EP1625685A1

EP1625685A1 - Method and arrangement for transmitting data by using ofdm and tdma

Info

Publication number: EP1625685A1
Application number: EP03700290A
Authority: EP
Inventors: Walter Braun
Original assignee: Ascom Powerline Communications AG
Current assignee: Current Communications International Holding GmbH
Priority date: 2003-01-28
Filing date: 2003-01-28
Publication date: 2006-02-15
Also published as: AU2003201593A1; WO2004068780A1

Abstract

The invention relates to a method for transmitting data in a network, wherein the transmitter/receiver devices for transmitting the data are connected to a line-bound transmission medium (5), especially a power supply network. Data is transmitted from lowlink transmitter/receiver devices to an uplink transmitter/receiver device (7) on several subcarriers which are provided by FDMA (Frequency Division Multiple Access) and preferably FDM (Frequency Division Multiplex) and which are orthogonal in relation to each other. Said subcarriers are further divided into time slots by TDMA (Time Division Multiple Access). Said method enables high spectral efficiency in the uplink and good utilisation of band width without increasing linearity demands on a receiver (12) used in the uplink transmitter/receiver device (7).

Description

METHOD AND SYSTEM FOR TRANSMISSION OF DATA IN _V ERWEND _UNG OF OFDM AND TDMA

Technical field

The invention relates to a method for transmitting data in a network with a superordinate transceiver and a plurality of subordinate transceivers, the transceivers mentioned for the transmission of the data being connected to a line-bound transmission medium, in particular a power supply network and data are transmitted from the subordinate transceivers to the superordinate transceiver on several subcarriers. The invention further relates to an arrangement for carrying out the method. State of the art

In networks for the transmission of data via a wired transmission medium, e.g. For example, a power supply network, a transmission protocol is often used, which provides that all data traffic is routed via higher-level transceivers, so-called masters. These masters control the distribution of the available transmission channels to the subordinate transceivers, the so-called slaves. For the transmission of data from the master to the slaves (downlink), the available frequency band can be divided into several subcarriers using FDM (Frequency Division Multiplex). These subcarriers can then, for. For example, individual slaves can be assigned so that multiple transmissions are possible independently of one another.

However, the different subcarriers often have a very different quality, e.g. B. because the lines from the master to different slaves have a different length or have different attenuation or because the slaves are technically different. In such a situation, OFDM (Orthogonal Frequency Division Multiplex) offers the possibility of adapting the modulation format, which determines the number of bits per transmitted symbol, to the signal / noise ratio of the individual transmission channels. The mutual influence of the subcarriers is minimized in that the subcarriers are orthogonal to one another. OFDM allows better use of the available frequency bands and therefore has a high spectral efficiency.

However, the use of different modulation formats with OFDM means that the signals transmitted on different subcarriers from the slaves to the master (uplink) have very different power levels. This means that a receiver with extremely good linearity is used in the master so that these signals can be further processed, in particular demodulated and decoded. However, such receivers have a complex structure and are correspondingly expensive. In order to avoid this effort and to be able to use simpler receivers, the same modulation format was previously used on all transmission channels. As a result, the poor quality of a transmission channel in the uplink determined the common modulation format for all transmission channels. As a result, however, bandwidth was lost for all transmission channels of better quality.

Presentation of the invention

The object of the invention is to provide a method belonging to the technical field mentioned at the outset which enables high spectral efficiency and good utilization of the bandwidth in the uplink without increased linearity requirements on the receiver.

The solution to the problem is defined by the features of claim 1. According to the invention, the plurality of subcarriers are provided by FDMA (Frequency Division Multiple Access) and preferably FDM (Frequency Division Multiplex), are orthogonal to one another and are further divided into time slots by TDMA (Time Division Multiple Access).

The subdivision of the subcarriers provided by FDMA and possibly FDM into several time slots leads to greater flexibility: the master can specify a certain modulation format for a certain number of time slots and a different one for a next number of time slots. It can also control the transmissions from the slaves, e.g. B. by assigning each slave certain time slots for transmission. The master can make this assignment, the choice of the modulation format and the number of time slots for which this should apply, e.g. B. adjust the current quality of the subcarrier.

The signals which are used on the various subcarriers for data transmission are orthogonal to one another by using OFDM (Orthogonal Frequency Division Multiplexing). Due to the orthogonality there is a certain spectral overlap the subcarrier allows, thereby the subcarriers can be arranged "closer" in the frequency spectrum and the spectral efficiency is improved.

It is also possible to use orthogonal and non-orthogonal subcarriers at the same time. However, care must be taken to ensure that overlaps of the subcarriers do not lead to interference, such as so-called adjacent channel crosstalk. Such disorders can e.g. B. can be avoided by controlling the allocation of the time slots accordingly, e.g. B. so that adjacent and not orthogonal subcarriers in the frequency band are not used simultaneously in the uplink.

Advantageously, the transmission power levels of the subordinate transceivers are set so that in a given time slot all data transmitted by the subordinate transceivers are received at the superordinate transceivers with approximately the same reception power levels. The power control for the slaves means that even with subcarriers or transmission channels with different attenuation, all signals of the slaves arrive at the master at approximately the same power level, so that the linearity of the receiver is not critical for reliable demodulation and decoding of the received signals , The reception power level at the master can be selected differently for different time slots or groups of time slots.

The power control can - as known from the prior art - z. B. done by a closed loop (closed loop power control). If the network is configured or can be configured so that the attenuation on all subcarriers used at the same time is approximately equal to all subordinate transceivers, an individual power control can be dispensed with.

The orthogonality of the data transmitted from the lower-level to the higher-level transceiver in principle means that there is no mutual interference between the data transmitted on different subcarriers. However, signals that are transmitted from different slaves to the master will always have certain frequency errors that lead to interference between the subcarriers. The frequency errors are continuously corrected by the slaves by making an estimate of the error based on the downlink signal. The accuracy of this estimate is decisive for the measure of remaining frequency errors and thus for the interference that occurs. The stronger the interference, in turn, the smaller the permissible performance difference between different subcarriers. In the event of large frequency differences, the interference which occurs can be reduced by leaving one or more subcarriers between subcarriers which are assigned to different slaves unused, so that the remaining interference is reduced to an admissible level. For receivers without increased linearity requirements, the permissible power difference between uplink signals from different slaves is typically between 10 dB and a maximum of 30 dB.

Different transmission rates are preferably used in different time slots. These result from different bit loading. The quality of the transmission channels changes over time. This is done z. B. by the different configuration of the network at different points in time, both in terms of the number and locations of the connected terminal devices and in terms of the current channel assignment and the currently transmitting or receiving transceivers. In addition, there are the usual fluctuations in transmission quality over time. If different transmission rates can be used in different time slots, these quality fluctuations can be taken into account.

Variable bit loading means that the number of bits that are transmitted with one symbol can be varied. This is made possible by various modulation methods, e.g. B. by PAM (Phase Amplitude Modulation), QAM (Quadrature Amplitude Modulation) or PSK (Phase Shift Keying). In general, transmissions with a higher number of bits per symbol and therefore a higher transmission rate are more susceptible to interference and require a transmission channel with higher quality than transmissions with a lower number of bits per symbol, ie the signal-to-noise ratio must be larger. In order to optimally take advantage of the variable transmission rates, the subordinate transceivers are preferably classified into two or more classes. The classification is based on the maximum possible transmission rate of the transmission channel from the subordinate transceiver (slave) to the superordinate transceiver (master). The data from transmitters / receivers of the same class are then transmitted in the same time slots if possible.

This leads to the fact that in a certain time slot predominantly transceivers transmit data to the master, which can use the same maximum transmission rate for this purpose. This transmission rate is then chosen as the maximum transmission rate for the transmission of all data in this time slot. Only if this means that the subcarriers are not yet "filled" in this time slot and capacity is still available can further transceivers, which could actually transmit on channels with a high transmission quality at a higher transmission rate, transmit some of their data. These transceivers reduce their transmission power accordingly, so that their power level does not significantly exceed that of the slower slaves. In this way, as few transceivers as possible are slowed down by slower transmission channels. The resulting loss of bandwidth is therefore minimized. Nevertheless, all signals from the slaves arrive at the master with approximately the same power level during the specific time slot. Demodulation and decoding is therefore possible even without increased demands on the linearity of the receiver.

The transmission is advantageously organized as follows: First, during one or more time slots, data of the transceivers are transmitted on all available subcarriers, the transmission channels of which correspond to the class with the lowest maximum transmission rate, namely with (this class assigned) lowest maximum transmission rate. This is followed, again during one or more time slots, by the data of the transceivers whose transmission channels correspond to a class with a higher maximum transmission rate. Accordingly, this data is transmitted at the higher maximum transmission rate. Depending on Number of classes and depending on the classification of the transmitting / receiving devices, this process is repeated until all data of the transmitting / receiving devices with the highest maximum transmission rate have also been transmitted.

The number of classes does not have to be constant and can be newly selected for each transmission cycle. It depends in particular on the number of subcarriers, the number of connected transceivers, the available transmission rates or the current data volume.

The number of subcarriers need not be constant and can e.g. B. selected differently for each transmission rate or adjusted to the parameters just mentioned.

The method is preferably implemented as follows:

a) A first number of time slots is determined, which is required for the transmission of all data of the transceivers in the class with the first, lowest maximum transmission rate;

b) this data is assigned to the available subcarriers during the first number of time slots;

c) in order to "fill up" further subcarriers during the first number of time slots, to which no data have yet been assigned, data from the transceivers of higher classes are assigned to these further subcarriers until all available subcarriers are "filled up";

d) the assigned data are transmitted during the first number of time slots on all available subcarriers, specifically at the first, lowest maximum transmission rate; e) next, a second number of time slots is determined, which is required for the transmission of all data in the class with the second, next higher transmission rate;

- f) this data is assigned to the available subcarriers during the second number of time slots;

g) further subcarriers, to which no data have yet been assigned, data of the transceiver devices of higher classes are again allocated during the second number of time slots until all available subcarriers have been filled;

h) the assigned data are transmitted on all available subcarriers during the second number of time slots, namely at the second transmission rate or a lower transmission rate;

i) steps e-h are repeated with the remaining data of the transceivers in the higher classes.

These repetitions of the individual steps continue until all the data of the transceivers in the class with the highest maximum transmission rate have been transmitted. It goes without saying that the transmission requirement in one class can be very low, so that the "filling" of the remaining time slots of the next lower class is sufficient to transmit all data of the class. In this case, the transfer of data of the next higher class is simply continued. Because lower transmission rates can also be used in step h, it is also possible to use subcarriers with a high noise level, which would not allow transmission at the current transmission rate. The subcarriers available in each case are those which permit the transmission of data at a given transmission rate. Subcarriers that would require a significantly higher reception sensitivity or a significantly higher reception power level even at the lowest transmission rate are not used for transmission. After the end of such a process cycle, ie as soon as the data of the highest class has also been transmitted, a new cycle follows, starting again with the lowest class. The lowest class for the new cycle can, however, correspond to a different maximum transmission rate, e.g. For example, because the quality of the transmission channels has generally improved or deteriorated, or because other transmitters / receivers register transmission requirements. There are various ways of determining which data is transmitted during a process cycle. Each transmitting ~ / receiving device can e.g. B. a certain maximum amount of data can be assigned, which can be transmitted during a cycle, or during a process cycle, all data are transmitted, which are in a buffer at a certain time.

To simplify matters, in step h of the method all assigned data can be transmitted at the second transmission rate (or the corresponding higher current transmission rate). As a result, in a given time slot, all subcarriers used have the same modulation format. Again, the modulation format can be selected differently in different time slots. If the noise levels in different transmission channels differ only slightly, the simplification results in at most a small loss of bandwidth or transmission capacity.

The data can be classified into two or more service classes. The data of the first service class are given priority, ie the data of the second service class are transferred later or at best at the same time as the data of the first service class. This "quality of service" concept allows selected transmitters / receivers or selected data to be given priority, e.g. B. because they perform important tax tasks or transmit important tax data. It is z. B. also possible to offer transmission capacity of different service classes at different prices. The integration of different service classes into the method according to the invention takes place, for. B. in a first process cycle all data of the first service class and only in a later process cycle all data of the second service class are transmitted.

An arrangement for the transmission of data in a network comprises a superordinate transceiver and a plurality of subordinate transceivers. Said transceivers include means for connection to a wired transmission medium, e.g. B. to a power supply network. The subordinate transceivers comprise means for transmitting data to the superordinate transceivers on a plurality of subcarriers, these subcarriers being provided by FDMA (Frequency Division Multiple Access) and preferably FDM (Frequency Division Multiplex), being mutually orthogonal and further divided into time slots by TDMA (Time Division Multiple Access).

Further advantageous embodiments and combinations of features of the invention result from the following detailed description and the totality of the patent claims.

Brief description of the drawings

The drawings used to explain the exemplary embodiment show:

1 shows a schematic representation of an arrangement according to the invention for connection to a power supply network;

Fig. 2 is a schematic representation of a higher-level according to the invention

Transmitting / receiving device;

3 shows a flow chart of a method according to the invention; 4 shows a possible assignment of data to time slots generated by a method according to the invention;

5 shows a possible assignment of data to time slots, generated by another method according to the invention.

In principle, the same parts are provided with the same reference symbols in the figures.

Ways of Carrying Out the Invention

Figure 1 is a schematic representation of an arrangement according to the invention for connection to a power supply network. A transformer station 1 is shown, in which the voltage of a medium-voltage line 3 is transformed with a transformer 2 into a low voltage, which can be tapped from a busbar 4. A low-voltage line 5 is connected to the busbar 4 and supplies, for example, a single building in a district with electricity. The low-voltage line 5 has an inductance 6.

A master station 7 is connected to the low-voltage line 5 and forms a communication network with the slaves 8.1, 8.2, 8.3, which are also connected to this low-voltage line 5. The communication network can also include a plurality of further slaves (not shown).

The field of application of the method and device according to the invention is not limited to data transmission in power supply networks. It includes other wired transmission media, such as conventional data cables or a broadband cable.

FIG. 2 shows a schematic representation of a higher-level transceiver (master station) according to the invention. The master station 7 is connected to the low-voltage line 5 for the transmission of data. The connection is made via a Interface 10, which in particular performs the electrical adaptation to the power supply network and couples signals in or out. On the other hand, master station 7 is on. a terminal 9, such as. B. a workstation, a personal computer, a terminal, a server or a router. A second interface 11 ensures the adaptation of the signal and the transmission protocol to the terminal 9. From the first interface 10, signals decoupled from the low-voltage line 5 reach the receiver 12, where they are demodulated and decoded if necessary, so that the received data is transmitted a buffer 13 and the second interface 11 can be transmitted to the terminal 9.

In the other direction, the terminal 9 transmits data via the interface 11 and further via a second buffer 14 to the transmitter 15. The data is possibly coded by this and a carrier is modulated onto it. The signals are coupled into the low-voltage line 5 via the interface 10.

The master station 7 also includes many other elements, in particular for controlling the data transmission in the respective network. Of these, only those are described below which are used to assign time slots according to the invention for data transmission from the subordinate transceivers to the master station 7 (uplink).

The signals received by the subordinate transceivers (slaves), partially processed, e.g. B. already demodulated, signals or measured parameters such. B. a signal-to-noise ratio are transmitted from the receiver 12 to a classifier 16. Based on these signals and / or parameters, this determines the transmission quality of the transmission channel between the sending /! Slave and the master station. The classifier 16 is also transmitted by the receiver 12 control data sent by the slaves. These contain in particular information about a technically possible maximum transmission rate, which is supported by the slave.

Based on the data received from all subordinate transceivers, the classifier 16 periodically classifies the transceivers by assigning each one Assigns class. Each class corresponds to a specific, maximum possible transmission rate. The class assignments are held in memory 17 until the next reclassification.

The receiver 12 stores the number of data that each slave wants to transmit to the master in the next process cycle in a further memory 18. The class assignments stored in the memory 17 and the amounts of data stored in the memory 18 are transmitted to a controller 19 which, based on this, carries out the assignment of the data to be transmitted to uplink time slots, as described below. The assignment for each slave, i. H. the information about the time slots and on which subcarrier a particular transceiver can send to the master station 7 is transmitted via the transmitter 15 to the individual slaves.

The master station according to the invention can also be structured differently, e.g. B. by additional, separate from the receiver, circuits for determining the signal-to-noise ratio or other parameters are provided. It is also possible to integrate different functions in one switching element. For classifying the slaves or for assigning the uplink time slots, means can also be used which are present in a conventional superordinate transceiver anyway, so that a customary master station can carry out the method according to the invention after reprogramming.

So that the transmission power of the slaves can be regulated in such a way that all signals of all transmitting slaves arrive at the master at approximately the same power level, the master station can include means for controlling the transmission power of the slaves. These can function on the basis of a closed control loop (closed-loop power control), as is known from the prior art. Optionally, the master can only transmit the transmission power and the desired power level for receiving the signals from the masters to the slaves. Based on the power level with which a signal sent from the master to a specific slave is received, this slave determines the attenuation of the transmission channel used and sets its transmission power so that the master receives the signals of the slave with the desired power level.

FIG. 3 shows a flow chart of a method according to the invention. It is assumed that data with transmission rates of 2, 4, 6, 8 or 10 bits / symbol (at a constant symbol rate) can be transmitted. Before the transmission, all subordinate transmitting / receiving devices (slaves) are classified (step 20). Depending on the maximum transmission rate from slave to master, each slave is assigned to a class K ₂ , K ₄ , K ₆ , K ₈ or K ₁₀ . Class K ₆ slaves support e.g. B. a maximum transmission rate on the subcarriers of the best quality of 6 bits / symbol etc.

The current transmission rate R is then set to the lowest supported transmission rate of 2 bits / symbol (step 21). In the class that corresponds to the current transmission rate, the (remaining) amount of data is now determined, i. i.e. how much data should still be transferred from the slaves to the master during the current process cycle (step 22). Either each slave reports the amount of data that is due for transmission to the master, or the communication is organized centrally by the master, so that the information about the amount of data to be transferred is known to the master anyway.

The data to be transmitted in a process cycle can be selected in various ways. Either all reported data are transmitted or each slave is entitled to a certain maximum quota of data that can be transmitted in one cycle. The remaining amount of data in a class K _R , hereinafter referred to as N _D (K _R ), is the number of data to be transmitted less data already transmitted in this class. No data has yet been transferred at the start of the process cycle; N _D (K _R ) for the class with the lowest transmission rate thus corresponds to the total number of data in this class.

Next, it is checked whether data should be transmitted in this class at all (step 23). If this is not the case (N _D (K _R ) = 0), the current transmission rate is changed to the next higher transmission rate is set (step 14) and the process is continued with this new transmission rate.

If data of the current class are waiting for transmission, the number N _s required time slots is now determined (step 25). For this purpose, the data to be transmitted are distributed to all available subcarriers, with at most the current transmission rate being used for transmission on all subcarriers.

The data pending for transmission in the current class K _R are now assigned to all subcarriers during the N _s time slots (step 26). As a rule, the N _s time slots are not completely filled by this. Either not all subcarriers are available for the transmission of data of the current class K _R or the number of data to be transmitted does not fit exactly into a whole number of time slots. The latter is avoided by using additional criteria when selecting the data to be transmitted, which means that the total amount of data to be transmitted always corresponds to a multiple of the capacity of a time slot. Especially if the transmission capacity of a time slot is large (e.g. because many subcarriers are available or because the transmission rate is large) or if the data quantities to be transmitted by individual slaves are small, it is difficult to completely fill up the N _s time slots.

Therefore, the remaining sub-carriers, which have been assigned no data during one or more of the N _s time slots, with data from slaves in the class K _R. filled with the next higher transmission rate R '(R -R + 2 in the example shown) (steps 27, 28). It should be noted that this additional data is also transmitted at the current transmission rate R at the most.

Under certain circumstances, all data of the next higher class K _R. are transmitted in the prohibiting subcarriers or subcarriers also stand for the transmission of data of the next higher class K _R. not available due to poor quality. If, after the first filling with the data of the next higher class, there is still capacity, ie subcarriers with non-filled time slots remain (Step 29), the filling process is repeated with data of an even higher class (in the example R + 4). The counter R 'is increased again by 2 bits / symbol in the example described (step 30). In extreme cases, the filling procedure can be continued with data of even higher classes.

Once all Ünterträgem associated data for all N time slots _s ,, _s time slots to the current transmission rate R by the participating slave (Class K _R and possibly higher grades) this data is transmitted to the master (step 31) during the N.

The steps 22-31 described are now carried out with the next pending transmission rate R + 2. A process cycle ends when all the data in all classes has been transferred. As a rule, a process cycle is followed directly by a next cycle in which a quantity of data is again determined for transmission.

The method according to the invention can also be implemented differently. So z. B. the filling of the remaining time slots when the selection of the data to be sent in a process cycle is carried out so that the data of each class completely fill a number of time slots and if data of each class K _R can be transmitted on all subcarriers. The method will also generally be carried out in a "pipeline", ie, while data is being transmitted at the rate R (step 31), the data of the next higher class are already being assigned (steps 22-29). It is also possible to assign data of an entire process cycle, to communicate this to the subordinate transceivers and only then to transmit all assigned data.

FIG. 4 shows a possible assignment of data to time slots and serves to explain the method according to the invention further. For this purpose, it is assumed that four subcarriers 33, 34, 35, 36 are available for the transmission of data from the slaves to the master with transmission rates of 2, 4, 6, 8 or 10 bits / symbol. For the sake of simplicity, the number of subcarriers is constant during the process cycle under consideration. In the method shown, which leads to the assignment shown in FIG. 4, the same modulation format, ie the same transmission rate, is used in a given time slot on all subcarriers used.

The subcarriers are each in time slots 33.1-33. n, ..., 36.1-36. n divided, with all time slots having the same duration and the subcarriers being synchronized with one another in such a way that the time slots start and end at the same time: The time slots 33.1, 34.1, 35.1 and 36.1 on the subcarriers 33, 34, 35 and 36 are therefore simultaneous.

The following data quantities (in an arbitrary unit) are available for transmission for the slaves of a certain class in the example shown:

with a maximum transfer rate of 2 bit / symbol (class K ₂ ): 18 with a maximum transfer rate of 4 bit / symbol (class K ₄ ): 78 with a maximum transfer rate of 6 bit / symbol (class K ₆ ): 56 with a maximum transmission rate of 8 bit / symbol (class K ₈ ): 48 with a maximum transmission rate of 10 bit / symbol (class. K ₁₀ ): 16.

It is further assumed that 2 data / 2 data units can be transmitted in a time slot with the lowest transmission rate (corresponding to 4 data units with 4 bit / symbol etc.). It should be noted that the numbers 2, 4, 6, 8, 10 shown in the figure, relating to the time slots of the individual subcarriers, do not indicate the transmission rate or the amount of data transmitted in a time slot, but rather the class in which the in this time slot is classified as a slave transmitting on this subcarrier.

For the transmission of the data in the deepest class K ₂ , 3 time slots are required, which form a first transmission section 37.1. However, the capacity of these 3 time slots on 4 subcarriers is 24 data units with the corresponding transmission rate of 2 bit / symbol. The remaining 6 data units can thus be allocated to the slaves of the next higher class K and can already be transmitted in the first transmission section 37.1, albeit with the lower transmission rate of 2 bits / symbol. For the transmission of the 6 data units of the next higher class K ₄ , 3 time slots of one subcarrier are claimed, that is, one subcarrier during the entire transmission section 37.1. Because the transmitting transceivers of class K _{4 could} actually work with the next higher transmission rate, this subcarrier can e.g. B. a frequency band can be assigned, the transmission quality for the other transceivers, in the lower class K ₂ , would not be sufficient. Therefore, the assignment shown is advantageous, in which first subcarriers are "filled" with data of the current class for the entire duration of the transmission section, so that u. U. other subcarriers can be assigned data of a higher class for the entire duration of the section.

72 data units remain from the amount of data in the next higher class K ₄ . For the transmission of the same with 4 bits / symbol in the transmission section 26.2, 5 time slots are required. Again there is a capacity of 8 data units. This is "filled" with data of the next higher class K ₆ .

Accordingly, 48 units of the data volume of this class K ₆ remain, which can be transmitted during exactly 2 subsequent time slots, which form section 26.3. In this case, no refilling takes place.

Two time slots are required for the transmission of the data volume in the next class K ₈ . There is a capacity of 16 data units (1 time slot each on 2 subcarriers with a transmission rate of 8 bit / symbol). This remaining capacity is just sufficient for the data of the slaves in the highest class with a maximum transfer rate of 10 bits / symbol.

Because there is no data to be transmitted in the highest class, there is no transmission of data with the highest transmission rate of 10 bits / symbol in this process cycle. Rather, the next procedural cycle follows after section 37.4, in which data has been transmitted with 8 bits / symbol. FIG. 5 again shows a possible assignment of data to time slots and serves to further explain another possible embodiment of the method according to the invention. Again, four subcarriers 33, 34, 35, 36 are available for the transmission of data from the slaves to the master with transmission rates of 2, 4, 6, 8 or 10 bits / symbol. The number of subcarriers is also constant during the process cycle under consideration. In contrast to the simplified method already described, in the method now shown, which leads to the assignment shown in FIG. 5, different modulation formats, ie also different transmission rates, can be used in a given time slot on different subcarriers. This method also allows optimal use of the available transmission channels if the different subcarriers have very different noise levels.

The subcarriers are again each in time slots 38.1-38. n, ..., 41.1-41.n, whereby all time slots have the same duration and the subcarriers are synchronized with each other in such a way that the time slots start and end at the same time: time slots 38.1, 39.1, 40.1 and 41.1 on the subcarriers 38, 39, 40 and 41 are therefore simultaneously.

It is assumed that the same amounts of data as in the previous example are available for transmission for the slaves of a certain class. The transmission rates are also defined identically (in the arbitrary unit). For the sake of simplicity, it is assumed that all subcarriers used by a slave have approximately the same attenuation.

In contrast to the first example, the four subcarriers 38, 39, 40, 41 have very different noise levels: the subcarrier 38 shows the lowest noise, with the subcarriers 39, 40 and 41 the noise level increases in this order. It is assumed that the data of a certain class K _R can be transmitted on the respective subcarriers with a maximum transmission rate according to the following table: subcarrier

- * - κτ>

The data in the deepest class K ₂ can only be transmitted on the two subcarriers 38, 39 of the best quality. Five time slots are therefore required, which form a first transmission section 42.1. The remaining fifth time slot on the subcarrier 39 and the time slots of the subcarrier 40 are assigned data from class K ₄ , which can be transmitted to these subcarriers with at least 2 bits / symbol. The time slots of the subcarrier 41 with the poorest quality are assigned data of class K ₆ because the quality of this subcarrier is not sufficient for the transmission of data of a lower class.

In the first transmission section 42.1, all data are transmitted with the lowest transmission rate of 2 bits / symbol.

66 data units remain of the amount of data in the next higher class K ₄ . Subcarriers 38 and 39 with 4 bits / symbol and subcarriers 40 with 2 bits / symbol are available for transmitting them. The capacity per time slot on all available subcarriers is therefore ten data units, so seven time slots are required. The remaining time slots of the subcarriers 40, 41 are filled with data of the next higher class K ₆ , which on the subcarrier 40 with 4 bits / symbol and on the Subcarrier 41 can be transmitted with 2 bits / symbol. In the second transmission section 42.2, data are transmitted in different time slots on different subcarriers with different transmission rates (4 bit / symbol or 2 bit / symbol). Due to the different noise level of the subcarriers and the different attenuation of the channels from the slaves to the master, it is nevertheless possible to adapt the transmission power of the slaves in such a way that the signals from the slaves arrive at the master with approximately the same reception power level during the entire transmission section.

24 units of the data volume of the next higher class K ₆ remain. The capacity per time slot for this data in the next transmission section 42.3 is 18 units (6 + 6 + 4 + 2), so two time slots are still required. In the remaining time slots of the subcarriers 40, 41, data of class K ₈ are transmitted at transmission rates of 4 bit / symbol or 2 bit / symbol.

In class K _{8 there} are still 36 units left. Two time slots are sufficient to transmit the same. The capacity of these two time slots remaining after the assignment of the class K ₈ data is also sufficient for the transmission of the class K ₁₀ data with transmission rates of ό bit / symbol on the subcarrier 40 or 2 bit / symbol on the subcarrier 41.

Because there is no data to be transmitted in the highest class, there is no transmission of data with the highest transmission rate of 10 bits / symbol in this process cycle. Rather, after Section 42.4, in which data with 8 bits / symbol and less have been transmitted, the next process cycle follows.

If different subcarriers used by the same slave have a very different damping, a refined class system can be used, different classifications being carried out for different subcarriers or groups of subcarriers. The method according to the invention also works in such modified class systems without any significant change. The assignment of the data to the individual time slots and Unterträgerπ can also be done differently, z. B. by assigning the data to time slots for the entire cycle by optimizing. A standard minimization algorithm such as steepest descent or a brute force solution can be selected.

In summary, it can be stated that a method is created which enables a high spectral efficiency and a good utilization of the bandwidth in the uplink without increased linearity requirements on the receiver.

Claims

claims

1. A method for transmitting data in a network with a higher-level transceiver (7) and several subordinate transceiver (8.1-8.3), said transceiver (7, 8.1-8.3) for transmitting the Data to a line-bound transmission medium (5), in particular a

Power supply network, are connected and data are transmitted from the subordinate transceivers (8.1-8.3) to the superordinate transceiver (7) on several subcarriers, which by FDMA (Frequency Division Multiple Access) and preferably FDM (Frequency Division Multiplex) are provided and are orthogonal to each other, these subcarriers by TDMA

(Time Division Multiple Access) are further divided into time ^slots'.

2. The method according to claim 1, characterized in that the transmission power levels of the subordinate transceivers (8.1-8.3) are set so that in a given time slot all data transmitted by the subordinate transceivers at the superordinate transceiver (7) can be received at approximately the same received power levels.

3. The method according to claim 1 or 2, characterized in that different transmission rates, which result from different bit loading, are used in different time slots.

4. The method according to claim 3, characterized in that the subordinate transceiver (8.1-8.3) according to a maximum possible transmission rate of a transmission channel of a subordinate transceiver (8.1-8.3) to the superordinate transceiver (7) in two or more classes are classified, and that data from transceivers of the same class are transmitted in the same time slots if possible.

5. The method according to claim 4, characterized in that the data of the transceiver, whose transmission channels correspond to the class with a first, lowest maximum transmission rate, are transmitted in one or more first time slots with the first, lowest maximum transmission rate , whereupon data of the transceivers whose transmission channels correspond to a class with a second, higher maximum transmission rate are transmitted in one or more second time slots with the second, higher maximum transmission rate.

6. The method according to claim 5, characterized by the following steps:

a) determining a first number of time slots which are required for the transmission of a first plurality of data, the first plurality of data corresponding to all data of the transceivers in the class with the first, lowest maximum transmission rate;

b) assignment of the first plurality of data to subcarriers during the first number of time slots;

c) assigning data of the transceivers in classes with higher maximum transmission rates than the first transmission rate to remaining subcarriers during the first number of time slots until all available subcarriers are assigned during the first number of time slots;

d) transmission of all assigned data at the first transmission rate during the first number of time slots;

e) determining a second number of time slots which are required for the transmission of a second plurality of data, the second plurality of data corresponding to all data of the transceivers in the class with the second, next higher transmission rate; f) assigning the second plurality of data to subcarriers during the second number of time slots;

g) assigning data of the transceivers in classes with higher maximum transmission rates than the second transmission rate to remaining sub-carriers during the second number of time slots until all available sub-carriers are assigned during the second number of time slots;

h) transmission of all assigned data at the second transmission rate or a lower transmission rate during the second number of time slots;

i) repetition of steps e-h with the remaining data of the transceiver in the classes with the higher transmission rates compared to the second transmission rate.

7. The method according to claim 6, characterized in that in step h all assigned data are transmitted at the second transmission rate, so that in a given time slot all subcarriers used have the same modulation format.

8. The method according to any one of claims 4 to 7, characterized in that data are classified into at least two service classes and that all data of a second service class are transmitted later or simultaneously as data of a first service class.

9. Arrangement for the transmission of data in a network, comprising a superordinate transceiver (7) and several subordinate transceivers (8.1-8.3), said transceivers (7, 8.1-8.3) for transmitting the data, means for connecting to a line-bound transmission medium (5), in particular a power supply network, and the subordinate transceivers (8.1-8.3) comprise means for transmitting data to the superordinate transceiver (7) on several subcarriers comprehensive sen, wherein the subcarriers are provided by FDMA (Frequency Division Multiple Access) and preferably FDM (Frequency Division Multiplex) and are orthogonal to one another and wherein these subcarriers are further divided into time slots by TDMA (Time Division Multiple Access).

10. The arrangement according to claim 9, characterized by means for adjusting the power level of the subordinate transceiver (8.1-8.3), so that in a given time slot all data transmitted by the subordinate transceiver at the superordinate transceiver (7 ) are received at approximately the same power level.

1 1. Arrangement according to claim 9 or 10, characterized in that different transmission rates, which result from different bit loading, are used in different time slots.

12. The arrangement according to claim 1 1, characterized by means (16) for classifying the subordinate transmitting / receiving devices (8.1-8.3) into two or more classes according to a maximum possible transmission rate of a transmission channel of a subordinate transmitter / receiving device (8.1-8.3) higher-level transceiver (7) and by means of transmission of data from transceivers of the same class, if possible in the same time slots.

13. The arrangement according to claim 12, characterized in that the data of the transceivers, whose transmission channels correspond to the class with the lowest maximum transmission rate, are transmitted in one or more first time slots with the lowest maximum transmission rate, whereupon data from the transceivers whose transmission channels correspond to a class with a higher maximum transmission rate are transmitted in one or more second time slots with the higher maximum transmission rate.

14. Arrangement according to claim 13, characterized by

a) means for determining a first number of time slots which are required for the transmission of a first plurality of data, the first plurality of data corresponding to all data of the transceivers in the class with the lowest maximum transmission rate;

b) means for assigning the first plurality of data to subcarriers during the first number of time slots;

c) Means for assigning data of the transmitting / receiving devices in a class with a next higher maximum transmission rate to remaining subcarriers during the first number of time slots until all subcarriers are assigned data during the first number of time slots.

d) means for transmitting all assigned data with the lowest maximum transmission rate during the first number of time slots.

15. Arrangement according to one of claims 12-14, characterized by means for classifying data in at least two service classes and by means for later or simultaneous transmission of all data of a second service class according to data of a first service class.