FR2943195A1 - Data transmitting method for synchronous wireless personal area network, involves applying source coding instead of another source coding to assembly of useful data to be transmitted during predetermined duration when data loss is detected - Google Patents

Abstract

The method involves applying a source coding instead of another source coding to an assembly of useful data to be transmitted during a predetermined duration when data loss is detected. The former source coding and the predetermined duration are set such that sum between a quantity of data resulting from an encoding along the former source coding of the assembly of useful data and a quantity of data necessary to cure the detected loss is lower than or equal to a quantity of data resulted from encoding along the latter source coding of the assembly of useful data during detection of loss. Independent claims are also included for the following: (1) a computer program product having computer program code instructions for implementing a data transmitting method (2) a computer readable storage medium having a set of instructions for implementing a data transmitting method (3) a data transmitting device.

Description

A data transmission method with retransmission and synchronization mechanisms in case of loss, transmitting device, computer program product and corresponding storage means. FIELD OF THE INVENTION The field of the invention is that of systems for transmitting data from a transmitting device (also called a source node) to a receiving device (also called a destination node). The present invention applies in particular, but not exclusively, in the context of a personal wireless network (or WPAN for synchronous synchronous, low latency and using the 60GHz radio band). This type of network typically comprises a source node transmitting video data (for example uncompressed) to a destination node. 2. TECHNOLOGICAL BACKGROUND In the remainder of this document, we will focus more particularly on the problematic existing in the case of high definition content (HD) transmission in a WPAN network, a problem faced by the inventors. of the present patent application. The invention is of course not limited to this particular case of application, but is of interest for any data transmission technique from a transmitting device to a receiving device (the network may be wired or wireless, and the data may be be video or other data) faced with a similar or similar problem (retransmission of lost data while managing synchronization between transmitter and receiver devices). Thanks to the latest technological advances from manufacturers, domestic audio / video applications have multiplied and become more popular in recent years.

Thus manufacturers now offer very attractive home-cinema systems offering High Definition video (HD) and multiple audio channels. The HDMI interface (for High Definition Multimedia Interface in English) is one of the technologies that has allowed manufacturers to offer High Definition (HD) audio / video systems. This wired interface offers a bandwidth of several Gbps and thus allows the transmission of uncompressed High Definition (HD) content between different audio / video equipment. The current challenge for many industrialists is to offer wireless transmission technology equivalent to the HDMI interface.

A WPAN network using the 60GHz radio band is particularly suitable for constituting a wireless transmission technology of uncompressed or weakly compressed HD contents. The 60GHz radio band is characterized by a large bandwidth and a high rate of attenuation of the signal in the air and through obstacles (walls, furniture, living beings). In addition, in some countries, this radio band is subject to strong regulation by the authorities which limits the power of emission (for example to 10W in the United States of America). As a result, the 60GHz radio band is well suited for very high speed data transmission within a limited radius. Advantageously, the source nodes and destination of the WPAN network at 60 GHz use agile antennas (also called smart antennas, or smart antennas in English) and the Beam Forming technique (beam formation). An agile antenna consists of a network of radiating elements distributed matrixically on a support. This type of antenna allows the implementation of the so-called Beam Forming technique, according to which each radiating element of the antenna is electronically controlled in phase and power to obtain a transmitting beam and / or receiving directional and more or less narrow. Thus, this type of antenna makes it possible to increase the transmission and / or reception gains in a given direction and to filter certain undesired directions in transmission and / or reception. In practice, the use of this type of antenna in a WPAN network allows both the reception of signals transmitted in direct line (LoS, for Line Of Sight in English) and signals reflected on obstacles (walls, furniture, reflectors ...). The 60GHz radio channel is a particularly random transmission channel. In addition to sensitivity to noise and interference inherent in any radio channel, the 60GHz radio channel is very sensitive to masking. Thus, a human passing through the direct view path (LoS) used for communication between a source node and a destination node will momentarily degrade the quality of transmission on the radio channel. To overcome the problems of degradation of the 60GHz radio channel, it is therefore necessary to implement specific mechanisms to improve the transmission of HD content. US20080253462A1 discloses a method and system for transmitting video data in a wireless system using the 60GHz radio band. In this patent application US20080253462A1, the video data comprising a first number of bits / pixel (that is to say resulting from a first source coding) are transmitted by a transmitting node using a first channel modulation and coding scheme. . The characteristics of the radio channel are exchanged between the receiving node and the transmitting node via a return channel. A dedicated module of the transmitter node analyzes the characteristics of the radio channel (BER (for Bit Error Rate in English), SNR (for Signal Noise Ratio in English, or Signal to Noise Ratio in French), request for retransmission ...). When a significant degradation of the radio channel is identified, the video data including a second number of bits / pixel (i.e. resulting from a second source coding) are then transmitted by the sending node using a second scheme. modulation and channel coding. The first number of bits / pixel is greater than the second number of bits / pixel. The second channel modulation and coding scheme provides a higher degree of disturbance robustness than the first channel modulation and coding scheme. The patent application US20080253462A1 therefore proposes to adapt the source coding together with the channel modulation and coding scheme according to the characteristics of the 60GHz radio channel. As the radio channel degrades, the sending node decides to switch to a second channel modulation and coding scheme that adds redundancy to make the transmission more robust to interference. In return, the bandwidth and the transmission rate on the radio channel being constant, the transmitting node must apply a second source code using fewer bits / pixel. The patent application US20080253462A1 therefore offers a solution for ensuring the quality of video data transmission when the 60GHz radio channel is degraded. In this document, it is mentioned that the dedicated module for analyzing the characteristics of the radio channel can identify the degradation of the radio channel, in particular via a radio packet retransmission request. However, this patent application does not specify a method for managing the retransmission of the lost radio packet (s) or method for providing video synchronization at the receiver. During a rapid degradation of the radio channel, it may indeed happen that one or more radio packets are not received correctly by the receiving node. In addition to adapting the transmission mode to the new conditions on the radio channel, it would be interesting for the sending node to also manage the retransmission of the lost radio packet or packets, while ensuring video synchronization at the receiver. It is particularly important in systems for transmitting audio and / or video data to ensure that the timing of the data at the consumer application level is maintained. In this sense, synchronization of the audio and / or video data at the receiver must be ensured. 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. More specifically, in at least one embodiment of the invention, an objective is to provide a technique for transmitting data from a transmitting device to a receiving device, making it possible, on detection of a data loss, for the transmission of data. data transmitter for correcting the detected loss, while ensuring synchronization of data at the receiving device. At least one embodiment of the invention also aims to provide such a technique, which is simple to implement and inexpensive. 4. DISCLOSURE OF THE INVENTION In a particular embodiment of the invention, there is provided a method for transmitting useful data, from a sending device to a receiving device, the transmitting device applying to said payload data, before transmission, a first source coding and then a first modulation scheme and channel coding. Following a detection of a data loss, the transmitting device performs a step of applying a second source code, in place of the first source code, to a set E of useful data to be transmitted for a first predetermined duration, the second source coding and the first predetermined duration being such that the sum between: a quantity of data resulting from an encoding according to the second source code of said set of useful data, and a quantity of data necessary to remedy said loss, is less than or equal to a quantity of data which would have resulted from an encoding according to the first source coding of said set E of useful data, if said loss had not been detected. The general principle of the invention therefore consists, in the event of detected loss, in making, during the first predetermined duration, a catch-up phase, making it possible to temporarily increase the speed of transmission of the useful data (in their state before application of the coding source and modulation scheme and channel coding). Thus, during the first predetermined period of time, the transmitting device can transmit not only the set E of useful data which it was intended to transmit during this first predetermined duration, but also data necessary to remedy the loss. Subsequently, two cases are detailed for this data necessary to remedy the loss: in one case, it is encoded data resulting from an encoding, with the second source coding, of useful data which has not could not be transmitted normally (for a second predetermined duration (detailed below) because of the retransmission of the lost useful data), and - in the other case, it is encoded data resulting from an encoding, with the second source coding of useful data corresponding to a retransmission of the lost data to which the loss corresponds or to data, such as redundancy data, making it possible to remedy the loss of the data. This particular embodiment of the invention is therefore based on a completely new and inventive approach allowing, on detection of a loss of data, the transmission by the transmitter of data intended to remedy the loss of data, while by synchronizing the data at the receiving device. Advantageously, following the detection of said data loss, the transmitting device performs a step of applying a second modulation scheme and channel coding for a second predetermined duration which precedes said first predetermined duration, said second modulation scheme and channel coding providing a more robust channel coding than said first channel modulation and coding scheme. Thus, before the first predetermined duration (that is to say before the catch-up phase), the transmitting device performs, during the second predetermined duration, a step making it possible to adapt the channel modulation and coding scheme to the new conditions of the radio channel . Indeed, we switch to the second modulation scheme and channel coding that makes the transmission more robust. Advantageously, following the detection of said data loss, the transmitting device applies: said second source coding and said second channel modulation and coding scheme, during said second predetermined duration; then - said second source coding and said first channel modulation and coding scheme during said first predetermined duration. Thus, it simplifies the processing since it is the same second source coding that is applied during the second and first predetermined periods. In summary, in this particular case, two phases of failover can be distinguished. A first switching phase on the second source coding (cs # 2) and the second channel modulation and coding scheme (sm / cc # 2), which makes it possible to adapt the channel modulation and coding scheme to the new conditions of the radio channel and adapt the source coding accordingly. The radio packet loss is an indicator of the (momentary or longer term) degradation of the radio channel. The sending device then decides to switch to the second channel modulation and coding scheme (sm / cc # 2) which adds redundancy to make the transmission more robust. In return, if the bandwidth and the transmission rate on the radio channel are constant, the transmitting device must apply the source coding (cs # 2) using less bits / pixel. The first switching phase is done over a duration t1 (second predetermined duration), which can be arbitrary or determined according to the evolution of the conditions on the transmission channel, but at least greater than or equal to the time required to retransmit the transmission. or lost radio packets using cs # 2 and sm / cc # 2 mode. A second switching phase on the second source coding (cs # 2) and the first channel modulation and coding scheme (sm / cc # 1), which temporarily increases the transmission speed of the useful data, to recover the time lost when retransmitting the lost radio packet (s) (that is, during the first failover phase). After the duration t1 has elapsed, the transmitter device estimates that the transmission conditions on the radio channel have improved. The transmitter device therefore reverts to the first channel modulation and coding scheme (sm / cc # 1) which adds less redundancy to the transmitted data. The second switching phase is done over a period t2 (first predetermined duration), during which the transmitting device continues to use the second source code (cs # 2) to temporarily transmit more useful data. This second switching phase therefore makes it possible to rebalance the buffers (also known as buffers) for transmitting Tx and receiving Rx respectively at the sending device and the receiving device, and thus for returning to initial transmission conditions. Returning to these initial transmission conditions will absorb any new packet loss and thus ensure synchronization. The duration t2 corresponds, for example, to the transmission time necessary to transmit the delayed radio packet or packets (by the retransmission or retransmissions) using the mode cs # 2 and sm / cc # 1. The duration t2 is a function of the number of delayed packets and the ratios cs # l / cs # 2 and (sm / cc # l) / (sm / cc # 2). Advantageously, the lost useful data to which said loss corresponds are retransmitted during said second predetermined duration. In addition, the data necessary to remedy the loss, which are transmitted during said first predetermined duration, are coded data resulting from an encoding with the second source coding useful data that could not be transmitted during said second duration predetermined because of the retransmission of lost useful data.

Thus, the lost data is first retransmitted during the second predetermined duration (t1) and then the delayed packets are transmitted during the first predetermined duration (t2). In this way, the order of transmission of the data is preserved, which facilitates the processing carried out by the transmitter and receiver devices. In addition, the processing performed by the transmitter and receiver devices is also facilitated by the fact that the second source encoding is applied to obtain the data necessary to remedy the loss. Advantageously, said first predetermined duration is determined as a function of a first ratio between: a first quantity of coded data, resulting from an encoding with the first source coding of a given quantity of data, and a second quantity encoded data, resulting from encoding with the second source encoding of said given amount of data. Thus, for a given (and fixed) bandwidth allocated to the communication of said user data from the transmitting device to the receiving device, the first predetermined duration (t2) is easy to determine.

Advantageously, the second predetermined duration corresponds to: a time slot for transmitting a predetermined number of useful data packets from the receiving device to the receiving device; or - an interval of time during which transmission conditions between the transmitting device and the receiving device are estimated such that data loss would occur if the first source coding was applied. Thus, the second predetermined duration (t 1) is easy to determine. Advantageously, said predetermined number of payload packets is determined as a function of a number of negative acknowledgments from the receiving device for said payload transmission. For example, the predetermined number of payload packets is proportional to the number of negative acknowledgments received by the transmitting device (from the receiving device). In an advantageous variant, following the detection of said data loss, the transmitting device applies said second source coding and said first modulation scheme and channel coding, during said first predetermined duration.

In this variant, the first channel modulation and coding scheme (sm / cc # 1) is retained for the first predetermined duration. There is therefore no second predetermined duration, prior to the first duration. This variant therefore makes it possible to rebalance the buffers (also known as buffers) of transmission Tx and reception Rx more rapidly than with the embodiment comprising the second predetermined duration (tl). On the other hand, it is based on the assumption that the channel does not remain degraded. If this assumption turns out to be false, and if the data transmitted during the first predetermined duration are again badly received, it suffices to reiterate the mechanism of this variant. In this case, for another first predetermined duration, the first channel modulation and coding scheme (sm / cc # 1) is retained, but a third source coding (cs # 3) whose compression ratio is higher than that is applied. that of the second source coding (cs # 2). Advantageously, the data needed to remedy the loss, which are transmitted during said first predetermined duration, are encoded data resulting from an encoding, with the second source coding, of useful data corresponding to a retransmission of the lost payload data to which said data corresponds. loss. The processing performed by the transmitter and receiver devices is facilitated by the fact that the second source encoding is applied to obtain the data necessary to remedy the loss.

Advantageously, following the detection of said data loss, the transmitting device performs a step of modifying parameters of a transmitting antenna of the transmitting device, for said transmission of useful data. Thus, it is anticipated a possible masking of a main radio communication path, of a consequent duration, by deciding to switch to a secondary radio communication path, for the transmission of useful data. Advantageously, said transmission of useful data is performed in a network comprising said transmitting device and said receiving device and implementing a timing according to a transmission cycle comprising: a first part, dedicated to a first mode of communication, for the exchange of data signaling data between said transmitting device and said receiving device; and a second part, dedicated to a second mode of communication, for said transmission of useful data from the transmitting device to the receiving device. The detection of said data loss comprises a step of reception, by the transmitting device, of at least a negative acknowledgment, transmitted by the receiving device during the first part of a current cycle, for a payload packet previously transmitted by the transmitting device during the second part of a previous cycle. Thus, the transmission of the acknowledgments (which are part of the signaling data) is guaranteed, regardless of the state of the channel for the transmission of the useful data. In another embodiment, the invention 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. This computer program product includes program code instructions for carrying out the aforesaid method (in any one of its various embodiments), when said program is run on a computer. In another embodiment, the invention relates to a computer readable storage means storing a computer program comprising a set of computer executable instructions for carrying out the above method (in any one of its different embodiments). In another embodiment, the invention relates to a transmitting device for transmitting useful data to a receiving device, the transmitting device applying to said payload, before transmission, a first source coding and a first modulation scheme and channel coding. The transmitting device comprises means, activated following a detection of a loss of data, making it possible to apply a second source coding, in place of the first source coding, to a set E of useful data to be transmitted during a first predetermined duration. the second source coding and the first predetermined duration being such that the sum between: - a quantity of data resulting from an encoding according to the second source code of said set E of useful data, and - a quantity of data necessary to remedy said loss, is less than or equal to a quantity of data that would have resulted from an encoding according to the first source coding of said set E of useful data, if said loss had not been detected. Advantageously, the transmitting device comprises means, activated following the detection of said data loss, making it possible to apply a second channel modulation and coding scheme for a second predetermined duration which precedes said first predetermined duration, said second modulation scheme and channel coding providing a more robust channel coding than said first channel modulation and coding scheme.

In a particular embodiment, the transmitting device comprises means, activated following the detection of said loss of data, making it possible to apply: said second source coding and said second channel modulation and coding scheme, during said second predetermined duration ; then - said second source coding and said first channel modulation and coding scheme during said first predetermined duration. The characteristics which the transmitting device possesses in at least one of the various possible embodiments of the invention are described below. Advantageously, the lost useful data to which said loss corresponds are retransmitted during said second predetermined duration. In addition, the data necessary to remedy the loss, which are transmitted during said first predetermined duration, are coded data resulting from an encoding with the second source coding useful data that could not be transmitted during said second duration predetermined because of the retransmission of lost useful data. Advantageously, said first predetermined duration is determined as a function of a first ratio between: a first quantity of coded data, resulting from an encoding with the first source coding of a given quantity of data, and a second quantity encoded data, resulting from encoding with the second source encoding of said given amount of data.

Advantageously, the second predetermined duration corresponds to: a time slot for transmitting a predetermined number of useful data packets from the receiving device to the receiving device; or - an interval of time during which transmission conditions between the transmitting device and the receiving device are estimated such that data loss would occur if the first source coding was applied. Advantageously, said predetermined number of payload packets is determined as a function of a number of negative acknowledgments from the receiving device for said payload transmission.

In an advantageous variant, the transmitting device comprises means, activated following the detection of said data loss, for applying said second source coding and said first modulation scheme and channel coding during said first predetermined duration. Advantageously, the data needed to remedy the loss, which are transmitted during said first predetermined duration, are encoded data resulting from an encoding, with the second source coding, of useful data corresponding to a retransmission of the lost payload data to which said data corresponds. loss. Advantageously, the transmitting device comprises means, activated following the detection of said loss of data, making it possible to modify parameters of a transmitting antenna of the transmitting device, for said transmission of useful data. Advantageously, said transmission of useful data is performed in a network comprising said transmitting device and said receiving device and implementing a timing according to a transmission cycle comprising: a first part, dedicated to a first mode of communication, for the exchange of data signaling data between said transmitting device and said receiving device; and a second part, dedicated to a second mode of communication, for said transmission of useful data from the transmitting device to the receiving device. In addition, for the detection of said data loss, the transmitting device comprises receiving means, for receiving at least one negative acknowledgment, transmitted by the receiving device during the first part of a current cycle, for a packet useful data previously transmitted by the transmitting device during the second part of a previous cycle. 5. LIST OF FIGURES Other features and advantages 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. first mode of communication between two nodes (one transmitter, the other receiver) of a WPAN network, in a particular embodiment of the invention; FIG. 2 illustrates a second mode of communication between the two nodes of FIG. 1; FIG. 3 describes a TDMA cycle used by the two nodes of FIGS. 1 and 2 to access the WPAN network and to communicate according to the first and second communication modes; FIG. 4 represents the functional block diagram of a node of the WPAN network, in a particular embodiment of the invention; FIG. 5 represents an example of data exchange between the transmitter / receiver nodes at the start of a data transmission, in a particular embodiment of the invention; FIG. 6 represents an example of data exchange between the transmitter / receiver nodes during retransmission of data, in a particular embodiment of the invention; FIG. 7a represents an example of data exchange between the transmitter / receiver nodes during a retransmission of data accompanied by a radio path change, in a particular embodiment of the invention; FIG. 7b, which corresponds to the continuation of FIG. 7a, represents an example of data exchange between the transmitter / receiver nodes during the return in the normal transmission conditions after retransmission of data and radio path change, in a particular embodiment of the invention; FIG. 8a is a representation, over several TDMA cycles, of the same data exchange as that represented in FIG. 6; FIG. 8b is a representation, over several TDMA cycles, of a first data exchange between the transmitter / receiver nodes, in an alternative embodiment of the invention; FIG. 8c is a representation, over several TDMA cycles, of a second data exchange between the transmitter / receiver nodes, in said variant embodiment of the invention; and FIG. 9 presents a flowchart of a particular embodiment of the method according to the invention, implemented in the transmitting device. 6. DETAILED DESCRIPTION In all the figures of this document, the elements and identical steps are designated by the same numerical reference. FIG. 1 illustrates a first mode of communication between two nodes (one transmitter 11, the other receiver 12) of a WPAN network, in a particular embodiment of the invention. In this first mode of communication, the nodes 11 and 12 exchange synchronization information and control and signaling information relating to the second communication mode (start / stop information), positive acknowledgment information. or negative (Ack / nAck), change of antenna parameters ...). This first mode of communication is characterized for example by a bidirectional transmission, a relatively low data rate (10-20Mbps) and a strong transmission robustness thanks in particular to the repetition of the transmitted data (for example three times) and a transmission scheme. channel coding comprising frequency multiplexing of OFDM type (for Orthogonal Frequency Division Multiplex in English, or for orthogonal frequency division multiplexing in French) and for protocol data robust binary coding for example according to a BPSK (for Binary Phase Shift) modulation English-language keying or binary phase shift keying), for encoding one bit per symbol and inserting redundancy data according to a convolutional code, for example Viterbi type with a ratio of 2/3. In this first mode of communication, the agile antennas of the nodes 11, 12 are advantageously parameterized to obtain relatively broad transmission / reception beams 13, 14. This configuration makes it possible to avoid the total interruption of the transmission when a mobile object (partial masking) interposes between the two nodes. For example, when an obstacle is positioned on the direct view path 15 between the nodes 11 and 12, the communication can always be done via the indirect path 16 where the radio signals are reflected on a surface 17 (wall, furniture .. .).

FIG. 2 represents the same two nodes 11, 12 of the WPAN network 10 when they are in the second communication mode. In this second mode of communication, the transmitting node 11 transmits HD video data to the receiving node 12. In normal operation, this data is not compressed.

This second mode of communication is characterized for example by a unidirectional transmission and a very high data rate (> 3Gbps). This very high data rate is achieved in particular by means of a channel coding scheme comprising an OFDM frequency multiplexing and a binary coding allowing a better bit rate (than the BPSK type coding), for example using a QAM modulation (for Quadrature Amplitude Modulation in English, or 16QAM, for encoding 4 bits per symbol and allowing the insertion of redundancy data according to a convolutional code, for example of Viterbi type with a 2/3 ratio. In this second mode of communication, the agile antennas of the transmitter nodes 11 and receiver 12 are advantageously set to obtain very thin transmission (18, 18 ') and reception (19, 19') beams. This antenna configuration makes it possible to concentrate the energy in a precise direction and thus to improve the signal-to-noise ratio (SNR) at the receiving node 12. FIG. 2 presents two radio path alternatives for the transmission of data from the transmitting node 11 to the receiving node 12: • by setting their respective antennas to obtain the transmit and transmit beams 18 and 19, the transmitting node can transmit data to the receiving node via the direct view path 15; By setting their respective antennas to obtain the transmit beam 18 'and receive beam 19', the transmitting node can transmit data to the receiver node via the indirect path 16 (where the radio signals are reflected on the surface 17). It should be noted that the agile antennas, in this second mode of communication, can be parameterized so that they form several beams simultaneously, thus allowing the formation of several simultaneous paths. In order to initiate a communication in this second mode of communication, the nodes 11 and 12 must identify the different possible radio paths as well as the associated antenna parameters. This identification is done during a discovery phase, for example during the first start of the WPAN network or at the request of the user. For example, during this discovery phase, the transmitting node 11 transmits a test signal of a predetermined duration for the different orientations of the transmission beam of its agile antenna. During this time, the receiving node scans the different orientations of the receive beam of its receiving antenna and performs a measurement of the level and / or quality of the signal received for each orientation. At the end of this discovery phase, the nodes 11 and 12 have identified and classified in a table the different possible radio paths with their associated antenna parameters. The classification may for example be based on a received signal level (RSSI for Received Signal Strength Indication in English) or a bit error rate (BER). FIG. 3 describes a TDMA cycle used by the two nodes of FIGS. 1 and 2 to access the WPAN network and to communicate according to the first and second communication modes. TDMA access mode (Time Division Multiple Access in English) is a multiplexing mode for transmitting several signals on a single radio channel. It is about time multiplexing, whose principle is to cut the available time and to establish a distribution between the different emitters. According to this technique, each node is assigned one or more time slots to transmit and / or receive. This time cutout is cyclically reproduced according to a time period called the TDMA cycle. The implementation of a TDMA mode implies a perfect synchronization between the different nodes of the network. This synchronization is done for example with respect to a radio packet transmitted at the beginning of the TDMA cycle by a particular node often called master (master in English). A complete TDMA cycle is a Super Frame. FIG. 3 shows the Super Frame (k) 24 and the beginning of the following Super Frame (k + 1). The duration of the Super Frame can for example be lms. As can be seen in FIG. 3, a period 26 (relatively short, for example of the order of 1 / 10th of the total duration of a TDMA cycle) at the beginning of the Super Frame is allocated to the first mode of communication. , the continuation (period 27) of the Super Frame is entirely devoted to the second mode of communication. In the period 26 dedicated to the first mode of communication, the transmitting node 11, which in the case described is also the master node, transmits its signaling data (synchronization, control, etc.) during the time slot (also called time interval or Time Slot in English) in the form of at least one radio packet. As mentioned above, the radio packets transmitted by the sending node 11 constitute the time reference of the WPAN network 10. The receiving node 12 maintains its synchronization with respect to the transmitting node 11, on reception of the transmitted radio packets in the period 26. During this period 26, the transmitting node 11 can transmit several times the same signaling data. It is then possible for the receiving node to receive several times the same signaling data, possibly using different antenna configurations for receiving radio packets carrying the same data. In this time slot 20, the transmitter transmits in particular the synchronization information of the video stream (data or signal called Video Sync), as signaling data.

Once this synchronization is established, the receiving node is able to determine its own time slot 21 for the transmission of its own radio packets. The next time slot 21 is allocated to the receiving node 12 for the transmission of its different signaling data (Ack / nAck, antenna parameters, etc.). For example, the receiving node 12 transmits its various signaling data three times. In this time slot 21, the receiving node 12 transmits notably the acknowledgment (Ack) or the non-acknowledgment (nAck) of the radio packet of the second communication mode of the Super Frame (k-1).

In this period 26 dedicated to the first mode of communication, it is possible to imagine the interleaving of the radio packets generated by the nodes of the system in order to reinforce the robustness of this first mode of communication. In addition, other time slots may possibly be assigned to other nodes present or joining the WPAN network 10. These nodes will also contribute to reinforcing the robustness of the first mode of communication, by repeating the data transmitted by the nodes 11, 12 and thus offering diversity (spatial, in addition to time) and increased redundancy. In the period 27 dedicated to the second communication mode, the sending node 11 uses the time slot 22 to transmit the HD video data to the receiving node 12. This period 27 occupies the largest part (for example of the order of 9 / 101th of the total duration of a TDMA cycle) of the Super Frame since it requires a high bandwidth to reach the bit rates required by the application (> 3Gbps), especially when these data are transmitted in an uncompressed way. FIG. 4 represents the functional block diagram 40 of a node of the WPAN network (that is to say one of the nodes 11 and 12 of FIGS. 1 and 2) and its interconnection with video equipment 30, in a mode particular embodiment of the invention. The video equipment 30 is connected to the node 40 of the WPAN network via a wired interface 31 of the HDMI type. The video equipment 30 can be an HD video source (Blu-ray player (registered trademark), Set top box HD ...) or HD display equipment (LCD HD screen, HD video projector ...).

Typically, the video equipment 30 supports uncompressed HD video resolutions up to 1080p 60Hz (1080 lines x 1920 pixels @ 60Hz). The data rate corresponding to this resolution is almost 3 Gbps when the three video components are sampled at the same frequency and quantized on 8 bits (4: 4: 4). The interconnection of the video equipment 30 with the node 40 via the wired interface HDMI 31 can transport this rate without resorting to compression of video data. The block 41 of the node 40 is connected to the HDMI interface 31. The block 41 performs the source coding / decoding functions necessary for the implementation of the present invention. When the node 40 is a transmitting node (11), the block 41 performs the source coding function. When the node 40 is a receiving node (12), the block 41 performs the source decoding function.

Block 41 uses buffer 42 (also called a buffer memory) to provide the transmit (Tx) or receive (Rx) buffering function necessary for carrying out the present invention. Thus, the block 41 must constantly buffer (that is to say put in the buffer 42) a number of video lines to ensure retransmission of radio packets and maintain video synchronization.

The number of buffered video lines in the buffer 42 is a multiple of the number of video lines transmitted during a Super Frame of lms. In the case of 1080p 50Hz video, 54 lines are transmitted during a Super Frame of lms. Block 41 must therefore buffer n * 54 lines in buffer 42, n being a function of the maximum latency allowed by the WPAN network and the retransmission capacity chosen. For example, if n = 5, the buffer 42 should be able to store 11.86Mbits (1080p 50Hz (4: 4: 4)). In the remainder of this document, the 54 video lines transmitted during a Super Frame will be called the Video Stream Portion (PFV). The source coding / decoding applied by the block 41 can be configured by the CPU (for Central Processing Unit in English) 47 via the address / data bus 50. Thus, depending on the conditions of the transmission over the radio channel, the CPU will manage the switching of block 41 from a first source coding (cs # 1) to a second source coding (cs # 2). The first source coding (cs # 1) can for example correspond to video data of 1080p 50Hz resolution for which the three video components are sampled at the same frequency and quantized on 8bits (4: 4: 4). In the case of encoding cs # l, 46080 bits / line will then be transmitted. The second source coding (cs # 2) may for example correspond to video data of 1080p 50Hz resolution for which the chrominance components are downsampled by a factor of 2 relative to the luminance Y (4: 2: 0). In the case of the cs # 2 encoding, 23040 bits / line will be transmitted. In this example, for the same image transmitted over the radio link, the cs # 2 source encoding will use half as much bandwidth as the cs # 1 encoding. The following relation can therefore be written: R (cs # 1) / (cs # 2) = 2 Other types of second source encoding are possible, such as, for example, a source encoding of applying a data compression algorithm. The source coding / decoding block 41 is connected to the MAC block 43. The MAC block 43 performs packet packetization / reception functions necessary for the present invention.

When the node 40 is a transmitting node (11), the block 43 retrieves the video data after the source coding performed by the block 41. The block 43 then performs the packetization of the video data to be transmitted according to the second communication mode. This packaging essentially consists of grouping the video lines to be transmitted in the next Super Frame within one or more radio packets. The number of radio packets used for the second communication mode is determined, for example, from signaling data transmitted according to the first communication mode. When the node 40 is a transmitting node (11), the block 43 also performs the recovery / packetization function of the data relating to the first communication mode. Thus, the block 43 retrieves the signaling data packets transmitted by the receiving node according to the first communication mode. Block 43 then analyzes the different signaling data (Ack / nAck, antenna parameters, etc.) coming from the receiving node. Block 43 also carries out the packetization function of the signaling information (commands or signals of Start / stop type, Video Sync, number of radio packets used for the second communication mode, etc.) transmitted according to the first mode of communication. to the receiving node 12. When the node 40 is a receiving node (12), the block 43 performs the function of recovering the video data packets transmitted by the transmitting node according to the second communication mode. When the video data packets are correctly received, the block 43 generates the acknowledgment corresponding to the number of the packet and transmits the data of the corresponding video lines to the source decoding block 41. When the video data packets are not correctly received, the block 43 generates the non-acknowledgment corresponding to the number of the packet. In this case, no new video data is transmitted to the source decoding block 41. When the node 40 is a receiving node (12), the block 43 also performs the recovery / packetization function of the signaling data relating to the first mode of communication. Thus, the block 43 retrieves the signaling data packets transmitted by the transmitting node according to the first communication mode.

Block 43 then analyzes the different signaling data (synchronization, start / stop ...) from the transmitter node. Block 43 also carries out the packetization function of the signaling information (Ack / nAck, antenna parameters, etc.) transmitted according to the first communication mode to the sending node. The MAC block 43 is also connected to the CPU block 47 via the address / data bus 50. When the node 40 is a transmitting node (11), this connection makes it possible in particular for the MAC block 43 to alert the CPU block 47 when a non-acknowledgment is received or when a change of antenna parameter is requested. According to the alert received from the MAC block 43, the CPU block 47 can then manage the configurations of the blocks 41, 44 and 45.

The MAC block 43 is connected to the block 44. The block 44 performs the channel coding / decoding and modulation / demodulation functions necessary for the present invention. When the node 40 is a transmitting node (11), the block 44 performs the channel and modulation coding functions. When the node 40 is a receiving node (12), the block 44 performs the functions of channel demodulation and decoding. The signaling data transmitted according to the first communication mode are always treated in the same way by block 44. For example, block 44 still uses convolutional coding 2/3, primary modulation of BPSK type and OFDM secondary modulation on 128 carriers to transmit these signaling data. This channel coding is well suited to the first communication mode of the WPAN network because it offers great robustness to the noise and interference present on the radio channel. The channel coding applied by the block 44 to the video data transmitted according to the second communication mode can be configured by the CPU 47 via the address / data bus 50. Thus, depending on the transmission conditions on the radio channel, the CPU will handle the switching of the block 44 from one channel coding to another, for example from a first channel coding (sm / cc # 1) to a second channel coding (sm / cc # 2). The first channel coding (sm / cc # 1) may for example use 2/3 convolutional code, 16-QAM type primary modulation and 512 carrier-based OFDM secondary modulation. An OFDM symbol resulting from the sm / cc # 1 channel modulation and coding scheme will be composed of 112 octets corresponding to 96 octets of useful data, 16 octets of redundancy, for example determined from algebraic code of Reed Solomon type. The second channel coding (sm / cc # 2) may for example use a 1/3 convolutional code, a QAM 16 type primary modulation and a 512 carrier OFDM secondary modulation. An OFDM symbol resulting from the sm / cc # 2 channel coding will be composed of 56 bytes corresponding to 48 bytes of useful data, 8 bytes of redundancy for example determined from the same algebraic code of the Reed Solomon type as previously.

Thus a data field of 96 octets (eg consisting of video data), will be transmitted by a single OFDM symbol when the first channel coding (sm / cc # 1) is used. This same 96-byte data field will be transmitted by two OFDM symbols when the second channel coding (sm / cc # 2) is used The sm / cc # 2 channel coding provides a higher degree of data protection than the sm / dc channel coding #l, adding redundancy. This increase in the degree of protection is to the detriment of the number of useful data (i.e., in this context, the number of data before application of the modulation scheme and channel coding) transmitted. Thus, the payload data rate transmitted using the sm / cc # 2 channel modulation and coding scheme will be two times lower than the payload data rate transmitted using the sm / cc # 1 channel modulation and coding scheme. The following relation can therefore be written: R (sm / cc # 1) / (sm / cc # 2) _ Msm # 1. Rcc # 1 4.2 / 3 = 2 Msm # 2.Rcc # 2, 4.1 / 3, Where : Rcc # 1 represents the Convolutional Code ratio of the sm / cc # 1 scheme; Rcc # 2 represents the ratio of the Convolutional Code of the sm / cc # 2 schema; Msm # 1 represents the number of bits / symbol in the modulation according to the sm / cc # 1 scheme; Msm # 2 represents the number of bits / symbol in the modulation according to the sm / cc # 2 scheme; The OFDM symbols resulting from the sm / cc # 1 and sm / cc # 2 channel encodings have the same transmission duration. Whatever the channel coding, the same number of OFDM symbols can therefore be transmitted in a Super Frame. Thus, for a 50Hz 1080p video, using the cs # 1 source coding and the sm / cc # l channel coding, 3240 OFDM symbols will be required to transmit the 54 lines of an image (= lms of video) in a Super Frame . In the example illustrating the present invention, the association cs # 1 and sm / cc # 1 is the configuration used by default to transmit the video data of the transmitting node 11 to the receiving node 12 according to the second communication mode.

For 1080p 50Hz video, using cs # 2 source coding and sm / cc # 2 channel coding, 3240 OFDM symbols will also be required to transmit the 54 lines of an image (= lms of video) in a Super Frame. The combination of the source coding cs # 2 and the sm / cc # 2 channel coding is advantageously used in the present embodiment to (re) transmit the video data from the transmitting node 11 to the receiving node 12 when the transmission conditions on the radio channel are degraded. This degradation of the radio channel can for example be identified by the reception of non-acknowledgment (nAck). In this configuration, channel coding makes it possible to add redundancy to the data to make the communication more robust to noise and interference and to adjust the source coding accordingly. The configuration cs # 2 and sm / cc # 2 is applied for a time t1 which can be predefined, more particularly so as to be greater than or equal to the time necessary to retransmit the lost radio packet (s).

Still for a 50Hz 1080p video, using the cs # 2 source coding and sm / cc # l channel coding, the transmission of 3240 OFDM symbols will transmit 2 * 54 lines of an image (= 2ms of video) into a Super frame. The combination of the source coding cs # 2 and the coding channel sm / cc # 1 advantageously makes it possible to temporarily increase the speed of transmission of the video data in order to recover the time lost during the retransmission of the lost radio packet (s). After the duration t1 has elapsed, the transmitting node estimates that the transmission conditions on the radio channel have improved. The transmitting node therefore reverts to the sm / cc # 1 modulation and coding scheme, which adds less redundancy to the transmitted data. During a period t2, the transmitter node will then continue to use the source code cs # 2 to temporarily transmit more video data. This configuration cs # 2 and sm / cc # 1 applied during the duration t2 thus allows to rebalance the video buffers Tx and Rx, respectively at the transmitter and the receiver, and thus to return to the conditions of initial transmissions. Returning to these initial transmission conditions will absorb any new radio packet loss and thus ensure video synchronization. The duration t2 corresponds to the transmission time required to transmit the delayed radio packets (by the retransmission or retransmissions) using the mode cs # 2 and sm / cc # l. The duration t2 is a function of the number of radio packets lost and the ratios cs # 1 / cs # 2 and (sm / cc # l) / (sm / cc # 2). In this description of a particular embodiment of the present invention, the association of the source coding cs # 1 and the coding channel sm / cc # 1 on the one hand, and the source coding cs # 2 and sm / cc channel coding # 2 on the other hand, are advantageously chosen so as to always transmit 1 PFV (54 lines of an image) per TDMA cycle. This results in equal ratios of cs # l / cs # 2 source coding and channel modulation and coding schemes (sm / cc # l) / (sm / cc # 2): R (cs # 1) / (cs # 2) = R (sm / cc # 1) / (sm / cc # 2) (1) Considering the equality represented by equation (1), the time t2 expressed in TDMA cycle number can be calculated by the following equation (2): ## EQU1 ## ) t2 = 25 Where: - N represents the number of lost radio packets, - LJ represents the default integer part The source coding cs # 2 and the duration t2 are such that the sum between: - a quantity of data resulting from a encoding according to the source coding cs # 2 of a set E of useful data to be transmitted during the duration t2, and - a quantity of data necessary to remedy the loss (in this example, it is radio packets delayed by the retransmission of lost packets), is equal to (or more generally less than or equal to) a This data would have resulted from encoding according to the source coding cs # 1 of this set E of useful data, if the loss had not been detected.

Block 44 is interconnected with physical interface block 45. Block 45 provides the functions necessary for the transmission / reception of data on the 60GHz radio link via the agile antenna 46. The CPU 47 can adjust the parameters of the agile antenna 46 via the address / data bus 50. Thus, during the discovery phase or when the radio channel is degraded, the CPU can adjust the parameters of the antenna 46 so as to test or select a new path radio. The CPU block 47 of the node 40 is connected to a ROM 48 and a RAM 49.

The ROM 48 contains a software program allowing, when it is executed by the CPU (using the RAM 49), the implementation of methods according to the present invention (for example according to the particular embodiment described above). , as well as the antenna parameters corresponding to the different radio paths identified during the discovery phase.

The RAM 49 is used for the execution by the CPU 47 of the aforementioned software program and for the processing of the various tasks performed by the CPU 47. FIG. 5 represents an example of data exchange between the transmitter node (11) and the receiving node (12) when starting a data transmission (video transmission in the present example), in a particular embodiment of the invention. In FIG. 5, the successive successive TDMA cycles are identified by the numbers 100, 110, 120, 130, 140, 150, 160 and 170. In each of the TDMA cycles, the lxxa numbers denote the exchange of signaling data between the nodes 11 and 12 according to the first mode of communication. The numbers 1xxb denote the transmission of HD video data (i.e. useful data) between the transmitting node 11 and the receiving node 12 according to the second communication mode. The data exchange begins with the TDMA 100 cycle. In 101_a, the transmitting node 11 transmits a data (or information) of synchronization corresponding to the Super Frame No. 0 (SF (0)). This synchronization information makes it possible for example to indicate to the receiver node 12 that the synchronization is well maintained while the video transmission phase has not started. In 102_a, the receiving node 12 also confirms the maintenance of good synchronization of the system. Nodes 11 and 12 will then keep this synchronization active on subsequent TDMA cycles. Thus, in the TDMA cycle 110, at 111_a, the transmitting node 11 transmits this active synchronization signaling data (or information) for the Super Frame SF (x). In 112_a, the receiving node 12 does the same. The data exchange continues with the TDMA 120 cycle, corresponding to SF Super Frame (i). At 121_a, the transmitting node 11 transmits signaling data to synchronize the receiver with respect to the start of the video sequence. The transmitting node 11 then starts reading the Video Stream Portion (0) (PFV (0)) and initializing its transmission buffer. In 122_a, the receiver emits an empty sequence. The next TDMA cycle 130 corresponds to SF Super Frame (i + 1). At 131_a, the transmitting node indicates that it does not yet have video data to transmit. The transmitter node 11 reads the PFV (1) and continues the initialization of its transmission buffer. At 132_a, the receiver 12 transmits an empty sequence to the transmitter 11. The sequence continues with the TDMA cycle 140 corresponding to the SF Super Frame (i + 2). At 141_a, the transmitting node announces the start of video transmission in cs # 1 and sm / cc # l mode. In parallel, the transmitter node 11 performs the reading of the PFV (2) and buffers it. At 142_a, the receiving node transmits an empty sequence. At 141 b, the sending node starts the transmission according to the second communication mode by transmitting the PFV (0) stored in the transmission buffer. At 142_b, the receiver 12 correctly receives the PFV (0) and then initializes its reception buffer.

The next TDMA cycle 150 corresponds to SF Super Frame (i + 3). At 151_a, the transmitting node announces transmission of the PFV (1) in cs # 1 and sm / cc # 1 mode. In parallel, the transmitter node 11 performs the reading of the PFV (3) and buffers it. At 152_a, the receiving node issues the acknowledgment of the PFV (0).

At 151 b, the transmitting node transmits the PFV (1) stored in the transmission buffer. At 152_b, the receiver 12 correctly receives the PFV (1) and continues the initialization of its reception buffer. The next TDMA cycle 160 corresponds to SF Super Frame (i + 4). At 161_a, the transmitting node announces transmission of the PFV (2) in cs # 1 and sm / cc # 1 mode. In parallel, the transmitter node 11 performs the reading of the PFV (4) and buffers it. In 162_a, the receiving node issues the acknowledgment of the PFV (1). At 161_b, the transmitting node transmits the PFV (2) stored in the transmission buffer. At 162_b, the receiver 12 correctly receives the PFV (2) and continues the initialization of its reception buffer. The sequence continues with the next TDMA cycle 170, corresponding to SF Super Frame (i + 5). At 171_a, the transmitting node 11 announces the transmission of the PFV (3) in cs # 1 and sm / cc # 1 mode. In parallel, the transmitter node 11 performs the reading of the PFV (5) and buffers it. In 172_a, the receiving node issues the acknowledgment of the PFV (2). At this stage, the receiving node 12 has completed the initialization of its reception buffer, it can then start sending to the consumer application (display) of the PFV (0). At 171_b, the transmitting node transmits the PFV (3) stored in the transmission buffer. In 172_b, the receiver 12 correctly receives the PFV (3) and stores it in its reception buffer. The consuming application is for example a data display element of the transmitted video stream. It may also be for example a recording element of this stream or more generally of continuous processing of this stream. These data could of course be of a different type than video (audio or any other), as long as these data need a continuous timed delivery. FIG. 6 represents an example of data exchange between the transmitting node (11) and the receiving node (12) during a retransmission of data, in a particular embodiment of the invention.

For better understanding, this same data exchange is also presented at the TDMA level in Figure 8a, which has the same numbering as Figure 6 described below. In FIG. 6, the successive successive TDMA cycles are identified by the numbers 300, 310, 320, 330, 340, 350 and 360. In each of the TDMA cycles, the numbers 3xx_a designate the exchange of signaling data between the nodes 11 and 12 according to the first mode of communication. The numbers 3xx_b denote the transmission of HD video data (that is to say useful data) between the transmitting node 11 and the receiving node 12 according to the second communication mode. The TDMA 300 cycle corresponds to the Super Frame j + 1 (SF (j + 1)). In 301_a, the transmitting node 11 announces the transmission of the PFV (n + 1) in mode cs # 1 and sm / cc # 1. In parallel, the transmitter node 11 performs the reading of the PFV (n + 3) and stores it in its transmission buffer. In 302_a, the receiving node 12 transmits the acknowledgment of the PFV (n). In parallel, the receiving node 12 sends the consumer application (display) the PFV (n-2). At 301 b, the transmitting node 11 transmits the PFV (n + 1) stored in the transmission buffer. In 302_b, the receiver 12 detects poor reception of the PFV (n + 1). The data exchange is then continued with the TDMA cycle 310 corresponding to the Super Frame j + 2 (SF (j + 2)). In 311_a, the transmitting node 11 announces the transmission of the PFV (n + 2) in mode cs # 1 and sm / cc # 1. In parallel, the transmitter node 11 performs the reading of the PFV (n + 4) and stores it in its transmission buffer. In 312_a, the receiving node 12 issues the non-acknowledgment of the PFV (n + 1) and sends the consumer application (display) in parallel with the PFV (n-1).

In 311_b, the transmitting node 11 transmits the PFV (n + 2) stored in the transmission buffer. In 312_b, the receiver 12 detects poor reception of the PFV (n + 2). The data exchange then continues with the TDMA cycle 320 corresponding to the Super Frame j + 3 (SF (j + 3)). In 321_a, the transmitting node 11 announces the retransmission of the PFV (n + 1) in mode cs # 2 and sm / cc # 2. In parallel, the transmitter node 11 performs the reading of the PFV (n + 5) and stores it in its transmission buffer. At 322_a, the receiving node 12 issues the non-acknowledgment of the PFV (n + 2). In parallel, the receiving node 12 sends to the consumer application (display) the PFV (n). At 321_b, the sending node 11 retransmits the PFV (n + 1) stored in its transmission buffer. At 322_b, the receiver 12 correctly receives the retransmitted PFV (n + 1).

The next TDMA cycle 330 corresponds to the Super Frame j + 4 (SF (j + 4)). In 331_a, the transmitting node 11 announces the retransmission of the PFV (n + 2) in mode cs # 2 and sm / cc # 2. In parallel, the transmitter node 11 performs the reading of the PFV (n + 6) and stores it in its transmission buffer. In 332_a, the receiving node 12 transmits the acknowledgment of the PFV (n + 1) and sends the consumer application (display) in parallel with the PFV (n + 1). At 331_b, the sending node 11 retransmits the PFV (n + 2) stored in its transmission buffer. At 332_b, the receiver 12 correctly receives the retransmitted PFV (n + 2). The next TDMA cycle 340 corresponds to the Super Frame j + 5 (SF (j + 5)). At 341_a, the transmitting node 11 announces the transmission of the two PFVs (n + 3; n + 4) in mode cs # 2 and sm / cc # 1. In parallel, the transmitter node 11 performs the reading of the PFV (n + 7) and stores it in its transmission buffer. At 342_a, the receiving node 12 transmits the acknowledgment of the PFV (n + 2) and sends the consumer application (display) in parallel with the PFV (n + 2). At 341_b, the transmitting node 11 transmits the two PFVs (n + 3; n + 4) stored in its transmission buffer. At 342_b the receiver 12 correctly receives the 2 PFVs (n + 3, n + 4) transmitted. The next TDMA cycle 350 corresponds to the Super Frame j + 6 (SF (j + 6)). At 351_a, the sending node 11 announces the transmission of the two PFVs (n + 5, n + 6) in the mode cs # 2 and sm / cc # 1. In parallel, the transmitter node 11 performs the reading of the PFV (n + 8) and stores it in its transmission buffer. At 352_a, the receiving node 12 issues the acknowledgment of the PFVs (n + 3; n + 4) and sends the PFV (n + 3) to the consumer application (display) in parallel. At 351_b, the transmitting node 11 transmits the two PFVs (n + 5; n + 6) stored in its transmission buffer. At 352_b, the receiver 12 correctly receives the two PFVs (n + 5, n + 6) transmitted.

The data exchange continues with the TDMA 360 cycle corresponding to the Superframe j + 7 (SF (j + 7)). In 361_a, the transmitting node 11 announces the transmission of the PFV (n + 7) in mode cs # 1 and sm / cc # 1. In parallel, the transmitter node 11 performs the reading of the PFV (n + 9) and stores it in its transmission buffer. At 362_a, the receiving node 12 transmits the acknowledgment of the two PFVs (n + 5, n + 6) and sends the PFV (n + 4) to the consuming application (display) in parallel. At 361_b, the transmitting node 11 transmits the PFV (n + 7) stored in its transmission buffer. In 362_b, the receiver 12 correctly receives the transmitted PFV (n + 7). FIG. 7a shows an example of data exchange between the sending node (11) and the receiving node (12) during a retransmission of data accompanied by a radio path change, in a particular embodiment of the invention. In FIG. 7a, the successive successive TDMA cycles are identified by the numbers 200, 210, 220, 230 and 240.

In each of the TDMA cycles, the numbers 2xx_a denote the exchange of signaling data between the nodes 11 and 12 according to the first communication mode. The numbers 2xx b denote the transmission of HD video data (that is to say useful data) between the transmitting node 11 and the receiving node 12 according to the second communication mode.

The TDMA 200 cycle corresponds to the Super Frame j (SF (j)). In 201_a, the transmitting node 11 announces the transmission of the PFV (n) in mode cs # 1 and sm / cc # l. In parallel, the transmitter node 11 performs the reading of the PFV (n + 2) and stores it in its transmission buffer. In 202_a, the receiving node 12 issues the acknowledgment of the PFV (n-1). In parallel, the receiving node 12 sends the consumer application (display) the PFV (n-3). In 201_b, the transmitting node 11 transmits the PFV (n) stored in the transmission buffer. In 202_b, the receiver 12 correctly receives the PFV (n) and stores it in its reception buffer. We then go to the TDMA cycle 210 corresponding to the Super Frame j + 1 (SF (j + 1)). At 211_a, the transmitting node 11 announces the transmission of the PFV (n + 1) in mode cs # 1 and sm / cc # 1. In parallel, the transmitter node 11 performs the reading of the PFV (n + 3) and stores it in its transmission buffer. At 212_a, the receiving node 12 transmits the acknowledgment of the PFV (n) and sends the consumer application (display) in parallel with the PFV (n-2).

At 211_b, the transmitting node 11 transmits the PFV (n + 1) stored in the transmission buffer. At 212_b, the receiver 12 detects poor reception of the PFV (n + 1). The data exchange is then continued with the TDMA cycle 220 corresponding to the Super Frame j + 2 (SF (j + 2)). In 221_a, the transmitting node 11 announces the transmission of the PFV (n + 2) in mode cs # 1 and sm / cc # 1. In parallel, the transmitter node 11 performs the reading of the PFV (n + 4) and stores it in its transmission buffer. In 222_a, the receiving node 12 issues the non-acknowledgment of the PFV (n + 1) and sends the consumer application (display) in parallel with the PFV (n-1). At 221_b, the transmitting node 11 transmits the PFV (n + 2) stored in the transmission buffer. At 222b, the receiver 12 detects poor reception of the PFV (n + 2).

The data exchange then continues with the TDMA cycle 230 corresponding to the Super Frame j + 3 (SF (j + 3)). In 231_a, the transmitting node 11 announces the retransmission of the PFV (n + 1) in mode cs # 2 and sm / cc # 2. In parallel, the transmitter node 11 performs the reading of the PFV (n + 5) and stores it in its transmission buffer. At 232a, the receiving node 12 issues the non-acknowledgment of the PFV (n + 2) and instructs the sending node to adjust its antenna parameters. In parallel, the receiving node sends the PFV (n) to the consuming application (display). In 231_b, the transmitter node 11 adjusts the parameters of its antenna Tx to use an alternative transmission path (according to the classification made during the aforementioned discovery phase) and retransmits the PFV (n + 1) stored in its transmission buffer.

In 232b, the receiver 12 adjusts the parameters of its Rx antenna to use this alternative transmission path and correctly receives the retransmitted PFV (n + 1). The next TDMA cycle 240 corresponds to the Super Frame j + 4 (SF (j + 4)). In 241_a, the transmitting node 11 announces the retransmission of the PFV (n + 2) in mode cs # 2 and sm / cc # 2. In parallel, the transmitter node 11 performs the reading of the PFV (n + 6) and stores it in its transmission buffer. At 242a, the receiving node 12 transmits the acknowledgment of the PFV (n + 1) and sends the consumer application (display) in parallel with the PFV (n + 1). At 241_b, the sending node 11 retransmits the PFV (n + 2) stored in its transmission buffer. In 242_b, the receiver 12 correctly receives the retransmitted PFV (n + 2).

FIG. 7b, which corresponds to the continuation of FIG. 7a, represents an example of data exchange between the transmitter / receiver nodes during the return in the normal transmission conditions after retransmission and radio path change, in one embodiment particular of the invention. In FIG. 7b, the successive successive TDMA cycles are identified by the numbers 250, 260, 270, 280 and 290. In each of the TDMA cycles, the numbers 2xx_a denote the exchange of signaling data between the nodes 11 and 12 according to the first mode of communication. The numbers 2xx_b denote the transmission of HD video data (that is to say useful data) between the transmitter node 11 and the receiver node 12 according to the second communication mode. The TDMA cycle 250 corresponds to the Super Frame k (SF (k)). In 251_a, the transmitting node 11 announces the transmission of the PFV (x-4) in mode cs # 2 and sm / cc # 2. In parallel, the transmitter node 11 performs the reading of the PFV (x) and stores it in its transmission buffer. At 252a, the receiving node 12 transmits the PFV (x-5) acknowledgment and sends the PFV (x-5) to the consuming application (display) in parallel. In 251_b, the transmitting node 11 transmits the PFV (x-4) stored in its transmission buffer. In 252_b, the receiver 12 correctly receives the transmitted PFV (x-4). The next TDMA cycle 260 corresponds to the Super Frame k + 1 (SF (k + 1)). In 261_a, the transmitting node 11 announces the transmission of the two PFVs (x-3; x-2) in mode cs # 2 and sm / cc # 1. In parallel, the transmitter node 11 performs the reading of the PFV (x + 1) and stores it in its transmission buffer. In 262_a, the receiving node 12 transmits the PFV (x-4) acknowledgment and sends the PFV (x-4) to the consuming application (display) in parallel.

In 261_b, the transmitter node 11 readjusts the parameters of its antenna Tx to return to the initial radio path. The transmitting node then transmits the two PFVs (x-3; x-2) stored in its transmission buffer. In 262_b, the receiver 12 also readjusts the parameters of its antenna Rx to return to the initial radio path. The receiving node 12 then correctly receives the two PFVs (x-3; x-2) transmitted. The next TDMA cycle 270 corresponds to the SuperField k + 2 (SF (k + 2)). In 271_a, the transmitting node 11 announces the transmission of the two PFVs (x-1; x) in the mode cs # 2 and sm / cc # 1. In parallel, the transmitter node 11 performs the reading of the PFV (x + 2) and stores it in its transmission buffer. At 272_a, the receiving node 12 issues the acknowledgment of the two PFVs (x-3; x-2) and sends the PFV (x-3) to the consuming application (display) in parallel. At 271b, the transmitting node transmits the two PFVs (x-1; x) stored in its transmit buffer. In 272_b, the receiver 12 correctly receives the two PFVs (x-1; x) transmitted.

The next TDMA cycle 280 corresponds to the k + 3 Superframe (SF (k + 3)). At 281a, the transmitting node 11 announces the transmission of the PFV (x + 1) in cs # 1 and sm / cc # 1 mode. In parallel, the transmitter node 11 performs the reading of the PFV (x + 3) and stores it in its transmission buffer. At 282a, the receiving node 12 issues the acknowledgment of the two PFVs (x-1; x) and sends the PFV (x-2) to the consuming application (display) in parallel. In 281_b, the transmitting node transmits the PFV (x + 1) stored in its transmission buffer. At 282_b, the receiver 12 correctly receives the transmitted PFV (x + 1). The data exchange continues with the TDMA 290 cycle corresponding to the k + 4 Superframe (SF (k + 4)). In 291_a, the sender node 11 announces the transmission of the PFV (x + 2) in mode cs # 1 and sm / cc # l. In parallel, the transmitter node 11 performs the reading of the PFV (x + 4) and stores it in its transmission buffer. In 292_a, the receiving node 12 transmits the acknowledgment of the PFV (x + 1) and sends the consumer application (display) in parallel with the PFV (x-1). At 291b, the transmitting node transmits the PFV (x + 2) stored in its transmit buffer. In 292_b, the receiver 12 correctly receives the transmitted PFV (x + 2).

Figures 8a, 8b and 8c are now described. FIG. 8a is a representation, over several TDMA cycles, of the same data exchange as that represented previously in relation to FIG. 6. FIGS. 8b and 8c are a representation, on several TDMA cycles, of a first and a second exchange of data between the transmitter / receiver nodes, in an alternative embodiment of the invention. In each of FIGS. 8a, 8b and 8c the execution of the steps of reading and storing the PFV elements originating from the video source is represented at 800. At each new TDMA cycle, the transmitter node 11 reads and stores in its buffer of issue a new portion of video stream (PFV). Each new video stream portion (PFV) is provided two TDMA cycles later to the source code application means; however, each PFV element remains stored in the transmission buffer until the transmitting node 11 receives the corresponding acknowledgment sent by the receiver node 12 according to the first communication mode. These PFVs coming directly from the video source are called data before source coding. In each of FIGS. 8a, 8b and 8c, the application of the source coding is represented at 801_a, 801_b and 801_c. In 801_a is represented the application of source coding performed systematically on each new portion of video stream (PFV). In 801_b and 801_c is shown the application of source coding performed only during a retransmission of data. The data encoded according to the source coding cs # 1 are called data after source encoding sc # 1. These are the 801_a data corresponding to the PFV (n + 1) and PFV (n + 2) data in Figures 8a and 8c, and to the PFV (n + 1), PFV (n + 2), PFV (n +) data. 5) and PFV (n + 6) in Figure 8b.

The data encoded according to the source coding cs # 2 are called data after source coding sc # 2. It is the data, in FIG. 8a, corresponding in 801_a to the data PFV (n + 3), PFV (n + 4), PFV (n + 5) and PFV (n + 6) and corresponding in 801_b to the data PFV ( n + 1) and PFV (n + 2). It is the data, in FIGS. 8b and 8c, corresponding in 801_a to the data PFV (n + 3) and PFV (n + 4) and corresponding in 801_b to the data PFV (n + 1) and PFV (n + 2).

The data encoded according to the source code cs # 3 are called data after source encoding sc # 3. It is the data, in FIG. 8c, corresponding in 801_a to the data PFV (n + 5) and PFV (n + 6), corresponding in 801_b to the data PFV (n + 3) and PFV (n + 4) and corresponding in 801_c to the PFV (n + 1) and PFV (n + 2) data.

In each of FIGS. 8a, 8b and 8c, access to the 60GHz radio channel according to the TDMA protocol is represented at 802. The signaling data transmitted according to the first communication mode are referenced 850. The data transmitted according to the second communication mode and encoded according to the sm / cc # 1 channel coding are called data after channel coding sm / cc # 1.

These are the data transmitted during the TDMA cycles 300, 310, 340 and 350 in FIG. 8a, as well as during all the TDMA cycles represented in FIGS. 8b and 8c. The data transmitted according to the second mode of communication and coded according to the coding channel sm / cc # 2 are called data after encoding channel sm / cc # 2. These are the data transmitted during TDMA cycles 320 and 330 in Figure 8a.

Of course, this representation is purely schematic, for educational purposes of simplified explanation of an embodiment of the invention. In fact, the information introduced by the various codings is not expressly distinguishable from the useful data when all of this data is transmitted over the communication channel. The illustration made in these figures aims to highlight changes in data quantities over the TDMA cycles, according to the principle of the invention. For simplicity of the figures, the first two TDMA cycles 300 and 310 are identical for the different cases to which correspond to Figures 8a, 8b and 8c.

The durations t2 for FIG. 8a, T for FIG. 8b, and T 'for FIG. 8c correspond to the first predetermined duration, that is to say the period of time during which a catch-up phase is carried out, as already mentioned. FIG. 8a is a representation, on several TDMA cycles, of the same data exchange as that represented in FIG. 6, between the transmitting node 11 and the receiving node 12. FIG. 8a resumes strictly the same numbering as FIG. 6, and FIG. 6 having been described in detail above, FIG. 8a will therefore not be described in more detail. FIG. 8b is a representation, over several TDMA cycles, of a first data exchange between the transmitter and receiver nodes, in an alternative embodiment of the invention. This variant consists in applying the configuration associating cs # 2 and sm / cc # 1 at the sending node 11 directly after reception of the non-acknowledgments at 312_a and 322'_a, respectively corresponding to the PFV (n + 1) and PFV (n 2). In this case, in the cycle 320 ', the PFV (n + 1) to be retransmitted is then transmitted at the same time as the new PFV (n + 3). Similarly, in the next cycle 330 ', the PFV (n + 2) to be retransmitted is transmitted at the same time as the new PFV (n + 4). This variant makes it possible to rebalance the transmit and receive buffers more quickly and thus to return to normal transmission conditions as soon as the TDMA cycle 340 '. The transmitting node can then return to the configuration associating cs # 1 and sm / cc # 1 as of the TDMA cycle 340 '. FIG. 8c is a representation, over several TDMA cycles, of a second data exchange between the transmitter / receiver nodes, in the same variant as that of FIG. 8b. It is assumed here that the PFVs (n + 1; n + 3) and PFV (n + 2; n + 4) transmitted in the cycles 320 "and 330" are also poorly received by the receiving node 12 (maintenance of the conditions of transmission degraded). The corresponding non-acknowledgments are thus transmitted by the receiving node 12 respectively at 332 "_a and 342" _a. In this case, the sending node 11 then switches to a new configuration associating a third source coding cs # 3 and the first modulation scheme and channel coding sm / cc # 1. The cs # 3 source coding uses even fewer bits / pixel than the cs # 2 coding, so that three times more video data can be transmitted in a TDMA cycle. Thus, in cycle 340 ", the PFVs (n + 1, n + 3) to be retransmitted are then transmitted at the same time as the new PFV (n + 5), and in the following cycle 350 n + 2, n + 4) to retransmit are transmitted at the same time as the new PFV (n + 6). FIG. 9 presents a flowchart of a particular embodiment of the method according to the invention, implemented in the transmitter device 11.

After a start step 400, the transmitting node proceeds to step 401, in which it maintains active system synchronization with the receiving node (sending signaling data for active synchronization in a first communication mode). In step 401, the sending node 11 also performs an initialization of its transmission buffer Tx. This step 401 corresponds to the TDMA cycles 100, 110, 120 and 130 of FIG. 5. In the following step 402, the transmitting node transmits video data in a second communication mode, by applying the source coding cs # 1 and the coding channel sm / cc # l. This step 402 corresponds to the TDMA cycles 140, 150, 160 and 170 of FIG. 5, as well as to the TDMA cycles 300 and 310 of FIG.

In the next step 403, the transmitting node detects whether it has received a Data Acknowledge (nAck) (PFV). If not, it returns to step 402 and continues to transmit video data in cs # 1 and sm / cc # 1 mode. If so, it goes to step 404. In step 404, the transmitting node changes mode and transmits video data, applying source code cs # 2 and sm / cc # 2 channel coding. This step 404 corresponds to the TDMA cycles 320 and 330 of FIG. 6. In the next step 405, the transmitting node detects whether a duration t1 (discussed in detail above with respect to FIG. 4) has elapsed since the switching from mode cs # 1 and sm / cc # 1 to mode cs # 2 and sm / cc # 2. If not, it returns to step 404 and continues to transmit video data in cs # 2 and sm / cc # 2 mode. If so, it goes to step 406. In step 406, the transmitting node changes mode and transmits video data, applying the source coding cs # 2 and the sm / dc modulation and coding scheme. #l. This step 406 corresponds to the cycles 340 and 350 TDMA of FIG.

In the next step 407, the transmitting node detects whether a duration t2 (discussed in detail above in connection with FIG. 4) has elapsed since switching from the mode cs # 2 and sm / cc # 2 to the mode cs # 2 and sm / cc # l. If not, it returns to step 406 and continues to transmit video data in cs # 2 and sm / cc # 1 mode. If so, it returns to step 402 and resumes transmitting video data in cs # 1 and sm / cc # 1 mode (case of the TDMA 360 cycle of FIG. 6).

Claims (16)

REVENDICATIONS1. A method for transmitting payload data from a transmitting device (11) to a receiving device (12), the transmitting device applying to said payload data, before transmission, a first source code (cs # 1) and then a first modulation and coding scheme channel (sm / cc # 1), said method being characterized in that, upon detection of a data loss (403), the transmitting device performs a step (406) of applying a second source code (cs # 2), in place of the first source coding, to a set E of useful data to be transmitted during a first predetermined duration (t2; T; T '), the second source coding and the first predetermined duration being such that the sum between: a quantity of data resulting from an encoding according to the second source code of said set E of useful data, and a quantity of data necessary to remedy said loss, is less than or equal to a quantity of data es that would have resulted from an encoding according to the first source coding of said set E of useful data, if said loss had not been detected. 2. Method according to claim 1, characterized in that, following the detection of said data loss (403), the transmitting device performs a step (404) of applying a second modulation scheme and channel coding (sm / cc # 2) for a second predetermined duration (t1) which precedes said first predetermined duration (t2), said second channel modulation and coding scheme providing a more robust channel coding than said first channel modulation and coding scheme. 3. Method according to claim 2, characterized in that, following the detection of said data loss (403), the transmitting device applies (404, 406): said second source coding and said second modulation scheme and channel coding during said second predetermined duration (t1); then - said second source coding and said first channel modulation and coding scheme, during said first predetermined duration (t2). 4. Method according to claim 3, characterized in that the lost useful data to which said loss corresponds are retransmitted during said second predetermined duration (tl), and in that the data necessary to remedy the loss, which are transmitted during said first predetermined duration (t2), are coded data resulting from an encoding with the second source coding useful data that could not be transmitted during said second predetermined duration (t 1) due to the retransmission of lost useful data. 5. Method according to any one of claims 1 to 4, characterized in that said first predetermined duration (t2) is determined according to a first ratio between: a first amount of coded data, resulting from an encoding with the first source encoding of a given amount of data; and - a second amount of encoded data resulting from encoding with the second source encoding of said given amount of data. 6. Method according to any one of claims 2 to 5, characterized in that the second predetermined duration (tl) corresponds to: a time interval for transmitting a predetermined number of useful data packets, from the receiving device to the receiving device; or - an interval of time during which transmission conditions between the transmitting device and the receiving device are estimated such that data loss would occur if the first source coding was applied. The method of claim 6, characterized in that said predetermined number of payload packets is determined based on a number of negative acknowledgments from the receiving device for said payload transmission. 25 8. Method according to claim 1, characterized in that, following the detection of said data loss, the transmitting device applies said second source coding and said first modulation scheme and channel coding during said first predetermined duration (T). 9. The method of claim 8, characterized in that the data necessary to remedy the loss, which are transmitted during said first predetermined duration (T), are encoded data resulting from encoding, with the second source coding. , of useful data corresponding to a retransmission of the lost useful data to which said loss corresponds. 10. Method according to any one of claims 1 to 9, characterized in that, following the detection of said loss of data, the transmitting device performs a step (231_b) of modifying parameters of a transmitting antenna. of the transmitting device, for said transmission of useful data. 11. Method according to any one of claims 1 to 10, characterized in that said transmission of useful data is performed in a network (10) comprising said transmitting device and said receiving device and implementing a timing according to a transmission cycle. (24, 25) comprising: - a first portion (26), dedicated to a first communication mode, for the exchange of signaling data between said transmitting device and said receiving device; and a second portion (27), dedicated to a second communication mode, for said transmission of useful data from the transmitting device to the receiving device, and in that the detection of said data loss comprises a reception step (403 ), by the transmitting device, at least one negative acknowledgment, transmitted by the receiving device during the first part of a current cycle, for a useful data packet previously transmitted by the transmitting device during the second part of a previous cycle. 12. 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 11, when said program is executed on a computer. A computer readable storage medium storing a computer program comprising a set of computer executable instructions for carrying out the method according to at least one of claims 1 to 11. 14. Transmitting device (11) for transmitting useful data to a receiver device (12), the transmitting device applying to said payload data, before transmission, a first source coding (cs # 1) then a first channel modulation and coding scheme (sm / cc # l), the transmitting device being characterized in that it comprises means, activated following a detection of a loss of data, making it possible to apply a second source coding (cs # 2) to the place of the first source coding, to a set E of useful data to be transmitted during a first predetermined duration (t2; T; T '), the second source coding and the first predetermined duration being such that the sum between: - a quantity of data resulting from encoding according to the second source code of said set of useful data, and - a quantity of data necessary to remedy said loss, is less than or equal to a quantity of data that would have resulted encoding according to the first source encoding of said set of useful data, if said loss had not been detected. 15. Transmitting device according to claim 14, characterized in that it comprises means, activated following the detection of said loss of data, for applying a second modulation scheme and channel coding (sm / cc # 2) during a second predetermined duration (t1) preceding said first predetermined duration (t2), said second channel modulation and coding scheme providing a more robust channel coding than said first channel modulation and coding scheme. 16. Transmitting device according to claim 15, characterized in that it comprises means, activated following the detection of said data loss (403), for applying: said second source coding and said second modulation scheme and channel coding during said second predetermined duration (t1); then - said second source coding and said first channel modulation and coding scheme, during said first predetermined duration (t2).