GB2518651A - Wireless geometry discovery - Google Patents

Abstract

A method for determining the geometry of an array of wireless communication devices, each including at least one antenna, from a set of possible geometries is disclosed. An antenna discovery protocol is performed at a first device in order to detect neighbouring devices 610. A further device is selected from the detected devices 615, and an antenna discovery protocol is performed at the further device in order to detect devices neighbouring the further device 620. A list of possible geometries is updated by eliminating geometries not compatible with results of previously performed antenna discovery protocols 625. One or more further devices may be iteratively selected from devices detected in previous antenna discovery operations, and the list of possible geometries updated until a single geometry is determined 630. The updating may be performed at the first device, or at one or more of the further devices. The antenna discovery protocol preferably comprises emitting one or more probe messages and receiving one or more feedback messages from neighbouring devices, if any. The probe messages may be emitted according to a predetermined set of antenna directions. The wireless communication devices preferably comprise video projectors which are networked to form a multi-projection system.

Description

WIRELESS GEOMETRY DISCOVERY

The present invention relates in general to wireless communication, and particularly, but not exclusively, to a multi-projection system comprising a plurality of wireless enabled video projector devices.

Certain multimedia applications, such as high-definition audio/video streaming, require transmission of uncompressed video at high data rate of about several Gbps (Gigabits per second), with low latency. Such rates are possible even in wireless with the use of new higher frequencies as the 57-66 GHz millimeter-wave unlicensed spectrum, referred to as 60 GHz millimeter wave technology. 60 GHz-based communication systems are widely studied (e.g. IEEE 802.11 Task Group; IEEE 802.15.3c standard; Wireless HD; WiGiG; etc.) and the research community proposes several solutions and methods to transport the audio and video applications with a desired quality of service (QoS).

One example of such high definition audio/video streaming application is the large high-resolution displays created by tiling together multiple video projectors, which are increasingly used for many applications, such as visualization, training, simulation and collaboration. For instance, each video projector displays a high definition video so as to form a large logical display wall. The video projectors may be wireless enabled, to allow wireless communication between video projector devices.

In such a system, referred to as a multi-projection system, an original image is partitioned according to the arrangement of the video projectors so as each video projector displays its own image (a sub-image of the original image) in order to display the original image in a large display. Consequently, the arrangement of each video projector is a key factor in the configuration of a multi-projection system. It includes the locations and the positions of each video projector. For instance, some instances of arrangements are array, dome, 8K, panoramic, wall display, dome, 360°.

However, this notion of arrangement can be improved by the introduction of additional parameters. One parameter is the relative orientation between two devices which can be expressed with angles for instance. Another parameter is the distance between the video projectors which can be expressed via the transmit power. Obviously, other parameters can be defined.

Consequently, a new notion, the geometry of a set of devices (the geometry of a multi-projection system with regard to projector devices) is introduced. This new notion includes the arrangement and specific parameters related to its arrangement. In this way, once the geometry of a set of devices is identified, a multi-projection system is able to adapt the video and to determine for instance how to cut the video in order to create sub-part, each sub-part being intended to one video projector.

US 8,123,360 discloses a method of geometry configuration, however this method relates only to a wired configuration and cannot be adapted to wireless devices.

It is an object of the invention to provide an improved method of determining the geometry of a set of wireless devices. In other words, it consists in identifying the geometry (of a set of wireless devices) among a list of predetermined geometries. Moreover, it is an object to perform this determination automatically, without for instance the intervention of a user who is in charge of the geometry of the system.

According to a first aspect of the invention there is provided a method of determining the geometry of an array of wireless communication devices, each device including at least one antenna, from a set of possible geometries, the method comprising performing, at a first device, an antenna discovery protocol to detect devices neighbouring said first device; selecting a further device from detected devices neighbouring the first device; performing, at said further device an antenna discovery protocol to detect devices neighbouring said next device; updating the list of possible geometries by eliminating geometries not compatible with results of previously performed antenna discovery protocols.

In this way, starting from an initial device, which may be selected randomly from among the devices, or may be user selected, the geometry of a local network of neighbouring devices is determined, by an antenna discovery protocol. This local geometry can be compared to a list of possible (global) geometries to test for compatibility, and eliminate global geometries which are not consistent with the determined local geometry.

Advantageously the method comprises iteratively selecting one or more further devices from neighbouring devices detected in a previously performed antenna discovery protocol, and performing an antenna discovery protocol at said one or more further devices, and updating the list of possible geometries until a single geometry is determined.

Hence the geometry discovery method can extend across the network, one device at a time, at each stage adding to the known local geometry. As the local geometry expands in size, the number of possible compatible global geometries will typically reduce, until only one feasible geometry remains. At this point the iterative process ends.

In embodiments, the result of an antenna discovery protocol comprises a set of antenna directions corresponding to communication paths between neighbouring devices. In this way a local geometry can be defined by a set of devices or nodes and a set of interconnections between nodes. The interconnections can be defined by the angle they make between neighbouring nodes.

In related embodiments each of the possible geometries may also be defined by information representing communication paths between devices.

In addition to an angle corresponding to the direction of a communication path between devices or nodes of a network, a measure of distance may also be included. The power corresponding to each communication path may be determined, and used to further define the network. The power typically corresponds to a transmission power required to establish communication between two devices, and acts as a measure of distance between two devices.

In one embodiment, updating the set of possible geometries is performed at the first device. Results of antenna discovery performed at other devices can be relayed back to the first device and used. Thus a local geometry is developed at the first device. Alternatively, the updating may be performed at another, further device. In such an embodiment, results from each antenna discovery are carried forward and incorporated to the determined local geometry dynamically, at a current device performing an antenna discovery.

The invention also provides a computer program and a computer program product for carrying out any of the methods described herein and/or for embodying any of the apparatus features described herein, and a computer readable medium having stored thereon a program for carrying out any of the methods described herein and/or for embodying any of the apparatus features described herein.

The invention extends to methods, apparatus and/or use substantially as herein described with reference to the accompanying drawings. Any feature in one aspect of the invention may be applied to other aspects of the invention, in any appropriate combination. In particular, features of method aspects may be applied to apparatus aspects, and vice versa. Furthermore, features implemented in hardware may generally be implemented in software, and vice versa. Any reference to software and hardware features herein should be construed accordingly.

Preferred features of the present invention will now be described, purely by way of example, with reference to the accompanying drawings in which: Figure 1 depicts for illustrative purposes a multi-projection system comprising multiple projection display apparatus; Figure 2 depicts two examples of antenna that may be included in a device, one having a single narrow main beam and the other having a single wide main beam.

Figure 3 illustrates the description and the storage of a network of devices.

Figure 4 illustrates an antenna discovery protocol process performed between a device and its neighbour(s).

Figure 5 illustrates messages sent during the antenna discovery protocol process between a first device and its neighbour(s).

Figure 6 is a flowchart showing the discovery protocol process of a multi-projection system.

Figure 7 is a flowchart showing the emission process of discovery protocol of a multi-projection system; Figure 8 is a flowchart showing the reception process of discovery protocol of a multi-projection system in centralized version; Figure 9 is a flowchart showing the reception process of the discovery protocol of a multi-projection system in distributed version; Figure 10 illustrates the discovery propagation message sent during the discovery protocol of a multi-projection system; Figure 11 illustrates the discovery feedback message sent during the discovery protocol of a multi-projection system in distributed version; Figure 12 is the flowchart showing the checking process of a predetermined geometry in order to know if it is compatible with the results of the previously performed antenna discovery protocols Figure 13 illustrates the the discovery protocol in the centralized and distributed version Figure 14 is a schematic block diagram of a computing device for implementing one or more embodiments of the invention.

Figure 1 depicts for illustrative purposes a multi-projection system 100 comprising multiple projection display apparatus 111, 112, 113 and 114 for projecting on a screen video frames delivered by a source device 101 (video source). The projection display apparatus is typically a video projector that projects a video stream but may encompass any type of projector such as for example a still image projector.

The delivery of image data is performed via communication paths forming a communication network. Each projector of the multi-projection system 100 receives image data either directly from the source device 101 over one of the communication paths 121, 122, 123 and 124, or indirectly from another projector by a further transfer (forward) over one of the inter-projectors communication paths 130, 131,132,133, 134 and 135.

According to one implementation example, the communication network is a full wireless communication network, i.e. all the communication links are wireless.

This wireless communication network may operate in the 57-66 GHz millimeter-wave unlicensed spectrum for example for providing the necessary bandwidth for the transport of the video data; particularly if this latter is high definition (HD) video data. Alternatively, direct communication paths 121 -124 are wired communication paths -e.g. HDMI -while inter-projectors communication path 130-135 are wireless communication paths.

An image to be projected is split into a plurality of sub-images. The number of sub-images per image is typically, but not necessarily, equal to the number of projectors in the multi-projection system in use. The size and shape of each sub-image is chosen so that a full composite image can be reconstructed when all the sub-images are projected by their corresponding projectors.

Relative to figure 1, the geometry of the multi-projection system 100 is a 2X2 array of 4 video projectors. The additional parameters are detailed in figure 3.

Figure 2 depicts two examples of antenna which can be integrated in one device such the devices 111,112,113 or 114: one has a single narrow main beam and the other has a single wide main beam. Such antennas (for which the notion of direction is used) coupled with the antenna discovery protocol described in figure 4, allow the relative position between two devices to be determined, a key parameter in the notion of geometry. A wireless communication path between two devices is established thanks to the use of antennas included in the said devices (module 1404 in figure 14). For each antenna, two modes may be defined: directional mode and wide mode.

In directional mode, the antenna focuses on the radio frequency in one direction.

Steering an antenna to a given orientation corresponds to controlling its parameters (for example the weighting coefficients associated with the elements of an antenna array) such that the radiation pattern, in case of emission, or the antenna sensitivity pattern, in case of reception, is accentuated in that given direction relatively to other directions. Figure 2a depicts an example of an antenna 210a which transmits a signal via a single narrow main beam 220a (angle equals 5°, measured at -3dBi from the maximum, where "dBi" represents a measure of antenna gain relatively to an isotropic antenna). The main beam gain is relatively high, for example 25 dBi. The antenna has thus different gain characteristics at different reception angles. A maximum gain is obtained in the direction of the main beam.

Alternatively, in wide mode, an antenna performs a transmission towards or reception from a plurality of transmission paths which can be affected simultaneously (covering simultaneously a plurality of directions). Figure 2b depicts an example of an antenna 210b having a single wide main beam 220b (angle equals 2100). The main beam gain is thus relatively small, approximately 4dBi. A wide beam or near omni-directional antenna can be implemented as a single dedicated antenna element. Alternatively, an antenna array comprising a plurality of antenna elements dedicated for directional transmission can be reused by activating a single antenna element among the array of antenna elements.

Often, it is not possible to cover all the directions even with a single antenna having wide beam. In order to overcome this issue, complementary antennas can be used in the same device, each one covers a part of the directions. For instance, in the case wherein a device has 2 antennas, the first antenna covers the directions between 0 and 180 degrees and the second antenna covers the direction between 181 and 360 degrees. Consequently, when a device has to send a message towards a given direction, it selects the antenna which covers it.

For purpose of simplification, the antennas of a same device are, in preferred embodiments, either all in wide or all in directional. Consequently, when the antennas of a device are all in wide, the device is considered as in omnidirectional' (as the antennas of a same device are complementary). In the same way, when the antennas of a device are all in directional, the device is considered as in directional'. Moreover, at a given time, a given antenna is either in emission, ready to send data or (exclusive) in reception, ready to receive data.

For purpose of simplification, the antennas of a device are, in an embodiment, either all in emission or all in reception. Consequently, when the antennas of a device are all in emission (resp. in reception), the device is considered as in emission' (resp. in reception').

Consequently, 4 antenna modes are introduced for a given device: * Emission/Wide: the device is in emission and in wide * Emission/Directional: the device is in emission and in directional * Reception/Wide: the device is in reception and in wide Reception/Directional the device is in reception and in directional Figure 3 illustrates how to describe and to store a geometry of a multi-projection system.

A geometry is described by an array 300 of dimension K*K (K rows, K columns), K being the number of devices in the system. It is referred to as geometry array.

A non-empty element of the array located at the row i and the column corresponds to the existence of a communication path between the device i and the device j. An empty element of the array means that there is no possible communication path in any directions between the two corresponding devices.

Each element 310 of the array 300 contains the transmitter antenna parameters used at device i and device j which are used to define the communication path between the two devices i and j.

In the rest of document, the considered transmitter antenna parameters are the antenna directions and transmit power but others may additionally or alternatively be used.

The identifier device considered in a geometry array is referred to as the geometry device identifier. It is different of the identifier device used at the beginning of the discovery algorithm based on for instance the MAC address of the source device or a function of its MAC address. In such a case, the discovery algorithm allows the geometry device identifier to be linked to the device identifier. In this way, when the link between the geometry device identifier and the device identifier is established, the corresponding line (corresponding to the (geometry device identifierf' line) is considered identified. At the beginning of the process, no line is identified. The line identification is performed by the step 625, and in particular the steps 1265 and 1275 described below.

For each geometry of the list of predetermined geometries there is associated a geometry array which is stored in the ROM 1403 of each device.

Relative to figure 1, for purpose of simplification, the distances between the devices 111, 112, 113 and 114 are supposed identical and a given transmit power can be fixed (for instance 0.01W). As for the antenna direction, the geometry array relative to figure 1 can be following one: Geometry 1 2 3 4 device identifier I X 0°/i 80° 270°/90° X 2 18070° X X 270°!90° 3 90°1270° X X 0°/i 80° 4 X 90°/270° 180°/0° X Each element (i,j) of this geometry array, d/d, corresponds to the antenna direction of the node i d and the antenna direction of the node j d to generate the communication path between the node i and the nodej.

Figure 4 illustrates the antenna discovery protocol process performed between a device and its neighbouring device(s). The objective of the antenna protocol discovery is to allow a device, referred to as the device A, to find its neighbour devices (if any) and the corresponding communication path(s) between the device A and its identified neighbouring devices. Such a process is defined as the antenna discovery protocol of the device A. Figure 4a introduces the antenna discovery process performed by the device A and figure 4b introduces the corresponding antenna process performed by the neighbouring devices of the device A. During this process, two types of messages are exchanged between the device A and its potential neighbours: probe and feedback messages. These two types of messages are illustrated with reference to Figure 5: figure 5a for probe message and Sb for feedback message.

First, the antenna discovery protocol process starts with step 405 which sets the antenna mode of the deviceAto EmissionlDirectional.

Next, in step 410, the device A selects a set of transmitter antenna parameters.

Let M be the size of the set. The set of transmitter antenna parameters considered in this process is the same as the transmitter antenna parameters used to define communication paths in the geometry array (described in figure 3) which characterizes a geometry. Consequently, in the rest of document, the considered transmitter antenna parameters are the antenna directions and transmit power but some others may be used.

According to a first embodiment, the selection is exhaustive by choosing a step between 0° and 360° and stepping through all directions according to this step size. For instance, if this step is 5°, the selected antenna directions are 0°, 50 100... 355° 3600. According to a second preferred embodiment, this selection is based only on the antenna directions contained in the geometry arrays stored in the ROM 1403 corresponding to the list of predetermined geometries. It means that the selection selects only antenna directions contained in the fields 310 of these geometry arrays. Although fewer antenna directions are tested, there is no impact on the results of the antenna discovery protocol (identification of the same neighbour devices and same communication paths because only the relevant (which allows to discover a potential communication path) ones are tested.

In the same way, the selection process may also be performed with the transmit power. According to a first preferred embodiment, the transmit power is based on a nominal predetermined value. According to a second preferred embodiment, this selection is based only on the transmit power values contained in the geometry arrays stored in the ROM 1403 relative the list of predetermined geometries. In the same way, the selection process may also be based on both antenna direction and transmit power.

Next, step 415 generates and sends M probe messages towards the M selected transmitter antenna parameters.

The probe message is illustrated in figure 5a. A probe message 500, sent by a device referred to as a source device, contains 2 parts, a data part 510 and a cyclic redundancy check (CRC) part 540 used to check the validity of the data part 510. The part 510 contains 2 fields. The field 515 corresponds to the identifier of the source device. The identifier of a device may be for example, the MAC address of the source device or a function of its MAC address. The field 520 corresponds to the transmitter antenna parameters used by the source device to send it. Consequently, it may include two fields, one dedicated to the antenna direction 525 (expressed in degree) and one dedicated to the transmit power 530 (expressed in Watt).

In this way, step 415 generates M probe messages. The values of the Km probe message are the following. The field 515 is equal to the MAC address of the device A and the field 520 is equal to the Kth selected transmitter antenna parameters at step 410. Next, the CRC part 540 is computed from the generated part 510. Once generated, step 420 sends the Ktt' probe message according to the Kth transmitter antenna parameters and performs the same operations for the (K+1)Th probe message up to the Mth message.

Step 420 sets the antenna mode to reception/omnidirectional in order to be ready to receive potential feedback messages from its neighbours.

At the beginning of the antenna discovery process, the other devices of the system which are potential neighbours of the device A set their antenna mode to Reception/Omnidirectional at step 450 in order to be ready to receive probe messages. When a device receives successfully (checking with the CRC) at least one piobe message 455, it means that the device is a neighbour of the device A. It is referred to as neighbouring device.

Step 460 selects the best probe message received by the neighbouring device among the M probe messages sent by the device A. This selection is based for instance on the received signal strength indication (RSSI) or the signal-to-noise ratio (SNR) measured by the neighbour device. The probe message received with the best RSSI (or SNR) is the selected one.

Next, step 465 sets the antenna mode of the neighbouring device to Emission/Directional.

Next, in step 470, the neighbouring device selects a set of transmitter antenna parameters. Let N the size of the set. This step is similar to step 410. Next, step 475 generates and sends N feedback messages towards the N selected transmitter antenna parameters at step 470.

A feedback message 550, sent by a device referred to as the destination device, is generated from a probe message sent by a device, referred to as the source device. It contains 2 parts, a data part 560 and a cyclic redundancy check pad 590 used to check the validity of the data part 560. The part 560 contains 4 fields.

The field 565 corresponds to the identifier of the source device. The field 570 corresponds to the transmitter antenna parameters used by the source device.

The field 575 corresponds to the identifier of the destination device. The field 580 corresponds to the transmitter antenna parameters used by the destination device.

In this way, step 475 generates N feedback messages (N is equal to M in basic case). The values of the Km message are the following. The field 565 and the field 570 are equal to the field 515 and 525 of the best received probe message selected in step 460. The field 575 is equal to the MAC address of the neighbouring device and the field 580 is equal to the Kth selected transmitter antenna parameters at step 470. Next, the CRC part 590 is computed from the generated part 560. Once generated, step 475 sends the Kth feedback message according to the Kth transmitter antenna parameters and perform the same operations for the (K+1)th feedback message up to the Nth feedback message.

Next, step 480 sets the antenna mode to reception/omnidirectional in order to be ready for the next antenna discovery process.

Next, when the device A receives successfully (checking with the CRC) at least one feedback message from a neighbouring device (425), it means that the device A has identified a neighbouring device and consequently a communication path between itself and this identified neighbouring device. This identification step 430 is based on the fields 565,570,575 and 580 of the received feedback message and the identified communication path is stored in the RAM memory 1402.

Figure 6 is a flowchart showing the discovery process of a multi-projection system. The objective is to identify a geometry according to a list of predetermined geometries. The storage of such geometries is described in figure 3. At the beginning of the process, a current list is set to this list of predetermined geometries. It represents the "compatible" (as explained in figure 12) predetermined geometries of the multi-projection system during the discovery process. At the end of the discovery process, this current list is equal to one element, the identified geometry of the multi-projection system.

The first step 605 consists in powering on all the devices of the multi-projection system. Next, in step 607, a first device, referred to as a start device, is selected.

This selection can be performed by the user by pressing down a button for instance.

Next, step 610 performs the antenna discovery protocol of the start device in order to find the neighboring device/s of the start device. This protocol is described with reference to figure 4. At the end of this step, a first list of communication paths is discovered.

Next, step 615 selects a device, referred to as next discovery device, among the neighbouring devices found in the previous performed antenna discovery protocols. Moreover, this next discovery device is a new device, meaning that each next discovery device is different at each iteration of the loop (steps 630- 615) of the discovery process. This selection is performed randomly among the neighbouring devices.

Next, step 620 requests the next discovery device to perform an antenna discovery protocol in order to find its neighboring device/s. At the end of this step, a new list of communication paths is discovered. Next, step 625 updates the current list of predetermined geometries by eliminating the predetermined geometries which are not compatible with the results of the previously performed antenna discovery protocols. Each predetermined geometry of the current list is tested successively. The step of updating the current list of predetermined geometries can optionally be performed a first time just after step 610 (referred as to step 612) after the first performed antenna discovery protocol. In such a case, the convergence of the algorithm can be accelerated (to the potential detriment of CPU load).

Next, step 630 checks whether the current list of geometries contains only a single element or not. If so, go to step 630. If not, return to step 615. Step 635 means the end of the algorithm, the remaining geometry of the list corresponds to the geometry of the multi-projection system.

Two particular implementations will be considered to perform this process (and in particular step 620). a centralized implementation and a distributed implementation.

These two implementations are described in the figures 7-9. They are described both from sender and receiver point of view. From sender side, the two versions are described in figure 7. From receiver side, the centralized version is described in figure 8 and the distributed version is described in figure 9. The choice of the version is fixed by the system.

In the centralized version, the start device remains the same device and the update of the current list of geometries is performed at that start device.

Consequently, it needs to acquire the results of the performed discovery protocols by the other devices. In the distributed version, the updating is performed at a device which changes at each iteration of the algorithm and consequently, the current status of the algorithm based on the previous performed discovery protocol, must be transferred from the previous start device to the new current start device.

Figure 7 is a flowchart showing the emission process of the discovery protocol of a multi-projection system.

At the start of the system, once all the devices are considered as powered on (step 605), each device checks whether it is the start device of the system or not.

If yes, go to step 730. If not, go to step 720. Step 720 sets its antenna mode to reception/omnidirectional in order to be ready to receive messages. Step 730 performs the discovery protocol of the start device as explained in figure 4. The exchange of discovery messages (between a device and its neighbours) described in figure 4 and 5 are not described in this figure.

Next, step 740 selects a next discovery device among the neighbour devices found in the previous performed antenna discovery protocols (this step corresponds to step 615). Next, step 750 generates and sends a discovery propagation message. This message differs according to the actual embodiment.

As for the distributed version, the discovery propagation message is described in figure ba. As for the centralized version, it is described in figure lOb.

As regards a distributed version, a discovery propagation message 1000 is generated and sent by a source device to a destination device. It contains the outputs of the previous performed discovery protocols. It contains 2 parts, a data part 1005 and a cyclic redundancy check part 1010 used to check the validity of the data part 1005. The data part 1005 contains 5 fields. The field 1015 corresponds to the identifier of the source device. The field 1020 corresponds to the identifier of the destination node. The field 1025 contains a list of communication paths. Each communication path 1030 is described as explained in the figure 3 with 4 parameters: a source device identifier 1035 and the transmitter antenna parameters 1040 of this source device, a destination device identifier 1045 and the transmitter antenna parameters 1050 of this destination device. The field 1060 contains the list of devices. Each device field 1065 is identified by 2 fields. The device identifier 1070 corresponds as explained before the MAC address of the source device or a function of its MAC address. The geometry device Id 1075 corresponds to the identifier device used in the stored geometries of the figure 3 corresponding to the device identifier 1070 (the link between the two identifiers is performed by the process explained in figure 12).

The last field 1077 represents the current list of predetermined geometries. It indicates for each predetermined geometry whether or not it is eliminated from the current list. For instance, one possible implementation is described. The size in bits of the field 1077 is 1, equal to the number of predetermined geometries, each bit of the field corresponding to a predetermined geometry. A bit is equal to I if its corresponding geometry was not eliminated by the step 625 during an iteration of the algorithm and 0 otherwise. The fields 1060 and 1077 are optional because it is possible to recover them from the field 1025. But these fields allow the step of updating 625 to be made faster.

Consequently, as regards a distributed version, step 750 generates the following discovery propagation message: the field 1015 corresponds to the identifier device of the start device, the field 1020 corresponds to the identifier device of the next discovery device selected at the step 740. The field 1025 contains the list of the discovered communication paths 1030 during the previous performed discovery protocols (610;620). The field 1060 contains also the list of discovered devices 1060 during the previous performed discovery protocols (610;620) and identified during the previous performed updating of the current list of geometries (step 625). Consequently, a device may be discovered by the discovery protocol but not identified yet. The field 1077 corresponds to the current list of geometries (updated at each iteration of the step 625).

As regards a centralized version, a discovery propagation message 1080 contains 2 fields: a data part 1085 and a cyclic redundancy check part 1086 used to check the validity of the data part 1085. The data part 1085 contains the path from a source device to destination device. The path is described by an ordered list of L fields 1090, L being the length of the path, wherein each field 1090 contains 2 fields 1094 and 1098. The ith field 1094 of the field 1090 corresponds to the ith device of the path and the ith field 1098 of the field 1090 corresponds to the transmitter antenna parameters to be applied by the in device in order to achieve the (i+1)th device of the path. An alternative of the field 1098 is that each device of the system stores in their memory the lists of their neighbours and the corresponding transmitter antenna parameters determined after the performing of the antenna discovery protocol described in figure 4.

Consequently, as regards a centralized version, step 750 generates the following discovery propagation message: the field 1085 corresponds to the path between the start device and the next discovery device. Each field 1090 is deduced from the list of discovered communication paths stored in the start device. Moreover, step 750 sends the generated discovery propagation message according to the antenna transmission parameters allowing to achieve the next device of the path between itself and the next discovery device. Next, step 760 sets the antenna mode of the start device to Reception/Omnidirectional.

Figure 8 is a flowchart showing the reception process of the discovery protocol of a multi-projection system in centralized version; Let Nd be the device performing this process.

At the reception of a message 800, step 805 checks if the message is a discovery propagation message or a discovery feedback message (the messages comprising the antenna discovery protocol between one device and its neighbours are not taken into account). If the message is a discovery propagation message, next step is 810. If not, next step is 860.

Step 810 analyses the field 1085 of the received discovery propagation message 1080 and checks whether the device Nd is in the discovery path. It means that it checks if its identifier is in one of the field 1094. If yes, let i be its position in the path (the ith element 1094) and the next step is 830. Otherwise go to step 815, end of the reception process.

Step 830 checks if the device Nd is the next discovery device corresponding to the last element 1094 of the field 1085. If yes, step 835, the discovery message has arrived at its destination which is the device Nd. If not, go to step 825. Step 825 relays the message to the next device of the discovery path according to the antenna transmission parameters contained in the i' field 1098 of the field 1085.

The next device corresponds to the (i+lf' device of the path.

Step 835 performs the antenna discovery protocol of the next discovery device as explained in figure 4. The exchange of antenna discovery messages (between a device and its neighbours) is described in figure 4 and 5 are not described in this figure. The output is a list of discovered communication paths.

Step 845 sets the antenna mode of the start device to Emission/Directional. Step 850 generates and sends a discovery feedback message which is described in figure 11.

With reference to Figure 11, a discovery propagation message 1100 is generated and sent by a source device and is directed to a destination device. It contains 2 parts, a data part 1105 and a cyclic redundancy check part 1110 used to check the validity of the data part 1010. The data part 1105 contains 2 fields. The data part 1115 contains the path from a source device to destination device. The path is described by an ordered list of L fields 1116, L being the length of the path, wherein each field 1116 contains 2 fields 1117 and 1118. The field 1117 of the field 1116 corresponds to the ith device of the path and the ith field 1118 of the field 1116 corresponds to the transmitter antenna parameters to be applied by the device in order to achieve (i+1)th device. An alternative of the field 1118 is that each device of the system stores in their memory the lists of their neighbors and the corresponding transmitter antenna parameters determined after the performing of the antenna discovery protocol described in figure 4. The data part 1120 contains a list of communication paths. Each communication path 1130 is described as explained in the figure 3 with 4 parameters: a source device identifier 1135 and the transmitter antenna parameters of this source device 1140, a destination device identifier 1145 and the transmitter antenna parameters of this destination device 1150.

Consequently, step 850 generates the following discovery feedback message: the field 1115 corresponds to the path between the next discovery device (the source device) and the start device (destination device). The fields 1120 correspond to the communication paths discovered by the next discovery device at step 835. Step 850 sends it according to the antenna transmission parameters allowing it to reach the next device of the path between the next discovery device and the start device (destination device).

Step 855 sets the antenna mode of the start device to Reception/Omnidirectional.

At the reception of a discovery feedback message, step 860 checks if the device Nd is in the discovery path. It means that it checks if its identifier is in one of the field 1115. If yes, let i be its position in the path (the th elements 1116) and the next step is 870. Otherwise go to step 865, end of the reception process. Step 870 checks if the device Nd is the start device corresponding to the last element 1116 of the field 1115. If yes, step 880, the discovery feedback message has arrived at its destination which is the device Nd. If not, go to step 875.

Step 875 relays the message to the next device of the discovery path according to the antenna transmission parameters contained in the ith field 1116 of the field 1115. The next device corresponds to the (i+l)th device of the path. Step 880 updates the current list of predetermined geometries from the communication paths discovered by the next discovery device. These ones are contained in the field 1120 of the discovery Feedback Message. This step corresponds to the step 625.

Next, step 885 checks whether the current list of predetermined geometries contains one element or not (step 630). If so, go to step 895. If not, go to step 890. Step 895 identifies the geometry; the remaining geometry of the list corresponds to the geometry of the multi-projection system (step 635).

Step 890 starts the discovery emission process described in figure 7 again.

Figure 9 is a flowchart showing the reception process of the discovery protocol of a multi-projection system in distributed version; Let Na be the device performing this process.

At the reception of a discovery message 900 (a discovery propagation message because there is no discovery feedback message in distributed version), step 905 is performed. Step 905 checks if the device Na is the next discovery device.

If yes, go to step 915. If not, go to step 910, end of the algorithm. This step consists in checking if the field 1020 of the received message is equal to the device identifier of the device Na. If yes, the test is positive. If not, the test is negative.

Step 915 updates the start device. The device Na becomes the new start device.

Next, step 920 performs the discovery protocol of the device Na (the new start node) as explained in figure 4. The exchange discovery messages (between a device and its neighbours) is described in figure 4 and 5 are not described in this figure. The output is a list of discovered communication paths.

Next, step 925 updates the current list of predetermined geometries from the communication paths discovered by the next discovery device. The current list of predetermined geometries is contained in the field 1025 of the received message. Moreover, this update (figure 12) may use the previous discovered paths which are contained in field 1025 of the received message and the list of identified devices which are contained in field 1060. Once retrieved this information, this step corresponds to the step 625. Step 930 checks whether the current list of geometries contains one element or not (step 630). If so, go to step 935. If not, go to step 940. Step 935 starts the discovery emission process, described in figure 7, again. Step 940 identifies the geometry; the remaining geometry of the list corresponds to the geometry of the multi-projection system (step 635).

Figure 12 is the flowchart showing the checking process of a predetermined geometry in order to determine whether or not it is compatible with the results of the previously performed antenna discovery protocols. These results correspond to a list of newly discovered communication paths having in common the same source device (start device or next discovery device).

First, step 1210 retrieves the geometry array (from the ROM 1403) corresponding to the tested predetermined geometry. All the lines of the geometry array are considered as not tested.

Step 1215 set a variable corresponding to the number of discovered paths, referred to as "Nb_discovered_paths", to zero. Step 1220 selects a line of the geometry array considered not tested (by this process) and not identified (the source device of the new discovered communicated path does not correspond to an identified line). The selected line is considered as tested.

Step 1225 checks whether the selected line is compatible with the new discovered paths. The check is positive if each new discovered communication path corresponds to an element of the selected line (same values of all the transmitter antenna parameters). In such a case, the next step is 1230, otherwise next step is 1235. Step 1230 increments the variable Nb_discovered_paths and stores the identified line. The next step is 1235.

Step 1235 checks whether an untested and unidentified line is available. If yes, return to step 1220. If not, go to step 1240. Step 1240 checks if the variable Nb_discovered_paths is equal to zero. If yes, go to step 1245. Else go to step 1250.

Step 1245 determines the tested geometry as not compatible and consequently the geometry is eliminated from the current list of predetermined geometries.

Step 1250 determines the tested geometry as compatible. The next steps are used to identify the devices in order to speed up the convergence of the algorithm.

Step 1255 checks if the variable Nb_discovered_paths is equal to one. If yes, go to step 1265. If not go to step 1260. Step 1260 stores the discovered paths. The device identification is not possible because there are several lines which may correspond to the new discovered paths. Step 1265 identifies the device corresponding to the line identified at step 1230. The line identifier, which is the geometry device identifier, corresponds to the identifier of the source device which is contained in the discovery messages. With these new identified lines, some new identified lines may be found.

Consequently, step 1270 retrieves the discovered path stored at step 1260 and performs steps 1225, 1230, 1235 and 1255 again, in order to check if the Nb_discovered_paths variable is equal to 1 or more. If it is equal to 1, the line is considered identified.

In this way, step 1275 identifies the devices corresponding to the line identified at step 1270.

Figure 13 is an illustration of the invention in the centralized and distributed version. In this example, the list of the 7 considered predetermined geometries is the following one: A 1X2 array of 2 video projectors, A 2X2 array of 4 video projectors, A 2X3 array of 6 video projectors, A 3X2 array of 6 video projectors, A 4X3 array of 12 video projectors, A 3X4 array of 12 video projectors, A 4X4 array of 16 video projectors, The current list of predetermined geometries is set to this list.

Once powered on all the devices (1305 up to 1380) of the multi-projection system 1300, the user selects the device 1350 as the start device. The start device 1350 performs a first antenna discovery protocol in order to find the neighbour device/s of the start device and the corresponding paths between the start device and its neighbours. This is performed by sending 4 probe messages towards the following directions Q°, 9Q0 180° and 360° (at a given power), these directions are contained in the list of the predetermined geometries. The devices 1355, 1330, 1345 and 1370 respond by a feedback message. Amongst these 4 found devices, the start device selects the device 1330 (random decision), referred as the next discovery device.

In the centralized version, the start device 1350 sends a discovery message 1080 with a field 1085 corresponding to the discovery path equal to 1350-1 330 toward 1330 (the direction 90° is known via the previous performed discovery protocol). At the reception of this discovery message, the device 1330 performs an antenna discovery protocol (with the same value as the first discovery protocol) in order to find its neighbour device/s and the corresponding paths between the next discovery next device and its neighbours. It discovers the devices 1335, 1310 and 1325. It generates and sends a discovery feedback message 1100 with a field 1115 corresponding to the discovery path 1330-1350 and a field 1120 corresponding to the discovered path discovered by the next discovery device 1330. The start device updates the current list of geometries by eliminating the geometries which are not compatible with the results of the previously performed discovery protocols. Thus it eliminates the 2X2 array, the 1X2 array, the 2X3 array, the 3X2 array, the 3X4 array. The current list contains now 2 predetermined list: the 4X3 array and the 4X4 array. The algorithm continues. The start device 1350 now selects the device 1310. Its sends a discovery message 1080 with a field 1085 corresponding to the discovery path equal to 1350-1330-1310 toward the direction 900. The device 1330 relays the message toward the direction 900. The device 1310 receives the discovery message, performs a discovery protocol, discovers the device 1305 and 1315 and the corresponding communication path. It sends a feedback discovery message with this information to 1350. This new information doesn't allow the actual geometry to be selected. Consequently, the start device selects a new device 1315 as next discovery device. This one discovers the device 1335 and 1320 and the corresponding communication paths. Once this information is received, the start device eliminates the geometry 4X3 and deduces that the geometry of the system is the 4X4 array.

In the distributed version, the start device 1350 sends a discovery message 1000 towards 1330 (the direction is known 90° via the previous performed discovery protocols). The field 1015 is equal to 1350. The field 1020 is equal to 1330. The field 1025 corresponds to the 4 discovered communication paths (1 350,0°,1 355,180°), (1350,90°, 1330,270°), (1350,1 80°,1 345,0°), and (1350,270°,1370,90°). The field 1060 is equal to 1355, 1330, 1345 and 1370.

The field 1077 is equal to a mask wherein all its bits (7 bits corresponding to the 7 predetermined geometries) have a value equal to 1 (no eliminated geometry).

The device 1330 becomes the start device. Next, it performs an antenna discovery protocol (with the same value as the first discovery protocol) in order to find its neighbor device/s and the corresponding paths between the start device and its neighbors. It discovers the devices 1335, 1310 and 1325. The start device updates the current list of geometries by eliminating the geometries which are not compatible with the results of the previously performed discovery protocols. So it eliminates the 2X2 array, the 1X2 array, the 2X3 array, the 3X2 array, the 3X4 array. The current list contains now 2 predetermined list: the 4X3 array and the 4X4 array. The algorithm continues. The start device 1330 sends a discovery message 1000 towards 1310 (the direction is known to be 90° via the previous performed discovery protocols). The field 1015 is equal to 1330. The field 1020 is equal to 1310. The field 1025 corresponds to 7 paths (1350,0°,1355,180°), (1350,900,1330,2700), (1350,1800,1345,00), (1350,2700,1370,900), (1 330,0°,1 335,180°), (1330,90°, 1310,270°), and (1330,1 80°,1 325,0°). The field 1060 is equal to 1355, 1330, 1345, 1370, 1335, 1310, 1325. The field 1077 is equal to a mask containing only 0000011 as value (the two values 1 corresponds to the 4X3 array and the 4X4 array). The device 1310 becomes the start device.

Next, it performs an antenna discovery protocol (with the same value as the first discovery protocol) in order to find its neighbor device/s and the corresponding paths between the start device and its neighbors. It discovers the devices 1305 and 1315. This new information is not sufficient to select the correct geometry.

Consequently, the start device selects a new device 1315. To do this, it sends a discovery message 1000 towards 1315 (the direction is known to be 0° via the previous performed discovery protocols). The field 1015 is equal to 1310. The field 1020 is equal to 1315. The field 1025 corresponds to the 9 discovered paths (1 350,0°,1 355,180°), (1 350,90°,1 330,270°), (1350,1 80°,1 345,0°), (1350,270°, 1370,90°), (1 330,0°,1 335,180°), (1330,90°, 1310,270°), (1330,180°,1 325,0°), (1310,0°,1315,180°) and (1310,180°,1305,0°). The field 1060 is equal to 1355, 1330, 1345, 1370, 1335, 1310, 1325, 1305 and 1315. The field 1025 is equal to a mask containing only 0000011 as value (the two values 1 corresponds to the 4X3 array and the 4X4 array). The device 1315 becomes the start device. Next, it performs an antenna discovery protocol (with the same value as the first discovery piotocol) in order to find its neighbor device/s and the corresponding paths between the start device and its neighbors. It discovers the devices 1335 and 1320. With this information, 1315 eliminates the geometry 4X3 and deduces that the geometry of the system is the 4X4 array.

Figure 14 is a schematic block diagram of a computing device 1400 for implementation of one or more embodiments of the invention. The computing device 1400 may be embodied in a video projector as the device 111, 112, 113 and 114 but also a device such as a micro-computer, a workstation or a light portable device. The computing device 1400 comprises a communication bus connected to: -a central processing unit 1401, such as a microprocessor, denoted CPU; -a random access memory 1402, denoted RAM, for storing the executable code of the method of embodiments of the invention as well as the registers adapted to record variables and parameters necessary such as the current list of predetermined geometries, the current discovered communication paths; the memory capacity thereof can be expanded by an optional RAM connected to an expansion port for example; -a read only memory 1403, denoted ROM, for storing computer programs for implementing embodiments of the invention such as the arrays of the predetermined geometries; -a network interface 1404 is typically connected to a communication network over which digital data to be processed are transmitted or received. The network interface 1404 can be a single network interface, or composed of a set of different network interfaces (for instance wired and wireless interfaces, or different kinds of wired or wireless interfaces). The wireless interface integrates the antenna 210a and 210b described in figure 2. Data packets are written to the network interface for transmission or are read from the network interface for reception under the control of the software application running in the CPU 901; -a user interface 1405 for receiving inputs from a user or to display information to a user; -a hard disk 1406 denoted HD -an I/O module 1407 for receiving/sending data from/to external devices such as a video source or display The executable code may be stored either in read only memory 1403, on the hard disk 1406 or on a removable digital medium such as for example a disk.

According to a variant, the executable code of the programs can be received by means of a communication network, via the network interface 1404, in order to be stored in one of the storage means of the communication device 1400, such as the hard disk 1406, before being executed.

The central processing unit 1401 is adapted to control and direct the execution of the instructions or portions of software code of the program or programs according to embodiments of the invention, which instructions are stored in one of the aforementioned storage means. After powering on, the CPU 1401 is capable of executing instructions from main RAM memory 1402 relating to a software application after those instructions have been loaded from the program ROM 1403 or the hard-disc (HD) 1406 for example.

Any step of the algorithm shown in Figure 4, 6, 7, 9 or 12 may be implemented in software by execution of a set of instructions or program by a programmable computing machine, such as a PC ("Personal Computer"), a DSP ("Digital Signal Processor") or a microcontroller; or else implemented in hardware by a machine or a dedicated component, such as an FPGA ("Field-Programmable Gate Array") or an ASIC ("Application-Specific Integrated Circuit") It will be understood that the present invention has been described above purely by way of example, and modification of detail can be made within the scope of the invention.

Each feature disclosed in the description, and (where appropriate) the claims and drawings may be provided independently or in any appropriate combination.

Claims (18)

1. CLAIMS1. A method of determining the geometry of an array of wireless communication devices, each device including at least one antenna, from a set of possible geometries, the method comprising: performing, at a first device, an antenna discovery protocol to detect devices neighbouring said first device; selecting a further device from detected devices neighbouring the first device; performing, at said further device an antenna discovery protocol to detect devices neighbouring said next device; updating the list of possible geometries by eliminating geometries not compatible with results of previously performed antenna discovery protocols.
2. 2. A method according to Claim 1, comprising iteratively selecting one or more further devices from neighbouring devices detected in a previously performed antenna discovery protocol, and performing an antenna discovery protocol at said one or more further devices, and updating the list of possible geometries until a single geometry is determined.
3. 3. A method according to any preceding claim wherein updating the set of possible geometries is performed at the first device.
4. 4. A method according to claim 3, wherein updating the set of possible geometries is performed based on results of an antenna discovery performed at at least one other device.
5. 5. A method according to any one of Claims 1 or 2, wherein updating the set of possible geometries is performed at one or more of said further devices.
6. 6. A method according to Claim 5, further comprising obtaining, by said one or more further devices, the set of possible geometries from a device which has previously updated the list.
7. 7. A method according to any preceding claim, wherein the result of an antenna discovery protocol comprises a set of discovered communication paths between neighbouring devices.
8. 8. A method according to any preceding claim, wherein each of said possible geometries is defined by information representing communication paths between devices.
9. 9. A method according to Claim 8, wherein said information comprises, for each path, a source device, a destination device, and an antenna angle corresponding to the direction of the communication path at each device.
10. 1O.A method according to Claim 9, wherein said information further comprises, for each path, a power corresponding to the communication path.
11. 11.A method according to any preceding claim, wherein the antenna discovery protocol comprises emitting one or more probe messages and receiving one or more feedback messages from neighbouring devices, if any.
12. 12.A method according to Claim 11, wherein the antenna discovery protocol comprises emitting probe messages according to a predetermined set of antenna directions.
13. 13.A method according to Claim 12, wherein said predetermined set of antenna directions correspond to antenna angles of communication paths included in said set of possible geometries.
14. 14.A method according to any preceding claim, further comprising updating said set of possible geometries after performing antenna discovery at the first device and before performing antenna discovery at a further device.
15. 15.A method according to any preceding claim, wherein said wireless communication devices comprise video projectors including at least one wireless communication module.
16. 16.A method according to Claim 15, wherein said array of video projectors are networked to form a multi-projection system.
17. 17.A computer readable program which, when executed by a computer causes that computer to perform the method of any preceding claim.
18. 18. Method or apparatus as hereinbefore described and with reference to the accompanying drawings.