CA1160773A - Method of clock rate matching plesiochronuous signals - Google Patents
Method of clock rate matching plesiochronuous signalsInfo
- Publication number
- CA1160773A CA1160773A CA000365170A CA365170A CA1160773A CA 1160773 A CA1160773 A CA 1160773A CA 000365170 A CA000365170 A CA 000365170A CA 365170 A CA365170 A CA 365170A CA 1160773 A CA1160773 A CA 1160773A
- Authority
- CA
- Canada
- Prior art keywords
- clock rate
- rate matching
- audio signal
- negative
- positive
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04J—MULTIPLEX COMMUNICATION
- H04J3/00—Time-division multiplex systems
- H04J3/02—Details
- H04J3/06—Synchronising arrangements
- H04J3/07—Synchronising arrangements using pulse stuffing for systems with different or fluctuating information rates or bit rates
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Synchronisation In Digital Transmission Systems (AREA)
- Time-Division Multiplex Systems (AREA)
- Electrophonic Musical Instruments (AREA)
- Ultra Sonic Daignosis Equipment (AREA)
Abstract
ABSTRACT
A method of clock rate matching a digitalised analogue signal (Tn0 to Tn31) to a data flow having a pulse frame structure which includes channels (ZK0 to ZK31) for the data and additional transmission capacity for signals (UR-Sy, Z1 to Z31) other than data, the digitalised analogue signal (Tn0 to Tn31) and data flow having mutually plesiochronous clock rates, said method including the steps of inserting digital bits forming words of the digitalised analogue signal (Tn0 to Tn31) into respective bit time slots of a plurality of consecutive channels (ZK1 to ZK3, ZK17 to ZK19) of each pulse frame, inserting into vacant bit time slots of the additional transmission capacity (ZK16) a clock rate matching signal (++,--) indicating whether the clock rates are effectively synchronous within predetermined limits or have a positive (++/++) or negative (--,--) deviation relative to each other outside said limits, adjusting, when said deviation is outside said limits, the distribution of words (Tn0 to Tn31) of the digitalised analogue signal in said plurality of consecutive channels (ZK1 to ZK3, ZR17 to ZK19) such that said clock rate matching signal (++,--) will again indicate effectively synchronous clock rates, said synchronism being indicated by the occurrence of alternate positive (++) and negative (--) deviation indicating clock rate matching signals, said positive deviation being indicated by consecutive occurrences of positive deviation indicating clock rate matching signals (++,++), and said negative deviation being indicated by consecutive occurrences of negative deviation indicating clock rate matching signals (--,--).
A method of clock rate matching a digitalised analogue signal (Tn0 to Tn31) to a data flow having a pulse frame structure which includes channels (ZK0 to ZK31) for the data and additional transmission capacity for signals (UR-Sy, Z1 to Z31) other than data, the digitalised analogue signal (Tn0 to Tn31) and data flow having mutually plesiochronous clock rates, said method including the steps of inserting digital bits forming words of the digitalised analogue signal (Tn0 to Tn31) into respective bit time slots of a plurality of consecutive channels (ZK1 to ZK3, ZK17 to ZK19) of each pulse frame, inserting into vacant bit time slots of the additional transmission capacity (ZK16) a clock rate matching signal (++,--) indicating whether the clock rates are effectively synchronous within predetermined limits or have a positive (++/++) or negative (--,--) deviation relative to each other outside said limits, adjusting, when said deviation is outside said limits, the distribution of words (Tn0 to Tn31) of the digitalised analogue signal in said plurality of consecutive channels (ZK1 to ZK3, ZR17 to ZK19) such that said clock rate matching signal (++,--) will again indicate effectively synchronous clock rates, said synchronism being indicated by the occurrence of alternate positive (++) and negative (--) deviation indicating clock rate matching signals, said positive deviation being indicated by consecutive occurrences of positive deviation indicating clock rate matching signals (++,++), and said negative deviation being indicated by consecutive occurrences of negative deviation indicating clock rate matching signals (--,--).
Description
The invention relates to a method of clock rate matching plesiochro-nous signals, particularly clock rate matching in respect of a digitalised ~converted into digital form) analogue signal to a data flow having a pulse frame structure which includes channels for the data and additional transmission capacity for signals other than data, the digitalised analogue signal and data flow having mutually plesiochronous clock rates. The data flow can be orga-nised in frames and super frames for telephone quality digital speech trans-mission with signalling bits for signalling transmission whose clock rate is plesiochronous to the clock rate of the digital audio signal, thus the two clock frequencies possessa slight deviation of e.g. 10 4 to 10 7. The digi-talised analogue signal may be an audio signal as referred to hereinafter.
The term "audio signal" is to be understood to include sound programs, such as transmitted from a concert hall to a broadcast transmitter, with a band-width of, for example, 7kHz or 15 kHz.
If the conversion of an analogue sound broadcast signal into digital form is controlled for example using a clock rate of a digital signal connec-tion for 2Mbit/s-signals (Siemens system DSV2, previously PCM30), the inser-tion of the audio signal data flow into the pulse frame of the DSV2 system and the co-exploitation of the possibly following further digital hierachy stages is non-problematical. However applications are conceivable in which the analogue/digital conversion (A/D conversion) of the audio signal takes place at a distance from the devices of the DSV2 system and for various reasons it is not possible to use the same clock rate. The A/D conversion must then be carried out using a locally produced clock rate which is plesiochronous to the clock rate of the DSV2 system.
The term "audio signal" is to be understood to include sound programs, such as transmitted from a concert hall to a broadcast transmitter, with a band-width of, for example, 7kHz or 15 kHz.
If the conversion of an analogue sound broadcast signal into digital form is controlled for example using a clock rate of a digital signal connec-tion for 2Mbit/s-signals (Siemens system DSV2, previously PCM30), the inser-tion of the audio signal data flow into the pulse frame of the DSV2 system and the co-exploitation of the possibly following further digital hierachy stages is non-problematical. However applications are conceivable in which the analogue/digital conversion (A/D conversion) of the audio signal takes place at a distance from the devices of the DSV2 system and for various reasons it is not possible to use the same clock rate. The A/D conversion must then be carried out using a locally produced clock rate which is plesiochronous to the clock rate of the DSV2 system.
- 2 -~ 160773 In these situations clock rate matching is necessary to achieve insertion of the digital audio signal into the pulse frames of the DSV2 system.
A plurality of clock rate matching methods are known for combining the data flows of a plurality of digital "sub-systems" having clock rates which are plesiochronous to one another to form a "super system". These methods may be split into those which do involve and those which do not involve information loss. If it were desired to use clock rate matching involving information loss in the audio signal insertion mentioned above, in order not to lose word synchronism it would be necessar~ to tolerate the omission or the repetition of an entire codé word from time to time. Investigations carried out by the present applicants have indicated that this measure would result in a fall in quality in the audio signal which would be unacceptable to the listener when the clock rate deviation exceeds 10 Clock rate matching methods which do not involve information loss in which the clock rate matching takes place bit-wise are known from the "Nachrichtentechnischen Fachher-ichten No. 42, 1972", PCM-Technik, VDE-Verlag GmbH Berlin-Charlottenburg, pages 235 to 244 and from the magazine "Elektronik", Vol. 6/1978, pages 78 to 83. In these methods, at an agreed point within the frame - generally of a "super system" - blank bits are interposed (positive clock rate matching) or information bits are gated out (negative clock rate matching). This procedure must be communicated by means of additional capacity available l 160773 in the pulse frame to the receiver which cancels the manipulation which has been effected at the transmitting end and reproduces the or-i-ginal clock rate. In addition to the procedures which use positive clock rate matching or negative clock rate matching and in which in the case of positive matching the clock frequency of the "sub system" is less than the clock frequency of the available data flow whereas in negative matching the clock frequency of the "sub system" is higher than that of the available data flow, the two possibilities can be combined as positive/
negative clock rate matching. In this case it is possible to compensate relative clock rate deviations in both positive and ne~ative directions.
The multiplex system which is disclosed in the "Taschenbuch der Fernmeldepraxis 1979" pages 13 to 41 is suitable both for digital transmission of audio signals and for'the combined transmission of audio signals and telephone signals in the pulse frame of the DSV2 system.
Fig. 1 ill~strates the ~nown structure of the pulse frame and of the super frame of a DSV2 system. This pulse frame of 125 ps duration contains 256 bits; this corresponds to a data flow of 2048kbit/s. The pulse frame is divided into 32 channels ZKO to ZK31. The first time channel ZKO serves alternately to transmit a message word and a frame code word (RSy). The seventeenth time channel ZK16 is provided mainly to transmit dialled characteristics Zl to Z15, Z17 to Z31 of the ihirty tele-phone channels Kl to K15, K17 to ~31. Sixteen consecutively transmitted ~ulse frames R0 to B15 fQnm one super fr~me. The synchronisation of the super frame is effected by means of a super frame synchronising word ~R-Sy in the seventeenth time channel ZK16 of the first pulse frame RO. For the sake of clarity the sixteen pulse frames of a super frame in Fig. 1 have been represented one below another.
The proposal disclosed in the "Taschenbuch der Fern-meldepraxis" permits mixed use of the DSV2 system for telephone channels and audio channels. In this case the first time channels ZKO and the seventeenth time channels ZK16 are not affected by the audio transmission.
In this digital audio transmission, channels of high quality possessing 15 kHz band width and channels of average quality possessing 7 kHz band width must be provided. In ~he former case the sampling frequency is 32 kHz and in the latter case 16 kHz. Eoth represent multiples of the sa~pling frequency of 8 kHz for teleFhone sign21s. Each sa~ple value with the protection bits in respect of the bit errors occurring on the transmission pa~h requires a word of 12 bits length. Therefore in one pulse frame four code words each comprising 12 bits must be accommodated for a 15 kHz audio channel and two code words each comprising 12 bits must be accommodated for a 7 kHz audio channel. Therefore six time channels ZKl to ZK3, ZK17 to ZKl9 each comprising eight bits are used for the transmission of a 15 kHz audio channel and three time channels ZKl to ZK3 are used for the transmission of a 7 kHz-audio signal. If further audio signals are to be transmitted, further time channels ZK4 to ZK6, ZK20 to ZK22 and ZK4 to ZK6 etc. are used. The bits of two consecutive code words are also interlocked in order to provide protection against double l 160773 errors on the transmission path. As a result a 15 kHz audio channel has an information flow of 384 kbit/s and an interface having a data rate of 384 kbit/s (in the case of a 7 kHz audio channel: 192 kblt/s) is provided between the coder/decoder device of the audio signal and the DSV2 system.
This known method of operation has the disadvantage that it merely provides a synchronous insertion of the information of coded audio channels into the frame of a DSV2 system. However this is not always possible such as for example:
- at a digital interface between a broadcasting studio and the digital network of the postal administration (or other authority) when the clock rates possess a permissible deviation of greater -than 10 7 from one another, - in national digital networks at the interfaces between parts of these networks which are not or not yet operating in synchronism and where the clock rate deviation is greater than 10 and - when digitaL audio signals are exchanged between international networks which although in themselves synchronous, are mutually plesiochronous with a clock rate deviation of greater than 10 7.
According to this invention there is provided a method for clock rate matching a digitalized audio signal to a data flow organized in pulse frames and super frames for telephone quality digital speech transmission and with characteristic bits for signalling transmission whose clock rate is plesiochronous to the clock rate of the digital audio signal, characterized in that in each super frame the clock rate matching occurs by word, i.e. an entire code word of the digital audio signal, and that in each super frame a positive and a negative clock rate matching signal is transmitted to the associated positions during transmission of the digital audio signal over several telephone channels of the characteristic bit, that with agreement of the clock rates of the data flow and of the digital audio signal, the clock rate matching signal is alternately positive and negative from super frame to super frame, that for a positive clock rate matching, the clock rate matching signals of two consecutive super frames are both positive and withln the second super frame -which contains the second of the positive clock rate matching signals directly following one another -a blank word is inserted at a predetermined point between two code words of the digital audio signal, and that for a negative clock rate matching, the clock rate matching signals of two consecut-ive super frames are both negative and freed positions of the characteristic bit in the super frame which follows the second negative clock rate matching signal are used with a code word gated out of the coded audio signal or with parts of code words gated out of the coded audio signal.
In an embodiment of the invention there is provided a method for clock rate matching a digital audio signal to a data flow organized in pulse frames and super frames for telephone quality digital speech transmission involving transmission of channel-associated signals as signalling bits in each pulse frame, the digital audio signal and data flow having plesiochronous clock rates, said method including the steps of inserting digital bits forming words of the audio signal into respec~ive bit time slots of a plurality of consecutive channels l 160773 of each pulse frame of a super frame, inserting into the vacant signalling bit time slots associated with the channels into which the audio signal has been inserted, a clock rate matching signal indicating whether the clock rates are synchronous or have a positive or a negative deviation relative to each other, adjusting the distribution of words of said audio signal in said plurality of consecutive channels when said deviation corresponds to one word length, such that said clock rate matching signal can indicate synchronous clock rates, said synchronism being indicated by the occurrence in alternate super frames of respective positive and negative deviation indicating clock rate matching signals, said positive indication being indicated by the occurrence in two consecutive super frames of positive deviation indicating clock rate matching signals, and said negative deviation being indicated by the occurrence in two consecutive super frames of negative deviation indicating clock rate matching signals.
Embodiments of this invention will now be described, by way of example, with reference to the accompanying drawings in which:-Figure 1 is a table illustrating a known form of super frame of a known data transmission system (Siemens system DSV2);
Figure 2 is a set of tables illustrating the structure of the data flow of a DSV2 system in the case of plesiochronous 7 kHz audio signal transmission in the synchronous state;
Figure 3 is a set of tables illustrating the structure of the data flow of the DSV2 system in the case of a positive clock rate matching of the digital 7 kHz audio signal; and 1 ~60773 Fig. 4 is a set of tables illustrating the structure of the data flow of the DSV2 system in the case of a negative clock rate matching of the digital 7 kHz audio signal.
Referring to Fig. 2, in each super frame of a trans-mission system embodying the invention, time channels ZKl to ZK3 are filled with 32 àudio signal words or code words Tnl to Tn32 each comprising 12 bits of a 7 kHz audio signal. If these three time channels ZKl to ZK3 are filled with the coded audio signal, the 3 times 4 bit time slots which become free in time channels ZR16 (originally conveying the dialled characteristics Zl to Z3 each comprising 4 bits) in pulse frames Rl, R2 and R3 can be used for clock rate matching. In each super frame the vacant 8 bit time slots in the pulse frames Rl and R2 are filled in alternate super frames with a positive and a negative clock rate matching signal ~+ and --. Thus each of the clock rate matching signals ++ and -- is divided between the pulse frames Rl and R2, whereas the remaining 4 bit time slots in the pulse frame R3 is filled with a signal word S which can be used for example to transmit alarms or for signalling purposes.
The synchronous state "no clock rate matching necessary"
need not be separately transmitted since the regular sequence ++, --, ++, -- etc. of alternately positive and negative clock rate matching signals does not trigger any reaction in the receiver.
Fig. 3 illustrates the case of a positive clock rate deviation, i.e. the clock frequency- of the audio signal is less than the corresponding clock frequency of the DSV2 system.
l 160773 If the time shift bPtween the two clock rates amounts to 12 clock periods, positive clock rate matching signals are trans-mitted in two consecutive super frames, and in the second super frame an information-free blank word LW is inserted at a position previously agreed upon with the receiver, here as the first code word following the frame code word RS~ in the pulse frame R3 of the second super frame. Thus only one code word of the 7 kHz audio signal is transmitted in the pulse frame R3 and within this super frame the number of transmitted code words is reduced by one to 31. This restores synchronism and in the following super frames positive and negative clock rate matching signals are transmitted alternately commencing with the negative clock rate matching signal.
In the case of negative clock rate deviation, illustrated in Fig. 4, the clock frequency of the audio signal is greater than the corresponding clock frequency of the DSV2 system. In order to signal this state, in two consecutive super frames negative clock rate matching signals are transmitted. Since a code word gated out of the audio signal must be additionally transmitted within a super frame, it is agreed that the super frame which follows the super frame with the second negative clock rate matching signal (the third super frame) should contain 33 code words.
In the third super frame, the bit time slots within the seventeenth time channel ZK16, which in the case of audio signal transmission are normally used for the clock rate matching signal and a signal word S, are filled in the pulse frame Rl with the 1 1~0773 first 4 bits of the code word ~n5, in the pulse frame R2 with the fifth to eighth bits of the code word Tn7, and in the pulse frame R3 with the last 4 bits of the code word Tn9. A resultant shift in the code word boundaries of the code words Tn5 to Tn9 S within the pulse frames R2 and R3 is cancelled out in the fifth pulse frame (not shown) and a tenth code word commences in the first bit time slot of the time channel ZKl in this pulse frame.
Here again, after the clock rate matching, positive and negative clock rate matching signals are transmitted alternately (commencing with the-positive signal) in the sub-sequent super frames.
It is also possible, in the case of positive clock rate matching, to insert 3 sets of 4 blank bits between code words within a super frame (e.g. 4 blank bits in three consecutive pulse frames) in order to achieve a less jerky clock rate matching than if a single 12-bit blank word were inserted.
It is equally possible in the case of negative clock rate matching to transmit a code word which is split into 3 sets of 4 bits and is gated out at the transmittingend within the time channel ZK16 and to provide that only following complete trans-mission is this code word inserted as a whole at the receiving end into the digital audio signal at a predetermined position.
Under these circumstances no shift occurs in respect of the code word boundaries within the pulse frames.
If it is assumed that a relative clock rate deviation of e.g. 5 . 10 5 occurs, under unfavourable circumstances clock rate matching need only occur every 0.316 s (approximately at spacing of 157 super frames) and in the event of negative clock ratematching the transmission of the signal word S need only be interrupted every 157 super frames.
In another possibility for negative clock rate matching, the transmission of the code word of the digital audio signal which has been gated out at the transmitting end is dispensed with, and the time of the clock rate matching is communicated to the receiver only by virtue of the two consecutive negative clock rate matching signals. The receiver replaces the gated out code word by an estimated value which is either e~ual to the previous code word, represents an extrapolation of two or more previous code words, or else is the mean value (geometric, arithmetic mean etc.) of the code words which are adjacent to the gated out code word. In this case there is no interruption of the trans-mission of the signal word S.
When a 15 kHz audio signal is transmitted, for examplein the six time channels ZKl,ZK2,ZK3,ZK17,ZK18 and ZK19, 12 bit time slots are additionally free at the positions normally occupied by the dialled characteristics Z17 to Zl9. These additional vacant bit time slots can assume functions other than transmission of the signal word S, such as for example the trans-mission of remote control signals.
A negative clock rate matching of the 15 kHz audio signal in accordance with the example illustrated in Fig. 4 also gives rise to a shift in the code word boundaries within the time channels ZK17 to ZKl9 as in pulse frame Rl 4 bits of the code word which is normally to be transmitted in time channel ZK17 and l 160773 in the first half of time channel ZK18 are transmitted as early as time channel ZK16. In pulse frame R2 following time channel ZK16 this code word-shift is increased to 8 bits and, in pulse frame R3 following time channel ZK16 the shift is increased to 12 bits, so that the beginning of the code word in time channels ZK17 and ZK18 conforms with the beginning of the time channel ZK17.
In the embodiments described above the clock rate matching signals ++ and -- are 8-position code words which are complement-ary to one another, so that their Hamming distance is 8. As aresult it is possible to correct up to three errors contained in these code words.
In terms of circuitry technology, use can be made of known modifiable stores (~linger) for the gating out and in of bit groups. The original clock rate assigned to the coded audio signal can be restored by means of a known phase locked loop (PLL) circuit with a low time constant in the regulating circuit.
The methods described above can be used for the clock rate matching of all digitalised analogue signals, in particular however for video signals and audio signals independently of the frame structure of the transmission system provided the frame structure possesses additional information capacity for the signalling of the clock rate matching signals.
The methods of clock rate matching described above allow the boundaries of the code words of the digitalised audio signal in the pulse frame to be left largely unchanged and thus permit rapid synchronisation at the receiving end in the event of transmission errors.
A plurality of clock rate matching methods are known for combining the data flows of a plurality of digital "sub-systems" having clock rates which are plesiochronous to one another to form a "super system". These methods may be split into those which do involve and those which do not involve information loss. If it were desired to use clock rate matching involving information loss in the audio signal insertion mentioned above, in order not to lose word synchronism it would be necessar~ to tolerate the omission or the repetition of an entire codé word from time to time. Investigations carried out by the present applicants have indicated that this measure would result in a fall in quality in the audio signal which would be unacceptable to the listener when the clock rate deviation exceeds 10 Clock rate matching methods which do not involve information loss in which the clock rate matching takes place bit-wise are known from the "Nachrichtentechnischen Fachher-ichten No. 42, 1972", PCM-Technik, VDE-Verlag GmbH Berlin-Charlottenburg, pages 235 to 244 and from the magazine "Elektronik", Vol. 6/1978, pages 78 to 83. In these methods, at an agreed point within the frame - generally of a "super system" - blank bits are interposed (positive clock rate matching) or information bits are gated out (negative clock rate matching). This procedure must be communicated by means of additional capacity available l 160773 in the pulse frame to the receiver which cancels the manipulation which has been effected at the transmitting end and reproduces the or-i-ginal clock rate. In addition to the procedures which use positive clock rate matching or negative clock rate matching and in which in the case of positive matching the clock frequency of the "sub system" is less than the clock frequency of the available data flow whereas in negative matching the clock frequency of the "sub system" is higher than that of the available data flow, the two possibilities can be combined as positive/
negative clock rate matching. In this case it is possible to compensate relative clock rate deviations in both positive and ne~ative directions.
The multiplex system which is disclosed in the "Taschenbuch der Fernmeldepraxis 1979" pages 13 to 41 is suitable both for digital transmission of audio signals and for'the combined transmission of audio signals and telephone signals in the pulse frame of the DSV2 system.
Fig. 1 ill~strates the ~nown structure of the pulse frame and of the super frame of a DSV2 system. This pulse frame of 125 ps duration contains 256 bits; this corresponds to a data flow of 2048kbit/s. The pulse frame is divided into 32 channels ZKO to ZK31. The first time channel ZKO serves alternately to transmit a message word and a frame code word (RSy). The seventeenth time channel ZK16 is provided mainly to transmit dialled characteristics Zl to Z15, Z17 to Z31 of the ihirty tele-phone channels Kl to K15, K17 to ~31. Sixteen consecutively transmitted ~ulse frames R0 to B15 fQnm one super fr~me. The synchronisation of the super frame is effected by means of a super frame synchronising word ~R-Sy in the seventeenth time channel ZK16 of the first pulse frame RO. For the sake of clarity the sixteen pulse frames of a super frame in Fig. 1 have been represented one below another.
The proposal disclosed in the "Taschenbuch der Fern-meldepraxis" permits mixed use of the DSV2 system for telephone channels and audio channels. In this case the first time channels ZKO and the seventeenth time channels ZK16 are not affected by the audio transmission.
In this digital audio transmission, channels of high quality possessing 15 kHz band width and channels of average quality possessing 7 kHz band width must be provided. In ~he former case the sampling frequency is 32 kHz and in the latter case 16 kHz. Eoth represent multiples of the sa~pling frequency of 8 kHz for teleFhone sign21s. Each sa~ple value with the protection bits in respect of the bit errors occurring on the transmission pa~h requires a word of 12 bits length. Therefore in one pulse frame four code words each comprising 12 bits must be accommodated for a 15 kHz audio channel and two code words each comprising 12 bits must be accommodated for a 7 kHz audio channel. Therefore six time channels ZKl to ZK3, ZK17 to ZKl9 each comprising eight bits are used for the transmission of a 15 kHz audio channel and three time channels ZKl to ZK3 are used for the transmission of a 7 kHz-audio signal. If further audio signals are to be transmitted, further time channels ZK4 to ZK6, ZK20 to ZK22 and ZK4 to ZK6 etc. are used. The bits of two consecutive code words are also interlocked in order to provide protection against double l 160773 errors on the transmission path. As a result a 15 kHz audio channel has an information flow of 384 kbit/s and an interface having a data rate of 384 kbit/s (in the case of a 7 kHz audio channel: 192 kblt/s) is provided between the coder/decoder device of the audio signal and the DSV2 system.
This known method of operation has the disadvantage that it merely provides a synchronous insertion of the information of coded audio channels into the frame of a DSV2 system. However this is not always possible such as for example:
- at a digital interface between a broadcasting studio and the digital network of the postal administration (or other authority) when the clock rates possess a permissible deviation of greater -than 10 7 from one another, - in national digital networks at the interfaces between parts of these networks which are not or not yet operating in synchronism and where the clock rate deviation is greater than 10 and - when digitaL audio signals are exchanged between international networks which although in themselves synchronous, are mutually plesiochronous with a clock rate deviation of greater than 10 7.
According to this invention there is provided a method for clock rate matching a digitalized audio signal to a data flow organized in pulse frames and super frames for telephone quality digital speech transmission and with characteristic bits for signalling transmission whose clock rate is plesiochronous to the clock rate of the digital audio signal, characterized in that in each super frame the clock rate matching occurs by word, i.e. an entire code word of the digital audio signal, and that in each super frame a positive and a negative clock rate matching signal is transmitted to the associated positions during transmission of the digital audio signal over several telephone channels of the characteristic bit, that with agreement of the clock rates of the data flow and of the digital audio signal, the clock rate matching signal is alternately positive and negative from super frame to super frame, that for a positive clock rate matching, the clock rate matching signals of two consecutive super frames are both positive and withln the second super frame -which contains the second of the positive clock rate matching signals directly following one another -a blank word is inserted at a predetermined point between two code words of the digital audio signal, and that for a negative clock rate matching, the clock rate matching signals of two consecut-ive super frames are both negative and freed positions of the characteristic bit in the super frame which follows the second negative clock rate matching signal are used with a code word gated out of the coded audio signal or with parts of code words gated out of the coded audio signal.
In an embodiment of the invention there is provided a method for clock rate matching a digital audio signal to a data flow organized in pulse frames and super frames for telephone quality digital speech transmission involving transmission of channel-associated signals as signalling bits in each pulse frame, the digital audio signal and data flow having plesiochronous clock rates, said method including the steps of inserting digital bits forming words of the audio signal into respec~ive bit time slots of a plurality of consecutive channels l 160773 of each pulse frame of a super frame, inserting into the vacant signalling bit time slots associated with the channels into which the audio signal has been inserted, a clock rate matching signal indicating whether the clock rates are synchronous or have a positive or a negative deviation relative to each other, adjusting the distribution of words of said audio signal in said plurality of consecutive channels when said deviation corresponds to one word length, such that said clock rate matching signal can indicate synchronous clock rates, said synchronism being indicated by the occurrence in alternate super frames of respective positive and negative deviation indicating clock rate matching signals, said positive indication being indicated by the occurrence in two consecutive super frames of positive deviation indicating clock rate matching signals, and said negative deviation being indicated by the occurrence in two consecutive super frames of negative deviation indicating clock rate matching signals.
Embodiments of this invention will now be described, by way of example, with reference to the accompanying drawings in which:-Figure 1 is a table illustrating a known form of super frame of a known data transmission system (Siemens system DSV2);
Figure 2 is a set of tables illustrating the structure of the data flow of a DSV2 system in the case of plesiochronous 7 kHz audio signal transmission in the synchronous state;
Figure 3 is a set of tables illustrating the structure of the data flow of the DSV2 system in the case of a positive clock rate matching of the digital 7 kHz audio signal; and 1 ~60773 Fig. 4 is a set of tables illustrating the structure of the data flow of the DSV2 system in the case of a negative clock rate matching of the digital 7 kHz audio signal.
Referring to Fig. 2, in each super frame of a trans-mission system embodying the invention, time channels ZKl to ZK3 are filled with 32 àudio signal words or code words Tnl to Tn32 each comprising 12 bits of a 7 kHz audio signal. If these three time channels ZKl to ZK3 are filled with the coded audio signal, the 3 times 4 bit time slots which become free in time channels ZR16 (originally conveying the dialled characteristics Zl to Z3 each comprising 4 bits) in pulse frames Rl, R2 and R3 can be used for clock rate matching. In each super frame the vacant 8 bit time slots in the pulse frames Rl and R2 are filled in alternate super frames with a positive and a negative clock rate matching signal ~+ and --. Thus each of the clock rate matching signals ++ and -- is divided between the pulse frames Rl and R2, whereas the remaining 4 bit time slots in the pulse frame R3 is filled with a signal word S which can be used for example to transmit alarms or for signalling purposes.
The synchronous state "no clock rate matching necessary"
need not be separately transmitted since the regular sequence ++, --, ++, -- etc. of alternately positive and negative clock rate matching signals does not trigger any reaction in the receiver.
Fig. 3 illustrates the case of a positive clock rate deviation, i.e. the clock frequency- of the audio signal is less than the corresponding clock frequency of the DSV2 system.
l 160773 If the time shift bPtween the two clock rates amounts to 12 clock periods, positive clock rate matching signals are trans-mitted in two consecutive super frames, and in the second super frame an information-free blank word LW is inserted at a position previously agreed upon with the receiver, here as the first code word following the frame code word RS~ in the pulse frame R3 of the second super frame. Thus only one code word of the 7 kHz audio signal is transmitted in the pulse frame R3 and within this super frame the number of transmitted code words is reduced by one to 31. This restores synchronism and in the following super frames positive and negative clock rate matching signals are transmitted alternately commencing with the negative clock rate matching signal.
In the case of negative clock rate deviation, illustrated in Fig. 4, the clock frequency of the audio signal is greater than the corresponding clock frequency of the DSV2 system. In order to signal this state, in two consecutive super frames negative clock rate matching signals are transmitted. Since a code word gated out of the audio signal must be additionally transmitted within a super frame, it is agreed that the super frame which follows the super frame with the second negative clock rate matching signal (the third super frame) should contain 33 code words.
In the third super frame, the bit time slots within the seventeenth time channel ZK16, which in the case of audio signal transmission are normally used for the clock rate matching signal and a signal word S, are filled in the pulse frame Rl with the 1 1~0773 first 4 bits of the code word ~n5, in the pulse frame R2 with the fifth to eighth bits of the code word Tn7, and in the pulse frame R3 with the last 4 bits of the code word Tn9. A resultant shift in the code word boundaries of the code words Tn5 to Tn9 S within the pulse frames R2 and R3 is cancelled out in the fifth pulse frame (not shown) and a tenth code word commences in the first bit time slot of the time channel ZKl in this pulse frame.
Here again, after the clock rate matching, positive and negative clock rate matching signals are transmitted alternately (commencing with the-positive signal) in the sub-sequent super frames.
It is also possible, in the case of positive clock rate matching, to insert 3 sets of 4 blank bits between code words within a super frame (e.g. 4 blank bits in three consecutive pulse frames) in order to achieve a less jerky clock rate matching than if a single 12-bit blank word were inserted.
It is equally possible in the case of negative clock rate matching to transmit a code word which is split into 3 sets of 4 bits and is gated out at the transmittingend within the time channel ZK16 and to provide that only following complete trans-mission is this code word inserted as a whole at the receiving end into the digital audio signal at a predetermined position.
Under these circumstances no shift occurs in respect of the code word boundaries within the pulse frames.
If it is assumed that a relative clock rate deviation of e.g. 5 . 10 5 occurs, under unfavourable circumstances clock rate matching need only occur every 0.316 s (approximately at spacing of 157 super frames) and in the event of negative clock ratematching the transmission of the signal word S need only be interrupted every 157 super frames.
In another possibility for negative clock rate matching, the transmission of the code word of the digital audio signal which has been gated out at the transmitting end is dispensed with, and the time of the clock rate matching is communicated to the receiver only by virtue of the two consecutive negative clock rate matching signals. The receiver replaces the gated out code word by an estimated value which is either e~ual to the previous code word, represents an extrapolation of two or more previous code words, or else is the mean value (geometric, arithmetic mean etc.) of the code words which are adjacent to the gated out code word. In this case there is no interruption of the trans-mission of the signal word S.
When a 15 kHz audio signal is transmitted, for examplein the six time channels ZKl,ZK2,ZK3,ZK17,ZK18 and ZK19, 12 bit time slots are additionally free at the positions normally occupied by the dialled characteristics Z17 to Zl9. These additional vacant bit time slots can assume functions other than transmission of the signal word S, such as for example the trans-mission of remote control signals.
A negative clock rate matching of the 15 kHz audio signal in accordance with the example illustrated in Fig. 4 also gives rise to a shift in the code word boundaries within the time channels ZK17 to ZKl9 as in pulse frame Rl 4 bits of the code word which is normally to be transmitted in time channel ZK17 and l 160773 in the first half of time channel ZK18 are transmitted as early as time channel ZK16. In pulse frame R2 following time channel ZK16 this code word-shift is increased to 8 bits and, in pulse frame R3 following time channel ZK16 the shift is increased to 12 bits, so that the beginning of the code word in time channels ZK17 and ZK18 conforms with the beginning of the time channel ZK17.
In the embodiments described above the clock rate matching signals ++ and -- are 8-position code words which are complement-ary to one another, so that their Hamming distance is 8. As aresult it is possible to correct up to three errors contained in these code words.
In terms of circuitry technology, use can be made of known modifiable stores (~linger) for the gating out and in of bit groups. The original clock rate assigned to the coded audio signal can be restored by means of a known phase locked loop (PLL) circuit with a low time constant in the regulating circuit.
The methods described above can be used for the clock rate matching of all digitalised analogue signals, in particular however for video signals and audio signals independently of the frame structure of the transmission system provided the frame structure possesses additional information capacity for the signalling of the clock rate matching signals.
The methods of clock rate matching described above allow the boundaries of the code words of the digitalised audio signal in the pulse frame to be left largely unchanged and thus permit rapid synchronisation at the receiving end in the event of transmission errors.
Claims (3)
PROPERTY OR PRIVILEGE IS CLAIMED ARE DEFINED AS FOLLOWS:
1. A method for clock rate matching a digitalized audio signal to a data flow organized in pulse frames and super frames for telephone quality digital speech transmission and with characteristic bits for signalling transmission whose clock rate is plesiochronous to the clock rate of the digital audio sig-nal, characterized in that in each super frame the clock rate matching occurs by word, i.e. an entire code word of the digital audio signal, and that in each super frame a positive and a negative clock rate matching signal is transmitted to the associated positions during transmission of the digital audio signal over several telephone channels of the characteristic bit, that with agreement of the clock rates of the data flow and of the digital audio signal, the clock rate matching signal is alternately positive and negative from super frame to super frame, that for a positive clock rate matching, the clock rate matching signals of two consecutive super frames are both positive and within the second super frame -which contains the second of the positive clock rate matching signals directly following one another -a blank word is inserted at a predetermined point between two code words of the digital audio signal, and that for a nega-tive clock rate matching, the clock rate matching signals of two consecutive super frames are both negative and freed positions of the characteristic bit in the super frame which follows the second negative clock rate matching signal are used with a code word gated out of the coded audio signal or with parts of code words gated out of the coded audio signal.
2. A method for clock rate matching according to claim 1, characterized in that for a negative clock rate matching at a predetermined point a code word of the digital audio signal is gated out at the transmitting end and replaced by an estimated value at the receiving end.
3. A method for clock rate matching according to claim 2, wherein said estimated value is the mean value of adjacent code words.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| DEP2947227.3 | 1979-11-23 | ||
| DE2947227A DE2947227C2 (en) | 1979-11-23 | 1979-11-23 | Method for clock adjustment for a digital audio signal to a data flow |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CA1160773A true CA1160773A (en) | 1984-01-17 |
Family
ID=6086688
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CA000365170A Expired CA1160773A (en) | 1979-11-23 | 1980-11-21 | Method of clock rate matching plesiochronuous signals |
Country Status (6)
| Country | Link |
|---|---|
| EP (1) | EP0029585B1 (en) |
| JP (1) | JPS5690646A (en) |
| AT (1) | ATE4263T1 (en) |
| CA (1) | CA1160773A (en) |
| DE (1) | DE2947227C2 (en) |
| NO (1) | NO803527L (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE2948435C2 (en) * | 1979-12-01 | 1984-09-06 | Aeg-Telefunken Ag, 1000 Berlin Und 6000 Frankfurt | Method for the transmission of up to four error-protected audio program signals in the pulse frame of the PCM 30 telephone system with the possibility of clock adjustment |
| DE3146317A1 (en) * | 1981-11-23 | 1983-07-28 | Philips Kommunikations Industrie AG, 8500 Nürnberg | Method for forming a time-division multiplex signal from broadband channels or analog signals |
| DE3146466A1 (en) * | 1981-11-24 | 1983-09-01 | AEG-Telefunken Nachrichtentechnik GmbH, 7150 Backnang | Method for trunking sound broadcast signals |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE2814351A1 (en) * | 1978-04-03 | 1979-10-11 | Siemens Ag | Frame structure for third order digital TDM system - with every 1056 signal bits preceded by 18 additional bits to combine 120 speech channels into rapid linear data flow |
| DE2826364C3 (en) * | 1978-06-16 | 1981-09-17 | Siemens AG, 1000 Berlin und 8000 München | Method for the transmission of error-protected audio program signals in the pulse frame of the PCM 30 telephone system |
-
1979
- 1979-11-23 DE DE2947227A patent/DE2947227C2/en not_active Expired
-
1980
- 1980-11-19 EP EP80107201A patent/EP0029585B1/en not_active Expired
- 1980-11-19 AT AT80107201T patent/ATE4263T1/en not_active IP Right Cessation
- 1980-11-21 CA CA000365170A patent/CA1160773A/en not_active Expired
- 1980-11-21 NO NO803527A patent/NO803527L/en unknown
- 1980-11-25 JP JP16575580A patent/JPS5690646A/en active Pending
Also Published As
| Publication number | Publication date |
|---|---|
| EP0029585B1 (en) | 1983-07-20 |
| NO803527L (en) | 1981-05-25 |
| ATE4263T1 (en) | 1983-08-15 |
| JPS5690646A (en) | 1981-07-22 |
| EP0029585A1 (en) | 1981-06-03 |
| DE2947227A1 (en) | 1981-05-27 |
| DE2947227C2 (en) | 1982-11-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP0216456B1 (en) | Multiplex structure | |
| US4397020A (en) | Error monitoring in digital transmission systems | |
| US3535450A (en) | Multiplex transmission method | |
| DE3642982A1 (en) | TRANSMISSION SYSTEM | |
| CA1208817A (en) | Fourth-order digital multiplex system for transmitting a plurality of digital signals at a nominal bit rate of 44 736 kbit/s | |
| US5717762A (en) | WACS-type mobile communication with a unified frame format | |
| US4755993A (en) | Transmission system using forward error correction | |
| CA1160773A (en) | Method of clock rate matching plesiochronuous signals | |
| US5228030A (en) | Time division communication system frame changeover arrangement | |
| EP0125802B1 (en) | Multiplexed digital data transmission | |
| CA1310436C (en) | Tdm data transmission system | |
| CA1157592A (en) | Method of clock rate matching plesiochronous signals | |
| US20030021270A1 (en) | Method of transporting frames of information between parts of a network through an intermediate network | |
| WO1982001094A1 (en) | Error monitoring in digital transmission systems | |
| AU5786280A (en) | Digital telephone system | |
| AU671631B2 (en) | Method for transferring ancillary information in a basic time-division multiplex system | |
| EP0252630A2 (en) | Transmission system using forward error correction | |
| JP2545538B2 (en) | Time division multiplexing transmission method | |
| JPS6223247A (en) | Digital time division multiplex system for transmitting video signal and narrow band signal | |
| US4847836A (en) | Circuit arrangement for synchronizing the units in the switching exchanges and repeaters of a time-division multiplex transmission system | |
| US6995695B1 (en) | System, method, and article of manufacture for synchronizing running disparity values in a computer and a data demapping device | |
| DE3339967C2 (en) | ||
| DE2948435A1 (en) | Transmission system for telephone network - sends up to four error-protected audio programme signals in pulse frame of telephone system and matches clocking | |
| JPH0252896B2 (en) | ||
| JPH04287589A (en) | Video signal multilplex system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| MKEX | Expiry |