DK163089B

DK163089B - PROCEDURE FOR TRANSMISSION AND RECEIVING DIGITAL INFORMATION SIGNALS

Info

Publication number: DK163089B
Application number: DK141984A
Authority: DK
Inventors: Werner Schreitmueller; Otto Klank
Original assignee: Ant Nachrichtentech; Telefunken Fernseh & Rundfunk
Priority date: 1983-03-07
Filing date: 1984-02-29
Publication date: 1992-01-13
Also published as: NO163589B; DE3308025A1; NO840838L; SG61590G; EP0118123A2; DE3477115D1; DK141984A; EP0118123B1; EP0118123A3; DK163089C; DK141984D0; NO163589C

Description

DK 163089 BDK 163089 B

Opfindelsen angår en fremgangsmåde til overføring og modtagelse af informationssignaler, navnlig til digital lydoverføring over satellitter, i henhold til indledningen til krav 1.The invention relates to a method for transmitting and receiving information signals, in particular for digital audio transmission over satellites, according to the preamble of claim 1.

5 Ved satellit-stereoradiofoni er det kendt, ad en overføringsvej, at overføre flere forskellige programmer, hvoraf et vilkårligt selektivt kan udvælges på modtagesiden. Overføringen sker f.eks. i tidsmultipleks. I dette tilfælde overføres der indenfor en såkaldt ramme, 10 dvs. en bestemt tidsperiode, tidsmæssigt efter hinanden følgende bitord for den venstre kanal i det første program, den højre kanal i det første program, den venstre kanal i det andet program, den højre kanal i det andet program, den venstre kanal i det tredje program osv. Ef-15 ter afviklingen af en sådan ramme begynder den sekventielle overføring påny med den venstre kanal i det første program osv.5 With satellite stereo radiophony, it is known, through a transmission path, to transmit several different programs, any of which can be selectively selected on the receiving side. The transfer occurs e.g. in time multiplex. In this case, within a so-called frame, i.e. a specific period of time, successive bit words for the left channel of the first program, the right channel of the first program, the left channel of the second program, the right channel of the second program, the left channel of the third program, etc. After the execution of such a frame, the sequential transfer resumes with the left channel of the first program, etc.

Ved den digitale informationsoverføring over satellitter optræder kun statistisk fordelte enkeltbitfejl 20 (Gauss'sk sus), således som angivet i "British Telecommunications; OTS and Digital 'Television Transmission Investigations of Error Statistics Relevant to the Design of Error Correctors for Television (20.11.80)".In the digital information transmission over satellites, only statistically distributed single bit errors 20 (Gauss's case) occur, as stated in "British Telecommunications; OTS and Digital" Television Transmission Investigations of Error Statistics Relevant to the Design of Error Correctors for Television (20.11.80 ) ".

Til korrektion af statistisk fordelte fejl, er cykliske 25 koder særligt egnede.For the correction of statistically distributed errors, cyclic 25 codes are particularly suitable.

Det er blevet foreslået (DE patentansøgning nr.It has been proposed (DE patent application no.

P 31 34 831.9), at anvende en BCH-kode 63/44 til fejlkonstatering og -korrektion. Med denne kode, kan man ved hjælp af en passende dekoderkobling f.eks. konstatere 30 fem fejl, hvoraf de to kan korrigeres ved hjælp af koden. De tre yderligere konstaterede fejl interpoleres med en passende kobling.P 31 34 831.9), to use a BCH code 63/44 for error detection and correction. With this code, by means of a suitable decoder coupling, e.g. find 30 five errors, both of which can be corrected using the code. The three additional errors found are interpolated with an appropriate coupling.

Ved faseforskelsmodulation opstår der ved demodulatorens udgang i overvejende grad dobbeltbitfejl, der 35 i utilladelig grad begrænser fejlkonstateringen eller fejlkorrektionen med blokkoder. Allerede tre overfø-In phase difference modulation, at the output of the demodulator, predominantly double-bit errors occur, which inadvertently limit the error detection or error correction with block codes. Already three transfers

DK 163089 BDK 163089 B

2 ringsfejl konstateres under visse omstændigheder ikke mere af denne kode, og to overføringsfejl bevirker en fejlkonstatering i stedet for en fuldstændig korrektion.In some circumstances, two ring errors are no longer detected by this code, and two transfer errors cause an error finding instead of a complete correction.

Som følge af udnyttelsen af forskellen fører en på over-5 føringsstrækningen forfalsket bit, nemlig i demodulatoren til to falske udsagn: dels henføres denne falske bit til en forudgående rigtig bit, dels henføres den efterfølgende rigtige bit til den forudgående falske bit.Due to the exploitation of the difference, a counterfeit bit on the transmission line, namely in the demodulator, leads to two false statements: first, this false bit is assigned to a preceding real bit, and secondly the subsequent correct bit is assigned to the preceding false bit.

Med opfindelsen tilsigtes der tilvejebragt en 1 0 forbedring af fejIkonstateringen og fejlkorrektionen ved en fremgangsmåde i henhold til indledningen til krav l, navnlig ved anvendelse af blokkoden 63/44 og faseforskelsmodulation .The invention aims to provide an improvement in error detection and error correction by a method according to the preamble of claim 1, in particular by using block code 63/44 and phase difference modulation.

Dette opnås ifølge opfindelsen ved en fremganga-15 måde i overensstemmelse med den kendetegnende del af krav 1.This is achieved according to the invention by a method according to the characterizing part of claim 1.

Hensigtsmæssige videreudviklinger af fremgangsmåden følge opfindelsen er angivet i underkravene.Suitable further developments of the method according to the invention are set out in the subclaims.

Fremgangsmåden ifølge opfindelsen udmærker sig 20 ved, at informationsbittene i en første information ved spredning er adskilt af bittene i andre informationer.The method according to the invention is distinguished by the fact that the information bits in a first information on spreading are separated by the bits in other information.

Dette er opnået ved, at datarækkerne i to eller flere BCH-blokke ved enkeltbit-multipleks fremgangsmåden er flettet ind i hinanden. Hvis der f.eks. i stedet for 25 enkeltbitfejl kun optræder dobbeltfejl, kan der, medens der hidtil kun kunne erkendes to dobbeltfejl og én dobbeltfejl kunne korrigeres, med en spredning af to BCH-blokke, erkendes og interpoleres fem dobbeltfejl eller korrigeres to dobbeltfejl. Ved dette eksempel er konsta-30 tering af en blokfejl på indtil ti bit mulig.This is achieved by interleaving the data rows in two or more BCH blocks by the single-bit multiplex method. For example, if instead of 25 single-bit errors occurring only double errors, while so far only two double errors and one double error could be corrected, with a spread of two BCH blocks, five double errors can be recognized and interpolated or two double errors corrected. In this example, a block error of up to ten bits is possible.

En indfletning af hele delrammen (fire BCH-blok-ke) forøger sikkerheden mod blokfejl. En ulempe er dog en forøget indtræfningssandsynlighed for de optrædende fejl. Indfletningen af kun to BCH-blokke udgør derfor et 3 5 gunstigt kompromis.Interleaving the entire subframe (four BCH blocks) increases security against block failure. A disadvantage, however, is an increased probability of occurrence of the occurring errors. The interleaving of only two BCH blocks therefore constitutes a favorable compromise.

Ved de oprindelige rammer uden indfletning kræves der en forarbejdningshastighed for demodulatoren på en 3In the original frame without interlacing, a processing speed of the demodulator of a 3

DK 163089 BDK 163089 B

datahastighed på 10,24 MHz. Ved indfletningen af to BCH-blokke falder datahastigheden til 5,12 MHz. Ved den halverede forarbejdningshastighed, kan der hensigtsmæssigt opbygges en demodulator i MOS-teknik.data rate of 10.24 MHz. When interleaving two BCH blocks, the data rate drops to 5.12 MHz. At the halved processing speed, a demo modulator in the MOS technique can conveniently be constructed.

5 Det foreslås, at overføre digitalsignalet fra senderen til modtageren i form af 4-fase-vinkelændringer eller -forskelle af bæreren. Dette sker på den måde, at der i senderen udføres en firetrins-faseforskelsmodulation og i modtageren en tilsvarende forskelsdemodula-10 tion, således som angivet i "Nachrichtentechnische Zei-tung, 29 (1976), hæfte 5, siderne 390-394". Ved en anden demodulationsfremgangsmåde, som den foreliggende opfindelse ligeledes kan anvendes på, benyttes der kohærent demodulation med differentiel afkodning. De ved denne 15 fremgangsmåde opnåede gunstigere systemforhold fås på bekostning af et større opbud.It is proposed to transmit the digital signal from the transmitter to the receiver in the form of 4-phase angular changes or differences of the carrier. This is done in that in the transmitter a four-stage phase difference modulation is carried out and in the receiver a corresponding difference demodulation, as stated in "Nachrichtentechnische Zei-tung, 29 (1976), booklet 5, pages 390-394". In another demodulation method to which the present invention may also be applied, coherent demodulation with differential decoding is used. The more favorable system conditions obtained by this method are obtained at the expense of a larger supply.

I modsætning til en fuldstændig kohærent modulation, hvor firtydigheden af 4-faseomkoblingen ved demodulationen skal elimineres ved passende koblingsforan-20 staltninger, således som dette er foreslået i DE-patent-ansøgning nr. P 31 25 894.8, er denne firtydighed ikke mere tilstede ved faseforskelsmodulation eller -afkodning. Informationen ligger i forskellen mellem to successive bit. Paseforskelsmodulationen er især egnet ved 25 digital lydoverføring i slukkeperioderne i et fjernsynssignal. Da bæreren ikke skal afledes i demodulatoren, skal der ved fjernsynsmultipleksfremgangsmåden med overføring af et digitalt fjernsynslydsignal i slukkeperioderne ved den differentielle modulation af signalet kun 30 afledes bittakten. En fortløbende kontrol med, om den modtagne ramme er rigtig synkroniseret, bortfalder.In contrast to a completely coherent modulation, where the clarity of the 4-phase switching on demodulation is to be eliminated by appropriate coupling means, as proposed in DE Patent Application No. P 31 25 894.8, this clarity is no longer present at phase difference modulation or decoding. The information lies in the difference between two successive bits. Passage difference modulation is particularly suitable for digital audio transmission during the off periods of a television signal. Since the carrier is not to be diverted into the demodulator, in the television multiplexing method of transmitting a digital television audio signal during the off periods by the differential modulation of the signal, only the bit rate must be derived. Continuous verification of whether the received frame is properly synchronized will lapse.

Opfindelsen forklares nærmere i det følgende under henvisning til den skematiske tegning, hvor fig. 1 viser opbygningen af en indfletning af to 35 BCH-blokke, fig. 2 opbygningen af en halvramme til 4-faseom-skiftning, og 4The invention is explained in more detail below with reference to the schematic drawing, in which fig. 1 shows the structure of an interlacing of two 35 BCH blocks; FIG. 2 shows the construction of a half-frame for 4-phase switching; and 4

DK 163089 BDK 163089 B

fig. 3 en kobling til BCH-blokudvælgelsen.FIG. 3 shows a link to the BCH block selection.

Fig. 1 viser opbygningen af en indfletning af to BCH-blokke, Ved 8 stereokanaler i én halvramme kendetegner et bogstav L en venstre kanal, og et bogstav R 5 en højre kanal af et stereoprogram. Li er f.eks. den venstre kanal i et første stereoprogram og R3 den højre kanal i et tredje stereoprogram. En aftastningsværdi består af 14 bit, der er opdelt i 11 højværdibit, der kort betegnes MSB, og tre lavværdibit, der kort betegnes 10 LSB. To stereoprogrammer er sammenfattet til en BCH-blok. En første BCH-blok består af stereoprogrammerne 1 og 2, en anden BCH-blok af stereoprogrammerne 3 og 4. De første 22 bit er MSB fra LI og L3. En første bit er bit nummer 1 i LI, en anden bit er bit 15 nummer 1 i L2 og derefter følger som tredj e bit, bit nummer 2 i LI, som fjerde bit, bit nummer 2 i L2 osv. Derpå følger de 22 MSB i R2 og R4. Til hver af de to BCH-blokke hører 19 prøvebit Fl-2 og F3-4. De slutter sig til de 88 MSB i LI til R4.FIG. 1 shows the structure of an interlacing of two BCH blocks, At 8 stereo channels in one half frame, a letter L characterizes a left channel and a letter R 5 a right channel of a stereo program. Li is e.g. the left channel in a first stereo program and R3 the right channel in a third stereo program. A scan value consists of 14 bits divided into 11 high-value bits, briefly referred to as MSB, and three low-value bits, briefly referred to as 10 LSBs. Two stereo programs are combined into one BCH block. A first BCH block consists of stereo programs 1 and 2, a second BCH block of stereo programs 3 and 4. The first 22 bits are MSB from LI and L3. A first bit is bit number 1 in LI, a second bit is bit 15 number 1 in L2 and then follows as third e bit, bit number 2 in LI, as fourth bit, bit number 2 in L2, etc. Then they follow 22 MSB in R2 and R4. Each of the two BCH blocks includes 19 sample bits Fl-2 and F3-4. They join the 88 MSB in LI to R4.

20 Som foreslået i DE patentskrift nr. 31 26 880 findes der i halvrammen informationsbit i form af kendetegnende bit for programtypen og programnummeret. Til prøvebittene F1-2/F3-4 for de to BCH-blokke slutter der sig et kendetegnende bit for en BCH-blok i denne 25 ramme ifølge opfindelsen. Efter disse kendetegnende bit Kl-2, K3-4 følger ialt 24 LSB for de to BCH-blokke, nemlig LSB for L1/L3, derpå R1/R3, samt endvidere L2/L4 og R2/R4.20 As proposed in DE patent specification 31 26 880, the semi-frame contains information bits in the form of characteristic bits for the program type and program number. To the sample bits F1-2 / F3-4 of the two BCH blocks, a characteristic bit of a BCH block in this frame according to the invention is joined. Following these characteristic bits Kl-2, K3-4, a total of 24 LSBs follow for the two BCH blocks, namely LSB for L1 / L3, then R1 / R3, as well as L2 / L4 and R2 / R4.

Denne opbygning gentages derpå til afslutningen 30 af halvrammen for stereoprogrammerne 5-8 så ialt 2 x 152 bit er indflettet i hinanden. Ved begyndelsen af en halvramme er der reserveret 16 bit til et synkronord og/eller særtjeneste. Ialt har denne halvramme en længde på 320 bit. Opbygningen af den anden halvramme til ste-35 reoprogrammerne 9 til 16 er identisk.This structure is then repeated until the end 30 of the half frame of the stereo programs 5-8 so that a total of 2 x 152 bits are intertwined. At the beginning of a half-frame, 16 bits are reserved for a synchronous word and / or special service. In total, this half frame has a length of 320 bits. The structure of the second half frame for control programs 9 to 16 is identical.

Fig. 2 viser en fuldstændig halvramme for 4-fase-omskiftningen. Halvrammen begynder med 16 ikke i hinan-FIG. 2 shows a complete half-frame for the 4-phase switch. The half frame starts with 16 not in each other.

Claims

A method for transmitting and receiving digital information signals, in particular for digital audio transmission over satellites in which data rows in a frame are arranged in succession, characterized in that an error correction system with block formation over two stereo channels or four mono channels is used, and that the data rows in each subframe of a 4-phase switched signal (4-PSK) are so intertwined over at least two blocks by a single-bit multiplex method that two consecutive bits for each sensing value for each subframe at least separated by a bit associated with another scan value.

Method according to claim 1, characterized in that the high-value bits (MSBs) for each of four DK 163089 B scan values are summarized in a cyclic block code.

Method according to claim 1 or 2, characterized in that a bit row consists of a cyclic block code, the corresponding low value bits and a bit for additional information.

Method according to claim 3, characterized in that two bit rows are interleaved in a single bit multiplex method.

Method according to claim 1, characterized in that the frame consists of two subframes with 320 bits each.

Method according to claim 1 or 5, characterized in that a subframe contains 16 sensing values associated with eight stereo channels or 16 mono channels or a combination of stereo and mono channels.

Method according to claim 6, characterized in that a scanning value consists of 14 bits, 20 divided into 11 high-value bits (MSB) and three low-value bits (LSB).

Method according to claim 1 or 2, characterized in that the first 16 bits of a subframe are reserved for special services and / or for a synchronous word.

Method according to claim 5, characterized in that a 63.44 BCH code is used as a cyclic block code.

Method according to claim 1, characterized in that the digital information signals are transmitted from the transmitter to the receiver in the form of four phase angle changes or differences of the carrier.

Method according to claim 10, characterized in that in the receiver a phase difference demodulation or a coherent demodulation is performed with subsequent differential decoding of the received signal.