Method and device for dynamic resource management in a mesh wireless telecommunications network

Abstract

The invention relates to a device and a method for dynamic resource management in a linear or star mesh wireless network, which makes it possible to determine the frequency bands to be used for the different links of the linear mesh wireless network between a rear base BA and the mission vehicle(s) VM, with a view to minimizing interference between the different links and optimizing the performance of the network. The method makes it possible to determine the reconfigurations of these frequency bands to be implemented within the linear mesh network in the event of the addition or removal of a relay when a VM is moved, and in the event of the addition or removal of a VM. Figure for abstract: Fig. 8

Description

Title of the invention: Method and device for dynamic resource management in a mesh wireless telecommunications network Field of invention

[0001] The invention is in the technical field of telecommunications networks, and relates more particularly to a method for dynamically managing resources in a mesh wireless telecommunications network interconnecting a rear base (BA) to a mission vehicle (VM) via a set of wireless relay (R) equipment deployed dynamically during the movement of the VM in order to maintain quality bidirectional connectivity between the BA and the VM. State of the Art

[0002] The technical problem addressed by the present invention is that of maintaining quality wireless connectivity between a rear base (BA) and a mission vehicle (VM) interconnected by a set of relays (R) forming a linear mesh wireless network between BA and VM.

[0003] Traditionally, in mesh wireless networks, for example WiFi mesh networks or others, the devices use omnidirectional antennas (i.e. capable of transmitting in all directions) and no frequency band management strategy is implemented. Thus, all the links in the network use the same frequency band and are therefore likely to interfere with each other. This interference between the different links in the network (sometimes called co-interference) causes a significant degradation in the quality of communications within the network, particularly when the amount of data to be transmitted increases. This is explained by the fact that several devices (BA, VM or R) are likely to use the same resources (i.e. a frequency band common to all) simultaneously, thus creating collisions (or interference) between the different packets transmitted by these devices.This prevents proper reception of these packets, and therefore a loss of associated data, and this may require retransmission of these packets, further increasing the network load.

[0004] Typically in a linear mesh wireless network where all the links use the same frequency band, it can be considered that the bandwidth of the network which is obtained will be at best divided by 3, that is to say at best equal to 1 / 3 of the nominal bandwidth of a link, and this in the most favorable case where the distance between the different devices (BA, VM, R) is close to the maximum range of the communication link.

[0005] In the case of a denser linear mesh wireless network, typically if the distance between the devices (BA, VM, R) is much less than the maximum range of the communication link, the flow rates are reduced even more significantly.

[0006] This can be illustrated on the linear mesh wireless network shown in [Fig. 1]. When the transmission of a data file (102) from a source, for example a BA, to a destination, for example a VM, is in progress on hop 'n', it is not possible to have transmissions in progress on hops 'n+1' and 'n+2'. In other words, at any given time, at most 1 hop out of 3 can be used to transmit a fragment. This is understandable if we consider that a simultaneous transmission from 2 consecutive relays would necessarily involve an interference problem at the level of these relays. In the same way, if we consider a simultaneous transmission from 2 relays separated from each other by an intermediate relay, for example relays 2 and 4 in [Fig.l], relay 3 which separates them simultaneously receives signals transmitted by its 2 immediate neighbors, and it can then decode neither one nor the other because each constitutes interference for the other. A "collision" occurred.

[0007] This degradation of quality of service is even more significant in a star-topology network, consisting of a central node (typically a BA) and several branches extending from this central node (typically several separate wireless mesh networks to different VMs). When all the links in this star-topology wireless mesh network use the same frequency band, the risk of interference between nodes is exacerbated, particularly between nodes near the BA, which then further degrades the quality of communications within the network. Thus, the greater the number of branches, the more the performance is degraded.

[0008] Solutions address the problem of quality of service degradation in linear mesh wireless networks.

[0009] Patent application [1] No. FR1362391 entitled “Deployment of an ad-hoc network” and patent application [2] No. FR2013143 entitled “Method and device for automatic deployment of a communication network”, each propose a different method for automatic deployment of a linear wireless mesh network between a rear base (BA) and a mission vehicle (VM) in order to maintain quality bidirectional connectivity between a BA and a VM and relying on a set of wireless relays dynamically deployed during the movement of the VM along a trajectory between the VM and B A. In application [1], the deployment is ensured by the VM, which “deposits” communication relays constituting the linear wireless mesh network, whereas in application [2] the communication relays are robotic and autonomous (RCRA) and capable of moving autonomously along a trajectory followed by the VM. However, the problem of management frequency bands to be used within the linear wireless mesh network in order to optimize its operation and performance is not addressed in these two documents.

[0010] Patent application [3] No. FR2207559 entitled “Method and device for automatic redeployment of the geographical coverage of a wireless telecommunications network”, proposes a method for interconnecting several VMs to the same BA using separate linear mesh networks associated with each of the VMs, thus forming a star-topology mesh wireless network. The VMs are each equipped with a radio base station designed to provide radio coverage in an area in the vicinity of the VM. This document addresses the problem of allocating radio frequencies associated with the radio base stations embedded on separate VMs in order to avoid interference situations between these base stations when the VMs are neighbors. However, the problem of managing the frequency bands to be used within the star-topology mesh wireless network (i.e.the “internal” network interconnecting the BA to the different VMs via the different relays) is not explicitly addressed.

[0011] Patent application [4] No. FR3087989 entitled “Method and device for wireless data transmission in networks”, proposes a method for controlling / coordinating the sending of messages within a linear mesh wireless network in order to reduce co-interference between the different links. The method assumes that the different links use the same frequency band (and can therefore potentially interfere with each other), and proposes a set of control messages between the nodes of the linear mesh wireless network allowing each to determine when it is entitled to transmit data. However, this solution cannot achieve a throughput greater than 1 / 3 the nominal bandwidth of a link. Furthermore, this method is not applicable to the case of mesh wireless networks with a star topology.

[0012] Patent application [5] No. FR1906976 entitled “Method and device for configuring vehicle-to-vehicle communication”, proposes a method for configuring a linear mesh wireless network between autonomous vehicles forming a convoy. Each vehicle in the convoy (i.e. each relay in the linear mesh wireless network) is equipped with two transmission antennas, assumed to be omnidirectional, one connected to the preceding vehicle / relay and the other to the following vehicle / relay. This solution proposes allocating different frequency bands to each of the different links in the linear wireless network in order to avoid any risk of interference between these links, and therefore to maximize the bandwidth available within the network. However, document [5] does not specify how to obtain these different frequency bands, and in particular it does not address the issue of resource sharing (share a predefined frequency band) between the different links of the linear mesh wireless network. Document [5] does not mention how to determine which frequency bands to allocate to each of the network links to avoid interference and optimize bandwidth. Furthermore, in document [5], since the number of vehicles to be interconnected is fixed, the topology of the linear mesh wireless network remains unchanged over time, and the issue of reconfiguring frequency bands used within a network when adding or removing a relay (here a vehicle) is not addressed, nor is the issue of frequency management addressed for star-topology mesh wireless networks.

[0013] Also, there is a need for a solution that overcomes the drawbacks of known approaches to address the general problem of managing frequency bands to be used within a linear mesh or star topology wireless network, interconnecting a rear base (BA) to a mission vehicle (VM) and relying on a set of wireless relays, dynamically deployed during the movement of the VM along a trajectory between the VM and BA, in order to maintain quality (bidirectional) connectivity between BA and VM.

[0014] Furthermore, in the particular context of the invention where a mission vehicle is moving, there is a need to be able to reconfigure the frequency bands used so that the performance of the network remains optimal, when the number of links evolves between a rear base and a mission vehicle, typically if a relay is added because the VM is moving away from the BA or if a relay is removed because the VM is moving closer to the BA.

[0015] The present invention meets these needs by proposing a solution for determining the frequency bands to be used for the different links of the network in order to minimize interference between the different links and optimize the performance of the network, i.e. maximize the throughput, minimize latencies and packet losses. Summary of the invention

[0016] An object of the present invention is a method for dynamically managing resources, in particular frequencies, in a linear mesh wireless network interconnecting a rear base (BA) to a mission vehicle (VM) and relying on a set of wireless relays (R) deployed dynamically during the movement of the VM, along the trajectory between the VM and the BA, with a view to maintaining quality connectivity between the BA and the VM.

[0017] Advantageously, the method of the invention can be applied whether the relays are deployed or deposited by a VM during its movement or whether they are autonomous robotic communication relays (RCRA) capable of moving autonomously along a trajectory followed by a VM.

[0018] Advantageously, the method of the invention makes it possible to determine frequency bands to be used for the different links of a linear mesh wireless network between a rear base BA and a mission vehicle VM, via different relays, with a view to minimizing interference between these different links and optimizing the performance of the network, i.e. maximizing the throughput, minimizing latencies and packet losses.

[0019] Another object of the present invention is to address cases where a rear base is common and simultaneously connected to several mission vehicles using separate linear mesh networks each associated with a VM. In this case, the method of the invention makes it possible to jointly determine frequency bands to be used for the different links of these multiple linear mesh wireless networks between a common BA and the different VMs (via the relays). Advantageously, this makes it possible to minimize interference between the different links and to optimize performance (maximize throughput, minimize latencies and packet losses) within this mesh wireless network forming a star topology.

[0020] Another object of the invention is to determine the reconfigurations of the frequency bands to be put in place within a linear mesh wireless network or a star-topology mesh wireless network, in the event of movement of a VM on a trajectory (and therefore with addition or removal of a relay), or in the event of addition or removal of a VM (and therefore addition or removal of a branch to the star-topology network).

[0021] The fields of industrial application of the invention are multiple, and target robotic solutions for inspection, surveillance, intervention for security and defense needs, first aid, or industrial needs such as nuclear dismantling. The present invention is of interest to suppliers of telecommunications solutions, particularly for constrained environments.

[0022] To obtain the desired results, a method is proposed for allocating frequencies in a mesh wireless telecommunications network, the network operating on an initial global frequency band to serve all communications and the different links within the network, the network comprising at least one rear base BA equipped with at least one antenna, one or more mission vehicles VM each equipped with an antenna for connecting to a relay in the direction of the BA, a plurality of relays R each equipped with two antennas, a first for connecting to a previous relay or to a VM and a second for connecting to a following relay or to a rear base.

[0023] The method comprises steps consisting of: - determining whether the network topology is a linear network topology having a single branch to interconnect a single mission vehicle to a rear base via at least one relay, or a star network topology having a plurality 'a' of branches to interconnect a plurality 'a' of different mission vehicles to a common rear base, at least one branch having at least one relay; - partition the initial global frequency band into a number of distinct sub-bands of the same width, the number of sub-bands depending on the network topology; and - cyclically and alternately allocating the sub-bands to the links of the branch or to the links of the plurality of branches, such that two consecutive links use different sub-bands.

[0024] The method may operate according to alternative or combined embodiments such as below.

[0025] In one embodiment, the step of partitioning the initial global frequency band consists for a linear network in partitioning said band into a basic sub-band and a complementary sub-band, or consists for a star network in partitioning said band into a basic sub-band and 'a' complementary sub-bands.

[0026] In one embodiment, the method further comprises, after the partitioning step if the network is star-shaped, steps consisting of: - constitute 'a' pairs of sub-bands with the baseband and the complementary sub-bands, such that each of the pairs includes the baseband and one of the complementary sub-bands; - assign each of the pairs of sub-bands to a different branch of the network; and - for each branch of the network, allocate cyclically and alternately, the two sub-bands of the pair corresponding to said branch to the different links of the branch, so that two consecutive links use different sub-bands and the base sub-band is not used by the antenna of the rear base serving the first link of said branch.

[0027] In one embodiment, the method further comprises, after the sub-band allocation step, steps consisting of: - determine a variation in the network topology; and - reallocate frequency sub-bands based on network topology and the nature of topology variation.

[0028] In one embodiment, the step of determining a topology variation for a linear network includes determining whether a relay is added or removed from the branch, or determining whether a mission vehicle is added to create a new branch.

[0029] In one embodiment, the step of determining a topology variation for a star network includes determining whether a relay is added or removed from a branch, or whether a mission vehicle is added or removed from the network.

[0030] In one embodiment, the method further comprises, if a relay Rn is added between the rear base and a relay Rn-1, steps consisting of: - allocate to the link between the new relay Rn and the previous relay Rn-1, the sub-band which was used by the link between the rear base and the previous relay Rn-1; - allocate to the link between the rear base and the new relay Rn the sub-band complementary to that of the link between the new relay Rn and the previous relay Rn-1; - establish the wireless link between the rear base and the new relay Rn; and - establish the wireless link between the new relay Rn and the previous relay Rn-1.

[0031] In one embodiment, the link establishment steps include steps of configuring and adjusting the power of the antennas of the rear base, the new relay Rn and the previous relay Rn-1.

[0032] In one embodiment, the link establishment steps further comprise steps of orienting the antennas of the rear base, the new relay Rn and the previous relay Rn-1.

[0033] In one embodiment, the method further comprises, if a relay Rn is removed between the rear base and a previous relay Rn-1, steps of establishing a wireless link between the rear base and the previous relay Rn-1, said steps of configuring and orienting the antennas of the rear base and the previous relay Rn-1.

[0034] In one embodiment, the method further comprises, if the topology variation is due to an addition or removal of a mission vehicle, steps consisting of partitioning the overall frequency band into distinct sub-bands of the same width, and for each branch of the network, allocating to the different links of the branch in a cyclic and alternating manner, the two sub-bands associated with said branch, so that two consecutive links use different sub-bands.

[0035] In one embodiment, the method further comprises, if the topology variation is due to an addition or removal of a relay from a branch, steps consisting of shifting said sub-bands used on the different links of said branch so that the base sub-band is not used by the antenna of the rear base serving this branch.

[0036] The invention also relates to a device comprising means for implementing the steps of the method of the invention.

[0037] According to alternative embodiments of the device of the invention, the mission vehicle is a land or air or amphibious or aquatic vehicle, remotely piloted or self-piloted (without human intervention).

[0038] In an advantageous embodiment, the wireless communication network is implemented according to an SDN-oriented architecture for "Software Defined Networking" according to the established Anglicism, comprising an SDN controller commonly having a "North Interface" and a "South Interface", which is set up at central system level (at a rear base or a server connected to the rear network).

[0039] The invention also relates to a computer program product which comprises non-transitory code instructions making it possible to carry out the steps of the method of the invention, when the program is executed on a computer. Description of the figures

[0040] Other characteristics and advantages of the invention will appear with the aid of the following description and the figures of the appended drawings in which:

[0041] [Fig.l] illustrates a linear mesh wireless network established between a rear base and a mission vehicle with relays;

[0042] [Fig.2] illustrates different communication configurations between a mission vehicle and a rear base;

[0043] [Fig.3] illustrates an example of a context for implementing the method of the invention for a linear mesh wireless network consisting of a rear base and autonomous relays deployed along a trajectory of a mission vehicle;

[0044] [Fig.4] illustrates another example of a context for implementing the method of the invention for a mesh wireless network with a star topology interconnecting via relays, a rear base common to several mission vehicles equipped with radio base stations responsible for providing additional network coverage to mobile equipment in their vicinity;

[0045] [Fig.5] illustrates an example of partitioning a frequency band B into several sub-bands;

[0046] [Fig.6] illustrates an example of allocation of frequency sub-bands for the case of a linear mesh wireless network interconnecting a rear base via relays to a mission vehicle;

[0047] [Fig.7] illustrates an example of allocation of frequency sub-bands for the case of a star-topology mesh wireless network between a common rear base and several mission vehicles;

[0048] [Fig.8] is a flowchart of the steps of the initial frequency band allocation phase according to the method of the invention, in the context of a mesh wireless network;

[0049] [Fig.9] is a flowchart of the steps for re-allocating frequency sub-bands for a linear mesh wireless network, in the case of adding a relay;

[0050] [Fig. 10] is a flowchart of the steps for re-allocating frequency sub-bands for a linear mesh wireless network, in the case of a relay removal;

[0051] [Fig. 11] is a flowchart of the steps for re-allocating frequency sub-bands for a star-topology mesh wireless network, in the case of adding or removing a mission vehicle;

[0052] [Fig. 12] is a flowchart of the steps for re-allocating frequency sub-bands for a star-topology mesh wireless network, in the case of adding or removing a relay. Detailed description of the invention

[0053] In the remainder of the document, for reasons of simplicity of description and not of limitation, the embodiments presented are described in a context of use where the relays are autonomous robotic communication relays (RCRA) capable of moving autonomously along a trajectory followed by a VM.

[0054] Those skilled in the art can derive the principles described to other contexts of use.

[0055] In general, the invention is applicable in all linear or star-topology wireless mesh networks, and in particular when the topology of these networks is likely to evolve through the addition or removal of relays and / or mission vehicles. Examples of application contexts are illustrated by Figures 2, 3 and 4.

[0056] [Fig.2] illustrates different communication configurations between a mission vehicle VM and a rear base B A.

[0057] In the context of the invention, the expression “mission vehicle” (VM) is to be taken in its broadest sense and can designate any mobile machine, whether it be a land vehicle (202), maritime vehicle (204, 206), air vehicle (208), a civil or military vehicle, etc., which moves to carry out a mission.

[0058] The VM can be a vehicle piloted by a person or remotely operated or even an autonomous vehicle capable of moving alone in its environment according to a mission objective (e.g. following a predefined trajectory in a given area).

[0059] The VM can be a flying vehicle such as a drone (remotely operated or autonomous), an airplane, a land vehicle, whatever its mode of locomotion (wheels, tracks), a floating vehicle (a boat, a drone), an underwater vehicle.

[0060] A mission vehicle can be in charge of any type of missions such as rescue missions for the inspection of an area having suffered an incident (fire, flood, earthquake), for the search and location of victims, etc.

[0061] The missions of a VM can be for security, such as monitoring an area (border), detecting and locating security alerts (unauthorized entries into an area).

[0062] Still without being exhaustive, the missions of a VM can be industrial missions, for example to take charge of the dismantling of a critical installation (such as a nuclear site).

[0063] In the context of the invention, a “rear base” (RB) can designate any system configured to be in communication with a mission vehicle.

[0064] The BA can also take different forms such as, for example, a server (212) in the internet or a cloud platform responsible for receiving (and potentially storing and / or processing) data collected and transmitted by the VM during its movement.

[0065] A BA may be a control and supervision center (210), for example in charge of supervising one or more autonomous drones (fleet of drones). In this context, the data exchanges between the BA and the VM(s) may be, for example, mission instructions (such as a trajectory to follow) transmitted to the drone(s) or even be data (videos, photos, event detection alerts) collected by the drone(s) and transmitted to the BA.

[0066] A BA can also be a control console (or a remote control) for teleoperation of a drone or robot allowing the drone / robot to be remotely piloted and a video stream captured by the drone / robot to be viewed.

[0067] Furthermore, a BA can also be fixed or mobile, for example associated with a boat (214) responsible for communicating with submarines.

[0068] Advantageously, any type of wireless transmission technology can be used between a BA and a VM, such as for example radio transmissions in any type of RF band (ad-hoc WiFi transmission), transmissions on the visible part of the electromagnetic spectrum (LiFi transmissions) or even acoustic transmissions (underwater).

[0069] [Fig.3] illustrates an example of a context for implementing the method of the invention for a linear mesh wireless network consisting of a rear base (304) and a set (306) of autonomous relays deployed along a trajectory of a mission vehicle (302).

[0070] [Fig.3] shows a situation where the VM which has already evolved on its trajectory since its departure where it was in direct communication with the BA, required the deployment of three communication relays RCRA(l) (306-1), RCRA(2) (306-2), RCRA(3) (306-3) in order to maintain quality wireless connectivity between itself and the BA.

[0071] A device (RCRA) according to the invention is an entity capable of moving autonomously along the trajectory followed by the VM and parking at a position which is calculated to allow establishing, as a communication relay, a multi-hop linear wireless communication between the VM and the rear base. An RCRA depending on the applications can be a terrestrial device, an aerial device performing a hovering flight such as a drone, a floating or underwater device.

[0072] According to alternative embodiments, the RCRAs can be equipped with any type of sensors or actuators potentially useful to the VM in carrying out its mission. The VM can then, thanks to the linear wireless network, collect information from these additional sensors, thus enabling it to “increase” its perception along its past trajectory. In the same way, the VM can then, again thanks to the linear wireless network, control the actuators of the RCRAs and thus “increase” its capacity for action along its past trajectory.

[0073] In one embodiment, the RCRAs are grouped in a zone called a reserve zone (308) from which they can leave to position themselves autonomously at the planned position which is calculated during the movement of the VM.

[0074] [Fig.4] illustrates another example of a context for implementing the method of the invention for a star-topology mesh wireless network interconnecting a common rear base (402) via relay sets to several mission vehicles (404-1, 404-2, 404-3) equipped with radio base stations (VMSSF), each in charge of offering additional network coverage to a set of mobile equipment (EM: 406-1, 406-2, 406-3) potentially moving over long distances, but where a fixed communications infrastructure usable by the EM does not exist.

[0075] In such a scenario, temporary network coverage is deployed automatically (without human intervention) in the environment where the mobile equipment is operating, this network coverage itself being self-reconfigurable in order to offer at all times the network connectivity required by the EMs while minimizing the necessary resources. The network coverage thus provided by the solution therefore has the characteristic of being itself mobile, and with variable coverage and extent.

[0076] The present invention is advantageous, for this type of star topology of the mesh wireless network, for maintaining quality of service in the aforementioned cases: of rescue missions (providing wired network coverage in a potentially very large disaster zone, and devoid of pre-existing communication infrastructure, to allow different mobile responders to carry out their mission thanks to this temporary wireless network with self-reconfigurable coverage); of security missions (surveillance of an area by a group of mobile responders); of industrial missions (providing temporary wireless network coverage and self-reconfigurable coverage on a construction site or for the dismantling of a critical installation).

[0077] In the context of the invention, a (VMSSF) is a vehicle remotely operated automatically (i.e. without human intervention) from a rear base (BA), or from a server located in a rear network (410) and connected to the BA. The VMSSF can thus be automatically repositioned at a given geographical point, for example by indicating the geographical coordinates of this targeted point or even indicating a trajectory to follow.

[0078] A VMSSF is equipped with a wireless base station allowing it to provide wireless network connectivity in a geographic area near its location. Any type of wireless transmission technology can be used for this wireless network coverage offered by the VMSSF, such as for example radio transmissions in any type of RF band (WiFi, 4G, 5G, etc.), transmissions on the visible part of the electromagnetic spectrum (for example LiFi transmissions) or even acoustic transmissions (underwater).

[0079] The implementation of the method of the invention assumes that the mesh wireless network (linear or star topology) for which dynamic resource management is sought, has only one predefined global frequency band B to serve all communications and the different links within the network. The present invention makes it possible, in an initial phase, to partition this initial global band B into multiple sub-bands and to allocate the sub-bands to the different links of the mesh wireless network.

[0080] [Fig.5] illustrates an example of partitioning a frequency band B into six sub-bands (Bb, Bl, B2, B3, B4, B5). According to embodiments, the sub-bands may be of the same width or of variable width, the width of a sub-band directly impacting the throughput that can be offered by it, i.e. the greater the width, the higher the throughput can be.

[0081] [Fig.6] illustrates an example of allocation of frequency sub-bands for the case of a linear mesh wireless network interconnecting a rear base via three relays to a mission vehicle. After determining and allocating two frequency sub-bands Bb and Bl according to the method of the invention, the communications between the mission vehicle VM and the relay RI are made on the frequency sub-band Bb, the communications between the relay RI and the relay R2 are made on the frequency sub-band Bl, the communications between the relay R2 and the relay R3 are made on the frequency sub-band Bb, and the communications between the relay R3 and the rear base BA are made on the frequency sub-band BL

[0082] There is thus an alternation of the defined sub-bands Bb and Bl, making it possible to avoid or minimize interference between the links, and to optimize the performance of the network.

[0083] According to the example of [Fig.6], the method for determining and allocating frequency sub-bands consists of partitioning the initial band B into two sub-bands, respectively named Bb (for base band) and BL

[0084] In one embodiment, the sub-bands are chosen to be of the same width in order to maximize the end-to-end throughput within the linear mesh wireless network.

[0085] The sub-bands are then allocated cyclically and alternately to the different links of the linear mesh wireless network, so that two consecutive links use different sub-bands. This alternating allocation makes it possible to avoid any risk of interference between the two interfaces of the same relay.

[0086] Each relay R of a mesh wireless network is equipped with two antennas, a first to connect said relay to the previous relay (towards the VM) or to the VM directly if it is the first relay, and a second to connect said relay to the next relay (towards the BA) or to the BA directly if it is the last relay.

[0087] The VM is also equipped with an antenna to connect to its first relay in the direction of the BA.

[0088] Finally, the BA is equipped with at least one antenna. In the case of a linear network, only one antenna is used. In the case of a star topology network, each antenna of the BA (the BA being common to several branches of the network) is associated with the communication towards the VM of a branch, and is then used to connect the BA to the first relay towards the corresponding VM.

[0089] In one embodiment, the antennas used are directional antennas.

[0090] The alternating allocation of sub-bands makes it possible, when using directional antennas, to avoid any risk of interference between two links using the same sub-band. Indeed, when omnidirectional antennas are used instead of directional antennas, the left antenna of a relay Rn presents a risk of interference with the right antenna of the relay Rn-1 (both using the same sub-band) when one of them transmits a message, thus preventing the other from correctly receiving other messages.

[0091] In the example of [Fig.6], if the equipment is equipped with omnidirectional antennas, there may be interference if the relay R2 transmits a message to the relay R3 at the same time as the relay RI transmits a message to the VM.

[0092] To overcome this problem, an appropriate procedure can be implemented to control the movement of the relays when moving the VM, where the distances between two consecutive devices (R or VM) are maximized (with the aim of limiting the number of relays as much as possible). As a result, the interfaces of two different devices (R or VM) using the same sub-bands are then all out of range of each other.

[0093] Such a procedure is also suitable for directional antennas with greater range.

[0094] In one embodiment, the method of the invention is implemented with equipment (BA, VM, R) having directional antennas. The directional antennas can be of two distinct types: directional antennas classics, and directional antennas implementing spatial filtering techniques or “beamforming” according to the established Anglicism.

[0095] "Conventional" directional antennas (or beam antennas) are antennas that radiate or receive greater power in specific directions, thus reducing interference from unwanted sources and increasing performance. High-gain antennas are also referred to as directional antennas with a focused, narrow radio beam, allowing for more precise targeting of radio signals. The most common are parabolic antennas, helical antennas, yagi antennas, and phased arrays of smaller antennas. These antennas can be fixedly oriented (particularly to interconnect two fixed points) or mobilely oriented by using a motor to dynamically orient the radiation beam in a certain direction (for example, for a satellite antenna positioned on a vehicle).

[0096] One embodiment of the invention uses conventional directional antennas with motorized mobile orientation.

[0097] Directional antennas with beamforming or pathforming (beamforming) implement a signal processing technique used in antenna and sensor arrays for the directional transmission or reception of signals. This is achieved by combining the elements of a phased array antenna in such a way that in particular directions the signals interfere constructively while in other directions the interference is destructive. Beamforming can be used on the transmitter side or the receiver side to achieve spatial selectivity. This type of antenna can locate its corresponding antenna by measuring the signal received on the different elements constituting the antenna (the "antenna array"). It can thus focus its beam in the direction of this corresponding antenna.

[0098] One embodiment of the invention uses beamforming type antennas capable of dynamically adapting their orientation towards a corresponding antenna.

[0099] The person skilled in the art will adopt the type of directional antenna to the context of the application, each type of directional antenna having advantages. Thus, if the use of a beamforming antenna offers numerous advantages in terms of integration (reduced hardware complexity) compared to a conventional directional antenna with motorized mobile orientation, the latter potentially makes it possible to achieve greater ranges in mobility situations.

[0100] A directional antenna allows waves to be focused in a well-defined direction, thus allowing (1) the risk of interference to be minimized (the emission zone then being better defined / targeted) compared to an omnidirectional antenna or a sector antenna, and (2) to increase the transmission range (the energy being focused in a defined / targeted direction), and therefore the quality of the connection.

[0101] In one embodiment, the method for determining and allocating frequency sub-bands for the case of a linear mesh wireless network can be supplemented by a method for dynamically managing the transmission power on the link between the BA and the last relay added (when moving away from the VM), so as to avoid any risk of interference between this new link and the penultimate link.

[0102] In the example of [Fig.6], when R3 is added to the trajectory, the implementation of the dynamic transmission power management method makes it possible to avoid any interference between the BA-R3 and R2-R1 links which, due to the alternating allocation, use the same frequency. Indeed, the addition of R3 is typically triggered as soon as the wireless signal propagation conditions between BA and R2 cross a threshold which soon no longer allows satisfactory communication quality to be ensured without the addition of a relay between them. In this particular case, when R3 is integrated into the linear mesh network, the distance between BA and R3 is potentially relatively small and therefore the BA antenna, if it transmits at full power, can generate a risk of interference with the R2 antenna which is turned towards the VM (and using the same sub-band as BA) since the two antennas are then potentially too close to each other.The steps of the dynamic transmission power management algorithm on the link between the BA and the last added relay are detailed with reference to [Fig.9].

[0103] A person skilled in the art can generalize the description of the determination and allocation of frequency sub-bands of the example of [Fig.6], to a linear mesh wireless network having a BA, a VM and a plurality of relays.

[0104] [Fig.7] illustrates an example of determining and allocating sub-bands of frequencies for the case of a star-topology mesh wireless network between a common rear base BA and four mission vehicles VM1, VM2, VM3, VM4.

[0105] In the chosen example, the trajectory of VM1 comprises three relays RI, R2, R3, the trajectory of VM2 comprises three relays R4, R5, R6, the trajectory of VM3 comprises two relays R7, R8, and the trajectory of relay VM4 comprises two relays R9, RIO.

[0106] The initial phase consists of partitioning the global B-band into 'a+1' distinct sub-bands of the same width, 'a' being the number of VMs to be connected to the same BA.

[0107] The sub-bands are designated by Bb for the basic sub-band, and B1 to Ba for the complementary sub-bands. In the case of [Fig.7], five sub-bands Bb, Bl, B2, B3 and B4 are defined, all of identical widths. The respective location of each of the sub-bands in the initial overall band B is not important, so the band B can be partitioned (from left to right of the band B) according to different ways of arranging the different sub-bands such as (Bb, Bl, B2, B3, B4) or (Bl, B2, Bb, B3, B4) or (B3, Bb, B2, B4, Bl) or any other arrangement.

[0108] It is then constituted 'a' pairs of sub-bands (Bb, Bl), (Bb, B2), ..., (Bb, Ba) such that each of the pairs includes the baseband Bb and one of the complementary sub-bands. Each pair is then associated with one of the branches of the star-topology mesh wireless network (centered on the common BA), each branch ensuring connectivity to a particular VM. It should be noted that an independent branch between the BA and a VM forms a linear mesh wireless network.

[0109] The sub-band allocation phase consists, for each branch of the wireless mesh network, in cyclically allocating to the different links of the linear wireless mesh network (corresponding to said branch) the two sub-bands assigned for said branch, such that two consecutive links use different sub-bands.

[0110] In one embodiment, the allocation is made such that the base sub-band Bb is not assigned to the BA and is not used by one of the antennas of the BA. In this way, each antenna of the BA uses one of the frequencies of the remaining sub-bands Bi (l<=i<=a).

[0111] Advantageously, the alternating allocation coupled with the use of directional antennas makes it possible to avoid any risk of interference between the different interfaces of the antennas of the B A. In addition, it makes it possible to avoid any interference between the equipment of the same branch. Finally, the alternating allocation of frequencies coupled with the use of directional antennas makes it possible to avoid any risk of interference between the different branches of a mesh wireless network with star topology.

[0112] In an advantageous embodiment, the method for determining and allocating frequency sub-bands for the case of a mesh wireless network with star topology can be supplemented by a frequency shift method, as soon as a relay is added or removed from a branch of the mesh network (typically due to the VM of said branch being moved away or closer to the BA).

[0113] This method makes it possible to avoid two antenna interfaces of the BA using simultaneously the base frequency Bb (and therefore to avoid them being in a situation of interference), as soon as relays on different branches are added or removed consecutively.

[0114] Taking [Fig.7] for example, considering the distance of the VM4 which leads to the need to add a new relay RI 1 between RIO (which is moving away) and the BA, this new relay RI 1 will then be configured in such a way that the new link RI 1-RIO uses the sub-band B4 and the new link BA-R11 uses the sub-band Bb of so as to avoid any interference at the VM4 branch, and interference with the other branches. However, if VM3 then moves away, implying the need to add a new R8bis relay between the BA and R8, this new R8bis relay would be configured so that the new R8bis-R8 link uses the B3 sub-band and the new BA-R8bis link uses the Bb sub-band. This would then generate an interference situation at the BA which would use the same Bb frequency on two of its antenna interfaces for two separate VMs (in this case on the BA-R11 and BA-R8bis links). The "frequency shift" process implemented whenever a relay is added or removed on a branch of the mesh network (and before any addition or removal of another relay on a different branch), makes it possible to overcome this problem.

[0115] The frequency shift consists, once a relay has been added or removed from a branch, in shifting the sub-bands used on the different links of this branch. Thus, if we take the previous example of the addition of the relay RI 1 on the branch of VM4, once RI 1 is added (and the link BA-R11 uses the sub-band Bb and the link RI 1-R10 uses the sub-band B4), the frequency shift operation makes it possible to reconfigure the branch of VM4 so that the link BA-R11 uses B4, the link RI 1-R10 uses Bb, the link R10-R9 uses B4, and the link R9-VM uses Bb. This makes it possible to shift all the sub-bands on the different links of the branch concerned. In this way, the BA is no longer in a position to directly use the Bb sub-band for one of its antennas, which then allows the addition or removal of another relay on any branch without risk of interference.

[0116] In one embodiment, the method for determining and allocating frequency sub-bands for the case of a mesh wireless network with a star topology, can be supplemented by a method for dynamically managing the transmission power on the link between the BA and the last relay added (when the VM is moved away), so as to avoid any risk of interference between this new link and the penultimate link. For example, when the relay RI 1 is added, the dynamic transmission power management operation makes it possible to avoid any interference between the links BA-R11 and R10-9 which use the same frequency.

[0117] [Fig.8] is a flowchart of the steps 800 of the initial frequency band allocation phase according to the method of the invention, in the context of a mesh wireless network having an initial frequency band B, interconnecting a rear base via at least one relay to at least one mission vehicle.

[0118] Advantageously, the method can be applied to linear mesh wireless networks and star topology mesh wireless networks, in order to minimize interference between the different links and optimize the performance of the network.

[0119] In a first step 802, the method makes it possible to determine the topology of the network considered, i.e. linear or star.

[0120] If the mesh wireless network is linear, having a rear base BA connected via at least one relay R to a single mission vehicle VM, the method continues with a phase 804 of initial allocation of frequencies for the linear network.

[0121] The initial configuration phase 804 consists in a first step 806 of partitioning the initial frequency band B into two sub-bands of the same width, a first basic sub-band Bb and a second complementary sub-band Bl. Then the method makes it possible in a following step 808, to allocate cyclically and alternately the two sub-bands to the current links of the network, so that two consecutive links use different sub-bands.

[0122] If it is determined that the mesh wireless network is of star topology, having a common rear base BA connected to several mission vehicles VM via Rx relays, the method continues with a phase 810 of initial allocation of frequencies for star network.

[0123] The initial configuration phase 810 consists in a first step 812 of determining the number 'a' of branches of the star network (or the number 'a' of different mission vehicles connected to the BA via multiple Rx relays).

[0124] In a following step 814, the method makes it possible to partition the global frequency band B into 'a+1' distinct sub-bands of the same width, comprising a first basic sub-band Bb, and 'a' complementary sub-bands (Bl, ..., Ba).

[0125] In a following step 816, the method makes it possible to constitute 'a' pairs of sub-bands with the base band: (Bb, Bl), (Bb, B2), (Bb, Ba).

[0126] The method continues with a step 818 consisting of assigning each of the pairs of sub-bands (Bb, Bl), (Bb, B2), (Bb, Ba) to a different branch of the network.

[0127] In a following step 820, the method makes it possible, for each branch of the network, to allocate in a cyclic and alternating manner, the two sub-bands of the pair corresponding to the branch to the different links of the branch so that two consecutive links use different sub-bands and that the base sub-band Bb is not used by the antenna of the rear base BA serving the first link of this branch.

[0128] Once the initial configuration phases are completed, the method makes it possible to determine in a following step 822 when a variation of the initial topology occurs.

[0129] A variation in topology, whether on the initial topology or on a subsequent current topology, may occur in different cases.

[0130] For a linear topology, a variation in topology can be linked to the movement of a mission vehicle which can then lead to the addition or removal of a relay on the trajectory of the VM, or it can be related to the addition of a new VM causing a change in nature from linear topology to star topology with the addition of a new branch with the new VM added.

[0131] For a star topology, a topology variation may be related to an addition or removal of a relay on the path of one of the VMs in the network, or it may be related to an addition or removal of a VM from the network.

[0132] When a topology variation is detected, the method continues with a phase 824 of re-allocation of the frequency sub-bands, which depends on the topology of the network and the nature of the variation in the topology.

[0133] Figures 9 to 12 illustrate the different cases of reallocation of frequency sub-bands.

[0134] [Fig.9] is a flowchart of the steps for re-allocating frequency sub-bands for a linear mesh wireless network, in the case of adding a relay R. The method 900 operates according to the following steps, which consist of:

[0135] Step 902: Determine when a relay Rn is added between a BA and a relay Rn-1, i.e. the relay Rn-1 which is in communication with the BA by a first link.

[0136] Step 904: Allocate to the link between the new Rn and the previous relay Rn-1, the sub-band which was used by the link between the BA and the Rn-1, for example the sub-band Bl.

[0137] Step 906: Allocate to the link between the BA and the new Rn the sub-band complementary to that which exists between the link between the Rn and the Rn-1, for example the basic sub-band Bb.

[0138] The method continues with a phase 908 of establishing the wireless link between the BA and the new relay Rn, and which comprises steps consisting of:

[0139] Step 910: Determine the orientation to be given to the antennas of the BA and the Rn to bring them towards each other. In a preferred embodiment, the antennas are directional antennas.

[0140] In the case of using beamforming antennas, the orientation is automatic.

[0141] In the case of the use of conventional directional antennas with mobile orientation motorized, the orientation is carried out by considering that the (fixed) position of the BA and the initial position of the Rn (when it is added to the network) are predefined and known a priori by each respectively. Also, each can determine the respective location of its correspondent and automatically orient its antenna in its direction.

[0142] Step 912: Once the antennas are oriented, the method makes it possible to configure the power of the antenna from Rn to the BA at the maximum power (Pmax), and thus to establish the link from the new relay Rn to the BA.

[0143] Step 914: The method makes it possible to configure the power of the antenna from BA to Rn at a predefined initial power (Pinit), corresponding to a power level sufficient to reach a relay Rn in its initial position without risk of interference with a relay Rn-1. The initial power Pinit can be predefined because the initial location of an Rn when it joins the network is also predefined. Step 914 establishes the link from the BA to the new relay Rn.

[0144] Advantageously, the reallocation method makes it possible to dynamically adjust the orientations of the antennas of a BA and of a new relay Rn when the Rn is moved, as well as the power of the BA antenna towards an Rn, according to the following steps 916 and 918.

[0145] Step 916: Dynamically adjust the power and orientation of the BA antenna towards an Rn when moving the Rn.

[0146] In the case of using beamforming antennas, the orientation is automatic.

[0147] In the case of the use of conventional directional antennas with mobile orientation motorized: When the Rn moves (typically when it moves away from the BA because the Rn-1 relay and the VM potentially move away), or periodically, the Rn communicates its location to the BA which uses this information to reorient its antenna towards the Rn.

[0148] For both types of antenna, the BA must adjust its transmission power so that the quality of the link from the BA to the Rn remains good, i.e. the power of the signal from the BA received by the Rn remains above a given threshold 'S+delta', S being a minimum threshold and delta a safety margin.

[0149] To do this, the BA adjusts its transmission power towards the Rn (Ptransmission-BA) in the following manner:

[0150] Pmission-BA (BARn) = (S+delta) x Pmax / Preceived-BA (RnBA), where

[0151] Preceded-BA (RnBA) is the power of the Rn signal received by the BA.

[0152] Step 918: Dynamically adjust the orientation of the Rn antenna towards the BA during of the displacement of the Rn.

[0153] In the case of using beamforming antennas, the orientation is automatic.

[0154] In the case of the use of conventional directional antennas with mobile orientation motorized: The location of the BA being fixed and predefined (and known to the Rn), the Rn can use this information to determine the new orientation to give to its antenna when it moves.

[0155] Once the wireless link has been established between the BA and the new Rn, the method makes it possible, in a following phase 920, to establish the wireless link between the relay Rn and the relay Rn-1, according to steps consisting of:

[0156] Step 922: Determine the orientation of the directional antennas of Rn and Rn-1 towards each other.

[0157] In the case of using beamforming antennas, the orientation is automatic.

[0158] In the case of the use of conventional directional antennas with motorized mobile orientation: The position of the Rn-1 is communicated by the BA to the Rn (the BA was until then connected to the relay Rn-1 and therefore received the location of the Rn-1, which for example was transmitted to it by the Rn-1 regularly. The Rn-1 knows the initial position of the new relay Rn (when it is added to the network) because this information is known and predefined. Also, thanks to this information, each of the relays Rn and Rn-1 reorients its antenna towards each other.

[0159] Step 924: Once the antennas are oriented, the method makes it possible to configure the power of the antenna from Rn to Rn-1 at the maximum power (Pmax). Step 924 makes it possible to establish the link from Rn to Rn-1.

[0160] Step 926: The method makes it possible to configure the power of the antenna from Rn-1 to Rn at the maximum power (Pmax), and thus to establish the link from Rn-1 to Rn.

[0161] Advantageously, the reallocation method makes it possible to dynamically adjust the orientations of the antennas of a new relay Rn and of a relay Rn-1 during the movement of the Rn or of the Rn-1, according to the following steps 928 and 930.

[0162] Step 928: Dynamically adjust the orientation of the antenna of the new Rn towards the Rn-1 when the Rn or the Rn-1 moves. When the Rn-1 moves, or periodically, the Rn-1 transmits its location to the Rn, which allows the Rn to readjust the orientation of its antenna towards the Rn-1.

[0163] When the Rn moves, as it knows the location of the Rn-1, it can readjust the orientation of its antenna in the direction of the Rn-1.

[0164] Step 930: Dynamically adjust the orientation of the antenna of the Rn-1 towards the Rn when the Rn or the Rn-1 moves. Similarly to step 928, when the Rn moves, or periodically, the Rn transmits its location to the Rn-1, which allows the Rn-1 to readjust the orientation of its antenna towards the Rn.

[0165] When the Rn-1 moves, as it knows the location of the Rn, it can readjust the orientation of its antenna in the direction of the Rn.

[0166] After a phase of re-allocation of the frequency sub-bands after the addition of a relay Rn, the method loops back to the step of determining a variation of the topology.

[0167] [Fig. 10] is a flowchart of the steps of re-allocating frequency sub-bands for a linear mesh wireless network, in the case of a removal of a relay R. The method 1000 operates according to the following steps, which consist of:

[0168] Step 1002: Determine when a relay Rn is removed from the path between a BA and a relay Rn-1. The case of a relay removal typically occurs when the VM returns towards the BA and therefore fewer relays are needed to ensure connectivity.

[0169] The method continues with a phase 1004 of establishing the wireless link between the BA and the relay Rn-1, which comprises steps consisting of:

[0170] Step 1006: the BA reconfigures its antenna to use the same sub-band as that which was previously used between the removed Rn and Rn-1 (the former neighbor of Rn).

[0171] Step 1008: Rn-1 continues to use this same sub-band to communicate with the BA instead of Rn.

[0172] The method then makes it possible to reorient the antennas of the BA and the Rn-1.

[0173] Step 1010: orientation of the BA antenna towards the Rn-1 which makes it possible to establish the link from the BA to the Rn-1. In the case of the use of beamforming antennas, the reorientation is automatic.

[0174] In the case of the use of conventional directional antennas with motorized mobile orientation, the reorientation is carried out in the following manner: the BA obtains the location of the Rn-1 (this information can be provided by the Rn to the BA for example) and the BA uses this information to orient its antenna in the direction of the Rn-1.

[0175] Step 1012: orientation of the antenna of the Rn-1 towards the BA which makes it possible to establish the link from the Rn-1 to the BA. The Rn-1 knowing the location of the BA which is fixed and predefined, it uses this information to reorient its antenna towards the BA.

[0176] After a phase of re-allocation of the frequency sub-bands after the withdrawal of a relay Rn, the method loops back to the step of determining a variation of the topology.

[0177] Advantageously, the method of the invention makes it possible to support mesh wireless networks with variable star topology in the cases of adding or removing a VM (and therefore adding or removing a new branch to the topology), or in the cases of adding or removing a relay on one of the branches when moving the VM associated with said branch. This method then makes it possible to minimize interference between the different links and optimize the performance of the network.

[0178] When a topology variation is determined, the method makes it possible to determine (step 1102) whether the topology variation relates to an addition or a removal of a mission vehicle VM or a relay.

[0179] [Fig. 11] is a flowchart of the steps for re-allocating frequency sub-bands, in the case of a topology variation relative to a VM, on an existing star topology by adding or removing a VM, or on an initial linear topology by adding a VM.

[0180] [Fig. 12] is a flowchart of the steps for re-allocating frequency sub-bands for a star-topology mesh wireless network, in the case of a topology variation relative to a relay.

[0181] Returning to [Fig. 11], if the topology variation is linked to a VM, the method makes it possible to determine whether it is the addition or removal of a VM (steps 1104 and 1114 beyond [Fig. 11]).

[0182] If it is the addition of a new mission vehicle VMa+1, the method continues with a phase of re-allocation of the sub-bands which comprises the following steps 1106 to 1112.

[0183] Step 1106: Partition the global band B into 'a+2' distinct sub-bands of the same width: Bb', Bl', Ba', Ba+1'.

[0184] Step: 1108: Constitute 'a+1' pairs of sub-bands with the basic sub-band Bb': (Bb', Bl'), (Bb', B2'), (Bb', Ba'), (Bb', Ba+1').

[0185] Step 1110: Associate each of the pairs of sub-bands (Bb', Bl'), (Bb', B2'), (Bb', Ba'), (Bb', Ba+1') with a different branch of the network.

[0186] Step 1112: For each branch of the network, allocate, in a cyclical and alternating manner, the two sub-bands associated with the branch, to the different links of the branch so that (1) two consecutive links use different sub-bands and (2) the sub-band Bb' is not used by the BA antenna serving this branch.

[0187] After the phase of re-allocation of the frequency sub-bands due to the addition of a new VM, the method can return to the step of determining a variation of the topology.

[0188] If the topology variation is linked to a withdrawal of a mission vehicle VMk, the method continues with a phase of re-allocation of the sub-bands on the branches of the remaining 'a-1' VMs, which comprises the following steps 1116 to 1122. This case only applies for star topologies.

[0189] Step 1116: Partition the global band B into 'a' distinct sub-bands of the same width: Bb”, Bl”, Ba-1”.

[0190] Step 1118: Constitute 'a-1' pairs of sub-bands with the base band Bb”: (Bb”, Bl”), (Bb”, B2”), (Bb”, Ba-1”).

[0191] Step 1120: Associate each of the pairs of sub-bands (Bb”, Bl”), (Bb”, B2”), (Bb”, Ba-1”) with a different branch of the network.

[0192] Step 1122: For each branch of the network, allocate, in a cyclical and alternating manner, the two sub-bands to the different links of the branch so that (1) two consecutive links use different sub-bands and (2) the sub-band Bb” is not used by the BA antenna serving this branch.

[0193] After a phase of re-allocation of the frequency sub-bands due to the removal of a VM, the method can return to the step of determining a variation of the topology.

[0194] If the topology variation is linked to a relay, the method allows (1200) a reallocation of the sub-bands.

[0195] The method then makes it possible to determine whether it is the addition or removal of a relay (steps 1202 and 1208 of [Fig. 12]).

[0196] If it is the addition of a new relay Rn to the branch of a mission vehicle VMi, the method continues with a phase of re-allocation of the sub-bands which comprises the following steps 1204 and 1206.

[0197] Step 1204: Integrate the Rn relay into the VMi branch according to the steps of the method described for [Fig.9].

[0198] Step 1206: Shift the sub-bands used on the different links of the VMi branch.

[0199] After a phase of re-allocation of the frequency sub-bands after the addition of a new relay, the method can return to the step of determining a variation in the topology.

[0200] If it is the withdrawal of a relay Rk from the branch of a mission vehicle VMi, the method continues with a phase of re-allocation of the sub-bands which comprises the following steps 1210 and 1212.

[0201] Step 1210: Remove the Rk relay from the VMi branch according to the steps of the method described for [Fig. 10].

[0202] Step 1212: Shift the sub-bands used on the different links of the VMi branch.

[0203] After a phase of re-allocation of the frequency sub-bands after the removal of a relay, the method can return to the step of determining a variation in the topology.

[0204] The various methods of the invention which have been described can be implemented at the level of one or more processors of a central system in the form of one or more computer programs which comprise non-transitory code instructions making it possible to carry out the steps of the methods of the invention, when this or these programs are executed.

[0205] Those skilled in the art understand that the invention is applicable without limitation to other implementation modes. In an advantageous embodiment, the invention is implemented via a network management architecture oriented towards “Software Defined Networking” (SDN). This architecture is based on an SDN controller implemented at the central system level (at the level of a BA or a server connected to the rear network) and capable of controlling, via its “South Interface”, the configurations of a set of SDN equipment: the BA, the relays and the VMs.

[0206] In this SDN implementation, the methods described above can be implemented in the form of an SDN Service (at the central system level) interfacing with the SDN Controller via its “North Interface”, so as to control the configurations of the various SDN devices reachable via the South interface of the SDN controller.

[0207] This SDN service is thus responsible for (1) managing the initial configuration (or allocation) of the frequency sub-bands to be used in a mesh wireless network (linear or star topology); (2) managing the reconfigurations (or reallocation) of the frequency sub-bands within the mesh wireless network (linear or star topology) when adding or removing a VM or a relay; (3) managing the orientation of the directional antennas of the different equipment (BA, relay, VM) within the mesh wireless network (linear or star topology) when these equipments are moved; and (4) to manage the transmission powers of the directional antennas of the different equipment (BA, relays, VMs) within the mesh wireless network (linear or star topology), and in particular those of the directional antennas of the BA when adding a new relay.

[0208] In the case of an implementation on a central system, the latter can be located at the level of the BA or at the level of a server connected to another network (called the “rear” network) to which the BA is itself also connected (this “rear” network therefore being a network different from the mesh wireless network associated with the BA).

[0209] It may also be noted that the implementation of the effective use of a frequency sub-band at the level of a link interconnecting two pieces of equipment (e.g. of the BA, relay, or VM type) requires that each of the two pieces of equipment concerned have its respective antenna system configured so that it uses said frequency sub-band.

[0210] As indicated above, in the case of the implementation of the methods of the invention via an SDN Service, the latter will thus be in charge of determining the configurations / allocations of the frequency sub-bands to be used at each moment for the different links of the network, also taking charge of these reconfigurations / reallocations as soon as the topology of the network evolves due to the addition or removal of a relay or a VM. In addition, this SDN Service will then rely on the South interface of the SDN controller so as to set up (“push”) the corresponding configurations of the antenna systems for the equipment concerned: i.e. configure the antenna systems of this equipment with the appropriate frequency sub-band (as determined by the SDN Service according to the methods of the invention).

[0211] Advantageously, depending on the targeted usage scenario, the frequency sub-bands allocated within the framework of the present invention may correspond to any type of waves / frequencies, such as for example and without limitation: - Acoustic waves (for example for underwater mesh wireless networks); - Radio waves (for example for terrestrial or aerial wireless networks); - Optical waves (e.g. for terrestrial or aerial wireless networks); - Etc.

[0212] The invention is also applicable regardless of the technologies, the communication protocols implemented within the mesh network, including for cases where several technologies, protocols are used within the same network (i.e. heterogeneous network): Technologies of the WiFi, Bluetooth, 3G / 4G / 5G / 6G type.

[0213] Furthermore, the sub-bands can also be used without limitation by all the techniques for sharing wireless resources and controlling access to the communication medium implemented at the level of a link such as for example and without limitation: - Two-way communications using CSMA / CA or CSMA / CD approaches for medium access control; - Bidirectional communications in FDD (Frequency Division Duplex) mode: in this case, the sub-band allocated to a given network link can itself be partitioned into two sub-bands, one for communications in one direction and the other for communications in the other direction; - Two-way communications in TDD (Time Division Duplex) mode: in this case, communications at a network link use the entire width of the same sub-band regardless of the direction of communications; however, each direction can only use the sub-band over a specific time slot reserved for that direction; - Etc.

[0214] A technical solution has been described which makes it possible to optimize the performance of wireless mesh networks, linear or with a star topology, and in particular when the topology of these networks is likely to evolve through the addition or removal of relays and / or mission vehicles VM. The performance gain is obtained by limiting interference between the different links of the network and by maximizing the bandwidth allocable to each of the links (under the assumption of a fixed overall bandwidth). This performance gain is then characterized firstly by an available end-to-end bandwidth between each of the nodes of the network, which is much higher than known approaches. The performance gain is also characterized by a lower packet / data loss rate, and by a reduced latency for data transmission.

[0215] The bandwidth gains of the present invention can be estimated as follows:

[0216] For a linear network: Increase in available bandwidth of at least 50% compared to state-of-the-art solutions. Indeed, the solution presented makes it possible to allocate to each link of the network a sub-band width equal to half of the global bandwidth (=B / 2) which is characterized by a bandwidth within the network of the order of 50% of the nominal bandwidth of a transmission link, where for example the solution presented in document [3] only allows at best to offer 33% of this bandwidth.

[0217] For a star network: Increase in available bandwidth of at least 300% compared to state-of-the-art solutions. Indeed, the solution presented makes it possible to allocate to each link of the network a sub-band width equal to l / (a+l) ('a' being the number of branches of the network), which is characterized by a bandwidth within the network of the order of l / (a+1) of the bandwidth of a transmission link. For comparison, the solution presented in document [3] can only offer at best a bandwidth within the network of the order of l / (3xa), because of significant interference at the BA level. Thus, whatever the size of the network (the number of branches and relays), the present invention provides a factor x3 on the improvement of the flow rates.

[0218] Furthermore, the embodiment with integration of directional antennas makes it possible to have increased transmission ranges (at constant power / energy consumption) for each of the network links, and therefore to reduce the number of relays required to allow a VM to reach a certain distance (hence a cost saving in deploying the solution), or even to reach greater distances with the same number of relays. The present invention thus provides a very significant performance gain for application scenarios such as those presented.

Claims

Claims

1. A method for allocating frequencies in a mesh wireless telecommunications network, the network operating on an initial global frequency band to serve all communications and different links within the network, the network comprising at least one rear base BA equipped with at least one antenna, one or more mission vehicles VM each equipped with an antenna for connecting to a relay in the direction of the BA, a plurality of relays R each equipped with two antennas, a first for connecting to a previous relay or to a VM and a second for connecting to a next relay or to a rear base, the method comprising steps consisting in: - determining (802) whether the topology of the network is a linear network topology having a single branch for interconnecting a single mission vehicle to a rear base via at least one relay,or a star network topology having a plurality 'a' of branches for interconnecting a plurality 'a' of different mission vehicles to a common rear base, at least one branch having at least one relay; - partitioning (806, 814) the initial global frequency band into a number of distinct sub-bands of the same width, the number of sub-bands depending on the network topology; and - cyclically and alternately allocating (808, 820) the sub-bands to the links of the branch or to the links of the plurality of branches, such that two consecutive links use different sub-bands.,

2. The method of claim 1 wherein the step of partitioning the initial global frequency band comprises for a linear network partitioning (806) said band into a basic sub-band and a complementary sub-band, or comprises for a star network partitioning (812, 814) said band into a basic sub-band and 'a' complementary sub-bands.

3. The method according to claim 1 or 2 further comprising after the partitioning step if the network is star-shaped, steps consisting of: - constituting (816) 'a' pairs of sub-bands with the baseband and the complementary sub-bands, such that each of the pairs includes the baseband and one of the complementary sub-bands; - assigning (818) each of the pairs of sub-bands to a different branch of the network; and - for each branch of the network, allocating (820) cyclically and alternately, the two sub-bands of the pair corresponding to said branch to the different links of the branch, so that two consecutive links use different sub-bands and the base sub-band is not used by the antenna of the rear base serving the first link of said branch.

4. The method according to any one of claims 1 to 3 further comprising after the sub-band allocation step, steps consisting of: - determining (822) a variation in the network topology; and - reallocating (824) the frequency sub-bands according to the network topology and the nature of the variation in the topology.

5. The method of claim 4 wherein the step of determining a topology variation for a linear network comprises determining (902, 1002) whether a relay is added or removed from the branch, or determining whether a mission vehicle is added to create a new branch.

6. The method of claim 4 wherein the step of determining a topology variation for a star network comprises determining whether a relay is added or removed from a branch, or whether a mission vehicle is added or removed from the network.

7. The method according to claim 5 or claim 6 further comprising, if a relay Rn is added between the rear base and a relay Rn-1, steps consisting in: - allocating (904) to the link between the new relay Rn and the previous relay Rn-1, the sub-band which was used by the link between the rear base and the previous relay Rn-1; - allocating (906) to the link between the rear base and the new relay Rn the sub-band complementary to that of the link between the new relay Rn and the previous relay Rn-1; - establishing (908) the wireless link between the rear base and the new relay Rn; and - establishing (920) the wireless link between the new relay Rn and the previous relay Rn-1.

8. The method of claim 7 wherein the link establishment steps comprise steps of configuring (912, 914, 924, 926) and adjusting (916) the power of the antennas of the rear base, the new relay Rn and the previous relay Rn-1.

9. The method of claim 7 or claim 8 wherein the link establishment steps further comprise steps (918, 928, 930) of orienting the antennas of the rear base, the new relay Rn and the previous relay Rn-1.

10. The method of claim 5 or claim 6 further comprising, if a relay Rn is removed between the rear base and a previous relay Rn-1, steps (1000) of establishing a wireless link between the rear base and the previous relay Rn-1, said steps of configuring (1006) and orienting (1010, 1012) the antennas of the rear base and the previous relay Rn-1.

11. The method according to claim 5 or claim 6 further comprising if the topology variation is due to an addition or removal of a mission vehicle, steps (1100) of partitioning (1106, 1116) the overall frequency band into distinct sub-bands of the same width, and for each branch of the network, allocating (1108, 1110, 1112, 1118, 1120, 1122) to the different links of the branch in a cyclic and alternating manner, the two sub-bands associated with said branch, so that two consecutive links use different sub-bands.

12. The method of claim 5 further comprising if the topology variation is due to an addition or removal of a relay from a branch, steps (1206, 1212) of shifting said sub-bands used on the different links of said branch so that the base sub-band is not used by the antenna of the rear base serving this branch.

13. A computer program product, said computer program comprising code instructions for carrying out the steps of the method according to any one of claims 1 to 11, when said program is executed on a computer.

14. A frequency allocation device in a mesh wireless telecommunications network, the network operating on an initial global frequency band to serve all communications and different links within the network, the network comprising at least one rear base BA equipped with at least one antenna, one or more mission vehicles VM each equipped with an antenna for connecting to a relay in the direction of the BA, a plurality of relays R each equipped with two antennas, a first for connecting to a previous relay or to a VM and a second for connecting to a next relay or to a rear base, the device comprising means for implementing the steps of the method according to any one of claims 1 to 11.

15. The device according to claim 14 wherein the communication network is implemented according to an SDN 'Software Defined Networking' oriented architecture comprising an SDN controller having a North Interface and a South Interface, set up at the level of a central system.

Images