JP2005223753A - Mobility management method and handover control method of network for mobile communication - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a mobility management method and a handover control method of a network for mobile communication which distributes the load of processing accompanying handover of a layer 2 (data link layer), introduces a flexible switching mechanism without the use of a stationary label table, and moreover participates handover for a wireless section also positively from the network side to deal with high-speed movement of a mobile terminal. <P>SOLUTION: With a network (layer 2 network) for mobile communication connected to an access point, handover processing between the mobile terminal moving at high speed and the network and mobility management following it are performed dispersively with each access switch and core switch by using an exchange of a control frame. A filtering database is constituted appropriately for switching, and movement prediction of the mobile terminal is performed by a simple method by performing an appropriate information exchange between access switches. The result of this movement prediction is conveyed to the mobile terminal through the wireless section from the network side, and handover processing is performed smoothly. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a mobility management method and a handover control method that are particularly necessary for establishing an appropriate communication path in a mobile communication network accessed using a wireless LAN technology.

Various standards for wireless LANs define communication means establishment methods up to the physical layer (layer 1) and data link layer (layer 2) of the ISO OSI reference model. On the other hand, in the network layer (layer 3), the Internet standard IP protocol is the most widely used protocol in the current data communication network.
The original IP protocol was insufficient in terms of mobility management in a situation where a terminal to which an IP address is assigned moves from the initial development assuming use in a wired network. However, until now, the standard has been expanded to deal with such problems, and this technology is generally called Mobile IP.

  Here, the term mobility management is used to mean that information associated with the position of a terminal that changes with time is managed in a situation where terminals involved in communication move. At this time, the information corresponding to the position and the information corresponding to the time may be explicitly associated with each other or may not be managed. The essence of mobility management here is that the location of the mobile terminal is grasped with the accuracy required to reliably send information to the mobile terminal at each time when the mobile terminal communicates. is there.

In the following description, the process of changing the belonging relationship to the network element that is performed as the mobile terminal continuously moves is referred to as handover.
In a network to which mobile IP is applied, when a mobile terminal belongs to a new IP subnetwork at a destination and is given a temporary IP address called a care-of address, this is transmitted to the network side in the form of registration. This corresponds to performing layer 3 handover. Here, the part indicating the network address in the IP address can be usually associated with a rough position. As a result, the layer 3 handover described above, that is, the mobile IP also has an aspect as mobility management.

In contrast, layer 1 and layer 2 handover has a strong meaning as establishment of communication means, and is not directly linked to the concept of mobility management. This is generally due to the fact that the identifier used in layer 2 has no correlation with the position.
It is assumed that the above-described layer 3 handover processing by mobile IP is independent of the layer 1 and layer 2 handover processing. In general, since handover of a lower layer precedes handover of an upper layer, mobile IP needs to wait for completion of handover of layer 1 and layer 2 before starting processing. In addition, the normal layer 3 process takes a relatively long time. For these reasons, it is difficult for the mobile IP to cope with the case where the mobile terminal moves at high speed.

For this situation, a method has been proposed in which information related to layer 2 handover is used as a trigger for starting layer 3 handover processing and processing is started at a stage before completion of layer 2 handover. However, since such a method does not reduce the layer 3 handover process itself, it can be said that it is difficult to cope with the high-speed movement of the mobile terminal.
It is also conceivable to reduce the handover process by adopting a network configuration in which layer 3 handover is not required and only handover to layer 2 is required. That is, a layer 2 switching network is constructed over a wide geographical area and used as a relay network for mobile communication. When a wireless LAN is used in the access section of this network, each access point (Access Point) is connected to a layer 2 switch port.

  In general, a layer 2 switch has a mechanism for switching a frame to an appropriate port. In order to cope with the situation where the mobile terminal moves at high speed in such a network, not only the above-described mechanism can be dynamically reconfigured but also the port that can reach the mobile terminal is communicated. It is necessary to recognize correctly at each time to do. This means that the concept of mobility management is also adopted in the layer 2 handover.

Based on the above situation, a conventional example will be described.
(First conventional example)
The first conventional example is to use a VLAN ID defined by IEEE 802.1Q as a layer 2 identifier and associate this with the location of the mobile terminal.
The VLAN ID is to be distinguished from the MAC address (also referred to as a physical address) that is most commonly used as a layer 2 identifier, and is referred to as a label here. At this time, switching by the label is performed in the network.

FIG. 30 is a diagram showing an overall configuration of a network 300 using such layer 2 label switching.
Each switch SW shown in FIG. 30 has a database (hereinafter referred to as a label table) that associates a label (VLAN ID) with a port of the switch SW, which is referred to at the time of switching. The label table does not change with the movement of the mobile terminal MH but does not change. That is, the label is statically associated with the port of the switch SW and further with the access point AP ahead. Further, the label is given so that each mobile terminal MH does not overlap.

  The mobile terminal MH needs to change the access point AP to which the mobile terminal MH belongs as the mobile terminal MH. At this time, the registration for establishing a connection relationship with the network 300 with respect to the access point AP to be the new affiliated destination Send a request. The access point AP that has received the registration request sends a label allocation request for the mobile terminal MH moving to the network 300, and receives a new label from the network 300 as a result. The access point AP notifies the mobile terminal MH of a new label, and thereafter, communication involving the mobile terminal MH is performed using the new label. The label controller 301 is responsible for label management in the network, and may be installed anywhere on the network 300.

As described above, in the method shown in FIG. 30, the label table held in each switch SW of the network 300 does not change with the movement of the mobile terminal MH, and does not change, the label is the port of the switch SW, It is statically associated with the access point AP ahead. Therefore, the label corresponds to position information in a broad sense. Further, since the label control device 301 recognizes the access point AP to which the moving mobile terminal MH belongs and assigns a label, the label control device 301 grasps the position of the mobile terminal MH as a result. . Therefore, this method is a method for performing intensive mobility management in the label control apparatus in association with the layer 2 handover (see Non-Patent Document 1).
(Second conventional example)
Similar to the first conventional example, the second conventional example has a network configuration and a handover method for performing mobility management of the mobile terminal MH moving at high speed without performing processing related to routing in the layer 3. A second conventional example will be described with reference to FIG.

FIG. 31 is a diagram showing the overall configuration of a network in the second conventional example. The network shown in FIG. 31 is a network for narrow area communication in which the range covered by one access point AP is narrower than that of a wireless LAN or the like, such as a non-stop automatic toll collection system at a tollgate on a highway. is there.
As shown in FIG. 31, in this network, the zone control devices 313 and 315 accommodate a plurality of access points AP belonging to the zones 312 and 314 managed by the zone control device 313 and 315, and the network control device 311 at a higher level controls all the zone controls. The devices 313 and 315 are accommodated.

  As shown in FIG. 31, in this network, when a connection is secured between a moving mobile terminal MH and a new access point AP, the access point AP registers the location of the mobile terminal MH with respect to the zone controller 315. Is sent to the network controller 311 from the zone controller 315. This location registration information includes not only information for identifying the mobile terminal MH but also time information at which registration was performed. That is, as the mobile terminal MH moves, the position information and time information are associated with each other, and these pieces of information are accumulated in the network control device one after another.

The network control apparatus 311 can predict the movement of the mobile terminal MH to some extent based on such information. When prediction is performed by the network control device 311, it is necessary for all access point APs belonging to one or a plurality of zones (312 and 314) predicted to be entered next by the mobile terminal MH based on the result. Information is multicast. By adopting a mechanism based on such prediction, processing time related to handover is shortened (see Patent Document 1).
(Third conventional example)
The third conventional example is a technique that includes a prediction element in handover processing in a network for narrow area communication. Specifically, a position detection device for detecting the position of the mobile terminal MH in advance is provided, The access point AP to be used next is predicted based on the detected information. (See Patent Document 2)
JP 2002-152113 A JP 2003-229801 A Kawakami, Iino, Suzuki, “Examination of Mobility Control in Wide Area Ethernet (Registered Trademark)”, IEICE Technical Report IA2002-24, pp.15-22,2002

In the first conventional example, in a layer 2 label switching network using a VLAN ID as a label, mobility management associated with layer 2 handover is centrally performed by a label control device. Therefore, there has been a problem of an increase in load due to the concentration of processing in the label control apparatus or inefficiency.
In addition, since the label table held by each switch is static, for example, in order to minimize frame loss during the handover process, a state of multicasting to several specific ports is created in a spot manner. However, there is a problem that flexible control is difficult.

In addition, for example, when a failure occurs in the label allocation due to a failure in the label control device, there is a problem that the connection between the mobile terminal and the network is impaired even if there is no abnormality in the layer 1 there were.
Furthermore, since layer 2 handover by label allocation does not have the effect of reducing handover processing in the radio section between the mobile terminal and the access point, it is difficult to cope with the situation where the mobile terminal moves at high speed on a train or the like. There was a problem.

  In the second conventional example, position information is associated with time information in a narrow area communication network, and a prediction element is included in the handover process. However, with this technique, the processing is concentrated on the network control device, and the load increases greatly when this technology is applied to a large-scale network. In addition, since multicasting of information is performed for all access points belonging to a plurality of prediction destination zones, when applied to a large-scale network, bandwidth usage efficiency is greatly impaired due to an increase in unnecessary traffic. For these reasons, there is a problem that it is difficult to apply this technology to a large-scale network.

  In the third conventional example, handover processing is predicted by detecting the position of a mobile terminal in advance in a narrow area communication network. However, when this technology is applied to a large-scale network that covers a wide geographical area, the accuracy of prediction deteriorates. Further, when the number of position detection devices is increased in order to increase the prediction accuracy, the number of devices that do not contribute to the original user traffic transfer process increases. For these reasons, there is a problem that it is difficult to apply this technology to a large-scale network.

  In order to solve the problem pointed out in the first conventional example, the present invention distributes the processing load associated with layer 2 handover and introduces a flexible switching mechanism that does not rely on a fixed label table. Objective. Furthermore, an object of the present invention is to provide a mobility management method and a handover control method for a mobile communication network that are actively involved from the network side in handover of a radio section in order to cope with high-speed movement of a mobile terminal.

  In addition, the present invention solves the problems pointed out in the second and third conventional examples, so that handover control can include a prediction element in a handover process without any problem even in a large-scale network over a wide geographical area. The purpose is to introduce the method.

  The mobility management method for a mobile communication network according to claim 1, comprising a mobile terminal, an access point that gives the mobile terminal a means for connecting to the network, and a filtering database, and is included in a search result and a frame of the filtering database. In a mobility management method for a mobile communication network comprising a layer 2 switch that forwards the frame to an appropriate port based on the layer 2 information to be transmitted between the mutually connected layer 2 switches. If the mobile terminal newly establishes a radio link with the access point in a certain time interval, transmits information on the necessary part of its own filtering database in the direction, it connects from the new access point Transmit that fact to the layer 2 switch The layer 2 switch that has received the actual article updates the filtering database related to the physical address unique to the mobile terminal, and grasps the port that transfers the user frame to the mobile terminal when the fact is confirmed. It is characterized by.

According to the first aspect of the present invention, since the filtering database is updated quickly, accurate mobility management of the mobile terminal is realized by each switch.
The mobility management method for a mobile communication network according to claim 2, comprising a mobile terminal, an access point that gives the mobile terminal means for connecting to the network, and a filtering database, and is included in a search result and a frame of the filtering database. In a mobility management method for a mobile communication network comprising a layer 2 switch that forwards the frame to an appropriate port based on the layer 2 information to be transmitted between the mutually connected layer 2 switches. If the mobile terminal newly establishes a radio link with the access point in a certain time interval, transmits information on the necessary part of its own filtering database in the direction, it connects from the new access point Transmit that fact to the layer 2 switch The layer 2 switch that has received the actual article updates the filtering database related to the layer 2 label that is uniquely given to the mobile terminal and used for communication over a radio link, triggered by the confirmation of the fact, It is characterized by grasping a port for transferring a user frame to the mobile terminal.

According to the second aspect of the present invention, since the filtering database is updated quickly, accurate mobility management of the mobile terminal is realized by each switch.
A mobility management method for a mobile communication network according to claim 3, comprising a mobile terminal, an access point that gives the mobile terminal a means for connecting to the network, and a filtering database, and is included in a search result and a frame of the filtering database. In a mobility management method for a mobile communication network comprising a layer 2 switch that forwards the frame to an appropriate port on the basis of layer 2 information to be transmitted, bidirectional or In one direction, when necessary information of a part of its own filtering database is transmitted at a certain time interval, and the mobile terminal newly establishes a radio link with the access point, from the new access point, The fact is transmitted to the layer 2 switch to which In the layer 2 switch, the filtering database related to the layer 2 label that is uniquely associated with the mobile terminal but is not used for communication over the radio link is updated by confirming the fact. And grasping a port for transferring a user frame to the mobile terminal.

According to the third aspect of the present invention, since the filtering database is updated quickly, accurate mobility management of the mobile terminal is realized by each switch.
5. The mobile communication network handover control method according to claim 4, comprising: a mobile terminal; an access point that provides a connection means to the network to the mobile terminal; and a filtering database, wherein the filtering database includes a search result and a frame. In a handover control method for a mobile communication network consisting of a layer 2 switch that forwards the frame to an appropriate port based on layer 2 information to be transmitted, the upper layer 2 switch has a filtering database owned by itself. By using the data related to the layer 2 information of the mobile terminal, the mobile terminal grasps the timing at which the mobile terminal should switch the radio link to a new access point, and the port to be used for the communication of the mobile terminal next. 2 switches When it is determined that the next switching process can be completed only with the layer 2 switch under its own, a switching process start instruction is issued from the port used for communication of the mobile terminal to the lower layer 2 switch. The lower layer 2 switch that outputs and receives the switching process start instruction further instructs the lower layer 2 switch or access point from the port currently used for communication of the mobile terminal at that time. Is transmitted to the mobile terminal via the access point that has established a radio link with the mobile terminal at that time. The layer 2 switch that has output the switching process start instruction is in the lower level from the next port to be used. The lower layer 2 switch that outputs the information necessary for the switching advance notice and the subsequent switching to the switch 2 and receives the information necessary for the subsequent switching and the subsequent switching, 2. To the switch or access point, transfer the information required for the next notice and the subsequent switching from the next port to be used, and repeat the transfer notice and the information necessary for the subsequent switching. The mobile terminal that has received the switching process start instruction establishes a radio link with the new access point by causing the access point to be used next to reach the switching advance notice and information necessary for the subsequent switching. Then, the mobility management process according to any one of claims 1 to 3 Until the fact that a new radio link has been established reaches the layer 2 switch that first output the switching start process, sequentially through the ports of several layer 2 switches used for communication of the mobile terminal. It is characterized by transferring.

According to the invention described in claim 4, the mobile terminal is actively involved from the network side with respect to the handover of the radio section, and executes the mobility management according to any one of claims 1 to 3, so that the mobile terminal Can cope with high-speed movement.
The mobile communication network handover control method according to claim 5 is the mobile communication network handover control method according to claim 4, wherein the layer 2 switch receives or first transmits a switching processing start instruction. In addition to receiving the fact of establishment of a new radio link, not only the port that the mobile terminal is using for communication but also the port that is determined to be used next is simultaneously addressed to the mobile terminal. When the mobile terminal changes the access point to which the radio link is established, the mobile terminal receives user frames from the old and new access points at the same time.

According to the fifth aspect of the present invention, it is possible to realize a bicast handover in which user frames are received simultaneously from two new and old access points, so that user frame loss can be reduced.
The mobile communication network handover control method according to claim 6, comprising: a mobile terminal; an access point that provides connection means to the network to the mobile terminal; and a filtering database, the search result of the filtering database being included in a frame In a handover control method for a mobile communication network consisting of a layer 2 switch that forwards the frame to an appropriate port based on layer 2 information to be transmitted, the upper layer 2 switch has a filtering database owned by itself. By using the data related to the layer 2 information of the mobile terminal, the mobile terminal grasps the timing at which the mobile terminal should switch the radio link to a new access point, and the port to be used for the communication of the mobile terminal next. 2 switches If it is determined that the next switching process cannot be completed with only the lower layer 2 switch, at the time, the port used for communication of the mobile terminal and one lower layer 2 switch are connected. A switching process start instruction is output to the upper layer 2 switch, and the other layer 2 switches belonging to the same hierarchy as the layer 2 switch and having a port to be used next are sent to the two corresponding layer 2 switches. The switching notice is transmitted through the horizontal link connecting the switches, and the fact that the new radio link is established is confirmed by the fact that the port of several layer 2 switches used for the communication of the mobile terminal and the The layer 2 switch that first outputs the switching start process via the ports at both ends sequentially, and the layer 2 switch that is one higher Wherein the forwarding to reach the pitch.

According to the invention described in claim 6, since the horizontal link is adopted, it is possible to promptly prompt the change of the access point to which the wireless link is established.
The handover control method for a mobile communication network according to claim 7 is the handover control method according to claim 6, wherein a new radio link is received after the layer 2 switch receives or first transmits the switching process start instruction. The user frame addressed to the mobile terminal is transferred not only to the port that the mobile terminal is using for communication but also to the port that is determined to be used next at the same time until the fact of establishment is received. The mobile terminal receives user frames simultaneously from the old and new access points when the access point to which the radio link is established is changed.

According to the seventh aspect of the invention, it is possible to realize a bicast handover in which user frames are received simultaneously from two new and old access points, so that user frame loss can be reduced.
The mobile communication network handover control method according to claim 8 is the mobile communication network handover control method according to any one of claims 4 to 7, wherein a switching process to be transmitted to the mobile terminal is started. In the instruction, an access point to which a new radio link is to be established is designated, and information necessary for establishing the radio link is notified in advance.

According to the eighth aspect of the invention, since an access point where a new radio link is to be established is designated and information necessary for establishing the radio link is notified in advance, the access point can be switched more quickly.
The handover control method for a mobile communication network according to claim 9 is the handover control method for a mobile communication network according to any one of claims 4 to 7, wherein the switching is performed when the radio link is switched. The mobile terminal that has received the processing start instruction notifies the access point that has established the radio link at that time of disconnection before making a connection request to the new radio link.

According to the ninth aspect of the invention, since the connection release is notified to the access point that has established the wireless link, it becomes possible to quickly delete unnecessary information regarding the mobile terminal in the filtering database.
The handover control method for a mobile communication network according to claim 10 is the old handover control method for a mobile communication network according to any one of claims 4 to 7, wherein the radio link is switched. When the layer 2 switch directly connected to the other access point receives the fact that a new radio link has been established, the layer 2 switch notifies the access point that the disconnection has already been made. And

According to the tenth aspect of the present invention, since it is notified that the connection has already been released, it becomes possible for the old access point to quickly delete unnecessary information regarding the mobile terminal.
The mobile communication network handover control method according to claim 11 is the mobile communication network handover control method according to any one of claims 4 to 7, wherein the filtering database has a data structure. An element representing a layer 2 identifier associated with the mobile terminal, an element representing a port number to which a user frame to which the layer 2 identifier is a destination should be transferred, an element representing a state relating to use of the port, An element for managing port usage time, which means a time when a port is actually used for user frame transfer related to the mobile terminal, an element indicating a timing at which a radio link switching process should be started, and the port The user frame transfer of the mobile terminal until just before being used for the user frame transfer of the terminal. And having an element representing the number immediately before use port means that is used, and a component for managing the validity of the combination of series of elements.

According to the eleventh aspect of the present invention, by using each element of the filtering database, it is possible to incorporate elements for predicting the location change of the mobile terminal in mobility management.
The mobile communication network handover control method according to claim 12 is the mobile communication network handover control method according to claim 11, wherein the filtering database includes, as a data structure, a moving speed of the mobile terminal, It includes elements represented by a number of speed classes classified according to the size, and the speed class is determined by searching a table that defines the correspondence between the past port usage times.

According to the twelfth aspect of the present invention, since mobile terminals are classified into speed classes, it is possible to increase the number of elements for predicting the position change of the mobile terminal and improve the simplicity of the prediction procedure in mobility management.
The mobile communication network handover control method according to claim 13 is the mobile communication network handover control method according to claim 11, wherein the filtering database is configured such that a moving speed of the mobile terminal is defined as a data structure. It includes elements represented by several classes classified according to its size, and the speed class is determined by a declaration from the mobile terminal.

According to the invention described in claim 13, since the mobile terminal is classified into the speed class, and the speed class is reported from the mobile terminal, in the mobility management, the position change prediction element of the mobile terminal is increased, and the prediction procedure is simplified. Can be improved.
The mobile communication network handover control method according to claim 14 is the mobile communication network handover control method according to claim 11, wherein the filtering database includes a port usage time included in a data configuration thereof. The start timing of the next radio link switching process is determined by calculation based on the managed element.

According to the invention described in claim 14, since the start timing of the next radio link switching process is obtained by calculation based on the element for managing the port usage time, the position change prediction of the mobile terminal can be quickly and accurately performed in mobility management. Can be done.
The mobile communication network handover control method according to claim 15 is the mobile communication network handover control method according to claim 11, wherein a past port use time and a next radio link are based on the filtering database. A table that associates the start timing of the switching process is created, and the start timing of the radio link switching process is determined by searching the created table.

According to the fifteenth aspect of the present invention, since the table for associating the past port use time with the start timing of the next radio link switching process is created, the mobile terminal position change prediction can be easily performed in the mobility management.
The mobile communication network handover control method according to claim 16 is the mobile communication network handover control method according to claim 12 or 13, wherein the mobile terminal speed class and next radio link switching are performed. A table for associating processing start timing is created, and the start timing of radio link switching processing is determined by searching the table.

According to the sixteenth aspect of the present invention, since the table that associates the speed class of the mobile terminal with the start timing of the next radio link switching process is created, the mobile terminal position change prediction can be easily performed in mobility management. .
The handover control method for a mobile communication network according to claim 17 is the mobile communication network handover control method according to any one of claims 11 to 13, wherein a plurality of access points are arranged in position. The order relation is matched with the order relation of the ports of the layer 2 switch to which the plurality of access points are connected, and the positional order relation of the plurality of lower layer 2 switches and the plurality of lower layer 2 switches are The port used immediately before the mobile terminal obtained by matching the order relationship of the ports of the upper layer 2 switch to be connected and referring to the filtering database and the port number used for communication of the mobile terminal at a certain point in time And the number of the port to be used for communication of the mobile terminal is grasped.

According to the invention described in claim 17, in the mobility management, the position change prediction of the mobile terminal can be performed quickly and accurately.
In general, in a layer 2 switching network connected to an access point of a wireless LAN, the present invention performs handover processing between a mobile terminal moving at high speed and the network and the accompanying mobility management by exchanging control frames. This is performed in a distributed manner by layer 2 switches. In addition, a filtering database for switching is appropriately configured, and movement prediction of a mobile terminal is performed by a simple method through appropriate information exchange between layer 2 switches. Further, the result of the movement prediction is transmitted from the network side to the mobile terminal through the wireless section, so that the handover process is performed more smoothly.

  In the present invention, it is assumed that the movement of the mobile terminal is a gradual change in the direction of movement and change in the movement speed. Such an assumption is established, for example, when a mobile terminal exists in a train.

According to the first to third aspects of the invention, even when the mobile terminal moves at high speed, accurate mobility management of the mobile terminal is realized by each switch.
According to the invention described in claim 4, since the handover is actively involved from the network side and the mobility management according to any one of claims 1 to 3 is executed, the mobile terminal moves at high speed. Can also deal with.

According to the invention of claim 5, since handover by bicast can be realized, user frame loss can be reduced.
According to the invention described in claim 6, since the horizontal link is adopted, it is possible to promptly prompt the change of the access point to which the wireless link is established.
According to the seventh aspect of the present invention, since handover by bicast can be realized, user frame loss can be reduced.

According to the eighth aspect of the present invention, since an access point where a new radio link is to be established is designated and information necessary for establishing the radio link is notified in advance, the access point can be switched more quickly.
According to the ninth aspect of the present invention, it becomes possible to quickly delete information related to unnecessary mobile terminals in the filtering database.

According to the tenth aspect of the present invention, it becomes possible to quickly delete unnecessary information regarding the mobile terminal at the older access point.
According to the eleventh aspect of the present invention, it is possible to incorporate a mobile terminal location change prediction element in mobility management.
According to the twelfth to seventeenth aspects of the present invention, the position change prediction of the mobile terminal can be performed quickly and accurately in mobility management.

In general, according to the present invention, when a layer 2 switching network is introduced as a relay network for mobile communication accessed by a wireless LAN over a wide geographical area, a handover required for establishing an appropriate communication path is required. Control and mobility management can be efficiently executed in a distributed manner in each layer 2 switch.
Furthermore, by including an element for predicting the movement of the terminal in the mobility management, a handover instruction can be issued from the network side to the mobile terminal, so that the handover processing time can be shortened. At this time, it is possible to reduce the frame loss by adopting a method of temporarily broadcasting the user frame. For this reason, even in a situation where the terminal moves at high speed, there is an effect that it is possible to provide a communication environment with a short communication interruption time.

Hereinafter, embodiments of the present invention will be described. In all the drawings for explaining the embodiments, the same reference numerals are given to those having the same function, and the repeated explanation thereof is omitted. The embodiments described below correspond to all the claims.
FIG. 1 is a diagram showing an overall configuration of a layer 2 switching network (hereinafter simply referred to as a network) for high-speed mobile communication used in the embodiment. As shown in the figure, in this network configuration, the types of switches SW are classified into two. That is, an access switch AccessSW (Access SW1, SW2, SW3) directly connected to the access point AP and a core switch CoreSW (Core SW1, SW2, SW3) positioned hierarchically higher than the access switch AccessSW in the topology. is there. The topology of the connection between these switches SW is a star type having one core switch CoreSW1 as the apex as shown in the figure.

The layer 2 switch described in the claims corresponds to, for example, “access switch AccessSW” and “core switch CoreSW”.
When the mobile terminal MH moves in such a network 10, the access points AP (AP1, AP2,...) Used for connection in the wireless section change with time, and are used for transferring frames related to the mobile terminal MH accordingly. Switches and switch ports also change. In the following description, when this state is expressed, the expression that the destination of the mobile terminal MH changes is used. According to this, it can be said that the mobile terminal MH sequentially switches the belonging destination of the access point AP in the wireless section and the belonging destination of the port of the switch SW of the network 10 as it moves. Such switching of the attribution destination corresponds to handover.

  The reason why the types of the switches SW are distinguished as described above is that the time scales of the handover involved are different. That is, in the access switch AccessSW, this occurs on a relatively short time scale, whereas in the core switch CoreSW, it occurs on a relatively long time scale. Such differences appear as differences in required processing speed and frequency.

  FIG. 2 shows the connection form of the switch SW shown in FIG. 1 including the ports in the switch SW. Each access point AP is connected to the port of the access switch AccessSW on a one-to-one basis. At this time, it is assumed that the numbers given to these ports are integers that monotonously increase (or decrease), such as 1, 2, 3, for example, as illustrated. Such a method of assigning port numbers is very common and natural. Similarly, it is assumed that a port number is assigned to the core switch CoreSW. In FIG. 2, the number of ports of each switch SW is the same, but this is not essential when the present invention is implemented, and may be different.

Further, the number of ports on the uplink side connected to the hierarchically higher switch SW need not be limited to one illustrated.
As to which access switch AccessSW each access point AP is connected to, the positional order relation of the arrangement of the access points AP and the order relation of the connection destination ports are matched. In order to do this, the positional order relationship of the access points AP needs to be clear.

For example, considering the case of performing mobile communication from the mobile terminal MH in the train, the access topology AP is assumed to be arranged along the track, and therefore satisfies the above requirements. Matching the port order relationship with the positional order relationship in this way is also applied to the connection between the access switch AccessSW and the core switch CoreSW.
As can be seen from the above description of the connection form, when the mobile terminal MH moves without abruptly changing the traveling direction, the belonging port of the mobile terminal MH changes sequentially with time. Moreover, the change of the port number at that time becomes a monotonous increase (or decrease).

The switch fabric 21 (21a, 21b, 21c) illustrated in FIG. 2 is a functional block that executes frame switching between ports. The filtering database 22 (22a, 22b, 22c) is a database that is referred to when this is performed.
The filtering database 22 is also used in a normal Ethernet (registered trademark) switch (Ethernet-SW) or the like. In the Ethernet (registered trademark) switch, the filtering database 22 is dynamically constructed by acquiring the MAC address of the transmission source included in a valid Ethernet (registered trademark) frame received at each port. The frame here refers to a frame that carries user traffic (hereinafter referred to as a user frame).

  In the switch SW of the present embodiment, when a frame transmitted from the access point AP is received by a port of the access switch AccessSW, or a frame transmitted from the hierarchically lower switch SW is present in the upper switch SW. When received at the port, the layer 2 identifier and the reception port number included in the frame are associated with each other and registered in the filtering database 22. Conversely, the layer 2 identifier to be used is not given from the network toward the mobile terminal MH.

The “update of the filtering database related to the physical address unique to the mobile terminal” described in claim 1 and the “update of the filtering database related to the layer 2 label” described in claims 2 and 3 are the same as those described above. Corresponds to “Register 22”.
The point that reception of a frame triggers registration in the filtering database 22 is a processing method similar to the MAC address acquisition method in the above-described Ethernet (registered trademark) switch, but is transmitted not only in the user frame but also in handover. The specific control frame is also used for registration. Further, the components of the filtering database 22 are also defined differently from those in the Ethernet (registered trademark) switch or the like in order to perform mobility management including the concept of handover prediction.

  Here, the layer 2 identifier used for registration in the filtering database 22 will be described. This may be a hardware specific MAC address, or a label such as a VLAN ID. In general, since the number of bits required for the expression of the label is smaller, the use of this has the advantage that the processing load is reduced when there is no problem in scalability.

  When using a MAC address, a unique address assigned to the mobile terminal MH is used. On the other hand, when a label is used, it may be fixedly given to the mobile terminal MH in advance, or the access point AP newly assigned to at the time of handover is associated with the MAC address of the mobile terminal MH each time. May be. In other words, the use range of the label may be included up to the mobile terminal MH (Claim 2), or may be limited to the access point AP (Claim 3).

The “physical address unique to the mobile terminal” described in claim 1 corresponds to, for example, a MAC address.
In addition to the MAC address unique to the mobile terminal, there is a method of providing another MAC address for encapsulation in the layer 2 frame. However, the MAC address for encapsulation should be classified as a label. is there. In the following description, since there is no difference depending on whether a hardware-specific MAC address or label is used as a layer 2 identifier, no particular limitation is imposed.

In this way, each switch SW itself constructs the filtering database 22, and based on that, the necessary handover under the switch SW and the mobility management process associated therewith are performed. That is, this can be regarded as distributed processing by each switch SW. The details will be described below.
First, a processing procedure related to handover will be described. FIG. 3 shows a control method for handover under the access SW. FIG. 3 shows an example in which only the unicast method is used without using the bicast method.

  As shown in FIG. 3A, it is assumed that the mobile terminal MH belongs to a certain access point AP and a port of the access switch AccessSW to which the access point AP is connected at a certain time. At this time, the user frame whose source or destination is the mobile terminal MH is transferred through the access point AP and the port. In FIG. 3A, this state is indicated by a thick solid line. Thereafter, with the passage of time, the mobile terminal MH moves in the illustrated moving direction. The access switch AccessSW knows the direction of this movement and the time when the next access point AP and port should be handed over. The method of grasping will be described later, but the description will be continued assuming this.

  As shown in FIG. 3B, the access switch AccessSW transmits a control frame called a reassociation trigger to notify the access point AP when the time to start the handover process is reached. To do. The access point AP that has received this control frame further transmits a control frame called a reassociation order to the mobile terminal MH.

  The mobile terminal MH that has received this determines that the handover process in the radio section should be started, and changes to an access point AP that is different from the access point AP that has been used until now, usually an access point AP located next to it. Request attribution. At this time, the mobile terminal MH may search for information on the next access point AP to be used on its own (claims 4 to 7) or use it next in a control frame of a reassociation order (Reassociation Order). The mobile terminal MH may be instructed including information on the access point AP to be used (claim 8). In general, the process of searching for information on a new access point AP takes time, so the method of omitting this as in the latter is preferable for realizing a smoother handover. However, in order to execute the latter method, for example, the moving direction of the mobile terminal MH is included in the control frame of the reassociation trigger (Reassociation Trigger), and the access point AP that has received the moving method uses the moving direction next. A procedure is required in which the access point AP to be determined is determined, and information on the access point AP is transferred to the mobile terminal MH using a control frame of a reassociation order. Alternatively, when the access switch AccessSW holds all the information of the access point AP under its control, the information of the access point AP to be used next may be included in the control frame of the reassociation trigger (Reassociation Trigger). Good. In any case, some devices on the network side need to share information on the access point AP. Hereinafter, such a method will be described without particular limitation.

  In FIG. 3C, the mobile terminal MH that has obtained the information of the access point AP to be used next sends a control frame of an association request (Association Request) to the access point AP, thereby giving a new attribution. Request. On the other hand, the access point AP that has received the association request returns a control frame of an association response to the mobile terminal MH. This control frame has a meaning of success if the request can be accepted, and failure if the request cannot be accepted. Note that a normal wireless LAN often includes a so-called authentication process at the time of such handover, but this may or may not be included in the present invention. Hereinafter, in order to simplify the explanation, the entire authentication process is omitted.

In the process in FIG. 3C described above, it is assumed that the physical communication means between the mobile terminal MH and the new access point AP has been established. However, at this stage, since the handover including layer 2 has not yet been completed, user frame transmission / reception cannot be executed. In FIG. 3C, this state is indicated by a dotted line.
On the other hand, when the new access point AP accepts the attribution request and returns a control frame of the association response (Association Response) with the content of success to the mobile terminal MH, the handover of the radio section is completed at this stage. Is ready for user frame transmission / reception. Further, the new access point AP transmits a control frame of a recomposition trigger (Recomposition Trigger) toward the access switch AccessSW, and the access switch AccessSW that has received this control frame updates the contents of the filtering database 22 described later. As a result, the mobile terminal MH updates the belonging relationship regarding the port of the access switch AccessSW, thereby completing the handover process. Thereafter, the user frame is correctly transferred using the new port and the access point AP. In FIG. 3D, the state where the handover process is completed is indicated by a solid line.

Note that “transmit the fact from the new access point to the layer 2 switch to which it is connected” described in claims 1 to 3 corresponds to, for example, a “recomposition trigger control frame”. .
Further, the “update of the filtering database” described in claims 1 to 3 corresponds to, for example, the “update of the filtering database 22” described above.

Further, the “switching process start instruction” described in claims 4 to 9 corresponds to, for example, a “reassociation trigger control frame”.
In FIG. 3, in the middle of the handover shown in (b) and (c), the user frame addressed to the mobile terminal MH continues to be sent via the old port and the old access point AP. That is, after the mobile terminal MH receives the control frame of the reassociation order and starts the handover process in the radio section, the switch SW receives the control frame of the recomposition trigger and receives the filtering database 22. Until this is rewritten, there is a frame loss.

In order to prevent this, a method called bicast is effective. This is a special form of multicast that uses two specific ports simultaneously for forwarding user traffic. A handover processing procedure using this is shown in FIG.
In FIG. 4, the control frame transmission procedure related to the handover and the contents included therein are the same as those shown in FIG. The difference is that, as shown in FIG. 4B and FIG. 4C, a bicast state is temporarily provided in the middle of processing the handover (as shown in the figure). , Thick line coming out of two ports). As shown in FIG. 4D, when all the handover processes are completed, the original unicast state is restored.

  When such a bicast method is used, it is assumed that the mobile terminal MH can simultaneously communicate with two access points AP. Because, if the mobile terminal MH does not communicate with the older access point AP after starting the handover process, even if the user frame is delivered from the network side to the access point AP, the mobile terminal MH This is because there is no frame transfer means. FIG. 4C shows a state where user frames are being transmitted / received to / from the older access point AP at the same time as handover processing is proceeding with the new AP.

5, 6, and 7 are sequence diagrams illustrating the control frame transmission / reception method related to the handover under the access switch AccessSW illustrated in FIG. 3 or 4.
5, 6, and 7, a solid bidirectional arrow represents a state in which the user frame is normally transferred. A dashed one-way arrow represents a control frame. In FIGS. 5, 6, and 7, each control frame is represented by one broken line arrow. However, these broken line arrows may actually transmit the same content frame a plurality of times.

  As described above, the access switch AccessSW predicts the start of handover and transmits a control frame of a reassociation trigger (Reassociation Trigger) to the current access point AP, and then the access switch AccessSW recomposes from the new access point AP. Until the control frame of the trigger (Recompositon Trigger) is received, it is the time during the handover process. In FIG. 5, FIG. 6, and FIG. 7, this time zone is shaded.

In FIGS. 5, 6, and 7, shading is added to two time zones. In this case, the shaded area in the first time zone indicates the handover process from the access point AP1 to the access point AP2, and the shaded area in the second time zone indicates the handover process from the access point AP2 to the access point AP3.
The difference between FIG. 3 and FIG. 4 is the difference between unicasting or bicasting in the time zone during the shaded handover, as described above. However, this difference is not clearly shown in FIGS.

  FIG. 5 shows a case in which the mobile terminal MH receives a request to start handover from the current access point AP1 (or AP2), that is, a control frame of a reassociation order (Reassociation Order). The control frame of “Response” is returned to the current access point AP1 (or AP2), which is different from FIGS. The control frame of the reassociation response means that the mobile terminal MH applies for the cancellation of the belonging relationship to the access point AP1 (or AP2) used so far. After this, a request for belonging to the next access point AP2 (or AP3) is made.

The “notification of cancellation” described in claim 9 corresponds to, for example, a “reassociation response control frame”.
In FIG. 6, the mobile terminal MH does not return a reassociation response control frame to the access point AP1 (or AP2) in order to cancel the belonging relationship. Instead, when the access switch AccessSW1 finally receives the control frame of the recomposition trigger (Recomposition Trigger) and the handover is completed, the access switch AccessSW1 belongs to the old access point AP1 (or AP2). A control frame for disassociation notification, that is, disassociation notification is sent. Since the transmission of the control frame should be performed after the handover process is completed, the control frame is shown outside the shaded time zone in FIG.

The “notification that connection has already been released” described in claim 10 corresponds to, for example, a “disassociation notification control frame”.
Further, “the fact that a new radio link is established” described in claim 10 corresponds to a “recomposition trigger control frame”.

  In FIG. 7, as in FIG. 5 and FIG. 6, it is not explicitly instructed to release the belonging relationship by the control frame. Here, it is assumed that due to aging, the belonging relationship that has lost its meaning will be automatically deleted. Aging is a process of erasing old data that has been confirmed to be unnecessary or wrong, and is also used in a normal Ethernet (registered trademark) switch (Ethernet-SW) or the like. According to this, normally, if a valid frame is not received for a certain period, data related to it is erased.

  One of the main features of the handover processing procedure described above is that, when the mobile terminal MH moves and changes the access point AP, the transmission / reception of the control frame does not cause a large time shift. The filtering database 22 of the access switch AccessSW is updated. That is, in a normal operation state, a handover in the radio section and a handover in the relay network section configured by the access switch AccessSW are always performed in pairs. As a result, it is possible to realize mobility management that reliably grasps the position of the MH at the time when the user frame addressed to the mobile terminal MH needs to be transmitted.

  Another main feature is that the network side instructs the mobile terminal MH to start handover. As will be described later, this is performed by introducing an element of prediction into mobility management. By such a procedure, it is possible to shorten the handover processing time in the radio section. This is because, although it generally takes time for the mobile terminal MH to determine the start of handover in the radio section, this time can be omitted.

  However, when implementing the present invention, considering that there is a possibility that “prediction is not always accurate” or “prediction is impossible in the first place”, the mobile terminal does not depend on an instruction from the network side. It does not exclude the handover start procedure based on the decision of the MH itself. That is, the start of the handover process is allowed both when it is caused by an instruction from the network side and when it is caused by the judgment of the mobile terminal MH itself.

However, the processing procedure performed as a result, that is, from the transmission of the control frame of the association request (Association Request) to the new access point AP from the mobile terminal MH to the reception of the control frame of the recomposition trigger (Recomposition Trigger) at the access switch AccessSW The processing procedure is the same for both.
The handover procedure under the access switch AccessSW has been described above. Actually, a handover that crosses between the access switches AccessSW or between the core switches CoreSW is necessary via the core switch CoreSW that is hierarchically higher. There is a case. The procedure in such a case will be described later. Here, first, the data configuration of the filtering database 22 will be described.

  8 and 9 are diagrams showing the data structure of the filtering database 22. The combination of the layer 2 identifier and the port number common to all of these is the most basic and essential component of the filtering database. The port states shown in FIGS. 8 and 9 are newly introduced in this embodiment in order to include a prediction element in mobility management.

Note that the “data relating to the layer 2 information of the mobile terminal in the filtering database owned by itself” described in claims 4 and 6 corresponds to the data configuration shown in FIGS. 8 and 9, for example.
Further, the handover control method described in claim 11 is realized by using, for example, the data configuration shown in FIGS.

FIG. 10 is a diagram illustrating an example of the port state definition method illustrated in FIGS. 8 and 9. In FIG. 10, two states of running and previous are defined as port states. In FIG. 10, Port (k−1), Port k, and Port (k + 1) indicate ports to be handed over.
As described above, the handover is completed when the new attribution destination is determined along with the movement of the mobile terminal MH, and at this time, a control frame of a recomposition trigger is transmitted to the port of the new attribution destination. When the access switch AccessSW correctly receives the control frame of the recomposition trigger (Recomposition Trigger), the access switch AccessSW transits to the running state. In other words, when the handover is completed and it is confirmed that the mobile terminal MH belongs to the own port, the port enters a running state. Through these procedures, in a normal operation state, when the mobile terminal MH moves and performs a handover, the port of the access switch AccessSW changes to a running state without a large time shift. To do.

  When the mobile terminal MH further moves, another port becomes a new destination of the mobile terminal MH, and the handover is completed. At this time, the state of this new belonging port becomes a running state, and the state of the original port becomes a previous state. The database component associated with the port in the previous state is used for the start prediction of the handover described below, and is not used for switching. Therefore, if it can be confirmed that this port is not used for communication of the mobile terminal MH for a certain period of time, the set of layer 2 identifiers and this port number, and the database components associated therewith are all deleted. Managing this process is an element that has been shown to be effective and is used for the aging data deletion process.

  While the mobile terminal MH belongs to a certain access point AP and further to the port of the access switch AccessSW ahead thereof, all user frames transmitted from this mobile terminal MH are received at this port. The component represented as “effectiveness” indicates a result of monitoring whether or not such a user frame is received within a certain time period. That is, if the frame is received even once during this time period, the validity can be confirmed.

In a normal operation, since a port in the previous state does not receive a user frame, the validity cannot be confirmed in the time period of validity confirmation and is deleted from the filtering database 22.
Here, it is assumed that the time period described above is sufficiently larger than the time interval of the handover that occurs as the mobile terminal MH moves. As a result, it is possible to prevent the filtering database 22 updated upon completion of the handover process from being deleted before the next handover occurs by the validity check process.

  However, even when such a time setting is performed, for example, when the mobile terminal MH stops moving in a state belonging to a certain port, and the validity check time period has passed without transmitting any frame, The validity cannot be confirmed even in a port in a running state. As a result, the data itself is directly deleted without going from the running state to the previous state as the port state. If there is no data, a frame to be transmitted to the mobile terminal MH cannot be identified as a port for transmission, and is therefore broadcast to all ports or discarded (this depends on the network operation policy). ). After that, when a user frame is transmitted from the mobile terminal MH and received at the port, or when the handover process is started again and a recomposition trigger is received, the port state is re-running. ) Return to the state.

  As described above, both the reception of the recomposition (Recomposition Trigger) at the completion of the handover and the reception of the user frame are permitted as the trigger for the port state to be in the running state. In addition, there is no restriction on the port state before entering the running state. On the other hand, in order to be in the previous state, it is a necessary condition that the previous port state is always in the running state.

  In FIG. 8 and FIG. 9, the component indicated as the previous used port number indicates that the port state of the port is in the running state at the same timing as the running state from the running state to the previous state. This means the port number changed to. That is, it is a port used for user frame transfer until just before. However, these two state changes do not always occur simultaneously. As described above, when the port used until immediately before is unknown, the port number is unknown.

When the previously used port number is correctly recognized, the port number to be used next can be determined by comparing this with the port number. As described above, this is because the positional order relationship of APs matches the order relationship of port numbers.
Note that the handover control method according to claim 17 is realized by using, for example, the port number determination method described here.

  The port use start time in FIG. 8 means a time when the port state of the port becomes a running state. The port use end time means a time when the port state of the port becomes a previous state. Therefore, the time corresponding to the difference between these times corresponds to the time when the port is in the running state and used for transferring the user frame. Hereinafter, this is referred to as port usage time.

  As shown in FIG. 8, it may be inappropriate to manage the port usage time based on the absolute time. In such a case, this can be substituted by measurement by a counter. FIG. 9 shows the structure of data when measured by the counter. The counter is set when the port is in a running state, and then the counter value is incremented according to the counter cycle. When the state finally reaches the previous state, the count is terminated, and the counter value at that time is used as the port usage time. In this case, the port usage time causes a measurement error of about twice the counter cycle at the maximum, but there is an advantage that the processing is generally simplified rather than using the absolute time.

  FIG. 11 is a diagram illustrating the relationship between the port usage time when a handover is performed under the control of one access switch AccessSW and the port state defined in FIG. In FIG. 11, Port (k-1), Port k, and Port (k + 1) indicate ports to be handed over. In FIG. 11, transmission of a control frame for a reassociation trigger (Reassociation Trigger) indicating start of handover and reception of a control frame for recomposition (Recomposition Trigger) indicating completion of handover are shown. As shown in FIG. 11, the control frame transmission of the reassociation trigger (Reassociation Trigger) is executed when the time or counter value managed in the access switch AccessSW reaches the handover prediction timing. Then, the management method of this predicted value is described next.

  12A and 12B are conceptual diagrams of a handover start timing prediction method. As described above, the port usage time is recorded in the port whose port state is the previous state. Although not explicitly shown in the database configuration shown in FIG. 8, it can be easily calculated by taking the difference between the port use end time and the start time. In the data corresponding to the port in the running state, if the previous port number is specified in the data structure, this indicates the port in the previous state that was used until immediately before. . Therefore, the port usage time recorded in this port can be referred to. Further, when it is assumed that the movement speed of the mobile terminal MH changes or the movement speed changes slowly (for example, when the mobile terminal MH exists in the train), the port usage time immediately before this is used, The usage time of the port currently in use can be estimated in advance by simple calculation. However, in order for the estimation by such calculation to be effective, it is necessary to add that the geographical range covered by one access point AP is not too large as a precondition. If both of the two conditions are satisfied, if the coverage of one access point AP (the distance along the direction of the mobile terminal MH when the direction of movement of the mobile terminal MH is determined) is almost the same, the mobile terminal MH It is reasonable to assume that the time it takes to pass this will not change significantly. Such an assumption is the basis for the above-mentioned "by simple calculation in advance".

  In this way, the usage time is estimated in advance, and by subtracting the estimated time required for the handover from the estimated usage time, the timing for starting the handover process from the currently used port to the next port to be used can be determined. It is determined. This timing may be represented by a counter value or a time. Here, the method of setting a fixed value in advance for the “time assumed to be required for handover” is the simplest. Alternatively, if the processing is allowed to be complicated, it is possible to include an element of prediction in the same way for the “time required for handover”. In other words, for a port in the previous state, the time required for the previous handover can be determined by taking the difference between the port usage time and the handover prediction timing. It is a technique of using.

Note that “determining the start timing of the next radio link switching process by calculation” described in claim 14 is realized by, for example, a timing prediction method illustrated in FIG. 12.
FIGS. 13A and 13B are conceptual diagrams illustrating another method for determining the handover prediction timing. The concept of FIG. 13 is similar to the concept of FIG. 12 in that the data of the previously used port in the previous state is referred to, but the timing of handover start is not calculated but calculated. The difference is that the search is performed using a special table prepared for this purpose (hereinafter referred to as a prediction table).

It should be noted that the “table for associating the past port usage time with the start timing of the next radio link switching process” described in claim 15 and the “table of the mobile terminal speed class and the next radio link switching process” described in claim 16 are described. The “table for associating timing” corresponds to, for example, the “prediction table”.
FIG. 13 shows a method of referring to another component called a speed class instead of the port usage time of the previous used port as a method of searching the prediction table. The data structure of the filtering database 22 when this is done is shown in FIGS.

Note that the handover control methods described in claims 12 and 13 are realized by using, for example, the data structures shown in FIGS.
The speed class is a classification of the mobile terminal MH into several classes according to the speed of movement. One method of determining which speed class belongs is to use the port usage time (counter value or time display) as a criterion. In order to execute this, the access switch AccessSW needs to have a mapping table for associating the port usage time with the speed class. This mapping table is set in advance from the geographical information of the arrangement of the access points AP.

Note that the “table defining the correspondence between the past port usage time and it” described in claim 12 corresponds to, for example, the mapping table described above.
Another method of determining which speed class belongs is that the mobile terminal MH transmits the speed class to the access switch AccessSW. This can be executed only when the mobile terminal MH can measure its own speed. In order to transmit the speed from the mobile terminal MH to the access switch AccessSW, the control frame of the association request (Association Request) and the control frame of the recomposition trigger (Recomposition Trigger) may be used in combination. Frames may be used.

Note that the “speed class is reported from the mobile terminal” described in claim 13 corresponds to, for example, a combination of the control frames described above.
In this manner, when the speed class of the port used immediately before can be referred to, the next handover start timing prediction can be executed in the form of searching the prediction table. Note that the essence here is that the prediction is performed not by calculation but by table search, so as long as this is satisfied, it is not always necessary to separately provide an element such as a speed class.

16 (a) and 16 (b) and FIG. 17 (a) and FIG. 17 (b) show not only the port used immediately before in order to improve the accuracy of the prediction method shown in FIG. 12 and FIG. It shows the method of referring to the data of the port used previously. Thus, by using two reference destinations, there is an advantage that an element of change in movement of the mobile terminal MH can be included in the prediction.
However, in order to increase the number of reference destinations in this way, in addition to the previous state, another port state indicating a past state is required. FIG. 18 is a diagram illustrating an example of the port state definition method illustrated in FIGS. 14 and 15, and here, a 2 previous state is newly defined. Just as the condition for transitioning to the previous state is the running state just before that, in order to transition to the 2 previous state, the previous state must be in the previous state. There must be.

  If the idea shown in FIG. 16, FIG. 17, and FIG. 18 is developed, a method of holding the state further back in the past and referring to it can be considered. This is easily possible by extending the method described above. However, if too many such past states are defined, a disadvantageous factor arises in that the scale of the database to be held increases and the processing load increases. Here, the extent to which the past state should be retained is not limited, but in the following description, for the sake of simplicity, it is assumed that only one previous state, that is, the previous state is managed.

  FIG. 19 newly defines a transitional state as an intermediate state between the running state and the previous state of the port state definition shown in FIG. FIG. 20 shows the relationship between the port state, port usage time, and handover prediction timing in this case. As shown in FIG. 20, the transitional state is the time from when the port of the access switch AccessSW transmits the control frame of the reassociation trigger (Reassociation Trigger) until the reception of the recomposition trigger (Recomposition Trigger). Corresponds to a belt. That is, it is a time zone during which the handover process is executed.

  As shown in FIG. 19 and FIG. 20, there is only one port in the running state and the previous state for the same layer 2 identifier, whereas in the transitional state. There are two old and new ports related to handover at the same time. Therefore, if transmission of a user frame from a port in a transitional state is permitted, a bicast state can be temporarily created. Conversely, in order to adopt the bicast method as shown in FIG. 4, it is necessary to introduce a transitional state.

Note that the handover control method described in claims 5 and 7 is realized by the port state definition method shown in FIGS. 19 and 20, for example.
However, the purpose of introducing such an intermediate state is not necessarily limited to the realization of bicast. For example, when the handover is completed, the combination of the layer 2 identifier and the port number and the data component of the filtering database 22 to be associated with the combination are not prepared, but these are prepared in advance when the handover is started. It may be desirable to create a data component.

  For this purpose, an intermediate state is introduced and bicasting is not performed. Instead of a transitional state, instead of a transitional state, as shown in FIG. 21, a handling over state and taking There is also a method of defining an over (taking over) state. The handing over state is an intermediate state taken by the port that was previously used in the handover process, while the taking over state is the intermediate state taken by the newly used port. is there. FIG. 22 shows the relationship between the state based on these definitions, the port usage time, and the handover prediction timing. When unicast is performed in the middle of the handover process, it is only necessary to transfer the user frame only to the port in the handing over state.

  In the port state definition method shown in FIG. 21, the state of the old and new ports is explicitly distinguished, but it is also possible to perform unicast while adopting a transitional state as shown in FIG. It is. In this case, referring to the previous port number existing in the data structure of the two ports related to the handover, one of the previous port numbers should match the number of the other port. However, the case where the previous port number is unknown is excluded. On the other hand, the other previous used port number should point to another port that is not involved in this handover process, so considering these, it is possible to determine which port will be the older or newer port. It is possible to do. Based on this, unicast is performed only on the older port. On the other hand, as shown in FIG. 21, even when the handing over state and the taking over state are explicitly distinguished, the user frame is transferred to the ports in these two states. If this is allowed, bicast can be easily realized.

  In the above description, a method of mobility management including handover that occurs under the control of one access switch AccessSW and prediction associated therewith has been described. In the following description, a procedure will be described in the case where a handover process across two access switches AccessSW is required via the core switch CoreSW that is one level higher in the hierarchy. Here, as an example of such a case, a case where a handover occurs between the access point AP3 under the access switch AccessSW1 and the access point AP4 under the access switch AccessSW2 in FIG.

Note that the handover control method described in claim 4 is realized by, for example, a handover processing procedure described below.
FIG. 23 shows a handover control method in such a case. In FIG. 23, thick solid lines and broken lines are used in the same meaning as in FIG.
In FIG. 23A, the mobile terminal MH belongs to a certain access point AP, the port of the access switch AccessSW to which it is connected, and the port of the core switch CoreSW to which this access switch AccessSW is connected at a higher level. Suppose you are. As the mobile terminal MH moves, handovers under the access switch AccessSW are sequentially executed, but at a certain point in time, a handover is required to change the belonging destination to another access switch AccessSW. FIG. 23B shows a point in time when such a handover is started.

  The start of such a handover is instructed by the core switch CoreSW. Even in the access switch AccessSW, since mobility management including prediction is performed by the method described so far, it can be recognized that a handover to the next port should be started, but the next port is connected to the own access switch AccessSW. When recognizing that it does not belong, it waits for an instruction from the upper core switch CoreSW. A method for recognizing the predicted timing of handover start in the core switch CoreSW will be described later.

  In FIG. 23B, when the core switch CoreSW determines that the handover is started, first, a control frame for a reassociation trigger is transmitted to the access switch AccessSW. Receiving this, the access switch AccessSW interprets the contents of the control frame, and sends the control frame of the reassociation trigger (Reassociation Trigger) from the port to which the mobile terminal MH that should perform the handover belongs to the access point AP. Send. Receiving this, the access point AP transmits a control frame of a reassociation order to the mobile terminal MH. After that, the handover process of the wireless section is completed, and the new access point AP transmits a recomposition trigger control frame to the port of the new access switch AccessSW. The process so far is the same as the processing procedure in FIG. Next, when the control port of the recomposition trigger (Recomposition Trigger) is received by the new port, the control frame of the recomposition trigger (Recomposition Trigger) is further transmitted to the port to which the upper core switch CoreSW belongs. . As a result, the filtering database 22 of the core switch CoreSW is updated, and all handover processes are completed. The processing so far is shown in FIG. Thereafter, as shown in FIG. 23D, the user frame is transferred using the port of the new access switch AccessSW and the new access point AP.

  In the description of FIG. 23B, the access switch AccessSW that is the destination of the control frame of the reassociation trigger (Reassociation Trigger) transmitted from the core switch CoreSW is currently used as the destination of the mobile terminal MH. And the access switch AccessSW to be a new destination. At this time, it is assumed that the contents of the frame include data necessary for configuring the filtering database of the new access switch AccessSW. More specifically, the predicted timing for performing the next handover, the port number used before, and the like. This means that necessary data is transmitted in the form of a control frame via the core switch CoreSW because the access switch AccessSW does not share a filtering database as a prediction material.

Note that “repeating transfer of a switching process start instruction” described in claim 4 corresponds to repeating transfer of a “reassociation trigger control frame”, for example.
Further, “information required for switching notice and subsequent switching” described in claim 4 corresponds to, for example, a “reassociation trigger control frame”. Further, “the fact that a new radio link has been established” described in claim 4 corresponds to, for example, a “recomposition trigger control frame”.

  FIG. 24 shows that a state in which bicast is temporarily performed is provided in the middle of the handover process. This is due to the same method as the introduction of bicast in FIG. The difference from FIG. 4 is that bicasting is also performed in two links between the core switch CoreSW and the old and new access switches AcessSW (FIGS. 24B and 24C). Such a bicast state occurs at the same timing as when a reassociation trigger is transmitted in each access switch AcessSW. Finally, when the filtering database 22 of all access switches AcessSW is correctly updated, the original unicast state is restored.

Note that the handover control method described in claim 5 is realized by, for example, the handover processing procedure described in FIG.
FIG. 25 is a sequence diagram showing a control frame transmission / reception method related to the handover control shown in FIGS. However, in FIG. 25, in order to focus on the exchange between the core switch CoreSW and the access switch AccessSW, the sequence from the access switch AccessSW to the mobile terminal MH is simplified. In this range, any one of the sequences shown in FIGS. 5, 6, and 7 is actually executed.

  A solid bidirectional arrow and a broken unidirectional arrow in FIG. 25 have the same meanings as in FIG. 5 and the like, and represent user frame transfer and control frame transfer, respectively. A shaded portion means a time zone during the handover process. The difference between the case of performing the unicast in FIG. 23 and the case of performing the bicast in FIG. 24 corresponds to the user frame transfer method in this time zone, but this is not explicitly shown in the figure.

  In addition to these, an arrow indicated by an alternate long and short dash line represents transfer of a control frame of SW advertisement (SW Advertisement) exchanged between mutually connected access switches AcessSW. A control frame of SW advertisement (SW Advertisement) is periodically transmitted, and has a role of transmitting the contents of the filtering database 22 owned by itself to another access switch AcessSW. Here, the periodic transmission does not necessarily mean that transmission is performed at a strictly constant time interval, as long as transmission can be performed at a certain time interval. This transmission time interval should be determined in consideration of the frequency of handovers, etc., so it is not specified here, but the band used for user frame transfer may be greatly eroded. The premise is that there is no.

Note that the control frame of SW advertisement (SW Advertisement) is “transmitting information of a necessary part of its own filtering database at a certain time interval in both directions or in one direction” described in claims 1 to 3. Corresponds to expression.
FIG. 25 shows a state in which a control frame for SW advertisement (SW Advertisement) is transferred in one direction from the access switch AcessSW in the lower hierarchy to the access switch AcessSW in the upper hierarchy. However, this is because, in the following description, the upper access switch AcessSW only needs to have information on the lower access switch AcessSW. However, in the implementation of the present invention, the control frame for SW advertisement (SW Advertisement) may be transferred in both directions or in one direction from the lower access switch AcessSW to the upper access switch AcessSW.

  Here, the filtering database 22 held by the core switch CoreSW is configured in the same manner as described in FIG. 8, FIG. 9 or FIG. However, it is allowed that the configuration is different between the access switch AcessSW and the core switch CoreSW. For example, the access switch AcessSW may have the configuration of FIG. 9 that displays the port usage time by the counter value, and the core switch CoreSW may have the configuration of FIG. 8 that displays by the absolute time. In addition, when the counter value is used in both, it is possible to provide a difference in the counter cycle. Such matters are determined in consideration of the frequency of handover occurring under the access switch AcessSW and the frequency of handover occurring under the core switch CoreSW. This is because the mobility management at the core switch CoreSW has a larger allowable value as the processing time as it is positioned higher in the network topology hierarchy.

  Further, in FIG. 25, the contents of the control frame of SW advertisement (SW Advertisement) do not have to describe the entire contents of the filtering database 22. For example, if it can be determined that it is not necessary to manage information of a port whose port state is a previous state by the upper access switch AcessSW, it is not necessary to send them in a frame. In this way, only information necessary for the handover prediction of the upper core switch CoreSW is periodically transmitted from the lower layer. The core switch CoreSW updates its own filtering database 22 from such information and makes an appropriate handover prediction based on it. When the handover prediction timing is reached, a reassociation trigger is transmitted to the lower access switch AcessSW.

  In FIG. 25, when the core switch CoreSW receives the control frame of the recomposition trigger (Recomposition Trigger) transmitted from the lower access switch AcessSW at the final stage of the handover, the recomposition is transmitted from here to the old access switch AcessSW. Trigger (Recomposition Trigger) is returned. This is to notify that the mobile terminal MH is no longer belonging to the old access switch AcessSW. The access switch AcessSW that has received this can further transmit a control frame for disassociation notification to the old access point AP, similarly to the operation in FIG. However, these procedures may be omitted. This is because each access switch AcessSW has a process of deleting unnecessary data due to aging, so that it can be expected that data will be deleted.

As described above, the handover based on the network configuration of FIG. 1 and the procedure of mobility management associated therewith are shown. In the following, these procedures based on another network configuration will be described.
Note that the handover control method described in claim 6 can be realized by, for example, a handover processing procedure described below.

  FIG. 26 is a diagram showing an overall configuration of the network 20 used for mobile communication. FIG. 26 is a layer 2 switching network based on a star topology basically having one core switch CoreSW1 as a vertex, as in FIG. 1 differs from FIG. 1 in that there is a link connecting between access switches AcessSW belonging to the same hierarchy on the topology. Hereinafter, this is referred to as a horizontal link.

Such a lateral link is not used for normal user frame forwarding. This is to prevent a loop from being formed topologically. The use of horizontal frames is limited to the transfer of control frames exchanged during handover.
FIG. 27 shows a handover procedure when a horizontal link exists. Similarly to FIG. 23, this corresponds to the case where two access switches AcessSW are involved via the core switch CoreSW, and bicasting is not performed.

  As described with reference to FIG. 23, when there is no horizontal link for control frame transfer, the handover start determination is performed by the upper core switch CoreSW. On the other hand, in FIG. 27, this determination is performed by the access switch AcessSW to which the mobile terminal MH currently belongs. At the start of handover, not only the control frame of the reassociation trigger is transmitted to the subordinate access point AP, but also to the upper core switch CoreSW using the normal link, and the horizontal link A control frame of a reassociation trigger (Reassociation Trigger) is transmitted to the access switch AcessSW to be used next. This state is shown in FIG.

  Thereafter, when the mobile terminal MH receives a control frame of a reassociation order sent from the access point AP, handover in the radio section is started. When this is completed, a recomposition (Recompositon Trigger) control frame is sent from the access point AP to the new belonging access switch AcessSW. The procedure so far is the same as described above. The new access switch AcessSW transmits a control frame of a recomposition trigger (Recompositon Trigger) to the upper core switch CoreSW. Thereby, the necessary update of the filtering database 22 is completed, and the handover is completed. FIGS. 27B and 27C show this state.

  Here, as shown in FIG. 27C, a control frame of a recomposition trigger (Recompositon Trigger) is transmitted from the new access switch AcessSW to the old access switch AcessSW. A horizontal link is used for control frame transfer between the access switches AcessSW at this time. This has the significance of explicitly indicating that the mobile terminal MH no longer belongs to the older access switch AcessSW.

Note that “the fact that a new radio link has been established” recited in claim 6 corresponds to, for example, a “recomposition trigger control frame”.
FIG. 28 shows a procedure for performing bicasting in the same handover as in FIG. When each access switch AcessSW receives the control frame of the reassociation trigger (Reassociation Trigger), it starts bicasting under the respective subordinates, and then receives the control frame of the recomposition trigger (Recompositon Trigger). Return to the cast state. Other procedures are the same as those in FIG.

Note that the handover control method described in claim 7 is realized by the handover processing procedure described in FIG. 28, for example.
FIG. 29 is a sequence diagram showing a control frame transmission / reception method related to the handover shown in FIG. 27 and FIG. As in FIG. 25, in FIG. 29, since the focus is on the exchange between the core switch CoreSW and the access switch AcessSW, the sequence in the range from the access switch AcessSW to the mobile terminal MH is simplified. However, in practice, any one of the sequences shown in FIGS. 5, 6, and 7 is executed in this portion. The meanings of the solid bidirectional arrow and the broken unidirectional arrow in FIG. 29 are the same as those in FIG.

  In FIG. 29, it is the access switch AcessSW that has been used so far that performs handover start prediction. In FIG. 25, since the transmission of the control frame of the reassociation trigger (Reassociation Trigger) is started by the core switch CoreSW first, in order for the access point AP to receive this control frame, it is necessary to go through the access switch AcessSW on the way. . However, in FIG. 29, not only the access switch AcessSW but also the access point AP can be directly delivered by one control frame transmission.

  Providing a lateral link in this way has the advantage that the transfer of the necessary control frames is faster and more efficient. In general, it is assumed that the further away from the mobile terminal MH, that is, the higher the hierarchy of the access switch AcessSW, the more difficult it is to make a prediction with fine accuracy. For this reason, it is advantageous from the viewpoint of prediction accuracy to assign the role of prediction to the lower access switch AcessSW.

  So far, the handover between the access switches AcessSW via the upper core switch CoreSW has been described. However, the present invention is not limited to this, and in addition to this, for example, the access belonging to more hierarchies such as when straddling two core switches CoreSW and straddling each lower access switch AcessSW. It may be necessary to perform a handover between the switches AcessSW. When the number of hierarchies involved increases, the access switch AcessSW that plays the role of handover prediction shifts to the access switch AcessSW of the higher hierarchy, but the concept of the control frame transfer procedure is the same.

  The present invention can be greatly utilized in the field of mobile communications.

It is a figure which shows a mobile communication network structure. It is a figure which shows the connection form of a switch including a port. It is a figure which shows the procedure of the handover control by an access switch (unicast). It is a figure which shows the procedure of the handover control by an access switch (unicast / bicast). It is a figure which shows the handover control sequence under an access switch (when performing a handover request from a mobile terminal). It is a figure which shows the hand-over control sequence under an access switch (when connecting connection cancellation | release notification is performed from an access switch). It is a figure which shows the handover control sequence under an access switch (when not performing a hand-over request response and a connection relationship cancellation | release notification). It is a figure which shows the data structure of a filtering database. It is a figure which shows the data structure of a filtering database. It is a figure which shows the definition method of a port state. It is a figure which shows the relationship between port use time and a port state when a handover is performed under one access switch. It is a conceptual diagram for demonstrating the timing estimation method of a handover start. It is a conceptual diagram for demonstrating the timing estimation method of a handover start. It is a figure which shows the data structure of a filtering database. It is a figure which shows the data structure of a filtering database. It is a conceptual diagram for demonstrating the timing estimation method of a handover start. It is a conceptual diagram for demonstrating the timing estimation method of a handover start. It is a figure which shows the definition method of a port state. It is a figure which shows the definition method of a port state. It is a figure which shows the relationship between port use time and a port state when a handover is performed under one access switch. It is a figure which shows the definition method of a port state. It is a figure which shows the relationship between port use time and a port state when a handover is performed under one access switch. It is a figure which shows the handover control via a core switch (when there is no lateral link between access switches and bicast is not performed). It is a figure which shows the handover control via a core switch (when there is no horizontal link between access switches and bicast is performed). It is a figure which shows the handover control sequence under a core switch (when there is no horizontal direction link between access switches). It is a figure which shows a mobile communication network structure. It is a figure which shows the handover control via a core switch (when a control frame is transferred by the horizontal link between access switches, and bicast is not performed). It is a figure which shows the handover control via a core switch (when transferring a control frame by the horizontal link between access switches, and performing a bicast). It is a figure which shows the handover control sequence under a core switch (when there exists a horizontal link between access switches). It is a figure which shows a 1st prior art example. It is a figure which shows the 2nd prior art example.

Explanation of symbols

10, 20, 300 Network 21 (21a, 21b, 21c) Switch fabric 22 (22a, 22b, 22c) Filtering database 301 Label controller 311 Network controller 312, 314 Zone 313, 315 Zone controller
CoreSW (Core SW1, SW2, SW3 ...) Core switch
AccessSW (Access SW1, SW2, SW3 ...) Access switch MH Mobile terminal AP (AP1, AP2 ...) Access point

Claims (17)

A mobile terminal; an access point that provides a connection means to the network to the mobile terminal; and a filtering database. Based on a search result of the filtering database and layer 2 information included in the frame, the frame is set to an appropriate port. In a mobility management method for a mobile communication network comprising a layer 2 switch for forwarding,
Transmits necessary part of its own filtering database at a certain time interval in two or one way between interconnected layer 2 switches,
When the mobile terminal newly establishes a radio link with the access point, it transmits the fact from the new access point to the layer 2 switch to which it is connected,
Upon receiving the fact, the layer 2 switch updates the filtering database related to the physical address unique to the mobile terminal in response to the confirmation of the fact, and grasps the port for transferring the user frame to the mobile terminal. A mobile communication network mobility management method characterized by the above. A mobile terminal; an access point that provides a connection means to the network to the mobile terminal; and a filtering database. Based on a search result of the filtering database and layer 2 information included in the frame, the frame is set to an appropriate port. In a mobility management method for a mobile communication network comprising a layer 2 switch for forwarding,
Between the mutually connected layer 2 switches, information of a necessary part of its own filtering database is transmitted at a certain time interval in both directions or in one direction.
When the mobile terminal newly establishes a radio link with the access point, the fact is transmitted from the new access point to the layer 2 switch to which it is connected;
The layer 2 switch that has received the fact updates the filtering database related to the layer 2 label that is uniquely given to the mobile terminal and used for communication over a radio link, triggered by the confirmation of the fact, A mobility management method for a mobile communication network, characterized by grasping a port for transferring a user frame to the mobile terminal. A mobile terminal; an access point that provides a connection means to the network to the mobile terminal; and a filtering database. Based on a search result of the filtering database and layer 2 information included in the frame, the frame is set to an appropriate port. In a mobility management method for a mobile communication network comprising a layer 2 switch for forwarding,
Between the mutually connected layer 2 switches, information of a necessary part of its own filtering database is transmitted at a certain time interval in both directions or in one direction.
When the mobile terminal newly establishes a radio link with the access point, the fact is transmitted from the new access point to the layer 2 switch to which it is connected;
In the layer 2 switch, with the confirmation of the fact, the filtering is performed on the layer 2 label that is uniquely associated with the mobile terminal but is not used for communication over a radio link, thereby updating the filtering database. A mobility management method characterized by grasping a port for transferring a user frame to a mobile terminal. A mobile terminal; an access point that provides a connection means to the network to the mobile terminal; and a filtering database. Based on a search result of the filtering database and layer 2 information included in the frame, the frame is set to an appropriate port. In a handover control method for a mobile communication network comprising a layer 2 switch for transferring,
The upper layer 2 switch uses the data related to the layer 2 information of the mobile terminal in the filtering database held by itself, and the timing at which the mobile terminal should switch the radio link to a new access point, Understand which ports should be used for communication of mobile terminals,
When it is determined that the layer 2 switch can complete the next switching process only with the layer 2 switch below it, the port used for communication of the mobile terminal is directed to the layer 2 switch below. Output a switching process start instruction,
The lower layer 2 switch that has received the switching processing start instruction transfers the switching processing start instruction from the port currently used for communication of the mobile terminal to the lower layer 2 switch or access point. ,
By repeating the transfer of the switching process start instruction, via the access point that has established a radio link with the mobile terminal at that time, communicate the switching process start instruction to the mobile terminal,
The layer 2 switch that first output the switching processing start instruction outputs the information necessary for the switching notice and the subsequent switching from the port to be used next to the lower layer 2 switch,
The lower layer 2 switch that has received the information necessary for the switching notice and the subsequent switching is necessary for the switching notice and the subsequent switching from the port to be used next to the lower layer 2 switch or the access point. Together with the information to be transferred,
By repeating the switching notice and the transfer of information necessary for the subsequent switching,
Next, let the access point used next reach the information required for the switch notice and subsequent switch,
When the mobile terminal that has received the switching process start instruction establishes a radio link with a new access point, the mobile terminal executes the mobility management process according to any one of claims 1 to 3.
Until the fact that a new radio link has been established reaches the layer 2 switch that first output the switching start process, it is assumed that forwarding is sequentially performed through the ports of several layer 2 switches used for communication of the mobile terminal. A handover control method for a mobile communication network, characterized by the following. The mobile communication network handover control method according to claim 4,
Only after the layer 2 switch receives or first transmits a switching processing start instruction until it receives the fact that a new radio link is established,
At the same time, a user frame addressed to the mobile terminal is forwarded not only to the port used for communication by the mobile terminal but also to a port determined to be used next, and the mobile terminal establishes a radio link destination. A handover control method for a mobile communication network, wherein user frames are simultaneously received from two new and old access points when an access point is changed. A mobile terminal; an access point that provides a connection means to the network to the mobile terminal; and a filtering database. Based on a search result of the filtering database and layer 2 information included in the frame, the frame is set to an appropriate port. In a handover control method for a mobile communication network comprising a layer 2 switch for transferring,
The upper layer 2 switch uses the data related to the layer 2 information of the mobile terminal in the filtering database held by itself, and the timing at which the mobile terminal should switch the radio link to a new access point, Understand which ports should be used for communication of mobile terminals,
If it is determined that the layer 2 switch alone cannot complete the next switching process, the layer 2 switch is at the lower level from the port used for communication of the mobile terminal at that time. A switching process start instruction is output to the layer 2 switch one level higher than the layer 2 switch,
A switching notice is transmitted to another layer 2 switch that belongs to the same layer as the layer 2 switch and has a port to be used next through a lateral link connecting the two layer 2 switches,
The fact that a new radio link has been established is output as a first switch start process via a number of layer 2 switch ports that are newly used for communication of the mobile terminal and the ports at both ends of the horizontal link. A handover control method for a network for mobile communication, characterized in that forwarding is performed until reaching the layer 2 switch that is one layer higher than the layer 2 switch. The handover control method according to claim 6, wherein
Only after the layer 2 switch receives or first transmits the switching process start instruction until it receives the fact that a new radio link is established,
At the same time, a user frame addressed to the mobile terminal is forwarded not only to the port used for communication by the mobile terminal but also to a port determined to be used next, and the mobile terminal establishes a radio link destination. A handover control method for a mobile communication network, wherein user frames are simultaneously received from two new and old access points when an access point is changed. The method for controlling handover of a mobile communication network according to any one of claims 4 to 7,
In the switching processing start instruction transmitted to the mobile terminal, an access point to which a new radio link should be established is specified, and information necessary for establishing the radio link is notified in advance. Handover control method. The method for controlling handover of a mobile communication network according to any one of claims 4 to 7,
Upon switching of the radio link, the mobile terminal that has received the switching processing start instruction notifies the access point that has established the radio link at that time before releasing a connection request to the new radio link. A handover control method for a mobile communication network. The method for controlling handover of a mobile communication network according to any one of claims 4 to 7,
When the layer 2 switch directly connected to the old access point receives the fact that a new radio link has been established upon switching of the radio link, the connection is already released from the layer 2 switch to the access point. A handover control method for a mobile communication network, characterized by: The method for controlling handover of a mobile communication network according to any one of claims 4 to 7,
The filtering database has the following data structure:
An element representing a layer 2 identifier associated with the mobile terminal;
An element indicating a port number to which a user frame to which the layer 2 identifier is a destination is to be transferred;
An element representing a state relating to use of the port;
An element for managing port usage time, which means time when the port is actually used for user frame transfer related to the mobile terminal;
An element indicating the timing for starting the radio link switching process;
An element indicating the number of a port used immediately before, meaning that the port was used for user frame transfer of the mobile terminal until just before the port was used for user frame transfer of the mobile terminal;
Elements that manage the effectiveness of these combinations of elements,
A handover control method for a mobile communication network, comprising: In the mobile communication network handover control method according to claim 11,
The filtering database includes, as a data structure, an element that represents the moving speed of the mobile terminal by several speed classes classified according to the size thereof, and the speed class indicates the correspondence between the past port usage time and it. A handover control method for a mobile communication network, characterized by being determined by searching a predetermined table. In the mobile communication network handover control method according to claim 11,
The filtering database includes, as a data structure, elements representing the moving speed of the mobile terminal by several classes classified according to the size thereof, and the speed class is determined by a declaration from the mobile terminal. A handover control method for a mobile communication network. In the mobile communication network handover control method according to claim 11,
The filtering database determines a start timing of a next radio link switching process by calculation based on an element for managing a port usage time included in the data configuration, and performs handover of a mobile communication network Control method. In the mobile communication network handover control method according to claim 11,
Based on the filtering database, a table that associates the past port usage time with the start timing of the next radio link switching process is created, and the start timing of the radio link switching process is determined by searching the created table A handover control method for a mobile communication network, comprising: In the mobile communication network handover control method according to claim 12 or 13,
A mobile communication network characterized by creating a table associating the speed class of the mobile terminal with the start timing of the next radio link switching process, and determining the start timing of the radio link switching process by searching the table Handover control method. The method for controlling handover of a mobile communication network according to any one of claims 11 to 13,
The positional order relationship of the plurality of access points is matched with the order relationship of the ports of the layer 2 switch to which the plurality of access points are connected, and the positional order relationship of the plurality of lower layer 2 switches, Match the order relationship of the ports of the upper layer 2 switch to which a plurality of lower layer 2 switches are connected,
A port to be used for communication of the mobile terminal by comparing the port number used for communication of the mobile terminal at a certain point in time with the number of the port used immediately before of the mobile terminal obtained by referring to the filtering database A handover control method for a mobile communication network, characterized in that:

Images