EP1845642B1 - Method and means for receiving a DVB-T broadcast signal - Google Patents

Description

The present invention relates to a method for receiving a DVB-T radio reception signal according to the features of the preamble of patent claim 1.

DVB-T transmission methods and associated data are well known (ETSI EN 300 744). The source coding is done with MPEG 2 or MPEG 4. Due to this data compression several programs can be transmitted in one channel. The Kanalbandbeite lies between 6 MHz and 8 MHz, depending on the frequency grid. The entire data of a multiplex is subjected to channel coding after channel coding. This is done by dividing the bandwidth into several individual carriers by means of a special modulation method, the COFDM method. Each of these individual carriers is then modulated again using one of the three modulation methods defined for DVB-T, QPSK, 16 QAM or 64 QAM.
The problem with the transmission of the DVB-T signals is the multipath propagation, which arises functionally in the emission due to multiple reflection. Multipath propagation is unavoidable and technically conditioned in terrestrial networks. This results in the receiver corresponding echoes or interference, which are to be compensated by the receiver or in the receiver.
To guarantee a certain degree of error protection, DVB-T incorporates multiple error protection. There is external error protection (block code), with 16 bytes of redundancy being added to each data packet consisting of 188 bytes.
Furthermore, a bit interleaving takes place against burst errors. In addition, an internal error protection is provided by means of a convolutional coding is made. This allows the robustness against disturbances to be set, as is necessary for the various intended reception paths, for example for mobile reception, portable reception or stationary reception of DVB-T signals. Furthermore, there is still an "inner interleaving" to increase the robustness of the transmission. However, it should be noted that if the COFDM single carrier mode is selected properly, the transmitter network and the receiver must be matched to each other. The data transmission then takes place in each case in packets with the symbol duration T _S. The usable symbol duration T _U is in a fixed ratio to the distance between the individual carriers to each other. The entire symbol duration T _S is divided into a so-called guard interval T _G and a useful range T _U. In principle, the guard interval has the task of giving the receiver a certain settling time until the useful data in the useful range of the signal can be evaluated and these can be decoded.

Out EP 1 276 289 A2 and EP 1 331 783 A2 In each case a method for receiving COFDM signals is known. With these methods, the evaluation window is pushed to the guard interval of the next symbol. If an interference interference is then detected, the evaluation window is pushed back in order to position it in the range of the usable symbol duration. In this case, however, the evaluation window is not advanced in the range of the guard interval of the symbol duration in order to avoid interference or to stop continuously occurring interference.

Out EP 0 827 303 A1 For example, a distributed station system is known for transmitting an FDMA signal. The individual symbols sent in this system have a different temporal positioning.

The object of the present invention is to use the guard interval intelligently for the signal reception and to improve and increase the usable duration of the signal and thus to improve the reception, in particular when used in mobile DVB-T receivers.

This object is achieved with the features of claim 1. Advantageous embodiments of the invention will become apparent from the further description and in particular the dependent claims.

The subject of this invention is concerned with the OFDM technique, more specifically with the intelligent use of guard interval and payload of the transmitted symbol. The methods of channel coding remain unaffected.

As already shown, the useful range is fixed when transmitting DVB-T signals. To evaluate the symbol duration and the useful signals of the useful range, the useful range must be taken from the symbol duration. In conventional receivers, this is done by detecting the beginning of the guard interval and decoding the signals of the useful range when the end of the guard interval is detected from this point in time.

The present invention takes a different approach. The symbol duration T _{S of} a DBV-T signal, which is composed of the guard interval and the useful range, is no longer considered static.
The useful range T _U can be located arbitrarily in the symbol duration T _S , ie also at the beginning of the symbol. T _U is in a fixed relationship to the distance of the subcarrier each other. It is the start of the entire symbol of the DVB-T signal detected .. It is immediately after the start of the recognition of the symbol with the decoding started and it is not waited until the Guard interval is completed. For this purpose, a first window which corresponds to the width of the useful area T _U is evaluated. This first window is adjusted accordingly for each symbol that is received. It is begun, so to speak, even before the complete reception of the guard interval with the decoding of the user data from the useful signal from the useful range. In this case, it is particularly advantageous that, if echoes are present in the symbol duration, they can be corrected. If, for example, there is a disturbance in the rear area of the symbol, in the useful area, the region of the symbol to be evaluated can be pushed forward by skillful displacement of the first window, as it were into the area of the guard interval. It can be assumed that the signals can already be decoded before the useful range. The evaluation of the data of the entire symbol takes place before the actual useful range is present.
This surprisingly simple procedure evaluates the useful signals without waiting for the actual useful range T _U. The evaluation window of the width T _U can be moved freely over the entire symbol duration T _S. Thus, part of the guard interval T _{G is} already used for decoding.

In a further advantageous embodiment of the invention is continuously, temporally offset, shifted the evaluation window over the entire symbol duration and evaluated. It is then created a statistic from which determines an optimal location of the evaluation window for the payload, which is then used for later reception. The basis for the statistics is the bit error rate after the convolutional decoding and the signal form of a symbol in the time domain and / or in the frequency domain.
The evaluation takes place in such a way that the data, which are respectively taken from the evaluation window, are stored and compared with one another. In particular, an error comparison can be made thereby and a stochastic evaluation of the received data can be carried out.

This can be a useful hint for a diversity evaluation and control of the reception. If, for example, in the case of mobile reception disturbances occur periodically over a period of time, a corresponding receiver using the proposed method can prediction or estimate these errors in advance or compensate for these errors in a predictive manner or with a prediction.
The stochastic data can be evaluated in particular with a Kalman algorithm and used for intelligent diversity.
In the described method, each symbol is correspondingly evaluated and the reception window is dynamically shifted over the symbol during reception.
Furthermore, interference is detected or evaluated by means of Viterbi decoding. Interference patterns that exist over time can thus be compensated.

In particular, when using a plurality of antennas, which receive the same signal, can thus be carried out advantageously an interference compensation.
For each antenna, the described method is performed. For this purpose, then several receiving units are provided. The received symbols are then compared with each other via another device. If the knowledge now exists that, for example, every thirtieth symbol is disturbed when received over the first antenna, upon further reception every 30th symbol in this antenna is blanked out and replaced by the data which originate from another antenna and correspond to this symbol , In so far can be done by an intelligent interconnection of the antennas and a corresponding signal processing, a compensation of errors.

Furthermore, the invention will be explained with reference to a concrete embodiment with reference to figures.

FIG 1

den Aufbau eines DVB-T Datensignals,

FIG 2

das Datensignal aus Figur 1 mit der Auswertung nach dem Stand der Technik über ein Zeitfenster,

FIG 3

das erfindungsgemäße Verfahren der Anmelderin, basierend auf dem DVB-T Datensignal gemäß Figur 1,

FIG 4

das modifizierte Verfahren der Anmeldung der Verschiebung des Empfangsfensters und

FIG 5

den schematischen Aufbau einer Empfangseinheit mit mehreren Empfangsantennen.

It shows:

FIG. 1

the construction of a DVB-T data signal,

FIG. 2

the data signal off FIG. 1 with the evaluation according to the prior art over a time window,

FIG. 3

the inventive method of the applicant, based on the DVB-T data signal according to FIG. 1 .

FIG. 4

the modified method of the registration of the shift of the receiving window and

FIG. 5

the schematic structure of a receiving unit with multiple receiving antennas.

In FIG. 1 the structure of a COFDM modulated data signal is shown schematically in the time domain, which is used in the DVB-T transmission. This data signal has a uniform symbol duration T _S. The symbol duration T _S is divided into the guard interval T _G and the useful range T _U. The guard interval T _G is used to wait until all signals (due to the multipath reception) lead to a stationary signal at the receiver after the beginning of the symbol, so that the signal evaluation can be started. A conventional receiver begins to decode the useful data from the useful range T _U after the end of the guard interval T _G.

In FIG. 2 this is shown in more detail. FIG. 2 again shows the COFDM modulated data signal, shown in the time domain, a temporal portion thereof, especially the symbol duration T _S and an evaluation window 1. The evaluation window 1 begins with completion of the guard interval T _G and ends with the end of the useful area T _U. In this evaluation window 1, the decoding of the data of the use area T _{U is} made.

In FIG. 3 now the invention of the applicant is shown. In FIG. 3 is shown that the evaluation window 2, which is the same length as the evaluation window 1 in FIG. 2 , does not begin with completion of the guard interval T _G , but extends into the guard interval T _G. The evaluation window 2, which is used here, is the same length of time as the evaluation window normally used, but it already extends into the guard interval T _G. This is particularly advantageous when there are problems. If, for example, there are disturbances in the useful range T _U at the end of this useful range T _U , then the evaluation window 2 can be correspondingly moved forward so that the disturbances are not in the evaluation window 2.

In a further advantageous embodiment of the invention, it is provided to start with two receivers initially earlier with decoding, the decoding with the first receiver and a corresponding evaluation window 2 already in the guard interval T _G begins, in the following with a second receiver the Decoding to start only with completion of the guard interval T _G and an evaluation window 1 and to analyze only the useful range T _U. By means of this parallel evaluation, any faults which lie outside an evaluation window can be compensated by combining the data from both evaluation windows.

In a further advantageous embodiment of the invention according to FIG. 4 is provided that by an intelligent shift of the evaluation window, there denoted by F1 to FN, the respective evaluation window F1 to FN can be moved freely over the entire symbol duration T _S. This shift occurs at equidistant intervals over time. The evaluation window F1 to FN can thus use the entire symbol duration T _S. After each reception of a symbol duration T _S , evaluation window F1 to FN is shifted in time. This takes place N times until the evaluation window N is present. Each signal, which by means of each of the Receive window F1 to FN is decoded and statistically analyzed. In this case, the optimum reception window F1 to FN can then be determined and this is used later for reception in the case of a static receiver. In this way, it is possible to compensate for disturbances that may be present in the useful range T _U by shifting the evaluation window F1 to FN within the symbol duration T _S.

Furthermore, according to FIG. 5 the schematic structure of a data terminal with multiple antennas A1 to AN at least one receiving units E and a control unit DE shown. The antennas A1 to AN are all connected in parallel with the reception to the receiving unit E. The receiving unit E is in communication with the control unit DE. The control unit DE controls the receiving unit E in such a manner that the receiving unit E uses the individual antennas A1 to AN respectively for reception.
The method according to the invention is now applied as follows, in particular in the case of a mobile receiver. The control unit DE controls the receiving unit E in such a way that over a predetermined period of time in each case with one of the antennas A1 to AN a reception is made. Thus, first of all, the control unit DE starts the receiving unit E in such a way that it receives the incoming signals, for example via the antenna A1 over a predefined period of time. After a predefined time, the control unit DE controls the receiving unit E such that the next antenna, for example the antenna A2 is used, etc., until the last reception via the antenna AN. During reception via each of the antennas A1 to AN, the evaluation window is shifted in accordance with the method and in accordance with the invention over the entire symbol duration TB of the received signal. The received data are statistically analyzed, in particular subjected to an error analysis. This is done as described by means of a Viterbi algorithm or another known stochastic evaluation algorithm. In this way, all antennas A1 to AN, which are connected, evaluated and recorded and evaluated in their receiving characteristics. In particular, in the case of mobile operation, this can also take place during the reception; in that respect, for each of the individual antennas A1 to AN and possibly for a plurality of available receivers, which are then controlled in parallel. The received data is stored and later compared and evaluated. In this way, it is possible over a period of time to obtain a statistical analysis of the quality of the received data signals.

If it now turns out that a periodic disturbance is present in a data signal, which is received in particular via one of the antennas A1 to AN, then this interference can be masked out by the method, since with predictive recognition of the presence of a disturbance, the reception via another Antenna takes place. The data terminal DE then controls the receiver E such that in each case it is predestined from the outset which of the antennas A1 to AN is used to receive the data signal. The receiving unit E thus switches on the basis of the predictively predetermined reception periods, wherein the specification originates from the control unit DE, the individual antennas A1 to AN on reception. This makes it possible to compensate for disturbances which are present periodically.

Claims (7)

1. Method to decode a digitally transferred radio signal which contains data packets, wherein the data packets are symbols and each have a symbol duration (T_S), wherein the symbol duration (T_S) is divided into a guard interval (T_G) and a user data region (T_U) and obtains decoded data by means of a decoding of data transferred in the symbol duration (T_S) within an evaluation window (2), wherein the start of each symbol is detected and the decoding is begun after the recognition of the same, such that the evaluation window (2) extends into the guard interval (T_G) of the symbol duration (T_S) of the symbol, wherein the decoding is repeated multiple times by the evaluation window (2) being shifted over the entire symbol duration (T_S),
   characterised in that
   the data obtained during the decoding, which is repeated multiple times, are stored, compared to one another and an optimal position of the evaluation window (2) within the symbol duration (T_S) is determined by means of an evaluation algorithm, wherein, in the case of the optimal position of the evaluation window (2), disturbances in the user data region (T_U) do not lie in the evaluation window (2) .
2. Method according to claim 1,
   characterised in that
   the shifting of the evaluation window (2) is carried out in equidistant temporal steps.
3. Method according to claim 1,
   characterised in that
   the radio receiving signal is a DVB-T radio signal.
4. Method according to one or more of the preceding claims,
   characterised in that
   the method is applied in mobile receivers.
5. Method according to one or more of the preceding claims,
   characterised in that
   in the case of several receiving antennae (A1 to AN) being present, a stochastic evaluation of the reception data occurs and in the case of periodic disturbances, the reception occurs respectively via the antenna (A1 to AN) which has the best reception according to the stochastic evaluation on the basis of the preceding reception data.
6. Method according to claim 5,
   characterised in that
   the best reception is determined by means of a Viterbi algorithm and/or a probability estimation of the symbol form in the time range and/or frequency range.
7. Device to implement the method according to claim 1 having at least one antenna (A1 to AN), at least one receiving unit, a memory and a control unit (DE), wherein the control unit (DE) monitors a bit error rate after a convolutional decoding and the signal form of a symbol in the time range and/or in the frequency range and determines disturbances or the evaluation thereof by means of Viterbi decoding.

Images