GB2139050A - Digital data transmission - Google Patents

Abstract

A digital data multiplex system transmits two packet types identified by respective packet type indicators comprising respective codes separated by a substantial Hamming distance. Advance warning of a change of packet type is given by transmitting in advance of the change one or a sequence of intermediate codes, in which some of the bits are altered from those of a defined one of the codes, but which is still closest in Hamming distance to that one of the codes. The intermediate codes are detected to give the advance warning, yet are correctable to the defined one type code.

Description

SPECIFICATION Digital data transmission It has been proposed in a data multiplex system based on the use of fixed length packets of data, that packets should be given different addresses in a packet header to indicate different types of or destinations for the data in that packet. For example in a data multiplex system proposed for use in transmitting digital sound signals as part of a television signal, different sound channels may carry a commentary in different languages and the commentaries will be handled in different circuits at the decoder.

In such a system the data packet may be preceded by a header the first active byte of which is known as the packet type byte. The packet type may, typically, take one of four different forms, termed respectively a Bl byte which indicates an identification block, BC1 or BC2 which indicate two different types of data packet, and a reference code which will for the purposes of this specification be termed BR.

When the Bl code is used, the packet contains control information relating to the sound service as it is or will be after the next change in configuration. Such changes in configuration are in general indicated by a change from the BC1 to BC2 code, or vice versa.

In order to assist discrimination between these four codes, even in the presence of errors, they are selected to be readily distinguishable. An example of four such codes, each consisting of 8 bits, is as follows: BI 0 0 0 0 0 0 0 0 BC1 1 1 0 0 0 1 1 1 BC2 1 1 1 1 1 0 0 0 BR 0 0 1 1 1 1 1 1 These four codes are selected so that they have Hamming distances of 5 or 6 from each other. The Hamming distance is the number of bits which have to change to transfer from one code to the other. This means that even where the byte contains two digits in error, a closest one of the four codes can always uniquely be detected.In the case of the four exemplary codes given above, the Hamming distances are in fact as follows: BR BC2 BC1 BI 6 5 5 BC1 5 6 BC2 5 It is known in data communication systems generally to use sets of address codes with 'Hamming distances' of three or more from each other in order to improve the reliability of address detection and decoding. Without this feature there is a signficant probability that, in the presence of bit errors in the received data, information may be ignored or even sent to the 'wrong' destination. The use of such distanced codes allowes received codes with one, or sometimes more than one, bit error to be 'corrected' to the 'nearest' admissible code. Alternatively these, and codes with perhaps even more errors, can be identified as erroneous and a suitable action can be taken.Of course, in extreme cases of multiple errors a corrupted code can approach, or even exactly imitate, another valid code and a mistake is made.

Methods for decoding such distanced codes are well known, in the past it has been common to use codes with a mathematical structure, based on modulo-2 arithmetic. This made it possible to interpret the codes using multiple 'parity' checks using 'exclusive-or' gates. The Hamming codes used in the United Kingdom teletext system are of this type. Alternatively a programmable read-only memory can be programmed with all possible received codes together with the corresponding outputs as 'courses of action'. This approach can, of course, be applied to the structured codes as well but it is equally applicable to all sets of distanced codes.

However, in practice it is found that in the system described above when there is a change from the BC1 to the BC2 code, or vice versa, this change is announced abruptly and there can be difficulties in ensuring that the decoder always responds properly to this change.

The invention is defined in the appended claims, to which reference should now be made.

In particular, in accordance with this invention advance warning of a change between two distanced codes can be given by transmitting in advance of the change an-intermediate code in which one or more of the bits is deliberately altered.

This invention is thus concerned not with the steady-state reception and interpretation of distanced codes, but with the correct response to an anticipated change between the use of one particular code and another code of the set, when the actual timing of the change cannot be foreseen. The exact timing of this change of code is, in the illustrative example, being used to synchronise a major change within the decoder, such as a change of audio bandwidth in a digital sound service, or the change to the next encryption key in a subscription television service, so a failure to resppnd to the change at exactly the right instant can give rise to serious imperfections in the received service.

The example illustrated allows advance warning of a change to be signalled by inserting a systematic sequence of different codes leading up to the change. These different 'preamble' codes are chosen to be 'close' to the current code and still 'far' from the other codes. This means that the normal error correction of a decoder will restore the current code, unless there are further errors present (in which case there is greater need for this advance indication). But the decoder can be programmed to recognise these particular 'preamble' codes in sequence and so build up a very reliable count-down to the event itself.

The code detector may have additional outputs for the preamble codes and these would be supplied to a code change detector, clocked at the incoming code rate. The arrival of the first preamble code (which, of course, is dependent on the current code) sets a counter in action, and the arrival of the second and subsequent codes allow the counter to decrement towards the 'switchover' event. The-logic can be arranged so that occasional lost codes do not cause the counter to stop or reset, so that the final decision is, for example, based on the evidence provided by, say, at least six out often preamble codes received in the correct sequence. The change detector also needs to respond in the normal way to an abrupt change from the current code to the next code if it is required to be compatible with transmissions not using this facility.In any case a long-term majority decision on the 'current' code in use will be needed to allow the decoder to start operation, or to recognise that a change had taken place even if it failed to detect that change owing to the presence of errors.

An example of the invention will now be given. Using the four distanced codes given above, a change from code BC1 to BC2 may take place over some ten packets in accordance with the following sequence: Change from BC1 to BC2 Incremental Hamming Distances Hamming from Steps Codes Distances BR BI BC1 BC2 10+ 1 1 0 0 0 1 1 1 5 5 0 6 2 9 1 1 0 0 1 1 1 0 5 5 2 4 4 8 1 1 0 1 0 1 0 1 5 5 2 4 4 7 1 1 1 0 0 0 1 1 5 5 2 4 4 6 1 1 0 1 0 1 1 0 5 5 2 4 4.

5 1 1 1 0 0 1 0 1 5 5 2 2 4 4 4 1 1 0 0 1 0 1 1 5 5 2 4 4 3 1 1 1 0 0 1 1 0 5 5 2 4 4 2 1 1 0 0 1 1 0 1 5 5 2 4 4 1 1 1 0 1 0 0 1 1 5 5 2 4 4 0 1 1 1 1 1 0 0 0 5 5 6 0 It will be seen that this sequence starts with the BC1 code and then cycles through nine intermediate codes all of which have a Hamming distance of 2 from code BC1, and a Hamming distance of 4 from code BC2. The Hamming distances from codes Bl and BR remain at 5 throughout.

A decoder equipped to respond to the code sequence then has advance warning of the change 9 packets before the change takes place as is indicated by the dashed line on the table.

The system is compatible for use with decoders which do not respond to the intermediate sequence. This happens because all the intermediate codes retain a Hamming distance of not greater than two from code BC1, and will therefore be decoded by a conventional decoder as code BC1.

When changing from code BC2 to BC1 the codes given in the following table may be used.

Change from BC2 to BC1 Incremental Hamming Hamming distances from Steps Codes Distances BR BI BC1 BC2 10 1 1 1 1 1 0 0 0 5 5 6 0 2 9 1 1 1 1 0 0 0 1 5 5 4 2 4 8 1 1 1 0 1 0 1 0 5 5 4 2 4 7 1 1 0 1 1 1 0 0 5 5 4 2 4 6 1 1 1 1 0 0 1 0 5 5 4 2 4 5 1 1 1 0 1 1 0 0 5 5 4 2 4 4 1 1 0 1 1 0 0 1 5 5 4 2 4 3 1 1 1 1 0 1 0 0 5 5 4 2 4 2 1 1 1 0 1 0 0 1 5 5 4 2 4 1 1 1 0 1 1 0 1 0 5 5 4 2 4 0 1 1 0 0 0 1 1 1 5 5 0 6 Again, the intermediate codes will be detected by an unmodified decoder as being codes BC2. A specially modified decoder can be programmed to respond to the intermediate codes so as to give a count down to the instant when the change to code BC1 takes place.

It is of course possible to shorten the intermediate code sequence by jumping from any desired one of the intermediate codes directly to the final code: In an extreme only a single intermediate code may be used.

Each of the intermediate codes differs from the next subsequent code by a Hamming distance of 4 which makes the intermediate codes readily distinguishable from each other.

It will be appreciated that the codes given above are simply by way of example. The codes could be modified in principle by adding any fixed byte, or by permuting the bit order, without affecting the operation of the system as a whole.

Claims (7)

1. A digital multiplex system in which at least two packet type indicators are used to distinguish different packet types, in which advance warning of a change between packet types is given by transmitting in advance of the change an intermediate code in which one or more of the bits are deliberately altered.

2. A system according to claim 1, in which a series of intermediate codes are employed in a defined sequence.

3. A system according to claim 1 or 2, in which the packet type indicators comprise codes separated by a minimum Hamming distance, and the intermediate codes are selected to be closest to a defined one of the packet type indicator codes.

4. A digital multiplex system in which at least two packet type indicators are used to distinguish different packets, the packet type indicators comprising respective distanced codes, and in which signalling information is given by transmitting an intermediate code in which one or more of the bits are deliberately altered from those of a defined one of the codes but which is still closest to the said one of the codes.

5. A digital multiplex system substantially as herein described by way of example with reference to the foregoing tables.

6. A digital multiplex transmitter for use in a system in accordance with any of the preceding claims for transmitting a signal including the said intermediate codes.

7. A digital multiplex receiver for use in a system in accordance with any of the preceding claims and responsive to the intermediate codes.