WO2002037772A1

WO2002037772A1 - Packet communication system, control method thereof and mobile radio communication system

Info

Publication number: WO2002037772A1
Application number: PCT/JP2001/009687
Authority: WO
Inventors: Johnson Oyama; Ina Widegren; Brian Williams
Original assignee: Telefonaktiebolaget Lm Ericsson (Publ)
Priority date: 2000-11-06
Filing date: 2001-11-06
Publication date: 2002-05-10
Also published as: AU2002211016A1

Abstract

To provide a packet communication system equipped with a plurality of gateway apparatuses with a one-to-one correspondence with external networks, capable of selecting an optimal gateway apparatus used during a packet communication considering the other end of communication, channel quality and other arbitrarily preset conditions. A one-to-one correspondence between address information (e.g., IP address) of a service supply terminal that exists on an external network and an optimal gateway apparatus (GGSN) is established by using information on network topology such as information on the geographical location of the terminal at the other end of communication on the external network and this correspondence is stored in a policy DB 39. The policy DB 39 is searched by using the IP address of the service supply terminal declared from a terminal 1 or acquired on a network side and a packet communication channel is set between the gateway apparatus obtained as a result of the search and the terminal 1.

Description

DESCRIPTION

PACKET COMMUNICATION SYSTEM, CONTROL METHOD THEREOF AND MOBILE RADIO COMMUNICATION SYSTEM

TECHNICAL FIELD

The present invention relates to a packet communication system and control method thereof, and more particularly, to a packet communication system and control method thereof using a mobile radio communication terminal.

BACKGROUND ART

Conventionally, when a communication is carried out by using a mobile radio communication terminal such as a cellular phone and PHS (personal handy phone system) , a communication channel is assigned to each call and the assigned call occupies the channel irrespective of whether a conversation is actually engaged or not (irrespective of whether there is an effective voice signal or not) .

On the other hand, these mobile radio communication terminals have acquired a capability other than voice communications, that is, data communications, packet communications, in which data is transferred in packets to effectively use data communication channels, came into use. As a result, most of mobile radio communication terminals currently in use have not only a conventional voice communication function but also a data communication function using packet communication. While mobile radio communication have been drastically spreading all over the world, there have also been growing problems such as shortage of radio resources and incompatibility between mobile radio communication systems being operated under different communication standards. With the growing diffusion rate, a demand for applications as faster data communication terminals is also increasing.

Thus, efforts for establishing standards of next- generation mobile ratio communication systems are being made on the worldwide scale. The IMT-2000 is a next- generation mobile radio communication system standard and this is roughly divided into two; DS-CDMA system and MC-CDMA system, whose standardization is being carried out by standardization organizations called "3GPP" (third generation partnership project) and "3GPP2", respectively.

FIG. 6 illustrates a configuration example of a packet communication network section in a mobile radio communication system using a system formulated by the 3GPP. This configuration is based on a radio packet network used in a GSM network, that is, so-called GPRS (General Packet Radio Service) system. The packet communication network comprises a base station subsystem 2 that includes a radio base station (RBS) 21 and a radio network controller (RNC) 22 and a packet switching core network 3. The base station subsystem 2 handles a radio connection and communication with a terminal (mobile station) 1 and is shared by a voice communication network, which is not shown. Moreover, in the base station subsystem 2, the RNC 22 is actually connected to a plurality of RBSs 21, each RBS 21 exercising control related to a radio connection with the mobile station 1 that exists in own service area of the RBS 21.

The packet switching core network 3 includes an SGSN (Serving GPRS Service Node) 31 that corresponds to an MSC (mobile service switching center) in the voice communication network and GGSNs (Gateway GPRS Service Nodes) 33 to 36 that correspond to GMSCs (Gateway of MSCs) . Each of GGSNs 33 to 36 is configured by a plurality of GGSNs, and functions as a gateway to predetermined one of the subnetworks 41 to 44.

The subnetworks 41 to 44 are the packet networks established to provide certain services. For example, networks of Internet service providers (ISPs) that provide connection services to the Internet, as a packet data network 51 to 54, and to networks of banks having servers to provide so-called "mobile banking services" correspond to these subnetworks. Each of the subnetworks 41 to 44 has a CSCF (Call State Control Function) that controls calls and acquires an IP address of the other end of communication, etc. and DNS (Domain Name System) to acquire the corresponding IP address from a URL or an e-mail address.

In FIG. 6, the subnetworks 41 to 44 are further connected to packet data networks 51 to 54 via gateway apparatuses, which are not shown, but the packet data networks 51 to 54 need not always be connected. The packet data networks 51 to 54 can also be connected to other networks. In the present invention, all networks that carry out communication via GGSNs, that is, the subnetworks 41 to 44, packet data networks 51 to 54 and networks that are connected via these networks are treated as external packet networks.

The packet switching core network 3 also has a GGSN database (GGSNDB) 38 that stores the correspondence between APNs (Access Point Names) and GGSNs of the subnetworks 41 to 44 and an HLR (Home Location Register) 32 that stores service subscription information of users and positional information, etc. By using this HLR 32, the SGSN 31 performs location control, billing control and communication control, etc. of the users who use packet communication services. The drawing shows only one base station subsystem 2 and one SGSN 31, but there is actually more than one base station subsystem and more than one SGSN. FIG. 7 shows an excerpt of the connection between the mobile radio communication system and subnetwork 41 in FIG. 6.

Thus, the packet switching core network 3 is connected to different subnetworks via a plurality of geographically distant GGSNs 331 to 334 and the terminal 1 communicates with the other end of communication that exists on the subnetwork 41 or the packet data network 511 to 513 (further including networks connected in multiple stages via the packet data network 511 to 513) connected via the subnetwork 41, for example, a server 5111 that exists on the packet data network 511, via any one of the GGSNs 331 to 334. Connection between the subnetwork 41 and packet data network 511 to 513 is made via the corresponding gateway apparatus 414 to 416.

Then, with reference to the sequence chart shown in FIG. 8 together, the procedure for using a service provided by the server 413 through the packet communication that exists on the subnetwork 41 from the terminal 1 will be explained as an example.

First, processing (attach) for carrying out a communication between the terminal 1 and SGSN 31 is performed (step S101) . This processing includes confirmation processing to check whether the terminal 1 is registered in the network or not. This confirmation is carried out by the SGSN 31 by referencing the HLR 32 using, for example, the subscriber number of the terminal 1.

Next, the terminal 1 sends a PDP context to establish a control signal (signaling) channel (step S102) . The PDP context includes an APN (Access Point Name) and PDP address, etc. The APN included in the PDP context for signaling channel setting is an APN that indicates connection to the CSCF 411.

Upon reception of this PDP context, the SGSN 31 inquires the GGSNDB 38 about the GGSN corresponding to the APN using the APN included in the PDP context (step S103) . The GGSNDB 38 arbitrarily selects one GGSN (e.g., GGSN 331) from among the GGSNs 331 to 334 stored beforehand as the GGSN corresponding to the received APN, that is, the subnetwork 41 and returns the address thereof to the SGSN 31 (step S104) .

The SGSN 31 sends the PDP context for signaling channel setting to the specified GGSN 331 (step S105) . In response to this request, the GGSN 331 performs predetermined processing and returns a response signal to the terminal 1 via the SGSN 31, and then a signaling channel between the GGSN 331 and the terminal 1 is established, making it possible to exchange a control signal between the terminal 1 and subnetwork 41. Then, the terminal 1 performs registration work with the CSCF 411 that exists on the subnetwork 41 by using the signaling channel (step S107), in order to decide address information necessary for communication between the terminal 1 and devices in the subnetwork 41. Then, the terminal 1 sends a connection request (INVITE message) including the other end of communication specification information to specify the other end of communication to provide a desired service (server 413) , for example, the e-mail address and URL, etc. to the CSCF 411 (step S108) .

The CSCF 411 inquires the DNS 412 using the other end of communication specification information included in the INVITE message, acquires the IP address of the server 413 and transfers the INVITE message to the server 413. When the server 413 accepts the connection request from the terminal 1, the server 413 sends back a response message including the own IP address via the signaling channel (step S109) .

The terminal 1 that has received this response message sends the channel setting request PDP context for data transmission including the APN and IP address of the server 413 to the SGSN 31 (step S110) . As in the case where the signaling channel^* is set, the SGSN 31 inquires the GGSNDB 38 by using the APN included in the data transmission channel setting request message (step Sill) . The GGSNDB 38 selects any one GGSN (e.g., GGSN

332) from among the GGSNs 331 to 334 stored beforehand as the GGSN corresponding to the received APN, that is, the subnetwork 41 and returns the address thereof to the SGSN 31 (step S112) . The SGSN 31 sends the PDP context for data channel setting to the specified GGSN 332 (step S113) and when a response signal from the GGSN 332 is sent back to the terminal 1 via the SGSN 31, a data transmission channel between the terminal 1 and GGSN 332 is established. Then, a data communication between the terminal 1 and the server 413 is carried out by using the GGSN 332 as a gateway.

DISCLOSURE OF INVENTION

In the above-described conventional mobile radio communication system, when a packet communication service with a server, etc. that exists on the subnetwork and packet data network is provided to the terminal 1, the decision as to which GGSN is used from among a plurality of GGSNs assigned as the gateway for the subnetwork to be connected is made under such conditions as in a predetermined order, one of GGSNs available at that time, etc., and conditions related to communication quality such as the other end of communication, the quality of the communication path and length of the communication path, intentions of the service provider and the mobile radio communication system operator are not considered at all.

For this reason, even if, for example, GGSNs assigned to the same subnetwork are geographically distant from each other and the distance of the communication path from the terminal to the server that provides services varies a great deal depending on the GGSN used, that condition is not considered and a communication may be performed through an unnecessarily long path. In this case, the communication quality may deteriorate or a great delay may be involved.

Moreover, when the quality and speed of the channel vary depending on the GGSN used, if the GGSN to be used is decided without considering these conditions, the GGSN that uses a slow channel may be selected even if the GGSN that uses a faster channel is available, which will result in inefficient data transmission. Furthermore, by rather taking advantage of this difference in channel quality, when the mobile communication system operator (telephone company, etc.) attempts to provide services such as discounting of communication fees to subscribers using channels with ordinary quality compared to using channels with high quality, the conventional connection procedure cannot support such services.

Furthermore, there is a case where it is desirable to limit a GGSN used according to circumstances, etc. of a service provider (bank, etc. providing mobile banking services) that provides services to subscribers using packet communication, the conventional connection procedure cannot support this, either. The present invention has been achieved by taking into account the points described above and it is an object of the present invention to provide a packet communication system and a control method thereof capable of selecting a GGSN used during packet communication considering the other end of communication, the quality of the channel and other arbitrarily preset conditions when the terminal issues a packet communication request. That is, a subject of the present invention consists in a packet communication system equipped with a subscriber terminal, a packet switching network that switches packet data transmitted/received by this subscriber terminal and a plurality of gateway apparatuses that enables this packet switching network to communicate with an external packet network, which provides the subscriber terminal with a mutual packet communication with a terminal at the other end of communication that exists on the external packet network, characterized by including address information acquiring means for acquiring address information of the terminal at the other end of communication from a packet communication request signal received from the subscriber terminal and channel setting means for selecting one specified gateway apparatus from the plurality of gateway apparatuses based on the address information of the terminal at the other end of communication and setting a channel to/from the subscriber terminal.

Furthermore, another subject of the present invention consists in a packet communication system equipped with a subscriber terminal, a packet switching network that switches packet data transmitted/received by this subscriber terminal and a plurality of gateway apparatuses that enables the packet switching network to communicate with an external packet network, which provides the subscriber terminal with a mutual packet communication with a terminal at the other end of communication that exists on the external packet network, characterized by including address information extracting means for extracting address information of the terminal at the other end of communication from a response message sent back to the subscriber terminal from the terminal at the other end of communication; channel setting means for selecting one specified gateway apparatus from the plurality of gateway apparatuses based on the address information of the terminal at the other end of communication and setting a channel to/from the subscriber terminal.

Furthermore, another subject of the present invention consists in a mobile radio communication system equipped with the packet communication system of the present invention. Furthermore, another subject of the present invention consists in a control method for a packet communication system equipped with a subscriber terminal, a packet switching network that switches packet data transmitted/received by this subscriber terminal and a plurality of gateway apparatuses that enables the packet switching network to communicate with an external packet network, which provides the subscriber terminal with a mutual packet communication with a terminal at the other end of communication that exists on the external packet network, characterized by including an address information acquiring step of acquiring address information of the terminal at the other end of communication from a packet communication request signal received from the subscriber terminal and a channel setting step of selecting one specified gateway apparatus from the plurality of gateway apparatuses based on the address information of the terminal at the other end of communication and setting a channel to/from the subscriber terminal.

Furthermore, another subject of the present invention consists in a control method for a packet communication system equipped with a subscriber terminal, a packet switching network that switches packet data transmitted/received by this subscriber terminal and a plurality of gateway apparatuses that enables the packet switching network to communicate with an external packet network, which provides the subscriber terminal with a mutual packet communication with a terminal at the other end of communication that exists on the external packet network, characterized by including an address information extracting step of extracting address information of the terminal at the other end of communication from a response message sent back to the subscriber terminal from the terminal at the other end of communication and a channel setting step of selecting one specified gateway apparatus from the plurality of gateway apparatuses based on the address information of the terminal at the other end of communication and setting a channel to/from the subscriber terminal.

BRIEF DESCRIPTION OF DRAWINGS

FIG. 1 illustrates a configuration example of a packet communication system according to an embodiment of the present invention; FIG. 2 shows a detailed view of a part of FIG. 1; FIG. 3 illustrates an example of policy information registered in a policy DB;

FIG. 4 is a sequence chart showing a connection procedure in a first embodiment; FIG. 5 is a sequence chart showing a connection procedure in a second embodiment; FIG. 6 is a drawing to explain a configuration example of a conventional packet communication system;

FIG. 7 shows a detailed view of a part of FIG. 6; and FIG. 8 is a sequence chart showing a conventional connection procedure.

BEST MODE FOR CARRYING OUT THE INVENTION [First Embodiment] With reference now to the attached drawings, the present invention will be explained based on preferred embodiments in detail below.

FIG. 1 is a block diagram showing a packet communication network part of a mobile radio communication system as an example of a packet communication system according to an embodiment of the present invention. As apparent from a comparison with FIG. 6, which shows the conventional configuration, the packet communication network in FIG. 1 has the same configuration except that a policy DB 39 has newly been provided. Likewise, FIG. 2 shows a detailed configuration of the connection with the subnetwork 41, which corresponds to the conventional configuration in FIG. 7. The SGSN 31, GGSN 331 to 334, CSCF 411, etc. are all provided with a necessary communication interface, general-purpose computer devices capable of executing a packet data communication protocol, that is, CPU, ROM, RAM and HDD, etc., and therefore can be implemented using general computer devices that provides necessary processing by the CPU executing programs pre-stored in the ROM and HDD, etc. Therefore, the packet switching core network 3 is basically configured by a combination of a known packet switch such as an ATM switch and computer devices making up the SGSN and GGSN.

The policy DB 39 newly provided in this embodiment is a database used to store policy information to determine the GGSN used to perform a packet communication with the subnetwork considering predetermined conditions and, in this embodiment, stores a table (FIG. 3) to identify an optimal GGSN based on the IP address of the connection target and information necessary to determine the table, for example, topology information (information on the network configuration (geographical location of the server, quality of the communication path in the network, speed, etc.)) of the subnetwork 41 to 44.

A table shown in FIG. 3 is created based on the topology information of the subnetwork and is created to make it possible to select a GGSN connected to the terminal of the connection target through the shortest path using the IP address of the connection target as a keyword. This example shows a case where the GGSN to be used is unconditionally determined by the IP address of the target, but in the case where more detailed control is performed for every subscriber (for example, in the case where billing control, etc. is performed for every subscriber according to the channel quality, etc.), it is also possible to further combine information that can identify the terminal such as subscriber number, etc. It is also possible to determine policy information by arbitrarily combining arbitrary conditions such as changing the GGSN to be selected according to the date/time, day of the week, etc. on which the service is provided.

As shown in FIG. 3, when the policy information is determined so as to have the shortest connection path, topology information of the subnetwork 41 to 44 (or packet data network connected via the subnetwork 41 to 44) is necessary. In this case, the operator of the mobile radio communication system must receive the topology information of each network from the subnetwork 41 to 44. This is because the information on the contents of the packet switching core network 3 such as the geographical location, channel quality and path of the GGSN corresponding to the subnetwork, configuration and communication quality of the subnetwork (or packet data network) and the topology information, which is information of the geographical location, etc. of the host computer that actually provides services are information not disclosed publicly, and determining policy information for unilateral convenience of the packet switching core network 3 or subnetwork 41 to 44 is likely to cause influences unexpected by the target network. Therefore, the topology information is acquired from the subnetwork 41 beforehand, the GGSN selection policy is determined based on the topology information and the policy information based on the determined selection policy is registered in the policy DB 39.

Next, the connection procedure according to this embodiment when a service supply request is issued from the terminal 1 will be explained by using the sequence chart shown in FIG. 4. In FIG. 4, it is assumed that a service requested from the terminal is provided by the server 413 that exists on the subnetwork 41. In FIG. 4, the same processing procedure as that in FIG. 8 above will be assigned the same reference numerals as those in FIG. 8 and repetition of the same explanations will be avoided.

First, the topology information is registered or updated in the policy DB 39 (step S201) . For example, in the case where the policy information is determined according to a policy that a GGSN is selected so that a connection with the server 413 on the subnetwork 41 is made through the shortest path, it is necessary to acquire the IP address of the service supply server 413 that exists in the subnetwork 41, geographical placement and configuration of the subnetwork 41 including information on the location thereof and topology information such as the correspondence between the APNs and subnetworks. Therefore, the CSCF 411 gives topology information of the subnetwork 41 stored in the policy DB 39 every predetermined time and the policy DB 39 updates the content (step S201) . Of course, it is also possible to register or update the content of the policy DB 39 through direct input by the operator who owns the packet switching core network 3. In this case, a communication between the CSCF 411 and policy DB 39 is not necessary.

The procedure after the establishment of a signaling channel with any one of the GGSNs 331 to 334 assigned as the gateway with the subnetwork 41 at the request of the subscriber terminal 1 of the mobile radio communication system, transmission of an INVITE message to the server 413 using this signaling channel until reception of a response message including the IP address of the server 413 (step S101 to step S109) is the same as the conventional procedure explained by using FIG. 8.

Then, the terminal 1 sends a data channel setting PDP context including the IP address of the server 413 acquired in step S109 (step S110) . The SGSN 31 acquires the IP address of the other end of communication included in the received PDP context and inquires (searches) the policy DB 39 (step S116) . The policy DB 39 selects a GGSN (suppose GGSN 332) to be used to connect to the inquired IP address from the correspondence table shown in FIG. 3 and notifies the address of the selected GGSN (step S117) .

Then, the SGSN 31 sends the data channel setting PDP context to the GGSN 332 specified from the policy DB 39 (step S118) . In response to this, when the GGSN 332 sends back a response signal to the terminal 1 and the terminal 1 receives this response signal, a data channel is established between the terminal 1 and GGSN 332 (step S114). Hereinafter, a packet communication is carried out between the terminal 1 and server 413 via the GGSN 332 (step S115) . In this case, since the policy information registered in the policy DB 39 is based on the policy that a connection should be made through the shortest path, the packet communication path between the terminal 1 and server 413 is shorter than the case where another GGSN is used, allowing a communication with an optimal delay characteristic, [second Embodiment]

In the first embodiment, the IP address declared from the terminal 1 is used to select a GGSN based on the policy. However, this method is not available for a system in which the IP address is not declared from the terminal 1. Moreover, since a GGSN is selected based on a declaration from the terminal 1, it is not possible to exclude a possibility that the declaration may be tampered.

A possible example of an illegal declaration is that after an INVITE message is sent to declare a packet communication with a desired terminal (e.g., server 413), the communication destination may be illegally changed to another server in the stage of data channel setting. In this case, if, for example, the subscriber of the terminal 1 has a contract of receiving a discount of the packet communication fee on condition that the subscriber will use services provided by the serve 413, the illegal declaration of the IP address as described above will cause the discounted communication fee to be applied even when services provided by other servers are used.

To solve this problem, this embodiment allows the network side to acquire the IP address of the connection target, making it possible to select an optimal GGSN based on the policy information even if the IP address of the other end of communication is not declared from the terminal 1 at the time of data channel setting or there is a possibility of the terminal 1 carrying out an illegal communication.

Then, by using the sequence chart shown in FIG. 5, the connection procedure when the subscriber terminal 1 issues a service supply request will be explained. In FIG. 5, it is assumed that the service requested from the terminal is provided by the server 413 that exists on the subnetwork 41. Moreover, in FIG. 5, the same processing procedure as that shown in FIG. 8 and FIG. 4 above will be assigned the same reference numerals as those in these figures and repetition of the same explanations will be omitted.

As apparent from a comparison with FIG. 4, which is the sequence chart in the first embodiment, this embodiment is characterized in that a GGSN used to set a data channel is determined by using the IP address of the connection target acquired at a node that exists on the path between the connection target (server 413) and the terminal 1. Any node can be used to acquire the IP address of the connection target as far as such the node can communicate with the policy DB 39. The following explanation will describe a case where the CSCF 411 acquires the IP address. As described above, in the case where a signaling channel is established between the target server and the terminal 1 and the target server accepts a connection request, a response message including the IP address of the target server is sent back to the terminal 1 (step S109) . Since the signaling channel is formed via the CSCF 411, the CSCF 411 can acquire the IP address of the target server 413 included in the response message.

In this embodiment, the CSCF 411 acquires the IP address of the target server (server 413) and the destination of the response message, that is, the address of the terminal 1 from this response message. The CSCF 411, then notifies the acquired IP address of the server 413 and the address of the terminal 1 to the policy DB 39 (step S119) . The policy DB 39 stores the address of the terminal 1 in a one-to-one correspondence with the IP address of the server 41 notified from the CSCF 411.

On the other hand, the terminal 1 that has received the response message from the target server sends a data channel setting PDP context to the SGSN 31 (step S110) . Then, the SGSN 31 extracts* the address of the terminal 1 included in the received PDP context and inquires the policy DB 39 (step S121) .

The policy DB 39 decides from the address of the terminal 1 that a data channel setting request for a communication with the server 413 has been issued, reads the IP address of the server 413 from the information stored above and decides a GGSN to be used based on the IP address similarly as the first embodiment. Then, the policy DB 39 notifies to the

SGSN 31 so that the determined GGSN 322 is used (step S122) . On the other hand, in the case where the address of the terminal 1 is not stored in the policy DB 39, the policy DB 39 decides that the operation of the terminal 1 is illegal and cancels the session with the terminal 1. The SGSN 31 sends the data channel setting PDP context to the notified GGSN 322 (step S123) and sets a data channel between GGSN 322 and terminal 1. Hereinafter, a packet communication is performed between the terminal 1 and the server 413 of the subnetwork 41. Thus, since this embodiment provides a system independent of declaration of the IP address from the user, even if the user declares an illegal IP address in the data channel setting PDP context, such illegality can be excluded. As shown above, since this embodiment selects a GGSN based on the IP address of the other end of communication acquired on the network side, it is possible to select an optimal GGSN for the system even in the case of a system in which the terminal 1 declares no IP address or in the case where the terminal 1 declares an illegal IP address. [Other Embodiment]

The above-described embodiments describe the case where a signaling channel and data channel can use independent GGSNs, but there can also be a system with a restriction that the GGSN used when the signaling channel is set must be used to set the data channel. In such a system, in the case where the terminal 1 declares the IP address of the other end of communication when the signaling channel is set, for example, when the IP address of the server that provides services to be used is already registered in the terminal 1, the data channels to be set thereafter can be set by using an optimal GGSN by applying the method of the first embodiment.

That is, when the signaling channel is set, inquiring the policy DB 39 from the SGSN 31 and setting the signaling channel by using the notified GGSN makes it possible to use the same GGSN even if a data channel setting request is issued thereafter, which results in a packet communication between the terminal 1 and target server using an optimal GGSN.

The above-described embodiments are described based on a configuration in which the policy DB 39 and CSCF included in subnetwork 41 to 44 are connected without any GGSN, but it is also possible to adopt a configuration with the policy DB 39 connected with the CSCF via a GGSN.

The above-described embodiments describe the GGSNDB 38 and policy DB 39 as individual DBs, but any specific mounting method can be used as far as the similar function can be provided. For example, both GGSNDB 38 and policy DB 39 can be mounted on the same HDD. As described above, in a packet communication system equipped with a plurality of gateway apparatuses, which has a one-to-one correspondence with external networks, the present invention can select an optimal gateway apparatus to be used during a packet communication considering the other end of communication, channel quality and other arbitrary preset conditions.

Claims

1. A packet communication system having a subscriber terminal, a packet switching network that switches packet data transmitted/received by said subscriber terminal, and a plurality of gateway apparatuses that enables said packet switching network to communicate with an external packet network and providing said subscriber terminal with a mutual packet communication with a terminal at the other end of communication that exists on said external packet network, comprising: address information acquiring means for acquiring address information of said terminal at the other end of communication from a packet communication request signal received from said subscriber terminal; and channel setting means for selecting one specified gateway apparatus from said plurality of gateway apparatuses based on the address information of said terminal at the other end of communication and setting a channel to/from said subscriber terminal.

2. The packet communication system according to claim

1, wherein said packet communication is carried out by using a signaling channel for control signal communication and a data channel for data communication; and said channel setting means sets said signaling channel as a channel to/from one gateway apparatus arbitrarily selected from among said plurality of gateway apparatuses and sets said data channel as a channel to/from said specified gateway apparatus.

3. The packet communication system according to claim

1, wherein said packet communication is carried out by using a signaling channel for control signal communication and a data channel for data communication; and said channel setting means sets said signaling channel as a channel to/from said specified gateway apparatus and sets said data channel as a channel to/from said specified gateway apparatus for which said signaling channel is set.

4. A packet communication system having a subscriber terminal, a packet switching network that switches packet data transmitted/received by said subscriber terminal, and a plurality of gateway apparatuses that enables said packet switching network to communicate with an external packet network and providing said subscriber terminal with a mutual packet communication with a terminal at the other end of communication that exists on said external packet network, comprising: address information extracting means for extracting address information of said terminal at the other end of communication from a response message sent back to said subscriber terminal from said terminal at the other end of communication; and channel setting means for selecting one specified gateway apparatus from said plurality of gateway apparatuses based on the address information of said terminal at the other end of communication and setting a channel to/from said subscriber terminal.

5. The packet communication system according to any one of claim 1 to claim 4, wherein said channel setting means selects said gateway apparatus corresponding to the shortest communication path with said terminal at the other end of communication as said specified gateway apparatus.

6. The packet communication system according to any one of claim 1 to claim 5, wherein the address information of said terminal at the other end of communication is an IP address.

7. The packet communication system according to any one of claim 1 to claim 6, wherein said subscriber terminal is a mobile radio communication terminal.

8. A mobile ratio communication system equipped with the packet communication system according to any one of claim 1 to claim 7.

9. A control method for a packet communication system equipped with a subscriber terminal, a packet switching network that switches packet data transmitted/received by this subscriber terminal and a plurality of gateway apparatuses that enables this packet switching network to communicate with an external packet network, which provides said subscriber terminal with a mutual packet communication with a terminal at the other end of communication that exists on said external packet network, comprising the steps of: an address information acquiring step of acquiring address information of said terminal at the other end of communication from a packet communication request signal received from said subscriber terminal; and a channel setting step of selecting one specified gateway apparatus from said plurality of gateway apparatuses based on the address information of said terminal at the other end of communication and setting a channel to/from said subscriber terminal.

10. The packet communication system control method according to claim 9, wherein said packet communication is carried out by using a signaling channel for control signal communication and a data channel for data communication; and said channel setting step sets said signaling channel as a channel to/from one gateway apparatus arbitrarily selected from among said plurality of gateway apparatuses and sets said data channel as a channel to/from said specified gateway apparatus.

11. The packet communication system control method according to claim 9, wherein said packet communication is carried out by using a signaling channel for control signal communication and a data channel for data communication; and said channel setting step sets said signaling channel as a channel to/from said specified gateway apparatus and sets said data channel as a channel to/from said specified gateway apparatus.

12. A control method for a packet communication system equipped with a subscriber terminal, a packet switching network that switches packet data transmitted/received by this subscriber terminal and a plurality of gateway apparatuses that enables this packet switching network to communicate with an external packet network, which provides said subscriber terminal with a mutual packet communication with a terminal at the other end of communication that exists on said external packet network, comprising the steps of: an address information extracting step of extracting address information of said terminal at the other end of communication from a response message sent back to said subscriber terminal from said terminal at the other end of communication; a channel setting step of selecting one specified gateway apparatus from said plurality of gateway apparatuses based on the address information of said terminal at the other end of communication and setting a channel to/from said subscriber terminal.

13. The packet communication system control method according to any one of claim 9 to claim 12, wherein said channel setting step selects said gateway apparatus corresponding to the shortest communication path with said terminal at the other end of communication as said specified gateway apparatus.

14. The packet communication system control method according to any one of claim 9 to claim 13, wherein the address information of said terminal at the other end of communication is an IP address.

15. The packet communication system control method according to any one of claim 9 to claim 14, wherein said subscriber terminal is a mobile radio communication terminal.