FI99064C - Audio transmission in a digital broadcast radio system - Google Patents

Description

99064

This invention relates to audio transmission in a digital broadcasting system in which services can be transmitted as a continuous stream and as data packets.

5 In the digital audio system DAB (Digital Audio Broadcasting) developed for efficient use of frequency bands, the transmission path is completely digital. The system is intended to replace the current commonly used analog broadcasting using frequency modulation. The DAB defines a multi-carrier digital radio channel suitable for transmitting both audio and data services. A fully digital transmission channel can be either a continuous data stream or a packet channel. Packet forwarding is more flexible and makes it easier to send data units of finite length. The DAB system is described in ETSI (European Telecommunication Standards Institute) standard 300 401, February, 1995.

15 From the user's point of view, the DAB system has the highest level of abstraction called ensemble, Figure 1. It contains all the services that are in a certain single frequency band. The change of ensemble takes place by tuning to another frequency band, just like channel selection in the current FM radio reception. The ensemble is divided into services, in the figure Alpha Radio 1, Beta Radio and Alpha 20 Radio 2 as an example. In addition, there may be data services, although they are not shown in the figure. Each service is further divided into service components. The service component can be transmitted on either an audio channel or a data channel. For comparison, •; j / that the FM radio has only one service and one service component (audio / speech) on each channel. At the lowest level, a transmission frame with a duration of either 24 ms or 96 ms, depending on the DAB- .. 25 mode, consists of three parts that are • · • · * * ... in chronological order. The first is a synchronization channel that does not include a service * *! luinformaatiota. Next is the Fast Information Channel FIC (Fast Information * · • · ·

••• j Channel), which is a fixed length for each mode. The last is the MSC

• ·. ’; ‘(Main Service Channel), which contains all subchannels. The location, size and number of subchannels within the MSC may vary, but the size of the MSC is fixed. The MSC contains up to 63 different audio and / or data subchannels. The subchannels are numbered so-called. based on sub-Channel ld from 0 to 62. In addition, the MSC may have an Auxiliary Information Channel (AIC) for 2 99064 information with a fixed channel number 63. The AIC may contain the same type of information as the FIC. The MSC information is transmitted time-separated so that in DAB mode I, the MSC part of the frame is divided into four parts, and each part is placed in successive transmission frames. These parts are called CIF (Common Interleaved Frame), so the MSC part of the Mode I transmission frame contains four CIFs. In other modes, interleaving is not used, so MSC is the same as CIF.

At the transmitting end, the service provider can provide not only audio services but also data services and e.g. multimedia services. The DAB operator generates a DAB transmission signal from the audio information and data provided by the service providers, comprising successive transmission frames as shown in the lower part of Figure 1.

At the receiver, the information channel FIC and the channel MSC containing audio and data services are separated from the transmission frame. The subchannels are separated, decoded and then routed for further processing. The subscriber gets to know from the received FIC channel what all the services of the received ensemble contain and can choose the desired service or services accordingly. By combining the service components of the subchannels according to the application program, e.g. the desired multimedia service can be assembled.

20 One of the advantages of a DAB system is that data capacity can be allocated dynamically to service providers. The capacity can be up to 1,728 Mbit / s. In this case, the data is sent in packets according to Figure 2a, which • • · · · consist of a header field, a data field and a checksum. The meanings of the fields described are in accordance with the DAB standard. The Packet Header 25 contains information about the packet length (Pkt Len), Continuity Index (Cont Ind), First / Last packet information (First / Last), Service Component Identification Address (Pkt Address), Command ) and the actual data field * ···. length (Data Len). The Data Field contains the actual data to be transmitted and, if necessary, padding bits. Finally comes the package checksum (Pack 30 CRC).

By combining the data fields of the packets in the receiver, a so-called

Data group, Figure 2 B. The packets are formed at the transmitting end from the Data Group 3 99064 by simply breaking it into parts and placing each part in the data field of the data packet. Usually, a data group consists of the data fields of several packets sent in succession. At its simplest, one packet is enough to form a data group.

5 The data group is formed according to Figure 3. The meaning of the abbreviations in the Data Group header and Session header fields is as shown in the table below: _Data Group header__Session header_ EXT FL extension flag LAST FL last CRC FL CRC flag SEG NUM segment number SES FL session flags RFA reserved for future applications DG TYPE data group type LEN IND length of next address field CONT IND continuity index ADDR FIELD end user address REP IND repetition index EXT FIELD extension field__ These header fields are followed by the actual data as well as the data group 10 checksum DG CRC.

The continuous audio stream is transmitted in frames, the structure of which is shown in Fig. 4. At the transmitting end at 16 kHz, the 16-bit PCM encoded audio samples are divided into subbands and the subband samples are encoded using the masking effect, e.g. in the case of a mono channel, a speed of about 100 kbit / s. The 4-byte header of the frame contains information for the decoder of the receivers, such as synchronization information and information about the bit rate and sampling frequency. The bit allocation field after the checksum tells how the bits are allocated to each subband of the audio field • · · * '20 containing 36 coded samples and which bits have been removed.

t »I

samples when utilizing the masking effect. The Scale Factor Selection Information field tells how the audio • · · 1: sample group is scaled (normalized) in the decoder. Then enter the field containing the actual audio bits. Its information corresponds to 24 ms audio. The field contains 36 coded subband audio samples divided into twelve triplets, each of which thus contains 3 subband samples. The four triplets 4 99064 thus correspond to 12 ms of audio. This is followed by padding bits if there are fewer audio bits than the length of the audio field. Finally, there are the X-PAD and F-PAD fields that convey Program Associated Data (PAD). This data is in sync with the audio of the frame. The 5 PAD bytes of successive frames form the so-called PAD channel.

The audio portion of multimedia is thought to be transmitted in audio frames, however, there may be some reasons to transmit audio in packet mode as well. In principle, for example, audio files could be sent as packets, which would first be stored in the receiver's memory and reproduced from the speakers in oi-10 time during a multimedia presentation. The advantage of such an audio transmission is that the transmission can be performed at any bit rate and thus did not have to be a fixed bit rate used in the transmission of audio frames, i.e. the transmission of an audio stream, for which a number of allowed bit rates are given in the DAB specification.

The disadvantage of such a transfer is that the audio file must be stored in memory if the transfer rate is lower than the bitstream bit rate, or it must be buffered if the rate is higher than the audio stream rate. In the former case, playback cannot start immediately when the file arrives, while when buffering is used, playback can start immediately. However, the disadvantage is not real, as for most possible uses there is no need 20 to send an audio file at the bit rate of the audio stream. The real problem is that the audio stream in the audio frames cannot be sent as packets because there is no mechanism to indicate the boundaries of the audio frames.

• · ««

However, there are some applications that require real-time audio packet transmission. By real-time is meant here that the 25-bit rate is the same for the transmission of audio stream and audio packets, and that the same information as in the audio frames could be separated from the packets in some way.

One application would be to locate bit errors in audio frames by comparing j ·· * to the checksum CRC of packets received from the received audio frame that transmit the same audio information and the checksum CRC of the Data 30 Group formed from the packets. In this way, the audio samples of the audio frames would be bypassed by the CRC check. Using packets for such a purpose would mean additional audio ancillary signaling, which may be temporarily r - »» il liat ti f .i d,.

5 99064 accepted. Of course, if you want to permanently protect audio transmission as well as packet transmission, it would of course be better to directly improve the error protection of the audio bits in the audio frames.

Another possible application that requires real-time transmission of audiopa-5 chains is to assign audio only to a specific target group. The application utilizes the addressability of packets. According to the table above, the ADDR FIELD field is reserved for the end user's address in the session header. It could be used to target audio information only to a specific group, so that real-time packet transmission can be used as an information-hub for different user groups. The DAB specification also defines the transmission of information on the high-speed information channel FIC. The specification describes the interruption of an ongoing broadcast-type service by a bulletin, but the delimitation can only be done by identifying the services that will be interrupted. It is not possible to limit the bulletin to certain recipients only.

It is therefore an object of the present invention to provide audio transmission in packets which allows both addressable audio transmission and the indication of audio frames from a packet stream.

The set goal is achieved as described in the independent claim.

The audio of the first embodiment of the invention is transmitted on a packet channel as a file. Proceed in such a way that the audio frame is placed in the data group; ·. the data field. In this case, in the file transfer protocol, one segment corresponds to the audio <I «« .'r. frame. Packets from the channel are normally assembled into a data group 4, the data field of which is a file segment, which in this case is an audio-25 frame. In order to set the audio frames received from the data field in the receiver in the correct chronological order, the file transfer protocol must be used. Since the transmission rate can be any, the receiver has frames stored. ···, netting or buffering before presentation.

According to another embodiment, the audio frames are transmitted at the packet terminal 30 as a continuous stream. The procedure is to place the audio frame in the data field of the data group. The baud rate is exactly the same as the baud rate of the audio channel. In this case, the individual data groups are sent at exactly the right 6 99064 rate, in which case, in the sense of transfer, it is an endless queue of data groups. Therefore, no file transfer protocol is required. Buffering does not need to be used, but the receiver presents the audio frame according to how packet channel power enters the data groups.

The invention is illustrated by the accompanying figures, in which Figure 1 shows a known DAB hierarchy, Figure 2a shows the structure of DAB packets, Figure 2b shows the structure of a data group from packets, Figure 3 shows the structure of a data group, Figure 4 shows a DAB audio frame and Figure 5 shows Use of IDG.

In order to clearly indicate the boundaries of the audio frames when transmitting audio information in packet form, according to the invention only and only one audio frame as a whole, i.e. including the PAD part, is always placed in one data group.

According to the first embodiment, the audio frames are transmitted in packet mode as an audio file, i.e. the audio, thus having a start, a duration and an end, 20 forms a single file. According to the principle of file transfer used in DAB, the file is divided into segments and each segment is placed in the data group of the data Group. The protocol is briefly described below. According to the invention, the segment is exactly the same as the audio frame. Since the audio is transmitted as a file, the file transfer protocol must be used in order for the receiver part 25 to place the data groups assembled from the packets in the correct order. The file transfer rate may be higher, equal to, or lower than the data transfer rate of the audio stream.

The file transfer protocol can be implemented according to the general basic principle proposed by the Eureka-147 project, where each segment • · · ‘···’ forms one data group. The consecutive segments of the file are numbered 30 consecutively so that the number of the first segment in the Session header is 0. The last segment of the file is indicated by the flag of the LAST field of the session header of the data Group formed from it. The receiver receives the data packets and forms Data groups from them. If its checksum indicates a transfer of 7 99064. bit errors, the receiver extracts the data packets of the relevant Data Group from the file retransmission.

In order for the receiver to be able to extract the correct files from the transmitted packet stream and to know the type of file 5 for file processing, the EUREKA-147 project has proposed that a special Information Data Group (IDG) be set up. It is a file transfer descriptor, i.e. it provides the necessary information about the file to which it refers and is multiplexed with file segments.

Figure 5 shows the idea of IDG in file transfer. One IDG is associated with 10 only one file. It is placed at least first in the packet stream associated with the file, i.e. at the beginning of the file transfer, but may also be in the middle of the packet stream, i.e. the IDG may occur during the file transfer or the IDG may be sent some time before the actual file transfer. In Figure 5, the files X, Y and Z are transferred and the IDGs referring to the 15 files are marked with corresponding letters. The important thing that can be done with IDG is included in its data field, which is called file descriptor. The file description can be used to inform the receiver in detail of the necessary information about the file to be transferred. The file description includes a field with so-called transfer parameters (T parameters). A T-20 parameter called file type indicates the type of file, allowing the receiver's application program to decide which algorithm to use to analyze the file and interpret its contents.

• · ·

The applicant proposes to add a new file type parameter called “DAB audio” so that the receiver knows that the incoming file indicated in the IDG • ·; 25 is an audio file.

• · c \ According to the DAB specification, information about services is transmitted on the Fast Information Channel (FIC). It indicates the • and · location and quality of the service components in the multiplex. The service description for each service is placed in a separate field with the parameter field ni-30 us service component description. One parameter is the service component type and this parameter is set to the parameter "file transfer".

99064 β From the FIC channel information, the receiver can know that this is a file transfer and from the IDG information that it is an audio file. The receiver is thus able to receive the file, decode the audio, and present it. Depending on the data transfer rate, the receiver will need to use 5 audio file recording or buffering.

According to another embodiment, the audio frames are transmitted in packet mode, but as a continuous audio stream. The baud rate is exactly the same as what the baud rate would be when using audio frames. At the transmitting end, the procedure is such that when the audio frames become a continuous stream, the audio frame is first placed in the data group, and the packets are then formed from the data group to be transmitted. The audio frame also includes a PAD field. Thus, it is not a file transfer as in the first embodiment, so no file transfer protocol is required. Thus, information data groups IDG are not required. The data group containing audio must not exceed the CIF limits, so the entire data group must be transferred in one common interleaved frame CIF. In other words, the da-tagroup must accompany one transport frame. If a transmission repetition is used at the data group level, then the data group and all its repetitions must be transmitted in one CIF, i.e. each repetition is transmitted in one CIF.

When creating a Data Group, a session header may not be required, but it is preferred to use it because it includes an ADDR FIELD field for the end user address. This assignment can be used to target the audio to be transferred to the desired recipient group. If the session header • · · * is used, then the SEG NUM field is used as a counter that is incremented .. by one for each data group. This helps the receiver to stay in sync 25, because if a datagroup does not come in or is incorrect, • · · pauses the audio frame during playback. The flag in the field LAST FL is always zero, because it is a continuous audio stream, even in packet mode.

• ♦

The data group may be interleaved with other packet mode service components transmitted on the same subchannel, but the data group must remain in one CIF. If the data group is interleaved with other service components that also transmit data groups comprising audio, 9 99064 and is transmitted on the same subchannel, all data groups must be transmitted in one CIF.

As in the first embodiment, the applicant proposes that on the fast information channel FIC (Fast Information Channel), a service component type in the service component description field is transmitted as a service component type parameter "DAB audio stream".

It will be apparent to one skilled in the art that as technology advances, the basic idea of the invention can be implemented in many different ways. The invention and its embodiments are thus not limited to the examples described above but may vary within the scope of the claims.

• ♦ · «· · • · • · •« · • · «• (

• · C

«•» · • * · · • ·:: • · ·

Claims (12)

99064 A method for transmitting audio in a digital broadcasting system, wherein the audio to be transmitted in a continuous stream is transmitted as audio frames, in which packet mode the information to be transmitted is placed in a data field of a data group (DG) 5 and the data group is divided for transmission into data fields an audio frame is placed in the data field. A method according to claim 1, characterized in that a certain number of consecutive audio frames form an audio file and the audio file is transferred according to the file transfer protocol of the system. Method according to Claim 2, characterized in that the transmission rate of the audio frames can be freely selected within the limits of the packet transmission rate of the system. A method according to claim 1, characterized in that successive data frames are formed from successive audio frames and that the transmission rate of data packets transmitted in a continuous queue is selected such that the transmission rate of an audio frame placed in a data group is the same as a continuous audio frame transmission rate. A method according to claim 4, characterized in that the entire data group is transmitted in a single transmission frame, wherein the data group is not time-interleaved into several transmission frames. A method according to claim 1, characterized in that the segment number field in the header of the data group acts as a counter which is incremented by one when transmitting the data group. * A method according to claim 1, characterized in that the address field in the header of the data group is used to assign a user group • · • * ··. A method according to claim 1, characterized in that the receiver is informed by the mechanisms according to the system that there is audio information in the data field of the data group. I j ·: ** A method according to claim 1, characterized in that the digital broadcasting system is DAB. A method according to claim 9, characterized in that the information that there is audio information in the data field of the data group is indicated in the “file type” field of the information data group (IDG). 99064 A method according to claim 9, characterized in that the information that the audio is transferred as a file is indicated on the high-speed information channel FIC by setting the service component as a type parameter "file transfer". A method according to claim 9, characterized in that the information that the audio is transmitted in packet mode as an audio stream is indicated on the high-speed information channel FIC by setting the service component as a type parameter "DAB audio stream. • II« · · • · • · · · · · «II • • · · · • ·:: ··· •» · 99064