METHOD FOR CALCULATING A WORST TRANSMISSION TIME, COMPUTER PROGRAM AND ASSOCIATED COMPUTER SYSTEM

Abstract

The process involves, with the worst transmission time initially taken as the current worst transmission time, repeating the following step to obtain a dependency tree of worst transmission times: - for each current worst transmission time, one or more worst transmission times are identified for one of the flows (V1, V2, V3) from its starting node (e1, e2, e3) to another node on its path, on which the current worst transmission time depends. The worst transmission time(s) thus identified become the current worst transmission time(s). After a certain number of iterations, each current worst transmission time is fixed to an arbitrary value, that is, independent of the worst transmission time(s) on which the current worst transmission time depends, and the desired worst transmission time is calculated by tracing back up the dependency tree from these arbitrary values.Figure for the abridged version: Fig. 1.

Description

Title of the invention: METHOD FOR CALCULATING A WORST TRANSMISSION TIME, ASSOCIATED COMPUTER PROGRAM AND COMPUTER SYSTEM Technical field of the invention

[0001] The present invention relates to a method for calculating a worst-case transmission time, as well as to an associated computer program and computer system. Technological background

[0002] Current avionics networks operate with the ARINC 664 protocol known as AFDX and with centralized networks in star topology. This operation induces a significant cost due to the use of central switches. Indeed, these switches represent a significant volume and mass within an avionics system, limiting the deployment of this type of network within light avionics systems (drone, helicopter, twin-engine, etc.).

[0003] One solution is to propose a distributed network architecture with a mesh topology that can provide satisfactory cost, weight and volume characteristics for lightweight avionics systems. However, such an architecture must also offer the same security characteristics, particularly in terms of meeting data transmission deadlines, as known systems.

[0004] For this, it is necessary to use a method for calculating upper bounds of worst transmission times for data flows crossing the network.

[0005] A known method is the trajectory approach in which, the worst transmission time sought being initially taken as the current worst transmission time, a reiteration of the following step to obtain a dependency tree of worst transmission times: - for each worst current transmission time, an identification of one or more worst transmission times of one of the flows from its starting node to another node on its path, on which the worst current transmission time depends, the worst transmission time(s) thus identified becoming the worst current transmission time(s).

[0006] However, due to its recursion, this method is not applicable to a complex network such as a mesh structure network.

[0007] It may thus be desirable to provide a method which makes it possible to overcome at least some of the aforementioned problems and constraints. Summary of the invention

[0008] A first method is therefore proposed for calculating a worst transmission time of a flow crossing a network in an aircraft with other flows, the network comprising nodes and links between the nodes, each flow following a predefined path in the network, the method comprising, the worst transmission time sought being initially taken as the worst current transmission time, a reiteration of the following step to obtain a dependency tree of worst transmission times: - for each worst current transmission time, an identification of one or more worst transmission times of one of the flows from its starting node to another node of its path, on which the worst current transmission time depends, the worst transmission time(s) thus identified becoming the worst current transmission time(s); the method being characterized in that, after a certain number of iterations, each worst current transmission time is set to an arbitrary value, i.e. independent of the worst transmission time(s) on which the worst current transmission time depends, and in that the worst transmission time sought is calculated by going up the dependency tree from these arbitrary values.

[0009] The invention may further comprise one or more of the following optional features, in any technically possible combination.

[0010] Optionally, the arbitrary value is greater than one hour, more preferably greater than one year, more preferably greater than 100,000 years, more preferably greater than 500,000 years, for example equal to 2A64-1 microseconds.

[0011] Also optionally, the method further comprises a calculation of a maximum travel time of the path of the flow for which the worst transmission time is to be calculated, and the arbitrary value is taken equal to this maximum time.

[0012] Also optionally, the nodes of the path of the flow whose worst transmission time is sought have output queues, and the maximum time is calculated by assuming these output queues are loaded to allow only one frame of the flow considered to enter.

[0013] A second method for calculating a worst-case transmission time is also proposed, comprising setting a number of iterations to a predefined value, then repeating the following steps: - a calculation of the worst transmission time according to the first method, with the number of iterations; then - an increase, for example an increment of one, in the number of iterations.

[0014] Optionally, the reiteration is stopped when a convergence rate reaches a predefined threshold.

[0015] Also optionally, the reiteration is stopped when the number of iterations reaches a predefined limit.

[0016] A method for calculating a worst case transmission time is also proposed, comprising: - a calculation of the worst transmission time according to a method according to the invention, with a certain arbitrary value; - a determination of whether the worst calculated transmission time increases or stagnates with each reiteration; - if the calculated worst transmission time increases or stagnates at each reiteration, the arbitrary value is increased, for example doubled, to restart the previous steps; - if the worst calculated transmission time decreases at each reiteration, the worst transmission time calculated at the last reiteration is provided as the result of the process.

[0017] A third method for calculating a worst transmission time is also proposed, comprising a calculation of the worst transmission time sought according to the first method or the second method, then a repetition of the following steps: - a setting of the arbitrary value at the worst calculated transmission time; then - a calculation of the worst transmission time sought according to the first method or the second method, with the arbitrary value as set in the previous step.

[0018] There is also provided a computer program downloadable from a communication network and / or recorded on a computer-readable medium, characterized in that it comprises instructions for executing the steps of a method according to the invention, when said program is executed on a computer.

[0019] There is also provided a computer system comprising a processing unit and a computer program according to the invention, for executing the steps of a method according to the invention, when said program is executed by the processing unit. Brief description of the figures

[0020] The invention will be better understood with the aid of the following description, given solely by way of example and with reference to the appended drawings in which: - [Fig. 1] is an example of an aircraft network in which the invention can be implemented, - [Fig.2] is a block diagram of a method for calculating a worst-case transmission time of a stream, - [Fig.3] is a dependency tree obtained by implementing the method of [Fig.2] to the network of [Fig.l], - [Fig.4] is a block diagram of another method for calculating a worst-case transmission time of a stream, - [Fig.5] is a block diagram of another method for calculating a worst-case transmission time of a stream, - [Fig.6] is a simplified view of a computer system designed to implement one or more of the methods of Figures 2, 4 and 5, and - [Fig.7] is a part of another dependency tree obtainable by a method according to the invention. Detailed description of the invention

[0021] The invention is intended to be applied in a data transmission network of an aircraft, such as the network 100 illustrated in [Fig.l].

[0022] Generally speaking, a network consists of nodes and links between these nodes. Flows cross the network along respective predefined paths. Each path is a sequence of network nodes connected by network links. It is possible for a flow to have several paths, for example to be used actively in parallel (frames are sent at the same time on the paths).

[0023] In the illustrated example, the network 100 comprises the nodes: el, e2, e3, S0, SI, S2, si, s2, s3, and three flows are considered, denoted respectively VI, V2, V3. The flow VI follows the path [el S0 SI S2 si]. The flow V2 follows the path [e2 SI S2 S0 s2]. The flow V3 follows the path [e3 S2 S0 SI s3].

[0024] The worst transmission delay of a flow to a node of its path is the delay of a frame of this flow from end to end, that is to say from the starting node to the node considered, in the most unfavorable circumstances. The end-to-end delay is the sum of the crossing times of each node of the path and the transmission times of the links used. In particular, the nodes may have output queues, so that the calculation of the worst transmission delay must take into account the presence or absence of the frames of the other flows in these queues to calculate the crossing times.

[0025] To calculate the worst transmission delay, there is a method called the "trajectory approach". The trajectory approach makes it possible to obtain a general formula giving the worst transmission time of a flow to a node on its path. There are several variants of the general formula, depending on the network parameters. However, in all cases, the general formula gives the worst transmission time of the flow considered as a function in particular of the worst transmission times of this flow and / or other flows to a certain node on their path. The trajectory approach jectory allows to calculate the worst transmission time of a single path at a time, so that, for a flow following several paths, the present invention is preferably used for each path.

[0026] For example, the general formula used may be as follows: (1+½ Vq+J + (|PZ|-1) LC, 1 \ L / i / / w I ; teP x i n / i where: i is the index of the flow Vi considered; last(i) is the node of the path at which the worst transmission time is calculated; F, is the set of flows (including the flow Vi) interfering with the flow Vi between its starting node and its last(i) node; Tj is the minimum delay between two frames of the flow Vj; Cj is the propagation time of a frame of the flow Vj on each link of the network (In this example, we consider that all the links of the network have the same flow rate.In the case where the links do not have the same transmission rates, we would have, by denoting I the link, Cj,l = Taille_maxj / bitrate_of_link_l); P / is the path of the flow Vi up to the node lastÇi); is the technical latency of each node, this technical latency including all the operations carried out in the node, in particular until the frame reaches the output queue (in the previous formula, this technical latency is taken identical for all the nodes); and Aij is the time interval in which a frame of the flow Vj must be emitted from the first output queue in common between the flows Vi and Vj along the path of Vi, to interfere with a frame of the flow Vi. .

[0027] Aij is given by: Atj^Smaxp*™J) is the first common output queue between flows Vi and Vj along Vi's path; Smin1- is the minimum time taken by a frame of flow Vi to travel from its starting node to node h of its path; M!- is the earliest time at which a frame of flow Vi can dequeue from node h's output queue; Smax1] is the maximum time taken by a frame of flow Vi to travel from its starting node to node h of its path.

[0028] Smax;1^ ^ and y) are given by: Smaxfirst^'^ = + C where ■ j) ~ 1 is 'c node preceding the node first(i, j) along the path of flow Vi.

[0029] In the case where i = j, and are directly calculable.

[0030] Thus, in summary, the worst transmission time of the flow Vi to the node

[0031]

[0032]

[0033]

[0034] last(î) of its path therefore depends, for each other interfering flow Vj, on the worst transmission times and With reference to [Fig.2], a method 200 for calculating a worst case transmission time of a stream will now be described. During a step 202, the worst transmission time sought is taken as the current worst transmission time. During a step 204, for each current worst transmission time, one or more worst transmission times necessary for calculating the current worst transmission time (i.e. on which the latter depends) are identified. Each worst transmission time necessary is that of a flow from its starting node to another node on its path. The worst transmission time(s) thus identified become the current worst transmission time(s). For example, a general formula from the trajectory approach, such as the one given above ([Math. 1]), is used to identify the worst-case transmission time(s) required.

[0035] Step 204 is repeated a certain number N of times, in order to obtain a dependency tree of worst transmission times, during so-called first-level iterations. The dependency tree thus comprises a root formed by the worst transmission time sought. The dependency tree further comprises nodes respectively formed by the worst transmission times on which the worst transmission time sought depends. Each node is located at a certain depth from the root, corresponding to the number of first-level iterations implemented to reach it.

[0036] At the end of this number N of first-level iterations, during a step 206, each current worst transmission time is set to an arbitrary value. An arbitrary value is a value independent of the worst transmission time(s) on which the current worst transmission time depends.

[0037] The arbitrary value is in particular a pessimistic bound, that is to say a value sufficiently large to ensure that it is greater than the worst replaced transmission time, for example the largest possible.

[0038] For example, the arbitrary value is chosen to be greater than one hour, more preferably greater than one year, more preferably greater than 100,000 years, more preferably greater than 500,000 years. For example, when the values ​​are recorded in 64 bits, including the value zero, the arbitrary value may be 264 - 1 ps.

[0039] In another example, the arbitrary value is calculated as the maximum travel time of the path of the flow whose worst transmission time is sought. For example, when the node crossing delay comes from the presence of queues of output queues from the nodes, the maximum travel time is calculated assuming that these queues are loaded to allow only one frame of the considered flow to enter.

[0040] During a step 208, the worst transmission time sought is calculated by going up the dependency tree from the arbitrary values.

[0041] To determine a suitable arbitrary value, it is possible to start from an arbitrary value that is not too high, for example one hour, and then to implement the method 200 successively several times, each time increasing the depth N of iterations. If the result decreases with the depth N, then the chosen arbitrary value is suitable, i.e. greater than the worst transmission time replaced. Otherwise, if the result increases or stagnates with the depth N, the chosen arbitrary value is not suitable, and a new higher arbitrary value is chosen instead and the method 200 is again implemented several times, each time increasing the depth N of iterations. For example, the new arbitrary value is taken to be equal to twice the previous arbitrary value.

[0042] An application of the method 200 to the network 100 of [Fig.l] will now be described, with reference to [Fig.3] which illustrates the dependency tree 300 obtained.

[0043] In this application, the worst transmission time of the flow VI (i = 1) to the node S2 is sought and the general formula indicated above is used. The worst transmission time sought is noted and initially taken (step 202) as the current worst transmission time.

[0044] Iteration 1 of step 204

[0045] The interfering flow(s) are identified. In this case, on the path of flow VI to node S2, flow V2 interferes at the output of node SI and flow V3 interferes at the output of node S0.

[0046] Thus, for the flow V2, the worst transmission time Wy^ depends on the worst transmission time VLV2)-1 _ y^si-i = Wyjct of the P'rc transmission time xTtTtïrst V2,V1)-1 TxrSl-1 WV2 =WV2 = wV2-

[0047] For the V3 flow, the worst transmission time depends on the worst transmission time ^&st(Vl,V3)-l _ ^SO-i _ and the worst transmission time wfast(VJV1)-1 = W»1 = WS2.

[0048] Each of the worst transmission times Wy2 and Wy\ is independent of other worst transmission times, i.e. it can be calculated without knowing other worst transmission times. Therefore, they are not retained in the dependency tree 300.

[0049] Thus, as illustrated in Figure 3, at the end of the first iteration, the de- pendance 300 is updated to indicate that the worst transmission time depends on the worst-case transmission times and Wy2 (of depth one in the dependency tree 300), the latter being taken as the current worst-case transmission times.

[0050] Iteration 2 of step 204

[0051] The worst current transmission time is considered.

[0052] On the path of flow V1 to node S0, flow V3 interferes at the exit of the node S0.

[0053] Thus, for the V3 flow, the worst transmission time depends on the worst transmission time 1,V3)-1 „ and the worst transmission time Xï rfil'SOV3.V1 )-l yxrSO-1 w v3 ■ = w V3 =w V3 -

[0054] The worst transmission time is independent of other worst transmission times, i.e. it can be calculated without knowing other worst transmission times. Therefore, it is not retained in the dependency tree 300.

[0055] Thus, as illustrated in Figure 3, at the end of the second iteration, the dependency tree 300 is updated to indicate that the worst transmission time Wyj depends on the worst transmission time Wy2 •

[0056] The worst current transmission time Wy2 is then considered.

[0057] On the path of flow V3 to node S2, flow V2 interferes at the exit of the node S2.

[0058] Thus, for the flow V2, the worst transmission time Wy2 depends on the worst transmission time ^3^2)-1 _ ^^2-1 ~ Wy3 ct P^e transmission time AA / fustl y )-l \x / S2-1 w v2 • =W V2 = w V2 -

[0059] The worst transmission time W^3 is independent of other worst transmission times, i.e. it can be calculated without knowing other worst transmission times. Therefore, it is not retained in the dependency tree 300.

[0060] Thus, as illustrated in Figure 3, at the end of the second iteration, the dependency tree 300 is updated to indicate that the worst transmission time W'^ depends on the worst transmission time

[0061] The worst transmission times Wy3 and Wy2 identified (of depth two in the dependency tree 300) are then taken as the current worst transmission times.

[0062] Iteration 3 of step 204

[0063] The worst current transmission time Wy2 is considered.

[0064] On the path of flow V2 to node SI, flow V1 interferes at the exit of the node SL

[0065] Thus, for flow VI, the worst transmission time depends on the worst transmission time W^1'— Wy^ — ct P'rc transmission time wÇrst(VtV2)-l = WS1-1 = wæ.

[0066] The worst transmission time wf is independent of other worst transmission times, i.e. it can be calculated without knowing other worst transmission times. Therefore, it is not retained in the dependency tree 300.

[0067] Thus, as illustrated in Figure 3, at the end of the third iteration, the dependency tree 300 is updated to indicate that the worst transmission time Wy? depends on the worst transmission time Wyy

[0068] The worst current transmission time Wy3 is then considered.

[0069] Its dependency has already been previously identified. Thus, as illustrated in Figure 3, at the end of the third iteration, the dependency tree 300 is updated to indicate that the worst transmission time depends on the worst transmission time WwL vZ

[0070] At this stage, it is possible to notice that there is a dependency loop: the worst transmission time Wy| depends on the worst transmission time which depends on the worst transmission time Wyl which depends on the worst transmission time Wyr. Thus, by continuing the dependency search in this way, the calculation of the worst transmission time Wy would continue indefinitely as illustrated in [Fig.3] by the dotted lines, and therefore could not succeed.

[0071] This is why, in the method 200, it is provided in step 206 to set each worst transmission time remaining at the end of the number N of first-level iterations to the arbitrary value. For example, when this number N of first-level iterations is equal to 3, the worst transmission times W® and Wy2 (of depth three in the dependency tree 300) are set to the arbitrary value.

[0072] During step 208, the worst transmission time sought is calculated by going up the dependency tree 300 from the arbitrary values, in particular using the general formula. Thus, in the example of FIG. 3, the worst transmission times Wy? and Wy of the second level of the dependency tree 300 are respectively calculated from the arbitrary value. Then, the worst transmission times Wy3 and W® of the first level of the dependency chain are calculated from the worst transmission times Wy2 and Wy of the second level of the dependency tree 300 previously calculated. Then, the worst transmission time Wyi is calculated from the worst transmission times W® and Wyi of the first level of the dependency chain previously calculated.

[0073] With reference to [Fig.4], a method 400 for calculating a worst-case transmission time of a stream will now be described.

[0074] During a step 402, the number N of first level iterations is set to a predefined value, for example 1.

[0075] The following steps 404 to 410 are then repeated during so-called second-level iterations.

[0076] During a step 404, the method 200 is implemented, using the number N of first-level iterations, to calculate the worst transmission time sought.

[0077] During a step 406, from the second iteration of second level, a convergence rate of the worst transmission time sought is calculated and compared to a threshold. The convergence rate is for example calculated from a difference between the worst transmission time calculated at the current second level iteration and the worst transmission time calculated at the previous second level iteration. For example, the convergence rate is taken equal to this difference. Alternatively, the convergence rate can be calculated from a ratio between the worst transmission time calculated at the current second level iteration and the worst transmission time calculated at the previous second level iteration. For example, the convergence rate is taken equal to this ratio. When the threshold is reached, the reiteration of steps 404 to 410 is stopped.

[0078] During a step 408, the number N of first-level iterations is compared to a limit. When this limit is reached, the repetition of steps 404 to 410 is stopped. This prevents the number N of first-level iterations from becoming too large. The method then provides the worst transmission time sought as calculated in step 406.

[0079] During a step 410, the number N of first-level iterations is increased, for example incremented by one. Thus, at the next iteration, the method 200 is implemented with a greater number of first-level iterations. The method then provides the worst transmission time sought as calculated in step 406.

[0080] Gradually increasing the number N of first-level iterations makes it possible to avoid using too high a number N of first-level iterations when this is not necessary, and therefore to save execution time.

[0081] With reference to [Fig.5], a method 500 for calculating a worst-case transmission time of a stream will now be described.

[0082] During a step 502, the worst transmission time sought is calculated by implementing the method 200 or the method 400, for example by taking a very high value as an arbitrary value.

[0083] The following steps 504 and 506 are then repeated, during so-called third-level iterations.

[0084] During a step 504, the arbitrary value is set to the worst calculated transmission time.

[0085] During a step 506, the worst transmission time sought is again calculated by implementing the method 200 or the method 400, with the arbitrary value as set in the previous step 504.

[0086] An example of implementation of the method 500 will now be described.

[0087] During step 502, the worst transmission time sought, denoted W, is calculated with the arbitrary value taken at a very high value. The result is denoted W(oo). At the first third-level iteration, the worst transmission time sought W is again calculated with the arbitrary value taken at W(oo). The result is denoted W(W(oo)). At the second third-level iteration, the worst transmission time sought W is again calculated with the arbitrary value taken at W(W(oo)) and so on. The third-level iterations are for example stopped using a convergence rate, as described previously.

[0088] With reference to [Fig.6], an exemplary computer system 600 designed to implement each of the methods 200, 400, 500 will now be described.

[0089] The computer system 600 comprises a data processing unit 602 (such as a microprocessor) and a main memory 604 (such as a RAM memory, from the English "Random Access Memory") accessible by the processing unit 602. The computer system 600 further comprises for example a network interface and / or a computer-readable medium, such as for example a local medium (such as a local hard disk 606) or a remote medium (such as a remote hard disk and accessible via the network interface through a communication network) or even a removable medium (such as a USB key, from the English "Universal Serial Bus", or a CD, from the English "Compact Disc" or a DVD, from the English "Digital Versatile Disc") readable by means of an appropriate reader of the computer system 600 (such as a USB port or a CD and / or DVD disk reader).A computer program 608 containing instructions for the processing unit 602 is recorded on the medium 606 and / or downloadable via the network interface. This computer program 608 is for example intended to be loaded into the main memory 604, so that the processing unit 602 executes its instructions to implement the method 200, 400 or 500.

[0090] Alternatively, all or part of the computer program 608 could be implemented in the form of an electronic circuit, for example micro-wired, not involving a computer program.

[0091] A part of another dependency tree that can be obtained by a method according to the invention is illustrated in Figure 7. The arbitrary value applied to the first level iteration N is noted A. The same worst transmission time can appear several times at the same depth in the dependency tree. This is the case of the worst transmission time Wa which appears twice at depth N-3. In this case, the worst transmission time is preferably calculated only once and its value is recorded. This recorded value is thus reused for the other occurrence(s) at the same depth. The same worst transmission time can also appear at different depths in the dependency tree. This is the case of the worst transmission time Wa which also appears at depth N1. In this case, the worst transmission time must be recalculated. Indeed, in the latter case, the arbitrary value A is applied at different distances from the worst transmission time Wa along the dependency tree, so that the value of this worst transmission time Wa must a priori be different.

[0092] In conclusion, it will also be noted that the invention is not limited to the embodiments described above. It will indeed appear to those skilled in the art that various modifications can be made to the embodiments described above, in light of the teaching which has just been disclosed to them.

[0093] In the detailed presentation of the invention which is made above, the terms used must not be interpreted as limiting the invention to the embodiments set out in the present description, but must be interpreted to include all equivalents whose prediction is within the reach of those skilled in the art by applying their general knowledge to the implementation of the teaching which has just been disclosed to them.

Claims

1.

2.

3.

4. Claims Method (200) for calculating a worst transmission time (W^) of a flow (VI) crossing a network (100) in an aircraft with other flows (V2, V3), the network (100) comprising nodes (el, e2, e3, S0, SI, S2, si, s2, s3) and links between the nodes (el, e2, e3, S0, SI, S2, si, s2, s3), each flow (VI, V2, V3) following a predefined path in the network (100), the method (200) comprising, the worst transmission time sought being initially taken (202) as the current worst transmission time, a reiteration of the following step to obtain a dependency tree (300) of worst transmission times: - for each worst current transmission time, an identification (204) of one or more worst transmission times of one of the flows (VI, V2, V3) from its starting node (el, e2, e3) to another node on its path, on which the worst current transmission time depends, the worst transmission time(s) thus identified becoming the worst current transmission time(s); the method (200) being characterized in that, after a certain number of iterations, each worst current transmission time is set to an arbitrary value, i.e. independent of the worst transmission time(s) on which the worst current transmission time depends, and in that the worst transmission time (W^j) sought is calculated by going up the dependency tree (300) from these arbitrary values. Method (200) according to claim 1, in which the arbitrary value is greater than one hour, more preferably greater than one year, more preferably greater than 100,000 years, more preferably greater than 500,000 years, for example equal to 2A64-1 microseconds. Method (200) according to claim 1 or 2, further comprising a calculation of a maximum travel time of the path of the flow (VI) of which the worst transmission time (W^) is to be calculated, and in which the arbitrary value is taken equal to this maximum time. Method (200) according to claim 3, in which the nodes (el, S0, SI, S2, si) of the path of the flow (VI) whose worst transmission time (iy®) is sought have output queues, and in which the maximum time is calculated by assuming these output queues loaded to allow only one frame of the stream (VI) considered to enter.

5. Method (400) for calculating a worst transmission time, comprising setting a number of iterations to a predefined value, then repeating the following steps: - calculating (404) the worst transmission time according to a method according to any one of claims 1 to 4, with the number of iterations; then - increasing (410), for example increasing by one, the number of iterations.

6. The method (400) of claim 5, wherein the reiteration is stopped when a convergence rate reaches a predefined threshold.

7. Method (400) according to claim 5 or 6, wherein the reiteration is stopped when the number of iterations reaches a predefined limit.

8. A method for calculating a worst-case transmission time, comprising: - calculating the worst-case transmission time according to a method (400) according to any one of claims 5 to 7, with a certain arbitrary value; - determining whether the calculated worst-case transmission time increases or stagnates at each reiteration; - if the calculated worst-case transmission time increases or stagnates at each reiteration, the arbitrary value is increased, for example doubled, to restart the previous steps; - if the calculated worst-case transmission time decreases at each reiteration, the calculated worst-case transmission time at the last reiteration is provided as a result of the method.

9. Method (500) for calculating a worst transmission time, comprising a calculation (502) of the worst transmission time sought according to a method according to any one of claims 1 to 7, then a reiteration of the following steps: - a fixing (504) of the arbitrary value at the worst transmission time calculated; then - a calculation (506) of the worst transmission time sought according to a method according to any one of claims 1 to 7, with the arbitrary value as set in the previous step (504).

10. Computer program (608) downloadable from a communications network and / or recorded on a computer-readable medium, characterized in that it comprises instructions for executing the steps of a method (200, 400, 500) according to any one of claims 1 to 9, when said program is executed on a computer.

11. A computer system comprising a processing unit (602) and a computer program (608) according to claim 10, for executing the steps of a method (200, 400, 500) according to any one of claims 1 to 9, when said program is executed by the processing unit (602).

Images