DE2005836C3 - Method for transmitting a large number of binary messages over a transparent channel - Google Patents

Description

_Zei tpunktes the modulation Longer solid import _Sendeta if _t is _S. These are those ^ _{rend those of gsmein} a _me

' _for the respective message N 1 to the m J ^ 8 ^ ^

_each message the other _{on the geraemsa} men Ubertra au & Μ

characterized in that the step each after the change in the modulation state in the message to be transmitted Has prevailing polarity and that after the transfer of the steps mentioned for specifying the time of the change in modulation state again pulses with the prevailing polarity are transmitted if the next change in modulation status does not occur in the shortest possible time Distance follows.

30th

35

It is a method for transmitting a large number of binary-coded messages from several communication channels originating messages over a common transparent channel according to the time division multiplex principle known (DT-PS 12 65 247), in which the shortest code element of a message several times is scanned by the send clock and a message in the form of every change in the modulation status a binary coded pulse group with an indication of when - calculated from last send clock pulse before the change - the modulation state change has taken place, where the transmission of the message at the same time as the change in the modulation state in the respective Message channel following transmission clock begins, in which also several S ^ ndetaktimpulse for the transmission the binary coded pulse group can be used together and within the binary encoded pulse group the steps for specifying the time of the change in modulation state a start step is prepended.

The channels for the transmission of the individual messages are transparent except for a latching error which is caused by the scanning pulses and which can reach a maximum value of δ. This means that messages with any coding and any transmission speed up to a maximum step speed, which depends on the sampling frequency, can be transmitted. The principle of the procedure for transferring over

^ of _{the modo} | _a ^B , _io ; _s .

becomes a binary coded one

Counts transmit clock pulses S in units of t If there is a change in the modulation status in the message between two transmit clock pulses, the counter is stopped, the count is converted into a binary code combination and switched from the subsequent transmit clock pulses S as pulse group G to the common transmission line.

Thus, a transmission signal is only given by the message N 1 on the common transmission path when a change in the modulation state occurs in the message. If the same modulation state prevails over a longer period of time, no transmission signal is switched to the transmission path. In Fig. 1, the count Z 1 corresponds to the temporal position of the first change in the modulation state of the message / V 1, which is transmitted with the next transmit clock pulses S in the form of the binary-coded pulse group G 1. The second change in the message N 1 is identified by the count value Z 2, the third change by the count value Z 3 and the associated pulse groups G 2 and G 3.

In a coding device, the binary-coded pulse group G 1 to G 3 is formed from each of the count values ZX to Z 3. If a change in the modulation state has occurred, a further change in the modulation state can nominally only occur after a step duration s has elapsed. The binary coded pulse group therefore needs first and must be transmitted by this point in time. The pulse groups G according to FIG. 1 (line c) consist of three different types of steps. The first step, the "starting step", always has one and the same polarity; in Fig. 1 is this z. B. "1" This step forms the start criterion and informs the recipient that the following steps are to be assessed as related messages. The following steps are used to specify the time of the change in the modulation status. In Fig. I these are the second, third and fourth step (101) which contain the counting result Z 1, Z 2 and Z 3 as a binary number. The general result is the number of zui

Specification of the counting result required steps lus the following:

Allotted to a step duration s η counting ranges - an counting range corresponds to the distance between white send clock pulses S -, in each of which checked it is checked whether there is a change in the modulation status, so the time between two transmit clock pulses is -. In the time - * - between two sand pulse pulses S lie

nt

Sampling intervals (Fig. 1.8 intervals), one of which may contain a modulation state change. These sampling intervals are counted in each case. Each The counting result requires a binary one for representation

Code characters, so that different code

signs are necessary. Any of these binary code characters must have Ib Γ -— J zeitbc.Himmer.de steps (Ib = logarithm to base 21.

The last step of the binary coded pulse group indicates the modulation status after the change, ie a binary "0" and a binary "1 <'are transmitted alternately in the pulse groups. In total, each pulse group comprises π = 2 + Ib (j ^ j steps. This relationship results from the condition that the binary coded pulse group must be transmitted to the end of the shortest step. The number of counting areas at the end of each with one Transmit clock pulse S one step of the binary coded pulse group is transmitted must be at least equal to the number of steps of the binary coded pulse group. To transmit each of these 11 steps, one of the transmit clock pulses S is necessary during the step duration s . The transmission begins with the change in the modulation state The following transmit clock pulse S. If no pulse group is transmitted, the polarity of the modulation state opposite to that of the start step is always transmitted for the corresponding transmit clock pulses S; in Fig. 1 this is the state "0".

During a step duration s of the original message N 1, one of the binary-coded pulse groups with η =

2 + Ib

From this we can develop η = jf (fi) if Λ is given or

if 11 is specified. The transmission channel capacity of the continuing channel must therefore be greater by the factor η = jf (δ) than that of the channel for transmitting the original message N 1. This increase is a result of the signal conversion through the scanning and coding of the times of the change in the modulation state.

Au "of the magazine" Electronics "from January 19, 1970, page 9 E to 11 E, is already a method for rihertraeen a variety of binary coded, from

messages originating from several news channels via a common transparent channel known according to the time division multiplex principle, in which the changes in the modulation state are also coded as binary pulse groups are transmitted to the receiver, in which the permanent states are coded be transmitted and a reference signal must be transmitted over a special channel, the one Assignment of the pulse sequences used to transmit the permanent states to the switching states the original message allowed. The transmission of the reference signal requires additional transmission capacity in the common transmission channel. This also applies to that in the not pre-published DT-OS 19 34 869 described method for coding asynchronous data signals in which the modulation state changes can also be transmitted in coded form. With this method, too, the permanent states are displayed as pulse sequences »10 10 ...« or >> 0 1 0 I. . . ." transfer. The assignment of these pulse trains to one of the two permanent states is determined by a reference signal, a clock pulse. f '.- which must also be transferred and as a result, transmission capacity is consumed.

The invention is based on the object of transforming the original message so that the required one Channel capacity of the transmission system is smaller than with the known limitation periods.

This object is achieved by the invention specified in the claim.

In contrast to the known method described, the start shrill announcing the binary coded pulse group already has the new polarity. This new polarity is also transmitted after the steps for specifying the time of the change in modulation state, if the next change in modulation state does not immediately follow the binary-coded pulse group. It follows that, in contrast to the known method, the transfer of a special step indicating the new polarity in each case is superfluous. During a step duration? of the original message N 1, only m = 1 + lbf ^) steps must be able to be transmitted in the new method in the transparent channel. From this the function / 11 = Z ₂ (A) can be developed, or it applies

Steps can be transferred.

The channel capacity of the continuing channel also only increases by the factor m = / ₂ ' ^Λ ) This smaller increase in the required channel

ss capacities are particularly noticeable when the scanning abs * and (is relatively large, i.e. with a large permissible raster distortion D.

During the transition from the known method to the method according to the invention, the number of pulses

(, o the binary coded pulse group kept constant will. In this case, the number of time-determining increases in the process of driving according to the invention the pulses of the binary coded pulse group urr a pulse. The scanning distance I can be demented speaking are reduced so that the Rastver distortion ή decreases. The distance of the send clock impulse S and the number of impulses to be transmitted per step length remain constant. It is

however, it is also possible to leave the scanning distance t unchanged and instead the distance of the Send clock pulses S to increase. The number of pulses to be transmitted per step length increases then off.

If the number of zcitbestirniiienden pulses remains the same during the transition from the known method to the invention, so that the binary-coded pulse group is only one pulse shorter, the distance between the transmission clock pulses S and the scanning distance f can be increased accordingly while the step length ι ο remains the same. The raster distortion also increases. However, this increase is less than if the number of pulses in the binary-coded pulse group is reduced by one pulse in the known method (see table below). The number of pulses to be transmitted per step length decreases by one pulse. The following table gives an overview of the permissible latching distortions Λ that can be achieved for η or m:

ft 1 2 3 4th 5 Λ - 50% / ο 6V ₄ % , 2V ₂ % m 1 2 3 4th 5

A 100% 25%

3V _e % 1 \ ' ₄ %

δ. =

Impulse group now only comprises four impulses, you do not have the last impulse to mark the polarity. The starting step each pulse group has the respective new polarity, and the impulses between the impulse groups have the prevailing one Polarity. As in the example of the known method according to FIG. 1 are also omitted here on a stride length 48 sampling pulses T. The Rastverriing

The table has the relationships given above

35 1 - _L

.ν 48

has therefore remained constant.

In this example, the full step duration s is not used to transmit the binary coded pulse group G. The transmission of the pulse group G would only require four transmit clock pulses S; However, as in the known method, n; 'ch F i g. 1 six pulses per step length are transmitted. It can be shown, however, that there are areas of permissible raster distortion in which the method according to the invention requires only half as many pulses as the known method. During the duration s n or s _{m of} a binary coded pulse group, η (known method) or m pulses are transmitted. During the longer step duration s

then b „= η ■ - or b _m = ■■ m · - pulses

s ' ^s "'

transfer. It applies that - ^ - is greater than 1 and

^a nm

b "is greater than 11 or b _{m is} greater than m . The step duration s corresponds to the raster distortion A = - ■
Duration s "or s _m, on the other hand, represents the raster distortion

underlying. With the same number of steps η = m of the binary-coded pulse groups, the method according to the invention thus achieves a raster distortion that is 50% lower than with the known method.

If no change in the modulation state occurs in the method according to the invention, then in the rhythm of the transmit clock pulses continuously transmit the currently applied polarity. Follow binary coded Pulse groups close together, so the transmission of the prevailing polarity between the Impulse groups are also omitted.

The invention is explained using the exemplary embodiments described in the figures. It shows

F i g. 2 shows a time diagram for a method according to the invention,

F i g. 3 a time diagram for the known method,

F i g. 4 shows a time diagram for one from the method according to FIG. 3 derived method according to the Invention and

F i g. 5 is a diagram showing the required Enlargement of the transmission channel capacity.

F i g. 2 shows the principle of the invention using the example of a time diagram. The original message N 1 is shown again in line α. Line b shows the sampling pulses T and the transmission clock pulses S, and line c shows the G1, G 2 and G 3 characterizing the counting results. Every π-2 ""

The conditions apply

This results in the <during the step duration transmitted impulses:

6. =

b- =

These relationships are in that. Diagram ii F i g. 5 shown. The quantity I is plotted on the ordinate and the raster distortion δ on the abscissa in each case on a logarithmic scale.

Let m be given a certain number. If the detent distortion δ is now increased by reducing the step duration s, the step don approaches the value s _m , the detent distortion the value 6 _m and the number b _{m of} the pulses transmitted per step duration s the value nu which is not undershot

can. If s == s _m , then Λ = n _m and h _m = in. If the step duration. «; If the number is further reduced, the number ι »must be reduced by 1, since the binary-coded pulse group must not be longer than the step number s. The number of time-determining pulses thus decreases by 1. Therefore, the number of transmit clock pulses c S to be sent out per step duration and thus also the value of b _{m must be} doubled. With the values Λ = t) _m and δ = <\ there is a change in the values »i and η by 1 and a jump in the curves b _m and b". The diagram shows that the value of H _n (known method) is equal to the value h _m . if the relation π - 1 = m holds. ! If, on the other hand, η = »ι, then />", ie the

required transmission channel capacity. twice as big as b _m . These areas, in which the method according to the invention only requires half as many pulses as the known method, are between 25 and 16r _M , as can be seen from the diagram. 8 ¹ Z, and 6 ¹ Z ₄ . 3 ' ₈ and 2 ¹ Z ₂ . I ¹ Z ₄ and I ¹ Z ₁₄ ⁰ Zo etc.

An example for Λ ^ l '~' ₄ % In = 80 M / Eigen F ι g. ? and 4. In FIG. 3, the known method is shown again with it - 5 and b "- 10. The bits. \. \\: are the time-determining impulses, the bit P is the fidelity bit. The method according to the invention derived therefrom with in - 5 and /> _m = 5 is shown in FIG. U 4. The Bil is the fourth time determining "pulse.

For this purpose 3 sheets of drawings

Claims (1)

Claim: Method for transmitting a large number of binary-coded messages from several communication channels originating messages via a common, transparent channel based on the time division multiplex principle, in which the shortest code element of a message is scanned several times by the send clock and with every change in the modulation status a message in the form of a binary coded pulse group with an indication of when - calculated from the last transmit clock pulse before the change - the modulation status change took place hat, whereby the transmission of the message at the same time as the modulation status , _s one of these channels is shown in the timing diagram in FIG. I. Line α shows the message N 1 with the shortest Step duration s. This message is sent with a clock T scanned at intervals ί (line b). The greater r is, m spacings t augci _{r the} greater is the detent error 6 = y in the