FR2931614A1 - Audio data channel restituting method for e.g. home cinema, involves selecting channel for substituting three audio data channels based on information representative of audio data channels, where channels is restituted by receiver devices - Google Patents

Abstract

The method involves detecting whether a data set of an audio data channel initially intended to a receiver device is erroneous. A channel for substituting three audio data channels is selected based on information representative of the audio data channels, where the data channels is restituted by receiver devices of a phonic diffusion system. The information representative of selection of the substitute channel is transmitted to the diffusion system. Independent claims are also included for the following: (1) a computer program product comprising program code instructions for implementing an audio data channel restituting method (2) a storage unit for storing the program code instructions (3) a receiver device for a phonic diffusion system.

Description

A method for outputting an audio data channel, computer program product, storage means and corresponding first receiver device. FIELD OF THE INVENTION The field of the invention is that of phonic reproduction systems.

More specifically, the invention applies to a multi-channel sound reproduction system comprising at least two sound channels, each channel carrying data in accordance with a dedicated speaker. For example, a 7.1-channel multi-channel sound reproduction system provides: - seven audio channels (front left, front center, front right, left side, right side, left rear, right rear) for speakers respectively satellites positioned around the user and; a channel dedicated to low frequencies, intended for a subwoofer, of a construction different from the other speakers and specialized in the restitution of this frequency range. The invention is particularly applicable when these satellite speakers are connected to the signal source by a digital wireless network using communication nodes. 2. TECHNOLOGICAL BACKGROUND Consider a multi-channel sound reproduction system (commonly referred to as home theater or home theater in English), in which each audio channel is intended to be reproduced by a satellite speaker and in which the source diffuses to the satellite speakers , by wireless communication, the digital data representing each of the audio channels.

During communications, transmission errors can occur, for example due to electromagnetic disturbances, and can interfere with the spatial rendering of the sound diffusion. Often such errors do not affect all the transmitted channels, and all communications (defined by a transceiver pair) in the wireless network are not necessarily affected by the same interference.

This phenomenon can occur when the different audio channels are multiplexed temporally (TDM for Time Division Multiplexing in English) or frequency (FDM for Frequency Division Multiplexing). Indeed, certain electromagnetic disturbances can be superimposed on a single transmission channel and affect only the data exchanges taking place on this transmission channel. Thus, in general, a loss of a set (or block) of data is often an isolated event. Mechanisms based on error correcting codes significantly reduce the rate of transmission errors. Nevertheless, in some cases, transmission errors that can not be repaired by the error correction code can appear. In this case, unwanted audible effects that significantly alter the user's spatial sound perception persist. Solutions for the concealment of the errors presented in the state of the art are based on a substitution of the missing audio data of a certain channel by data coming from another channel. For example, in a stereo system, the data in the left channel is replaced by the data in the right channel. There are also solutions based on substituting the missing audio data of a certain channel with data preceding the lost data and coming from the same channel as the lost data.

It should be noted that an inappropriate choice of a substitute audio channel, for example by inverting two audio channels, may strongly disturb the user's perception of the sound diffusion, especially the spatial rendering. Indeed, multi-channel sound reproduction systems are most often used in addition to a visual reproduction device, such as a screen or a video projector, and are intended to immerse the user in the displayed scene with a adapted management of phonic reproduction. For example, when broadcasting a scene taken from a movie, object movements are enhanced by a sound effect with a spatial dimension, obtained by synchronized broadcasting of audio channels (from a predetermined set speaker). An inappropriate choice of a substitute audio channel will then distort the desired spatial rendering by the sound broadcast and give a different impression of movement than that from the video display. A patent document DE 3,638,922 proposes a first solution allowing the concealment of uncorrected errors (non-repairable errors) for a system comprising only two audio channels. In the case of a non-repairable error in one of the two stereo channels (left / right), the solution described in this document proposes to replace the missing data of this channel with the data of the other channel. In the case of a large difference in signal amplitudes (or more generally a low correlation) between the two channels, the missing data is replaced either by repeating the previous data of the same channel, or by silence. The amplitude of the replaced signal is adapted, and a sliding transition (also called cross-fade) is applied between the original signal and the substitution signal, progressively decreasing the power of the original signal while progressively increasing the power of the signal. substitution signal at the previous level, the two signals being superimposed. Finally, in order to reduce the probability of ending up with the data of the two erroneous channels at the same time, the transmitter is able to apply a time offset between the channels, this offset being subsequently compensated by the receiver. This first known solution is based on the observation that the spatial sound perception is subject to a certain inertia, that is to say that brief changes in the representation of the direction and distance of the sound sources remain undetectable when they are applied for a reduced lapse of time and when their application frequency is rather low. However, the solution described in DE 3,638,922 discloses no guidance on the choice of the substitution channel when there are more than two audio channels. Indeed, the description is limited to the case of two stereophonic channels. Therefore, only one choice of substitution channel is possible as soon as a non-repairable error is detected. In addition, the patent document DE 3,638,922 does not seek to provide a solution to respect as much as possible the expected spatial dimension of the sound diffusion.

A second solution described in US patent 6,006,173, US 6,351,728 and US 6,490,551 proposes an improvement of the solution described in DE 3,638,922. In the context of this second solution, the data to be transmitted undergo a reduction process. Error detection on reception is then done in the reduced data. Therefore, the treatments disclosed in DE 3,638,922 can be applied by spectral value or subband rather than over the entire signal. This second solution allows the concealment of errors by using the processed subband signal or spectral values of the channel in question, as well as the untreated subband signals of the adjacent channels, the latter signals being used only if they are not used. have not themselves been affected by interference. However, this second solution does not give any indication on the choice of the substitution channel when there are more than two audio channels because, similar to the patent document DE 3,638,922, this second solution is limited to the case of two stereophonic channels. . Therefore, only one choice for the substitution channel is possible. In addition, these US patent documents 6,006,173, US 6,351,728 and US 6,490,551 do not seek to provide a solution to respect as much as possible the expected spatial dimension of the sound diffusion. A third solution described in US Pat. No. 6,985,856 proposes to make a choice between the use of inter-frame interpolation (within the same audio channel) or inter-channel interpolation following an evaluation of the distance. Euclidean signals in question. The substitution channel is selected from predetermined preferential channels. For example, in the aforementioned case of a type 7.1 audio reproduction system as described in FIG. 1, the user is placed in the center of the sound reproduction system. It is in this case surrounded by seven speakers, each channel of each speaker being named according to the relative position of each speaker with respect to the user and as follows: a channel C corresponding to a central speaker; an FL channel corresponding to a front left speaker; an SL channel corresponding to a left lateral speaker; a BL channel corresponding to a left rear speaker; - a BR channel corresponding to a right rear speaker; an SR channel corresponding to a right lateral enclosure; an FR channel corresponding to a right front speaker; - a SW channel corresponding to a subwoofer. For example, the substitution channel to replace the BR channel of the right rear speaker corresponds to the BL channel of the left rear speaker, and vice versa. In US patent application document 2005 / 0,182,996 is disclosed a technique (frequently used in voice synthesis) called formant, that is to say based on energy maxima of the sound spectrum. Filtering parameters are determined for each speaker by correlating the output of this filter with the signal that is intended for the speaker in question, by using a monophonic signal in excitation of the filter. This monophonic signal is constructed as a weighted sum of all the audio channels, the signal received by the pre-processing speaker containing all the audio channels. In the presence of errors on an audio channel, or even disappearance of an audio channel, keeping the filter parameters previously established, the error (loss of its transported) is first concealed for a certain time by applying the monophonic sound to the filter and using the output of the filter as a sound signal to be restored by the speaker. The excitation signal must then be adapted by reducing the weighting associated with the erroneous channel and new filtering parameters must be estimated accordingly. Although the filters dampen the inversion effect, such a phenomenon may occur, hence the need to silence the speaker if the signal received by it contains too many wrong audio channels.

Although these third and fourth solutions are able to provide an indication of the choice of the substitution channel when there are more than two audio channels, a channel inversion is still possible. Indeed, in no way do these two solutions propose to avoid that a first erroneous data channel is replaced by a second data channel, itself erroneous, and using said first data channel as a substitution channel.

Thus, these US Pat. No. 6,985,856 and US Pat. No. 2005 / 0,182,996 do not attempt to provide a solution that makes it possible to respect as much as possible the expected spatial dimension of the sound diffusion. OBJECTIVES OF THE INVENTION The invention, in at least one embodiment, has the particular objective of overcoming these various disadvantages of the state of the art. The present invention aims in particular to provide a technique for minimizing unwanted audible effects and to preserve as far as possible the spatial sound perception of the user, when residual errors remain in the audio channels after transmission in a system of multi-channel phonic reproduction. Another objective is to provide such a technique that is simple to implement and inexpensive. 4. DISCLOSURE OF THE INVENTION In a particular embodiment of the invention, there is provided a method of rendering, by a first receiving device of a sound broadcasting system comprising a plurality of receiving devices, a channel of audio data from a set of at least three audio data channels received by said first receiving device, at each receiving device initially for a given one of said audio data channel of at least three audio data channels. This method is remarkable in that it comprises an audio data channel selection phase to be restored comprising the following steps: first detection that a set of data of the audio data channel initially intended for said first receiver device is erroneous; in the case of a first positive detection, selection of a substitution channel, based on information representative of the channels, of said set of at least three audio data channels, restored by the receiving devices of said plurality of receiving devices. Thus, for a receiver node whose audio channel to be restored comprises an erroneous audio data block, the method according to the invention compensates for this error by substitution with data from another channel, while taking into account the audio channels. which are restituted in the rest of the multi-channel sound reproduction system. Thus, it is possible to avoid the permutations of channels between left and right, and between front and rear in such multi-channel sound reproduction systems such home cinema for example. Thus, it is also possible to avoid the overrepresentation of an audio channel in a multi-channel sound reproduction system. An audio channel that is overrepresented creates a disruption in the distribution of sound power and a feeling of acoustic imbalance is then perceived by the user. Advantageously, said selection step comprises a verification sub-step that said substitution channel is not initially intended for a second receiver device which, by channel substitution, restores the channel originally intended for said first receiver device. For example, if the left rear speaker is already using the BR channel originally intended for the right rear speaker, then the right rear speaker can not select the BL channel originally intended for the left rear speaker. Channel inversions are thus avoided. The sound spatial perception at the level of the user is in this case preserved. Advantageously, in the case of negative verification, the selected substitution channel is a synthesis audio data channel representative of a silence. Thus, when the substitution of the lost data of a given audio channel proves inadequate to preserve the expected spatial dimension of the sound diffusion, the erroneous data are replaced by data of synthesis of the silence. Advantageously, said step of selecting a substitution channel, called the first substitution channel, comprises the following steps: classifying said set of at least three audio data channels according to a predetermined loop sequence; second detection that at least one other receiver device uses a second substitution channel corresponding to an audio data channel which is neighbor, in a determined direction of travel of said sequence, of the audio data channel initially intended for said other receiver device; in the case of a second positive detection, selecting said first substitution channel, taking an audio data channel that is adjacent, according to said determined direction of travel of said sequence, of the audio data channel initially intended for said first receiving node. Thus, the method makes it possible to avoid not only a channel inversion but also the use of the same audio data channel to make several substitutions. For example, if the left rear speaker is already using the BR channel originally intended for the right rear speaker, then the right side speaker will not be able to use the BR channel originally intended for the right rear speaker, but instead will use the channel FR originally intended for the right front speaker.

In a particular embodiment of the invention, said step of selecting said first substitution channel comprises the following steps: third detection that no other receiver device uses a substitution channel; in the case of a third positive detection, selection of said first substitution channel according to at least one predetermined criterion. Thus, if no other device or receiving node uses a substitution channel, the choice of said first substitution channel is made by taking an audio data channel which is adjacent to the audio data channel initially intended for said receiving device or node. Indeed, in this case, there is no risk of channel inversion. The choice can then be made among the immediate adjacent audio data channels of said receiver device, either in a predefined manner or by correlation determination of audio signals between the audio channel to be substituted and the neighboring audio channels: for example, the first channel Substitution is the one whose signal shows the best correlation with the signal of the audio data channel associated with said receiving node. In another particular embodiment of the invention, in the case of a second positive detection, said step of selecting said first substitution channel comprises the following steps: selecting a first neighboring audio data channel, said current neighbor channel, which neighbor, according to said determined direction of travel of said sequence, the audio data channel initially intended for said receiving device; fourth detection that the data of said current neighbor channel are erroneous;

- in case of fourth positive detection, selection of a new data channel

neighboring audio, which is neighbor, according to said determined direction of travel of said

sequence, of the current neighbor channel.

Thus, if the audio data belonging to the current neighbor channel is also erroneous, a new neighbor channel running further and providing correct data is then selected.

Advantageously, said step of selecting a substitution channel is followed by a step of transmitting, in said sound broadcasting system, information representative of the selection of the substitution channel performed by said first receiver device.

Thus, other devices or receiving nodes know the current state (substitution or not) of the device or receiving node considered. If these devices or nodes have to substitute the channel which is initially intended for it, an appropriate selection of the substitution channel is made possible thanks to this knowledge of the current state of the other devices or receiving nodes.

In another embodiment, the method further comprises the following steps:

fifth detection that a set of audio data of the audio data channel initially intended for said first receiving node is correctly received;

in case of fifth positive detection, selection of said audio data channel

initially intended for the first receiver device.

As soon as the receiving device or node no longer receives erroneous data blocks for the audio channel initially intended for it, it then processes the data of the audio data channel associated with said receiving node again.

Advantageously, in the case of a fifth positive detection, the method comprises a step of transmitting, in said sound broadcasting system, information representative of a selection stoppage of the substitution channel performed by said first receiver device.

Thus, when the receiving device or node no longer substitutes the audio channel that is initially intended for it, this information is broadcast to other receiving devices or nodes, in order to update the information relating to the current state of the others. receiving devices or nodes.

The invention also relates to a computer program product downloadable from a communication network and / or recorded on a computer readable medium and / or executable by a processor, and characterized in that it includes program code instructions for implementing the method according to the invention, when said program is executed on a computer.

The invention also relates to a storage medium, possibly completely or partially removable, readable by a computer, storing a set of instructions executable by said computer to implement the method according to the invention.

The invention also relates to a first receiving device of a sound broadcasting system comprising a plurality of receiving devices, and enabling the reproduction of an audio data channel from a set of at least three audio data channels received by said first device receiver, each receiving device being initially intended for a given audio data channel of said set of at least three audio data channels.

This first receiver device is remarkable in that it comprises:

a first detection means that a set of data of the audio data channel initially intended for said first receiver device is erroneous;

means for selecting a substitution channel, as a function of information representative of the channels, of said set of at least three audio data channels, restored by the receiving devices of said plurality of receiving devices.

Advantageously, this first receiver device comprises verification means that said substitution channel is not initially intended for a second receiver device which, by channel substitution, restores the channel originally intended for said first receiver device.

Also, this first receiver device comprises means for selecting a synthetic audio data channel representative of silence in the event of negative verification.

Advantageously, this first receiver device comprises for the selection of a substitution channel, called the first substitution channel:

means for classifying said set of at least three audio data channels according to a predetermined loop sequence; a second detection means that at least one other receiver device uses a second substitution channel corresponding to an audio data channel which is neighbor, in a determined direction of travel of said sequence, of the audio data channel initially intended for said other receiving device;

means for selecting said first substitution channel, by taking an audio data channel which is neighbor, according to said determined direction of travel of said sequence, of the audio data channel initially intended for said first receiving node.

In addition, this device comprises:

a third detection means that no other receiver device uses a substitution channel;

means for selecting said first substitution channel according to at least one predetermined criterion.

Advantageously, in the case of a second positive detection, this first receiver device furthermore comprises:

Means for selecting a neighboring neighboring audio data channel, said neighbor channel, which is neighbor, according to said determined direction of travel of said sequence, of the audio data channel initially intended for said receiver device;

fourth means of detecting that the data of said current neighbor channel are erroneous;

means for selecting a new neighboring audio data channel, which is neighbor, according to said determined direction of travel of said sequence, of the current neighboring channel;

Advantageously, this first receiver device comprises means 30 for transmitting, in said sound broadcasting system, information representative of the selection of the substitution channel performed by said first receiver device. Also, this device comprises: - a fifth detection means that a set of audio data of the audio data channel initially intended for said first receiving node is correctly received; means for selecting said audio data channel initially intended for the first receiver device. In addition, this first receiver device comprises means for transmitting, in said sound broadcasting system, information representative of a selection stop of the substitution channel performed by said first receiver device. 5. LIST OF FIGURES Other features and advantages of embodiments of the invention will appear on reading the following description, given by way of indicative and nonlimiting example, and the appended drawings, in which: FIG. schematically a wireless multi-channel sound reproduction system embodying the present invention; FIG. 2 diagrammatically illustrates the structure of a node constituting the communication interface of an enclosure (or receiver node) belonging to the wireless multi-channel sound reproduction system of FIG. 1; FIG. 3 illustrates an error detection algorithm implemented by a receiver node of the wireless multi-channel sound reproduction system of FIG. 1; FIG. 4 illustrates an audio channel selection algorithm, according to a particular embodiment of the invention, implemented by a receiver node of the wireless multi-channel sound reproduction system of FIG. 1; FIG. 5 illustrates an information management algorithm for channel substitutions, according to a particular embodiment of the invention, implemented by a receiver node of the wireless mufti-channel phonic reproduction system of FIG. 1 ; - Figure 6 shows a diagram of a generic node adapted to implement each method according to a particular embodiment of the invention. DETAILED DESCRIPTION The present invention relates to a technique for minimizing undesired audible effects and preserving as far as possible the spatial sound perception of the user. Subsequently, an exemplary implementation of the invention is described for a multi-channel sound reproduction system comprising at least three channels of audio data, each channel carrying audio data for at least one given speaker. More particularly, the exemplary implementation of the invention is a type 7.1 system, where each channel carries audio data for a dedicated speaker. Figure 1 illustrates a type 7.1 wireless multi-channel reproduction system. This system comprises a WSC 110 (Wireless Surround Controller) wireless controller (or transmitter node), broadcasting an audio signal to all the satellite speakers (or receiver nodes) 100 to 107 via communication nodes WAS 100a to 107a (for Wireless Active Speaker). In an alternative embodiment of FIG. 1, the WAS nodes are integrated in the enclosures to which they are each associated. The term WAS satellite speakers or WAS nodes will be used interchangeably. All nodes are equipped with transmit and receive antennas (it is possible for a single antenna to perform the function of the receiving antenna and the transmitting antenna). In more detail, the user is placed in the center of the sound reproduction system. It is in this case surrounded by eight speakers 100 to 107.

Each audio channel is named according to the relative position of the speaker to which it corresponds with respect to the user and as follows: a channel C corresponding to a central speaker 100; o an FL channel corresponding to a front left speaker 101; an SL channel corresponding to a left lateral enclosure 102; o a BL channel corresponding to a rear left speaker 103; o a BR channel corresponding to a right rear speaker 104; an SR channel corresponding to a right lateral enclosure 105; an FR channel corresponding to a right front speaker 106; o a SW channel corresponding to a subwoofer 107. The satellite speakers 100 to 106 are speakers covering a wide band of acoustic frequencies (from 100 Hz to 20 kHz approximately) locatable by human hearing. Consequently, the physical location of these speakers with respect to the user and with respect to a possible video screen (not shown) is very important in order to ensure a good spatial rendering of the sound and a good audio-visual coherence. In a particular embodiment of the present invention, the satellite speakers 100 to 106 are of identical construction. In general, the smaller the distance between two satellite speakers, the closer the signals of the respective audio channels are. For this reason, in the case where the signal dedicated to a certain speaker turns out to be unusable due to transmission errors, and a substitution of the data is envisaged using another audio channel, it is preferable to choose an audio channel. corresponding to a neighboring speaker as a substitution channel, in order to minimize the impact of the substitution on the perception of the user. For example, at the speaker 104 (rear right), the choice is possible between the channel corresponding to the speaker 103 (left rear) or the channel corresponding to the speaker 105 (right side). Audio channels corresponding to more distant speakers, for example dedicated to the speaker 102 (left-hand side) or to the speaker 106 (front right) are used in rarer cases where the closer channels prove to be unusable at their turn. In this case, the impact on the user's spatial perception may be greater. The speaker 107 representing the subwoofer is of a construction different from the other speakers. This speaker 107 is specialized in the reproduction of low frequencies (<100 Hz approx). These frequencies are difficult to locate by human hearing. For this reason, the physical location of the subwoofer is of lesser importance. The audio channel corresponding to the speaker 107, also called LFE channel, is generally not interchangeable with other audio channels because it only carries low frequencies. Therefore, this audio channel does not participate in the channel substitution process.

FIG. 2 schematically illustrates the structure of one of the WAS receiver nodes 100 to 106 (104 in the example of FIG. 2) belonging to the wireless multi-channel sound reproduction system of FIG. 1. The radio signal coming from the receiving antenna 220 (and whose multiplexing is performed, for example, in a TDM time division or FDM frequency division mode) is processed and demultiplexed by a processing module 200. This processing module 200 extracts the seven audio channels C, FL, SL, BL, BR, SR, FR respectively corresponding to the speakers 100 to 106 and a control CTRL signal. This control CTRL signal is more fully described later.

Once extracted, the seven audio channels C, FR, SR, BR, BL, SL, FL are then respectively processed by detection and error correction units 210, 211, 212, 213, 214, 215 and 216. For reasons of limiting the cost of manufacturing such speakers, it is possible for the processing module 200 to output only a reduced number of channels from the seven audio channels C, FR, SR, BR, BL, SL, FL transmitted by the WSC node. This makes it possible to limit the memory capacity necessary for the implementation of the invention. The units 210 to 216 are of identical construction and perform an error detection and correction algorithm for detecting and correcting an erroneous audio data block received for a given audio data channel.

The amount of errors detected for a given channel is further determined by error correcting modules 210a to 216a of the units 210 to 216 respectively. This error detection and correction algorithm is described in more detail below with reference to FIG. 3. After executing the detection and error correction algorithm, each of the units 210 to 216 delivers: the data audio, for which possible correctable reception errors have been corrected; and an auxiliary signal indicating whether there are uncorrectable errors in the delivered audio data.

An audio channel selection unit 204 executes an algorithm for selecting an audio channel from among all the audio channels received by the receiving node: this selected audio channel may be, depending on the operating conditions, the audio channel intended for the node receiver or a substitution channel. This selection algorithm is described in more detail in connection with FIG. 4. The audio channel selection unit 204 then delivers a selected signal which is: - either a main signal, corresponding to the audio data channel intended for the node receiver, if all transmission errors present in the data destined for the receiving node have been corrected; or a substitution signal, corresponding to a substitution channel, if all the transmission errors present in the data intended for the receiving node could not be corrected. Note that in the case of cross-fading processing, the main signal is also outputted from the channel selection unit 204 when a substitution signal is selected by the channel selection unit 204. .

When the main signal has been selected by the audio channel selection unit 204, it is decoded by a unit 205a and then processed by a cross-fade unit 223. When a substitution signal has been selected by the audio channel selection unit 204, it is decoded by a unit 205b and then processed by the cross-fade unit 223. The rest of the treatment is identical to the treatment performed on the main signal. In the case where no cross-fade is applied, the cross-fade unit 223 is absent from the data processing chain and the units 205a and 205b can be directly connected to the conversion unit 206, and this in function of the selection made by the audio channel selection unit 204. The selected signal is converted from digital to analog by a conversion unit 206. The selected signal is then filtered by a filter 206a and amplified by an amplifier 207. A loudspeaker 222 then restores the sound corresponding to the selected signal.

The control signal CTRL coming from the unit 200 is processed by a storage unit 201 making it possible to update a channel utilization chart 202 by applying the algorithm described in relation with FIG. Table 202 is taken into account by channel selection unit 204 during the selection operation and can be updated by it. The information (selection or not of a substitution channel) is transmitted to a transmission unit 203 to ensure the dissemination of information to other satellite speakers (or receiver nodes) using an antenna 221 This treatment will be described in more detail in connection with FIG.

FIG. 3 illustrates the error detection and correction algorithm implemented by a receiver node of the wireless multi-channel sound reproduction system of FIG. 1. This algorithm is unrolled independently by each of the units 210 to 216.

In a first step 300, the audio signal comprising all the audio data channels for the satellite speakers (seven in the example) is received by a current receiving node. In a next step 301, for each audio channel, a data block (from the demultiplexer 200) is extracted so as to be able to apply an error correction algorithm in a next step 302. By way of example in no way the system uses a Reed-Solomon error correction code, associated with a CRC code (for Cyclic Redundancy Check in English) or an LDPC code (for Low). Density Parity Check in English or French low density parity code) which has the property of being a detector and error corrector. In both cases, the WSC sending node 110 adds redundancy information of a predetermined size to each block of data to be transmitted. The method of decoding the correction code, well known to those skilled in the art, can then have two possible outcomes, tested by the use of an error detector code in a step 303.

In the most favorable case, all the erroneous bytes of a block of data could be corrected (or there were no transmission errors). In this case, the processing continues with a step 304 indicating that the block after treatment is valid. In a less favorable case, the erroneous bytes of this data block are too numerous compared to the capacity of the redundancy information. The correction fails in this case. The processing then continues with a step 305 signaling that the block of data obtained after processing is erroneous. In step 304, the valid data block (once all redundancy information has been extracted) is passed to the audio channel selection unit 204, while signaling that it is exploitable audio data. On the other hand, in step 305, the audio channel selection unit 204 is informed of receipt of an erroneous audio data block on the analyzed audio data channel. In both cases, the processing of the data block at one of the units 210 to 216 stops in a step 306. FIG. 4 illustrates the algorithm of the audio channel selection method implemented by a node multi-channel wireless reproduction system receiver of FIG. 1. More precisely, FIG. 4 describes the different steps of the method of selection, by a receiver node, of an audio data channel from a set of at least three channels of audio data included in a radio signal received by said receiving node. Typically, the radio signal is transmitted from a sender node to a set of at least three receiver nodes, each receiver node being associated with a separate audio data channel of said set of at least three audio data channels. The algorithm of FIG. 4 is executed by the audio channel selection unit 204 upon reception of data blocks (correct or erroneous) belonging to the audio data channel (initially) for the considered receiving node. In a step 400, the audio channel selection unit 204 receives a block of audio data to be analyzed.

A step 401 then makes it possible to test the state (correct or erroneous) of the block of audio data to be analyzed (also called current data block). Then, in a step 402 or 410 is tested the error state of at least one audio data block previously stored and preceding the current data block. Four cases are then conceivable: if the current data block (step 401) is erroneous and a set of a predetermined number of previous data blocks was also erroneous (step 402), then the processing continues in a step 409. - if the current data block (step 401) is erroneous and a set of a predetermined number of previous data blocks was correct (step 402), then the processing continues in a step 403. - if the current data block (step 401) is correct and a set of a predetermined number of previous data blocks was erroneous (step 410), then processing proceeds to a step 411. - if the current data block ( step 401) is correct and a set of a predetermined number of previous data blocks was also correct (step 410), then processing continues in a step 412. Note that when processing a first data blockaudio received after initialization of the system, it is considered in tests 402 or 410 that no erroneous block has previously been received by the corresponding receiving node. Note also that the predetermined number of previous data blocks analyzed may be equal to a block of data. Step 403 consists in checking, in Table 202, whether another receiving node is already applying audio channel substitution following receipt of an erroneous data block. In the case of a positive verification in step 403, processing continues in a step 404 of determining which channel the other receiving node uses in substitution for the channel that is (originally) intended for it. Preferably, according to a particular embodiment of the invention (as described hereinafter with respect to steps 405 and 406), the substitution takes place with a neighboring channel. Step 404 then consists in determining which neighbor channel the other receiving node uses in substitution for the channel which is (initially) intended for it. Note that by neighboring channel is heard an immediate neighbor channel, left or right, seen from the user, for a given receiving node. For example, for the receiving node 104 of FIG. 1, its right-hand neighbor channel is associated with the receiving node 103 and its left-hand neighbor channel is associated with the receiving node 105. Thus, if it is the left-neighbor channel said other receiving node, then a next step 405 consists in choosing, for the current receiving node, the neighbor channel (immediate) left as a substitution channel. For example, if the current receiving node is the node 104 of FIG. 1, the neighbor (immediate) left channel of the current receiving node corresponds to the receiving node 105. If it is the right neighbor channel of said other receiving node , then a next step 406 consists of choosing, for the current receiving node, the right neighboring (immediate) channel as a substitution channel. For example, if the current receiving node is the node 104 of FIG. 1, the right neighbor (immediate) channel of the current receiving node corresponds to the receiving node 103. The general principle of steps 404, 405 and 406 therefore consists of selecting a a substitution channel from a set of at least three channels of audio data, the selection of a substitution channel being performed by: obtaining channel information indicating for each of the other receiving nodes whether the channel in use is the audio data channel initially intended for said other receiving node or a substitution channel; and * by checking at least the following condition based on said channel information: said substitution channel should not be the audio channel initially intended for another receiving node which uses as the substitution channel the audio channel initially intended for said receiving node L ' The invention thus makes it possible to avoid creating an inversion between two speakers which would significantly modify the spatial effect compared with that to be rendered by the sound diffusion system.

For example, if the left rear speaker is already using the BR channel of the right rear speaker, then the right rear speaker does not select the BL channel of the rear left speaker. The sound spatial perception at the level of the user is in this case preserved. The invention also proposes to classify the set of at least three audio data channels in a predetermined loop sequence. In this case, the selection of the substitution channel is performed by taking an audio data channel which is neighbor, according to a determined direction of travel of said sequence, of the audio data channel associated with said receiving node. Interestingly, the direction of travel of said sequence is determined following detection that at least one other receiving node uses a substitution channel (step 404) corresponding to an audio data channel which is neighboring, in a given direction of said sequence, of the audio data channel initially intended for said other receiving node. Thanks to this predetermined loop sequence, the channel inversions are then avoided because the selection of each substitution channel among the neighboring channels (right or left seen by the user) is always performed in an identical direction of travel. In the case of a negative verification during step 403, the processing continues in a step 407 consisting in selecting, according to at least one predetermined criterion, for the current receiving node, the substitution channel corresponding to the immediate neighboring nodes (of left and right) of the current receiving node. For example, the neighbor channel of the current receiving node whose signal shows the best correlation with the signal of the channel associated with said current receiving node is selected. Of course, in this case, the channel selection unit 204 then has a signal correlation determining unit of the channel for the current receiving node and its neighboring receiving nodes. Another example is to select the neighbor channel among predetermined preferential channels (for example, the BL channel to replace the BR channel, the BR channel to replace the BL channel, the FL channel to replace the SL channel, the FR channel to replace the SR channel, ...).

Another possibility consists in selecting the channel presenting the least errors among the neighboring channels, this quantity of errors being determined by the error-correcting modules 210a to 216a of the detection and error correction units 210 to 216 respectively . In an alternative embodiment, the choice of an adjacent channel of the current receiving node is performed, if necessary, iteratively.

More precisely, if the neighbor channel initially selected during steps 405, 406 or 407 is also erroneous, then a new adjacent audio data channel is selected, which is, in the predetermined direction of travel of the predetermined loop sequence, current neighbor channel. This selection of a new audio data channel is in particular performed iteratively by taking the new neighbor audio data channel as the current neighbor channel; Otherwise, if the neighboring channel initially selected in steps 405, 406 or 407 is not erroneous, then this channel is chosen as the substitution channel. After the steps 405, 406 or 407, the processing then continues in a step 408 for storing and broadcasting information relating to the selected substitution channel. Then, a step 409 is executed. More specifically, during step 408, the information on the selected substitution channel is stored locally in table 202. This information is then broadcast to the other receiving nodes by radio using a transmission unit 203 associated with an antenna. 221 (Figure 1).

It should be noted that it is also possible for the exchanges of this information to take place between the WAS receiving nodes 100 to 107 and the WSC 110 sending node, according to a centralized control protocol. In this case, the transmitter node WSC 110 indicates, in correspondence with the audio channels that it broadcasts, the audio channel reproduced by each speaker or receiving node. The receiving nodes send their WSC channel node 110 their channel substitution information. The transmitting node WSC 110 indicates in the broadcast frame on the communications network which audio channel each receiving node of the network uses, namely the channel which is initially intended for them or an identified substitution channel. Thus each receiving node can update the table 202, as described below in connection with FIG. 5, with the information coming from the other receiving nodes of the network.

In step 409, the substitution channel selected in one of steps 405 to 407 is used for sound reproduction. In an alternative embodiment of this step 409, the channel substitution is not systematic. It is first verified that the substitution channel satisfies at least one predetermined condition (for example, if the audio data of the selected substitution channel does not have a large difference in signal amplitude with the audio data of the channel associated with the receiving node current). In case of positive verification, the audio data channel associated with said receiving node is substituted by said substitution channel.

In case of negative verification, the audio data channel initially intended for said receiving node is replaced by silence. It is also possible to replace the data of the audio data channel initially intended for said receiving node with silence, when the other audio channels available at the receiving node are not suitable for the substitution. For example, if an audio channel suitable for substitution is the channel initially intended for a receiving node which itself already substitutes its own audio channel with that initially intended for the considered receiving node, this new substitution would be inappropriate. channels in sound reproduction). It is then desirable to substitute the data of the audio channel initially intended for the considered receiver node with synthetic data representing a silence. This data may for example be predefined and stored in an internal memory unit 204 audio channel selection. The processing of the selected substitution channel is then continued in a conventional manner with the units 205a, 205b, 223, 206, 206a and 207.

In the preferred embodiment, the invention proposes for example to equalize the signal amplitude as well as a transition smoothing with the preceding data block (audio channel initially intended for the receiving node) by using a method of cross-linked, for example using the mechanisms described in patent document DE 3,638,922 already mentioned. For this purpose, the channel selection unit 204 adjusts a cross-fade control digital signal 223 which progressively goes from zero to one or from one to zero. This ensures a transition adapted to the sound reproduction. In a step 412, an inverse channel substitution is performed: the receiving node switches from a substitution channel (previously used) to the audio data channel initially intended for the current receiving node. The initial configuration (before possible substitutions), for the considered receiving node, is then found. In the same manner as for switching the audio data channel initially intended for the current receiver node to a substitution channel, the switching of a substitution channel to the audio data channel initially intended for the current receiving node can implement a equalization of the signal amplitude as well as transition smoothing, so as to ensure a smooth transition to sound reproduction. Step 411 makes it possible, before executing step 412, to memorize and broadcast information concerning the cessation of the substitution.

In particular, during step 411, the table 202 is updated by deleting the information on the use of the substitution channel previously used. This suppression is also broadcast to the other satellite speakers by radio using the transmission unit 203 and the antenna 221 (Figure 2). FIG. 5 illustrates an algorithm for the channel substitution information management protocol implemented by a node of the wireless multi-channel sound reproduction system of FIG. 1. This algorithm illustrates the various updating operations. of the table 202 by the control unit 201. In a first step 500, the unit 201 waits for the reception of the control messages by the radio channel from other receiving nodes (or the transmitter node WSC 110 in the case of the use of a centralized control protocol). Two cases can then be envisaged: on reception (step 501) of a message (from unit 204 selecting channels) informing the choice of a substitution channel by another node, the substitution channel and the corresponding node are stored in Table 202 (step 503); on reception (step 502) of a message informing of the stop of use of a substitution channel following the execution of step 411 of FIG. 4, the substitution channel and the corresponding node are erased of Table 202 (Step 504). In both cases, the processing performed by the unit 201 stops in a step 505 after one of the aforementioned steps 503 and 504 has been executed. FIG. 6 illustrates an implementation of a generic node 600 adapted to implement the methods according to a particular embodiment of the invention. This generic node 600 may in particular be connected to any means for storing images, videos or sounds that deliver to the generic node 600 multimedia data before forming an audiovisual data broadcasting source in the communications network. Thus, the generic node 600 comprises a communication bus 1002 to which are connected: a central processing unit 603 (which is for example a microprocessor referenced CPU); a read-only memory 604 (referenced ROM for Read Only Memory in English); a random access memory 606 (referenced RAM for Random Access Memory in English), functioning as a main memory, a work area, etc., of the central processing unit 603. The capacity of the RAM 606 can be extended by a Optional RAM connected to an extension port (not shown); a keyboard 610, a mouse 611 and a screen 608 optionally allowing a user to interact with the generic node 600; A communication interface 618 with the wireless communication network. The generic node optionally includes a hard disk 612 and a floppy disk drive (or other removable data storage medium) 614. The communication bus 602 allows communication between the different modules of the generic node 600. More generally, thanks to the communication bus 602, the central unit 603 is able to communicate instructions to any module included in the generic node 600 directly or through another module of the generic node 600.

After power-on, CPU 603 is able to execute instructions from RAM 606 in connection with a software application after these instructions have been loaded from, for example, program ROM 604 or 612 hard drive or a removable data storage medium. Such a software application, when it is executed by the central unit 603, causes all or some of the steps of the flowcharts illustrated previously in FIGS. 3 to 5 to be carried out. It should be noted that all the steps of FIGS. 5, including the various embodiments, can be implemented as a computer program. However, for performance reasons, it is advantageous to perform certain functions (such as decoding and error correction functions) by dedicated hardware.

Claims (20)

REVENDICATIONS1. A method of rendering, by a first receiving device of a sound broadcasting system comprising a plurality of receiving devices, an audio data channel from a set of at least three audio data channels received by said first receiving device, at each receiver device being initially intended for a given audio data channel of said set of at least three audio data channels, said method being characterized in that it comprises an audio data channel selection phase to be restored comprising the following steps the first detection (401) that a set of data of the audio data channel initially intended for said first receiver device is erroneous; in the case of a first positive detection, selection (405, 406, 407) of a substitution channel, based on information representative of the channels, of said set of at least three audio data channels, restored by the reception devices of said plurality of receiving devices. 2. Method according to claim 1, characterized in that said selection step comprises a verification substep that said substitution channel is not initially intended for a second receiver device which, by channel substitution, restores the channel initially. intended for said first receiving device. 3. Method according to claim 2, characterized in that, in the case of negative verification, the selected substitution channel is a synthetic audio data channel representative of a silence. 4. Method according to any one of claims 1 to 3, characterized in that said step of selecting a substitution channel, said first substitution channel, comprises the following steps: - ranking said set of at least three channels audio data according to a predetermined loop sequence; second detection that at least one other receiver device uses a second substitution channel corresponding to an audio data channel which is neighbor, in a determined direction of travel of said sequence, the audio data channel initially intended for said other receiving device; in the case of a second positive detection, selecting (405, 406) of said first substitution channel, taking an audio data channel which is neighbor, according to said determined direction of travel of said sequence, of the audio data channel initially intended for said first device receiver. 5. Method according to any one of claims 1 to 4, characterized in that said step of selecting said first substitution channel comprises the following steps: third detection (403) that no other receiver device uses a channel of substitution ; in the case of a third positive detection, selection (407) of said first substitution channel according to at least one predetermined criterion. 6. Method according to any one of claims 4 and 5, characterized in that, in the case of a second positive detection, said step of selecting said first substitution channel comprises the following steps: selecting (405, 406) a first an adjacent audio data channel, said current neighbor channel, which is neighbor, according to said determined direction of travel of said sequence, of the audio data channel initially intended for said receiver device; fourth detection that the data of said current neighbor channel are erroneous; in the case of a fourth positive detection, selecting a new neighbor audio data channel which is neighbor, according to said determined direction of travel of said sequence, of the current neighboring channel. 7. Method according to any one of claims 1 to 6, characterized in that said step of selecting a substitution channel is followed by a step of transmitting, in said sound broadcasting system, information (408). ) representative of the selection of the substitution channel performed by said first receiver device. 8. Method according to any one of claims 1 to 7, characterized in that it further comprises the following steps: fifth detection (410) a set of audio data of the audio data channel initially intended for said first receiving device is correctly received, in the case of a fifth positive detection, selection (412) of said audio data channel initially intended for the first receiver device. 9. Method according to claim 8, characterized in that in case of fifth positive detection, the method comprises a step (411) of transmission, in said voice broadcast system, information representative of a selection stop of the substitution channel performed by said first receiver device. Computer program product downloadable from a communication network and / or recorded on a computer-readable medium and / or executable by a processor, characterized in that it comprises program code instructions for the implementation of the Method according to at least one of Claims 1 to 9, when said program is executed on a computer. 11. Storage medium, possibly completely or partially removable, readable by a computer, storing a set of instructions executable by said computer to implement the method according to at least one of claims 1 to 9. A first receiver device of a sound broadcasting system comprising a plurality of receiver devices, said first receiver device for outputting an audio data channel from a set of at least three audio data channels received by said first device receiver, each receiving device being initially intended for a given audio data channel of said set of at least three audio data channels, said first receiving device being characterized in that it comprises: a first detection means that data set of the audio data channel initially intended for said first receiver device, is erroneous; means for selecting a substitution channel, as a function of information representative of the channels, of said set of at least three audio data channels, restored by the receiving devices of said plurality of receiving devices. 13. First receiving device according to claim 12, characterized in that it comprises means for verifying that said substitution channel is not initially intended for a second receiver device which, by channel substitution, restores the channel originally intended for auditing. first receiver device. 14. First receiving device according to claim 13, characterized in that it comprises means for selecting a synthetic audio data channel representative of a silence in case of negative verification. 15. The first receiver device according to any one of claims 12 to 14, characterized in that it comprises for the selection of a substitution channel, said first substitution channel: - means for classifying said set of at least three channels of audio data in a predetermined loop sequence; a second detection means that at least one other receiver device uses a second substitution channel corresponding to an audio data channel which is neighbor, in a determined direction of travel of said sequence, of the audio data channel initially intended for said other device receiver; means for selecting said first substitution channel, by taking an audio data channel which is neighbor, according to said determined direction of travel of said sequence, of the audio data channel initially intended for said first receiving device. 16. The first receiving device according to any one of claims 12 to 15, characterized in that it comprises: a third detection means that no other receiver device uses a substitution channel; means for selecting said first substitution channel according to at least one predetermined criterion. 17. First receiving device according to any one of claims 15 and 16, characterized in that it comprises, in case of second positive detection; - Means for selecting a first neighbor audio data channel, said current neighbor channel, which is adjacent, in said determined direction of travel of said sequence, the audio data channel initially intended for said receiving device; fourth means for detecting that the data of said current neighbor channel are erroneous means for selecting a new neighbor audio data channel which is neighbor, according to said determined direction of travel of said sequence, of the current neighbor channel; 18. First receiving device according to any one of claims 12 to 17, characterized in that it comprises means for transmitting, in said sound broadcasting system, information representative of the selection of the substitution channel performed by said first receiver device. The first receiver device according to any one of claims 12 to 18, characterized in that it comprises: a fifth detection means that a set of audio data of the audio data channel initially intended for said first receiving device is correctly received ; means for selecting said audio data channel initially intended for the first receiver device. 20. First receiving device according to claim 19, characterized in that it further comprises means for transmitting, in said sound broadcasting system, information representative of a selection stoppage of the substitution channel carried out by said first signal transmission system. receiving device.