WO2009125125A1 - Technique for connecting a node to a subnetwork by way of an access point in a communication network - Google Patents

Abstract

The invention relates to a technique for connecting a node (MN) to a subnetwork (SN1, SN2) by way of an access point (PA1-PA3) in a communication network (RC), implemented by the node. A method of connection comprises: - a step of selecting a new access point from among a plurality of detected access points, - a step of detecting a subnetwork modification, characterized in that it furthermore comprises a step of receiving by the node an item of information broadcast by the new access point, said item of information being representative of a subnetwork to which the new access point is connected and in which the step of detecting a subnetwork modification is performed by means of the item of information received.

Description

 Technique for connecting a node to a subnet through an access point in a communication network

The invention relates to a technique of connecting a node to a subnetwork via an access point in a communication network.

Here we are dealing with mobility management, at the network level, of mobile nodes in a communication network such as the Internet network. A mobile node may be an IP terminal or a mobile router capable of receiving terminals. Mobile nodes are able to detect wireless access points while traveling. These wireless access points are themselves connected to IP subnets. An IP subnet provides access to the Internet and administers a set of IP addresses, hereafter referred to as the IP prefix, that can be allocated to nodes connected to the subnet. The mobile node during its movement detects a new access point and decides, according to its own criteria, to connect to this new access point. It performs an inter-access point transfer between the access point to which it was connected to the new access point. Then, if the IP subnet to which the new access point is connected is different from the one to which the previous access point is connected, it acquires the network prefix of the new IP subnet and requests the allocation of an address. depending on this network prefix. These actions allow it to establish a connection to the IP network through the new access point.

The time required to detect an IP subnet change is an important part of the interrupt time observed for the exchanges between the mobile node and the IP network. "Mobile IP" protocols, as defined by the IETF, for "Internet

Engineering Task Force ", in its IPv4 version by RFC 3344 and the draft document titled ietf-mip4-rfc3344bis-05.txt and in its IPv6 version by RFC 3775, provide methods for detecting subnetwork modification IP based on ICMP message scanning, for "Internet Control Message Protocol." An IP network access router broadcasts ICMP announcement messages, called "Router Advertisement", comprising an IP subnet network prefix and a field indicating a validity period of the announcement message. The mobile node, once connected, listens continuously to the announcement messages sent. The broadcasting of the announcement messages can be implemented according to two methods. A first method is based on spontaneous periodic sending of announcement messages by the access router to which the terminal connects. As long as the validity period of the ad message has not expired, a new ad message with a new network prefix is ignored. Then, after the time expires, following detection of a network prefix change, the node initiates a procedure to obtain a network address from the new IP subnet. If the time between two announcement messages is too long, the time required to detect the IP subnet change is too long. If the time between two announcement messages is too short, the node can quickly detect an IP subnetwork modification but in return, the access network load increases, which is not desirable especially for points wireless access.

A second method relies on the use of a "Router Solicitation" solicitation message transmitted by the mobile node to the access router at the expiration of the validity period of the announcement message. Thus, the access router transmits to the mobile node an announcement message upon receipt of the node's request message. This method also presents the problem of waiting for the expiry of the validity period and leads to the issuance of possibly unnecessary messages in case of absence of modification of IP subnet.

There is therefore a need for a technique that allows a mobile node to decrease the time required to detect a subnetwork change during an access point change without increasing the access network load.

The invention addresses this need by providing a method of connecting a node to a subnet through an access point in a communication network, implemented by the node, the method comprising: a step of selecting a new access point from among a plurality of detected access points, a step of detecting a subnetwork modification; characterized in that the method further comprises a step of reception by the node of information broadcast by the new access point, said information being representative of a subnet to which the new access point is connected and in which which, the step of detecting a subnetwork modification is performed using the information received.

The node may be a terminal or a router capable of receiving terminals.

The node may be mobile, i.e., it is likely to move in a radio environment. It can also be fixed, the access point being changed by a user of the node, a first access point, to which the node is for example connected in a wired manner, to a second access point, to which the node wants to connect for example by radio.

Thus, the node obtains information representative of the subnet to which the new access point is connected and can, by comparison with such information received from the old access point, quickly determine whether there has been a change in subnet and if it is necessary to obtain a new address in the new subnet. This makes it possible not to wait for a timer to expire to obtain this information.

This method is also advantageous because it does not require the taking into account of information relating to the IP protocol layer in the physical layer.

It should also be noted that the method does not induce additional load of the radio channels, the information being broadcast according to existing mechanisms on a broadcast radio channel.

In addition, the access points being grouped into sets according to their respective subnetwork, the information broadcast by the new access point identifies a set to which the new access point belongs.

In this case, the information broadcast is an identifier representing a set of access points, these being grouped according to the subnet to which they are connected. It is very easy to compare the identifiers of the respective sets to which the old and new nodes belong. In addition, in case of detection of a subnetwork modification, the method comprises a step of obtaining an address in a subnet to which the new access point is connected.

The procedure for obtaining an address can be carried out using the ICMP protocol or by any other mechanism for obtaining an address, for example using the DHCP protocol (for "Dynamic Host Configuration Protocol").

Thus, the node can immediately trigger a procedure for obtaining a new address in the subnet to which the new access point is connected if it is different from the subnet to which the old access point is connected. . This makes it unnecessary to wait before initiating the procedure to obtain an address in the new subnet and perform this procedure only when necessary. This also reduces the time required for a connection to the communication network following a subnetwork modification.

In one embodiment, the selection step takes into account, when selecting a new access point, a possible modification of the sub-network, the step of detecting a sub-network modification being carried out beforehand. to this one.

The criteria used during the selection of an access point can thus be modified to take into account a possible modification of sub-network. Thus, it may be chosen to favor certain sub-networks over others.

Moreover, during the step of selecting a new access point, at least one access point is favored for which the information broadcast is identical to that broadcast by the access point to which the node is connected. .

Thus, it may be chosen not to make a change to an access point causing a subnetwork change as long as it is possible to remain on an access point connected to the same subnet.

The invention also relates to a node of a communication network, able to connect to a sub-network via an access point and change access point, an access point being connected to a subnet, including: means for selecting an access point in a plurality of access points detected by the node;

means for detecting a sub-network modification; characterized in that the node further comprises means for receiving information broadcast by an access point, said information being representative of the subnet to which the access point is connected and wherein, the detection means of a subnet change process the received information.

The invention also relates to an access point of a communication network, able to allow access to nodes to a sub-network, comprising means for broadcasting information to nodes, characterized in that the means broadcast are arranged to broadcast further information representative of the sub-network.

The invention also relates to a communication system in a communication network comprising:

at least one node as described above; a plurality of access points to the network as described above.

The invention further relates to a computer program comprising instructions for implementing the connection method according to claim 1 when the program is executed by a processor.

The invention also relates to a signal supporting a message broadcast by an access point to nodes in a communication network, said message comprising information representative of a subnet to which the access point is connected.

The invention will be better understood with the aid of the following description of a particular embodiment of the method of the invention, with reference to the appended drawings in which: FIG. 1 represents an architecture of a communication network; FIG. 2 represents the steps of the method according to a particular embodiment of the invention; FIG. 3 represents a device according to a particular embodiment of the invention; FIG. 4 represents an information element intended to carry information representative of an IP subnet.

In Figure 1 is shown an MN node moving in a radio environment. The mobile node MN may be an IP terminal or a mobile router capable of receiving terminals. It is able to detect wireless access points PA1, PA2, PA3 during its movement represented by an arrow. These wireless access points PA1, PA2 are themselves connected to an IP subnet SN1 via an access router RA1 for accessing an RC communication network. The wireless access point PA3 is connected to an IP subnet SN2 via an access router RA2 for accessing the communication network RC. The physical layers used for the establishment of the connection with the network RC can be of different types of wireless technologies: cellular, satellite, IEEE 802.11b, .... One places oneself thereafter in the particular case of points d wireless access according to the IEEE 802.11b specification. An access router provides access to the RC communication network and can manage one or more subnetworks. It administers a set of IP addresses, hereafter called network prefix, which can be allocated to nodes connected to the subnet. The MN node connects to a subnet through an access point. It is placed later in the case where the mobile node MN is connected to the access point PAl by radio and obtained an address from the access router RAl.

The wireless access points PA1, PA2, PA3 broadcast respectively on a radio channel messages comprising various information relating to access. According to the invention, the messages furthermore comprise information representative of the subnet to which the new access point is connected, as represented in FIG. 4. On the latter, is represented an information element comprising an identifier EI. , information of length L of the information element and a field denoted Info, intended to contain the information representative of the sub-network. In a first variant of the particular embodiment, the access points are grouped together in sets according to their respective subnetwork and the information broadcast by the access points identifies a set to which they belong.

For example, in the case of access points that comply with the IEEE 802.11b specification, they regularly issue Beacon frames on a common broadcast radio channel. These "Beacon" frames include access point information, such as an SSID access point identifier, supported data rates, supported services, etc. The information element can be added to the content. of this frame, or be provided in information elements that can be parameterized by an access point constructor ("Vendor Specifies"). In the first variant of the particular embodiment and in the example of FIG. 1, the access points PA1 and PA2 make it possible to access the same subnet and therefore belong to the same set, denoted El. They all broadcast both a message comprising in the Info field an identifier of the set El. The access point PA3 provides access to another subnet and belongs to a second set, noted E2. It broadcasts a message including in the Info field an identifier of the set E2. The access points PA1, PA2 and PA3 therefore comprise means for broadcasting information representative of the connection subnetwork.

The method of connecting the MN mobile node to a subnet through an access point in a communication network will now be described in connection with Figure 2.

In a waiting EO step, the node is waiting for a timer to expire.

In a step E2, denoted "Det PA" in FIG. 2, the MN node performs radio power measurements on a set of detected access points. Then, in a test step, the MN node checks whether a new access point has been detected with sufficient received power.

If the set of access points does not include new access points, the method proceeds to step E6. In the opposite case, in a reception step E4, denoted "Rec Info PA" in FIG. 2, the node receives one or more information broadcasted by the new access point (s). Broadcast information is representative of the subnet to which a new access point is connected.

In a selection step E6, denoted "PA sel" in FIG. 2, the node selects, from among a set of access points that it has detected, an access point according to its own criteria, by example the level of reception of the different access points.

If, after the selection step E6, it is decided by the MN node to remain on the current access point, the process returns to the waiting step E0.

If, after the selection step E6, it is decided by the node MN to connect to a new access point selected, in a step E8 of establishing the link between the node and the point of access. Access, denoted "Conn PA", the node establishes a radio link with the new access point selected.

Then, once this link is established, in a test step ElO, denoted "Comp" in FIG. 2, the node detects a possible subnetwork modification from the information received from the new access point to the step E4 and so, if it is necessary to obtain a new address in the subnet to which the access point is connected.

For this purpose, it can compare this received information with that received from the access point to which it was previously connected. It implements a step of detecting a subnetwork modification. In the first variant of the particular embodiment, the MN node can directly compare the respective identifiers of the sets of access points to which the old and new access points belong.

If it is not necessary to obtain a new address from the subnet, the process returns to the waiting EO step. In the opposite case, the node MN implements a detection procedure for obtaining a new address. In a step E12, denoted "Send RS", the MN node sends to the access router to which the access point is connected a request message to obtain the network prefix of the subnet administered by the access router. . This can be an ICMP message for "Internet Control Message Protocol", such as "Router Solicitation". In a step E14, denoted "Recep RA" in FIG. 2, the node receives a message comprising the network prefix used in the new subnet. This can be an ICMP message such as "Router Advertisement".

In a step E16, denoted "Obt ad" in FIG. 2, the node selects an address from the set of addresses defined by the network prefix and performs the procedures necessary for its allocation. At the end of this step E16, the node has established a connection with the communication network via the access point and the sub-network. The process then returns to the waiting step E0.

The steps E12 to E16 are more particularly described in the case of the implementation of the ICMP protocol but can also be implemented using any other protocol to obtain an address in a subnet, for example to using the DHCP protocol.

In a second variant of the particular embodiment, the MN node performs the ElO step of testing or detecting a sub-network modification prior to the step of selecting a new access point from among a plurality of access points received. At the end of this test step ElO, he determined a possible change of sub-network. The criteria specific to the node used during the selection step E6 can be modified in order to take into account the possible change of sub-network. He may choose to favor a change to a particular subnet. It may also choose to privilege, during the selection, an access point for which a subnetwork change is not necessary, that is to say an access point for which the information broadcast is identical to that broadcasted by the access point to which the node is connected. Thus, it is possible to take into account when selecting an access point information to avoid a change of sub-network if there is an access point to stay on the same subnet. This also makes it possible to reduce an interruption time during an access point change by avoiding the procedures necessary to obtain an address and thus to reestablish a connection with the RC communication network. The other steps of the process are unchanged. By way of nonlimiting example, in the particular case shown in FIG. 1, the node MN is connected via the access point PA1 to the first subnet. Then, during its displacement materialized by the arrow, in a step E6, the MN node selects the access point PA2. It is connected to the same subnet as the access point PAl, which is identified during the step ElO. It is not necessary to obtain a new address. Then, the MN node continuing its movement detects the access point PA3 then selects it, the access point PA2 no longer having a sufficient level. The access point PA3 is connected to the second subnet, which is detected during the step ElO. The MN node then immediately initiates a procedure for obtaining an address in the second subnet, without waiting for the expiration of the validity period of the "Router Advertisement" message received from the first subnet.

In the second variant of the particular embodiment, by way of nonlimiting example, the node MN can receive the two access points PA2 and PA3, the reception level of the access point PA3 being greater than that of the point d PA2 access. However, the MN node identifies during the detection step ElO using the broadcast information received respectively from these two access points that this change of access point causes a subnetwork modification. The MN node then privileges in the selection step E6 PA2 access point to avoid an address obtaining procedure as the received radio power is sufficient to allow a link of good quality.

A node 100 according to a particular embodiment will now be described in relation to FIG.

Such a node is able to connect to a subnet through an access point and change access point. It includes:

a module 102, noted "Mrecep" in FIG. 3, for receiving information broadcast by an access point, said information being representative of the subnet to which the access point is connected; a module 104, denoted "Mdet" in FIG. 3, for detecting a subnetwork modification, arranged to detect a subnetwork modification as a function of information received by the module 102;

a module 106, denoted "Mobt" in FIG. 3, for obtaining an address in a subnet to which an access point is connected;

a module 108, denoted "Msel" in FIG. 3, for selecting an access point; arranged to select, according to its own criteria, an access point in a set of access points detected by the node.

Optionally, the selection module 108 is furthermore arranged to take into account in the selection any change detected by the module 104 for detecting a subnetwork modification.

The modules 102, 104, 106, 108 are arranged to implement the previously described connection method. These are preferably software modules comprising software instructions for executing the steps of the connection method described above, implemented by a processor of a node of a communication network.

The invention therefore also relates to:

a computer program comprising instructions for implementing the connection method as described above when this program is executed by a processor;

a recording medium readable by a node on which is recorded the computer program described above.

The software modules can be stored in or transmitted by a data carrier. This may be a hardware storage medium, for example a CD-ROM, a magnetic diskette or a hard disk, or a transmission medium such as an electrical signal, optical or radio, or a telecommunications network.

The invention also relates to an access point of a communication network, able to allow access to nodes to a sub-network, comprising means for broadcasting information to nodes, the broadcasting means being arranged to further disseminate information representative of the sub-network. The invention also relates to a communication system in a communication network comprising:

at least one node as described above;

a plurality of network access points, as described above.

The method can also be implemented for a fixed terminal, when following a user action, the connection interface passes from one access point to another, for example an access point by wired link to an access point over a radio link.

Claims

A method of connecting a node (MN) to a subnet (SN1, SN2) via an access point (PA1-PA3) in a communication network (RC) implemented by the node, the method comprising: a step (E6) of selecting a new access point from among a plurality of detected access points, a step (ElO) for detecting a subnetwork modification; characterized in that the method further comprises a step (E4) of reception by the node of information (Info) broadcast by the new access point, said information being representative of a sub-network to which the new point of access The access is connected and wherein the step of detecting a subnetwork modification is performed using the received information. 2. The connection method according to claim 1, wherein the access points being grouped into sets according to their respective subnetwork, the information broadcast by the new access point identifies a set to which the new point of access. access belongs. 3. Connection method according to claim 1, comprising, in case of detection of a subnetwork modification, a step (E 16) for obtaining an address in a subnet to which is connected the new point d 'access. 4. Connection method according to claim 1, wherein, the selection step takes into account when selecting a new access point a possible modification of the sub-network, the step of detecting a change. subnet is performed prior to this. The connection method as claimed in claim 4, wherein during the step of selecting a new access point, at least one access point is favored for which broadcast information is identical to that broadcast by the access point to which the node is connected. Node (MN, 100) of a communication network, able to connect to a subnetwork through an access point and change access point, an access point being connected to a subnet, including: means (108) for selecting an access point in a plurality of access points detected by the node; means (104) for detecting a sub-network modification; characterized in that the node further comprises means (102) for receiving information (Info) broadcast by an access point, said information being representative of the subnet to which the access point is connected and in which , the means for detecting a subnetwork modification process the received information. 7. Access point (PA1-PA3) of a communication network, able to allow access to nodes to a sub-network, comprising means for broadcasting information to nodes, characterized in that the means broadcast are arranged to broadcast further information representative of the sub-network. 8. Communication system in a communication network comprising: at least one node according to claim 6; a plurality of network access points according to claim 7. 9. Computer program comprising instructions for implementing the connection method according to claim 1 when the program is executed by a processor. A signal supporting a message broadcast by an access point to nodes in a communication network, said message comprising information (Info) representative of a subnet to which the access point is connected.