WO2008081137A2

WO2008081137A2 - Communication switching method, access point, network controller and associated computer programs

Info

Publication number: WO2008081137A2
Application number: PCT/FR2007/052483
Authority: WO
Inventors: Benoît Fourestie
Original assignee: France Telecom
Priority date: 2006-12-22
Filing date: 2007-12-11
Publication date: 2008-07-10
Also published as: WO2008081137A3

Abstract

A communication switching method in a wireless network (1) comprising a network controller (8) and access points (3,4), wherein the controller is linked to the access points, comprising the following step: switching an established communication with a wireless terminal (2) through a current access point (3), by transferring, upon receipt of a command from the network controller (8), radio resources of a radio link descending from the current access point to the terminal first attributed to the current access point (4), for continuing communication through the new access point with the same radio resource.

Description

COMMUNICATION SWITCHING METHOD, ACCESS POINT, NETWORK CONTROLLER, AND COMPUTER PROGRAMS

ASSOCIATED

The present invention relates to the field of mobile telecommunications networks. These networks include several wireless access points often called "base stations", and hereinafter referred to as access points, each associated with a cell or radio coverage area. The present invention applies more particularly to the communication switching established with a wireless terminal and transmitted by such a network.

An access point implements communications transmitted by the network for a terminal located in its cell by means of radio resources selectively allocated to this access point. Indeed, the radio resources allocated to an access point A are different from the radio resources allocated to at least one access point adjacent to the access point A, that is to say an access point whose cells cut off or adjoin the cell associated with the access point A, in order to avoid the creation of interference between the radio signals received or transmitted by this neighboring access point and those received or transmitted by the access point A .

The radio resources allocated to an access point are those needed to establish the physical radio link between a terminal and an access point and to allow the transport of data on this link between the access point and the terminal. They may include, but are not limited to, frequency channels for uplink and downlink links, one of a frequency band, a periodic time interval, or a code.

The allocation of radio resources by access point is generally fixed in a network.

The overall radio resources of the network generally include in particular one or more frequency bands. These global resources are selectively distributed among the different access points. Their use can be distributed among access points over time, for example in time division multiple access (TDMA) systems. In this case, within a time cycle, a periodic time slot is devoted to each access point for a given frequency. The use of these global resources can be frequency-shared between the access points, as in frequency division multiple access (FDMA) systems: a sub-band or channel is assigned to each access point. The use of global resources may be code-distributed, as in code division multiple access (CDMA) systems, in which each access point is assigned code. spreading (and possibly scrambling code) that distinguish the signals concerning it by encoding them and then filtering them using codes.

In first-generation cellular networks, mobile terminals could not move from one cell to another cell while maintaining communication. At the boundary between the two cells, a communication established in the first cell between a terminal and the network was interrupted during the cell change.

The functionality of switching from one cell to another without switching off a communication, called "handover", appeared in second generation networks (2G) for real-time type data. In GSM networks for example, when a mobile terminal, in communication with the network via a current cell, approaches the border between the current cell and a new cell, the situation is signaled to the controller of network that reserves for the terminal some of the resources previously allocated by the network to the new cell. This handover procedure is based on the measurement by the terminal of pilot channels transmitted continuously by the access points and the provision of these measurements by the terminal to the network. When the pilot channel of the new access point is received for a sufficiently long time with a power greater than that of the pilot channel of the current access point increased by a margin, the decision of handover is taken by the network. Then, simultaneously, the radio link between the current access point and the terminal implementing the communication between the network and the terminal by means of radio resources allocated to the current access point, is cut off and a radio link between the new access point and the terminal, implementing the communication between the network and the terminal, is established by means of radio resources allocated to the new access point.

In third generation networks (3G), to ensure that the procedure is performed correctly before cutting a radio link, the technique of "soft handover" was introduced. In this technique, the terminal can be in simultaneous communication with several access points. At the boundary between two cells, a radio link remains established between the current access point and the terminal, implementing communication between the network and the terminal by means of radio resources previously allocated to the current access point and another A radio link is established between the new access point and the terminal, implementing the same communication by means of radio resources previously allocated to the new access point.

The link with the current access point is cut off, for example, when the corresponding radio reception level falls below a given threshold. This procedure reduces communication interruptions in 3G networks when changing access points.

EP 0 500 775 discloses a communication transfer technique in a CDMA network, according to which, a wireless terminal being in communication with the network via a current access point, this terminal sends to the network controller a message indicating a a new access point and a first spreading code used by the terminal to encode the data on the uplink, i.e. data from the terminal and to the network. The network controller transmits to the new access point this first spreading code characterizing the data from the terminal. The new access point is thus able to exploit the data transmitted by the terminal to the current access point. There is therefore a single radio uplink (to transmit data from the terminal), shared between the two access points.

In parallel with the operation by the terminal of the data coming from the current access point by means of a second spreading code indicated by the current access point in its establishment channel, the terminal operates in in addition to data from the new access point, using a third spreading code indicated by the new access point in its settlement channel. Thus during this communication transfer procedure, a downlink (to transmit data to the terminal) is established between the current access point and the terminal, and in parallel, a downlink is established between the new access point and the terminal, until the radio links to the current access point are cut by the network controller.

For non-real-time data traffic, 2G (GPRS) and 3G networks, as well as networks such as WLAN, do not allow handovers. In these networks, the terminal selects a new access point and then needs to identify with this new access point to gain access to the network. These procedures are called "selection / reselection procedures".

In addition, these handover or soft handover techniques have certain disadvantages. Indeed, signaling data is exchanged with the terminal as part of the access point change procedure. Radio resources are thus used to transmit signaling data and not the communication signals themselves, which results in a loss of spectral efficiency or a loss of quality (for example, the technique of "bit theft" in a GSM network). On the other hand, the reception level measurements made by the terminal in these procedures consume energy and thus contribute to emptying the terminal's battery.

In the soft handover procedure, the same useful information is transmitted twice, on the two downlinks established in parallel. For terminals using broadband services, this overconsumption of radio resources is particularly troublesome.

When the residence times of the terminals in the cells are of short duration (for example for fast-moving terminals or for small cells), the selection times of a new access point, of resynchronization, and the delays introduced for the triggering of handover or soft handover procedures further reduce the time of useful presence of a mobile in a cell, and the useful rates observed can be very low (we mean by useful time or flow useful the time or the rate used for the implementation of the communication of the useful data, that is to say data other than those relating to procedures of handover or soft handover). This is the case, for example, for GPRS or WLAN networks in which the procedures for selecting a new access point are particularly slow, and are not suitable for mobiles subject to rapid movement, for example in high speed (see the paper "Influence of the Velocity on the handover delay associated with a radio-signal-measurement-based handover decision", Emmelmann TKN Technical Report, Technical University Berlin, April 2005). These discontinuities of flow or quality can have particularly critical effects on high OSI layers. In particular, the Transmission Control Protocol (TCP) layer is based on an adaptation of the bit rate as a function of the performance of the radio link ("slow start" or "congestion avoidance" procedures). When the rates are sharply reduced, algorithms for decreasing the application rates are implemented, up to the momentary interruption of transmissions. The sudden decrease due to the progress of a handover can in this case have the consequence of interrupting the communication. A rate adaptation procedure is implemented which results in a zero flow initially, this flow being gradually increased to correspond to the rate found on the radio link ("slow start" procedure). This procedure leads to a sudden drop in the useful rate which takes a significant time to resume a normal value. The communication will need to be reset in the new cell.

Algorithms, such as "SACK" or "I-TCP", make it possible to reduce the number of untimely cuts, but can not provide a solution to variations in flow rates.

In existing WiFi type networks (802.11x type standards), a moving terminal must perform a disassociation procedure with the current access point before initiating an association procedure with the new access point. . These procedures have been established for data services and use the IAPP protocol ( "Inter Access Point Protocol") as recommended by IEEE I ¹ ( "Institute of Electrical and Electronics Engineers"). Handover deadlines are in practice too long to support real-time applications, for example of the VoIP type. Another problem is that the unassociated terminal does not know if resources are available in the new access point. If no resources are available, or if these resources are insufficient, the communication of the terminal with the network is degraded and can even be interrupted (for example for services with strong time constraints such as Voice over IP).

Moreover, resource planning in the prior art is fixed: in fact, the radio resources are allocated to the access points (codes or frequencies), and this allocation does not depend on the temporal variations of the traffic.

When a train passes for example, a large number of terminals occur in a cell crossed by the train. To ensure the flow of traffic, significant radio resources allocated to the cells along the train path would therefore be required. However, the resources allocated to a cell are used only when the train passes through it, and otherwise unused. Frequently, when changing access points by terminals on board a train, there are problems of overflow and saturation, while the resources are unused in the surrounding cells.

There is therefore a need for a technique for changing access points that does not have the drawbacks of the techniques of the prior art.

According to a first object, the invention proposes a method of switching communications in a wireless network comprising a network controller and access points, the controller being connected to the access points, comprising the following step:

- switching an established call with a wireless terminal via a current access point, transferring on command of the network controller, radio resources of a downlink from the current access point to the terminal beforehand assigned to the current access point, from the current access point to a new access point, for communication to continue through the new access point with the same radio resources.

Such a communication switching method makes it possible to continue, without interruption, the communication between the network and the terminal, and without the need to reset the connection to the network by the terminal.

The transfer of radio resources corresponds to a selective reallocation of these resources to the new access point.

Indeed, the new access point transmits communication data on a downlink established with the wireless terminal and which is similar, as regards the radio resources implemented, the downlink previously established with the wireless terminal. terminal by the current access point. During the communication switching operation, there is therefore only one downlink radio link established between the terminal and the network. Compared to the techniques of the prior art, the communication switching method according to the invention considerably reduces, or even eliminates, according to the embodiments, the consumption of radio resources necessary to implement the cell change, and therefore reduced. in the same way the decrease of the useful flow.

In addition, the problems encountered in cells located along the path of a train (ie non-use of resources in cells in which the train is not saturation in other cells) are also overcome by a following method the first object of the invention. Indeed, the radio resources move at the same time as the train.

In one embodiment, the reassigned radio resources correspond to at least one of a frequency band, a periodic time slot, and a spreading code. Indeed, according to the multiplexing modes used, each of these elements characterizes a radio resource used selectively by a wireless access point of a network among the neighboring wireless access points. In one embodiment, when a plurality of communications to a plurality of wireless terminals are implemented, in a first phase, through the current access point by means of a set of radio resources that are selectively allocated to it, the method reassigns said set of radio resources to the new access point, for the implementation of communications to said plurality of terminals, via the new access point. by means of said set of transferred radio resources. This arrangement is particularly advantageous when terminals move in groups, for example when terminals are on board the same train.

According to another embodiment, the communications intended for the terminals located in the train can be concentrated in a router and this router exchanges with the access points. In this case, the method according to the invention controls that the router communicates with the access point providing the best coverage. This access point can then have all available radio resources, since the cells of the previous and next access points are empty. In both of these modes, for example, the reallocation of radio resources is triggered if more than half of the plurality of wireless terminals satisfy a given condition. The condition to be verified is, for example, that the new access point is more suitable than the current access point for a terminal (for example greater proximity of the new access point to a terminal, signal of better quality etc.). This arrangement makes it possible to serve more than half of the terminals with the most suitable access point.

In one embodiment, an identification address of the current access point is further transferred to the new access point for identification of the new access point by said reassigned address. This feature allows the direct reorientation of the communication flows after switching, without the new access point having to be known to a network controller as the current access point to the network for the terminal. In one embodiment, the location of the wireless terminal is determined and the reassignment of the radio resources is performed according to at least said determined location. Decisions to transfer radio resources from one access point to another are thus made without any mobile intervention. The transmission capacity on the radio resources is therefore not used at all to transmit information relating to the cell change.

In one embodiment, the reallocation of the radio resources is performed according to a signal sent by the terminal to the network indicating a better level of reception by the terminal of a signal from the new access point relative to at a reception level of a signal from the current access point. In such a case, it is the terminal that signals the need to change access point. However, the information exchanged with the mobile for the change of access point is very small and very low consumption of transmission capacity.

In one embodiment, the method further transfers from the current access point to the new access point information specific to the communication with the terminal and previously stored by the current access point. This information relates, for example, to the authentication of the terminal, and / or to parts of the communication stream to be transferred to the terminal, and / or to encryption data, and / or to a type or state of the communication session and / or charging data. Thus, no reconfiguration is necessary during the change of cell whatever the type of communication engaged between the network and the terminal. The switching of the communications takes place in a completely transparent manner for the terminal, without the need for provision of elements by the terminal when entering the new cell. Communication continues to flow smoothly. According to a second object, the invention proposes a base station able to constitute a wireless access point to a radio network for at least one wireless terminal, said base station comprising:

means for implementing at least one communication to a wireless terminal by means of radio resources selectively allocated to said base station;

means for, on receiving a transfer control signal, supplying a second wireless access point to the network, data identifying said radio resources for the purpose of implementation by the second access point of at least one communication to the terminal by means of said radio resources.

According to a third object, the invention proposes a network controller adapted to control at least a first and a second wireless access point to a network, said controller comprising:

means for, when the first access point implements at least one communication to a wireless terminal by means of radio resources selectively allocated to said access point and following the selection of the second access point as a new network access point for said terminal, sending control signals to the first and second access points for transferring said radio resources from the first access point to the second access point.

According to a fourth object, the invention proposes a computer program to be installed in a base station able to constitute a wireless access point, for at least one wireless terminal, to a radio communication network, said program comprising instructions for implementing, during a program execution by means of processing the base station, the following steps, when at least one communication to a wireless terminal is implemented by the base station by means of radio resources selectively allocated to said base station:

receiving a transfer control signal from said radio resources;

after receiving said transfer control signal, providing a second wireless access point to the network, data identifying said radio resources for the implementation by the second access point of at least one communication to the terminal by means of said radio resources.

According to a fifth object, the invention proposes a computer program to be installed in a network controller adapted to control at least a first and a second wireless access point to a network, said program comprising instructions for implementing , during a execution of the program by processing means of said controller, the next step, when the first access point implements at least one communication to a wireless terminal from the network by means of radio resources selectively allocated to said point access: - following the selection of the second access point as a new network access point for said terminal, sending control signals to the first and second access points to transfer said radio resources from the first point of access access to the second access point. Other features and advantages of the invention will become apparent on reading the description which follows. This is purely illustrative and should be read in conjunction with the attached drawings in which:

- Figure 1 shows a system in a first embodiment of the invention. ; - Figure 2a to 2d shows a system in a second embodiment of the invention.

In Figure 1 is a system 10 in a first embodiment of the invention.

This system 10 comprises a plurality of mobile terminals of which only one, the terminal 2, is represented, a plurality of wireless access points to a network 1. Among these access points, only the access points 3 and 4 are shown in Figure 1.

The access point 3 is adapted to implement WiFi type radio communications between the network 1 and wireless terminals located within the radio coverage cell 5 which is associated with the access point 3. buffer 9 for temporarily storing the communication data to be transmitted to a terminal in the cell 5 when they can not be transmitted immediately. Similarly, the access point 4 is adapted to implement WiFi type radio communications between the network 1 and wireless terminals located within the radio coverage cell 6 which is associated with the access point 4. It comprises a buffer memory 11 intended to store temporarily the communication data to be transmitted to a terminal located in the cell 6 when they can not be transmitted immediately.

The network 1 further comprises a network controller 8 connected to the plurality of wireless access points of the network 1. Moreover, the wireless access points are interconnected by a connection 7, for example by cable, optical fiber or equivalent, and secure. In one embodiment, this link

7 is compliant with the IAPP protocol.

Consider that a communication is currently implemented by the access point 3 between the network 1 (generally, the communication is transmitted, for example, to a telephone connected to the network 1) and a terminal

2 wirelessly located in the cell 5, by means of a radio link comprising an uplink (to transmit the exchanges from the terminal and to the access point 3) and a downlink (to transmit the exchanges from from access point 3 to the terminal).

Communication data streams are therefore transmitted on these links.

The terminal 2 moves within the cell 5 towards the access point 4. When the access point 4 is selected as new network access point 1 for the terminal 2, the network controller 8 sends a command at the access point 3 requiring the transfer of the radio link from the access point 3 to the access point 4.

In response to this command, the access point 3 sends on the link 7 to the access point 4 the following data:

the identifier or identifiers of the access point 3 comprising: an identifier for identifying the access point by the network. For example, in this case, it is an identifier SSID type ("Service Set Identifier") that identifies both the access point and the terminal 2 by the network;

. the physical address MAC which uniquely identifies the access point 3, and which is present in the data flows exchanged between the access point 3 and the terminal 2 in the cell 5, and the IP configuration of the point d access 3 (ie its IP logical address, as well as its subnet mask and its gateway); in this case the IP and MAC addresses further identify the terminal 2;

the indication of the WiFi operation mode characterizing the overall frequency band and the bit rate (for example 802.11 b, 802.11 g, etc.); the technical characteristics defining the radio resources used by the access point 3, that is to say defining the upstream radio link and the downlink radio link dedicated to the communications between the access point 3 and the terminal 2: in particular the channels or frequencies of transmission of these links; the values of the transmission parameters such as, for example, that of the threshold RTS which determines the minimum packet size for sending a request to transmit in the form of an RTS signal ("Request To Send"), the binary value of the CTS parameter, that of the DTIM interval defining the wake-up interval in power saving mode, the value of the fragmentation threshold indicating the maximum size that a packet can reach before being fragmented, etc. ;

the data necessary for the encryption (for example of the WEP or WPA type) of the signals to be transmitted on the downlink and those necessary for the decryption of the signals coming from the terminal 2; the physical address MAC, the logical IP address and the authentication data of the terminal 2, as well as the status of the communication in progress with the terminal 2;

the content of the buffer 9 of the access point 3 relating to the communication in progress with the terminal 2. The access point 4 receives this data sent on the link 7.

According to a particular embodiment, the network controller 8 then sends a command or commands to the access points 3 and 4, controlling the following simultaneous operations:

stopping the implementation by the access point 3 of the communication with the terminal 2 via the upstream and downstream radio links identified in the data sent to the access point 4; and the implementation by the access point 4 of the communication with the terminal 2 via the upstream and downstream radio links defined by these data received by the access point 4.

The communication data streams (for example TCP or IP type) coming from the network and intended to be transmitted to the terminal 2 are automatically directed towards the access point 4, since it is now identified by the same identifiers as it was previously the access point 3.

In an embodiment where the identity of the access point has not been transferred at the same time as the radio resources, the controller also controls the reorientation to the new access point 4 of the data stream coming from the network and to communicate to the terminal 2, so that the communication is delivered by the new access point 4.

The access point 4 and the access point 3 each send a message to the controller 8 indicating that the transfer of resources went well (or if the transfer failed).

Thus the uplink and downlink radio links implemented by the access point 4 are the same as those previously implemented by the access point 3. They are no longer used by the access point 3. The radio resources previously assigned to the access point 3 are transferred to the access point 4 at the same time as the terminal 2 leaves the access point 3 for the access point 4.

No signaling exchange between the terminal 2 and the access points is therefore necessary to effect such a change of access point while maintaining the communication. The access point change is transparent to the user of the terminal 2.

The elements transmitted from a current access point to a newly selected access point have the function of allowing the new access point to take over the implementation of the communications with the terminal 2 in a manner that is transparent for this terminal. , without modification of the transmission parameters, nor of the characteristics of the radio links rising and falling etc. Depending on the technologies implemented, the types of elements transmitted from a current access point to a newly selected access point vary somewhat. For example, in some cases, the radio sub-band IDs are sufficient to identify the radio resources (FDMA multiplexing). In other cases, time slot identifiers, or spreading codes (CDMA), are required (TDMA multiplexing).

According to the CDMA technique, the transmitted digital symbols are multiplied by spreading codes. Orthogonal or quasi-orthogonal spreading codes are allocated to different physical channels sharing the same carrier frequency, in order to allow each receiver to detect the sequence of symbols that is intended for it, by multiplying the received signal by the spreading code. corresponding.

In a UMTS system implementing the CDMA technique, the spreading code used on a downlink is the product of a scrambling code ("scrambling code") assigned to the access point (also called "Node B"). ) transmitter and a channel code ("channelization code"). The data to be transmitted on the downlink are therefore spread by a channel code and coded by a scrambling code.

Scrambling codes are orthogonal from one access point to another. Channel codes are different within a cell. Channel codes within a cell are organized as a spreading tree.

In one embodiment of a method according to the invention, in the case of a UMTS wireless technology, in addition to the elements defining the terminal 2, the elements transmitted by the current access point to the new point of access include for example:

- name and identifier of Node B or current access point;

- MAC physical address and IP logical address of the current access point;

- the channel codes currently used by the cell and the terminal on the downlink and uplink; the authentication data of the terminal 2;

- the contents of the buffer memory;

the scrambling code (used by the current access point) used in the exchanges between the current access point and the terminal 2; the power transmitted on the downlink between the current access point and the terminal 2 and the status of the pilot channel, in particular the time interval between two beacon sends.

Thus, the new access point will use in the implementation of the communications with the terminal 2 after the transfer, the scrambling and channel codes used before the transfer by the current access point in its communications with the terminal 2. Thus, in particular, the spreading code used on the downlink by the current access point for communications to the terminal before the communication switch is identical to the spreading code used on the downlink by the new data point. access for communications to the terminal after communication switching.

The selection of a new access point can be made in different ways depending on the embodiments. In one embodiment, the location of the terminal is known to the network 1, for example the network controller 8, which then selects the best access point to the network 1 for the terminal 2 as a function of the location thereof. relative to the location of the access points.

The location of the terminal can be provided by GPS-based satellite positioning systems for example, or by sensors of various technology. Location can also be determined using triangulation location techniques, based on arrival time

(TOA) or received signal power. Or when a terminal is on board a train, the speed of the train can be measured by the train or predicted, and the location of the terminal is determined according to this speed.

In an embodiment where the position of the terminals is unknown to the network 1, the selection of a new access point is performed using a pilot channel access point on which the latter broadcasts a beacon signal. When a terminal determines that the level of reception of a signal on the pilot channel of an access point other than its current access point is higher than the level of reception of a signal on the pilot channel of its current access point, the terminal transmits a signal to the network 1, for example to the network controller 8. Upon receipt of such a signal, the controller 8 triggers the transfer of the radio resources assigned to the terminal from the current access point to the access point associated with the pilot channel received at the best level by the terminal. In the case where the position of the terminal is known to the network, the change of access point according to the invention will not involve any consumption of radio resources dedicated to this change. In the case where the signal must send a signal to indicate that a change is necessary, the volume of resources used is minimal and does not lead to a significant variation in useful bit rate.

In the example commented with reference to FIG. 1, the resources implemented for the terminal 2 have been transferred independently of the resources allocated to the implementation of communications between the network and other terminals present in the cell. another embodiment of the invention, the radio resources, which are in use by the current access point 3 to implement on the uplinks and downwards the communications of all the terminals of the cell 5 with the network, are transferred in block to the new access point 4, with the information of the buffers, the data relating to transmission, authentication, encryption, etc., as well as the data flows corresponding to these communications . This arrangement is particularly useful in the case where terminals move at a high speed and in a group, which results in non-stationary traffic in the cells covering the trajectory of the group of terminals. This arrangement is also useful when the network is not able to allocate resources per terminal to an access point, but only to access points.

In one embodiment, in such a case of access point change for a group of terminals, the resource transfer is performed if a certain percentage of terminals of the group verifies a given condition. For example, if a new access point selected is the best access point in terms of signal reception for more than half of the terminals in the group, the transfer is made. Figures 2a to 2d illustrate an embodiment of the invention implementing a transfer of all radio resources of an access point implemented for a group of terminals. The elements indicated by identical references in different figures are similar elements.

The system 100 shown in FIG. 1 represents a railroad track 15. A mobile communication network, not shown, is able to provide wireless communications on the path corresponding to this railroad track 15, via N wireless access points to the network bi, b ₂ , ..., b _N included in base stations and able to communicate by radio according to the WiFi standard. Only a few of the access points bi, b ₂ ,..., Bw are shown in FIGS. 2a-2d. The access points are able to communicate with each other via a link 16 in accordance with the AIPP protocol. Each wireless access point b, is associated with a respective radio cell Zj, i = 1 to N. In this case, the radio cells of the access points are of identical size and shape.

A train 17 moves on the track 15 in the direction indicated by the arrow F. The train 17 carries users of mobile terminals, distributed in the train 17. A transmitter 18 located at the front of the train transmits continuously to a network controller, not shown, the speed V of the train, which we assume here constant.

The length L of the train is known. It corresponds to the global radio coverage of M cells. Each cell associated with an access point therefore covers an area of the railroad of length equal to L / M.

A terminal therefore traverses a cell in a time Δt = L / (M * V). At a time t = 0 corresponding to FIG. 2a, the radio coverage of the train 17 is ensured by the access points b ₁ to b _M. Each access point b ,, i = 1 to M, then exchanges, by means of a set of respective radio resources, respective data streams within the cell associated with it, with the terminals of a group of terminals embedded in the train, which are located within this cell. The access points bj, for j = M + 1 to N are in standby. The radio resources used by an access point b ,, i = 1 to M, are different from those assigned to its immediately adjacent access points bj.-i (except for i = 0) and bj ₊ i (except for i = M), for i = 1 to M. A respective radio resource set comprises a uplink radio link and a downlink radio link for each terminal in the group for which communication is with the network via the access point b. ,.

Thus, for example with reference to FIG. 2a, for i = k, access to the network of the terminal group Gi is provided by the access point b _k by means of the set of radio resources R ₁ (link frequencies rising f11 M i ₁ and downlink frequencies f11 ', ..., M i ₁ '). The location of the group of terminals Gi is represented by lines at an angle. Access to the network of the group of terminals G ₂ is provided by the access point bk ₊₁ by means of the set of radio resources R2 (uplink frequencies f21, ..., f2i ₂ with f2i ≠ f1j for ie [1, i ₂ ] and I [1, H]; downlink frequencies f21 ', ..., f2i ₂ ' and f2i '≠ f1j' for ie [1, i ₂ ] and j ε [1, J ₁ ],). The location of the group of terminals G ₂ is represented by dots. The access point radio cell b _k is the cell Z _k and the radio cell of the access point bκ ₊₁ is the cell Zk ₊₁ .

FIG. 2b shows the configuration of the train 17 with respect to the track 15 and the access points at a time t = Δt / 2.

When the time reaches t = Δt, as represented in FIG. 2c, the train has traveled a distance equal to the crossing of a cell. The train 17 is now covered by the M cells Z ₂ to Z _{M +} i respectively associated with the access points b ₂ to b _{M + 1} . From t = 0 and up to this time t = Δt, the communications with the network were implemented for the terminals of the group of terminals G1 by the access point b _k and the communications with the network were implemented to the terminals of the group of terminals G2 by the access point bk ₊₁ .

As soon as the time t reaches the value Δt, the network controller controls the transfer of all the radio resources used, from t = 0 to Δt, by each access point bi, i = 1 to M, towards the point access bj ₊₁ . The data streams provided by the network, from t = 0 to Δt, to the access point bi, i = 1 to M, are further provided at the access point b, ₊₁ . At t = Δt, each terminal in the This changes the network access point, and the radio resources (including uplink and downlink frequencies) and transmission characteristics used to implement the communications between the terminal and the network are kept unchanged. The access point change is transparent to the user of the terminal.

Thus, when t exceeds Δt, the access point DM ₊ I becomes active and the access point bi is put on standby: this access point does not give access to the network to any terminal of the train 17. The points of access access b _{M + 2} to b _N remain in standby vis-à-vis the terminals present in the train 17. For example, the radio resources R ₁ used from t = 0 to Δt by the access point b _k are transferred to the access point b _{k +} i in the manner indicated, for example, above with reference to FIG. 1. All the elements necessary for this transfer are provided by the access point b _k to the access point bk ₊ i by the link 16. the network controller simultaneously controls the stop by the access point b _k of the transmission of the communications intended for the terminals of the group d and the use of the set of resources R ₁ of a on the other hand it controls the beginning of the transmission by the access point b _{k + 1} , by means of said set of radio resources R ₁ , communications intended for the terminals of group G ₁ . In addition, the data streams intended to be communicated to the terminals of the group G ₁ are now provided at the access point b _{k + 1} .

In such a case where all the radio resources allocated to an access point are transferred, the access point may have, for all terminals to which it gives access to the network at a given moment, a single identifier of SSID type, a single MAC address and a single IP address, these identifiers being transferred in block to the new access point and thus becoming the identifiers of this new access point.

In the same way, the radio resources R ₂ used from t = 0 to Δt by the access point b _{k + 1} are transferred at t Δt to the access point b _{k + 2} associated with the coverage area Z _{k + 2} . All the elements necessary for this transfer are provided by the access point b _k to the access point b _{k + 1} via the link 16. In addition, the data streams intended to be communicated to the terminals of the group G ₂ are now provided at the access point b _{k + 2} . The communications terminals G ₂ group with the network are implemented by the access point b _{k + 2} by means of radio resources R ₂ dèst = Δt.

When t reaches the value Δt, the network controller also simultaneously controls the stop by the access point b _{k +} i of the transmission of the communications intended for the terminals of the group G ₂ and the use of the set of resources R ₂ and the beginning of the transmission by the access point b _{k + 2} , by means of said set of radio resources R ₂ , communications intended for the terminals of the group G ₂ .

FIG. 2d shows the system 100 just after the transfer of the radio resources carried out at t = Δt.

A similar process is implemented after each new period Δt, the speed of the train being constant. When the speed of the train is not constant, the durations between the changes of access points are adapted according to the speed of the train 17. In such an embodiment of the invention, the radio resources are in fact affected. groups of terminals and follow them when moving. There is no longer any problem for the cells covering the trajectory of the saturation period train followed by a period of non-exploitation of the radio resources dedicated to them. There is no interference because the radio resources allocated to a first active access point are different from the radio resources allocated to a nearby active access point, ie an active access point whose cells cut or adjoin the cell associated with the first access point.

In one embodiment, communications to the terminals in the train may be concentrated in a router and this router will exchange with the access points. In this case, in one embodiment of the invention, the access point with which the router communicates is the access point providing the best radio coverage. This access point can then have all available radio resources, since the cells of the previous and next access points are empty.

When the transfer of the radio resources is implemented for a group of wireless terminals, the method also transmits to the new access point the state of the spreading tree indicating the spreading codes. currently used in downlinks by the current access point.

In one embodiment of the invention, the steps implemented by an access point and / or those implemented by a controller are consecutive to the execution of instructions of respective computer programs by means of processing of said access point and / or said controller.

Thus, the invention makes it possible to change cells in a wireless communications network by considerably reducing the transmission capacity and / or time consumption relative to the prior art generated by the change of cells. The variations of useful rate are thus canceled or negligible.

A terminal in communication with the network is assigned a downlink of given characteristics which is invariant during a change of access points. The corresponding radio resources are therefore no longer dedicated to an access point among a set of neighboring access points, but dedicated to a terminal in communication. They follow it during a radio communication initiated via an access point and continuing through a new access point. There is no longer need for signaling exchanges between the new access point and the previous access point, in particular to define the characteristics of a new downlink to be established between the new access point and the terminal.

By virtue of such a dynamic assignment of the radio resources to the access points, the invention also makes it possible to solve the problems of saturation and non-exploitation encountered consecutively on certain saturation access points of resources, particularly in the case of networks intended to provide communications to terminals moving in groups, for example on board trains.

Claims

A method of switching communications in a wireless network (1) comprising a network controller (8) and access points (3,4), the controller being connected to the access points, comprising the steps of: - Switching an established communication with a wireless terminal (2) via a current access point (3), transferring radio resources of a downlink radio link to the controller (8) of the network from the current access point to the terminal previously assigned to the current access point, from the current access point to a new access point (4), for the communication to proceed via the new access point with the same radio resources,

- further transfer an identification address of the current access point (3) to the new access point (4) for the identification of the new access point by said transferred address.

The method of claim 1, wherein the transferred radio resources comprise at least one of a frequency band, a periodic time slot, and a spreading code.

The method of claim 1 or claim 2, wherein a plurality of communications to a plurality of wireless terminals (Gi) are in progress through the same current access point (bk). by means of a set of radio resources (Ri) previously allocated, the method transfers said set of radio resources to the new access point (bk ₊ i), to continue the communications via the new access point to destination of said plurality of terminals, by means of said set of transferred radio resources.

The method of claim 3, wherein the transfer of radio resources (Ri) is performed if more than half of the plurality of wireless terminals (Gi) satisfy a given condition.

5. Method according to one of the preceding claims, wherein the location of the wireless terminal (2) is determined and the transfer of radio resources is performed according to at least said determined location.

6. Method according to one of the preceding claims, wherein the transfer of radio resources is performed according to a signal sent by the wireless terminal (2) to the network (1) indicating a better level of reception by the wireless terminal of a signal from the new access point (4) with respect to a reception level of a signal from the current access point (3)

A method according to any one of the preceding claims, wherein the method further transfers from the current access point (3) to the new access point (4) information specific to the communication established with the wireless terminal ( 2) previously stored by the current access point, said information being related to at least one of the authentication of the terminal, parts of the communication stream to be transferred to the wireless terminal, encryption data, a type or a state of the communication session and charging data.

A base station adapted to constitute a wireless access point (3) to a radio communication network for at least one wireless terminal (2), said base station comprising: means for implementing at least one communication intended for the wireless terminal by means of radio resources allocated to said base station;

means for, on receiving a transfer control signal, supplying a second wireless access point (4) to the network, data identifying said radio resources, for the purpose of implementation by the second point; to access at least one communication to the terminal by means of said radio resources.

9. Base station according to claim 8, further comprising means for stopping the implementation of at least one communication to the terminal by means of the radio resources in a manner almost simultaneously with the implementation by the second data point. access of at least one communication to the terminal by means of the radio resources identified by the data provided by the base station.

The base station (b _{k +} O according to claim 8 or claim 9, further comprising:

means for receiving from a third wireless access point to the network, data identifying other radio resources used previously for the implementation by the third access point of at least one communication intended for another wireless terminal; and

means for, on reception of a control signal, implementing at least one communication destined for the other terminal by means of the other radio resources identified by the received data, in a manner that is almost simultaneous with the stopping by the third access point of the implementation of at least one communication to the other terminal.

A network controller (8) adapted to drive at least a first (3) and a second (4) wireless access point to a network (1), said controller comprising:

means for, when the first access point (3) implements at least one communication from the network to a wireless terminal (2) by means of radio resources previously allocated to said access point and further selecting the second access point (4) as a new network access point for said wireless terminal, sending control signals to the first and second access points for transferring said radio resources from the first point of access; access to the second access point.

12. Computer program to be installed in a base station capable of constituting a wireless access point (3), for at least one wireless terminal (2), to a radio communication network (1), said program comprising instructions for implementing, during a program execution by the base station processing means, the following steps, when at least one communication to a wireless terminal from the network is implemented by the base station by means of radio resources allocated to said base station:

receiving a transfer control signal from said radio resources;

following receipt of said transfer control signal, supplying a second wireless access point (4) to the network, data identifying said radio resources for the implementation by the second access point of at least one communication to the terminal from the network by means of said radio resources.

A computer program to be installed in a network controller (8) adapted to drive at least a first (3) and a second (4) wireless access point to a network, said program including instructions for implementing during the execution of the program by means of processing said controller, the next step, when the first access point implements at least one communication to a wireless terminal (2) from the network by means of radio resources allocated to said access point: - continued selecting the second access point as a new network access point for said terminal, sending control signals to the first and second access points for transferring said radio resources from the first access point to the second access point; access.