WO2021053137A1

WO2021053137A1 - Method for marking a broadcast signal, method for identifying a marking signal, corresponding computer program products and corresponding devices

Info

Publication number: WO2021053137A1
Application number: PCT/EP2020/076098
Authority: WO
Inventors: Frédéric PIROT
Original assignee: Enensys Technologies
Priority date: 2019-09-19
Filing date: 2020-09-18
Publication date: 2021-03-25
Also published as: FR3101220B1; FR3101220A1

Abstract

The invention relates to a method for marking a broadcast signal. Such a marking method comprises: - obtaining (E200) the broadcast signal; and - obtaining (E210) a marking signal. The marking method furthermore comprises marking (E230) the broadcast signal by adding the marking signal to the broadcast signal. The marking signal is dependent on at least one Zadoff-Chu sequence.

Description

DESCRIPTION

TITLE: Method for marking a broadcast signal, method for identifying a marking signal, computer program products and corresponding devices

Field of the invention

The field of the invention is that of the broadcasting of information in a television broadcasting network. More precisely, the invention relates to a method for marking a signal broadcast in such a network.

The invention thus finds numerous applications, notably, but not exclusively, in digital television broadcasting networks including in particular DVB-T / T2 (standing for “Digital Video Broadcasting - Terrestrial”), ISDB-T (standing for “Digital Video Broadcasting - Terrestrial”). "Integrated Services Digital Broadcasting - Terrestrial"), ATSC-3 (from the English "Advanced Television Systems Committee"), DAB (from the English "Digital Audio Broadcasting"), DMB (from the English "Digital Media Broadcasting" ), or DTTB (standing for “Digital Television Terrestrial Broadcasting”).

Prior art and its drawbacks

In the remainder of this document, attention is paid more particularly to describing an existing problem in the field of information dissemination in a digital television broadcasting network of the SFN type (for "Single Frequency Network" in English), to which have been described. confronted by the inventors of the present patent application. The invention is of course not limited to this particular field of application, but is of interest for the marking of any type of broadcast signal of which it is advantageous to be able to determine the origin (eg acoustic, optical transmission, etc.) .

The impulse response of a radio propagation channel, or CIR (for "Channel Impulse Response" in English), is the response of the propagation channel in question to an impulse waveform emitted by a transmitter at a given time. Such a pulse is received directly and / or in the form of replicas by a receiver after a propagation delay in the channel in question.

For example, if the propagation channel is of the single-path type (eg when the transmitter and the receiver are in direct view, i.e. without there being any reflection on a third object along the path between the transmitter and receiver), a pulse emitted by the transmitter generates a single pulse received by the receiver. If, on the other hand, the propagation channel is of the multi-path type (eg via reflections on surrounding objects), an emitted pulse will generate a train of pulses received by the receiver. Moreover, in an SFN network where several transmitters transmit the same data at the same time and on the same frequency, the same phenomenon is observed at the level of a receiver placed in a coverage zone of each of the transmitters in question.

Finally, multipaths can take place within an SFN network, which then gives rise to the observation of even more complex CIRs as illustrated in Figs. 1a and FIG. lb.

For example, the terminal 100 (for example an SFN network monitoring equipment) of FIG. receives it three times, via three separate paths 120a, 120b and 120c, a pulse emitted simultaneously at time t0 by the first 110a and second 110b transmitters.

In this way, the CIR obtained (FIG. Lb) at the level of the terminal 100 presents: at the instant t1 a first peak 150a corresponding to the first path 120a in direct view between the second transmitter 110b and the terminal 100. The first path 120a is in fact the shortest among the three paths 120a, 120b and 120c; at the instant t2 a second peak 150b corresponding to the second path 120b between the second emitter 110b and the terminal 100 via a reflection on the building 130; and at time t3 a third peak 150c corresponding to the third path 120c in direct view between the first emitter 110a and the terminal 100, the third path 120c being the longest among the three paths 120a, 120b and 120c.

However, the first 110a and second 110b transmitters transmitting the same signals, the terminal 100 does not have the means to know whether the different peaks 150a, 150b and 150c that it observes result from the different reflections encountered by a signal emitted by a single one. transmitter, or if several transmitters have transmitted the same signal.

In this context, methods for marking the signals transmitted in such an SFN network have been proposed. More particularly, by marking the broadcast signal in a differentiated manner as a function of the transmitter, a receiver is able to determine the origin of the signal received by knowing the marking associated with each transmitter that it receives.

For example, the DAB protocol provides for such a possibility of marking the signal transmitted via the so-called T11 information (standing for “Transmitter Identification Information”). More particularly, the Tll consists in sending a code on a few sub-carriers during the null symbol at the start of the frame. This system makes it possible to detect the presence of Tll in a signal received according to the DAB protocol. However, when receiving echoes from several transmitters, as in the case of an SFN, it is impossible to reliably associate a given echo with a Tll. Indeed, the power and the power spectral density of the signal are insufficient to accurately time stamp a T11 detected in a noisy environment. Alternatively, the DVB-T2 protocol provides for the marking of the signal broadcast via a specific signal which takes place within the data frames instead of part of the useful data. This technique is therefore intrusive because it decreases the total throughput available for the useful data.

The ATSC3 protocol provides for a marking process at the modulator level. More particularly, a specific code by direct sequence spread spectrum is sent during the preamble of the frame, synchronously. This method has the advantage of being independent of the data symbols transmitted, and therefore does not reduce the total bit rate available for the payload data.

However, this method suffers from several limitations:

The power of the marking signal, relative to the power of the useful signal, must be strong enough to guarantee detection;

Since the signature is generated via a spread spectrum by direct sequence, its spectral occupancy is equal to the sampling band, which, in a constrained environment where the spectral mask is severely specified, can cause certain problems.

There is thus a need for a marking technique which does not have the drawbacks of the prior art.

More particularly, there is a need for a marking technique exhibiting fine temporal resolution. This is in order to precisely associate a given peak of the impulse response of the channel with a transmitter of the broadcasting network, e.g. when a different marking signal is used for each transmitter of the network.

There is a need for such a technique not to degrade the throughput of the payload data.

There is also a need for the detection of the marking to be robust without degrading the demodulation performance of the payload data of the broadcast signal.

Disclosure of the invention

In one embodiment of the invention, there is provided a method for marking a broadcast signal. Such a marking method comprises: obtaining the broadcast signal; and obtaining a marking signal.

The tagging method further comprises tagging the broadcast signal by adding the tagging signal to the broadcast signal. The labeling signal is a function of at least one Zaddoff-Chu sequence. Thus, the invention proposes a new and inventive solution for adding a marking signal (also called a signature or watermark signal) to a broadcast signal, eg as intended to be broadcast in a television broadcasting network (eg of the SFN type. , for “Single Frequency Network” in English).

More particularly, the intercorrelation properties of the Zaddoff-Chu sequences make it possible, for example, on reception, to identify the signals received from different transmitters of a television broadcasting network, eg when a different marking signal is used for each transmitter in the network.

Likewise, the autocorrelation properties of the Zaddoff-Chu sequences make it possible, on reception, to have fine temporal resolution in the identification of the marking signal. This makes it possible to precisely associate a given peak of the impulse response of the channel with a transmitter of the broadcasting network, e.g. when a different marking signal is used for each transmitter of the network.

Moreover, the marking of the broadcast signal via the addition of the marking signal makes it possible not to use useful parts of the broadcast signal initially intended to convey other data in order to convey the marking signal. In this way, the broadcast payload data rate is not degraded.

According to one embodiment, the marking signal is a signal of the OFDM type (standing for “Orthogonal Frequency Division Multiplex”) comprising Nfft subcarriers. Said at least one Zaddoff-Chu sequence is represented by a series of Nzc samples. Obtaining the labeling signal comprises: mapping the Nzc samples representative of said at least one Zaddoff-Chu sequence to Nfft samples, with Nzc less than or equal to Nfft, the mapping delivering a vector of Nfft mapped samples intended to modulate the sub-carriers. The mapping preserves the ordering of the Nzc samples representative of said at least one Zaddoff-Chu sequence within the vector of Nfft mapped samples; and a Fourier transform of the vector of Nfft mapped samples delivering a vector of Nfft transformed samples.

The labeling signal is a function of the Nfft samples transformed.

Thus, the application of the Zaddoff-Chu sequence in the frequency domain makes it possible to precisely control the spectrum of the labeling signal obtained.

According to one embodiment, when Nzc is strictly less than Nfft, the mapping comprises a zeroing of the Nfft minus Nzc samples having a value different from the Nzc samples representative of said at least one Zaddoff-Chu sequence among the Nfft mapped samples.

Thus, the number Nzc of samples of the Zaddoff-Chu sequence can be chosen to be different from the number of points of the Fourier transform in order to optimize the performance and the computational load of the processing operations. For example, it can be chosen a prime number Nzc of samples of the Zaddoff-Chu sequence in order to obtain optimal performance in terms of autocorrelation and intercorrelation, and a power of two for the number Nfft of points of. the Fourier transform in order to implement the latter according to an efficient fast Fourier transform algorithm.

According to one embodiment, during the implementation of the mapping, the Nzc mapped samples having a value equal to the Nzc samples of said at least one Zaddoff-Chu sequence occupy a central position in the vector of Nfft mapped samples.

According to one embodiment, obtaining the marking signal comprises applying an apodization window to the vector of Nfft mapped samples delivering a vector of Nfft apodized samples. The Fourier transformation is applied to the vector of Nfft apodized samples, instead of the vector of Nfft mapped samples, delivering the vector of Nfft transformed samples.

Thus, the side lobes of the spectrum of the labeling signal obtained are optimized.

According to one embodiment, obtaining the labeling signal comprises a duplication of the Nfft samples transformed a plurality of times delivering a plurality of replicas of the Nfft samples transformed. The labeling signal is a function of a concatenation of the replicas of the transformed Nfft samples.

Thus, a redundancy of the transmitted data is obtained, allowing a processing gain on reception. In this way, good detection of the marking signal is obtained even though the ratio of the broadcast signal as such to noise is maintained at a level allowing negligible degradation of the demodulation performance.

According to one embodiment, the marking method comprises an adjustment of an amplitude of the marking signal relative to an amplitude of the diffusion signal before the addition of the marking signal to the diffusion signal.

Thus, the signal-to-noise ratio obtained during the processing on reception (the marking signal being seen as noise with respect to the broadcast signal) is adjustable.

According to one embodiment, the addition of the marking signal to the diffusion signal is carried out synchronously, at least one given sample of the marking signal being added with a sample of predetermined rank of the diffusion signal. Thus, the marking signal can for example be synchronized with the data of the broadcast signal used for the synchronization of the frame. In this way, the marking signal can be temporally associated in a fine manner with the different paths of the propagation channel in which the diffusion signal is broadcast.

According to one embodiment, the marking method comprises broadcasting the marked broadcast signal by a transmitter of an SFN television broadcasting network.

In one embodiment of the invention, there is provided a method for identifying at least one marking signal added to a corresponding broadcast signal, said at least one marking signal being a function of at least one sequence of Correspondent Zaddoff-Chu. Such an identification method comprises, for a received signal comprising said at least one broadcast signal: obtaining a signal portion representative of said at least one marking signal from the received signal; obtaining at least one expected marking signal; a correlation performed between, on the one hand, the signal portion and, on the other hand, said at least one expected marking signal delivering at least one corresponding correlation result; and identification of said at least one marking signal from at least said at least one correlation result and at least one predetermined identification rule.

Said at least one expected labeling signal is a function of at least one expected Zaddoff-Chu sequence.

For example, a broadcast signal given in the received signal was broadcast by a corresponding transmitter in a broadcast network. In the case where several transmitters are in play, and when the device receiving the received signal is geographically located in a coverage area of several of the transmitters in question, the received signal results from the superposition of the broadcast signals transmitted by the corresponding transmitters.

In this case, and as discussed above in relation to the labeling method, the cross-correlation properties of the Zaddoff-Chu sequences allow for example an identification of the broadcast signals received from different transmitters of the broadcast network, eg when a different marking signal is used by a network transmitter.

Furthermore, the autocorrelation properties of the Zaddoff-Chu sequences make it possible, on reception, to have fine temporal resolution in the identification of the marking signal. This makes it possible to precisely associate a given peak in the impulse response of the channel to a transmitter of the broadcasting network, eg when a different marking signal is used for each transmitter of the network.

According to one embodiment, said at least one marking signal is an OFDM type signal comprising Nfft subcarriers and obtaining a signal portion representative of said at least one marking signal implements an extraction of a block of Nfft samples of the received signal. The signal portion representative of said at least one marking signal is a function of the block of Nfft samples.

According to one embodiment, said at least one marking signal is an OFDM type signal comprising Nfft subcarriers and obtaining a signal portion representative of said at least one marking signal implements: an extraction of N consecutive blocks of Nfft samples of the received signal, with N an integer greater than or equal to 2; and a coherent summation of samples of the same rank from each of the N consecutive blocks delivering a block of Nfft summed samples.

The signal portion representative of said at least one marking signal is a function of the block of Nfft summed samples.

When the broadcast signal (s) has been labeled by implementing a duplication of the Nfft samples of the labeling signal, redundancy is obtained between each consecutive block of Nfft samples. The useful data being for their part random from one block of Nfft samples to another, the coherent summation thus leads to a gain in processing on the marking signal (s) making it possible to bring out the latter with respect to the diffusion signal. .

According to one embodiment, obtaining said at least one expected marking signal implements a step of obtaining a marking signal (according to any one of the aforementioned embodiments in relation to the marking method) based on said at least one expected Zaddoff-Chu sequence.

According to one embodiment, said at least one marking signal is an OFDM type signal comprising Nfft subcarriers and obtaining a signal portion representative of said at least one marking signal implements: an extraction of a block of Nfft samples of the received signal; and a Fourier transform of the block of Nfft samples of the received signal delivering a corresponding block of Nfft transformed samples.

The signal portion representative of said at least one labeling signal is a function of the block of Nfft transformed samples. According to one embodiment, obtaining a signal portion representative of said at least one marking signal implements a Fourier transformation of the block of Nfft summed samples delivering a block of Nfft transformed samples. The signal portion representative of said at least one labeling signal is a function of the block of Nfft transformed samples.

According to one embodiment, said at least one marking signal is an OFDM type signal comprising Nfft subcarriers and obtaining a signal portion representative of said at least one marking signal implements: an extraction of N consecutive blocks of Nfft samples of the received signal, with N an integer greater than or equal to 2; a Fourier transformation of each of the N consecutive blocks of Nfft samples of the received signal delivering N corresponding blocks of Nfft transformed samples; and a coherent summation of samples of the same rank from each of the N blocks of Nfft transformed samples delivering a block of Nfft summed samples.

Thus, the coherent summation here carried out in the frequency domain also leads to a gain in processing on the marking signal (s) making it possible to bring out the latter with respect to the broadcast signal as discussed above in the case of a consistent summation in the time domain.

According to one embodiment, the marking signal is a signal of the OFDM type comprising Nfft subcarriers and said at least one expected Zaddoff-Chu sequence is represented by a series of Nzc samples. Obtaining the labeling signal comprises a mapping of the Nzc samples representative of said at least one expected Zaddoff-Chu sequence to Nfft samples, with Nzc less than or equal to Nfft, the mapping delivering a vector of Nfft mapped samples intended to modulate the sub-carriers. The mapping preserves the ordering of the Nzc samples representative of said at least one expected Zaddoff-Chu sequence within the vector of Nfft mapped samples. The expected labeling signal is a function of the Nfft mapped samples.

According to one embodiment, when Nzc is strictly less than Nfft, the mapping comprises a zeroing of the Nfft minus Nzc samples having a value different from the Nzc samples representative of said at least one Zaddoff-Chu sequence expected among the Nfft mapped samples . According to one embodiment, during the implementation of the mapping, the Nzc mapped samples having a value equal to the Nzc samples of said at least one expected Zaddoff-Chu sequence occupy a central position in the vector of Nfft mapped samples.

According to one embodiment, obtaining the expected labeling signal comprises applying an apodization window to the vector of Nfft mapped samples delivering a vector of Nfft apodized samples. The expected labeling signal is a function of the vector of Nfft apodized samples.

According to one embodiment, the correlation is carried out between, on the one hand, NZc samples of the signal portion and, on the other hand, the Nzc samples representative of said at least one expected Zaddoff-Chu sequence delivering at least one corresponding correlation result. Thus, the computational load is reduced.

According to one embodiment, the identification of said at least one marking signal comprises a comparison of said at least one correlation result with a predetermined threshold. At least one expected marking signal is considered to be a candidate marking signal when the corresponding correlation result is greater than the predetermined threshold.

According to one embodiment, when for a plurality of expected marking signals, a plurality of candidate marking signals is obtained during the implementation of the correlation and identification steps, the predetermined identification rule belongs to the group comprising : no marking signal is considered to be identified; each candidate marking signal is considered to identify a marking signal in the received signal; and a candidate tagging signal whose corresponding correlation result is extreme among the correlation results associated with the candidate tagging signals is considered to identify a tagging signal in the received signal.

According to one embodiment, obtaining a signal portion is carried out a plurality of times delivering a corresponding plurality of different signal portions representative of said at least one marking signal. Correlation and identification are implemented for each signal portion of the plurality of different signal portions providing a plurality of candidate marking signals. The identification further implements a statistic based on the plurality of candidate marking signals delivering a plausible marking signal. The predetermined identification rule corresponds to the fact that the probable marking signal is considered to identify a marking signal in the received signal. The invention also relates to at least one computer program, comprising program code instructions for implementing at least one method as described above, according to any one of its various embodiments, when it is run on a computer.

In another embodiment of the invention, there is provided a device for marking a broadcast signal. Such a device comprises a reprogrammable computing machine or a dedicated computing machine configured to: obtain the broadcast signal; and obtaining a marking signal.

The reprogrammable computing machine or the dedicated computing machine is also configured to mark the broadcast signal by adding the marking signal to the broadcast signal. The labeling signal is a function of at least one Zaddoff-Chu sequence.

Such a marking device is in particular suitable for implementing the marking method according to the invention (according to any one of the various aforementioned embodiments). Thus, the characteristics and advantages of this device are the same as those of the marking method described above. Therefore, they are not detailed further.

In another embodiment of the invention, there is provided a device for identifying at least one marking signal added to a corresponding broadcast signal. Said at least one labeling signal is a function of at least one corresponding Zaddoff-Chu sequence. Such a device comprises a reprogrammable computing machine or a dedicated computing machine configured, for a received signal comprising said at least one broadcast signal, to: obtain a signal portion representative of said at least one marking signal from the received signal ; obtain at least one expected marking signal; performing a correlation between, on the one hand, the signal portion and, on the other hand, said at least one expected marking signal delivering at least one corresponding correlation result; and identifying said at least one marking signal from at least said at least one correlation result and at least one predetermined identification rule.

Such an identification device is in particular capable of implementing the identification method according to the invention (according to any one of the various aforementioned embodiments). Thus, the characteristics and advantages of this device are the same as those of the identification method described above. Therefore, they are not detailed further.

List of Figures

Other aims, characteristics and advantages of the invention will emerge more clearly on reading the following description, given by way of simple illustrative example, and not limiting, in relation to the figures, among which:

[Fig. 1a], already discussed above in the section “Prior art and its drawbacks”, illustrates a terminal receiving a signal via three paths in an SFN network;

[Fig. lb], already discussed above in the section “Prior art and its drawbacks”, illustrates the CIR corresponding to the propagation channel as seen by the terminal of FIG. the via the signals it receives;

[Fig. 2] illustrates the steps of a method for marking a broadcast signal according to one embodiment of the invention;

[Fig. 3a] illustrates an exemplary implementation of the mapping step of the method of FIG. 2; [Fig. 3b] illustrates an exemplary implementation of the step of applying an apodization window of the method of FIG. 2;

[Fig. 3c] illustrates an example of implementation of the marking step of the method of FIG. 2; [Fig. 4] illustrates the functional blocks of a device for marking a broadcast signal according to one embodiment of the invention;

[Fig. 5a] illustrates the steps of a method for identifying at least one marking signal according to a first embodiment of the invention;

[Fig. 5b] illustrates the steps of the method for identifying at least one marking signal according to a second embodiment of the invention; and

[Fig. 6] illustrates the functional blocks of a device for identifying at least one marking signal according to one embodiment of the invention.

Detailed description of embodiments of the invention

The general principle of the invention is based on the marking of a broadcast signal (eg a signal of the OFDM type such as broadcast for example in a digital television broadcasting network, eg of the SFN type according to a standard such as DVB-T / T2 , ISDB-T, ATSC-3, DMB, DTTB, etc.) by adding a marking signal. Such an addition makes it possible, among other things, not to use, in order to convey the marking signal, parts of the broadcast signal initially intended for the payload data. In this way, the broadcast payload data rate is not degraded. Furthermore, the labeling signal is a function of at least one Zaddoff-Chu sequence. Thus, the intercorrelation properties of the Zaddoff-Chu sequences allow, for example on reception, an identification of the signals received from different transmitters of a television broadcasting network, eg when a different marking signal is used for each. network transmitter. Likewise, the autocorrelation properties of the Zaddoff-Chu sequences make it possible to have a fine temporal resolution in the identification of the labeling signal. This makes it possible to precisely associate a given peak of the impulse response of the channel with a transmitter of the television broadcasting network, eg when a different marking signal is used for each transmitter of the network.

We now present, in relation to FIG. 2 the steps of a method for marking a broadcast signal according to one embodiment. Certain steps of the process of FIG. 2 are furthermore discussed in more detail in relation to FIGS. 3a, Fig. 3b and Fig. 3c.

During a step E200, a broadcast signal is obtained. For example, the I and Q samples of the in-phase and quadrature branches representing the real and imaginary parts of the complex envelope of the broadcast signal are supplied to the marking device 400 (as described below in relation to FIG. 4). In other implementations, the device 400 receives the data to be broadcast and itself generates the broadcast signal (e.g. the aforementioned I and Q samples). During a step E210, a marking signal is obtained. More particularly, the labeling signal is a function of a Zaddoff-Chu sequence.

For example, the Zadoff-Chu sequence is represented by Nzc samples generated using the following formula:

With: ne [0: N _zc [an integer representing the index of the sequence sample; ue [0: N _zc [an integer representing the sequence number; and <7 th Z.

Thus, the modulus of each sample is unitary. Moreover, the Fourier transform of a Zadoff-Chu sequence is another Zadoff-Chu sequence. The cross-correlation of two different sequences (ie two sequences of different number u) of Nzc samples is constant and equal to 1 / ^ jNzc when Nzc is a prime number. Finally, the autocorrelation of a Zadoff-Chu sequence is non-zero only for a cyclic shift equal to n * Nzc, with n an integer. For example, when the broadcast signal is a signal such as broadcast in a digital television broadcasting network (eg of the SFN type according to a standard such as DVB-T / T2, ISDB-T, ATSC-3, DMB, DTTB, etc. ), different Zadoff-Chu sequences are associated with different transmitters in order to be able to identify on reception the origin of a received marking signal. For this purpose, one can for example: a) Choose different Zadoff-Chu sequences for each transmitter by adjusting the value of u if the value of Nzc is sufficiently large; b) Choose Zadoff-Chu sequences with the same root u for each emitter but by adding a predetermined circular permutation of the sequence in question (with the interesting property, in this case, of perfectly orthogonal sequences); or c) Make a combination of a) and b).

When the labeling signal is an OFDM type signal comprising Nfft subcarriers, obtaining the labeling signal implements a step E210a of mapping the Nzc samples representative of the Zaddoff-Chu sequence to Nfft samples, with lower Nzc or equal to Nfft. As illustrated in Fig. 3a, the mapping delivers a vector of Nfft mapped samples intended to modulate the subcarriers. Each (complex) sample of the Zadoff-chu sequence is mapped to a different carrier. The mapping is here carried out on contiguous sub-carriers and preserves the ordering of the Nzc samples representative of the Zaddoff-Chu sequence within the vector of Nfft mapped samples.

In the embodiment illustrated in FIG. 3a, Nzc is strictly less than Nfft and the mapping comprises a zeroing of the Nfft minus Nzc samples having a value different from the Nzc samples representative of the Zaddoff-Chu sequence among the Nfft mapped samples.

Thus, the number Nzc of samples of the Zaddoff-Chu sequence can be chosen to be different from the number of points of the Fourier transform implemented to generate the OFDM symbols. This makes it possible to optimize the performances and the computational load of the treatments. For example, it can be chosen a prime number Nzc of samples of the Zaddoff-Chu sequence in order to obtain optimal performance in terms of autocorrelation and intercorrelation, and a power of two for the number Nfft of points of. the Fourier transform in order to implement the latter according to an efficient fast Fourier transform algorithm.

Furthermore, in the embodiment of FIG. 3a, when performing the mapping, the Nzc mapped samples having a value equal to the Nzc samples of the Zaddoff-Chu sequence occupy a central position in the vector of Nfft mapped samples. Returning to Fig. 2, obtaining the marking signal comprises a step E210b of applying an apodization window to the vector of Nfft mapped samples delivering a vector of Nfft apodized samples. The implementation of this embodiment is illustrated for example in FIG. 3b in which the minimum or maximum frequency subcarriers are gradually attenuated by the application of the apodization window.

Obtaining the labeling signal also comprises a step E210c of Fourier transformation of the vector of Nfft apodized samples delivering a vector of Nfft transformed samples. The labeling signal is thus a function of the Nfft samples transformed. The application of the Zaddoff-Chu sequence in the frequency domain makes it possible to precisely control the spectrum of the labeling signal obtained.

In the embodiment of FIG. 2, the step E210b of applying the apodization window makes it possible to control the secondary lobes of the spectrum of the marking signal obtained in order to comply with the pollution constraints of the adjacent channels. In other embodiments, such a window is not applied and step E210b is not implemented. In this case, during step E210c, the Fourier transformation is applied to the vector of Nfft mapped samples, instead of the vector of Nfft apodized samples, delivering the vector of Nfft transformed samples.

Returning to Fig. 2, obtaining the labeling signal comprises a step E210d of duplicating the Nfft samples transformed a plurality of times delivering a plurality of replicas of the Nfft samples transformed. The labeling signal is thus a function of a concatenation of the replicas of the transformed Nfft samples.

Thus, a redundancy of the transmitted data is obtained, allowing a gain in processing on reception via a coherent summation of the samples of the same rank of each of the replicas of the Nfft samples transformed as detailed below in relation to step E500b of the method of Fig. 5a and Fig. 5b.

In other embodiments not illustrated, such a duplication step E210d is not implemented and no processing gain is obtained on reception.

In certain embodiments, such a marking signal is calculated beforehand and stored in memory. In this case, step E210 corresponds for example to a reading of the marking signal in the memory in question. In other embodiments, the samples of the marking signal are calculated on the fly as needed.

In other embodiments, several Zaddoff-Chu sequences are used to obtain the same labeling signal, eg in order to robustify the labeling obtained. For example, step E210a, optionally in conjunction with step E210b, is applied a plurality of times with a different Zaddoff-Chu streak each time. A plurality of blocks of Nfft samples is thus obtained to which a Fourier transformation is applied. The labeling signal is thus a function of a concatenation of the blocks of Nfft transformed samples. A processing gain on reception is again obtained on reception via a coherent summation of the samples of the same rank of each of the blocks of Nfft samples transformed by taking into account the Zaddoff-Chu sequences present in each block.

Returning to Fig. 2, the marking method comprises a step E220 of adjusting the amplitude of the marking signal relative to the amplitude of the broadcast signal.

In other embodiments not illustrated, such an adjustment step E220 is not implemented and the broadcast signal is marked with the marking signal as obtained during the implementation of the step E210.

Returning to Fig. 2, the marking method comprises a step E230 of marking the broadcast signal by adding the marking signal to the broadcast signal.

The marking of the broadcast signal via the addition of the marking signal to the broadcast signal makes it possible not to use, in order to convey the marking signal, parts of the broadcast signal initially intended for the payload data. In this way, the broadcast payload data rate is not degraded.

Moreover, in the case where the duplication step E210d is implemented, the different replicas of the transformed Nfft samples are added to different useful data. Thus, during the coherent summation of the samples of the same rank of each of the replicas of the Nfft transformed samples (as detailed below in relation to step E500b of the method of FIGS. 5a and FIG. 5b), the useful data are summed up. destructively and not coherently in amplitude and phase as do the samples of the same rank of each of the replicas. Thus, the processing gain is obtained. In this way, good detection of the marking signal is obtained even though the ratio between the power of the marking signal relative to the power of the broadcast signal is maintained at a level allowing demodulation of the payload data without apparent degradation.

In the embodiment illustrated in FIG. 3c, the addition of the marking signal to the broadcast signal is carried out synchronously. For example, the samples of the labeling signal are added to the corresponding samples of the diffusion signal so that a given sample of the labeling signal is added with a sample of predetermined rank of the diffusion signal. In the case described above in relation to step E200 in which the device 400 receives the I and Q samples of the complex envelope of the broadcast signal, the device 400 receives, for example, synchronization information, a synchronization signal, etc., indicating the instant of insertion of the samples of the marking signal with the corresponding samples of the broadcast signal. For example, the marking signal is synchronized with the data of the broadcast signal used for the synchronization of the frame. In the example of FIG. 3c, such data used for the channel estimation are located at the start of the frame N + 1 following the frame N. In this way, the marking signal is temporally associated in a fine manner with the different paths of the propagation channel in which the broadcast signal is broadcast.

Fig. 4 shows an example of the structure of a device 400 for marking a broadcast signal according to one embodiment of the invention. More particularly, such a device 400 allows the implementation of the method of FIG. 2. The device 400 comprises a random access memory 403 (for example a RAM memory), a processing unit 402 equipped for example with a processor, and controlled by a computer program stored in a read only memory 401 (for example a memory. ROM or hard drive). On initialization, the code instructions of the computer program are for example loaded into the random access memory 403 before being executed by the processor of the processing unit 402.

This Fig. 4 illustrates only one particular way, among several possible, of producing the means included in the device 400, so that it carries out certain steps of the method detailed above, in relation to FIG. 2 (in any of the different embodiments). Indeed, these steps can be performed either on a reprogrammable computing machine (a PC computer, a DSP processor or a microcontroller) executing a program comprising a sequence of instructions, or on a dedicated computing machine (for example a set of logic gates such as an FPGA or ASIC, or any other hardware module). In the case where the means included in the device 400 are produced with a reprogrammable computing machine, the corresponding program (that is to say the sequence of instructions) can be stored in a removable storage medium (such as by for example a floppy disk, a CD-ROM or a DVD-ROM) or not, this storage medium being partially or totally readable by a computer or a processor.

In some embodiments, the device 400 is located at the level of a transmitter of a television broadcasting network, e.g. of the SFN type according to a standard such as DVB-T / T2, ISDB-T, ATSC-3, DMB,

DTTB, etc. For example, device 400 is included in the first 110a or the second 110b transmitter.

In some embodiments, the device 400 is included in a modulator located at the level of such a transmitter. We now present, in relation to FIG. 5a the steps of a method for identifying at least one marking signal according to a first embodiment of the invention. In this first embodiment, the correlation implemented during step E510 described below is implemented in the time domain.

Such a method is implemented by an identification device 600 (as described below in relation to FIG. 6) processing a received signal comprising one or more broadcast signals. The received signal is for example delivered to the device 600 by a receiver synchronized in time and in frequency on the radiofrequency channel used by the television broadcasting network.

More particularly, the broadcast signal (s) are marked by implementing the marking method described above in relation to FIG. 2 (according to any one of the aforementioned embodiments). Thus, the broadcast signal or signals each convey a corresponding marking signal.

For example, the identification device 600 is included in the terminal 100 and the received signal comprises the broadcast signals broadcast by the first 110a and second 110b transmitters. Each of the broadcast signals in question carries a corresponding marking signal in order to be able to identify the transmitter at the origin of a peak in a CIR analyzed at the level of the terminal 100.

Returning to Fig. 5a, during a step E500, a signal portion representative of the marking signal or signals is obtained from the received signal.

To do this, when the marking signal or signals are of the OFDM type comprising Nfft subcarriers, during a step E500a, N consecutive blocks of Nfft samples are extracted from the received signal, with N an integer greater than or equal to 2.

During a step E500b, a coherent summation of samples of the same rank of each of the N consecutive blocks delivers a block of Nfft summed samples. The signal portion representative of the marking signal (s) is a function of the block of Nfft samples summed.

Thus, when the broadcast signal (s) have been marked by implementing a duplication of the Nfft samples of the marking signal (cf. step E210d described above in relation to FIG. 2), redundancy is obtained between each block. consecutive Nfft samples. The useful data conveyed by the diffusion signal (s) being for their part random from one block of Nfft samples to another, the coherent summation thus leads to a gain in processing on the marking signal (s) making it possible to bring out the latter. with respect to the broadcast signal (s).

In other embodiments not illustrated, the coherent summation step E500b is not implemented. In this case, a single block of Nfft samples is extracted from the received signal. The part signal representative of the marking signal or signals is a function of the block of Nfft samples in question.

Returning to Fig. 5a, one or more expected marking signals are obtained. In this first embodiment in which the correlation implemented during step E510 described below is implemented in the time domain, the expected marking signal (s) are obtained by implementing step E210 (according to any of the aforementioned embodiments) described above in relation to FIG. 2. However, in the context of the identification method, step E210 (according to any one of the aforementioned embodiments) is implemented on the basis of the expected Zaddoff-Chu sequence (s) instead of and instead. of the Zaddoff-Chu sequence (s) used in the labeling process.

For example, the same embodiment of step E210 is implemented both in order to obtain the marking signal at the level of the marking method described above in relation to FIG. 2 and the identification method described herein. In this way, the overall performance is optimized. In other variants, different embodiments of step E210 are implemented in the marking method and in the identification method. For example, the step E210b of applying an apodization window is implemented only at the level of the marking method in order to control the secondary lobes of the marking signal, and not at the level of the identification method in order to reduce the computational load.

During a step E510, a correlation is carried out between, on the one hand, the signal portion and, on the other hand, the expected marking signal (s) delivering one or more corresponding correlation results. As described above, in this first embodiment of the identification method, the correlation implemented during step E510 is implemented in the time domain.

During a step E520, the marking signal (s) conveyed by the broadcast signal (s) received are identified on the basis of at least: the correlation result (s); and one or more predetermined identification rules.

More particularly, in certain embodiments, step E520 comprises a comparison of the correlation result (s) with a predetermined threshold. An expected marking signal is considered to be a candidate marking signal when the corresponding correlation result is greater than the corresponding predetermined threshold.

In some embodiments, when for a plurality of expected tagging signals, a plurality of candidate tagging signals is obtained when performing the steps Correlation E510 and identification E520, the predetermined identification rule belongs to the group comprising: no marking signal is considered to be identified (several candidate marking signals being considered as an eliminating factor, eg when a single signal of marking was actually expected in practice even if several were tested); each candidate marking signal is considered to identify a marking signal in the received signal; and a candidate tagging signal whose corresponding correlation result is extreme among the correlation results associated with the candidate tagging signals is considered to identify a tagging signal in the received signal.

In certain embodiments, the step E500 of obtaining a signal portion is implemented a plurality of times delivering a corresponding plurality of different signal portions representative of the marking signal or signals. The correlation steps E510 and identification E520 are implemented for each signal portion of the plurality of different signal portions delivering a plurality of candidate marking signals. The identification step E520 also implements a statistic based on the plurality of candidate marking signals delivering a plausible marking signal. The predetermined identification rule corresponds to the fact that the probable marking signal is considered to identify a marking signal in the received signal.

We now present, in relation to FIG. 5b the steps of the method for identifying at least one marking signal according to a second embodiment of the invention. Unlike the embodiment described above in relation to FIG. 5a, in this second embodiment, the correlation implemented during step E510 ′ described below is implemented in the frequency domain.

Thus, step E500 ′ of obtaining a signal portion representative of the marking signal (s) comprises steps E500a and E500b of FIG. 5a (according to any one of the aforementioned embodiments). In addition, step E500 ′ comprises a step E500c of Fourier transformation of the block of Nfft summed samples delivered by the implementation of step E500b. The implementation of step E500c delivers a block of Nfft transformed samples. The signal portion representative of the marking signal (s) is a function of the block of Nfft samples transformed.

In other embodiments of step E500 ′, the coherent summation is performed after the Fourier transformation. In other words, the Fourier transformation step E500c is performed before coherent summation step E500b. In these embodiments, step E500 'thus implements: step E500a of extracting N consecutive blocks of Nfft samples of the received signal, with N an integer greater than or equal to 2; step E500c of Fourier transformation of each of the N consecutive blocks of Nfft samples of the received signal delivering N corresponding blocks of Nfft transformed samples; and step E500b of coherent summation of samples of the same rank of each of the N blocks of Nfft transformed samples delivering a block of Nfft summed samples.

The signal portion representative of the marking signal (s) is a function of the block of Nfft samples summed.

The coherent summation carried out in the frequency domain also leads to a gain in processing on the marking signal (s) making it possible to bring out the latter with respect to the broadcast signal according to the same principle as discussed above in the case of a summation. consistent in the time domain.

In other embodiments not illustrated, the coherent summation step E500b is not implemented. In this case, a single block of Nfft samples is extracted from the received signal. A Fourier transform of the block of Nfft samples extracted from the received signal delivers a corresponding block of Nfft transformed samples. The signal portion representative of the marking signal (s) depends on the block of Nfft transformed samples in question.

Returning to Fig. 5b, one or more expected marking signals are obtained by implementing a step E210 ′.

More particularly, the step E210 ′ comprises the mapping step E210a (according to any one of the aforementioned embodiments) described above in relation to FIG. 2, but with an implementation of the step in question on the basis of the expected Zaddoff-Chu sequence (s) instead of the Zaddoff-Chu sequence (s) used in the labeling process. Thus, the expected marking signal (s) implemented in the embodiment of FIG. 5b are frequency signals.

In certain embodiments, step E210 ′ further comprises step E210b (according to any one of the aforementioned embodiments) of applying an apodization window as described above in relation to the Fig. 2. Thus, the performance of the correlation implemented during step E510 ′ is improved in the case where the marking signal to be identified has been obtained by implementing the marking method according to an embodiment also comprising the apodization step E210b. During a step E510 ′, a correlation is carried out between, on the one hand, the signal portion and, on the other hand, the expected marking signal (s) delivering one or more corresponding correlation results. Thus, in this second embodiment of the identification method, the correlation implemented during step E510 ′ is implemented in the frequency domain.

In embodiments, when Nzc is strictly less than Nfft, the correlation is carried out between, on the one hand, NZc samples of the signal portion and, on the other hand, the Nzc samples representative of the Zaddoff sequence (s). -Chu expected delivering at least one corresponding correlation result. Thus, the computational load of the correlation is reduced.

Returning to Fig. 5b, the identification method according to the second embodiment illustrated in this figure comprises the identification step E520 (according to any one of the aforementioned embodiments) described above in relation to FIG. 5a.

Fig. 6 shows an example of the structure of a device 600 for identifying at least one marking signal according to one embodiment of the invention. More particularly, such a device 600 allows the implementation of the method of FIGS. 5a and Fig. 5b. The device 600 comprises a random access memory 603 (for example a RAM memory), a processing unit 602 equipped for example with a processor, and controlled by a computer program stored in a read only memory 601 (for example a ROM memory or a hard disc). On initialization, the code instructions of the computer program are for example loaded into the random access memory 603 before being executed by the processor of the processing unit 602.

This Fig. 6 illustrates only one particular way, among several possible, of producing the means included in the device 600, so that it carries out certain steps of the method detailed above, in relation to FIGS. 5a and Fig. 5b (in any of the different embodiments). Indeed, these steps can be performed either on a reprogrammable computing machine (a PC computer, a DSP processor or a microcontroller) executing a program comprising a sequence of instructions, or on a dedicated computing machine (for example a set of logic gates such as an FPGA or ASIC, or any other hardware module). In the case where the means included in the device 600 are produced with a reprogrammable computing machine, the corresponding program (that is to say the sequence of instructions) can be stored in a removable storage medium (such as by for example a floppy disk, a CD-ROM or a DVD-ROM) or not, this storage medium being partially or totally readable by a computer or a processor.

In some embodiments, the device 600 is located at a terminal, for example the terminal 100. In some embodiments, the device 600 is included in equipment for monitoring a digital television broadcasting network, eg of the SFN type according to a standard such as DVB-T / T2, ISDB-T, ATSC-3, DMB, DTTB , etc.

Claims

1. A method of marking a broadcast signal, said method comprising:

- obtaining (E200) of said broadcast signal; and

- obtaining (E210) a marking signal, characterized in that it further comprises a marking (E230) of said broadcast signal by adding said marking signal to said broadcast signal, said marking signal constituting screw noise - with respect to the diffusion signal, and in that said labeling signal is a function of at least one Zaddoff-Chu sequence represented by at least one sample (Nzc), at least one sample representing the Zaddoff-Chu sequence being mapped to at least one subcarrier of the tagging signal.

2. The method of claim 1 wherein said marking signal is an OFDM type signal (for "Orthogonal Frequency Division Multiplex" in English) comprising Nfft subcarriers, wherein said at least one Zaddoff-Chu sequence is represented by a series of Nzc samples, and wherein said obtaining said labeling signal comprises:

a mapping (E210a) of the Nzc samples representative of said at least one Zaddoff-Chu sequence to Nfft samples, with Nzc less than or equal to Nfft, said mapping delivering a vector of Nfft mapped samples intended to modulate said sub-carriers, said mapping preserving the ordering of said Nzc samples representative of said at least one Zaddoff-Chu sequence within said vector of Nfft mapped samples; and

a Fourier transformation (E210c) of said vector of Nfft mapped samples delivering a vector of Nfft transformed samples, said marking signal being a function of said Nfft transformed samples.

3. The method of claim 2 wherein, when Nzc is strictly less than Nfft, said mapping comprises a zeroing of Nfft minus Nzc samples having a value different from said Nzc samples representative of said at least one Zaddoff-Chu sequence among said. Nfft mapped samples.

4. The method of claim 2 or 3, wherein said obtaining said marking signal comprises an application (E210b) of an apodization window to said vector of Nfft mapped samples delivering a vector of Nfft apodized samples, said Fourier transformation being applied to said vector of Nfft apodized samples, instead of said vector of Nfft mapped samples, delivering said vector of Nfft transformed samples.

5. Method according to any one of claims 2 to 4 wherein said obtaining said labeling signal comprises a duplication (E210d) of said Nfft samples transformed a plurality of times delivering a plurality of replicas of said Nfft transformed samples, said labeling signal being function of a concatenation of said replicas of said Nfft transformed samples.

6. Method according to any one of claims 1 to 5 comprising an adjustment (E220) of an amplitude of said marking signal relative to an amplitude of said broadcast signal before said addition to said broadcast signal.

7. Method according to any one of claims 1 to 6 wherein said addition of said marking signal to said broadcast signal is carried out synchronously, at least one given sample of said marking signal being added with a sample of predetermined rank of said signal. broadcast.

8. A method of identifying at least one marking signal added to a corresponding broadcast signal, said at least one marking signal being a function of at least one corresponding Zaddoff-Chu sequence, characterized in that it comprises , for a received signal comprising said at least one broadcast signal:

- obtaining (E500, E500 ') of a signal portion representative of said at least one marking signal from said received signal;

- obtaining (E210, E210 ') of at least one expected marking signal;

a correlation (E510, E510 ') carried out between, on the one hand, said signal portion and, on the other hand, said at least one expected marking signal delivering at least one corresponding correlation result; and

- an identification (E520) of said at least one marking signal from at least said at least one correlation result and at least one predetermined identification rule, and in that said at least one expected marking signal is a function of 'at least one expected Zaddoff-Chu streak.

9. The method of claim 8 wherein said at least one marking signal is an OFDM type signal (for "Orthogonal Frequency Division Multiplex" in English) comprising Nfft subcarriers, and wherein said obtaining a signal portion representative of said at least one marking signal implements: an extraction (E500a) of N consecutive blocks of Nfft samples of said received signal, with N an integer greater than or equal to 2; and

a coherent summation (E500b) of samples of the same rank of each of said N consecutive blocks delivering a block of Nfft summed samples, said signal portion representative of said at least one marking signal being a function of said block of Nfft summed samples.

10. The method of claim 8 or 9 wherein said obtaining said at least one expected marking signal implements a step of obtaining (E210) a marking signal according to any one of claims 2 to 4 on the. base of said at least one expected Zaddoff-Chu sequence.

11. The method of claim 8 wherein said at least one marking signal is an OFDM type signal (for "Orthogonal Frequency Division Multiplex" in English) comprising Nfft subcarriers, and wherein said obtaining a signal portion representative of said at least one marking signal implements:

- an extraction (E500a) of a block of Nfft samples of said received signal; and

a Fourier transformation (E500c) of said block of Nfft samples of said received signal delivering a corresponding block of Nfft transformed samples, said signal portion representative of said at least one marking signal being a function of said block of Nfft transformed samples.

12. The method of claim 9 wherein said obtaining a signal portion representative of said at least one marking signal implements a Fourier transformation (E500c) of said block of Nfft summed samples delivering a block of Nfft transformed samples, said block. signal portion representative of said at least one marking signal being a function of said block of Nfft transformed samples.

13. The method of claim 8 wherein said at least one marking signal is a signal of the OFDM type (for “Orthogonal Frequency Division Multiplex”) comprising Nfft subcarriers, and wherein said obtaining of a signal portion representative of said at least one marking signal implements:

an extraction (E500a) of N consecutive blocks of Nfft samples of said received signal, with N an integer greater than or equal to 2;

- a Fourier transformation (E500c) of each of said N consecutive blocks of Nfft samples of said received signal delivering N corresponding blocks of Nfft transformed samples; and

a coherent summation (E500b) of samples of the same rank of each of said N blocks of Nfft transformed samples delivering a block of Nfft summed samples, said signal portion representative of said at least one marking signal being a function of said block of Nfft summed samples .

14. The method of any one of claims 11 to 13, wherein said at least one expected Zaddoff-Chu sequence is represented by a sequence of Nzc samples, and wherein said obtaining said labeling signal comprises mapping the Nzc samples. representative of said at least one expected Zaddoff-Chu sequence towards Nfft samples, with Nzc less than or equal to Nfft, said mapping delivering a vector of Nfft mapped samples intended to modulate said subcarriers, said mapping preserving the ordering of said Nzc samples representative of said at least one expected Zaddoff-Chu sequence within said vector of Nfft mapped samples, said expected labeling signal being a function of said Nfft mapped samples.

15. The method of claim 14 wherein, when Nzc is strictly less than Nfft, said mapping comprises a zeroing of Nfft minus Nzc samples having a value different from said Nzc samples representative of said at least one expected Zaddoff-Chu sequence among said Nfft mapped samples.

16. The method of claim 14 or 15, wherein said obtaining said expected marking signal comprises applying an apodization window to said vector of Nfft mapped samples delivering a vector of Nfft apodized samples, said expected marking signal being a function of of said vector of Nfft apodized samples.

17. A method according to any one of claims 8 to 16 wherein said identification of said at least one marking signal comprises a comparison of said at least one correlation result with a predetermined threshold, at least one expected marking signal being considered as one. candidate marking signal when the corresponding correlation result is greater than said predetermined threshold.

18. The method of claim 17 wherein, when for a plurality of expected marking signals, a plurality of candidate marking signals is obtained during the implementation of said correlation and identification steps, said predetermined identification rule belongs to the group comprising:

- no marking signal is considered to be identified; each candidate marking signal is considered to identify a marking signal in said received signal; and

a candidate marking signal whose corresponding correlation result is extreme among the correlation results associated with said candidate marking signals is considered to identify a marking signal in said received signal.

19. The method of claim 17 wherein said obtaining of a signal portion is carried out a plurality of times delivering a corresponding plurality of different signal portions representative of said at least one marking signal, wherein said correlation and said identification are implemented for each signal portion of said plurality of different signal portions delivering a plurality of candidate marking signals, said identification further implementing a statistic from said plurality of candidate marking signals delivering a marking signal probable, and wherein said predetermined identification rule corresponds to the fact that the probable marking signal is considered to identify a marking signal in said received signal.

A computer program product comprising program code instructions for implementing a method according to any one of claims 1 to 19, when said program is executed on a computer.

Device (400) for marking a broadcast signal, said device comprising a reprogrammable computing machine (402) or a dedicated computing machine configured to:

- obtaining said broadcast signal; and

obtaining a marking signal, characterized in that said reprogrammable computing machine or said dedicated computing machine is also configured to mark said broadcast signal by adding said marking signal to said broadcast signal, said marking signal constituting screw noise - With respect to the diffusion signal, and in that said labeling signal is a function of at least one Zaddoff-Chu sequence represented by at least one sample (Nzc), at least one sample representing the Zaddoff-Chu sequence being mapped to at least one subcarrier of the tagging signal.

20. Device (600) for identifying at least one marking signal added to a corresponding broadcast signal, said at least one marking signal being a function of at least one. corresponding Zaddoff-Chu sequence, characterized in that it comprises a reprogrammable computing machine (602) or a dedicated computing machine configured, for a received signal comprising said at least one broadcast signal, in order to: - obtain a portion of signal representative of said at least one marking signal from said received signal;

- obtain at least one expected marking signal;

performing a correlation between, on the one hand, said signal portion and, on the other hand, said at least one expected marking signal delivering at least one corresponding correlation result; and

- identify said at least one marking signal from at least said at least one correlation result and at least one predetermined identification rule, and in that said at least one expected marking signal is a function of at least one Zaddoff-Chu streak expected.