EP3643072A1

EP3643072A1 - Method and device for generating a transport stream, broadcast method and site, and computer program therefor.

Info

Publication number: EP3643072A1
Application number: EP18730321.9A
Authority: EP
Inventors: St phane LETHIMONIER
Original assignee: Enensys Teamcast
Current assignee: Enensys Teamcast
Priority date: 2017-06-21
Filing date: 2018-06-05
Publication date: 2020-04-29
Also published as: WO2018234033A1; US20200280379A1; FR3068194A1

Abstract

The invention relates to a method for generating a transport stream intended to be distributed to a plurality of broadcast sites, comprising the following steps: - modulating (23) a source signal, delivering a modulated signal, - sampling said modulated signal, delivering a stream of complex samples, - framing (24) the stream of complex samples, delivering said transport stream, characterised in that said framing step (24) comprises: - dividing the stream of complex samples into frames, - inserting, in each frame, time information associated with the group of successive complex samples making up the frame for the time synchronisation of said broadcasting sites.

Description

Method and equipment for generating a transport stream, method and broadcast site, and corresponding computer program

1. Field of the invention

The field of the invention is that of the distribution and dissemination of information, in a distribution and digital broadcasting network comprising at least one fixed reference site and a plurality of broadcast sites. More precisely, the invention proposes a solution for the temporal synchronization of the different broadcasting sites.

Here, the term "fixed reference site" means an entity making it possible to format contents and to distribute them in a distribution network. For example, such an entity is a head-end, located in a content creation studio.

The term "broadcast site" means an entity for receiving distributed content in the distribution network, and disseminate particular to individual receivers. For example, such an entity comprises at least one exciter (in English "exciter"). Conventionally, broadcast sites are located on separate geographical sites.

The invention applies more particularly, but not exclusively, to SFN ("Single Frequency Network") networks, regardless of the broadcast standard used:

DVB-T or DVB-T2 (in English "Digital Video Broadcasting - Terrestrial", in French "digital-terrestrial television broadcasting");

T-DMB (in English "Terrestrial Digital Multimedia Broadcasting", in French "terrestrial digital multimedia broadcasting");

ATSC (Advanced Television Systems Committee), including ATSC 3.0;

ISDB-T (Integrated Services Digital Broadcasting - Terrestrial, in French "digital terrestrial integrated services broadcasting");

DAB (in English "Digital Audio Broadcasting", in French "digital sound broadcasting");

etc.

2. Prior Art

In the following, with reference to FIG. 1, an example of a distribution network according to the ATSC 3.0 standard, implementing a fixed reference site, located for example in FIG. a content creation studio, and a plurality of SD1, SD2, SDN broadcast sites located on separate geographical sites.

At the studio level, the source data to be distributed 11 (source 1, source 2, source i, for example data, audio, and / or video services, etc.), provided by one or more service providers, is 12. For example, the source data is compressed and formatted so that at the broadcast sites each physical layer modulator can perform deterministically modulation. This pre-processing step may in particular be implemented in a broadcast gateway of a headend.

In particular, the source data is encapsulated in baseband packets. These baseband packets, with signaling information and synchronization information obtained by taking into account a universal reference time (UTR), such as the GPS signal, are distributed to the SD1 broadcast sites, SD2, SDN via an interface STL (in English "studio-to-transmitter link", in French "link studio - site of diffusion"). Baseband packets, along with signaling and synchronization information, can also be encapsulated in Physical Layer Pipes before distribution.

In particular, STL packets, including baseband packets and signaling information, are carried in STL-TP transport packets over an Ethernet, satellite, and so on link.

The source data is thus managed centrally at the studio level, in order to create a unique transport signal, distributed to all the broadcast sites, allowing in particular a deterministic treatment at the different broadcasting sites. The structure of such a transport signal is particularly detailed in the document "ATSC Candidate Standard: Scheduler / Studio to Transmitter Link" - Document S32-266rl6 - September 30, 2016.

The distribution path 13 between the studio and the broadcasting sites SD1, SD2, SDN may be a satellite link, microwave, optical fiber, etc., possibly with transmission over IP.

Each broadcast site SD1, SD2, SDN receives the transport signal, possibly delayed, and implements a process for resynchronizing the complex samples obtained at the output of the physical layer modulator of each broadcast site, taking into account the time universal reference, and a radiofrequency transmission of re-synchronized complex samples. In particular, each broadcast site SD1, SD2, SDN implements a modulator / exciter 141, 151, 161, delivering a radio frequency signal, and a power amplifier 142, 152, 162 of the radiofrequency signal. Each modulator / exciter 141, 151, 161 comprises in particular:

a physical layer modulator, incorporating time synchronization, delivering a stream of complex samples, and

an exciter, integrating a quadratic modulator (also called I / Q modulator), delivering a radiofrequency signal.

In addition, in the context of terrestrial digital broadcasting, SFN technology is conventionally used to improve the coverage of the territory / of a geographical area and to make up for gray areas linked to disruptive elements in the diffusion (mountain, hills, valleys, tall buildings, ...). It also makes it possible to reduce the number of frequencies used, and consequently to release certain frequency ranges, as well as to optimize the transmission powers.

For example, in the distribution network illustrated in FIG. 1, the transmitters of the broadcasting sites SD1, SD2, SDN transmit a radio frequency signal modulating the same frequency f1.

This SFN technology, very effective, implies a perfect synchronization, in time and frequency, of the sites of diffusion between them. Thus, to synchronize the broadcast sites belonging to the same SFN plate, it is necessary to provide a very precise time and frequency reference to each broadcast site.

3. Presentation of the invention

In one embodiment, the invention provides a method of generating a transport stream for distribution to a plurality of broadcast sites comprising the steps of:

modulation of a source signal, delivering a modulated signal,

modulated signal sampling, delivering complex samples,

framing complex samples, delivering the transport stream,

the framing step comprising:

the distribution of the complex samples in at least one frame,

inserting, in each frame, at least one temporal information for the temporal synchronization of the broadcasting sites.

The invention according to this embodiment thus proposes to move, at the fixed reference site, a part of the modulation processing conventionally implemented at the level of the fixed reference site. Physical layer modulators of each broadcast site.

More specifically, this embodiment of the invention proposes to transport, in the transport stream distributed by the fixed reference site, to the distribution sites, complex samples, also called samples I and Q, or samples 1 / Q.

By deporting part of the modulation processing at the head of the network, the complexity, and therefore the cost, of the exciters of each broadcasting site is reduced.

In particular, in order to ensure a temporal synchronization of the broadcasting sites, at least one temporal information is associated with a complex group of samples, also called a frame. The insertion of such temporal information in a frame makes it possible in particular to ensure SFN operation of the broadcast sites receiving the transport stream. In addition, the insertion of time information into the complex sample frame allows for synchronous distribution of time information and complex samples.

Frequency synchronization of broadcast sites can be provided in a conventional manner, from a reliable reference signal such as GPS.

According to a particular embodiment, the temporal information inserted in a frame corresponds to the instant of transmission of the first complex sample of the frame by the transmitters of the broadcast sites. According to another embodiment, the temporal information inserted in a frame corresponds to the output time of the SFN adapter of the different broadcast sites. In another embodiment, the invention relates to equipment for generating a corresponding transport stream.

The technique of generating a transport stream according to the invention can therefore be implemented in various ways, in particular in hardware form and / or in software form.

Another embodiment of the invention relates to a method for disseminating data, implemented at a broadcasting site, comprising the following steps:

receiving a transport stream, comprising at least one frame carrying complex samples, representative of a source signal, and at least one temporal information for the temporal synchronization of the broadcast site with at least one remote broadcast site;

detecting, in the transport stream, the first complex sample of a frame;

extracting, from the transport stream, complex samples of the frame;

detecting, in the transport stream, the temporal information associated with the frame; emission of the first complex sample of the frame at the instant defined by the temporal information. Such a method, implemented at the level of the broadcasting sites, is in particular intended to receive a transport stream generated by the method of generating a transport stream described above.

In particular, the temporal information carried by the transport stream can be used by the broadcast site to determine the transmission time of the radio frequency signals, or the output instant of the SFN adapter of the broadcasting sites, and thus ensuring the temporal synchronization of the radiofrequency signals broadcast by the different broadcasting sites receiving the same transport stream.

In another embodiment, the invention relates to a corresponding broadcast site.

The diffusion technique according to the invention can therefore be implemented in various ways, in particular in hardware form and / or in software form.

For example, at least one step of the technique of generating a transport or diffusion stream according to one embodiment of the invention can be implemented:

on a reprogrammable calculation machine (a computer, a processor for example DSP (in English "Digital Signal Processor"), a microcontroller, etc.) executing a program comprising a sequence of instructions,

on a dedicated computing machine (for example a set of logic gates such as an FPGA (English Field Programmable Gate Array) or an ASIC (in English "Application-Specific Integrated Circuit"), or any other hardware module). In particular, the computer program can use any programming language, and be in the form of source code, object code, or intermediate code between source code and object code, such as in a partially compiled form, or in any other desirable form.

Accordingly, an embodiment of the invention also aims to protect one or more computer programs having instructions adapted to the implementation of the methods of generating a transport stream or data broadcast as described herein. when one or more programs are executed by a processor, as well as at least one computer-readable information medium comprising instructions from at least one computer program as mentioned above.

An embodiment of the invention also relates to a broadcasting system comprising equipment for generating a transport stream and at least two transmission sites. broadcast as described above, wherein the broadcast sites are configured to transmit a radio frequency signal on the same frequency.

4. List of figures

Other features and advantages of the invention will appear more clearly on reading the following description of a particular embodiment, given as a simple illustrative and nonlimiting example, and the appended drawings, among which:

FIG. 1, described in relation with the prior art, presents a block diagram of a distribution network according to the ATSC 3.0 standard;

FIG. 2 illustrates a distribution network according to one embodiment of the invention; FIG. 3 illustrates a first example of a transport packet according to one embodiment of the invention;

FIG. 4 illustrates a second example of a transport packet according to one embodiment of the invention;

Figures 5 and 6 respectively show the simplified structure of a fixed reference site and a broadcast site according to one embodiment of the invention.

5. Description of embodiments of the invention

The invention is set in the context of a digital distribution and broadcasting network comprising at least one fixed reference site and a plurality of broadcast sites, wherein part of the physical layer modulation processing is implemented at from the reference site. This results in a flow of complex samples (also called I and Q samples, or I / Q samples), to be distributed to a plurality of broadcast sites.

More specifically, the general principle of the invention is based on the insertion of at least one synchronization time information associated with a group of complex samples obtained at the output of the physical layer modulation (implementation at the reference site level ).

FIG. 2 illustrates an example of a distribution network according to one embodiment of the invention, based on the ATSC 3.0 standard. Of course, the invention is not limited to this broadcast standard, and can be implemented with any digital broadcast standard, notably allowing SFN operation of broadcast sites.

The distribution network illustrated in FIG. 2 comprises a fixed reference site, located for example in a content creation studio, and a plurality of broadcasting sites, for example. for example, two SD1 and SD2 broadcast sites, implanted at distinct geographical sites and belonging to the same SFN plate.

5.1 Method implemented on the fixed reference site side

A) General principle

Such a fixed reference site, also called equipment for generating a transport stream, implements the method of generating a transport stream according to one embodiment of the invention.

Thus, at the studio level, the source data to be distributed 21 (source 1, source 2, source i, for example of the data, audio, and / or video services type, etc.), provided by one or more service providers, can 22. For example, the source data is encoded, multiplexed and ordered in a coding / multiplexing / scheduler block. The pre-processing step 22 may in particular be implemented in a broadcast gateway.

The source data, possibly pre-processed, are then modulated in a physical layer modulation block, delivering a modulated signal. Such a physical layer modulation block notably implements a transformation of the source data, possibly pre-processed, from the frequency domain to the time domain, for example by means of an inverse fast Fourier transform (IFFT for "inverse fast Fourier transform"). ").

The complex samples (samples I and Q) of the sampled modulated signal are framed 24 in a SFN / ST2L adaptation block, also called SFN / ST2L adapter or ST2L interface, generating a single transport stream to be distributed to different broadcast sites SD1, SD2 via a distribution network 25 (satellite link, microwaves, optical fiber, etc., possibly with IP or ASI transmission).

More specifically, the framing of complex samples comprises:

the distribution of the complex samples in at least one frame,

In other words, the ST2L interface makes it possible to transport the complex samples generated by a studio-level physical layer modulator (for example an ATSC 3.0 modulator) to a set of exciters of the broadcasting sites, with the information necessary to the temporal synchronization of these complex samples, for the purpose of SFN operation of the broadcast sites. In particular, the transformation of the frequency domain to the time domain being implemented in the physical layer modulator located at the studio level, each broadcasting site directly obtains complex samples (I and Q), which makes it possible to overcome the implementation of a transformation from the frequency domain to the time domain at the level of each broadcasting site .

Thus, SFN adaptation is performed by cutting the flow of complex samples into "frames" (or groups of successive complex samples) and associating each frame with at least one temporal information, also called timestamp or "timestamp". Such temporal information corresponds, for example, to the time at which this frame must be broadcast by each of the broadcast sites, either at the time of transmission of the first complex sample of the frame, or according to another example in FIG. SFN adapter output time of the broadcast sites.

Optionally, the complex samples can be compressed before the framing.

According to a first example, the complex samples are distributed in frames corresponding to conventional physical frames of a signal according to the ATSC 3.0 standard, also called ATSC 3.0 frames. In this case, it is easy to detect, on the broadcasting site side, the beginning of a complex sample frame.

According to a second example, the complex samples are distributed in "virtual" frames, of arbitrary length.

The length of the frames can be fixed (i.e. all the frames have the same length). In this case, the length of the frames may depend on the sampling frequency of the modulated signal, in order to obtain an integer number of complex samples per frame. For example, if the sampling rate is the frequency of the ATSC 3.0 system clock of 6.912 MHz, the modulated signal is sampled at 6.912 Msps (mega samples per second). Complex samples can therefore be distributed in one-second frames (each comprising 6,912,000 complex samples) - corresponding to an integer number of system clock periods, or in half-second frames (each comprising 3 456,000 complex samples), etc.

The length of the frames may also be variable. In this case, an indicator and / or a pointer can be used to mark the start, end, or length of a frame.

In another particular embodiment, a frame is fragmented to distribute its complex samples in one or more transport packets. For example, these transport packets "ST2L packets" are noted. The transport packet or packets, or directly the complex samples, may possibly be encapsulated in one or more IP packets, in particular in the payload part of the IP packets. The temporal information associated with a group of complex samples can be inserted in the header of one of the IP encapsulation layers. For example, the time information can be inserted in the "Timestamp" field of the RTP header. The length of the transport packets can be fixed (ie all the transport packets have the same length). In this case, a length advantageously chosen to have an integer number of complex samples per transport packet, and an integer number of transport packets per frame.

The length of the transport packets can also be variable. In this case, an indicator and / or a pointer may be used to mark the beginning, end, or length of a transport packet.

For example, such transport packets are of the MPEG-TS type, that is to say they have the same structure as the MPEG-TS packets. In other words, ST2L packets are of MPEG-TS packet type. Thus, each transport packet contains 188 bytes. The encapsulation of complex samples in transport packets makes it possible in particular to reuse modules conventionally used for transporting or processing MPEG-TS packets, such as modules implementing an error-correcting coding, etc. The use of transport packets of 188 bytes ensures compatibility of the proposed technique with the SMPTE-2022 standard.

In particular, a transport packet includes a reserved field for the time information (for example on 3 bytes) and a field carrying the complex samples (for example on 180 bytes).

According to a particular embodiment, making it possible to avoid the transport of the time information in all the transport packets of a frame, the reserved field for the temporal information of the transport packet carrying the first complex sample of the gate frame the time information, and the reserved field for the time information of the other transport packets is empty or non-existent.

Thus, if a frame comprises a single transport packet, the reserved field for the temporal information carries the time information.

For example, the structure of at least one transport packet is equivalent to the structure of an MPEG-TS packet, and includes:

a header including: o a synchronization field,

an error indicator indicating whether the complex samples carried by the transport packet are reliable,

a start indicator indicating whether the transport packet carries the first complex sample of a frame,

a payload comprising:

o a reserved field for temporal information,

a field carrying the number of the transport package, and

o a field carrying complex samples.

In the following, with reference to FIGS. 3 and 4, two examples are presented for framing complex samples.

In these two examples, it is considered that the frames comprise transport packets having a structure equivalent to that of an M PEG-TS packet, each transport packet carrying 188 bytes. According to these two examples, it is therefore considered that encapsulates complex samples in M PEG-TS type packets.

B) First example

Figure 3 illustrates the structure of a transport packet carrying complex samples, according to a first example.

According to this first example, the complex samples are mapped to 180 bytes, among the 188 bytes of the transport packet. Each component (I or Q) of a complex sample is coded on 16 bits. Each complex sample (l / Q) is thus coded on 4 bytes (2x16 bits), and each transport packet can carry 45 complex samples, on 180 bytes.

More precisely, the first byte, byte 1, is u n synchronization byte (0x47).

Byte 2, denoted FFP (in English "Frame Flag and Pointer", in French "indicator of frame and pointer"), includes:

an error indicator, on a bit, for example FFP (7), indicating whether the complex samples carried by the transport packet are "clean" or "reliable" (ie if the broadcast site can broadcast them). For example, if FFP (7) = 1, the complex samples carried by this transport packet will not be taken into account by the different broadcast sites (output of the "mute" exciter), because considered corrupt, and if FFP (7) = 0, the complex samples carried by this packet of transport will be taken into account by the different sites of diffusion, because considered as reliable; a start flag, on a bit, for example FFP (6), indicating whether the transport packet carries the start of a frame, ie the first complex sample of a frame. For example, if FFP (6) = 0, this transport packet does not carry the first sample of a frame, and if FFP (6) = 1, this transport packet carries the first sample of a frame;

a pointer from 0 to 44, for example FFP (5..0), indicating the position of the first complex sample of a frame, among the 45 complex samples of the transport packet. This pointer is used if FFP (6) = 1 (which means that this transport packet carries the start of a frame). For example, a pointer value of 0 indicates that the first complex sample of the transport packet is the first complex sample of a frame, and a pointer value of 44 (0x2C) indicates that the last complex sample of the packet of transport is the first sample of a frame.

The use of such a pointer therefore makes it possible to separate the frames from the transport packets. In other words, the complex samples of a transport packet may belong to different frames, and the number of transport packets in a frame is not necessarily an integer. Such a pointer thus makes it possible to delimit a frame, if the number of transport packets in a frame is not whole.

Byte 3, noted SEC, carries part of the temporal information, corresponding to the entire part of the time information ("absolute part of the timestamp"). In particular, if the time information corresponds to a transmission time, this byte indicates the number of the second (0 to 59) at which the first sample of a frame will have to be sent at the broadcast sites. This field is used if FFP (6) = 1 for this transport packet (which means that this transport packet carries the start of a frame). Otherwise, this field is empty, or reserved for transporting another type of information.

Bytes 4 to 6, denoted TS [23..0], carry part of the temporal information, corresponding to the fractional part of the temporal information ("relative part of the timestamp"). In particular, if the time information corresponds to a transmission instant, these bytes indicate the fraction of a second (0 to 0x6977FF), within a period of the system clock (for example at the frequency 6.912 MHz according to the standard ATSC 3.0) to which the first sample of a frame will have to be transmitted at the broadcasting sites. This field is used if FFP (6) = 1 for this transport packet (which means that this transport packet carries the start of a frame). Otherwise, this field is empty or reserved for transporting another type of information. The use of these two fields SEC and TS [23..0] thus makes it possible to obtain, at the level of the broadcasting sites, very precise information concerning the instant of emission. The temporal information, based for example on a system clock at 6.912 MHz for the fractional part, allows a simple and exact calculation at the SFN adapter of the broadcasting sites (references 261, 262 in FIG. 2). This fractional part is reset at each synchronization pulse (for example one pulse per second, or 1 pps for "Puise Per Second", resulting from a reference signal, for example of the GPS type). The whole part (SEC) is for example extracted from the UTC time.

Note that the distribution of this whole part to the broadcast sites is not mandatory, as shown below in connection with the second embodiment.

Bytes 7 to 8, denoted Rfu (in English "reserved for future use", in French "reserved for future use") are left empty in this first example.

Bytes 9 to 188 are used for the transport of complex samples, each complex sample being carried by 4 bytes (2x16 bits). For example, bytes 9 and 10 carry the component I of the complex sample n, and the bytes 11 and 12 the component Q of the complex sample n. Bytes 13 and 14 carry component I of the complex sample n + 1, and bytes 15 and 16 the component Q of the complex sample n + 1. Bytes 185 and 186 carry the component I of the complex sample n + 44, and the bytes 187 and 188 the component Q of the complex sample n + 44.

If the sampling rate is 6.912 Msps, the payload of the transport stream thus obtained ("usefui net bitrate"), at the level of the physical layer, is of the order of 6.912 × 32, or about 221 Mbps, and the overall rate of the transport stream at the level of the physical layer ("row bitrate") is of the order of 6.912x32x188 / 180, or about 231 Mbps.

Such a rate is compatible with a distribution of the transport stream over an Ethernet link.

Thus, the transport packets can in particular be grouped by seven, and the groups of seven transport packets encapsulated in an RTP packet, to transmit these RTP packets on an Ethernet interface. To do this, RTP packets can be encapsulated in IP packets, as described in SMPTE-2022. The set of protocols used for data distribution is IQ / TS / RTP / UDP / IP / ETH.

Concerning the RTP encapsulation, it is noted that the time stamp conventionally present in the header of the RTP packets is not necessary, since time information for the synchronization of the broadcasting sites is already present in the transport packets. The sequence number ("sequence number") conventionally provided in the header of the RTP packets can be used to detect a disordering or loss of RTP packets.

A Forward Error Correction (FEC) module, for example according to the SMPTE-2022-2 standard, can be implemented to protect the distribution over an IP network. In a variant, such an FEC module is not implemented. There is therefore neither reconstruction of the lost RTP packets, nor re-scheduling on the broadcast site. If RTP packets are lost (or unscheduled), they can be replaced by stuffing, that is, by null complex samples, at the broadcast sites.

Concerning the UDP / IP encapsulation, it is noted that the destination multicast IP address and the destination port number U DP are configurable by the user.

In particular, at the level of IP encapsulation, a checksum verification mechanism can be used to detect, on the broadcast site side, an error on the content and to delete a transport packet considered as corrupt / or replace the complex samples carried by this transport packet by stuffing at the broadcast sites.

For example, the chosen encapsulation generates an IP stream at 238 Mbps, for an overall bit rate of 231 Mbps (MPEG-TS bit rate) and a payload of 221 Mbps, for a sampling rate of 6.912 Msps (IQ bit rate).

It should be noted that, according to this first example, the overall bit rate obtained does not allow direct distribution of the transport packets, bearing the complex samples, on an ASI interface (in English "Asynchronous Serial Interface", in French "asynchronous serial interface"). .

Therefore, a second example is presented hereinafter allowing a direct distribution of the transport packets on an ASI interface.

C) Second example

Figure 4 illustrates the structure of a transport packet carrying complex samples, according to a second example.

According to this second example, the complex samples are mapped to 180 bytes, among the 188 bytes of the transport packet. Each component (I or Q) of a complex sample is coded on 12 bits. Each complex sample (l / Q) is thus coded on 3 bytes (2x12 bits), and each transport packet can carry 60 complex samples on 180 bytes. By reducing the number of bits used to encode a complex sample, one can increase the number of complex samples per transport packet, for the same transport packet length.

More specifically, the structure of a transport packet according to this second example is as follows:

byte 1, is a synchronization byte (0x47) - as for the first example;

Byte 2, marked FFP, includes - as for the first example:

an error flag, on a bit, for example FFP (7), indicating whether the complex samples carried by the transport packet are "clean" or "reliable" (i.e. if the broadcast site can broadcast them);

a one-bit start flag, for example FFP (6), indicating whether the transport packet carries the start of a frame, i.e. the first complex sample of a frame.

According to this second example, the FFP bits (5..0) of the octet 2 may be empty. As a variant, the FFP bits (5..0) of the byte 2 carry a pointer of 0 to 59 indicating the position of the first complex sample of a frame, among the 60 complex samples of the transport packet, as for the first example .

Byte 3, noted SEC, may not be used according to this second example.

Bytes 4 to 6, denoted TS [23..0], carry part of the temporal information, corresponding to the fractional part of the temporal information. In particular, if the time information corresponds to a transmission instant, these bytes indicate the fraction of a second (0 to 0x6977FF), within a period of the system clock (for example at the frequency 6.912 MHz according to the standard ATSC 3.0) to which the first sample of a frame will have to be transmitted at the broadcasting sites. This field is used if FFP (6) = 1 for this transport packet (which means that this transport packet carries the start of a frame). Otherwise, this field is empty, or reserved for transporting another type of information.

In the particular case of frames having a duration of one second and a system clock at the frequency of 6.912 MHz, the fractional part of the time information is identical for each frame (the TS field [23..0] indicating the fraction of a second within a period of the system clock, modulo a second).

Alternatively, to avoid the distribution of an empty field TS [23..0] if FFP (6) = 0, it is possible to code a complex sample on the three bytes 4 to 6. According to this variant, for the samples complex of a frame, the transport packet carrying the first complex sample of the frame (FFP (6) = 1) can carry the time information on bytes 4 to 6, and 60 complex samples on bytes 9 to 188. The other transport packets (FFP (6) = 0) can carry 61 complex samples, a complex sample on bytes 4 to 6 and 60 complex samples on bytes 9 to 188. According to this therefore, the transport packets do not all carry the same number of complex samples.

According to this second example, to improve the reliability / security of distributed complex samples, a counter is added to bytes 7 and 8. More precisely, bytes 7 and 8, denoted SEQ [15..0], carry the number of the packet. in the transport packet sequence, coded on 16 bits (ie modulo 65536 the transport packet number is double-byte coded). Such a field makes it possible to count the transport packets, and can be used to detect, at the level of the broadcast sites, a sequence break (disordering in the transport packets, loss of a transport packet, etc.).

Bytes 9 to 188 are used for the transport of complex samples, each complex sample being carried by 3 bytes (2x12 bits). For example, bytes 9 to 11 carry the component I and the component Q of the complex sample n. Bytes 12 to 14 carry the I component and the Q component of the n + 1 complex sample. Bytes 186 to 188 carry the I component and the Q component of the n + 59 complex sample.

If a sampling rate of 6.912 Msps is considered, the complex samples can be divided into frames having a length of 1 second, or 6 912 000 complex samples per frame. Thus, the 6,912,000 complex samples of a frame can be encapsulated in 115,200 transport packets, each carrying 60 complex samples. In this way, the number of complex samples per transport packet is whole, and the number of transport packets per frame is whole. For example, only the first transport packet carries a start flag FFP (6) equal to 1 (which means that this first transport packet carries the first sample of the frame) and the time information for the synchronization of the transport sites. diffusion (TS field [23..00]). In this case, it is not necessary to manage a pointer indicating the position of the first complex sample of a frame, since the first complex sample of a frame corresponds to the first complex sample of the first transport packet (equivalent to FFP [5..0] = 0).

In particular, the presence of an error indicator and SEQ field [15..0] allows the distribution of a robust transport stream to errors.

If the sampling rate is 6.912 Msps, the useful bitrate of the transport stream thus obtained ("useful net bitrate") at the level of the physical layer is of the order of 6.912x24, ie about 166 Mbps, and the overall flow rate of the transport stream at the level of the physical layer ("row bitrate") is of the order of 6.912x24x188 / 180, or about 173.3 Mbps.

Such a bitrate is compatible with a distribution of the transport stream on an interface

ASI.

Such a rate is also compatible with a distribution of the transport stream over an Ethernet link. In this case, an RTP / U DP / IP encapsulation can be implemented, as described in connection with the first example or in the SMPTE-2022 standard. The implementation of a FEC module to protect the distribution over IP network is optional, since information on the order of transport packets is already present in the SEQ field [15..00] of each packet of transport (number of the transport packet in the sequence).

5.2 Method implemented on the broadcasting site side

Returning to FIG. 2, the transport stream thus generated is distributed to the various broadcasting sites SD1, SD2 via the distribution network 25.

Each broadcast site may implement the data broadcast method according to one embodiment of the invention. In particular, the post-processing related to the power amplification of the radiofrequency signal for broadcasting remains managed by the broadcasting sites.

Each broadcast site SD1, SD2 therefore receives the transport stream and implements a process for resynchronizing the complex samples at each broadcast site. Indeed, each broadcast site receives a delayed version of the transport stream, due to the latency between the fixed reference site and the broadcast site and / or jitter. In the particular case of an SFN broadcast, each broadcast site must transmit a multiplex at the same time and at the same frequency. It is therefore necessary for each broadcast site to reconstruct and resynchronize the complex samples carried by the transport stream before broadcasting.

The data dissemination method implements the following steps, at each SFN plate diffusion site:

receiving the transport stream as described above, comprising at least one frame carrying complex samples, representative of a source signal, and at least one temporal information for the temporal synchronization of the broadcast site with at least one remote broadcast site;

detecting, in the transport stream, the first complex sample of a frame;

extracting, from the transport stream, complex samples of the frame;

detecting, in the transport stream, the temporal information associated with the frame; emission of the first complex sample of the frame at the instant defined by the information time.

For example, each broadcast site SD1, SD2 comprises an ST2L / SFN synchronization adaptation block 261, 262, implementing the steps of receiving the transport stream, detecting the first complex sample of a frame and information. time associated with the frame, and extraction of all the complex samples of the frame.

For example, as they are extracted, complex samples are stored in a buffer. Thus, at the time of transmission defined in the time information, the first complex sample of the frame can be broadcast. Such a buffer notably makes it possible to absorb the latency and / or the jitter of the different broadcasting sites. The writing of complex samples in buffer memory can be more or less fast, depending on the broadcast site. On the other hand, the reading of the complex samples is implemented at a regular rate, and common to the different broadcast sites, depending on the system clock.

Each broadcast site SD1, SD2, also implements a quadratic modulator (also called I / Q modulator) 271, 272, delivering a radio frequency signal, and a power amplifier 281, 282 of the radiofrequency signal.

For example, the adaptation block ST2L / SFN synchronization 261 (respectively 262) and the quadratic modulator 271 (respectively 272) present at the broadcast site SD1 (respectively SD2) belong to an exciter of the broadcast site SD1 (respectively SD2).

In particular, the quadratic modulator 271, 272, present at each diffusion site, implements a quadrature modulation of a transmission carrier, with the complex samples extracted from the transport stream, and a digital analog conversion CNA. The transmission step thus makes it possible to transmit the transmission carrier modulated by the first complex sample at the instant defined by the time information.

In the context of the examples described above, according to which we consider that the samples of a frame are distributed in transport packets having a structure equivalent to that of an MPEG-TS packet, the detection the first complex sample of a frame implements the extraction of the transport packets associated with this frame, and the detection of a start flag, for example FFP (6).

The transport packet carrying the indicator FFP (6) = 1 carries the first sample of this frame. For example, the FFP pointer (5..0) indicates the position of the first complex sample of the frame. It is thus possible to detect the first complex sample of a frame, and to extract the other complex samples from this frame (following samples). We can also detect and extract the temporal information associated with this frame, in the TS field [23..00] and possibly in the SEC field.

From the temporal information extracted from a first frame, it is possible to temporally synchronize the flow of complex samples in accordance with the value of this temporal information, for example by emitting, at each diffusion site, the first complex sample from the first frame to the instant defined by the time information. It can then be verified that the time synchronization is always correct thanks to the temporal information extracted from the following transport packets, or subsequent frames.

In particular, if the error indicator FFP (7) associated with a transport packet is equal to 1, the complex samples carried by this transport packet will not be taken into account by the different broadcasting sites, and possibly replaced. by stuffing (complex samples with zero value).

Similarly, if an error in the number of the transport packets is detected by the SEQ [15..00] field of the transport packets (disordering the transport packets or the loss of one or more transport packets), the Broadcasting (including the exciter) can reorder the transport packets, or replace the complex samples of lost transport packets with stuffing (zero-valued complex samples), or decide not to broadcast the data to protect the broadcast site.

In particular, these different fields FFP (7) and SEQ [15..00] make it possible to signal to the broadcast site, in particular to the exciter, a problem at the level of a transport packet, and to inform the amplifier of this problem, so that it does not seek to amplify the power of the corrupted or not received complex samples.

5.3 Devices

Finally, with reference to FIGS. 5 and 6, the simplified structure of a generating equipment for a transport stream and a broadcasting site according to one embodiment of the invention is presented.

As illustrated in FIG. 5, an equipment for generating a transport stream according to an embodiment of the invention comprises a memory 51 (comprising, for example, a buffer memory) and a processing unit 52 (equipped for example with less a processor, FPGA, or DSP), driven or pre-programmed by an application or a computer program 53 implementing the method of generating a transport stream according to an embodiment of the invention. At initialization, the code instructions of the computer program 53 are for example loaded into a RAM memory before being executed by the processing unit 52. The processing unit 52 receives as input at least one piece of content. distribute (source 1, source 2, source i). The processing unit 52 implements the steps of the generation method described above, according to the instructions of the computer program 53, to generate a transport stream comprising at least one frame carrying complex samples, representative of a source signal. , and at least one temporal information for the temporal synchronization of the broadcasting sites.

To do this, according to one embodiment, the processing unit 52 is configured to: modulate a source signal, delivering a modulated signal,

sample the modulated signal, delivering complex samples, and

to frame the complex samples in:

o distributing the complex samples in at least one frame,

o inserting, in each frame, at least one temporal information for the temporal synchronization of the broadcasting sites.

As illustrated in FIG. 6, a broadcasting site according to one particular embodiment of the invention comprises a memory 61 (comprising for example a buffer memory) and a processing unit 62 (equipped for example with at least one processor, FPGA , or DSP), driven or pre-programmed by an application or a computer program 63 implementing the data broadcasting method according to one embodiment of the invention.

At initialization, the code instructions of the computer program 63 are for example loaded into a RAM memory before being executed by the processing unit 62. The processing unit 62 receives as input a transport stream. The processing unit 62 implements the steps of the data diffusion method described above, according to the instructions of the computer program 63, for extracting and resynchronizing the complex samples, so as to ensure the temporal synchronization of the broadcasting sites.

To do this, according to one embodiment, the processing unit 62 is configured to: receive a transport stream, comprising at least one frame carrying complex samples, representative of a source signal, and at least one temporal information for time synchronization of the broadcast site with at least one remote broadcast site;

detecting, in the transport stream, the first complex sample of a frame;

extracting, from the transport stream, the complex samples of the frame; detecting, in the transport stream, the temporal information associated with the frame; and transmitting the first complex sample of the frame at the instant defined by the time information.

5.4 Variants

An example embodiment of the invention according to ATSC 3.0 has been described above. Of course, other standards of diffusion can be envisaged. The number of complex samples per frame or per transport packet can thus vary, for example according to the frequency of the system clock.

Similarly, the implementation of the method for generating a transport stream at a fixed reference site has been described, and the implementation of the method of broadcasting data at a broadcasting site. Of course, certain steps (for example the calculation and storage of complex samples) can be implemented in the "cloud" (in French "cloud"), by one or more remote servers, communicating for example through the Internet.

The implementation of certain operations in the "cloud" makes it possible in particular to simplify the equipment of the distribution network, and in particular the implementation of the physical layer modulation. This provides a more flexible architecture of the distribution network.

Note also that, in the two examples described for the structure of a transport packet, bytes 2 to 4 were used to facilitate the detection of frames and optimize the generation of a transport stream. However, these bytes 2 to 4 can be released in a particular embodiment of the invention, to ensure the compatibility of the transport packets with the MPEG-TS standard.

Claims

A method of generating a transport stream for distribution to a plurality of broadcast sites,

characterized in that it comprises the following steps:

modulation (23) of a source signal, delivering a modulated signal,

sampling said modulated signal, delivering a stream of complex samples, framing (24) said stream of complex samples, delivering said transport stream, and in that said framing step (24) comprises:

splitting said stream of complex samples into frames,

inserting, in each frame, a temporal information for the temporal synchronization of said broadcast sites, associated with the group of successive complex samples composing said frame.

2. Method according to claim 1, characterized in that said time information inserted in a frame corresponds to the instant of transmission of the first complex sample of said frame by the transmitters of said broadcast sites.

3. Method according to any one of claims 1 and 2, characterized in that the length of a frame depends on the sampling frequency of said modulated signal.

4. Method according to any one of claims 1 to 3, characterized in that said complex samples of a frame are distributed in transport packets.

5. Method according to claim 4, characterized in that a transport packet comprises a reserved field for said time information and a field carrying complex samples of said frame.

6. Method according to claim 5, characterized in that, for a frame:

the reserved field for the time information of the transport packet carrying the first complex sample of the frame carries said time information, and the reserved field for the time information of the other transport packets is empty.

7. Method according to any one of claims 5 and 6, characterized in that at least one of said transport packets also comprises:

a header including:

o a synchronization field,

o a start indicator indicating whether the transport package carries the first complex sample of a frame,

a payload comprising:

o said reserved field for said time information,

o a field carrying the number of the transport package,

o said field carrying said complex samples.

The method according to any one of claims 5 to 7, characterized in that at least one of said transport packets further comprises a pointer indicating the position of the first complex sample of a frame in the transport packet.

9. Method according to any one of claims 4 to 8, characterized in that said transport packets are of type M PEG-TS.

10. A method of disseminating data, implemented at a broadcast site, characterized in that it comprises an adaptation phase (261, 262) comprising the following steps:

receiving a transport stream, comprising frames each carrying complex samples, representative of a source signal, and time information for the temporal synchronization of said broadcast site with at least one remote broadcast site, associated with the group of successive complex samples composing said frame;

detecting, in said transport stream, the first complex sample of one of said frames;

extracting, from said transport stream, complex samples of said frame;

detecting, in said transport stream, said temporal information associated with said frame;

and a transmission phase of said first complex sample of said frame at the instant defined by said time information.

11. Method according to claim 10, characterized in that it comprises a quadrature modulation step (271, 272) of a transmission carrier, with said complex samples extracted from said transport stream,

and in that said transmission phase transmits said transmission carrier modulated by said first complex sample at the instant defined by said time information.

Apparatus for generating a transport stream for distribution to a plurality of broadcast sites, comprising at least one processor operatively coupled to a memory, and configured to: modulating a source signal, delivering a modulated signal,

sampling the modulated signal, delivering a flow of complex samples, and framing said flow of complex samples, delivering said transport stream, by: cutting the flow of complex samples into frames,

o inserting, in each frame, at least one temporal information for the temporal synchronization of said broadcast sites, associated with the group of successive complex samples composing said frame.

13. Data broadcast site, comprising at least one processor operatively coupled to a memory, and configured to:

receiving a transport stream, comprising frames each carrying complex samples, representative of a source signal, and temporal information for the temporal synchronization of said broadcast site with at least one remote broadcast site, associated with the group of complex samples successive components of said frame; detecting, in said transport stream, the first complex sample of one of said frames;

extracting, from said transport stream, the complex samples of said frame;

detecting, in said transport stream, said temporal information associated with said frame; and

transmitting said first complex sample of said frame at the instant defined by said time information.

14. Computer program comprising instructions for implementing a method according to any one of claims 1 to 11 when the program is executed by a processor.