Cooperative retransmission method in an omamrc system with joint resource allocation and selections of the sources to be helped

Abstract

The present invention relates to a transmission method intended for an OMAMRC telecommunication system with M sources (S 1 , ...,S M ), optionally L relays and one destination, M ≥ 2, L ≥ 0. In such a solution, the transmission rate allocation and the selection of the sources to be helped are carried out by frame. Such an implementation is referred to as FLA (fast link adaptation), ("link" also being referred to as "channel" in the remainder of the document). The state of the channel may be different from one frame to another. Dynamic transmission rate allocation by frame provides the best performance. The implementation of such a solution assumes the knowledge, by the destination, of the CSI data for all the channels of the system, that is to say the channels between the sources, between the relays, and between the relays and the sources.

Description

[0001] DESCRIPTION

[0002] TITLE: Cooperative retransmission method in an OMAMRC system with joint resource allocation and selection of sources to be assisted

[0003] Field of invention

[0004] The present invention relates to the field of digital communications. Within this field, the invention relates more particularly to the transmission of coded data between at least two sources and a destination with relaying by nodes which may be relays or sources.

[0005] It is understood that a relay does not have a message to transmit. A relay is a node dedicated to relaying messages from sources while a source has its own message to transmit and can also in certain cases relay messages from other sources i.e. the source is said to be cooperative in this case.

[0006] There are many relaying techniques known by their Anglo-Saxon names: "amplify and forward", "decode and forward", "compress-and-forward", "non-orthogonal amplify and forward", "dynamic decode and forward", etc.

[0007] The invention applies in particular, but not exclusively, to the transmission of data via mobile networks, for example for real-time applications, or via, for example, sensor networks.

[0008] Such a sensor network is a multi-user network, consisting of several sources, several relays and a destination that can use an orthogonal multiple access scheme of the transmission channel between the sources and the destination, noted OMAMRC ("Orthogonal Multiple-Access Multiple-Relay Channel" according to Anglo-Saxon terminology).

[0009] According to this scheme, orthogonality between source and relay transmissions is achieved by time multiplexing in the form of disjoint time slots.

[0010] Prior art and its drawbacks

[0011] An OMAMRC telecommunication system has M sources, possibly L relays and one destination, M > 2, L > 0 with an implementation of a time orthogonal multiple access scheme of the transmission channel that applies between the nodes taken from the M sources and the L relays. The maximum number of time slots per transmitted frame is M + T max with M intervals allocated during a first phase to the successive transmission of the M sources and T U sed T max intervals for one or more cooperative transmissions allocated during a second phase to one or more nodes selected by the destination according to a selection strategy.

[0012] The known OMAMRC transmission system comprises at least two sources, each of which can operate at different times either exclusively as a source or as a relay node. The system may optionally also comprise relays. The node terminology covers both a relay and a source acting as a relay node or as a source. The system considered is such that the sources can themselves be relays. A relay is distinguished from a source because it has no message of its own to transmit, i.e. it only retransmits messages from other nodes.

[0013] The channels between the different nodes of the system are subject to slow fading and white Gaussian noise. Knowledge of all the channels of the system (CSI: Channel State Information) by the destination is not always available. Indeed, the channels between sources, between relays, between relays and sources are not directly observable by the destination and their knowledge by the destination requires a very important exchange of information between the sources, the relays and the destination. To limit the cost of feedback overhead, most often only information on the channel distribution / statistics (CDI: Channel Distribution Information) of all channels, e.g. average quality (e.g. average SNR, average SINR) of all channels, is assumed to be known by the destination in order to determine the bitrates allocated to the sources.

[0014] A transmission method implemented in such an OMAMRC system distinguishes three phases, an initial phase and, for each frame to be transmitted, a ère phase and a 2 nde phase. The transmission of a frame takes place in two phases which are possibly preceded by an additional phase called the initial phase.

[0015] In a first variant, during the initialization phase, the destination determines an initial rate for each source by taking into account the average quality (for example SNR) of each of the system's channels.

[0016] The destination estimates the quality (e.g. SNR) of the direct channels: source to destination and relay to destination using known techniques based on the use of reference signals. The quality of the source-source, relay-relay and source-relay channels is estimated by the sources and relays using, for example, the reference signals. The sources and relays transmit the average qualities of the channels to the destination. This transmission occurs before the initialization phase. Since only the average value of the quality of a channel is taken into account, its refresh occurs on a long time scale, i.e. over a time that allows the rapid variations (fast fading) of the channel to be averaged. This time is of the order of the time required to travel several tens of wavelengths of the frequency of the transmitted signal for a given speed. The initialization phase occurs, for example, every 200 to 1000 frames.The destination returns the initial flow rates it has determined to the sources via a return path. The initial flow rates remain constant between two occurrences of the initialization phase.

[0017] In the first phase, the M sources successively transmit their message during the M time slots using modulation and coding schemes determined from the initial bit rates. During this phase, the number of channel uses (channel use, i.e. resource element according to 3GPP terminology) is fixed and identical for each of the sources.

[0018] In the second phase, messages from the sources are transmitted cooperatively by the relays and / or by the sources. This phase lasts a maximum of T max timeslots. During this phase, the number N2 of channel uses is fixed and identical for each of the selected nodes (sources and relays).

[0019] During the first phase, independent sources broadcast their messages in the form of coded information sequences to a single recipient. Each source broadcasts its messages at the initial rate. The destination communicates its initial rate to each source via very limited rate control channels. Thus, during the first phase, the sources each transmit their respective messages in turn during timeslots, each dedicated to a source.

[0020] Sources other than the one transmitting and possibly relays, of the "Half Duplex" type, receive successive messages from the sources, decode them and, if selected, generate a message only from the messages from the sources decoded without error.

[0021] The selected nodes then access the channel orthogonally in time to each other during the second phase to transmit their generated message to the destination.

[0022] The destination can choose which node should transmit at any given time.

[0023] In a second variant, during the initialization phase, an initial flow rate is assigned to each source For each source i, the destination calculates the different possibilities of assigning a flow rate assuming that the flow rates assigned to the other sources j with j #= i are fixed. The flow rate assigned to source i is the one that maximizes the spectral efficiency of the system given the flow rates assigned to the other sources j in the system. These steps are repeated until there is no more change in the value of the flow rate assigned to source i.

[0024] In the first phase, the M sources successively transmit their message during the M time slots using modulation and coding schemes respectively determined from the bit rates assigned to them during the initialization phase. During this phase, the number channel use (ie resource element according to 3GPP terminology) is fixed and identical for each of the sources.

[0025] In the second phase, messages from the sources are transmitted cooperatively by the relays and / or by the sources. This phase lasts a maximum of T max time slots. During this phase, the number N2 of channel uses is fixed and identical for each of the participating nodes (sources and relays).

[0026] During the first phase, independent sources broadcast their messages in the form of coded information sequences to a single recipient. Each source broadcasts its messages at the initial rate. The destination communicates its initial rate to each source via very limited rate control channels. Thus, during the first phase, the sources each transmit their respective messages in turn during timeslots, each dedicated to a source.

[0027] Sources other than the one transmitting and possibly relays, of the “Half Duplex” type, receive successive messages from the sources, decode them and generate a message only from the messages from the sources decoded without error.

[0028] Nodes that have decoded the message from a source identified by the destination without error then access the channel orthogonally in time between them during the second phase to transmit their generated message to the destination.

[0029] Although such solutions allow to improve the average spectral efficiency (utility metric) within the system considered under the constraint of respecting an individual quality of service (QoS) per source, it is desirable to try to further improve the performance of such a system.

[0030] The present invention meets this objective.

[0031] Statement of the invention

[0032] The invention meets this need by proposing a method for transmitting successive messages forming a frame intended for an OMAMRC (“Orthogonal Multiple-Access Multiple-Relay Channel”) type telecommunications system with N nodes and a destination (D), the N nodes comprising M sources i, i.e. {1, ..., M] and L relays (r ± ... , r L ) with M > 2, L > 0, a frame comprising M time slots allocated to the successive transmission of a message by the M sources and T used time slots allocated to at least one cooperative retransmission of a redundancy of a message transmitted by at least one source in one of said M time slots.

[0033] Such a method is particular in that it comprises the following steps implemented by the destination (D) for a current frame:

[0034] - selection, from a plurality of ordered sets of source identifiers A, of an ordered set of sources A comprising Tused identifiers of sources for which cooperative retransmission is planned by all nodes knowing this source, a source identifier being associated with a corresponding time interval among the T used time intervals allocated to cooperative retransmission, - transmission, to the N nodes and prior to the successive transmission of a message by the M sources, of information relating to an allocation of transmission rates per source Rt(A) determined for the ordered set of sources A selected,

[0035] - transmission, to the N nodes and prior to the cooperative retransmission of a redundancy of a message, of information representing the selected set of sources A.

[0036] Such a method makes it possible, for a given frame, to select the sources whose messages will be retransmitted and to allocate transmission rates to all the sources jointly, unlike what is done in the state of the art. Indeed, in the solutions of the state of the art, either the allocation of transmission rates is determined by assuming that certain nodes of the system will retransmit a redundancy of a message sent by a source; or the destination chooses, following the transmission phase, at least one source whose message must be retransmitted and asks all the nodes having decoded this message to retransmit a redundancy of this message.

[0037] By jointly selecting the sources whose messages will be retransmitted and allocating transmission rates to all sources, the spectral efficiency of the system is improved compared to state-of-the-art solutions.

[0038] In this solution, the destination has CSI data for all channels in the system, i.e., the channels between sources, between relays, and between relays and sources. This allows the determination of the ordered set of sources and the corresponding transmission rate allocation that maximizes the spectral efficiency of the system.

[0039] In a particular embodiment, the selection is based on a criterion relating to a spectral efficiency associated with the set A and to an allocation of transmission rates Rt(A) associated with it.

[0040] According to one aspect of the transmission method, the allocation of transmission rates Ri(A) is transmitted to the N nodes prior to the successive transmission of a message by the M sources.

[0041] Thus, all nodes in the system have information relating to the allocation of transmission rates prior to any transmission of messages by the sources.

[0042] More particularly, and in order to reduce the complexity of the selection process of A, a sequential strategy can be used. In such a sequential strategy, the determination of the ordered sets of source identifiers A comprises the following steps: a) initialization of an ordered set 40= 0, b) generation of a current ordered set of sources A J+1 by adding, to the previous ordered set of sources Aj, a source i for which a cooperative retransmission is planned during a corresponding time interval among the T maxtime intervals, such as Aj +1 = (Aj, i), said source i being selected from the M sources of the system according to the ordered set of sources Ap c) repetition of step b) until T is obtained max ordered sets of source identifiers A.

[0043] Such an iterative method allows to determine the ordered sets of sources for each T used E {1, ... , T max} or iteration which present a most efficient transmission flow allocation according to a given criterion and depending on the result of the previous iterations. The case T US ed = 0 is obvious and corresponds to an empty source set 40= 0.

[0044] More specifically, the source i selected from the M sources of the system according to the ordered set of sources Aj is the source which maximizes the spectral efficiency

[0045] For a given ordered set of sources A, the allocation of transmission rates R[(A) is carried out in the following manner: - for a source i whose identifier is not included in A, then R L (34) = I i D Yes i D represents the mutual information between source i and destination (D),

[0046] - for a source i whose identifier is included in A, then knowing that where R t is the d initial transmission rate of source i, ni represents the number of cooperative retransmissions from source i in A , the lower bound I i D + a ■ ni ■ I i D corresponds to a situation in which no other node than source i knows the message transmitted by source i, the upper bound I i D + a • n i • J* i D corresponds to a situation in which all nodes know the message transmitted by source i with J* L Dthe equivalent mutual information between source i and destination (D) when all nodes know the message transmitted by source i, and where 0 È r used (34) represents a cut-off event worth 0 when the destination cannot decode the message sent by source i in the current frame and worth 1 when the destination decodes the message sent by source i in the current frame.

[0047] The possible values ​​of the transmission rates are limited based on the decoding data from the different nodes in the system. This allows for faster convergence towards the transmission rate value that meets the allocation conditions.

[0048] More specifically, the allocation of transmission rates Rt(A) is obtained by setting an initial value of R i 0 defined as being — ■ - — - —■ - — , if for R t 0 , Gold used (34)

[0049] = 0, then the lower bound takes the value R i 0 , if for R i 0 , 0^ (4) = 1, then the upper bound takes the value R i 0 , the following iteration is then implemented until the difference between a current value of the upper bound and a current value of the lower bound is less than threshold: , if for then the lower bound takes the value R i a , if for = 1, then the upper bound takes the value R i a .

[0050] Such an approach is simple to implement and converges quickly.

[0051] According to another characteristic of the transmission method, the selection of the ordered set of sources A intended to be transmitted to the N nodes is carried out in the following manner: u s represents the spectral efficiency of the current frame and where a edrepresents a ratio between the number of transmission channels available in the T used time slots and the number of transmission channels available in the M time slots and where B represents a set of ordered sets of sources A.

[0052] In an example of realization

[0053] In an exemplary embodiment, B represents the set of ordered sets of sources A for T used time intervals such as The information relating to the selected ordered set of sources A transmitted to the different nodes of the system prior to the retransmission phase comprises, in one example, a first tuple of dimension T used (ie, T used -tuple) identifying the sources belonging to the selected ordered set of sources A and at least a second M-tuple n = (n 1 ; ... , n M) representing the number of times n, that a source i is intended to be retransmitted during the T US ed time intervals (if source i does not belong to  then n, = 0).

[0054] The information relating to the selected ordered set of sources A transmitted to the different nodes of the system prior to the retransmission phase comprises, in another example, at least one integer representing both the sources belonging to the selected ordered set of sources A and the number of times that a source identified in the ordered set of sources A is intended to be retransmitted during the T used time intervals.

[0055] The information transmitted is not cumbersome to transmit and does not induce additional load in the system.

[0056] The invention further relates to an OMAMRC telecommunication system with N nodes and a destination (£>), the N nodes comprising M sources i, i E {1, ... , M] and L relays (r^ ... , r L ) with M > 2, L > 0, suitable for implementing the transmission method which is the subject of the invention.

[0057] The invention finally relates to a computer program product comprising program code instructions for implementing a method as described above, when executed by a processor.

[0058] The invention also relates to a computer-readable recording medium on which is recorded a computer program comprising program code instructions for executing the steps of the method according to the invention as described above.

[0059] Such a recording medium may be any entity or device capable of storing the program. For example, the medium may include a storage medium, such as a ROM, for example a CD ROM or a microelectronic circuit ROM, or a magnetic recording medium, for example a USB flash drive or a hard disk.

[0060] On the other hand, such a recording medium may be a transmissible medium such as an electrical or optical signal, which may be conveyed via an electrical or optical cable, by radio or by other means, so that the computer program contained therein is remotely executable. The program according to the invention may in particular be downloaded over a network, for example the Internet.

[0061] Alternatively, the recording medium may be an integrated circuit in which the program is incorporated, the circuit being adapted to execute or to be used in the execution of the method which is the subject of the aforementioned invention.

[0062] List of figures

[0063] Other aims, characteristics and advantages of the invention will appear more clearly on reading the following description, given as a simple illustrative, and non-limiting, example, in relation to the figures, among which:

[0064] [fig. 1]: this figure represents an embodiment of the invention described in the context of an OMAMRC system,

[0065] [fig. 2]: this figure represents a transmission cycle of a frame,

[0066] [fig. 3]: this figure represents the different steps of the transmission method which is the subject of the invention implemented by the system of figure 1 each frame transmitted within the system, [fig. 4]: this figure represents a destination belonging to an OMAMRC telecommunications system with M sources, possibly L relays and a destination, M > 2, L > 0 according to an embodiment of the invention. Such a destination is capable of implementing the transmission method according to figure 3.

[0067] Detailed description of embodiments of the invention

[0068] The general principle of the invention is based on a method for selecting the sources whose messages will be retransmitted and for allocating transmission rates to all the sources jointly. By jointly selecting the sources whose messages will be retransmitted, or aided sources, and allocating transmission rates to all the sources, the spectral efficiency of the system is improved. More particularly, in the present solution the allocation of transmission rates and the selection of the sources to be aided are carried out per frame; such an implementation is called FLA for "Fast Link Adaptation" or fast link (or channel in the rest of the Document). Since the state of the channel may be different from one frame to another, the allocation of rates in dynamic transmission per frame offers the best performance.Implementing such a solution requires the destination to know the CSI data for all channels in the system, i.e., the channels between sources, between relays, and between relays and sources.

[0069] We now present, in relation to [Fig. 1], an embodiment of the invention described in the context of an OMAMRC system with the support of the diagram of [Fig. 2] which illustrates a transmission cycle of a frame.

[0070] This system includes M sources which belong to the source set "S' = {s 1; L relays that belong to the relay set SR = {r 1; ... , r L ] and a destination D. By convention, it is considered that s ( = i Yi E {1, ... , M} and r k = M + k V / ce {1, ... , L}.

[0071] Each source i in the set «S» communicates with the unique destination with the help of other sources (user cooperation) and cooperating relays.

[0072] To simplify the description, the following assumptions are made subsequently on the OMAMRC system: the sources, the relays are equipped with a single transmitting antenna; the sources, the relays, and the destination are equipped with a single receiving antenna; the sources, the relays, and the destination are perfectly synchronized; the sources are statistically independent (there is no correlation between them); all the nodes transmit with the same power; use is made of a CRC code assumed to be included in the K s information bits from each source to determine whether a message is correctly decoded or not; the links between the different nodes suffer from additive noise and fading. The fading gains are fixed during the transmission of a frame carried out for a maximum duration M + T max time intervals, but can change independently from frame to frame. Tmax > 1 is a system parameter; a frame comprises M time slots allocated to the successive transmission of a message by the M sources and T used < T max time slots allocated to at least one cooperative retransmission of a redundancy of a message transmitted by at least one source in one of said M time slots, the M time slots define a transmission phase and the T used time intervals define a retransmission phase, the instantaneous quality of the direct channel / link in reception (CSIR Channel State Information at Receiver) is available at the destination, sources and relays; returns are error-free (no errors on control signals).

[0073] Nodes include relays and sources that can behave like a relay when not sending their own message.

[0074] The nodes, M sources and L relays, access the transmission channel using an orthogonal multiple access scheme in time or frequency, which allows them to listen to the transmissions of other nodes without interference. The nodes operate in a "half-duplex" mode.

[0075] [Fig. 3] represents the different steps of the transmission method which is the subject of the invention implemented by the system described above. The steps described below are implemented for each frame transmitted within the system considered.

[0076] Knowing that during the retransmission phase, at each time interval, a single source is helped by a plurality of nodes and knowing that the retransmission phase has a maximum number of time intervals T max The following assumption is made: during a retransmission phase, there is a finite number of possible selections of sources to be helped.

[0077] In a step El, the destination D determines a plurality of ordered sets of sources A to be assisted in each time interval among the T max possible time intervals.

[0078] For this, the destination D generates for a first time interval T o , a first ordered set of sources 40 not identifying any source to help. Such an ordered set of sources 40 is therefore a scalar of zero value.

[0079] Then, for a second time interval T lt the destination determines a first source i among the M sources of the system and generates a second ordered set of sources A ± of dimension 2 including the coordinate of the ordered set of sources 40 determined for the time interval T o and the coordinate determined for the second time interval T ± such as A ± = (0, S1 ).

[0080] In order to determine the source i intended to be helped during the time interval T lt the destination (D) determines, in a step E2, a transmission rate allocation Ri(A) that is as efficient as possible for the ordered set of sources A considered. Thus, the destination D calculates for a source i an unavailability event 0i r used (.4). Such an unavailability event for source i is expressed as follows:

[0081] Yes i D represents the mutual information between source i and destination (D), represents the mutual information between the message sent by source i and destination (D) through the equivalent channel taking into account all active relay nodes towards destination (D) after the l iemeselection of source i with l E {1, ... , Hj} and where n, represents the number of selections of source i. Indeed, the equivalent mutual information only depends on the number l of times that a given source i has been helped, in other words, the index l refers to the number of retransmissions of a given source i.

[0082] Equivalent mutual information / i £) (Z) is obtained by identifying the set of relay nodes that helped source i, i.e. that decoded the message sent by source i at the end of the l — l ieme retransmission. To find out if a relay node has helped source i at the end of the l ieme retransmission, an unavailability event is performed by taking node j as the destination is calculated, i.e. the set of relay nodes having helped source i at the end of the l — l iemeretransmission transmits the message sent by source i thus defining an equivalent channel to node j having for mutual information Node j cannot help source i at the end of an n ieme retransmission if and only if [R t > I t j +

[0083] The optimal transmission rate allocation Ri(A) for a given ordered set of sources A is the transmission rate allocation guaranteeing that an unavailability event Oi7T US e d (^4) is not declared, that is, used

[0084] Two situations then appear: a first situation in which a second situation in which argmax R t .

[0085] Indeed, the set of messages known either by decoding or by definition (or "decoding set" in English) by a given relay node depends on the state of the channel and the transmission rate allocated to source i. Consequently, the equivalent mutual information depends on the transmission rate R t allocated.

[0086] It appears that the transmission rate value R t for a given source i can be limited depending on the decoding sets of the different relay nodes. More precisely, the value of the flow rate R t for a given source i in the case where n, =£ 0 is limited between a minimum value I i D + a ' n i ' k,D et une maximum value I i D + a ■ ni ■ J* i D -

[0087] The minimum value I i D + a ■ ni ■ I i D corresponds to a situation in which no relay node helps source i during the retransmission phase.

[0088] The maximum value I i D + a ■ ni ■ ]* i D corresponds to a situation in which all relay nodes help source i during the retransmission phase and where J* i £) represents the equivalent mutual information between source i and destination (D) when all relay nodes help source i.

[0089] In a first implementation, the allocation of transmission rates Rt(A) is obtained by applying a binary search algorithm. Such an algorithm is represented in Appendix 1.

[0090] Thus, an initial value of R t 0 is defined as being— - — - — - - - . If for

[0091] R i 0 , Oi,r used (X) = 0 / then the lower bound (left in the algorithm) takes the value R i 0 , if for R i 0 , Oi,r used (-4) = 1, then the upper bound (right in the algorithm) takes the value R i 0 .

[0092] The following iteration a is then implemented until the difference between a current value of the upper bound and a current value of the lower bound is less than threshold e: if for? u , O CTused (4) = 1, then the lower bound takes the value R i a , if forÆ u , O CTused (4) = 0, then the upper bound takes the value R i a In other words, when the search window is small enough, the algorithm terminates.

[0093] In a second implementation, the allocation of transmission rates Rt(A) is also obtained by applying a binary search algorithm. However, the values ​​of the transmission rates are chosen from a finite set of values ​​R = {R o ■ ■■ > RMCS) with R o < ■■■ < R MCS - Such an algorithm is represented in Appendix 2.

[0094] In practice, there is always a family of modulation and coding schemes where possible transmission rate values ​​are predefined. In other words, although in the present solution the transmission rates do not depend on a predefined set of transmission rates, the present solution can nevertheless be implemented in realistic scenarios where a predefined set of transmission rates is adapted. Here, it is not necessary to search among all the transmission rate values ​​to determine the optimal transmission rate value. For this, the binary search algorithm can be adapted to scan through the set of possible transmission rate values.

[0095] In an initialization phase, the value of e is set to E = — R,\

[0096] Then, an initial value of R i 0 is defined as being If for i0, O i T used(-4) = 0, then the lower bound

[0097] 2

[0098] (left in the algorithm) takes the value R t 0 , if for R i 0 , O i T (34) = 1, then the upper bound

[0099] (right in the algorithm) takes the value R i 0 .

[0100] The following iteration a is then implemented until the difference between a current value of the upper bound and a current value of the lower bound is less than threshold e: upper bound a-1 — lower bound a-1

[0101] Ri, a = if for R i a , Oi Tused (34) = 0, then the lower bound takes the value R i a , if for R i a , O i T (34) = 1, then the upper bound takes the value R i a .

[0102] The transmission rate allocation R^A) selected for the ordered set of sources A is

[0103] Ri G*) = argmin (lower boundlast iteration- r) rER such that r <borne inférieure dernière iteration

[0104] This process is repeated again for a third time interval T2, the destination determines a third source s? among the M sources of the system and generates a third ordered set of sources 42 of dimension 3 comprising the coordinate of the ordered set of sources 40 determined for the time interval T o and the coordinate s, determined for the second time interval T ± and the coordinate s? such that 42= (0, s,, Sy).

[0105] This ordered set of sources 42 is obtained by determining the optimal transmission flow allocation R t (342) such that:

[0106] Rt(A2) = argmaxR;,

[0107] KjelK knowing that: 0 CTused (42) = 0.

[0108] The process is carried out as many times as necessary until T is obtained maxordered sets of sources A, such that s t ) where Aj is the ordered set of sources and Aj_1 the previous ordered set of sources.

[0109] So, in an example where T max = 2, and the system includes two sources s, and s2, then the possible ordered sets of sources A to help are: for T used = 0 (no source is helped), the ordered set of sources A o is then a zero-valued scalar such that A o = 0, for T used = 1, either source s, is selected to be helped or source s2, the ordered set of sources A ± is then an ordered set of size 2 such that A ± = (0, sJ = (s^, or A ± = (0, s2) = (s2), for T used = 2, either source SL is selected twice to be helped, or source s2 is selected twice to be helped, or source SL and source s2 are each selected once, the ordered set sources A2 is then an ordered set of size 3 such that A2 = ( s 1; If), >12= (S2,S2), OR ^2= (Si,S2) = (s2,Si).

[0110] Indeed, in the case where destination D chooses to help source SL and source s2 during the same retransmission phase, it does not matter which one is helped first. In other words, destination D chooses to help source SL during the first time interval constituting the retransmission phase and then source s2 during the second time interval constituting the retransmission phase, this does not impact the system performance. The order in which the different sources are retransmitted is therefore not important because it is the number of time intervals allocated to the same source i among the T max possible time intervals that matter.

[0111] Once the set of ordered sets of sources A have been determined by the destination during the different iterations of steps E1 and E2, the destination (D) selects, in a step E3, the ordered set of sources A which offers the highest possible spectral efficiency for the system considered. Thus the selected ordered set of sources A is the ordered set of sources which satisfies:

[0112] A = argmax T]f rame (R(A),A) represents the spectral efficiency of the frame considered.

[0113] The number of time intervals T used constituting the retransmission phase is given by the size of the ordered set of sources A selected during step E3.

[0114] An example of an algorithm for selecting the ordered set of sources A that provides the highest possible spectral efficiency for the system under consideration is shown in Appendix 3.

[0115] In this third algorithm, T used is initialized to T used = 0 A is initialized to A = 0, and- n, = 0 V ie {1, ... , M}. In this case (0) is the transmission rate belonging to the MCS family which is lower and closest to I i D . We note the associated spectral efficiency is expressed as being:

[0116] Similarly, for n, = 0, Rt(A) is the transmission rate belonging to the lowest MCS family and closest to I i D .

[0117] The proposed algorithm is then confronted with a complexity problem due to the exponential number of possible allocations e S Tused . For a value of T used given, the number of ordered sets of possible sources A is T^ sed . Knowing that the order of the sources is not important, and that only the number of retransmissions for each source counts, the number of possible ordered sets A is reduced to

[0118] Although such a reduction is interesting, it can be the source of problems in implementing the algorithm presented in Appendix 3.

[0119] Subsequently, the t ème element of the T used -tuple A = (s 1; ■■■ > $T used ) is denoted [4] t = a t = s t . In this context a T used -tuple whose elements all belong to the same set {1,..,M} can also be considered as a vector.

[0120] To avoid these implementation problems, a sequential allocation strategy can be implemented. Thus, when allocating the T used -tuple A, the sources a t are allocated sequentially which leads to a convenient allocation where there is no need for an exponential search on the T used-tuples A. In such a case, and when calculating the optimal flow rate corresponding to the set A, only one flow rate is then updated; that of the source

[0121] An example of a sequential allocation algorithm is shown in Appendix 4.

[0122] In a step E4 prior to the transmission phase, the destination (D) transmits to the N nodes of the system information relating to the allocation of transmission rates R^A) corresponding to the selected ordered set of sources A.

[0123] Such information may consist of an index representative of the allocated bitrate from a set of MCS modulation and coding schemes.

[0124] So, if the possible flow rates are R = {R o = 0, R lt ... , R MCS}, et ' eflow allocated to source i is Rt(A) = Rfc for k E {0, ...,MCS}, then it is sufficient to transmit the index k identifying this flow.

[0125] In a step E5 implemented during the transmission phase, the destination (D) receives messages sent by the M sources.

[0126] In a step E6 prior to the retransmission phase, the destination (D) transmits to the N nodes of the system information relating to the selected ordered set of sources A, thus indicating to each node of the system the number of time intervals T used constituting the retransmission phase and the identity of the sources to be helped.

[0127] Such information may consist, in a first example, of a first T used -tuple identifying the sources belonging to the selected ordered set of sources A.

[0128] So, for T used = T max= 2, and still with two sources to help, source 1 and source 2, if it is decided to help source 1 twice, then A = (1,1) and n = (2,0), if it is decided to help source 2 twice then, A = (2,2) and n = (0,2) and finally if it is decided to help source 1 once and source 2 once then A = (1,2) and n = (1,1).

[0129] Such information may consist, in a second example, of the transmission of an integer identifying the sources to be assisted and indicating the number of retransmissions per source.

[0130] So, for T used = T max = 2, and still with two sources to help, source 0 and source 1, if it is decided to help source 0 twice, A = (0,0), in this case the integer representing this situation is 0 + 0 = 0.

[0131] If it is decided to help source 1 in a first time interval, then source 0 in a second time interval, then A = (1,0), in this case the integer representing this situation is 1 + 0 = 1. If it is decided to help source 0 in a first time interval, then source 1 in a second time interval, then A = (1,0), in this case the integer representing this situation is = 0 + 2 = 2.

[0132] Finally, if it is decided to help source 1 twice, A = (1,1), in this case the integer representing this situation is 1 + 2 = 3.

[0133] More generally, the integer identifying the sources to be helped and indicating the number of retransmissions per source is Q such that Q = q0 1 where q t + 1 (I

[0134] {0, ... , T used — 1}) is the identifier of a source taking its value in the set {1,..,M} associated with the i eme+ 1 retransmission time interval taking its value from the set {1, ... , T used}.

[0135] In a particular implementation of step E6, the selected ordered set of sources A is transmitted with the corresponding transmission rate allocation R^A) prior to the transmission phase.

[0136] Finally, in a step E7 implemented during the retransmission phase, the destination (D) receives at least one redundancy of the messages sent by the T used aided sources.

[0137] Knowledge of the CSI data for all the channels in the system is important in this solution, which determines for each possible ordered set of sources, an optimal transmission rate allocation. The latter is based on knowledge of the unavailability events of each source constituting an ordered set of sources. And since the solution is based on a parallel retransmission where all the relay nodes that have decoded a given source are activated to help the source in question, knowledge of the CSI data for all the channels in the system makes it possible to determine the "decoding sets" of the relay nodes in order to know which relay nodes are activated and what is the equivalent channel for a given ordered set of sources.

[0138] In other words, for each ordered set of sources, a different equivalent channel is obtained for each retransmission time slot. These equivalent channels allow the calculation of the equivalent mutual information and consequently the unavailability events.

[0139] For example, in a (2, 2, 1)-OMAI\ / IRC, the set of sources is S = {1,2}, and the set of relays is R = {3,4} and the destination is D. After the transmission phase, the decoding sets of the relay nodes and the destination are:

[0140] 5i,o = {1}< 52, O = {1}, 5 3 0 = {1}, S4, O = {2}, S DiQ = <p

[0141] If the destination chooses to help source 1 during the first time interval of the retransmission phase, we see that the first three relay nodes will be activated. The equivalent channel corresponding to these relay nodes results in equivalent mutual information expressed as follows: J 1 D (l) = log2(1 + 2p=i | h^DP) where h i D represents the channel gain in the link between relay node i and destination D. The destination used the relay nodes' decoding sets to calculate Ï 1 D (X) to identify which relay nodes to include in the summation due to the equivalent channel.

[0142] In order to identify which relay nodes are active in the first retransmission, the destination determines the decoding sets of each relay node after the transmission phase using the equations [R } > The destination then calculates / i,7 based on the gain of the indirect link channel h} j between source 1 and relay node j. For this, the destination uses the CSI data relating to the indirect channels. A similar procedure is implemented for the T U sed time intervals of the retransmission phase.

[0143] [Fig. 4] represents a destination belonging to an OMAMRC telecommunications system with M sources, possibly L relays and a destination, M > 2, L > 0 according to an embodiment of the invention. Such a destination is capable of implementing the transmission method according to Figure 3.

[0144] A destination may comprise at least one hardware processor 41, a storage unit 42, and at least one network interface 43 which are connected to each other through a bus 44. Of course, the constituent elements of the destination may be connected by means of a connection other than a bus.

[0145] The processor 41 controls the operations of the destination. The storage unit 42 stores at least one program for implementing the method according to an embodiment of the invention to be executed by the processor 41, and various data, such as parameters used for calculations performed by the processor 41, intermediate data of calculations performed by the processor 41, etc. The processor 41 may be formed by any known and suitable hardware or software, or by a combination of hardware and software. For example, the processor 41 may be formed by dedicated hardware such as a processing circuit, or by a programmable processing unit such as a central processing unit (Central Processing Unit) which executes a program stored in a memory thereof.

[0146] The storage unit 42 may be formed by any suitable means capable of storing the program(s) and data in a computer-readable manner. Examples of the storage unit 42 include non-transitory computer-readable storage media such as semiconductor memory devices, and magnetic, optical, or magneto-optical recording media loaded into a read-write unit.

[0147] Network interface 43 provides a connection between the destination and all nodes in the system.

[0148] 16

[0149] Annex 4

[0150] 5

Claims

CLAIMS 1. Method for transmitting successive messages forming a frame intended for an OMAMRC (“Orthogonal Multiple-Access Multiple-Relay Channel”) type telecommunications system with N nodes and a destination (D), the N nodes comprising M sources i, i E {1, ... , M] and L relays ( r i, with M > 2, L > 0, a frame comprising M time slots allocated to the successive transmission of a message by the M sources and T used < T max time slots allocated to at least one cooperative retransmission of a redundancy of a message transmitted by at least one source in one of said M time slots, said method comprising the following steps implemented by the destination (D) for a current frame: - selection, from a plurality of ordered sets of source identifiers A, of an ordered set of sources A comprising T usedidentifiers of sources for which cooperative retransmission is planned by all nodes knowing this source, a source identifier being associated with a corresponding time interval among the T used time slots allocated to cooperative retransmission, - transmission, to the N nodes and prior to the successive transmission of a message by the M sources, of information relating to an allocation of transmission rates per source Ri(A) determined for the ordered set of sources A selected, - transmission, to the N nodes and prior to the cooperative retransmission of a redundancy of a message, of information representing the selected set of sources A.

2. Transmission method according to claim 1 in which the selection of the ordered set of sources A intended to be transmitted to the IV nodes is carried out in the following manner: Z / li Kif l-Oi T ) where nJ rame = - i - ' 11 7 represents the spectral efficiency of the current frame, a represents ' M+aT usea a ratio between the number of transmission channels available in the T used time slots and the number of transmission channels available in the M time slots and where B represents a set of ordered sets of sources A.

3. Transmission method according to claim 1 or 2 in which the determination of the ordered sets of source identifiers A comprises the following steps: a) initialization of an ordered set 40= 0, b) generation of an ordered set of current sources Aj +1 by adding, to the previous ordered set of sources Aj, a source i for which a cooperative retransmission is planned during a corresponding time interval among the T max time intervals, such as Aj +1= (Aj, i), said source i being selected from the M sources of the system according to the ordered set of sources Ap c) repetition of step b) until T is obtained max ordered sets of source identifiers A.

4. Transmission method according to claim 3 in which the source i selected from the M sources of the system according to the ordered set of sources Aj is the source which maximizes the spectral efficiency 5. Transmission method according to any one of claims 1 to 3, in which, for a given ordered set of sources A, the allocation of transmission rates Rt(A~) is carried out in the following manner: - for a source i whose identifier is not included in A, then R L (4) = I i D Yes i D represents the mutual information between source i and destination (D), - for a source i whose identifier is included in A, then Rt(A) = argmax R^ , knowing that O CTused (4) = 0 where R t is the initial transmission rate of source i, ni represents the number of cooperative retransmissions of source i in 4 , a represents a ratio between the number of transmission channels available in the T used time slots and the number of transmission channels available in the M time slots, the lower bound I i D + a ■ ni ■ I i D corresponds to a situation in which no other node than source i knows the message transmitted by source i, the upper bound I i D + a ■ n t ■ J* t D corresponds to a situation in which all nodes know the message transmitted by source i with J* i D the equivalent mutual information between source i and destination (D) when all nodes know the message transmitted by source i, and where 0; r used(4) represents a cut-off event worth 0 when the destination cannot decode the message sent by source i in the current frame and worth 1 when the destination decodes the message sent by source i in the current frame.

6. Transmission method according to claim 5 in which an initial value of R i 0 being defined as being (4) = 0, then the lower bound takes the value R i 0 , if for R t 0 , 0 È r used (4) = 1, then the upper bound takes the value R i 0 , the next iteration is then implemented until the difference between a current value of the upper bound and a current value of the lower bound is less than threshold: upper bound a-1 — lower bound a-1 Ri, a = if for R i a , 0 CTused (4) = 0, then the lower bound takes the value R i a , if for R i a , 0 CTused(4) = 1, then the upper bound takes the value R i a .

7. Transmission method according to claims 2 and 3 in which B = {40,4!, ... , A TMax}.

8. Transmission method according to claim 2 in which B represents the set of ordered sets of sources 4 for T used time intervals such that B = {4 e {1, ... , M] Tused : T used e {0, ...,T max}}.

9. Method according to any one of claims 1 to 8 wherein said information relating to the selected ordered set of sources 4 comprises a first T used -tuple identifying the sources belonging to the selected ordered set of sources 4.

10. Method according to any one of claims 1 to 8 wherein said information relating to the selected ordered set of sources 4 comprises at least one integer Q such that is the identifier of a source taking its value in the set {1,..,M} associated with i eme + 1 retransmission time interval taking its value from the set {1, ... , T used}.

11. OMAMRC telecommunication system with N nodes and one destination (D), the N nodes comprising M sources i, i E {1, and L relays (ÎJ ■■■> r L ) with M > 2, L > 0, suitable for implementing a transmission method according to one of claims 1 to 10.

12. A computer program product comprising program code instructions for implementing a transmission method according to claim 1, when executed by a processor.

Images