CN100433658C

CN100433658C - Method and system for implementing automatic load redistributing

Info

Publication number: CN100433658C
Application number: CNB2006101093460A
Authority: CN
Inventors: 唐宗全
Original assignee: Huawei Technologies Co Ltd
Current assignee: Global Innovation Polymerization LLC; Gw Partnership Co ltd
Priority date: 2006-08-10
Filing date: 2006-08-10
Publication date: 2008-11-12
Anticipated expiration: 2026-08-10
Also published as: CN1905486A

Abstract

The invention discloses a method for implementing automatic load reallocation, comprising: when CN node starts unloading, informing RAN node; RAN node sets status corresponding to the CN node according to unloading information carried in the received notice and returns response to the CN node, and when the response the setting result information carried in the response received by the CN node shows success, entering the unloading status. And the invention also discloses a system for implementing automatic load reallocation. By setting unloading start/stop threshold, it automatically starts the unloading course according to the current load status, and besides, by unloading course starting interaction of Iu interfaces, it shortens delay caused by artificial interference and larges reduces maintenance costs of manpower and material resources.

Description

Method and system for realizing automatic load redistribution

Technical Field

The invention relates to a load redistribution technology, in particular to a method and a system for realizing automatic load redistribution under an Iu-Flex networking.

Background

The Iu-Flex feature is supported by the third generation partnership project (3GPP) release 5(REL5), and a Radio Access Network (RAN) node, such as a Radio Network Controller (RNC) or a Base Station Controller (BSC), may connect a plurality of Core Network (CN) nodes, such as a Mobile Switching Center (MSC) in a Circuit Switched (CS) domain and a Serving GPRS Support Node (SGSN) in a Packet Switched (PS) domain. For Iu-Flex (routing functionality for intra domain connection of RAN nodes to multiple CN nodes), meaning presumably "routing functions applied to connections between RAN nodes and multiple CN nodes within the same CN domain", those skilled in the art typically use Iu-Flex directly. Several areas covered by the RAN constitute a pool area, and a pool area is usually served by several CN nodes in parallel, or by one CN node. When a pool area is served by several CN nodes in parallel, these CN nodes constitute a CN node pool, i.e. MSC pool or SGSN pool, which is the concept of a pool area in general.

Fig. 1 is a schematic diagram of a network topology supporting Iu-Flex, and as shown in fig. 1, a RAN node service area, i.e., a set of all cells controlled by an RNC, which are commonly connected with a plurality of MSCs/SGSNs, respectively form a CS pool area/a PS pool area, where the corresponding set of MSCs/SGSNs is an MSC pool/an SGSN pool.

In order to identify a CN node in a CN node pool, each CN node is allocated one or several Network Resource Identifications (NRIs), and the allocated NRIs are configured in the CN node through an operation and maintenance (O & M) operation. When a CN node allocates a Temporary Mobile Subscriber Identity (TMSI) or a packet temporary mobile subscriber identity (P-TMSI) to a certain user terminal (UE), the NRI identity of the CN node is carried in the TMSI/P-TMSI and sent to the UE, so that the UE is assigned with a CN providing service, and the CN can be called as a service CN. Here, if the CN node is the MSC, the TMSI is allocated to the UE; if the CN node is the SGSN, the P-TMSI is allocated to the UE.

Since both the MSC and the SGSN in the pool area can provide services for the user, in the UE access process, when the RNC receives the initial direct transfer message from the UE, the MSC/SGSN node is selected, that is, one of the MSC or the SGSN is selected based on the routing parameters in the initial direct transfer message, such as the IDNNS information, and the configuration information in the RNC. In the 3GPP specification, the function corresponding to the procedure of determining the serving CN node is referred to as a non-access stratum (NAS) node selection function (NNSF), and is also commonly referred to as initial NAS message routing.

Here, the IDNNS information carried in the initial direct transfer message may be derived from TMSI, or P-TMSI, or International Mobile Subscriber Identity (IMSI), or International Mobile Equipment Identity (IMEI). If there is TMSI/P-TMSI available, the UE preferentially derives routing parameters from the TMSI/P-TMSI; if no TMSI/P-TMSI is available, the UE calculates routing parameters based on the IMSI; if a GSM subscriber identity module/universal subscriber identity module (SIM/USIM) does not exist, i.e. the UE cannot acquire the IMSI, the UE calculates routing parameters based on the IMEI. Before the UE sends the initial direct transfer message, the UE fills in the IDNNS information, i.e. the TMSI/P-TMSI routing parameters, or the IMSI routing parameters, or the IMEI routing parameters, based on the above-mentioned principle.

To obtain the TMSI/P-TMSI routing parameters, the relevant specifications extend the TMSI/P-TMSI with 10 bits (bit) of the 32-bit TMSI/P-TMSI, i.e. bit14 to bit23, allocated for NRI use, Table 1 is the position of the NRI in the TMSI/P-TMSI. The UE derives NRI from TMSI/P-TMSI as IDNNS information to be carried in the initial direct transfer message; in the RNC, when an operator performs network planning, the operator has already configured configuration information of respective corresponding NRI for all the MSCs/SGSNs. Thus, the RNC can search the serving CN node according to the NRI in the initial direct transfer message and the configuration information in the RNC, and complete the routing of the UE initial NAS message.

TABLE 1

In order to obtain IMSI/IMEI routing parameters, the UE calculates a V value according to the following formula (1), where div represents division and mod represents modulo operation, and takes the V value as IDNNS information. The value of V is an integer value falling within the range [0..999 ]; and in the RNC, the corresponding relation between different ranges of the V value and different CN nodes is configured according to the network planning of an operator. Thus, the RNC can search the serving CN node according to the V value in the initial direct transfer message and the corresponding relation in the RNC, and complete the routing of the UE initial NAS message.

V＝(IMSI/IMEI div 10)mod 1000 (1)

In addition, there are two main cases that indicate that no valid routing parameters are carried in the initial direct transfer message from the UE: (1) the RNC can not find the corresponding CN node according to the NRI, or can find the corresponding CN node but the CN node is unavailable; (2) the initial direct transfer message carries the routing parameter of IMSI/IMEI, but the RNC can not find the corresponding CN node according to the V value, or finds the corresponding CN node but the CN node is unavailable. In this case, the RNC selects an available CN node in its own configuration information or corresponding relationship as a serving CN node according to a load balancing principle and performs routing.

In the normal network operation process, due to the roaming behavior of the mobile user, for example, a certain number of users often enter a roaming pool area carrying the TMSI/P-TMSI allocated by the MSC/SGSN in the home CN node pool, and the RNC in the roaming pool area cannot find the corresponding CN node for routing according to the TMSI/P-TMSI carried by the UE, and at this time, the RNC can only perform routing according to the load balancing principle. Thus, over time, the cumulative effect of these roaming users leaving/entering the pool area may result in a severe imbalance in the load of different CN nodes in the roaming pool area. Thus, the 3GPP REL6 specification proposes a load reallocation function, enabling the higher loaded CN nodes to transfer a portion of their users to other CN nodes within the pool area.

The load redistribution function introduces two important identities: null-NRI and Non-broadcast location area identifier/routing area identifier (Non-broadcast LAI/RAI). The null-NRI is a unique identifier of a Public Land Mobile Network (PLMN) range, is uniformly planned in the whole network by an operator, has no fixed corresponding relation with a CN node, and needs to be configured in the RNC and the CN node; the Non-broadcast LAI/RAI has uniqueness in a pool area and has one-to-one correspondence with CN nodes in the pool area, wherein the Non-broadcast LAI is in one-to-one correspondence with a certain MSC, and the Non-broadcast RAI is in one-to-one correspondence with a certain SGSN. All CN nodes in the pool area need to configure the corresponding relation between other CN nodes in the pool area and Non-broadcastLAI/RAI. For the detailed definition of null-NRI and Non-broadcast LAI/RAI, see the relevant provisions of the 23.236 specification, and will not be described in detail here.

According to the 3GPP specification, when the network operation mode (NMO) is '2', namely a Gs interface does not exist between the MSC and the SGSN, the load redistribution process of the MSC and the SGSN is independently executed, and the processing processes are basically the same; when NMO is 1, that is, there is a Gs interface between the MSC and the SGSN, load reallocation of the MSC is initiated by the corresponding SGSN on the Gs interface. The method for implementing the existing load reallocation is described below by taking the load reallocation process of the MSC when NMO is 2 as an example, and the CN node is an MSC node at this time, which includes the following aspects:

(1) on the MSC, a network management personnel executes an O & M command to initiate a load redistribution process, so that an MSC node enters an unloading (off-load) state; meanwhile, on the RNC, it is also necessary to cooperatively execute an O & M command through network management personnel, and set the state of the MSC node corresponding to entering the offloading state to a state corresponding to offloading.

(2) Next, after the duration is about two Location Area (LA) update cycles or longer, for the UE performing LA update, the MSC node in the offload state carries a new TMSI including null-NRI and a Non-broadcast LAI in an LA update accept message returned to the UE to trigger the UE to immediately initiate a LA update request again, fill null-NRI in the IDNNS information carried in the LA update request, and carry the Non-broadcast LAI in the LA update request; and executing TMSI reallocation process to the remaining users, namely reallocating null-NRI and Non-broadcast LAI, wherein the remaining users mainly refer to UE in a connected state all the time, when the UE as the remaining users leaves the connected state, an LA update request is immediately initiated, and null-NRI and Non-broadcast LAI are filled in IDNNS information carried in the LA update request.

(3) After receiving the LA update request, the RNC parses the routing parameters carried in the LA update request, and finds that the request carries null-NRI, then searches for the MSC currently in the offload state, and according to the relevant specification, before performing load reallocation, a maintenance worker sets the state of the MSC performing the offload operation to the offload state on the RNC through an O & M command, where the number of MSCs in the offload state may be one or more. Then, the RNC performs NNSF functions on all MSC nodes that are available and in a non-offload state through O & M commands for LA update requests carrying null-NRI, and sends NAS messages to these MSC nodes in a non-offload state.

(4) The MSC node that receives the NAS message learns the MSC node to which the UE belongs before initiating the LA update according to the Non-broadcastLAI carried in the NAS message, and acquires necessary context, such as an IMSI, an integrity protection key, an encryption key, and other information, from the MSC node.

When the CN side needs to perform load reallocation, the reason why the RAN side maintenance personnel cooperate to perform corresponding settings on the CN node corresponding to the offloading is that, based on the specification of the 3GPP REL' 623.236 for the load reallocation process, the user cannot be allocated to the CN node performing offloading, and in order for the RNC to know whether the CN node is in an offloading state, the RAN side maintenance personnel must simultaneously set the status of the CN node and indicate that the corresponding CN node is in an offloading state.

However, since the CN side network management system and the RAN side network management system are usually separated and maintained by maintenance personnel in different departments of the operator, when the CN side maintenance personnel redistribute the load, the RAN side maintenance personnel must be notified by a communication means, which undoubtedly increases the maintenance cost in terms of manpower and material resources; more importantly, during the cooperative operation process, the maintainers of the CN side network management system and the RAN side network management system are likely to have operation errors, which will cause the failure of the whole load redistribution process, and reduce the quality of load redistribution, thereby introducing serious consequences.

Disclosure of Invention

In view of the above, the main objective of the present invention is to provide a method for implementing automatic load reallocation, which can ensure the quality of load reallocation and save the maintenance cost.

Another object of the present invention is to provide a system for realizing automatic load redistribution, which can ensure the quality of load redistribution and save the maintenance cost.

In order to achieve the purpose, the technical scheme of the invention is realized as follows:

a method for realizing automatic load redistribution, which comprises the following steps:

A. when a core network CN node starts unloading, a radio access network RAN node is informed;

the RAN node sets the state corresponding to the CN node according to the unloading information carried in the received notification, and returns a response to the CN node;

C. and when the setting result information carried in the response received by the CN node shows success, the CN node enters an unloading state to realize automatic load redistribution.

Presetting an unloading starting threshold for the CN node to judge whether to start the unloading process;

the method for starting unloading by the CN node in the step A comprises the following steps: and when the current load of the CN node is higher than a preset unloading starting threshold, starting an unloading process.

The method for notifying the RAN node in step a includes:

the CN node carries unloading information comprising an unloading mark indicating starting/stopping unloading and a global mark of the CN node in a preset unloading request message and sends the unloading information to the RAN node; the unloading request message is interacted through an Iu interface, and the Iu interface is an interface between the CN node and the RAN node;

the unload flag is shown activated.

The step B specifically comprises the following steps:

B1. the RAN node sets the state of the CN node corresponding to the received global CN node identification as an unloading state, and returns a preset unloading response message to the CN node, and the setting result information carried in the unloading response message is displayed as success;

the offload response messages are interacted via an Iu interface, which is an interface between the CN node and a RAN node.

The offloading information further includes: the empty network resource identifies null-NRI.

The step B specifically comprises the following steps: the RAN node judges whether the received unloading request message carries null-NRI or not, if the received unloading request message does not carry null-NRI, the RAN node sets the state of the CN node corresponding to the received global CN node identification as an unloading state and returns a preset unloading response message to the CN node, and the setting result information carried in the unloading response message shows that the unloading response message is successful;

if the information is carried, the RAN node judges whether the null-NRI is consistent with the null-NRI pre-configured by the RAN node, if not, the RAN node returns an unloading response message to a CN node corresponding to the received global CN node identifier, the setting result information carried in the unloading response message shows that the information is failed, and the failure reason is null-NRI inconsistency; and if so, setting the state of the CN node corresponding to the received global CN node identifier as an unloading state by the RAN node, returning an unloading response message to the CN node, displaying that the setting result information carried in the unloading response message is successful, and then executing the step C.

Presetting an unloading stop threshold for the CN node to judge whether to stop the unloading process; after the CN node enters the offload state, the method further includes:

and when the current load of the CN node is lower than a preset unloading stop threshold and is in an unloading state, stopping the unloading process.

The method for stopping the unloading process comprises the following steps: after the CN node reaches the unloading stop condition, the unloading information comprising an unloading mark indicating the start/stop of unloading and the global mark of the CN node is carried in an unloading request message and sent to the RAN node;

the unload flag is displayed as stopped;

and the RAN node sets the state of the CN node corresponding to the received global CN node identification as a non-unloading state and returns an unloading response message to the CN node.

The method further comprises the following steps: existing operation maintenance O & M commands to start/stop offloading are retained in the CN node and/or RAN node.

A system for realizing automatic load redistribution comprises a Core Network (CN) node and a Radio Access Network (RAN) node; wherein,

the CN node is used for sending an unloading starting notice to the RAN node when unloading is started; entering an offload state upon receiving a response from the RAN node and the response indicating success;

and the RAN node receives the unloading starting notification from the CN node, sets the state corresponding to the CN node and returns a response to the CN node.

The CN node is also used for sending an unloading stopping notification to the RAN node;

the RAN node receives the unloading stopping notification from the CN node, sets the state of the corresponding CN node to be a non-unloading state, and sends an unloading response message to the CN node.

Existing operation maintenance O & M commands to start/stop offloading are retained in the CN node and/or RAN node.

And the CN node and the RAN node belong to the same pool area.

The CN node is: a Mobile Switching Center (MSC) node or a Serving GPRS Support Node (SGSN) node;

the RAN node is as follows: a radio network controller RNC or a base station controller BSC.

The CN nodes belong to the same or different operators.

It can be seen from the above technical solutions that, the method of the present invention automatically starts/stops the offloading process according to the current load state by configuring the start/stop offloading threshold, and simultaneously, through the interaction of the start/stop offloading process of the Iu interface, no matter the CN node automatically or manually starts the offloading process, the intervention of the maintenance personnel on the RAN side is not needed, so that not only is the delay introduced by manual intervention shortened, but also the risk introduced by the cross-department collaborative operation error of the operator is avoided, and the maintenance cost of manpower and material resources is greatly reduced.

Furthermore, the method of the invention supplements the existing method, and better realizes the load redistribution.

Drawings

FIG. 1 is a schematic diagram of a network topology supporting Iu-Flex;

FIG. 2 is a flow chart of the method of the present invention.

Detailed Description

The core idea of the invention is as follows: when the CN node starts unloading, the RAN node is informed; and the RAN node sets the state corresponding to the CN node according to the unloading information carried in the received notification, returns a response to the CN node, and enters the unloading state when the setting result information carried in the response received by the CN node shows success.

In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and preferred embodiments.

Fig. 2 is a flowchart of the method of the present invention, according to the protocol specification, the corresponding relationship between null-NRI, the global CN node identifiers of all CN nodes in the CN pool area, and Non-broadcast LAI is configured on the CN node, and the method of the present invention includes the following steps:

step 200: when the CN node initiates offloading, the RAN node is notified.

In this step, the method of whether the CN node, such as the MSC node or the SGSN node, starts the offloading is: and presetting an unloading starting threshold for the CN node to judge whether to start the unloading process, and starting the unloading process when the current load of the CN node is higher than the preset unloading starting threshold. For example, when the load is represented by a class, the unloading start threshold is a certain class.

Meanwhile, the CN node notifies the RAN node with a newly set OFF-LOAD REQUEST message through an Iu interface with the RAN node such as RNC or BSC. The unloading information carried in the unloading request message comprises an unloading mark indicating the start/stop of unloading and a global identifier of the CN node, wherein the unloading mark takes the value of start and the global identifier of the CN node is the global identifier of the CN node. The offload information may further include null-NRI.

Step 201: and the RAN node sets the state corresponding to the CN node according to the unloading information carried in the received notification, and returns a response to the CN node.

After receiving the unloading request message, the RAN node firstly checks whether null-NRI is carried in the unloading request message, if the null-NRI is not carried, the RAN node defaults that null-NRI configured in the CN node is consistent with self configuration, the state of the CN node corresponding to the global CN node identifier carried in the unloading request message is set to be an unloading state, an unloading response (OFF-LOADRESPONSE) message is returned to the CN node, and the setting result information carried in the unloading response message shows that the unloading request message is successful;

if carrying null-NRI, the RAN node further judges whether the carried null-NRI is consistent with the null-NRI configured by the RAN node, if not, the RAN node returns an unloading response message to a CN node corresponding to the global CN node identification carried in the received unloading request message, the setting result information carried in the unloading response message shows that the unloading response message fails, and the failure reason is that the null-NRI is inconsistent; and if the CN node is consistent with the global CN node identifier, the RAN node sets the state of the CN node corresponding to the global CN node identifier carried in the received unloading request message as an unloading state, returns an unloading response message to the CN node, and displays the setting result information carried in the unloading response message as success.

In this step, the offload response message is a new message on the Iu interface between the CN node and the RAN node.

Step 202: and entering an unloading state when the setting result information carried in the response received by the CN node shows success.

After receiving the unloading response message from the RAN node, the CN node checks the setting result information carried in the unloading response message, and if the unloading response message is displayed as successful, the CN node enters an unloading state; otherwise, checking the failure reason, and if the failure reason is that null-NRI is inconsistent, giving an alarm, such as displaying the alarm through a network management system.

It should be noted that if there are more than one RAN node corresponding to a CN node in a pool area, the CN node needs to perform the automatic load reallocation procedure shown in fig. 2 with all the RAN nodes in the pool area.

Further, the method of the invention also comprises the following steps: an unloading stop threshold for the CN node to determine whether to stop the unloading process is preset, for example, when the load is represented by a class, the unloading stop threshold is a certain class. When the current load of the CN node is lower than a preset unloading stop threshold and is already in an unloading state, stopping the unloading process, which specifically includes:

after the CN node reaches the unloading stop condition, after waiting for a short delay time, such as 5 seconds, sending an unloading request message to the RAN node, wherein the unloading information carried in the unloading request message comprises an unloading mark with a value of stopping and a CN node global identifier; it should be noted that the delay time is mainly set to prevent the ping-pong effect, the delay time length depends on the implementation of the load control by different manufacturers, and the length of the delay time is not limited in the present invention.

After receiving the unloading request message, the RAN node sets the state of the CN node corresponding to the global CN node identifier as a non-unloading state because the unloading mark is displayed as stop, and then returns an unloading response message to the CN node.

Further, the method of the invention also comprises the following steps: existing start/stop offload commands are retained in the CN node, and offload may be started/stopped via O & M commands, as in the case where the CN node is loaded normally.

In addition, in some RAN nodes or all RAN nodes which do not support Iu interface offload interaction process, the existing start/stop offload command is retained to set the state of the corresponding CN node to be the offload state.

According to the method, the CN node automatically starts/stops the unloading process according to the current load state through threshold configuration, so that the delay introduced by manual intervention is shortened, the risk introduced by manual operation errors is eliminated, and the maintenance cost of manpower and material resources is greatly reduced.

Through the interaction of the start/stop unloading process of the Iu interface, no matter the CN node automatically or manually starts the unloading process, the intervention of maintenance personnel at the RAN side is not needed, the risk caused by manual operation errors is eliminated, more importantly, the cross-department cooperative operation of operators is avoided, and the maintenance cost is greatly saved.

The technical method is also suitable for multi-core network operators CN (MOCN), namely all CN nodes connected with RAN nodes belong to different operators, each CN operator has a corresponding null-NRI, the range of Iu-Flex is only effective in the range of the operator, and the RAN nodes respectively execute the method in the range of each operator.

According to the description of the method of the present invention, the present invention also provides a system for realizing automatic load redistribution, which comprises a core network CN node and a radio access network RAN node; the CN node is used for sending an unloading starting notice to the RAN node when unloading is started; entering an offload state upon receiving a response from the RAN node and the response indicating success; and the RAN node receives the unloading starting notification from the CN node, sets the state corresponding to the CN node and returns a response to the CN node.

The CN node also comprises: and stopping the unloading process, sending an unloading stopping notification to the RAN node, receiving the unloading stopping notification from the CN node by the RAN node, setting the state of the corresponding CN node as a non-unloading state, and sending an unloading response message to the CN node.

Existing operation maintenance O & M commands to start/stop offloading are retained in the CN node and/or the RAN node.

The above description is only exemplary of the present invention and should not be taken as limiting the scope of the present invention, and any modifications, equivalents, improvements and the like that are within the spirit and principle of the present invention should be included in the present invention.

Claims

1. A method for realizing automatic load redistribution is characterized by comprising the following steps:

A. when a core network CN node starts unloading, informing a radio access network RAN node;

2. The method according to claim 1, wherein an offload initiation threshold for the CN node to determine whether to initiate an offload process is preset;

3. The method of claim 1,

the method for notifying the RAN node in step a includes:

the unload flag is shown activated.

4. The method according to claim 3, wherein step B specifically comprises:

5. The method of claim 3, wherein the offloading information further comprises: the empty network resource identifies null-NRI.

6. The method according to claim 5, wherein step B specifically comprises:

the RAN node judges whether the received unloading request message carries null-NRI or not, if the received unloading request message does not carry null-NRI, the RAN node sets the state of the CN node corresponding to the received global CN node identification as an unloading state and returns a preset unloading response message to the CN node, and the setting result information carried in the unloading response message shows that the unloading response message is successful;

7. The method according to claim 1, wherein an offloading stop threshold for the CN node to determine whether to stop the offloading process is preset; after the CN node enters the offload state, the method further includes:

8. The method of claim 7, wherein the method of stopping the unloading process is:

after the CN node reaches the unloading stop condition, the unloading information comprising an unloading mark indicating the start/stop of unloading and the global mark of the CN node is carried in an unloading request message and sent to the RAN node;

the unload flag is displayed as stopped;

9. The method of claim 1, further comprising: existing operation maintenance O & M commands to start/stop offloading are retained in the CN node and/or RAN node.

10. A system for realizing automatic load redistribution is characterized in that the system comprises a Core Network (CN) node and a Radio Access Network (RAN) node; wherein,

11. The system of claim 10, wherein the CN node is further configured to send a stop offload notification to the RAN node;

12. The system according to claim 10 or 11, characterized in that existing operation and maintenance O & M commands to start/stop offloading are retained in the CN node and/or RAN node.

13. The system according to claim 10, wherein said CN node and RAN node belong to the same pool area.

14. The system according to claim 13, wherein the CN node is: a Mobile Switching Center (MSC) node or a Serving GPRS Support Node (SGSN) node;

15. The system according to claim 10, wherein said CN nodes belong to the same or different operators.