CZ2000672A3 - Data transmission system with relay stations between the source and destination stations - Google Patents

Abstract

A decentralized digital data transmission system has a large number of distributed stations which can only engage in direct data communication with neighbouring stations. Stations located between a source station and a target station function as relay stations. Station to station transmission occurs in each case on different channels. In each station signals received by a reception channel are converted onto a different transmission channel. Conversion of a bit flow or symbol flow arriving at said reception channel is carried out on a bit-by-bit or symbol-by-symbol basis so that data is transmitted in a continuous flow without long delays occurring during conversion as is the case for transmission of data packets. A coupling matrix (KM) is used for conversion. Each channel (C1...Cn) is divided into a plurality of sub-channels (CH-1...CH-8). Transmission channels are allocated in such a way that each station uses wherever possible only sub-channels (SC1...SC8) located in the same channel.

Description

Technical field

The present invention relates to a data transmission system which provides digital data transmission, including, for example, speech.

BACKGROUND OF THE INVENTION

DE 33 37 648 C2 discloses a decentralized data transmission network with a plurality of separate stations in which direct data transmission occurs only between neighboring stations. A special routing determines the transmission path between the source and destination stations, and then the data is transmitted from station to station. The stations broadcast on one channel, which is used only for the connection of two stations.

However, a correspondingly adapted transmission power is necessary for this.

Furthermore, data transmission between the stations of the packet data transmission network is known from the Internet. In this case, the data is grouped into packets and these packets are then separately transmitted over the data link that is most convenient at the moment. Such packet transmission results in significant delays that correspond to at least the time required to transmit a single packet. For a telephone system, such a packet transmission is disadvantageous for the delays associated with this transmission. In particular, in a decentralized data transmission network, in which the transmission is always station-to-station, the delays would be summarized according to the number of stations involved in the transmission.

SUMMARY OF THE INVENTION It is an object of the present invention to provide such a decentralized digital data transmission system which allows the use of different transmission channels between two stations, while guaranteeing extremely low delays.

SUMMARY OF THE INVENTION

The data transmission system according to the invention is characterized in that at each station the signals received by the reception channels are transmitted by symbols to at least one transmission channel which is different from the reception channel. This means that each symbol stream arriving at a reception channel is converted to transmission channels. In doing so, to some extent, information packets are created which form only one single symbol. In the simplest case, the symbol is a bit. However, it may also consist of a plurality of bits belonging together, such as the letter symbol 8 bits, for example. In the subchannel, the number of bits of one symbol location is constant during transmission. At the beginning of transmission, the number of bits is determined for one • 99 · «» · ·

symbol depending on the required or required transmission quality level. Conversion in the form of symbols means that at each station only a delay of the symbol stream corresponding to the size of the symbol location is required. This delay is due to the fact that there is normally no synchronization of the sequence of symbols on the inbound and outbound channels, so a certain waiting time is required before the outbound signal can be sent synchronously with the broadcast channels. However, this delay is minimal. In practice, it involves about one to two symbol locations. The delays of the individual stations are added together. However, due to the small delay of each individual station, the resulting overall delay on the transmission path is still acceptable.

According to a preferred embodiment of the invention, the transmission channels are divided into subchannels, each of which is suitable for transmitting a single symbol stream, wherein the symbols of all the subchannels of a single transmission channel are transmitted synchronously over time. This means that each station can receive incoming signals on all channels. Outgoing subchannels can be transmitted concentrated to one single channel or to several selected channels. In each channel that consists of a predetermined number of subchannels, the symbol locations of all the subchannels are transmitted in time synchronous. Each subchannel is suitable for unidirectional data flow. The data arrives at the station in each subchannel of the entire channel system as a continuous data stream without breaking into "frames" or "packets". As a result, the data stream requires no labels or other defining elements. Conversely, each data stream symbol, upon its reception, is converted to the subchannel selected for transmission and transmitted in synchronization with the broadcast channel as soon as possible.

Assigning subchannels to a single transmission channel between two stations occurs by dynamically adapting the transmission frequency range to the amount of information to be transmitted. This means that the number of subchannels per channel can vary.

Preferably, the channels to be transmitted are allocated to one station such that all transmitting subchannels of that station lie within a few channels. This greatly reduces the number of channels to be used. It should be noted that if one station transmits on one channel, this channel must not be used by neighboring stations to avoid interference or other disturbances. If only one subchannel is used per channel, the entire channel is reserved for that station. Therefore, preferably all connections that go through a particular station are divided into subchannels that are all part of the same channel.

In order to reduce the likelihood of data link errors, in a preferred embodiment of the invention, the contents of the time-synchronously transmitted symbol locations of the subchannels of one channel are associated with the respective error correction bits, and error correction is then performed at the receiving station. Known methods such as the PEC (Self-Correction Code Security) method or the ARQ (Automatic Repeat Feedback) method can be used as an error correction method. A peculiarity in the present case is that the contents of time-synchronously transmitted symbol locations of the subchannels of one channel are used to correct errors, with information contents that are completely independent of each other flowing in the individual subchannels. This means that error correction is performed on the basis of bits belonging to different information and located only randomly at mutually coincident channel locations.

The application of the error correction method is effective only if the proportion o is to be undershot

binary errors of about 10 '. For higher bit error rates, simple error detection is sufficient to provide information about the quality of the connection between the two stations involved.

Such error detection may be accomplished by means of a redundant security appendage (e.g., a parity bit), the security appendage being assigned to time-synchronously transmitted symbol locations of all subchannels of a single channel. Alternatively, or in addition, after each transmission of a predetermined number of symbol locations of one subchannel for each subchannel, one error detection bit corresponding to successive information contents of the subchannel may be formed, wherein the error detection occurs at the receiving station.

BRIEF DESCRIPTION OF THE DRAWINGS

In the following, one embodiment of the invention is explained in more detail with reference to the drawings, in which Fig. 1 shows a part of a data transmission system with differently located stations; Fig. 2 shows an example of a connection of a source station to a destination station; 4 shows an example of the data streams that run through the switching matrix of FIG. 3, and FIG. 5 shows successive symbol locations of one channel with error correction bits and error detection bits.

DETAILED DESCRIPTION OF THE INVENTION

The data transmission system consists of a plurality of differently located stations S, each station representing a subscriber station. Each station contains receiving and transmitting devices. Two 12.8 MHz frequency bands are available for radio data transmission. The two frequency bands are separated by a duplex separation. One frequency band is labeled as uplink and the other as downlink. In the coupling assembly, an upper connection channel is used for connection in one direction and a lower connection channel is used for connection in the other direction, so that both directions are completely disconnected in frequency.

In this exemplary embodiment, both 12.8 MHz frequency bands are divided into a total of 1280 channels at 20 kHz spacing. Of these channels, some channels are used as information channels for the junction assembly and other purposes. Each station can receive and transmit on each of the available channels.

According to FIG. 1, it is assumed that a connection is to be established between the source station Sol and the destination station S65. This connection is via stations S60 and Sol, which act as relay stations. In addition, the connection from station S62 to station S64 is routed via station S60.

The example of the connection assembly shown in FIG. 2 assumes that information is transmitted from the source station Sol by channel C1, information is transmitted from station S6Q to station S63 by channel C25, and information is transmitted from station S63 to destination station S65 by channel 02. Station S60 here specially monitored, it also transmits via channel C25 for source station Sol and station S64.

Routing, that is to say the selection of the stations through which the connection is to be made and the selection of the channels concerned, is also ensured by the dialogue which the participating stations have between themselves. The routing and the coupling assembly are not the subject of the present invention.

FIG. 3 shows the switching matrix KM which each station contains. In this exemplary embodiment, for the sake of simplicity, it is assumed that each symbol location, represented here as a square, forms one bit.

Each station contains one channel register CR-1 ... CR-n for each of the channels C1 ... Cn. The CR-1 channel register comprises eight information symbol locations 1 ... 8, each of these symbol locations corresponding to one SC subchannel. Thus, channel C1 is divided into eight subchannels 1 ... 8. Each subchannel has a bandwidth of 20 kHz, with frequencies of all subchannels 1 ... 8 following each other immediately. A one-way data connection can be realized by one SC subchannel.

Fig. 3 shows the time position rasters of the symbol transmissions to the channel 1 register for subchannels 4 and 5 of channel C1. The transmission is performed at a frequency of 20 kHz by a continuous symbol stream.

The symbols that are received with time synchronization (here bits) from the subchannels of one channel come to one of the receive buffer registers ER1 ... ERn, where they are transmitted to the corresponding channel register CR-1 ... CR with a delay of two symbol times. -n. The channel registers CR-1 ... CR-n are always connected to the respective sensor matrix columns. The sensing matrix comprises n rows and m columns, each row and each column being assigned to a different subchannel, respectively. other frequency. The KM sensor matrix lines correspond to one subchannel respectively. one transmission frequency. One channel register CR-1 ... CR-n is designated for each channel, which contains one symbol location for each of the subchannels 1 ... 8. The symbol locations of all channel registers on the transmit side are connected to the scan matrix lines KM. Each channel register CR-1 ··· CR-n on the transmit side is assigned a transmit buffer register SR1 ... SRn.

* 00 «* 0 •

4 * ♦ ··· • Í Í • 0 ··

The sensor matrix KM is designed as an integrated circuit, whereby the nodes of the connection points of one row and one column can be connected by respective activation signals. The corresponding node remains closed during one connection.

In the illustrated embodiment it is assumed that the information which are received 5 sub-channel no. 1 channel Cl., ^{M and} j 'being sent further subchannel no. 2 channel C25. A switched node KP is located at the respective intersection of the sensing matrix, so that the bit located at location 1 of the CR-1 channel reception register is transferred to location 2 of the transmitting channel CR-25 for channel C25.

In the same way, the signals received by channel # 4 of channel C2 are transmitted to 10 channel # 6 of channel C25 and transmitted by this channel.

FIG. 4 shows an example of the timing of the signals that are received by channels C1, C2 and C3. At the station S60 of FIG. 1 and FIG. 2, the signals received here to be redirected to C25 are converted. For station S60, it has previously been established through a dialogue between neighboring stations that channel C25 is available for data transmission.

As now shown in FIG. 2, station S6Q receives from station Sol via channel C1 the data to be forwarded to station S63. For this reason, this data is converted to channel C25 at station S60. At station S63, the same data is then transferred to another channel, e.g., channel C12, and then transmitted to the destination station S65. Station S60 monitored here receives signals from channel S62 from channel S62 in the selected example. These signals are to be forwarded to station S64 and channel C25 is also selected for this purpose. Finally, signals from station S60 are still to be transmitted to station Sol, for which a different channel sub-channel C25 is selected. Everything transmitted by station S60 is implemented in channel C25, but in different subchannels.

Fig. 4 shows the conversion of those data at station S60 that is received by stations Sol and stations S62 in channels C1 and C2. This data is converted to channel C25 into different subchannels. 4, it can be seen that the symbol points transmitted in the channels C1, C2 and C3 are offset in time relative to each other for a maximum of one symbol point duration.

The data is therefore held in the CR-1 ... CR-n channel registers (Fig. 3) until the corresponding symbol location is received for all channels. This is followed by a synchronization to the output channels in the KM sensor matrix.

In addition to the symbol locations of the subchannels 1 ... 8 that carry the information, an additional three bit locations for the error correction bits A, B, C are also assigned to each channel. The contents of these additional bit locations are evaluated in the receive buffer registers ER1. ERn and used to correct errors in information bits that have been synchronously received within one channel. Only the corrected information bits are then inserted into the respective CR-1 ... CR-N channel register.

• I * v 9 ’* * 9 · v v 9 9 9 9 9 9 9 9 9 9 9 9

In the transmit registers SR1 ... SRn, error detection bits A, B, C are added to the eight information symbols of one channel before the entire number of bits is transmitted. These error detection bits are generated by an error detection algorithm according to the contents of the information symbol locations. In the same way, ie by applying the algorithm, error correction is performed after receiving the total signal.

In Fig. 5, the waveforms of the individual symbol locations are shown, with the numbers 1 ... 8 indicating the information symbol locations and showing the subchannels. These subchannels have different frequencies. The frequency f increases with increasing sequence number, from left to right in FIG. The last subchannel (channel "8") is assigned three bit error correction locations A, B, C.

In the embodiment shown in FIG. 5, an additional symbol point P is added to the channel for a total of eight consecutive symbols, which also includes a parity bit for each subchannel for error detection. The addition of error detection bits and error correction bits as well as the evaluation of these bits based on the information contents are performed separately for each data link. Frequency conversion is not involved in these additional bits.

As an alternative to the above embodiment, in which the frequency assignment to the subchannels is fixed, the frequency assignment to the subchannels may be changed after each symbol step. This ensures that the interference source cannot permanently disturb such a subchannel.

Claims (4)

PATENT CLAIMS A data transmission system with variously spaced stations (S), which, on individually selectable transmission channels, can only communicate directly with adjacent stations, the symbol stream consisting of successive symbols being transmitted by this channel and the stations located between the source station and the destination station have the function of relay stations, characterized in that in each station (S), the received signals are converted from the respective receiving channel to a transmitting channel that is different from the receiving channel, wherein the receiving channel, the transmitting channel being realized in the form of symbols. Data transmission system according to claim 1, characterized in that the transmission channels are divided into subchannels (SC), each of which is suitable for transmitting a symbol stream, and that the symbols of all the subchannels (SCI ... SC8) of one transmission channel are transmitted time synchronously. Data transmission system according to claim 2, characterized in that bits (A, B, C) are added to the time-synchronously transmitted symbol locations of the subchannels (SC1 ... SC8) of one channel, according to the information contents of these symbol locations, wherein the station detects and / or corrects errors. Data transmission system according to claim 2 or 3, characterized in that the error detection and / or error correction is realized by a predetermined number of symbol locations in each individual subchannel.