FR2948462A1 - Wireless communication device i.e. home theatre, position determining method for wireless personal area network, involves selecting reference group, and determining position of device of network based on selected group - Google Patents

Abstract

The method involves constituting a set of groups of a preset number of wireless communication devices among an assembly of wireless communication devices, and determining a differential angle which is the difference between estimations of absolute angles obtained for two devices of the group. A reference group is selected among the groups according to a criterion which is the difference between a sum of differential angle determined for each of the devices and a reference value. A relative position of the device of the network is determined based on the selected reference group. Independent claims are also included for the following: (1) a computer program product comprising program code instructions for implementation of a position determining method (2) a storage medium comprising instructions for implementation of a position determining method (3) a device for determining positions of devices in a wireless communication network.

Description

A method of discovering device positions of a wireless communication network, computer program product, storage means and corresponding discovery device. FIELD OF THE INVENTION The field of the invention is that of communication networks, and more particularly wireless communication networks with programmable antennas, such as, for example, wireless home communication networks using the radio band. at 60 GHz. More specifically, the invention relates to a technique for determining positions of devices included in a wireless communication network. 2. BACKGROUND TECHNOLOGY The home networks or PAN networks (for "Personal Area Networks" in English) can be wired (as is the case for networks of the USB, Ethernet or IEEE 1394 type) but can also rely on the use of a wireless medium. This is called home wireless network (or WPAN network for "Wireless Personal Area Networks" in English). The IEEE 802.15.1, UWB, ZigBee (IEEE 802.15.4), IEEE 802.1 or IEEE 802.15.3 Bluetooth standards are among the most widely used protocols for this type of network. Radio transmission systems currently use a wide range of transmission frequencies, typically between 2.5 GHz and 60 GHz. These frequencies are particularly well suited for very high speed data transmission in a limited radius, for example as a means of connectivity between the different elements of a home cinema type communication network. The use of transmit and receive antennas can also play a crucial role in the quality of communication for such home wireless networks. An isotropic antenna is a radiating antenna with the same physical characteristics in all directions of space. A selective antenna is an antenna whose radiated energy is voluntarily distributed unequally in space. Certain directions are favored in order to obtain a better gain. We then speak of radiation lobes. A radiation pattern of an antenna makes it possible to visualize these lobes of radiation in the three directions of space, that is to say in the horizontal plane or in the vertical plane including the most important lobe. A signal processing based on beamforming (beamforming) is a technique based on the use of transmission boards or reception sensors that control the orientation and / or sensitivity of an antenna as a function of beamforming. a radiation pattern. When receiving a radio signal, this technique increases the sensitivity of the receiver device in a desired direction and thus reduce the sensitivity of the antenna for interference or highly noisy areas. When transmitting a radio signal, this technique increases the power of the radio signal in one or more desired direction (s). Such communication systems, however, have a high sensitivity to the phenomena of interference and masking of radio communication links. In the context of a static configuration of a home network, these phenomena can be caused by the presence of objects such as furniture, plants for example, or by the presence of living beings within the coverage area. of the home network. In the state of the art, several known techniques are known for determining the relative position of communication devices within a wireless network. The most common known techniques are generally based on: the transmission of an identification signal and positioning of equipment according to the GPS positioning principle (for Global Positioning System in English or global positioning system in French); or - the estimation of radio link budgets for determining the distances between several pieces of equipment; or else - the estimation of the propagation time of a radio signal between two distant equipments making it possible to determine the distance between these two equipments. Another known technique, based on a triangulation localization mechanism, can also be used in certain wireless networks, such as GSM networks (for Global System for Mobile Communications, in English or the Global System for Mobile Communication in French), This second known technique relies on the detection of a mobile communication device (such as a radiotelephone for example) in the vicinity of communication devices of the relay antenna type (such as, for example, mobile base station or BTS (for Base Transceiver Station) With the knowledge of the coverage area of the network to which each relay antenna belongs, a triangulation mechanism is implemented from a set of three antennas relays detected as being closest to the mobile device, to define a geographic area in which is located the mobile device to locate.

Nevertheless, all these aforementioned known techniques are not usable in the desired context (ie in a WPAN type communication network). Indeed, in the context of a WPAN communication system, radio waves very frequently undergo reflections and / or removals during their propagation in a given medium. These phenomena of reflection and absorption are more particularly due to the various elements (furniture, plants, metal objects, humidity (high oxygen concentration), ...) that constitutes the medium in which the communication network is implemented, which very often causes a change in the balance of radio links and therefore does not allow the use of such a technique to determine the position of network devices. Indeed, this known technique requires the use of a communication network in a medium said free field, that is to say that is not (or little) sensitive to interference and masking phenomena. This technique is therefore not suitable for WPAN communication systems because it is not controlled in such a medium. As for the known technique of location by triangulation, it requires prior knowledge of the coverage area associated with each of the relay antennas. It also requires the implementation of a step of detecting the relay antennas closest to the device to be located. On the other hand, if this known triangulation localization technique is transposed into the desired context, since the position of the relay antennas is not available, it is not possible to know the associated coverage areas.

The geographical area in which the mobile device to be located is located can not therefore be determined. In addition, the step of detecting the relay antennas located closest to the device to be located is not guaranteed in the desired context because it is also based on a radio link budget whose disadvantages have been developed above. OBJECTIVES OF THE INVENTION The invention, in at least one embodiment, has the particular objective of overcoming these various disadvantages of the state of the art. More specifically, in at least one embodiment of the invention, an objective is to provide a technique for discovering the relative position of devices included in a wireless communication network. At least one embodiment of the invention also aims to provide such a technique for determining the position, absolutely, of devices included in a wireless communication network.

Another objective of at least one embodiment of the invention is to provide such a technique which makes it possible to determine a distance between at least two devices of the network, even in the case where they can not communicate directly. Yet another object of at least one embodiment of the invention is to provide such a technique that minimizes errors relating to position estimates. A complementary objective of at least one embodiment of the invention is to provide such a technique which relies solely on means conventionally used for the transmission of data in a wireless communication network, in other words a technique that is simple to implement and inexpensive. 4. DISCLOSURE OF THE INVENTION In a particular embodiment of the invention, there is provided a method of discovering positions of devices included in a wireless communication network. Such a method comprises the steps of: obtaining, for each device of a set of devices included in the network, an estimation of at least two absolute angles of line of sight, each between an estimated axis of line of sight between said device and another device of the network and secondly a predefined reference axis; constituting a plurality of groups of a predefined number of devices among the devices of said set; determining, for each device of each of the groups constituted, at least one differential angle which is the difference between two absolute angle estimates obtained for two devices of said group constituted and with respect to the same predefined reference axis; selecting, from the groups constituted, a reference group according to a criterion which is, for each of the groups constituted, the difference between on the one hand a sum of differential angles, each determined for one of the devices of said group constituted , and on the other hand a reference value; determining the relative position of at least one device of the network according to the selected reference group.

The general principle of this particular embodiment of the invention therefore consists in determining, according to the characteristics of programmable antennas used for the implementation of communication links between devices of a wireless communication network, the relative position of one or more devices relative to other network devices.

The location of a device of the network is more particularly carried out using a group of reference devices whose geometric shape they constitute (each device representing a vertex of this geometric shape) has the strongest coherence with a predefined geometric shape. In other words, this particular embodiment provides for selecting, from among all the groups of devices constituted, the one for which the determined differential angles have the greatest relevance in determining the relative positions of the devices of the network. Thus, this particular embodiment of the invention is based on a completely new and inventive approach based on the use of programmable antenna characteristics of the devices and which requires, beforehand, no measurement of distance between the devices of the network. .

It should be noted that the notion of absolute line of sight angle should not be considered solely as a receiving antenna orientation angle, but can also be considered as an antenna orientation angle. in emission. In other words, no distinction is made between an antenna angle of reception or transmission. Advantageously, said step of selecting a reference group is performed according to at least one additional criterion belonging to the group comprising: for each of the groups constituted, an overall level of communication quality associated with the differential angles each determined for one of the devices of said group constituted; for each of the groups constituted, an overall level of uncertainty associated with the differential angles each determined for one of the devices of said group constituted.

By one or more of these additional criteria, it is possible to refine the selection of the reference group among all groups of devices that have been previously formed. It should be noted that this list of criteria is not exhaustive. According to an advantageous characteristic, said step consisting in determining the relative position of at least one device of the network as a function of the selected reference group comprises the steps of, for a device to be located which does not belong to the reference group: each of the devices of the reference group, obtaining an absolute angle of view angle estimate between said device to be located and said device of the reference group; determining the relative position of said device to be located according to the estimates of absolute angles of line of sight obtained for the devices of the reference group. Determining the relative position of a device to be located that is not included in the reference group can therefore be achieved by using each of the devices constituting the reference group. This makes it possible, iteratively, to successively determine the relative position of several devices to be located that do not belong to the reference group. According to an alternative embodiment, said step consisting in determining the relative position of at least one device of the network as a function of the selected reference group comprises the steps of, for a device to be located that does not belong to the reference group: for each of the devices of the reference group, obtaining an absolute angle of view angle estimate between said device to be located and said device of the reference group; - determining at least two differential angles which are each the difference between two estimates of absolute line-of-sight angles obtained for two devices of said reference group and with respect to a same predefined reference axis; determining the relative position of said device to be located according to the differential angles determined for the device to be located.

In this variant embodiment, the device to be located is used to determine its own relative position. This makes it possible, iteratively, to successively determine the relative position of several devices to be located that do not belong to the reference group. Preferably, it comprises a step of introducing, in said reference group, at least one device whose relative position has been determined. Incrementally, it is therefore possible to integrate in the reference group a device, whose relative position has been determined, to form a new reference group. This new reference group having a number of data (absolute angles of line of sight, differential angles, overall level of communication quality, overall level of uncertainty, ...) more important, makes it possible to estimate more precisely the relative position of other devices of the network which one seeks to locate. In other words, errors in position determinations are minimized in this way.

Preferably, said number of devices included in each of the constituted groups is three, and in that said reference value is 180 degrees. This geometric shape makes it possible to obtain position measurements quickly and efficiently. Advantageously, an operation of obtaining an absolute line of sight angle estimate of a first device with respect to a second device is performed by analyzing a radiation pattern of a receiving or transmitting antenna of said first node. Because absolute line of sight angle estimates are obtained by simple analysis of the antenna radiation pattern, only conventionally used means for data transmission are therefore implemented. In other words, this embodiment of the invention is therefore simple to implement and inexpensive. Advantageously, it comprises the steps of: obtaining a distance value between two devices whose relative positions have been determined; determining an absolute position of at least one device whose relative position has been determined as a function of said obtained distance value. Thus, from the knowledge of one or more relative positions of network devices and an absolute distance between two devices whose relative positions are also known, it is possible to deduce the absolute position of this or these network devices. . By this method, it is therefore possible to know the absolute position between two devices of the network even in the case where they are not able to communicate (because of masking or interference for example). In another embodiment, the invention relates to a computer program product downloadable from a communication network and / or recorded on a computer readable medium and / or executable by a processor. This computer program product includes program code instructions for carrying out the aforesaid method (in any one of its various embodiments), when said program is run on a computer.

In another embodiment, the invention relates to a computer readable storage means, possibly totally or partially removable, storing a computer program comprising a set of instructions executable by a computer to implement the aforementioned method (in any of its different embodiments). In another embodiment, the invention relates to a device position discovery device included in a wireless communication network. Such a discovery device comprises: first means of obtaining, for each device of a set of devices included in the network, an estimate of at least two absolute angles of line of sight, each included between an estimated line of sight between said device and another device of the network and a predefined reference axis; means for constituting a plurality of groups of a predefined number of devices among the devices of said set; first means for determining, for each device of each of the groups constituted, at least one differential angle which is the difference between two absolute angle estimates obtained for two devices of said group constituted and with respect to the same reference axis predefined; selection means, among the groups constituted, of a reference group as a function of a criterion which is, for each of the groups constituted, the difference between on the one hand a sum of differential angles, each determined for one devices of said group constituted, and secondly a reference value; second means for determining the relative position of at least one device of the network as a function of the selected reference group. According to an advantageous characteristic, said second means for determining the relative position of at least one device of the network as a function of the selected reference group themselves comprise, for a device to be located which does not belong to the reference group: second means for obtaining, for each of the devices of the reference group, an absolute line of sight angle estimate between said device to be located and said device of the reference group; Third means for determining the relative position of said device to be located as a function of the estimates of absolute line of sight angles obtained for the devices of the reference group. According to an alternative embodiment, said second means for determining the relative position of at least one device of the network as a function of the selected reference group themselves comprise, for a device to be located that does not belong to the reference group: second means for obtaining, for each of the devices of the reference group, an absolute angle of view angle estimate between said device to be located and said device of the reference group; third means for determining at least two differential angles, each of which is the difference between two estimates of line-of-sight absolute angles obtained for two devices of said reference group and with respect to a same predefined reference axis; fourth means for determining the relative position of said device to be located according to the differential angles determined for the device to be located. Preferably, said number of devices comprised in each of the groups constituted is three, and in that said reference value is 180 degrees. 5. LIST OF FIGURES Other features and advantages of the invention will appear on reading the following description, given by way of indicative and nonlimiting example, and the appended drawings, in which: FIG. 1 illustrates an example a wireless communication network in which can be implemented the discovery method according to a particular embodiment of the invention; FIG. 2 schematically shows the structure of a communication device implementing the discovery method according to a particular embodiment of the invention; FIG. 3A schematically illustrates a programmable antenna of a communication device; Figure 3B illustrates an exemplary radiation pattern of a programmable antenna having no angular ambiguity; FIG. 3C illustrates an example of a radiation pattern of a programmable antenna presenting an angular ambiguity; FIG. 4 represents a communication scheme in which a method for determining, for a device, an absolute angle of view line with respect to another device, according to a particular embodiment of the invention is carried out ; FIG. 5 represents a communication diagram in which is implemented a method of determining, for a device, a differential angle with respect to two other devices, according to a particular embodiment of the invention; FIG. 6 schematically illustrates a method of selecting a group of reference devices constituting a predefined geographical form, according to a particular embodiment of the invention; FIG. 7 schematically illustrates a method for determining relative positions of devices included in a reference group, according to a particular embodiment of the invention; FIG. 8 represents a communication scheme in which a method for determining the relative position of a device not belonging to the reference group illustrated in FIG. 7 is implemented, according to a first embodiment of the invention. ; FIG. 9 represents a communication scheme in which a method for determining the relative position of a device not belonging to the reference group illustrated in FIG. 7 is implemented, according to a second embodiment of the invention. ; FIG. 10 presents a flowchart of an algorithm for determining the relative positions of all the devices of the network, according to one particular embodiment of the invention; FIG. 11 presents a flowchart of an algorithm for determining the absolute positions of all the devices of the network, according to a particular embodiment of the invention. 6. DETAILED DESCRIPTION In connection with FIG. 1, an example of a wireless communication network 100 in which the discovery method is implemented, according to a particular embodiment in accordance with the invention, is presented. More particularly, the network 100 of FIG. 1 illustrates a wireless communication system, of the home cinema type for example, comprising a source device 112 having two transmission antennas 110 and 111, as well as a plurality of sending and receiving devices. receivers 120, 130, 140, 150, 160, each device being able to behave alternately as a transmitting device and a receiving device and having only one antenna for transmitting and receiving radio data signals. In addition, some nodes can play the role of relay node, that is to say that they retransmit on the network data they have previously received from another node. The network devices are all interconnected by radio communication links 170. Indeed, even if the radio signals can be broadcast in all directions, some relay nodes or receivers may not be able to detect these radio signals due to the presence of obstacles or the directivity of the antennas. A radio link is not necessarily present between a transmitting device and any other receiving device of the network. In order to limit this masking phenomenon and in order to provide information redundancy (whose objective is to facilitate the decoding of data by a receiver device), relay devices are defined in the network to relay the data transmitted by a device. transmitter device. By way of example, the device 120, considered here as a receiver, can potentially receive up to six copies (or copies) of the same original data packet, either directly from the source device 112 or via the devices relay 130, 140, 150, 160, in the case where the relay policy implemented within the network allows it.

It should be noted that the communication mode described below, which is particularly well suited to WPAN communication networks, is given for illustrative purposes only. It is clear that other modes of communication can be envisaged, without departing from the scope of the invention. The network 100 is based, in a conventional manner, on a time division multiple access mode (also called TDMA mode, for Time Division Multiple Access). This is a time division multiplexing, the principle of which is to cut out the available time in several time slots (also called time slots in English) or speech times which are allocated to the different devices of the network 100. Cyclically ( network cycle), each network device has a speaking time during which it transmits its data.

When a device transmits data, all other nodes can listen to it (for example, using their directionally configured antennas with positive gain). Each speech time can carry zero, one or more data blocks of a data stream, depending on the flow rate of this data stream, and more generally, depending on the data actually to be transmitted.

Each data block can itself be divided into a plurality of packets or symbols. The network relay devices 100 also retransmit data blocks (so-called relay blocks) to other devices of the network so that, at the end of a network cycle, the data blocks received in the different intervals of time represent different copies of the original data block. This is called Network Mesh Relay. Thus, each device is able to receive data from any other device in the network 100, whether in an aligned (i.e., directly) or non-aligned (i.e. say via one or more relay devices). To do this, the devices of the network 100 share a table describing a given sequence of time slots of the network cycle. Access to the network, during a network cycle, is therefore defined by this given sequence, the nodes communicating according to this sequence. This table may also include information relating to the duration of the time slots, as well as the formats of the packets exchanged. The structure of a communication device 200 implementing the discovery method, according to a particular embodiment according to the invention, is now presented schematically in relation to FIG.

More specifically, the communication device 200 can be integrated a central device dedicated exclusively to the implementation of the invention. By central device (also sometimes called control device) is meant a device of the network implementing centrally the implementation algorithms of the invention described hereinafter with reference to FIGS. 10 and 11. The device communication device 200 comprises: a Random Access Memory (RAM) 202 operating as a main memory; a calculation block 201 (denoted c for micro-controller in English) or a CPU (for Control Process Unit in English) whose capacity can be extended by an optional RAM connected to an expansion port (not shown on FIG. Figure 2). The CPU 201 is able to execute instructions when the communication device 200 is turned on from the ROM 203. After the power is turned on, the CPU 201 is able to execute the memory RAM 202 relating to a computer program, once these instructions loaded from the ROM 203 or external memory (not shown in Figure 2). Such a computer program, if executed by the CPU 201, causes the execution of the algorithms described hereinafter with reference to FIGS. 10 and 11; a block 205 (denoted RF-FE for RF Front-End in English) responsible for adapting the signal at the output of a baseband block 206 (denoted BB for Base-Band in English) before being transmitted by the through an antenna 204 (detailed further in connection with Figure 3A). For example, adaptation can be achieved by frequency translation and power amplification processes. Conversely, the block 205 also allows the adaptation of a signal received by the antenna 204 before its transmission to the baseband block 206. The baseband block 206 is responsible for modulating and demodulating the digital data exchanged with the baseband 206. block 205; an input / output interface block (denoted I / O If for Input / Output Interface in English) 211 connected to a communication network 212.

FIG. 3A schematically illustrates a programmable antenna 204 of a communication device, also known as a smart antenna or a smart phase antenna array, such as considered in the context of the present invention.

This programmable antenna 204 consists of an array of radiating elements or elementary antennas 301 to 306 distributed on a given support and is controlled, via the link 321, by the communication device 200. Each of the radio signals emitted and / or received, via these radiating elements 301 to 306, are controlled in phase and in power using the different phase shifters and / or amplifiers referenced 311 to 316. In the case of a transmitting device, the signal 320 supplied by the device The radiating elements 301 to 306 transmit the signal that can be out of phase and / or amplified by the respective phase shifters and / or amplifiers 311 to 316.

In the case of a receiver device, the radio signals received by the different radiating elements 301 to 306 can respectively be phase-shifted and / or amplified by the respective phase shifters and / or amplifiers 311 to 316. The programmable antenna 204 then provides a signal 320 to the communication device 200 which is the sum of the signals from the elements (phase shifters and / or amplifiers) 311 to 316. In the context of the present invention, it should be noted that the use of this type of programmable antenna allows to adjust, for each of the radiating elements 301 to 306, the phase and / or the power of the signals necessary to obtain a given radiation pattern. According to this radiation pattern, incident signals are favored in a desired direction and eliminated in undesired directions. The implementation of such a programmable antenna is performed by means of a correspondence table allowing, for an antenna orientation angle (corresponding to a given incident signal), to match a set of antenna parameters. (phase and / or power) to be applied to the different radiating elements of the programmable antenna.

In the following description, the set of antenna parameters, the set of parameters (such as phase and / or power) to be applied to the radiating elements of a programmable antenna in order to obtain a diagram of desired radiation.

During the implementation of the communications within the network 100, antenna parameters are selected and applied to the antennas of the different devices of the network 100 to allow the establishment of communication links between these devices. The application of these antenna parameters makes it possible to obtain antenna configurations adapted to the situation of the communications between the different devices of the network. To obtain a set of antenna parameters enabling the establishment of a communication link between the devices of a given pair of transmitter and receiver devices, it is possible to successively test several sets of predefined antenna parameters, in other words several predefined antenna configurations (more or less complex radiation patterns) or by carrying out an antenna scan for example in a predetermined angular sector (this antenna scan can be carried out by the transmitting device and / or the device receiver). These sets of antenna parameters correspond in this case to a set of antenna orientation angles for the transmission or reception of incident signals in certain directions. As an illustrative example, the predefined antenna parameters may correspond to a set of antenna orientation angles that can take any angle value, in steps of 5 degrees, in an angular sector between 0 degrees and 180 degrees. Among these sets of antenna parameters tested, the transmitting device (or receiver) selects at least one set of antenna parameters making it possible to implement a communication link with the receiving device (or transmitter), depending on the level of the antenna. reception quality measured by the receiver device during antenna scanning (for example between 0 degrees and 180 degrees). The level of reception quality can be measured for example using a RSSI (Received Signal Strength Indication) level measurement.

Preferably, the transmitting device (or receiver) selects the set of antenna parameters for which the reception quality level, measured by the receiver device, is the highest. The set of antenna parameters thus selected corresponds to an optimal antenna orientation angle for the establishment in line of sight (that is to say in an aligned manner) of a communication link between the transmitting devices. and receiver concerned. In the remainder of the description, absolute angle of line of sight is understood to mean an antenna orientation angle between, on the one hand, a predefined reference axis and, on the other hand, an estimated line axis. between devices of a pair of transmitter and receiver devices for establishing, with a sufficient level of communication quality, a communication link between these devices. Since the aforementioned line of sight axis is estimated (by analysis of the radiation pattern of a receiving or transmitting antenna), the absolute line of sight angle obtained is therefore also an estimate. In the remainder of the description, for the sake of simplification, no distinction is made between the absolute angle of line of sight and the estimate thereof. In this case, the determined absolute line-of-sight angle value is equal to the value of the antenna orientation angle corresponding to the set of antenna parameters selected for that angle.

It should also be noted that the programmable antenna, as shown in FIG. 3A, contains only six radiating elements. The number of radiating elements represented is deliberately limited, as a purely educational description, so as not to overload the figure and the associated description. FIG. 3B illustrates an example of a radiation pattern of a programmable antenna having no angular ambiguity. This radiation pattern makes it possible to visualize the gain of a programmable antenna, as a function of the orientation angle of the incident signal (transmitted or received), and more particularly for which a set of antenna parameters has been applied in order to obtain a high selectivity at an orientation angle of 85 degrees with respect to a predefined reference axis 370. This set of parameters gives the antenna a gain in the main lobe 330 which is very much greater than that of the lobes In this example, the radio signal transmitted or received by the programmable antenna is an 85-degree incident signal. There is only one incident signal possibility here since only one lobe is present on the radiation pattern. Thus, the antenna parameters selected to obtain an 85-degree incident signal have no ambiguity (or uncertainty) as to the determination of the absolute line-of-sight angle. Therefore, the absolute line of sight angle, obtained by the device implementing the configuration of its antenna with respect to another device of the network for which a communication link can be established, is 85 degrees with respect to reference axis 370. FIG. 3C is an example of a radiation pattern of a programmable antenna with angular ambiguity. This radiation pattern is used to visualize the gain of a programmable antenna for which a set of antenna parameters has been applied in order to obtain a high selectivity at an orientation angle of 45 degrees with respect to the reference axis 370. This set of parameters gives the antenna a gain in the main lobe 350 which is slightly greater than that of the secondary lobe 360 located at 135 degrees.

In this example, the radio signal received or transmitted by the programmable antenna may be either a 45-degree incident signal, a 135-degree incident signal, or a combination of the two 45-degree and 135-degree incident signals. Therefore, and contrary to the case illustrated in FIG. 3B, the antenna configuration obtained by applying a set of antenna parameters making it possible to obtain a 45-degree incident signal presents an ambiguity (or uncertainty) as to the determination of the absolute angle of line of sight, that which can be either 45 degrees or 135 degrees, or a combination of the two angles. Thus, for each set of parameters applied to the antenna of a device of the network, a correspondence table is established comprising the following elements: absolute angle value of the desired incident signal; information relating to the presence or absence of ambiguity; - values of absolute angles of line of sight possible. For the implementation of the discovery method according to one embodiment of the invention, it should be noted that this correspondence table is stored in the memory of a central device of the network.

FIG. 4 represents a communication scheme in which a method for determining, for a first device 410, an absolute line of sight angle 404 with respect to a second device 420, according to a particular embodiment of FIG. the invention. It should be noted that the device 410 can be indifferently transmitter or data receiver, the device 420 having a reciprocal role (receiver or data transmitter). In this figure, a communication link 402 can be established between the two devices 410 and 420 by means of the programmable antenna of the device 410. The set of antenna parameters used here for the device 410 makes it possible to obtain a Optimal antenna radiation to establish communication with the device 420. More specifically, the antenna of the device 410 is configured at a narrow radiation angle 401 (in this case referred to as a directional or selective antenna configuration) and orientation of the narrow radiation beam 401 is represented at an absolute line of sight angle 404 between, on the one hand, a predefined reference axis 403 passing through the device 410 and, on the other hand, the estimated line axis 402 between the devices 410 and 420 for which the power level measured by the receiving device is the highest compared to the device 420. The set of antenna parameters used for the device f 420 makes it possible to obtain an omnidirectional (or quasi-omnidirectional) configuration of antenna 400 (case of an isotropic radiation diagram). From the determination of the parameter set of the antenna of the device 410 enabling the establishment of the communication link 402 between the two devices 410 and 420, the correspondence table makes it possible to obtain, for the device 410, an estimation of at least an absolute angle of line of sight with respect to the device 420.

FIG. 5 represents a communication diagram in which a method is used for determining, for a device 410, a differential angle with respect to two other devices, according to a particular embodiment of the invention. In this figure, the devices 410 and 420 each have a configuration strictly identical to that presented above in relation to FIG. 4. The device 430 has meanwhile an antenna configured in a quasi-omnidirectional manner (radiation pattern isotropic), as for that of the device 420. For the device 410, a second set of antenna parameters has been selected to establish a communication with the device 430. The antenna of the device 410 is configured in a directional manner according to a narrow radiation angle 501 and oriented at an absolute line of sight angle 504 between the reference axis 403 passing through the device 410 and the estimated line of sight axis 502 between the devices 410 and 430 for which the level of the power measured by the receiver device is the highest with respect to the device 430. From the determination of the second set of parameters of the antenna of the device 4 10 enabling the establishment of the communication link 503 between the two devices 410 and 430, the correspondence table makes it possible to obtain, for the device 410, an estimate of at least one absolute angle of line of sight with respect to the device 430 The two absolute angles 404 and 504 of line of sight obtained with respect to the same reference axis 403 then make it possible to determine a differential angle 505 by making the difference between these two absolute angles. It should be noted that this differential angle 505 inherits the properties of the absolute line of sight angles from which it was determined. Thus, the ambiguity as to the determination of a differential angle is true if the ambiguity as to the determination of one of the two absolute angles of line of sight is true. A global level of uncertainty (or ambiguity) is then associated with each determined differential angle which is a function of the number of absolute angles taken into account in the calculation of the differential angle and of the ambiguity which is theirs. associated. In the case of FIG. 5, two absolute angles 404 and 504 are to be considered in order to estimate the level of uncertainty associated with the differential angle 505. Also, an overall level of communication quality is subsequently associated with each differential angle determined, this quality level being defined from the power levels measured by the receiving device for the establishment of communication links with each of the other network devices involved in the calculation of the differential angle considered. In the case of FIG. 5, only the power levels received by the receiver device via the communication links 402 and 502 are to be taken into consideration to estimate the overall quality level associated with the differential angle 505. As For example, for a set of two power levels measured by a receiver device, it is considered that: - if each power level measured by the receiver device is greater than a predetermined threshold S1, the overall level of communication quality associated with the differential angle is considered good; if one of the two power levels measured by the receiving device is below the predetermined threshold Si, the overall level of communication quality associated with the differential angle is considered to be average; if each power level measured by the receiver device is below the predetermined threshold S1, the overall level of communication quality associated with the differential angle is considered to be bad. The global levels of uncertainty and communication quality thus make it possible to characterize the differential angle measurements and to select those with the greatest relevance for the determination of the relative positions. In other words, among the set of determined differential angles, such a characterization makes it possible to select the differential angles for which the errors of measurements (or estimates) of position of the devices are minimized. It should be noted that these estimation errors are derived in particular from the characteristics specific to the antennas (for example the secondary lobes), to the possible multiple communication paths of radio transmission and / or to the phenomenon of masking by the presence of obstacles in the coverage area of the network.

Figure 6 schematically illustrates a method of selecting a group of reference devices constituting a predefined geographical form, according to a particular embodiment of the invention. According to a particular embodiment, each network device determines the set of possible differential angles with respect to the other devices of the network and transmits the result of these measurements to a central device of the network. The central device of the network performs a correlation of these results to verify the consistency of the previously determined differential angles with respect to a geometric shape whose number of vertices is predefined.

The central device then establishes a list of possible combinations of devices for constructing the predefined geometric shape from a predetermined number of network devices, each device representing a vertex of that geometric shape. According to a particular embodiment of the invention, this predefined geometric shape is a triangle (set of three devices). According to alternative embodiments of the invention, the predefined geometric shape may be a quadrilateral, a tetrahedron or a polygon having a remarkable property based on the values of the angles constituting its vertices, ... Among the set of possible combinations of devices, the central device then determines the group of devices having the highest coherence with the predefined geometric shape. In the case of a triangular geometric shape, the central device selects the best combination of devices from the following criteria: - the difference, in absolute value, between the sum of the three differential angles associated with the tested triangle and the value of 180 degrees ; the overall level of communication quality associated with the three differential angles of the triangle tested; the overall level of uncertainty associated with the three differential angles of the tested triangle.

It should be noted that this list of selection criteria is not exhaustive.

Two possible cases of combination of these criteria are described below as purely illustrative examples. In a first case, the central device is loaded for example to select only the group of devices for which the difference in absolute value between the sum of the three differential angles associated with the tested triangle and the value of 180 degrees is minimal. In case of equality of results for several groups of devices, the central device selects from among these groups the group of devices for which the overall levels of communication quality and uncertainty associated with the three differential angles of the tested triangle are the best. . Again, in case of equality, a weighting may be applied to at least one of the aforementioned criteria until a single group of devices is obtained. In a second case, the central device is responsible for preselecting first of all the groups of devices for which the difference, in absolute value, between the sum of the three differential angles associated with the tested triangle and the value of 180 degrees is more a predetermined margin of error, then from among these preselected groups, to select the one for which the overall levels of communication quality and uncertainty associated with the three differential angles of the tested triangle are the best. In case of equality for several groups of devices, it may be decided to reduce the predetermined margin of error, or to apply a weighting on at least one of the aforementioned criteria, until a single group of devices . The group of devices thus constituted is subsequently called reference group (or reference triangle in the case where the desired geometric shape has three vertices).

Figure 6 illustrates one of these combinations. The device 410 determines the differential angle 505 between the devices 420 and 430, the device 420, the differential angle 506 between the devices 430 and 410, and the device 430, the differential angle 507 between the devices 410 and 420. The reference group, consisting of devices 410, 420, 430, was selected from among the network devices that best met the criteria listed above.

An example of a reference triangle is illustrated below in relation to FIG. 7. FIG. 7 diagrammatically illustrates a method for determining relative positions of devices included in a reference group, according to a particular embodiment of the invention. It is more particularly in this case to determine the relative position of each of the three devices 706, 707, 708 included in the reference triangle 710. The device 706 determines the differential angle 702 between the devices 707 and 708, the device 707 , the differential angle 700 between the devices 706 and 708, and the device 708, the differential angle 701 between the devices 706 and 707. The reference group 710, consisting of the devices 706, 707 and 708, has been determined for example because the sum of the calculated differential angles 700, 701, 702 is closest to the value of 180 degrees (out of the set of determined differential angles). From the differential angles 700, 701, 702 determined for the reference triangle 710, the central device is able to determine the relative dimensions of this reference triangle 710 by the sine law. We consider the following notations, for the reference triangle 710: - A, the differential angle 700; B, the differential angle 701; C, the differential angle 702; L, the distance 703 between the devices 706 and 707; M, the distance 704 between the devices 707 and 708; N, the distance 705 between the devices 706 and 708.

In addition, the distance 703 between the devices 706 and 707 is arbitrarily considered to be the reference metric. According to the sine law applied to the reference triangle 710, we obtain the following trigonometric relation (1): L / sin (B) = M / sin (C) = N / sin (A) (1). distance L (reference metric) as unitary, the following relative distances are obtained: M = sin (C) / sin (B) N = sin (A) / sin (B) The relative dimensions of the reference triangle 710 are thus determined. From the determination of the relative dimensions of the reference triangle 710, it is possible to determine the relative position of a device included in this reference triangle 710 with respect to another device also included in this reference triangle 710. Indeed a first device 706 is considered to be the center of an orthonormal coordinate system whose abscissa axis coincides with the predefined reference axis of its antenna (corresponding to the axis 403 of FIGS. 4 and 5). One seeks to know the relative position of another device 707 with respect to the first device 706. The following notations are considered: X and Y, the coordinates of the device 707 to be located in the orthonormal coordinate system of the device 706; D, the distance between the two devices 706 and 707; - Ai, the absolute line of sight angle of the device 706 included between the predefined reference axis of its antenna and the estimated line of sight line with the device 707.

We then obtain the following trigonometric relations: X = D. cos (Ai) Y = D. sin (AI) The relative position (X, Y) of the device 707 included in the reference triangle 710 can therefore be determined from these relationships.

Thus, it is possible to determine each of the relative positions of the devices included in the reference triangle 710. It is therefore possible to determine, with a high degree of confidence, the relative distances of a part of the devices of the communication network (triangle reference 710). This reference triangle 710 can therefore serve as a basis for determining thereafter the relative position of network devices that do not belong to this reference triangle.

FIG. 8 represents a communication scheme in which is implemented a method for determining the relative position of a device not belonging to the reference group illustrated in FIG. 7, according to a first embodiment of the invention .

After having determined a reference group consisting of three devices according to the principle described above in relation to FIGS. 6 and 7, it is possible to determine the relative position of a device to be located which does not belong to the reference group 710. This determination method is performed according to the reference group as follows.

In FIG. 8, the device which does not belong to the reference group and which one seeks to locate, is referenced 810. It should be noted that this device 810 can be indifferently transmitter or receiver in the establishment of the links of 821, 822 and 823 with the devices 706, 707 and 708, the latter having a reciprocal role to that of the device 810. In the case of Figure 8, the antenna used by the device 810 is configured in a quasi-omnidirectional manner 800. Antenna parameters used by devices 706, 707, and 708 for obtaining directional antenna patterns 811, 812, and 813, and radiation beams oriented such that they allow for the establishment of link patterns. communication 821, 822 and 823. From the correspondence table stored in the memory of the central device, it then obtains an estimate of the absolute angles 801, 802 and 803 of line of sight, for the devices. 706, 707 and 708, respectively, with respect to the device to be located 810. Each of these absolute line of sight angles is between the predefined reference axis associated with the device for which the absolute angle is calculated and the estimated axis line of sight between this device and the device to be located 810. From the knowledge of each of the absolute angles 801, 802 and 803 of line of sight, as well as the relative position of each of the three devices 706, 707 and 708 of the reference triangle, the central device calculates the equations of lines each associated with a line of sight axis between a device of the reference triangle and the device to be located 810. Each equation on the line (also hereinafter referred to as line equation) view) is determined by using as a center of orthonormal reference a device of the reference triangle. The line of sight equations are defined as follows: * Y1 = X1.sin (Ai), in the orthonormal center coordinate system the device 706 whose abscissa axis coincides with the predefined reference axis of its antenna with Ai the absolute line-of-sight angle of the device 706 with respect to the device to be located 810; * Y2 = X2.sin (A2), in the center orthonormal coordinate system the device 708, whose abscissa axis coincides with the predefined reference axis of its antenna, with A2 the absolute angle of the line of view of the device 708 with regard to the device 810. The central device then performs a change of reference (by rotation and reference translation), for each of the previously determined line of sight equations, in the same orthonormal frame in order to be able to determine the intersection . In the case of FIG. 8, the intersection of the lines determined from the lines of view of the devices of the reference triangle with respect to the device to be located 810 makes it possible to estimate the position of this device 810. According to a first embodiment In particular, a line-of-sight equation is determined for each device belonging to the reference group. According to a second particular embodiment, only the lines of sight, for which the equation has been determined from absolute line of sight angles presenting no ambiguity and a good level of communication, are taken into account for determining the relative position of the device to be located.

The intersections of these lines make it possible to estimate several probable relative positions of the device to be located. Thus, several probable points of location of the device are obtained. In this case, the position of the device to be located can be determined, for example, by taking the center of a cloud of points representative of all the intersections of the line of sight lines obtained.

In the case of FIG. 8, it is therefore the reference group that is used to determine the relative position of the device 810.

Briefly below is the calculation of a point of intersection between two lines considered. The general equation of a line is generally written as: Y = mx + b where m is the slope of the line, and b is its intercept.

By knowing two points of a line defined by their coordinates (xi 'y ~) and (x2' y2), it is then possible to determine the values m and b. (Y2 -YI) Thus, m = (x2 - x) and b being deduced then from the equation y = mx + b We now consider two straight line segments of equations y = mix + bI and y = m2x + b2 (mi ùm2) The values of bl and b2 are then deduced from the equations y = mix + bI and y = m2x + b2 It should be noted that simi = m2, the two line segments considered are parallel and therefore do not have intersection. In a particular embodiment, the central device, once the relative position of a known device (such as device 810), adds it to the reference group to form a new reference group. In the context of the example of FIG. 8, if the device 810 is introduced into the reference group, the latter will now comprise a set of four devices. Incrementally, it is therefore possible to integrate in a reference group a device whose relative position has been determined. The new reference group thus obtained having a number of data (absolute angles of line of sight, differential angles, power levels and associated levels of uncertainty, ...) more important, it therefore allows to estimate, so more precisely, the relative position of other devices of the network which one seeks to know their position. The risks of estimation errors are thus minimized incrementally.

The coordinates of the point of intersection of these two segments then satisfy the following relation: y = mi x + b1 = m 2 x + b2, ie (mi - m2) x = b2 - bi Finally, we obtain the following relation: x = (b2-bi) FIG. 9 represents a communication scheme in which a relative position determination method of a device not belonging to the reference group illustrated in FIG. 7 is implemented, according to a second embodiment of FIG. embodiment of the invention.

As described above in relation to FIG. 8, it is for the central device to determine the relative position of a device that does not belong to the reference group 710. The device to be located which does not belong to the Reference group is referenced 910. The respective antenna parameters used by the devices 706, 707 and 708 for obtaining quasi-omnidirectional antenna patterns 931, 932, 933, 812 (isotropic radiation pattern). The device 910 performs a scan of its antenna configured in a directional manner. I1 then applies several sets of antenna parameters in order to be able to generate beams of radiation 901, 902, 903 oriented so that they allow the establishment of communication links 921, 922 and 923, with a quality level of sufficient communication, with the devices 708, 707, 706 respectively. It is recalled that the device 910 can be indifferently transmitter or receiver in the establishment of the communication links 921, 922 and 923, the devices 706, 707 and 708 having a reciprocal role to that of the device 910. In this case, the device 910 determines the absolute angles of line of sight between the reference axis of its antenna and the estimated line of sight axes with respect to each device (706, 707, 708) included in the reference triangle and transmits the results of these measurements to the device central network. From the stored correspondence table and the absolute angles thus determined, the central device then determines, for the device 910, two differential angles 940 and 950 with respect to the devices 708 and 707 and the devices 707 and 708 respectively. These two determined differential angles 940, 950 then enable it to deduce the differential angle 961 (between the line of sight axes between the devices 707 and 708 and the devices 708 and 910) and the differential angle 962 (included between the line of sight axes between devices 706 and 707 and devices 706 and 910), using the following trigonometric resolution. We first consider the following notations: - let x be the value of the differential angle 962 (unknown); either y is the value of the differential angle 961 (unknown); either a, the value of the differential angle 940 (known); or 13, the value of the differential angle 950 (known); - 0, the value of the differential angle 700 (known); or b, the relative distance (known) between the device 708 and the device 706; either a, the relative distance (known) between the device 707 and the device 706; or m, the relative distance (unknown) between the device 910 and the device 708. The sum of the angles of a quadrilateral whose vertices are the devices 910, 708, 707, 706 being equal to 360 degrees, it is possible to deduce the The following expression x + y + a + (3 + 0 = 360 degrees) gives the following equation (I): x + y = 360uau (3oo (I) After applying the sine law to the triangle whose vertices are the devices 708, 706, 910 give the following expression: m / b = sin (x) / sin (a) The relative distance sought m can therefore be deduced from the following equation (II): m = b.sin (x) / sin (a) (II) After applying the law of sinuses to the triangle whose vertices are the devices 707, 706 and 910, we obtain the following expression: m / a = sin (y ) / sin ((3) The relative distance sought m can therefore be deduced from the following equation (III): m = a.sin (y) / sin ((3) (III) From the equations ( II) and (III), the following expression can be written: sin (x) / sin (y) = (a.sin (a)) / (b.sin (13)) After simplification of the expression above, we deduce the following equation (IV): 30 sin (x) / sin (y) = b '/ b (IV) with: b' = a.sin (a) / sin ((3). Since [sin (x) / sin (y)] + 1 = [b '/ b] + 1 gives [sin (x) + sin (y) / sin (y)] = (b' + b) / b and [sin (x) / sin (y)] - 1 = [b '/ b] -1 give [sin (x) -sin (y) / sin (y)] = (b'-b) / b we deduce that: (b'-b) / (b '+ b) = [sin (x) - sin (y)] / [sin (x) + sin (y)] Now, according to the law tangents, we note that (b'-b) / (b '+ b) = tan /2.dixy] / tan /z.[xyy] From this expression, it is therefore possible to determine the value (xy), since the values b, b 'and (x + y) are known from the equations (I) and (IV). Knowing (x + y) and (x-y), one can thus easily obtain the values of x and y corresponding to the differential angles referenced 961 and 962, using a system of equations with two unknowns. From the knowledge of the differential angles 961 and 962 and the relative position of the devices 706 and 708, the central device determines the equations of lines each corresponding to a line of sight of the device 910 with regard to the devices 706 and 708.

FIG. 9 illustrates the case where the devices are represented so as to constitute a quadrilateral for which the device 910 is situated in the differential angle 700. Indeed, since the device 910 has been able to determine differential angles 940 and 950 and since the sum of these differential angles 940 and 950 is in this case necessarily lower than the value 180 °, it can therefore be considered that the device 910 is located in the differential angle 700 of the device 707. From these equations on the right and according to the principle described above in relation to Figure 8, the central device is able to determine the point of intersection of these lines. The relative position of the device 910 with respect to the reference group can therefore be determined.

It should be noted that, in this second particular embodiment, it is the device whose position is sought to be determined which is used to implement the steps of determining the relative position of a device of the network by report to the reference group. FIG. 10 shows a flowchart of an algorithm for determining relative positions of all the devices of the network, according to a particular embodiment of the invention.

In a first step 1000, the central device obtains, for each possible pair of transmitter and receiver devices of the network, the power levels measured by the torque receiving device as a function of the antenna parameters applied to the transmitting device and / or the device receiver. This is more particularly the antenna scanning step (s) which aims to determine the antenna parameters to be applied to establish the communication links between each pair, as described in connection with Figure 3A. In a step 1001, the central device obtains, for each device of the network, an estimate of one or more absolute angles of line of sight with respect to one or more other devices of the network. In a step 1002, the central device establishes a list of possible combinations of devices making it possible to construct a predefined geometrical shape from a predetermined number of devices of the network, each device representing a vertex of this geometrical shape (for example three devices in the case of a triangular geometrical shape). Then, for each of the devices of the constituted groups, it determines one or more differential angles with regard to other devices of the same group. It is recalled that a differential angle is obtained by making the difference between two absolute angles previously obtained for two devices of the group considered with respect to the same antenna reference axis passing through the device for which the differential angle is sought. The central device also obtains a global level of uncertainty (or ambiguity) and an overall level of communication quality associated with each determined differential angle. In a step 1003, the central device is able to select, from all the groups of devices constituted, a reference group according to selection criteria. These criteria are discussed above in relation to FIG. 6. According to a particular embodiment, the central device selects a group of devices for which the difference, in absolute value, between the sum of the determined differential angles and a predetermined value is the weaker.

This predetermined value is for example 180 degrees for the case of a triangle or 360 degrees in the context of a quadrilateral.

In this way, the selection phase of a reference group is more efficient. The central device then determines the relative positions of the devices included in the reference group according to the principle described above in relation to FIG. 7. The reference group is then used as a basis for the determination of relative positions of other devices of the network ( not belonging to the group). In a step 1004, the central device uses the relative positions of the devices included in the reference group to determine the relative positions of devices not belonging to the reference group.

According to a first embodiment of the invention (according to the principle detailed above in relation to FIG. 8), the central device determines the relative position of the device to be located as a function of the absolute angles of line of sight obtained for this purpose. compared to other devices in the reference group. According to a second embodiment of the invention (according to the principle detailed above in relation to FIG. 9), the central device determines the relative position of the device to be located according to the differential angles determined for this device, which are each the difference between two absolute angles obtained for two devices of said reference group and with respect to the same predefined reference axis.

According to another embodiment, in the case where the central device has measures making it possible to obtain differential angles exploitable by the two aforementioned embodiments, the latter can implement each of these two embodiments to determine the position of the same network device. The central device is then able to estimate the coordinates of the device to be located from an average of these two results (as for example using a calculation of centroid). The central device having thus determined the relative position of a device not belonging to the construction group, then tests, in a step 1006, whether all the devices of the network have all been located in a relative manner. If the test result is positive, then the algorithm ends at step 1005.

If the result of the test is negative, step 1004 is repeated for each device whose relative position is to be determined.

It should be noted that, according to a particular embodiment of the invention, each device whose position has been determined can be added, as and when, to the reference group in order to form a new reference group. Advantageously, the central device may seek to introduce, iteratively, a known relative position device for which the differential angles have the lowest overall level of uncertainty and / or the highest overall level of communication quality. . At the end of this iterative process (step 1005), the central device has a list containing all the relative positions of the network devices relative to the reference group whose metric (relative distance L between the devices 706 and 707 for example) is one of the sides of the geometric form that represents the reference group. A flow chart of an algorithm for determining the absolute positions of all the devices of the network, according to a particular embodiment of the invention, is now presented in relation to FIG. 11. After knowing all the relative positions of the network devices according to the determination method illustrated above in relation to FIG. 10, the central device obtains, in a step 1101, information relating to an absolute distance between two devices of the device. network whose central device knows the relative positions. This information can be provided for example by a user of the communication network which performs a distance measurement between two devices of the network and informs the central device. This information can also be obtained by the central device itself thanks to a particularity of a given equipment. For example, as shown in FIG. 1, the network 100 comprises a device 112 equipped with two communication devices 110 and 111, these two devices being spaced apart by a predefined absolute distance. The two communication devices of such equipment may have a particular identifier which is exchanged during an initialization phase of the network 100 and which allows the central device to identify these devices among all the network devices. and thus locate the predefined distance separating them from one another. As an illustrative example, this particular equipment may be a flat screen having two communication devices (two programmable antennas for example), each device being located on a vertical upright of its chassis. This particular equipment can also be an audio / video type source equipment having two communication devices, each being located at the front of each end. In a step 1102, the central device recalculates the relative positions of the set of devices by performing a marker change taking as the center of the marker the position of one of the two devices for which it knows the absolute distance. To determine the set of absolute positions of the devices of the network, the central device must know the real value of the metric used. This metric is equal to D / R, where D is the known absolute distance and R is the relative distance between two devices previously determined. From the knowledge of all the relative positions of the network devices and the absolute distance between two devices of the network, the central device is able to deduce the absolute position of all devices. In a last step 1103, the central device has a list containing all the absolute positions of the devices of the network. 5 10 15 25 30

Claims (14)

REVENDICATIONS1. A method of discovering positions of devices in a wireless communication network (100), the method being characterized in that it comprises the steps of: - obtaining (1001) for each device of a set of devices included in the network, an estimate of at least two absolute line-of-sight angles, each between an estimated line of sight line between said device and another device of the network and an axis of view. predefined reference; constituting (1002) a plurality of groups of a predefined number of devices among the devices of said set; determining, for each device of each of the groups constituted, at least one differential angle which is the difference between two absolute angle estimates obtained for two devices of said group constituted and with respect to the same predefined reference axis; selecting (1003), from the groups constituted, a reference group (710) according to a criterion which is, for each of the groups constituted, the difference between on the one hand a sum of differential angles, each determined for one of the devices of said group constituted, and secondly a reference value; determining (1004) the relative position of at least one device of the network according to the selected reference group (710). 2. Method according to claim 1, characterized in that said step of selecting (1003) a reference group (710) is performed according to at least one additional criterion belonging to the group comprising: for each of the groups constituted, an overall level of communication quality associated with the differential angles each determined for one of the devices of said group constituted; for each of the groups constituted, an overall level of uncertainty associated with the differential angles each determined for one of the devices of said group constituted. 3. Method according to any one of claims 1 to 2, characterized in that said step of determining (1004) the relative position of at least one network device according to the reference group (710) selected comprises steps consisting of, for a device to be located which does not belong to the reference group (710): for each of the devices of the reference group (710), obtaining an absolute angle of view angle estimate between said device to be located and said reference group device (710); determining the relative position of said device to be located as a function of the estimates of absolute line of sight angles obtained for the devices of the reference group (710). 4. Method according to any one of claims 1 to 2, characterized in that said step of determining (1004) the relative position of at least one network device according to the reference group (710) selected comprises steps consisting of, for a device to be located which does not belong to the reference group (710): for each of the devices of the reference group (710), obtaining an absolute angle of view angle estimate between said device to be located and said reference group device (710); - determining at least two differential angles which are each the difference between two estimates of line-of-sight absolute angles obtained for two devices of said reference group (710) and with respect to a same predefined reference axis; determining the relative position of said device to be located according to the differential angles determined for the device to be located. 5. Method according to any one of claims 1 to 4, characterized in that it comprises a step of introducing into said reference group (710), at least one device whose relative position has been determined. 6. Method according to any one of claims 1 to 5, characterized in that said number of devices included in each of the groups constituted is three, and in that said reference value is 180 degrees. 25 7. Method according to any one of claims 1 to 6, characterized in that an operation of obtaining an absolute angle of view angle estimate of a first device with respect to a second device is performed by analysis of radiation pattern of an antenna receiving or transmitting said first node. 8. Method according to any one of claims 1 to 7, characterized in that it comprises the steps of: - obtaining (1101) a distance value between two devices whose relative positions have been determined; determining (1102) an absolute position of at least one device whose relative position has been determined as a function of said obtained distance value. 9. Computer program product, characterized in that it comprises program code instructions for implementing the method according to at least one of claims 1 to 8, when said program is executed on a computer. A computer readable storage medium storing a computer program comprising a set of computer executable instructions for carrying out the method according to at least one of claims 1 to 8. Apparatus device discovery device included in a wireless communication network (100), the discovery device being characterized in that it comprises: - first obtaining means, for each device of a set of devices included in the network, an estimate of at least two absolute angles of line of sight, each between an estimated line of sight between said device and another device of the network and other part of a predefined reference axis; means for constituting a plurality of groups of a predefined number of devices among the devices of said set; first means for determining, for each device of each of the groups constituted, at least one differential angle which is the difference between two absolute angle estimates obtained for two devices of said group constituted and with respect to the same reference axis predefined means for selecting, from among the groups constituted, a reference group (710) according to a criterion which is, for each of the groups constituted, the difference between on the one hand a sum of differential angles , each determined for one of the devices of said group constituted, and secondly a reference value; second means for determining the relative position of at least one device of the network as a function of the selected reference group (710). 12. Discovery device according to claim 11, characterized in that said second means for determining the relative position of at least one network device according to the selected reference group (710) themselves comprise, for a device to locating not belonging to the reference group (710): second means for obtaining, for each of the devices of the reference group (710), an absolute angle of view angle estimate between said device to be located and said reference group device (710); third means for determining the relative position of said device to be located as a function of the estimates of absolute line of sight angles obtained for the devices of the reference group (710). 13. Discovery device according to claim 11, characterized in that said second means for determining the relative position of at least one network device according to the selected reference group (710) themselves comprise, for a device to locating not belonging to the reference group (710): second means for obtaining, for each of the devices of the reference group (710), an absolute angle of view angle estimate between said device to be located and said reference group device (710); third means for determining at least two differential angles which are each the difference between two estimates of absolute line of sight angles obtained for two devices of said reference group (710) and with respect to the same predefined reference axis ; fourth means for determining the relative position of said device to be located according to the differential angles determined for the device to be located. 30 14. Discovery device according to any one of claims 11 to 13, characterized in that said number of devices included in each of the groups formed is three, and in that said reference value is 180 degrees.