CN101146316A - A method for reallocating load - Google Patents

Abstract

The invention discloses a method for implementing load redistribution. The method comprises: a RAN node receives an initial direct transfer message, wherein the routing parameter type carried by the initial direct transfer message is of direct routing type; and the RAN node selects a servicing CN node from CN nodes in non-unloaded state when CN nodes in unloaded state exist in the belonging domain. In the inventive method, when the routing parameter type carried by the initial direct transfer message is of direct routing type, the RAN excludes the CN nodes in unloaded state and selects the servicing CN node from the CN nodes in non-unloaded state only. Therefore, on one hand, the invention can increase the efficiency of load redistribution and save network resources, on the other hand, the invention can reasonably realize the load redistribution and the selected servicing CN node is valid. Additionally, the servicing CN node is selected based on the load balance principle with regard of the five situations when the routing parameter type is of direct routing type, thereby ensuring the load balance between the CN nodes.

Description

A kind of method that realizes that load heavily distributes

Technical field

The present invention relates to the heavy distribution technique of load, refer to especially under a kind of Iu-Flex networking, realize the method that load heavily distributes.

Background technology

Third generation partner program (3GPP) version 5 (REL5) support the Iu-Flex characteristic, Radio Access Network (RAN) node such as radio network controller (RNC) or base station controller (BSC), can connect mobile switching centre (MSC) in a plurality of core net (CN) node such as circuit switching (CS) territory and the service GPRS support point (SGSN) in packet switching (PS) territory, here MSC is also referred to as MSC server (MSC Server), hereinafter adopts MSC.For Iu-Flex (Routing functionality forintra domain connection of RAN nodes to multiple CN nodes), the meaning is meant " being applied to the routing function that is connected between a plurality of CN nodes within RAN node and the same CN territory ", and those skilled in the art are direct use Iu-Flex usually.A pool area is formed in the zone that several RAN cover, and a pool area provides service concurrently by several CN nodes usually, also can provide service by a CN node.When a pool area provided service concurrently by several CN nodes, it was MSC pond or SGSN pond that these CN nodes are promptly formed a CN node pool, the notion of the pool area that Here it is says usually.

CS territory and PS territory are fully independently, and permission CS enables Iu-Flex in the territory but PS does not enable Iu-Flex, and vice versa.

Fig. 1 is a network topology structure schematic diagram of supporting Iu-Flex, as shown in Figure 1, the RAN node service area that connects some MSC/SGSN jointly is the set of all sub-districts of RNC control, constitutes CS Chi Qu/PS Chi Qu respectively, wherein, the set of corresponding MSC/SGSN is exactly MSC pond/SGSN pond.

For CN node of sign in a CN node pool, each CN node is assigned with one or several network resource identifiers (NRI), and by Operation and Maintenance (O﹠amp; M) operation is configured in the NRI that distributes in the CN node.When the CN node is certain user terminal (UE) distribution Temporary Mobile Subscriber Identity (TMSI) or packet TMSI (P-TMSI), simultaneously the NRI of this CN node sign is carried at and sends to this UE among the TMSI/P-TMSI, specify a CN that service is provided so just for this UE, this CN can be called service CN.Here, if the CN node is MSC, then be UE distribution T MSI; If the CN node is SGSN, then distribute P-TMSI for UE.

Because MSC and SGSN in the district of pond all can provide service for the user, therefore, in the UE access procedure, when RNC receives initial direct-sending (Initial Direct Transfer) message from UE, carry out the selection of MSC/SGSN node, promptly select (IDNNS, Intra Domain NAS node Selector) information as NAS node in the zone based on the routing parameter in the initial direct-sending message, and the configuration information among the RNC, select one of them MSC or SGSN.In the 3GPP standard, the function of determining the process correspondence of service CN node is called Non-Access Stratum (NAS) node selection function (NNSF), is called initial NAS message route usually again.

Here, the IDNNS information of carrying in the initial direct-sending message can be from TMSI, or P-TMSI, or IMSI International Mobile Subscriber Identity (IMSI), or derives in the International Mobile Station Equipment Identification (IMEI).If there is available TMSI/P-TMSI, then UE preferentially derives routing parameter from TMSI/P-TMSI; If there is not available TMSI/P-TMSI, then UE calculates routing parameter based on IMSI; If it is that UE can't obtain IMSI that GSM subscriber identity module/universal subscriber identity module (SIM/USIM) does not exist, then UE calculates routing parameter based on IMEI.Before UE sent initial direct-sending message, UE will be based on mentioned above principle, and filling in IDNNS information is the TMSI/P-TMSI routing parameter, or the IMSI routing parameter, or the IMEI routing parameter.

According to the current version standard, effectively routing parameter has 6 kinds, and is as follows:

1) derivation is from routing parameter localPTMSI, TMSIofsamePLMN and the tMSIofdifferentPLMN of TMSI/P-TMSI.

The 3GPP standard is expanded 32 TMSI/P-TMSI at the Iu-Flex characteristic, and wherein 23～14 (from high to low) bit is used to identify NRI, the NRI variable-length, and maximum length is 10bit.If UE derives routing parameter from TMSI/P-TMSI, then routing parameter is the bit23 among the TMSI/P-TMSI, bit22 ..., bit14,10bit altogether.During from TMSI/P-TMSI derivation routing parameter, the LAI/RAI when being assigned with TMSI/P-TMSI according to UE under the sub-district, and LAI/RAI under the current service cell can obtain three kinds of different route-types:

If LAI/RAI is identical with the affiliated LAI/RAI of current service cell under the sub-district when 1. UE is assigned with TMSI/P-TMSI, then the routing parameter type is localPTMSI;

If LAI/RAI is different with current service cell LAI/RAI under the sub-district when 2. UE is assigned with TMSI/P-TMSI, but PLMN is identical, then the routing parameter type is TMSIofsamePLMN;

If PLMN is different with the affiliated PLMN of current area under the sub-district when 3. UE is assigned with TMSI/P-TMSI, then the routing parameter type is tMSIofdifferentPLMN.

2) derivation is from routing parameter iMSIcauseUEinitiatedEvent and the iMSIresponsetopaging of IMSI.

If UE derives routing parameter from IMSI, then routing parameter is IMSIV value=(IMSI div10) mod 1000, but according to current calling is caller or called, the routing parameter type is different: if caller, then the routing parameter type is iMSIcauseUEinitiatedEvent, if called (triggered by paging), then routing parameter is iMSIresponsetopaging.

3) derivation is from the routing parameter iMEI of IMEI.

If UE derives routing parameter from IMEI, then routing parameter is IMEIV value=(IMEI div10) mod 1000, and the routing parameter type is iMEI.

According to 3GPP standard 25.331 and 23.236, the RAN node carries out the route of UE initial direct-sending message according to following rule, realizes NNSF:

1) the routing parameter type is iMEI.

The RNC that enables Iu-Flex is provided with the corresponding relation of IMEIV value and CN node in advance, and IMEIV value total size is 0～999.Suppose that IMEIV value scope is configured to corresponding relation between the MSC: 0～399 corresponding MSC1,400～799 corresponding MSC2,800～999 corresponding MSC3, and the IDNNS parameter of carrying in the initial direct-sending message of UE indication routing parameter type is that iMEI, routing parameter are 800 (i.e. routing parameter=1100100000 of 10 bits sign), and then RNC selects the service MSC of MSC3 as UE.

2) the routing parameter type is iMSIcauseUEinitiatedEvent and iMSIresponsetopaging.

According to standard 25.413, when not comprising TMSI in MSC issues the beep-page message of RNC, RNC can start timer TNNSF, and preserves the IMSI that is carried in the beep-page message and the corresponding relation of CN node before timer TNNSF is overtime.

If the IDNNS parameter of carrying in the initial direct-sending message of UE indication routing parameter type is iMSIresponsetopaging, then RNC will search corresponding CN node according to IMSI, and after searching successfully with it as service CN node.

Enable the pre-configured IMSIV value of RNC of Iu-Flex and the corresponding relation of CN node, IMSIV value total size is 0～999.If the IDNNS parameter of carrying in the initial direct-sending message of UE indication routing parameter type is iMSIcauseUEinitiatedEvent, then processing mode is the processing mode of iMEI with the route parameter type.

3) the routing parameter type is localPTMSI, TMSIofsamePLMN and tMSIofdifferentPLMN.

Enable the pre-configured NRI length of RNC of Iu-Flex, and the corresponding relation of NRI and CN node.The NRI in CS territory and PS territory configuration is independently, i.e. CS territory NRI length and value, and between the NRI length in PS territory and the value without any constraint, can be the same or different.

According to standard, RNC uses the NRI length coupling routing parameter in corresponding CN territory according to the CN domain identifier in the initial direct-sending message, searches corresponding CN node according to the NRI that obtains again.With the CS territory is example, suppose that CS territory NRI with corresponding relation between the MSC is: NRI=1 and MSC1 are corresponding, NRI=2 is corresponding with MSC2, NRI=3 is corresponding with MSC3, NRI length is 3bit simultaneously, then when routing parameter=298 (binary form is shown 0100101010), begin to get 3bit from a high position and promptly 010 obtain NRI=2, search the corresponding relation of default NRI and CN node, can get target CN node is MSC2, and promptly MSC2 is service MSC.

In addition, when occur the unavailable as fault of selecteed CN node or according to the NRI that obtains can not find the CN node of correspondence, in the time of maybe can't obtaining NRI, according to standard, to select the CN node by the load balancing principle, the load balancing principle is based on certain algorithm, load-balancing algorithm depends on the realization of particular vendor equipment, comprises that mainly pro-rata and poll distribute dual mode.For the RAN node, the status of each MSC in the CN pond is identical, and difference only is the capacity difference of each CN node.For the load balancing (Load Balancing) that keeps each node in the CN pond, the REL5 standard has proposed the load balancing principle, hope is in the long-time running of network, the load of each the CN node in the CN pond be through reasonable distribution promptly be balanced, but do not provide the specific implementation method of load balancing principle.

Because the RNC of different vendor is to the implementation method difference of load balancing, may cause after operation after a while, the load of each MSC is unbalanced in the MSC pond, need transfer on the lighter MSC of other load by the certain customers on the manual intervention MSC that load is heavier.Simultaneously, when under other special scenes as to certain MSC, upgrading, also need to migrate out the part or all of user on this MSC, therefore, the REL6 standard has proposed load and has heavily distributed (LoadRe-Distribution) function, makes the higher MSC of load a part of user wherein can be transferred to other MSC in the district of pond.

In order to realize the heavy distribution function of load, standard has been introduced two signs: empty NRI (null-NRI) and Non-broadcast Location Area Identification/Routing Area Identifier (Non-broadcast LAI/RAI).Wherein, null-NRI is the unique identification of public land mobile net network (PLMN) scope, by operator's unified planning in whole network, do not have fixing corresponding relation between this null-NRI and the CN node, but need in RNC and CN node, dispose this null-NRI information; Non-broadcast LAI/RAI has uniqueness in the district of a pond, and and the pond district in the CN node between have one-to-one relationship, wherein, in the CS territory, Non-broadcast LAI is corresponding one by one with certain MSC, and in the PS territory, Non-broadcast RAI is corresponding one by one with certain SGSN.All CN nodes all need to dispose the corresponding relation of other CN node and Non-broadcast LAI/RAI in this pond district in the district of pond.Can no longer describe in detail here referring to 23.236 standard relevant regulations about specific definition null-NRI and Non-broadcast LAI/RAI.

Null-NRI and common NRI do not have essential distinction, just its value has specific implication and promptly shows and have the MSC that is in unloading (off-load) state, for the UE initial direct-sending message that has comprised Null-NRI in TMSI, RNC will foreclose the MSC that is in unloaded state when selecting MSC.Whether be in unloaded state in order to allow RNC obtain MSC, needing to propose by specific mode such as standard can be by carrying out O﹠amp on RNC; The M operation, the state of MSC is set to unloading.

Non-broadcast LAI is a specific LAI such as a PLMN ID+LAC, and Non-broadcast LAI does not broadcast in the sub-district that RNC controlled, and promptly UE can not obtain Non-broadcast LAI by read system message.Non-broadcast RAI is similar to Non-broadcast LAI.

Unload the MSC that user soon shifts away for needs, can on MSC, carry out O﹠amp by specific mode such as standard proposition; The M operation starts after the unloading operation, will take following action:

(1) for the UE that carries out location area updating, MSC will accept to carry in the message Null-NRI and the Non-broadcast LAI of distribution at the location area updating that returns to UE; After UE obtained Non-broadcast LAI, because Non-broadcastLAI must be a new LAI for UE, so UE can initiate location area updating immediately once more, and the Null-NRI that carries acquisition in initial direct-sending message is as routing parameter.

(2) for new calling, MSC will accept and handle calling, and carry out the heavy assigning process of TMSI, distribute Null-NRI and Non-broadcast LAI; After the end of conversation, UE will be the Null-NRI of its distribution according to MSC, initiate the location area updating process immediately.

(3) for ongoing calling, when MSC starts unloading operation, can carry out the heavy assigning process of TMSI to the UE that is in the communication process immediately, distribute Null-NRI and Non-broadcastLAI; Behind the UE end of conversation, will be the Null-NRI of its distribution according to MSC, initiate the location area updating process immediately.

In heavy assigning process, also just mean at startup unloading operation on the MSC: the user should be transferred on other MSC as early as possible.

According to the 3GPPREL6 regulation and stipulation, RNC supports to carry out O﹠amp; Certain MSC status that the M operation connects is set to unloaded state, simultaneously, when RNC receives the initial direct-sending message that carries Null-NRI, RNC will check whether there is the MSC that is in unloaded state, and select service MSC among the MSC of and non-unloaded state available at all based on the load balancing principle.

Fig. 2 is in the CS territory, when carrying Null-NRI in the initial direct-sending message, existing RNC selects the schematic diagram of MSC, as shown in Figure 2, suppose that MSC1 is set to unloaded state, MSC2 and MSC3 are not available and be in non-unloaded state, and then RNC is at MSC2 and MSC3, select service MSC according to the load balancing principle, the initial direct-sending message that carries Null-NRI is distributed to MSC2 or MSC3.

In the heavy assigning process of load, standard is only at the situation of carrying Null-NRI in initial direct-sending message at present, the clear and definite treatment principle of RNC, the MSC that is about to be in unloaded state gets rid of outside range of choice, selects service MSC in the MSC that is in available and non-unloaded state.But do not provide processing method for following situation, as follows:

The routing parameter type of 1) carrying in initial direct-sending message is localPTMSI or tMSIofsamePLMN, but does not exist when unavailable with NRI (non-Null-NRI) corresponding MSC or corresponding MSC;

2) when the routing parameter type of carrying in the initial direct-sending message is tMSIofdifferentPLMN;

The routing parameter type of 3) carrying in initial direct-sending message is iMSIcauseUEinitiatedEvent or iMEI, but when not existing the MSC corresponding with IMSI/IMEI V value or corresponding MSC unavailable;

The routing parameter type of 4) carrying in initial direct-sending message is iMSIresponsetopaging, but when searching the failure of IMSI and MSC corresponding relation;

5) when the routing parameter type of carrying in the initial direct-sending message is spare2 or spare1, wherein routing parameter type spare2 or spare1 represent the unallowed type of current version.

Because the existing unqualified processing method at RNC under above-mentioned five kinds of situations of standard, so RNC may still select to be in the MSC of unloaded state as service MSC.Fig. 3 is in the CS territory, when not carrying Null-NRI in the initial direct-sending message, existing RNC selects the schematic diagram of MSC, as shown in Figure 3, suppose that MSC 1 is set to unloaded state, MSC2 and MSC3 are available and are in non-unloaded state, owing to do not carry NullNRI in the initial direct-sending message, then RNC can be in MSC1, MSC2 and MSC3, select service MSC according to the load balancing principle, and because the current unloaded state that is in of MSC1, very big may be to select MSC1 that the unloading back loading lightens as service MSC, clearly, the method for this selection service MSC is irrational.Herein, the routing parameter type with above-mentioned five kinds of situations is called direct route-type.

And, if location area updating process, be that the NAS message of carrying in the initial direct-sending is location area updating (Location area update) message, then Xuan Ding service MSC need handle the location area updating process, and accepts to distribute in the message Null-NRI and Non-broadcast LAI at location area updating; If process is set up in conversation, the NAS message of promptly carrying in the initial direct-sending message is to set up (setup) message, and then MSC need carry out the heavy assigning process of TMSI, also must handle conversation etc. simultaneously.In a word, still be selected as serving MSC, reduced the efficient that load heavily distributes, and wasted Internet resources owing to started the MSC of offloading functions.

In addition,,, do not provide route processing method, cause the load of CN node unbalanced so possibly, illustrate as follows for above-mentioned five kinds of direct route-types according to 3GPP standard 25.331 and 23.236:

Fig. 4 is the embodiment schematic diagram to the route of direct route-type, as shown in Figure 4, supposes the boundary at PLMN-A and PLMN-B, and the NRI length of PLMN-A network is 3bit, and the NRI of MSC1, MSC2 and MSC3 is configured to 5,6 and 7 respectively; The NRI length of PLMN-B network is 4bit, and the NRI of MSCx, MSCy and MSCz is configured to 10,11 and 12 respectively.

User in the PLMN-B zone comes distribution T MSI by MSCx, MSCy, MSCz respectively, because NRI length is 4bit, therefore, the bit23 of user's TMSI in the PLMN-B zone, bit22, bit2 1 and bit20 may be for 1010 being decimal system NRI=10, or 1011 be decimal system NRI=11, or 1100 is decimal system NRI=12.When the TMSI user who carries the PLMN-B region allocation moves in the PLMN-A zone, and when making a call, suppose that the routing parameter type among the IDNNS is tMSIofdifferentPLMN, high 4 of routing parameter is possible be 1010 (NRI=10), or 1011 (NRI=11), or 1100 (NRI=12), corresponding MSCx, MSCy or MSCz respectively.According to existing standard, when the RNC among the PLMN-A receives message related to calls, will use NRI length 3 coupling routing parameters, promptly obtain the high 3bit in the 10bit routing parameter.Obviously, be 1010 from high 4bit, or 1011, or the result who gets high 3bit in 1100 the routing parameter is 101, or 101, or 110, promptly obtain NRI=5 or 6, search corresponding CN node, not MSC1 then, be exactly MSC2.

As can be seen, along with the user moves to the PLMN-A zone from the PLMN-B zone, will be routed to MSC1 from the user among MSCx, the MSCy, will be routed to MSC2 from the user among the MSCz, find out easily, caused the load between MSC1, MSC2, the MSC3 very unbalanced so undoubtedly.

Summary of the invention

In view of this, main purpose of the present invention is to provide a kind of method that realizes that load heavily distributes, and can choose reasonable serve the CN node, improves the efficient that load heavily distributes, conserve network resources.

For achieving the above object, technical scheme of the present invention specifically is achieved in that

A kind of method that realizes that load heavily distributes, this method comprises:

Radio Access Network RAN node receives initial direct-sending message, the routing parameter type of carrying in initial direct-sending message is direct route-type, and exist in the pool area under the RAN node when being in the core net CN node of unloaded state, in being in the CN node of non-unloaded state, select service CN node.

Described direct route-type comprises:

The routing parameter type is localPTMSI or tMSIofsamePLMN, and does not exist the CN node corresponding with network resource identifier NRI or corresponding CN node unavailable;

Perhaps, the routing parameter type is tMSIofdifferentPLMN;

Perhaps, the routing parameter type is iMSIcauseUEinitiatedEvent or iMEI, and does not exist the CN node corresponding with the IMSI/IMEIV value or corresponding CN node unavailable;

Perhaps, the routing parameter type is iMSIresponsetopaging, and searches IMSI International Mobile Subscriber Identity IMSI and the failure of CN node corresponding relation;

Perhaps, the routing parameter type is unallowed spare2 or spare1 type.

The method of described selection service CN node is: adopt the load balancing principle to select available CN node to be service CN node.

Described load balancing principle is pro-rata, or poll distributes.

Also comprise before this method: described RAN node is safeguarded O﹠amp by executable operations; The M operation, the state that unloads the CN node under self in the pool area is set to unloaded state.

Described RAN node is: radio network controller (RNC), or base station controller BSC;

Described CN node is: moving exchanging center MSC, or service GPRS support point SGSN.

As seen from the above technical solution, in the inventive method, when the routing parameter type of carrying in the initial direct-sending message is direct route-type, the RAN node is when selecting service CN node, the CN node that will be in unloaded state forecloses, and only in being in the CN node of non-unloaded state, select available CN node as service CN node, and improved the efficient that load heavily distributes on the one hand, saved Internet resources; Realized reasonably also that on the other hand load heavily distributes, can use during the service CN node selected.

Simultaneously, be five kinds of situations of direct route-type for the routing parameter type, adopt the load balancing principle to select service CN node, guaranteed the internodal load balancing of CN.

Description of drawings

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

Fig. 2 is in the CS territory, and when carrying Null-NRI in the initial direct-sending message, existing RNC selects the schematic diagram of MSC;

Fig. 3 is in the CS territory, and when not carrying Null-NRI in the initial direct-sending message, existing RNC selects the schematic diagram of MSC;

Fig. 4 is the embodiment schematic diagram to the route of direct route-type;

Fig. 5 is the flow chart that the present invention realizes the embodiment that load heavily distributes.

Embodiment

Core concept of the present invention is: the RAN node receives initial direct-sending message, the routing parameter type of carrying in initial direct-sending message is direct route-type, and exist in the pool area under self when being in the CN node of unloaded state, in being in the CN node of non-unloaded state, select service CN node.

For making purpose of the present invention, technical scheme and advantage clearer, below with reference to the accompanying drawing preferred embodiment that develops simultaneously, the present invention is described in more detail.

Fig. 5 is the flow chart that the present invention realizes the embodiment that load heavily distributes, and as shown in Figure 4, specifically may further comprise the steps:

Step 500～step 502:RAN node receives initial direct-sending message, judges whether the routing parameter type of carrying in the initial direct-sending message is direct route-type, if, then enter step 503, otherwise handle according to existing procedure, withdraw from flow process of the present invention.

In this step, directly route-type is and mentions in the background technology:

LocalPTMSI or tMSIofsamePLMN, and do not exist CN node corresponding or corresponding CN node unavailable with NRI, or tMSIofdifferentPLMN, or iMSIcauseUEinitiatedEvent or iMEI, and do not exist CN node corresponding or corresponding CN node unavailable with IMSI/IMEI V value, or iMSIresponsetopaging, and search IMSI and the failure of CN node corresponding relation, or spare2 or spare1.

Step 503:RAN node judges under self whether have the CN node that is in unloaded state in the pool area, if exist, enters step 504; Otherwise enter step 505.

According to regulation and stipulation, RAN node such as RNC or BSC can be by carrying out O﹠amp; The M operation, the state that unloads the CN node under self in the pool area is set to unloading.

Step 504:RAN node is selected service CN node in being in the CN node of non-unloaded state.

When the routing parameter type of carrying in the initial direct-sending message is direct route-type, the RAN node is when selecting service CN node, the CN node that will be in unloaded state forecloses, select available CN node as service CN node and be intended to be in the CN node of non-unloaded state, improve the efficient that load heavily distributes on the one hand, saved Internet resources; Realized reasonably also that on the other hand load heavily distributes, can use during the service CN node selected.

Step 505:RAN node adopts the load balancing principle to select service CN node in all CN nodes.

In this step, be five kinds of situations of direct route-type, adopt the load balancing principle to select service CN node, guaranteed the internodal load balancing of CN for the routing parameter type.

Why adopt load balancing be because:

1) derives from IMSI (caller) or IMEI at routing parameter, but can not find the situation of the CN node of correspondence according to IMSI/IMEI V value.

Because IMSI/IMEI V value scope is determined 0～999, usually as long as will set up corresponding relation between each numerical value in 0～999 and the CN node, but along with operations such as the construction of network, dilatation, capacity allotments, may need to revise at any time IMSI/IMEI V value and the internodal corresponding relation of CN, operate more loaded down with trivial details, and make a fault easily, some IMSI/IMEI V value is interval to be disposed with the internodal corresponding relation of CN such as omitting; The client does not think the configuration of corresponding relation between maintaining IMS I/IMEI V value and CN node on the other hand, as long as can reach the purpose of load balancing between the CN node.

Therefore, when as long as RNC can not find corresponding CN node according to IMSI/IMEI V value, still can also carry out route according to the principle of load balancing, like this, can evade the adverse consequences that the configuration error causes on the one hand, allow the user not need to be concerned about IMSI/IMEI V value and the internodal corresponding relation of CN on the other hand, improved efficiency of operation greatly.

2) derive from IMSI (called) at routing parameter, but search the situation that IMSI-CN node corresponding relation is failed according to IMSI.

The reason that the corresponding relation failure occurs may be the weak point of TNNSF timer duration than actual needs, perhaps other abnormal cause.When this situation occurring, be the initial direct-sending message that can not abandon UE, so should select service CN node based on the principle of load balancing.

3) derive from TMSI/P-TMSI at routing parameter, and the routing parameter type is the situation of tMSIofdifferentPLMN.

Owing to not in same PLMN scope, in fact go to select the CN node to lose practical significance, make that finally load is extremely unbalanced between the CN node by NRI.Reasonably processing mode should be to adopt the principle that the present invention is based on load balancing to select service CN node.

4) at early protocol version UE in initial direct-sending message, carry the coding uncertain IDNNS parameter situation.

Because RNC can't determine whether the IDNNS parameter effectively is that the IDNNS parameter is uncertain coding according to the UE protocol version information, according to the coding rule of the current protocol version routing parameter type that the back obtains of decoding is the unallowed type of current version, be spare2 or spare1, this is, can be by select service CN node according to the load balancing principle, with the early stage UE of compatibility.

The above is preferred embodiment of the present invention only, is not to be used to limit protection scope of the present invention, all any modifications of being made within the spirit and principles in the present invention, is equal to replacement, improvement etc., all should be included within protection scope of the present invention.

Claims (6)

1. a method that realizes that load heavily distributes is characterized in that, this method comprises:

Radio Access Network RAN node receives initial direct-sending message, the routing parameter type of carrying in initial direct-sending message is direct route-type, and exist in the pool area under the RAN node when being in the core net CN node of unloaded state, in being in the CN node of non-unloaded state, select service CN node.

2. method according to claim 1 is characterized in that, described direct route-type comprises:

The routing parameter type is localPTMSI or tMSIofsamePLMN, and does not exist the CN node corresponding with network resource identifier NRI or corresponding CN node unavailable;

Perhaps, the routing parameter type is tMSIofdifferentPLMN;

Perhaps, the routing parameter type is iMSIcauseUEinitiatedEvent or iMEI, and does not exist the CN node corresponding with IMSI/IMEI V value or corresponding CN node unavailable;

Perhaps, the routing parameter type is iMSIresponsetopaging, and searches IMSI International Mobile Subscriber Identity IMSI and the failure of CN node corresponding relation;

Perhaps, the routing parameter type is unallowed spare2 or spare1 type.

3. method according to claim 1 is characterized in that, the method for described selection service CN node is: adopt the load balancing principle to select available CN node to be service CN node.

4. method according to claim 3 is characterized in that, described load balancing principle is pro-rata, or poll distributes.

5. method according to claim 1 is characterized in that, also comprise before this method: described RAN node is safeguarded O﹠amp by executable operations; The M operation, the state that unloads the CN node under self in the pool area is set to unloaded state.

6. according to each described method of claim 1～5, it is characterized in that described RAN node is: radio network controller (RNC), or base station controller BSC;

Described CN node is: moving exchanging center MSC, or service GPRS support point SGSN.

Images