Method and apparatus for telephone systems for facilitating data traffic between the subscriber stations and a central data processor

Circuit arrangement for telephone systems with additional data traffic between the individual subscriber stations and a central data processing system.

For data transmission in telephone systems, it is already known to provide data terminals at the respective subscriber stations which are connected to the connecting line by means of special switchover devices and which enable Laten5 transmission via the telephone connection path. Such terminal devices are generally very complex and in many cases not necessary at all, for example not when a central data processing system should be able to be selected from several subscriber stations in order to receive certain automatic announcements, such as account balances , Stock quotes, inventory levels, etc., or to place orders. In these cases, the telephone itself can rather be used as a terminal, the data being entered by means of the dialing device, while the data output by the data processing system takes place acoustically in the form of sounds or speech and can therefore be answered via the telephone handset.

A disadvantage of such data services with acoustic

2o data output by the data processing system is that the often requested document about the dialog with the data processing system is missing. The invention solves this problem in that the acoustic data output from the data processing system to the subscriber station

VPA 9/610/1125 Kt / Stl

- 2 • 1 data to be sent for the purpose of controlling a printing unit at the

Information supplemented in a code suitable for controlling the printing unit can be sent via the same connection path.

Based on this general solution principle, there are now a number of possible solutions, one of which is very simple: an acoustic and a digital one

1o. Output channel of the data processing system are coupled to the same connection line of the telephone system and both output channels are activated one after the other. The conversion of the acoustically output information into digital control characters for the printing unit control can be carried out per program by the data processing system itself.

By combining acoustic and digital output channels, however, the number of output channels of the data processing system that can be controlled by the telephone system becomes

2o significantly reduced. It is therefore more advantageous if in the case of sound or analog speech output parallel to the sound or speech generator, a conversion device converting the control signals for the sound odei * speech generator into the acoustic data equivalent code characters, which is also coupled on the output side to the connection path to the subscriber station Interface of the output channel of the data processing system is connected. In this case, only a single output channel of the data processing system is required. As a consequence of the restriction to sound or analog speech output, the information supplied to the sound or speech generator by the data processing system is converted directly into equivalent code characters for the printing unit control without Schvd. and with relatively little effort. The same interface can also be used to transfer the information.

VPA 9/610/1125

......

·-·/» ..

«Y -« »·

<img file="LU65853A1_D0001.tif" />

<img file="LU65853A1_D0002.tif" />

?·'&lt; *' ·

A particularly advantageous solution according to the invention is, however, that in the case of sound or digital speech output, the signals required for controlling the Druckvzerka at the subscriber station are likewise generated by the sound or speech generators assigned to the output channel, which are sent after transmission the equivalent code characters for the printing mechanism are sent to the acoustic data. The sound and speech generators are used twice in this case, namely once for the acoustic Dafcenausga.be

1o on the other hand for the generation of digital code characters for the printing unit control. The data is implemented by the data processing system, so that additional conversion devices are also not required. Those to be sent out by the sound or speech generators

Code characters are generated by simply pressing the individual frequency channels. In order for the drawing steps to be as long as possible, it is expedient, according to a development of the invention, that when a digital speech generator is used, the frequency of the pitch generator

2o is chosen so high that, with reference to the spectral channels used, the settling and decay time of the channel filter (s) is negligibly small compared to the frame frequency that defines the step duration of the transmission code for the printing unit control for controlling the speech generator by the data processing system.

Further details of the invention will be explained in more detail below with reference to an exemplary embodiment shown in the drawing: FIG. 1 shows an overview circuit diagram. It

3o is divided into three parts, namely the subscriber station

TIN, the switching center VST and the data processing system DVA.

The subscriber station TLN consists of a known one

Telephone set FG with keypad selection, which at the same time as

VPA 9/610/1125

- 4 1 data terminal is used. Furthermore, a further data input device DS, for example a hole, is via the changeover switch U. Strip transmitter or data reader, can be connected to the telephone line, whereby a telephone system of the telephone set is also activated. A printing unit DE is permanently coupled to the connecting line via a line coupler 1I &lt;ί, so that the printing unit can also be controlled when the telephone station is used as a data terminal. The connecting line of the subscriber station is via a subscriber circuit TS

1o connected to the switching network KN of the switching center VST, which, in a manner known per se, is used in connection with the central control ZST to establish the respectively desired connection. Outgoing transmissions UE are also shown, via which the connection to the data processing device DVA runs.

The data processing system is divided into the actual processing unit DV, a channel controller K-ST with the output channels K1 to Kx and the individual channels individually assigned channel units KE for. the input and output of data via the telephone connection lines L1 to Ln.

The channel units KE each consist of an input part EIN for receiving the data sent by one of the subscriber stations TLN, an output unit A .... and a common input / output buffer P. The output units can be used as voice output units A-SP or, depending on their intended use but as output units for digital data, for example the output unit · A_rD, be trained.

3o The in traffic between a subscriber station TLN and the

Data processing system DVA takes place roughly in the following way: The subscriber reaches the inputs of the data processing system by dialing a predefined telephone number.

The data processing system then reports with a

Continuous tone. The choice of γ / other digits indicates which one

VPA 9/610/1125

- 5 1 special program is desired. These digits serve the system program of the data processing system as an elimination criterion for loading and starting the desired program.

The participant has a dialogue with the desired program. recorded, the keyboard serves as an input device for further information. The data processing system answers via the speaker. In the simplest Pall, certain tones are sufficient as an answer. In many cases, it is expedient that the data processing system answers directly with spoken words. For this purpose, it controls a digital or an analog voice output device via the channel control.

In addition to the acoustic data output, according to the invention there is the possibility of additionally controlling a printing unit DE at the subscriber station TIN, which produces a record of the data dialog carried out. According to one possible solution, the telephone connection line 11 is connected to a second channel unit KE via a line coupler LK

2o printer output part AD and a second channel K2 of the channel controller K-ST connected. Both channel units working in this way on one and the same connecting line are controlled one after the other by the data processing device DV, and thus the output data are sent one after the other in different representations over the connected telephone line.

Another option for controlling the printing unit at the subscriber station is shown with reference to the Pn line Ln.

5o In this case only one channel Kx of the channel control K-ST and one channel unit KE are required. With this channel unit. however, a separate conversion device UM is coupled.

This consists of a buffer memory P, a handler W for converting the supplied data into equivalent code characters and a transmitter S, which is coupled to the telephone line via a line coupler LK. Lie control

VPA 9/610/1125

<img file="LU65853A1_D0003.tif" />

the conversion device UM is expediently carried out via the same

Interface S of the channel unit, via which the output section • A-SP is also controlled. Both devices, channel unit KE and conversion device UM can work together in such a way that all the data supplied to the output part A-SP are simultaneously fed to the input buffer P of the conversion device, where they remain stored until the acoustic data output has ended. Although this requires increased storage capacity in the conversion device, it relieves the load on the data processing system, since it does not need to control the conversion device separately. The solution inequality shown is

- '·' but only applicable to sound output or analogue

Spraohausgabe, in which the digital information supplied to the speech output A-SP by the data processing device LV represents addressers) with which the speech components recorded analogously on a carrier can be selected in the form of syllables and words.

The cheapest option for controlling a printing unit at the subscriber station is sound or digital voice output, since the output parts can also be used as transmitters for controlling the printing unit. The prerequisite for this is that the tones or tone combinations when the sound is output or the information determining the control signals of the θ 25 digital speech output is converted earlier into equivalent code characters, which then determine the keying of the individual frequency channels. This conversion can be carried out by the data processing device DV itself without difficulty.

3o

To better understand this process when using digital speech outputs, the basic structure and the mode of operation of such a speech output are first explained in more detail with reference to FIGS. 2 and 3. The block diagram according

Fig. 2. shows a signal distributor SV as input stage VPA 9/610/1125

- 7 k;

<img file="LU65853A1_D0004.tif" />

of the speech output section A-SP, to which the necessary information is fed via the input / output buffer P. Spectral channels 1 to n, each with a channel part SK .. ·, and a close-coupled filter F ... so5 are connected to the signal distributor 'like a pitch pulse generator THG. In addition to the pitch generator, a further noise generator RG is provided. Both generators can be connected to the individual channel units SK1 to SKn via switch S as required.

The outputs of the individual filters F1 to Fn work on one

1o common output amplifier V with output A.

The voice output is controlled in such a way that, for example, every 20 ms from the data processing device, a pulse frame with 48 bits is sent to the input / output buffer P.

is handed over. 3 is divided into eight words W1 to W8, each with six bits B1 to B6, which are optionally transferred to the signal distributor SV of the output unit A-SP and together characterize the current speech spectrum. The word W1 gives with the

2o Dit T1 to T6 for voiced sounds the height of the basic speech frequency and thus the pitch. Unvoiced sounds are characterized, for example, by the pattern 000111, which, when the switch S is flipped, causes the noise generator RG to be connected to the spectral channels. The words W2 to W8 contain information about the current power in relation to the individual spectral channels. For example, the three bits SK11 to SK13 in word W2 relate to spectral channel 1 and bits SKn1 to SKn3 in Y / location 8 relate to spectral channel n.

3o

The information supplied to the pitch pulse generator THG controls the frequency of the pitch pulse generator and the pulse width via a digital-to-analog converter.

The pulses generated by it provide the excitation energy. VPA 9/610/1125 for the voiced sounds. The information supplied to the spectral units SK .... regulates the amplitude of the pulses supplied via the switch S via a digital-to-analog converter, from which spectrum the respective one subsequently

Channel frequency is filtered out.

Due to the fact that it is possible to use different spectral channels in the time grid defined by the pulse frame frequency, only one or even no spectral channel

To excite 1o, digital code symbols can also be sent instead of speech elements. Both serial and parallel codes can be used. The selection (of the individual character channels) suitably depends on the other boundary conditions, which are given, for example, by the type of transmission channel, the dialing method used and other system properties.

The height of the pitch pulse frequency can also have an influence, namely if the individual harmonics of the pitch pulse frequency

If they are too close to one another, there is a risk that several harmonics, albeit damped differently, can pass through a filter, so that the amplitude modulation is increased. A uniform output signal with a large amplitude can be achieved with reference to a selected spectral channel if the decisive harmonic of the fundamental pitch frequency lies in the middle of the band of the selected channel and is therefore not damped.

The height of the pitch pulse frequency resulting at the highest pitch also has an influence on the stride length of the individual drawing elements. At a pitch pulse frequency of 200 Hz, for example, the start of excitation of the selected channel can fluctuate by a maximum of 5 ms. Since this fluctuation is possible both when the channel filter swings in and out, the drawing steps are of the same length despite the default

Pulse frame frequency not easily guaranteed. UnitÿPA 9/610/1125 • Λ.

//3 ·&lt;·'&lt;·&amp;

y.

.γ &lt;·. ·, Λ. &lt; '···: ·

<img file="LU65853A1_D0005.tif" />

Long stride lengths can only be achieved if the settling and decay times of the channel filters remain small compared to the pulse frame frequency. However, this is only the pall if the pitch pulse frequency is chosen to be as high as possible at the highest pitch. For example, if you choose a pitch pulse frequency of 500 Hz, the time tolerance of the channel excitation is reduced to a maximum of ms. The fluctuations in time associated therewith are therefore considerably less and readily allow one

1o reliable evaluation with a fixed clock pattern.

The following recoding phases result from the internal representation code of the data processing device DV. The character to be output in the internal code is first converted into a character of the transmission code. In the case of a serial character, each bit of the transmission code character is then replaced by a pulse frame as shown in FIG. 3. With the parallel code, however, a single vocoder pulse frame is sufficient for each character.