so is configured, if make that the 4th matrix (M4) does not have full rank (RH), then carries out following steps:

-said the 4th parameters R is set is R=Rmax,

-confirm that Wucan counts n, said Wucan is counted n and is confirmed according to following formula from the difference of the order RH of the quantity L of middle symbol and said the 4th matrix (M4)

n≤L-RH(M4)，

-the said second parameters R max is set is Rmax=Rmax+n, and

-impel repeating step g) to h), wherein substitute said second matrix (M2) and use said the 4th matrix (M4),

J) the tenth device has full rank (RH) then carries out following steps if it is configured to said the 4th matrix (M4):

Through the said first and second parameter (Rmin; Rmax) scope of identification code code sign (CC) uniquely; Wherein, For the said source of error-free ground reconstruct symbol (Q), with the said coding code sign of the form request of at least one error correction symbol (FKS) (CC), and transmit by said dispensing device (SV) through said receiving equipment (EV).

14. equipment as claimed in claim 13 (V) is characterized in that, the 11 device (ML11), and it so is configured, and makes can carry out like the described additional method step of one of claim 2 to 10.

15., it is characterized in that said equipment (V) is integrated in the terminal equipment (EG) like claim 13 or 14 described equipment (V).

16. equipment as claimed in claim 15 (V) is characterized in that, said terminal equipment (EG) is the portable set according to GSM standard, UMTS standard or DVB standard.

Images