KR20040087096A - Load distribution method in mobile communication system - Google Patents

Abstract

PURPOSE: A load distributing method in a mobile communication system is provided to prevent an increase of a load in a specific GGSN(Gateway GPRS Service Node) by distributing the load to a different GGSN that can provide the same packet service. CONSTITUTION: An SGSN(Serving GPRS Support Node) receives load distribution information from plural GGSNs and checks load state and resource usage state information of each GGSN(S40,S41). When a packet service request is generated through an MS(Mobile Station) of a mobile communication subscriber, the SGSN inquires IP addresses of the GGSNs that can be available for providing the requested packet service, a DNS(Domain Name Server) searches IP addresses of GGSNs that can provide the corresponding packet service, and the SGSN receives a GGSN IP address response message including information on the IP addresses from the DNS(S42). The SGSN selects a suitable GGSN according to the load state and resource usage state information and transmits a PDP context generation request message to the GGSN(S43).

Description

Load balancing method in mobile communication system {LOAD DISTRIBUTION METHOD IN MOBILE COMMUNICATION SYSTEM}

The present invention relates to a mobile communication system. In particular, when a packet service request is generated from a mobile subscriber, the SGSN of the GGSN received from the DNS according to the load distribution information received at predetermined time intervals from the GGSN interworking in the same GPRS backbone network Selecting a low-load GGSN to provide the packet service among IP addresses, thereby preventing the load from increasing to a specific GGSN and distributing the load to other GGSNs that can provide the same packet service. To a dispersion method.

1 is a block diagram showing the structure of a conventional next-generation General Packet Radio Service (GPRS) network. As shown therein, a mobile station (MS) 11-1, 11-2, a base station (NodeB) 12-1, 12-2), Radio Network Controller (RNC) (13-1, 13-2), Serving GPRS Support Node (SGSN) (14-2, 14-2), Home Location Register (HLR) (15-1, 15) 2), Domain Name Server (DNS) (16), Gateway GPRS Service Node (GGSN) (17-1 to 17-4), GPRS Backbone Network (18), Internet Network (19), Intranet (Intranet) ) Is composed of a network (20).

In general, a mobile subscriber in a GPRS network requests a packet service through MSs 11-1 and 11-2 corresponding to a conventional mobile communication terminal, and receives a base station (NodeB) 12-1 and 12-2 and an RNC. (13-1, 13-2) allocates the radio resources while managing radio resources for the MSs 11-1, 11-2 in the corresponding area.

In addition, SGSNs 14-1, 14-2, HLRs 15-1, 15-2, DNS 16, and GGSNs 17-1 through 17-4 are in the GPRS backbone network 18. SGSN (14-1, 14-2) is a GPRS service support system, in conjunction with the RNC (13-1, 13-2) and GGSN (17-1 to 17-4) to request a packet service It is in charge of location registration function, authentication and packet call request for MS 11-1, 11-2 of mobile subscriber.

The HLRs 15-1 and 15-2 are databases for storing subscriber information of mobile communication subscribers and managing location information (circuit / packet) for incoming routing. This function informs the GGSNs 17-1 to 17-4 of the SGSNs 14-1 and 14-2 where the MSs 11-1 and 11-2 are currently located for paging.

DNS 16 then informs SGSNs 14-1 and 14-2 of the IP addresses of the GGSNs 17-1 to 17-4 that will provide the packet service requested by the mobile subscriber. The GGSNs 17-1 to 17-4 allocate and release IP addresses to MSs 11-1 and 11-2 of mobile subscribers requesting packet services, and perform Internet network 19 and intranet network ( 20) serves as a gateway to match.

Meanwhile, referring to FIG. 2, a packet service setting process in a conventional GPRS packet network will be described. When a mobile subscriber wants to request a packet service through the MS 11, the MS 11 to the SGSN 14 in a PDP context. Send an activation request message.

Then, a radio channel connection is attempted between the MS 11 and the RNC 13 and between the RNC 13 and the SGSN 14 and a bearer channel is set up, thereby forming a radio channel for packet service.

In addition, the SGSN 14 receiving the packet service request may be configured by an APN (Access Point Name) for setting up a packet service transmitted when the MS 11 transmits a PDP context activation request message or by a corresponding mobile subscriber when subscribing a service. Using the APN for setting the registered packet service, the DNS 16 inquires of the IP addresses of the GGSNs 17-1 and 17-2 capable of providing the packet service requested by the mobile subscriber.

That is, the SGSN 14 sends a GGSN IP address request message to the DNS 16 to query the IP addresses of the GGSNs 17-1 and 17-2 to provide packet services according to the APN, and accordingly the DNS (16) retrieves the IP addresses of all GGSNs registered with the same APN as the inquired APN, and then recalls the GGSN IP address response message including information on the IP addresses of all GGSNs 17-1 and 17-2. Send to SGSN 14.

Here, the GGSN IP address response message includes information on a plurality of IP addresses for the corresponding GGSNs 17-1 and 17-2, and the SGSN 14 receiving the GGSN IP address response information includes information on the plurality of IP addresses. By selecting any GGSN1 17-1, a PDP context creation request message for establishing a traffic path between the RNC 13, the SGSN 14, and the GGSN1 17-1 is selected from the selected GGSN1 17-1. -1).

At this time, if the randomly selected GGSN1 (17-1) is in an overloaded state, the GGSN1 (17-1) transmits a PDP context creation rejection message to the SGSN 14 to notify that the traffic path can not be established, The SGSN 14 transmits the PDP context activation rejection message to the MS 11 of the mobile subscriber again, thereby failing to connect the packet call.

That is, when there are IP addresses of a plurality of GGSNs 17-1 and 17-2 having the same APN, the SGSN 14 receiving the information on the IP addresses of the GGSNs 17-1 and 17-2. In the absence of information on the load status of the GGSNs 17-1 and 17-2, any GGSN1 17-1 is selected from the IP addresses of the plurality of received GGSNs 17-1 and 17-2. Since the selected GGSN1 17-1 is overloaded and the other GGSN2 17-2 is unable to provide packet service, the packet call connection fails or the GGSN1 ( It provides a low quality service due to packet service with 17-1).

In the conventional technology as described above, the DNS of the GGSN to the IP address of the GGSN to provide the packet service from the SGSN in response to the packet service request of the mobile communication subscriber IP of the plurality of GGSN registered as the APN inquired from the SGSN When the address is delivered to the SGSN, the SGSN selects an IP address of any GGSN and requests packet service to the corresponding GGSN in the absence of information on the transferred state of the GGSN. Although present, the packet service request is rejected by the GGSN under an arbitrarily selected overload state, or there is a problem of providing a low quality service.

Accordingly, the present invention has been proposed to solve the above-described problems, and when a packet service request is generated from a mobile subscriber, SGSN is applied to load distribution information received at predetermined time intervals from a GGSN interworking in the same GPRS backbone network. Therefore, by selecting a low-load GGSN to provide the packet service among the GGSN IP addresses received from DNS, it is possible to prevent the load from increasing to a specific GGSN and distribute the load to another GGSN that can provide the same packet service. The purpose is to provide a way to.

1 is a block diagram showing the structure of a conventional next-generation GPRS network.

Figure 2 is a flow chart showing the flow of each signal of the packet service configuration process in the conventional GPRS packet network.

Figure 3 is a flow chart showing the flow of each signal in the load balancing method in the mobile communication system of the present invention.

4 is a flowchart illustrating an operation process of a load balancing method in a mobile communication system of the present invention.

5 is a flowchart illustrating a process of SGSN selecting a GGSN in a load balancing method in a mobile communication system of the present invention.

*** Description of the symbols for the main parts of the drawings ***

11-1,11-2: MS 12-1,12-2: Base station

13-1,13-2: RNC 14-1,14-2: SGSN

15-1,15-2: HLR 16: DNS

17-1 to 17-2: GGSN 18: GPRS backbone network

19: Internet network 20: Intranet network

The present invention for achieving the above object, the system operation, SGSN receives load distribution information from a plurality of GGSN at predetermined time intervals to check the load status and resource usage status information of each GGSN; When the packet service request is generated from the mobile subscriber, the SGSN selecting a GGSN to provide the packet service from among the IP addresses of the GGSNs transmitted from the DNS according to the previously identified load and resource usage information. It features.

Hereinafter, an embodiment according to the present invention will be described in detail with reference to the accompanying drawings.

3 is a flow chart showing the flow of each signal in the load balancing method in the mobile communication system of the present invention, the configuration of which is the same as the conventional configuration of FIG.

That is, the present invention provides a mobile station (MS) 11-1, 11-2, a base station (NodeB) 12-1, 12-2, a radio network controller (RNC) 13-1, 13-2, Serving GPRS Support Node (SGSN) (14-2, 14-2), Home Location Register (HLR) (15-1, 15-2), Domain Name Server (DNS) 16, Gateway GPRS Service Node (GGSN) 17-1 to 17-4, the GPRS backbone network 18, the Internet network 19, and the intranet network 20, and their functions are the same as in the prior art.

In general, the echo request message (GTP-C Echo Request Message) and the echo response message (GTP-C) between the SGSNs 14-1 and 14-2 and the GGSNs 17-1 to 17-4 when the system is in normal operation. Sending and receiving status confirmation messages such as echo response messages at predetermined time intervals, such as SGSNs 14-1 and 14-2 and GGSNs 17-1 to 17-4 that operate on the same GPRS backbone network 18. Check whether the node is operating normally.

However, in the present invention, the GGSNs 17-1 to 17-4 provide an operation definition area in the echo response message to load distribution information on the current response status, available IP address, etc. in the echo response message during operation of the system. Field), transmitted to the SGSNs 14-1 and 14-2, and linked to the same GPRS backbone network 18, SGSNs 14-1 and 14-2 and GGSNs 17-1 to 17-4. The current load status and resource usage status of the CTSNs can be known to other SGSNs 14-1, 14-2 and GGSNs 17-1 to 17-4.

In addition, the SGSNs 14-1 and 14-2 receive the load distribution information, check the load state and resource usage state information of each GGSN 17-1 to 17-4, and then request the packet service of the mobile subscriber. When it occurs, the RNC 13, SGSN 14, and the selected GGSN 17-1 to 17-4 are selected by selecting the low load GGSNs 17-1 to 17-4 according to the previously checked load state and resource usage state information. Performs a PDP context creation procedure for establishing a traffic path between).

Here, the load state information refers to information on which load state the GGSNs 17-1 and 17-2 are in. The load state information is classified into a predetermined level so that information on a level to which a current load corresponds, Or information about the current load rate.

The resource usage state information refers to information in the form of a list of available IP addresses for all APNs managed by the GGSNs 17-1 and 17-2, or information on the number of available IP addresses.

Meanwhile, referring to FIG. 3 to FIG. 5, a packet service setting process in a GPRS network will be described. In the next generation GPRS network, the SGSN 14 and the GGSN 17-1 and 17-2 may perform the echo request message and the system operation. Since the echo response message checks whether the other node is operating normally at predetermined time intervals, and the echo response message transmitted from the GGSNs 17-1 and 17-2 includes load distribution information, SGSN The 14 may receive the load distribution information from the plurality of GGSNs 17-1 and 17-2 to check the load state and resource usage state information of each GGSN 17-1 and 17-2 (S40). S41).

Thereafter, when a packet service request occurs through the MS 11 of the mobile subscriber, the MS 11 transmits a PDP context activation request message to the SGSN 14.

Then, a radio channel connection is attempted between the MS 11 and the RNC 13, between the RNC 13 and the SGSN 14, and a bearer channel is set up, thereby forming a radio channel for packet service.

In addition, the SGSN 14 receiving the packet service request may be configured by an APN (Access Point Name) for setting up a packet service transmitted when the MS 11 transmits a PDP context activation request message or by a corresponding mobile subscriber when subscribing a service. Using the APN for setting the registered packet service, the DNS 16 queries the IP address of the GGSNs 17-1 and 17-2 capable of providing the packet service requested by the mobile subscriber.

That is, the SGSN 14 sends a GGSN IP address request message to the DNS 16 to query the IP addresses of the GGSNs 17-1 and 17-2 to provide packet services according to the APN, and accordingly the DNS (16) retrieves all IP addresses of the GGSNs in which the inquired APNs are registered to provide the corresponding packet service, and the SGSN 14 includes information on the IP addresses of the retrieved GGSNs 17-1 and 17-2. A GGSN IP address response message is received from the DNS 16 (S42).

Here, the GGSN IP address response message includes a plurality of information on the IP address of the GGSN (17-1, 17-2) that can provide the packet service, the SGSN 14 receives this step (S41) Selects a specific GGSN (17-1, 17-2) according to the load state and resource usage state information previously identified (S43), traffic between the RNC 13 and SGSN (14) and GGSN1 (17-1) A packet call is connected by sending a PDP context creation request message for establishing a traffic path to the selected specific GGSNs 17-1 and 17-2.

In the present invention, it is assumed that the IP address of the GGSN registered as the APN according to the packet service request of the user is GGSN (17-1, 17-2), of which GGSN1 (17-1) is in a high load state, GGSN2 (17) -2) assumes a low load.

Therefore, in the above, the SGSN 14 selects the GGSN2 17-2 according to the previously confirmed load state and resource use state information and transmits a PDP context creation request message.

Here, referring to the process of the SGSN 14 selects the GGSN2 17-2, the SGSN 14 first loads received from each GGSN 17-1, 17-2 at predetermined time intervals during system operation. From the distributed information, the load state and resource use state information of each GGSN 17-1, 17-2 is checked.

Then, when a packet service request of a mobile subscriber occurs, the SGSN 14 determines whether the previously identified load state and resource usage state information is load state information of each GGSN 17-1, 17-2. In the case of load state information (S50), the most recently received load state information is selected from the received load state information (S51).

In this case, the load state information may be divided into predetermined levels of loads processed by the GGSNs 17-1 and 17-2, and information on a level corresponding to a current load, or information on a current load ratio may be used. Since it is included, SGSN 14 can select the GGSN2 (17-2) that is the lowest load state to provide the packet service from the selected recent load state information (S52).

On the other hand, in the step (S50), if the predetermined load state and resource use state information is information on the available IP address (S53), the resource use state information is the respective GGSN (17-1, 17-2) Information in the form of a list of available IP addresses for all APNs managed by the AP, or information on the number of available IP addresses is included.

Accordingly, the SGSN 14 transmits the IP for the APN requested by the mobile subscriber to the packet service through the list of available IP addresses or the number of available IP addresses for all ANPs of the received GGSNs 17-1 and 17-2. Among the addresses, the GGSN2 17-2 having the largest number of available IP addresses for the APN of the packet service is selected (S54).

As described above, in the present invention, the SGSN receives load distribution information at predetermined time intervals from GGSNs interworking in the same GPRS backbone network to check load state and resource usage state information of each GGSN, and the mobile communication subscriber. When a packet service request is generated from the GGSN, the GGSN to provide the packet service is selected from the IP address of the GGSN received from the DNS according to the information on the load and resource usage information. By providing a packet service by distributing the load to other GGSNs that can provide the same packet service, it is possible to stably operate the system and to provide a high quality packet service to a mobile subscriber.

Claims (5)

During operation of the system, SGSN receives load distribution information from a plurality of GGSNs at predetermined time intervals to check load state and resource usage state information of each GGSN; When the packet service request is generated from the mobile subscriber, the SGSN selecting a GGSN to provide the packet service from among the IP addresses of the GGSNs transmitted from the DNS according to the previously identified load and resource usage information. A load balancing method in a mobile communication system, characterized by the above-mentioned. The method of claim 1, wherein the load distribution information is included in a status confirmation message reported from the GGSN to the SGSN during operation of the system. The method of claim 1 or 2, wherein the load distribution information comprises at least one of load status information of each GGSN and information on available IP addresses. The method of claim 1, wherein the selecting of the GGSN comprises: determining whether the previously determined load state and resource usage state information is load state information of each GGSN, and selecting the most recently received load state information; ; And selecting the GGSN which is the lowest load state to provide the packet service from the selected recent load state information. The method of claim 1, wherein the selecting of the GGSN comprises: when the previously determined load state and resource use state information is information on an available IP address, available IP for the APN of the corresponding packet service among the received IP addresses of the GGSN. The method of claim 1, further comprising selecting a GGSN having the largest number of addresses.