FR2803151A1 - Wireless inter-communication method involves storing addresses of specific data packets transmitted between networks, in wireless access point - Google Patents

Abstract

Addresses including in specific data packets transmitted from wireless network to cabled network, are stored in wireless access point. Data packets are transmitted from cabled network to wireless network through the access point based on data packet addresses and addresses stored in access point. Independent claims are also included for the following: (a) Method of learning address of wireless communication terminal; (b) Datapackets transmission method; (c) Wireless access point; (d) Wireless telecommunication network base station; (e) Wireless terminal adapter; (f) Wireless mobile station; (g) Data packets filtering method; (h) Data packets filtering device; (i) Computer program for implementing address learning method; (j) Computer program product for implementing learning method

Description

The present invention relates to a method and an automatic address learning device in a telecommunications network, and to systems implementing them.

The invention belongs to the field of interconnection of wireless networks with wired networks.

More specifically, the invention is here described in application to the learning of the MAC <I> ("Medium </ I> Access ControP") address for a wireless access point and a wireless terminal adapter.

The European standard EN 301 649, 1999, DECT <I> ("Digital </ I> Enhanced Cordless Telecommunications") describes the use of DECT radio technology for interconnection with equipment or networks Ethernet type (IEEE 802.3 standard).

Document US Pat. No. 5,339,316 discloses a wireless local area network system which comprises - a wired local area network and - at least one wireless local area network cooperating with the wired network through an access point which has a gateway function.

When a source station in the wireless network transmits a data packet to the wired network, the access point stores the source address of the packet. This backup of the source address is made for each packet received from a wireless terminal by the access point.

It may be useful to learn the source address on each packet in case there is high mobility in the wireless network - new terminals joining the wireless network, - terminals moving from one cell from the wireless network to another, - etc.

However, in a system where the network equipment without is not very mobile, as is the case for example for digital copiers, the frequency with which the equipment is likely to be changed is very low. So in this situation, reading the source address on each packet is a disadvantage.

Also known from US-A-5,812,531 is a method and apparatus for interconnecting a wireless network with a wired network. In order to obtain the address of all wireless nodes, the method uses a topology determination message that is regularly broadcast in the wireless network to all terminals.

This method has the disadvantage of consuming a significant radio bandwidth and is especially suitable for an environment where wireless equipment is very mobile.

The present invention aims to overcome the above disadvantages.

For this purpose, the present invention proposes a method for learning the address of at least one communication terminal connected to a wireless telecommunications network, this learning being carried out by a wireless access point connecting said wireless network. to a wired telecommunications network, remarkable in that it comprises - an activation operation, of activating an automatic learning mode of the address when establishing a radio connection between the communication terminal and the access point, so that the access point performs read operations of storing the original address of the first data packet received from the communication terminal over the aforementioned radio connection - and - an operation disabling, of disabling the automatic learning mode after these storage read operations. The invention has many advantages, in particular - it allows a reduction in the processing time spent for automatic learning the address; it enables real-time updating of the filtering information of the data packets coming from the wired network before transmission on the radio medium (this advantage is explained below); - it constitutes a feasible solution without additional electronic component, by software means, hence a low cost <B>; </ B> - it is a solution independent of the filtering method chosen (by software or specific electronic component) ; - It is a solution specially adapted to allow to provide positioning mobility to heavy and low mobility equipment, such as digital copiers, that is to say that it gives great flexibility to change these equipment. location if needed.

Thus, the invention makes it possible to optimize the processing time of a packet received on the wireless access point radio port for learning the source address, in order to perform the filtering function.

In addition, the invention makes it possible, in a wireless network, to filter data from a wired higher speed network and is adapted to a system where the wired access point is for communicating with adapters. wireless terminal, which can connect only one network termination equipment, for example, a printer, a computer, a digital copier.

The present invention makes it possible to reduce the processing time of each packet, by reading the source address of a device connected to the wireless network, not for each packet, but only when it is necessary, ie say when a wireless network device is likely to have been changed. This improves the performance of the entire system.

According to a particular characteristic of the method of the invention, the radio connection between the communication terminal and the access point is established by a wireless terminal adapter connecting the communication terminal to the wireless network.

Thus, the invention allows, at a wireless terminal adapter, establishing a radio link with the access point whenever the terminal connected to the adapter is likely to have changed. In addition, the access point activates the learning function of the MAC source address each time it receives a connection from a wireless terminal adapter, and disables the learning function after learning the information. source address from the first packet received from the adapter.

According to one particular feature, the wireless access point communicates with the wireless adapter using DECT digital radio communication technology.

The use of this technology is particularly well suited here.

According to a particular characteristic, the communication terminal is a digital copier or a computer or a printer or any other device.

Thus, the invention allows the wireless interconnection of such equipment, which are relatively static or fixed.

According to one particular characteristic, the wired telecommunications network is of Ethernet type and the terminal communicates with the wired network via an Ethernet interface.

Indeed, the present invention is particularly well suited to apply in an Ethernet context.

For the same purpose as mentioned above, the present invention also proposes a packet data transmission method, originating from a wireless telecommunications network to a wired telecommunications network, via a wireless access point, which is remarkable. in that it implements a learning method as above, and in that it comprises, when receiving a data packet to be transmitted, a test operation of determining whether the mode automatic learning of the address of the communication terminal sending packet data is enabled; if and only if the learning mode is activated, reading and storing operations of the address of the communication terminal sending the packet, then deactivating the learning mode; and a packet transmission operation on the wired network. Still for the same purpose, the present invention further proposes a wireless access point allowing access to a wireless telecommunications network, the wireless network comprising a plurality of communication terminals respectively connected to the wireless network by a wireless network. plurality of wireless terminal adapters having no own address, the wireless access point being adapted to receive data packets and having a radio link controller, the access point being characterized in that it comprises in addition to a module for activating the learning of the source address of the first data packet received over a radio connection, when the radio link controller indicates that the connection is a new connection established with one of the plurality of adapters wireless terminal, and a module to disable learning, when the source address has been learned, from the first packet received on this con radio connection. Still for the same purpose, the present invention also proposes a base station in a wireless telecommunications network, remarkable in that it comprises a wireless access point as above.

Still for the same purpose, the present invention further provides a wireless terminal adapter in a wireless telecommunications network having a wireless access point, which adapter has no own address and is connected to a wireless terminal. communication, this adapter being remarkable in that it further comprises - a module for establishing a radio connection with the wireless access point when the network link between a communication terminal and the adapter goes from the inactive state to the active state; and - a module for breaking a radio connection with the wireless access point when the network link changes from the active state to the idle state. Thus, the adapter signals to the access point that the terminal connected to the adapter may have changed, no additional exchange of radio messages being necessary. This optimizes the use of radio bandwidth.

According to a particular characteristic, the adapter further comprises a network link controller.

This characteristic has the advantage of allowing the adapter to easily determine whether or not there is an active terminal on the network.

According to a particular characteristic, the network link controller comprises - module for identifying the state of the network link between the terminal and the adapter; and a module for signaling a change of state of the network link. According to a particular characteristic, the module for signaling a change of state of the network link makes it possible to cause an interruption in the processing carried out by the adapter on an interrupt bus.

This feature has the advantage of being particularly simple to achieve and provides status change information in real time. According to a particular characteristic, the adapter comprises a module for storing the state, active or inactive, of the network link.

This feature has the advantage of allowing the adapter to easily determine whether or not there is an active terminal connected to the adapter.

Still for the same purpose, the present invention further provides a mobile station in a telecommunications network without being notable in that it comprises a wireless terminal adapter as above or is connected to the wireless network via a wireless terminal adapter as above.

As noted above, the mobile station can be a digital copier or a computer or printer.

The invention also aims at - a means of storage of information readable by a computer or microprocessor retaining instructions of a computer program, allowing the covering of a learning method as above, and - a means partially or totally removable computer-readable information storage device that can be read by a computer or a microprocessor retaining instructions from a computer program for covering a learning method as above.

The invention also relates to a computer program comprising instruction sequences for implementing a learning method as above.

The particular features and advantages, not explained above, of the transmission method, the wireless access point, the base station, the adapter, the mobile station, the information storage means and of the computer program being the same as those learning process according to the invention, they are not recalled here.

Other particular features and advantages of the invention will appear on reading the following detailed description of particular embodiments, given as non-limiting examples. The description refers to the accompanying drawings, in which - Figure 1 schematically shows a local wireless telecommunications network, which can be connected to a wired local area network via an access point without wire; FIG. 2 schematically represents the structure of an Ethernet MAC data frame; - Figure 3 schematically shows the structure of a wireless terminal adapter according to the invention, in a particular embodiment; FIG. 4 schematically represents the structure of a wireless access point according to the invention, in a particular embodiment. FIGS. 5A and 5B are flowcharts illustrating the operation of an access point without wire according to the invention, in a particular embodiment; FIGS. 6A, 6B, 7A and 7B are flowcharts illustrating the operation of a wireless terminal adapter according to the invention, in a particular embodiment; and FIG. 8 schematically shows the information flows exchanged by a communication terminal, the wireless terminal adapter associated therewith, a wireless access point and a wired network, such as those illustrated on FIG. Figure 1, when the method of the invention is implemented, in a particular embodiment.

FIG. 1 represents a wireless local area network 107 able to interconnect with a wired local area network 104 via a wireless access point 102.

The wireless local area network 107 consists of several terminal equipments 100, 103 having a communication interface with the network.

In a preferred embodiment of the invention, terminals 100 103 are digital copiers with an Ethernet interface. However, the context of a wired Ethernet network is given by way of non-limiting example; the invention applies both to interconnection with any other type of wired network, for example. for example, a typical ring token ring (in English "token ring").

Each terminal is connected to the wireless local area network via a wireless terminal adapter 101.

A terminal adapter allows only one terminal to be connected.

According to a preferred embodiment of the invention the wireless terminal adapter is connected to the terminal by an interface type 10 or Base T (Ethernet).

The number of wireless LAN terminal adapters is limited by the capacity of the wireless access point.

According to a preferred embodiment of the invention, many terminal adapters 101 supported by the access point 1 are limited to one.

According to a preferred embodiment of the invention, the wireless communication between the access point 102 and the wireless terminal adapters 101 uses the DECT digital radio communication technology specified by ETSI, EN 300 175, 1991.

According to a preferred embodiment of the invention, the wired local network 104 is an Ethernet type network having a rate of 10 and / or 100 Mbitls. The devices 105, 106 and 108 are terminals connected to the wired network 104. For example, the terminal 105 is a print server and the terminals 106 and 108 are computers.

DECT radio technology achieves a maximum throughput of 552 kbps while wired network throughput can reach 100 Mbps. The wireless access point 102 can not therefore transfer on the radio medium all the data circulating on the wired network without risk of losing data. It is necessary to filter the data packets from the wired network before transmitting over the wireless network. Only packets intended for the terminals of the wireless network 100, 103 must be transferred to the radio medium.

Packet filtering is done at level 2 of the OSI ("Open SysfemsInferconnection") model layer, that is, at the <I> ("Medium </ I> Access ControP") layer. Ethernet network.

FIG. 2 shows the structure of an Ethernet data frame designated by the reference 200 in the drawing (defined by the IEEE 802 standard). This frame is composed of several fields. For the present invention, the first two fields 201 and 202 are particularly The "DA" field 201 contains the destination MAC address of the packet, on 6 bytes, and the "SA" field 202 contains the source MAC address, or original address of the packet, on 6 bytes.

The field "L" contains the length of the data contained in the frame, on 2 bytes. The "Info" field contains the actual data, on a number of bytes ranging from 0 to 1500. The Tad "field contains binary bits, up to 46 bytes, so that the total number of bytes contained in the fields "DA", "SA", "L", "Pad" and "FCS" is equal to 64. The "FCS" field (English <I> Trame Check </ I> Sequence ", frame verification sequence ) contains a control sequence for detecting transmission errors.

When the access point receives from the wired network 104 a data packet whose destination address 201 corresponds to the MAC address of one of the communication terminals 100 or 103 of the wireless network 107, this packet must then to be transmitted on the radio medium. On the other hand, when the access point receives a packet whose destination address 201 does not correspond to the MAC address of one of the terminals 100 or 103, this packet must not be taken into account, unless the address Destination 201 corresponds to a "broadcast" address ("BroadcasP") for which the access point performs filtering of the data that reaches it.

To be able to perform this filtering function, the access point must know the MAC addresses of the terminals connected to the local network without the invention proposes a method of automatic learning (that is to say without intervention of an operator) the MAC address, which allows you to build a filter table at the wireless access point. The invention is independent of the filtering method chosen (that the filtering function is carried out totally by software means implemented by a microprocessor, or that the filtering function is performed by a specific component adapted to the CAM <I> type. ("Content </ I> Addressable Memory ').

On the other hand, since the access point supports a single MAC address per wireless terminal adapter (equivalent to a single address per radio connection, since there is a radio connection to the access point by adapter ), which is the address to be taken into account when learning (we will see later that the access point reads the source address on the first received data packet), the invention applies only when the The wireless terminal adapter does not have its own MAC address, otherwise the access point would not "see" the MAC address of the communication terminals. This lack of MAC address specific to the wireless adapter has several advantages: it simplifies the implementation, reduces the cost of implementation brings a gain in processing time.

FIG. 3 shows the electronic circuit of a wireless terminal adapter 101, using the European DECT standard, which can be connected to a 10 or 100 Mbit Ethernet terminal.

The wireless terminal adapter operates in the DECT mode as a portable portion (i.e. mobile station). This wireless terminal adapter can be integrated in a computer-type equipment, printer, digital copier, etc., or connected to one of the equipment mentioned above.

According to a preferred embodiment of the invention the wireless terminal adapter is connected to a terminal having an Ethernet interface.

electronic circuit 101 comprises, connected between an address and data bus 307 - a radio unit 301, provided with an antenna, - a link controller 302, - a microprocessor 303, - a ROM ROM 304, and - a RAM RAM 305.

Furthermore, a power supply 306 provided for supplying all the components of the electronic circuit 101 (the electrical connections relating to the power supply are not shown).

By "link controller" means here any means to determine if there is an active terminal on the network, namely, a non-extinguished terminal. Such a means is known to those skilled in the art and will not be described in detail here.

In the particular embodiment described here, the link controller 302 is an Ethernet type controller, which manages the Ethernet interface for communicating with a terminal of the wireless network 100, 103. In addition, by an interrupt bus 308, the Ethernet controller 302 makes it possible to signal to the microprocessor 303 any change of state of the Ethernet connection, as explained further in connection with FIGS. 7A and 7B. The Ethernet controller 302 contains two "lT Status" and "lT Mask" registers defined below.

All the components illustrated in FIG. 3 are well known to those skilled in the field of processor communication circuits and, more generally, information processing circuits. They are not detailed here. The electronic circuit 101 and, in particular, the microprocessor 303, is adapted to exploit all the capabilities of the DECT standard known before the present invention.

The microprocessor is, furthermore, adapted to implement the method of the invention and, in particular, the flowchart illustrated in FIGS. 6A, 6B, 7A and the random access memory 305 includes registers which, for convenience, carry each the same name as the data they contain, in particular - a <I> "Link </ I> EthemeP 'register, in which the state of the Ethernet link is stored, which can be set to ACTIVE or INACTIVE, depending on whether whether or not there is an active communication terminal (that is to say turned on) connected to the terminal adapter 101, and - <I> "Modification" register, </ I> which indicates, when its state is TRUE, that a modification of the Ethernet link has been signaled by interrupting the interrupt bus 308.

ROM 304 contains registers which, for convenience, each have the same name as the data they contain, in particular - a "Program 9" register, in which the operating program of the microprocessor 303 is stored.

<B> Figure 4 </ B> represents the electronic circuit of the wireless access point 102, using the European DECT standard, which can be connected to an Ethernet network with a flow rate of 10 or 100 Mbitls.

The wireless access point operates in the DECT mode as a fixed part (i.e. base station).

The electronic circuit 102 comprises, connected to each other by an address and data bus 407, a radio unit 401 provided with an antenna, a link controller 402, a microprocessor 403, a ROM 404, and a RAM 405 RAM.

Moreover, a power supply 406 is provided for supplying all the components of the electronic circuit 102 (the electrical connections relating to the supply are not shown).

Link controller 402 is, in the particular embodiment described here, an Ethernet type controller, which manages the Ethernet interface for communicating with the wired network 104.

All the components illustrated in FIG. 4 are well known to those skilled in the field of processor communication circuits and, more generally, information processing circuits. They are therefore detailed <B> "</ B> The electronic circuit 102 and, in particular, the microprocessor 403, is adapted to exploit all the capabilities of the DECT standard known from before the present invention.

The microprocessor is further adapted to implement the method of the invention and, in particular, the flowchart shown in FIGS. 5A and 5B.

The RAM 405 includes registers which, for convenience, each have the same name as the data they contain, in particular - a register <I> "Filter", </ I> containing a table with as many elements as there are wireless terminal adapters connected to the wireless network 107; this table establishes a correspondence between MAC addresses and radio addresses (that is to say, identifying a specific radio connection); and a register <I> "Read Address", </ I> containing a table with as many Booleans as there are wireless terminal adapters connected to the wireless network 107; the booleans in this register are TRUE if activation of address learning is required for the corresponding wireless terminal adapter, or FALSE otherwise.

The read-only memory 404 includes registers which, for convenience, each have the same name as the data they contain, in particular a "Program2" register, in which the operating program of the microprocessor 403 is kept.

wireless access point 102 filter, that is, the <I> "Filter" register, </ I> is initialized after power on, so that the access point transmits on the medium radio only the frames intended for general broadcasting are the frames whose destination address 201 (see FIG. 2) is equal to FF-FF-FF-FF-FF-FF in hexadecimal.

In order to be able to exchange data between the access point and one or more wireless terminal adapters, a radio connection must first be established between the terminal adapter and the access point. This is the wireless terminal adapter that initiates the radio connection with access point.

Thus, after initializing the filtering function of the wireless access point 102, no frame from the wired network 104 is transmitted to the radio medium until a wireless connection is established to a wireless terminal adapter 101 with the wireless access point 102.

FIG. 5A shows the processes executed by the microprocessor 403 of the access point 102 when, during a step 500, a radio connection is received by the access point 102 from one of the wireless terminal adapters 101, this adapter being designated here by ATi.

The microprocessor 403 then performs an activation operation which consists in setting to TRUE the element situated at the position corresponding to the adapter ATi in the table <I> "Read Address" </ I> of the RAM This element is noted Read Address [ATi] in Figure 5A. This operation activates the auto-learning mode of the address.

The <B> Figure 5B </ B> represents the processes executed by the microprocessor 403 of the access point 102 when, during a step 502, packet of data is received over a radio connection from the adapter ATi.

The microprocessor 403 then performs a test 503 on the value of the element of the table <I> "Read Address" </ I> located at the position corresponding to the ATi adapter.

If this element is TRUE then a read, update and deactivate operation 504 described below is performed.

If this element is FALSE, the microprocessor 403 then directly executes a transmission operation 505 described hereinafter. If the element is TRUE, the read, update and deactivate operation 504 reads the source address 202 (see FIG. 2) received packet (i.e. the MAC address of the communication terminal). which issued the packet), then update the wireless access point filter 102 with this new address. If no address was previously associated with the adapter ATi in the filter, this address is then initialized with value of the source address of the packet being processed. If an address was already associated with the ATi adapter, this address remains unchanged in the case where the address read in the received packet is identical to it. Otherwise, this address is then replaced by the new address read in the packet being processed.

Then, the microprocessor 403 deactivates the automatic learning function of the address, by setting FALSE in the random access memory 405 the element of the array <I> "Read </ I> Address" corresponding to the ATi adapter .

The processing phase corresponding to the automatic learning of the terminal address associated with the ATi wireless adapter is then completed. The transmission operation 505, which consists in transmitting the packet to the Ethernet controller 402, is then executed. The Ethernet controller 402 is then responsible for transmitting this packet on the wired network 104.

The processes that have just been described in connection with FIGS. 5A and 5B enable the terminal address learning mode to be activated each time a new radio connection is established between the point of arrival of the terminal. access and a terminal, via the wireless adapter of the latter, - to disable the learning mode of the address as soon as the terminal address has been read from the source address of the first packet received on a given connection.

Thus, the source address is read only on the first packet received on a given radio connection, which reduces the following packet processing time. <B> Figure 6A </ B> represents the processes performed on the microprocessor 303 of the wireless terminal adapter 101 after the power is turned on.

After an initialization phase 600, described below in conjunction with FIG. 7A, the microprocessor 303 checks the state of the Ethernet link activity during a test 601, by testing the value of the variable <I> "Link </ I> EthemeP 'contained in the RAM 305. If the test 601 positive <I> (" Link </ I> Ethemef' = ACTIVE), it means that <I> there </ I> an active communication terminal (ie, turned on) connected to the terminal adapter. connection operation 602 is then performed, which consists in establishing radio connection between the adapter 101 and the access point If the test 601 is negative, nothing happens as the Ethernet link does not become active.

FIG. 6B represents the processing performed by the microprocessor 303 of the wireless terminal adapter 101 after the Ethernet controller 302 has reported to the microprocessor 303 a change in the state of the Ethernet link, this change state being signaled by TRUE setting the <I> "Modification" </ I> variable contained in a RAM register 305 (step 603).

The microprocessor then performs a test 604, to know the variable value <I> "Link </ I> EthemeY." The result of the test 604 is positive when the Ethernet link has moved from the idle state to the active state. If the test 604 is positive, a connection operation 605, which consists in establishing a connection with the access point 102, is then performed If the result of the test 604 is negative, a disconnection operation 606, which consists in breaking the established connection with the access point 102, is performed.

After any of the connection 605 or disconnect 606 operations, the microprocessor 303 performs an update step 607, which resets the <I> "change" variable to </ I> FALSE, indicating that the take account of the change of state was made and allowing to wait for a new notification of the change of the Ethernet link.

The processes described above with reference to FIGS. 6A and 6B make it possible to detect in real time the presence or absence of an active terminal (powered up) connected to a terminal adapter 101, and of establish a radio connection with the access point only when there is an active terminal connected to the terminal adapter 101.

combination of the processes illustrated in FIGS. 5A, 5B, 6A and 6B makes it possible to activate the automatic learning of the address of a terminal each time a new radio connection is established, that is to say to whenever the equipment connected to the wireless terminal adapter is likely to have been changed - either after the terminal adapter is powered on or after the disconnection and re-connection of the Ethernet cable between the terminal 100 or 103 and the wireless terminal adapter 101.

As described with reference to FIG. 3, the Ethernet controller 302 of the wireless terminal adapter 101 contains two "1T Stafus" registers: This register contains various information fields making it possible to identify the state of the Ethernet controller. In particular, the <I> "Link" </ I> field is used to identify the state of the link: if this field contains a logical "1", this means that the link is active, otherwise the link is inactive. Other fields in the "lT Status" register may identify the Ethernet packet receive, end of packet transmission, and other information necessary to control the Ethernet controller by the microprocessor 303; "1T Mask" <I>: </ I> This register contains various information fields enabling the interruption on the interrupt bus 308 to be validated or not for different types of events. In particular, an <I> "Link" </ I> field is used to validate the generation of an interrupt during a change in the state of the link. Other fields can be used to validate the generation of an interrupt when receiving an Ethernet packet or the end of transmission of a packet by the Ethernet controller. In general, if a field of the register "1T <I> Mask" </ I> contains a logical "1", it means that the interruption is validated.

Fig. 7A shows the processing performed by the microprocessor 303 of the wireless terminal adapter 101 during the initialization phase 600 shown in Fig. 6A. The microprocessor checks the status of the Ethernet link activity by first reading the "lT Status" register (read operation 701). Then, during a test 702, the microprocessor 303 tests the value of the field <I> "Link" </ I> of this register. If the test 702 is positive (Field <I> "Link" = </ I> 1), it means that there is an active terminal connected to the terminal adapter. An update operation 703 is then performed, which consists of setting the variable <I> "Link </ I> EthemeP '<I> to </ I> the value ACTIVE.If the test 702 is negative, an operation of update 704 is then performed, which consists of setting the variable <I> "Link </ I> Ethemet" <I> to </ I> the value INACTIVE.

After any of the update operations 703 or 704, the microprocessor 303 performs an interrupt enable operation 705 which consists of 'setting, in the register' / T <I> mask ', </ I> the field <I> "Link" to the </ I> logical value "1" so as to validate the interrupt on the interrupt bus 308, so that it indicates any change of state of the Ethernet link.

Fig. 7B shows the processing performed by the microprocessor 303 of the terminal adapter 101 when an interrupt is received from the Ethernet controller 302.

During a step 706, such an interruption is signaled by the interrupt bus 308. A read operation 707 is then performed, which consists of reading the register "T Status". Then, during a test 708, the microprocessor 303 tests the value of the field <I> "Link" </ I> of this register. If the test 708 is positive (Field <I> "Link" = </ I> 1), it means that there is an active terminal connected to the terminal adapter. An update operation 709 is then performed which consists of setting the variable <I> "Link </ I> EfhemeP '<I> to </ I> the value ACTIVE.If the test is negative, a setting operation of day 710 is then performed, which consists of setting the variable <I> "Link </ I> Ethemef '<I> to </ I> the value INACTIVE.

After any of the update operations 709 or 710, the microprocessor performs a link state change operation 711, which consists of setting the <I> "Modification" </ I> variable to TRUE, thereby allowing it to start. the treatments described above using the flowchart of Figure 6B. FIG. 8 represents an example of the packet flow exchanged between the wired network 104 and the terminal 103 connected to the wireless network 107. The same description would apply for a packet exchange between the wired network 104 and the terminal 100, illustrated in FIG. Figure 1, also connected to the wireless network After initialization of the adapter 101, it detects that the Ethernet link is active so it can establish an 800 connection with access point 102. After receiving the connection of the 101, the access point 102 activates the source address automatic learning mode (activation operation 501 shown in Figure 5A).

After power on, the access point filter initialized to transmit broadcast-type packets. By way of nonlimiting example, packet 801, received from the wired network, which is for example a packet from the terminal 105 to the terminal 106 also connected to the wired network, is not transmitted to the wireless network 107. On the other hand, a packet 802, which is a broadcast-type packet (the broadcast address being symbolically designated by the word "All" in Figure 8), is transmitted to the wireless adapter 101, which then transmits it to the terminal 103.

For example, the packet 802 may be an ARP request packet ("Address Resolution Protocol"), containing the network address of the terminal 103. The terminal 103 must then reply to this message with a packet 803 of the ARP response type at the source of the packet 802, that is to say terminal 105 connected to the wired network The packet 803 is transmitted by the adapter 101 to the access point 102. As the automatic mode of learning the source address is activated the access point 102 reads the packet source address 803 and initializes the filter with this address, then it deactivates the automatic learning, thus the source address of the next packets received on the same connection will not be analyzed ( read, update and disable operation 504 shown in Fig. 5B) The packet 803 is then transmitted by the access point 102 to the wired network 104.

Then, a packet 804 from the wired network terminal 105 to the terminal 103 may be transmitted on the radio medium because the access point filter 102 has been programmed with the address of the terminal 103. The packet 805 is the response of the terminal 103 to the packet 804.

after the sending of the packet 805, the user disconnects the Ethernet cable between the wireless adapter 101 and the terminal 103, the adapter 101 then disconnects from the access point 102 (disconnection 806). If a packet 807, whose destination address is that of the terminal 103, received by the access point 1 from the terminal 105 of the wired network, this packet is not transmitted because the connection with the terminal 103 n is more active.

If the user then reconnects a new terminal 109 in place of the terminal 103, the adapter 101 then establishes a new connection 808 with the access point 102. The access point 102 then activates the learning mode again. automatic source address. At this time, the filter always contains the address of the terminal 103. Thus, if the same terminal 103 is subsequently reconnected to the adapter 101, the system can refit immediately, without the need for new training.

In the nonlimiting example given here, after being turned on, the newly connected terminal 109 in place of the terminal 103 sends a broadcast message 809, which is transmitted to the access point 102 via the adapter 101. Upon receipt of the packet 809, the access point 102 reads the source address (that of the terminal 109), which is different from that contained in the filter (which has stored the address of the terminal 103). The old address (that of the terminal 103) is therefore replaced by the new address (that of the terminal 109). Thus, the filter is updated in real time. A terminal having left the wireless local area network 107 is thus quickly erased from the filter.

 After reading the address of the terminal 109, the learning mode is deactivated again (operation 504 shown in FIG. 5B) and the packet 809 is transmitted to the wired network 104. Next, the terminal 105 of the wired network and the terminal 109 can exchange packets, such as packets 810 and 811, illustrated by way of non-limiting example in FIG. 8.

Claims (11)

<U> CLAIMS </ U> A method of learning the address (202) of at least one communication terminal (100, 103) connected to a wireless telecommunications network (107), said learning being performed by a wireless access point (102) connecting said wireless network (107) to a wired telecommunications network (104), characterized in that it comprises - an activation operation (501), of activating an automatic learning mode of said address (202) when establishing a radio connection between said communication terminal (100, 103) said access point (1 so that the access point (102) performs operations of reading and storage of the the original address of the first data packet received from said communication terminal (100, 1 on said radio connection; and - a deactivation operation (504) of disabling said automatic learning mode after said read operations and of me morisation. The method according to claim 1, characterized in that said radio connection between said communication terminal (100, 103) and said access point (102) is established by a wireless terminal adapter (101) connecting said terminal communicating (100,103) to said wireless network (107). The method of claim 1 or 2, characterized in that said wireless access point (102) communicates with said wireless adapter (101) using the DECT digital radio communication technology. 4. Method according to claim 1, 2 or 3, characterized in that said communication terminal (100, 103) is a digital copier or a computer or printer or any other device. 5. Method according to any one of claims 1 to 4, characterized in that said wired telecommunications network (104) is Ethernet type and said terminal (100, 103) communicates with the wired network (104) via an Ethernet interface. A method of transmitting packet data from a wireless telecommunications network (107) to a wireline telecommunications network (104) through a wireless access point (102), characterized by: implement a learning method according to any one of claims 1 to 5, and in that it comprises, on receiving a data packet to be transmitted - a test operation (503), consisting in determining if the automatic learning mode of the address of the transmitting communication terminal of said data packet is activated; if and only if said learning mode is activated, operations (504) for reading and storing the address of the communication terminal transmitting said packet, and then deactivating said learning mode; and - a transmission operation (505) of said packet on the wired network (104). A wireless access point (102) for accessing a wireless telecommunications network (107), said wireless network (107) having a plurality of communication terminals (100, 103) respectively connected to said wireless network wire (107) by a plurality of wireless terminal adapters (101) having no own address, said wireless access point (102) being adapted to receive data packets and having a radio link controller ( 401), said access point being characterized in that it further comprises - means for activating the learning of the source address of the first data packet received over a radio connection, when said radio link controller (401 ) indicates that said connection is a new connection established with an adapter (ATi) among said plurality of wireless terminal adapters (101), and - means for disabling said learning, when said source address has been learned, from udit first packet received said radio connection. Base station in a wireless telecommunications network, characterized in that it comprises a wireless access point (102) according to claim 7. A wireless terminal adapter (101) in a wireless telecommunications network (107) having a wireless access point (102), said adapter (101) having no own address and being connected to a terminal communication device (100, 103), said adapter (101) being further characterized by means for establishing a radio connection with said wireless access point (102) when the network link between a mobile terminal communication and said adapter (101) switches from the inactive state to the active state, and - means for breaking a radio connection with said wireless access point (102) when said network link changes from the active state to the active state the inactive state. 10. The adapter of claim 9, characterized in that it further comprises a network link controller (302). An adapter according to claim 10, characterized in that said network link controller (302) includes - means (I / I) to identify the network link state between a terminal and said adapter. (101); and means ("! T Mask") to signal a change of state of the network link. 2. An adapter according to claim 10 or 1, characterized in that said means <I> ("! T Mask") </ I> to signal a change of state of the network link can cause an interruption in the processing carried out by said adapter (101) on an interrupt bus (308). 3. An adapter according to claim 10, 11 or characterized in that it comprises means <I> ("Link </ I> EthemeY ') for storing the active or inactive status of the network link. wireless telecommunications network, characterized in that it comprises a wireless terminal adapter (101) according to any one of claims 9 to 13 or is connected to said wireless network via a terminal adapter without wire (101) according to any one of claims 9 to 13. Mobile station according to claim 14, characterized in that it is a digital copier or a computer or a printer.