METHOD FOR CONSIDERING THE SUBSCRIBER TYPE IN MOBILITY AND RADIO RESOURCE MANAGEMENT DECISIONS IN A RADIO ACCESS NETWORK

Abstract

In order to efficiently allow mobility and radio resource management decisions on mobile networks the deciding entity (typically the radio access network) needs some understanding about the type of mobile user. In current cellular systems the radio network has only limited to null information about subscription dependent details of the mobile users. The proposed invention relates to a mobile system where a generic information container is defined which is applicable per UE in order to convey subscription related information from the core network to the radio access network which can be used to steer the user to an appropriate cell or for any other Radio Resource Management (RRM) related procedure/decision. It is proposed to define an information element (herein referred as “subscriber type”) which allows an indication or pointer to a set of rules which are locally defined in involved no'des (e.g. CN SGSN, MSC, routers etc) and describe the “handling rules” to be applied in the RAN for each user of the mobile system independently.

Description

BACKGROUND OF THE INVENTION

With the ongoing standardisation of 3GPP long term evolution (LTE or E-UTRAN) and WiMAX (IEEE802.16) air interfaces, in addition to existing standards (GSM/GPRS and UMTS) an increasing diversity of radio access technologies (RATs) will be present in future, Operators will require sophisticated mobility mechanisms to simultaneously steer user to appropriates RATs while also spreading load across the different technologies.

In today's wireless systems like GSM/GPRS or UMTS, typically UE-controlled cell reselection algorithms currently operate independently of any subscriber-specific or service considerations [for UMTS e.g. in 3GPP TS 25.304]. All UEs are handled in the same way based on parameter broadcast on the BCCH [for UMTS e.g. in 3GPP TS 25.331], and hence follow the same cell reselection strategies. Similar as for the idle mode the decision criteria for both mobility related and radio resource related decisions are independent from any subscriber or user type information. So in general the decisions is based on terminal capabilities and to some extend on current service usage or specific information provided by the CN during the service establishment (e.g. so-called “service based handover” which gives an indication from the CN to the RAN that a certain bearer might be handed over to a specific RAT or not). This all relates for the demand of providing similar service equally for all users of the mobile system.

With the advent of a demand for increased datarates and the cost reduction implied by specific tariffs of the mobile operators an increasing demand is foreseen for user as well as service differentiation for example in mobility decisions. Qualifying data by means of Quality of Service (QoS) can provide some means to differentiate the subscribers but mobility related decisions (especially while being in idle mode) can't reliably be taken based on this kind of information designed for differentiating traffic of different services using prioritised packet delivery rather than independent tailored to control mobility based upon the type of user.

3GPP have already specified access control classes to restrict access to mobile networks to a set of subscribers belonging to a set of access control classes during severe network congestion. It is proposed that a similar kind of user differentiation is available within the RAN for mobility decisions so that users can be spread appropriately across RATs.

In addition to load distribution across RATs, similar mechanisms may also be required to distribute load across different carriers of different bandwidths on the same RAT.

SUMMARY OF THE INVENTION

It is proposed to allow a procedure where an entity of the CN (e.g. aGW in case of SAE/LTE) informs the radio access network (e.g. the eNB in case of LTE/SAE) on registration of a UE or transition from LTE_IDLE to LTE_ACTIVE state about subscriber specific information which in turn should be considered in the radio access network for mobility and radio resource management decisions, both in idle and active mode. The information is preferably defined as an integer number which is associated with the specific handling of the UE defined by a set of rules. These rules will be defined locally in a eNB (or any other involved node) by means of configuration files (e.g. using mainstream dataformats like XML, etc) and will be applied by the eNB during the mobility and radio resource management.

DETAILED DESCRIPTION OF THE INVENTION

The signalling from the CN (200) to the RAN (201) at registration of a UE or during the transition from LTE_IDLE to LTE_ACTIVE state (see FIG. 1) will include a new “Subscriber Type” information element (in 301) that can be used in mobility or other radio resource management decisions by the network (302). It could also be used to signal towards that UE that the UE has to follow only a specific part of the system information broadcast (on the BCCH) in order to allow user differentiation also in terminal based decisions (e.g. for cell reselection purpose).

The coding of the information element should flexible enough to allow its application to a wide range of mobility applications. To allow sufficient granularity it is recommended that at least 16 classes are defined,

One example of the use of the subscriber type field for load balancing between UMTS and LTE RATs is shown in FIG. 2. Assuming the LTE load is greater that a certain threshold, it is desirable to spread users across the different RATs based upon both the services used and the subscriber type.

Some services demanding especially-high data rates or minimum latency are handled more appropriately on the more higher performing LTE RAT, independent of subscriber type. However, the majority of services are well handled on both RATs, and the RAT selection for these is predominantly based upon subscriber type. In this example, 8 classes of subscriber type are shown (number restricted for illustrative purposed only), and for 3 subscriber types, most services are selected to be handled on the UMTS rather than the LTE RAT. Making subscriber type the dominant criterion rather than service reduces the amount of inter-RAT mobility triggered by different service use, and also allows a more consistent RAT selection policy in IDLE and ACTIVE modes (FIG. 2).

The number of subscriber type classes selected for handling on the UMTS or LTE RATS can be made dependent on the load levels in both RATs. In the case of low LTE load, it is beneficial to handle all users on the better performing LTE RAT. At higher loads, it is possible to alter the number of subscriber type classes directed to the different RATs depending upon the relative load on the two RATs. Another example of the subscriber type information used in mobility decisions of a cellular network is the case of network sharing or national roaming. In such a case typically on RAT (for example LTE) is shared among different mobile operators while the technologies are not shared to the legacy reasons (e.g. UMTS or GSM). Typically the operator is interested getting the own subscriber back the the own network whenever possible. E.g. if the coverage of the shared LTE network is left then the own subscriber should be preferably handed over to the right target network though both might fulfil the requirements for a HO target in terms of radio conditions. In such a case a “subscriber type” indication received form the CN could be used to bring the correct subscriber in the correct network. In today implementations this requirement can only be fulfilled by the analysis of an operator specific identity available in the RAN (namely the IMSI available in the RNC in UMTS, but not in the BSC of GSM networks). Having a RRM rule defined in the decisions nodes for always performing a handover of subscriber type X (from PLMN B) back to PLMN B whenever possible would be very flexible and could be enhanced/changed depending on any subscription information available in the CN without requiring any change to existing RAN implementations. Also security would benefit for the fact that no subscriber specific information is available in certain nodes of a network which might be vulnerable (e.g. the complete IMSI of a subscriber which can identify his identity). Having only an abstract “subscriber type” information and a defined “handling rule” available would ensure the subscribers confidentiality in all cases.

Another application for the usage of a “subscriber type” information could be related to the specific usage conditions of a terminal in a cellular network. E.g. today in all cellular network that decision functions in the network/RAN do not have any information how a specific terminal is used. E.g. it is not possible for the network with simple means to detect e.g. if a terminal is a stationary device such as a part of a machine or a wireless gateway router etc. The only means today to get this information is either to introduction of a specific signalling or the usage of radio measurements to determine if a terminal is stationary or not. Specifically the operation of the network with stationary wireless gateway routers (providing access of small home or office networks wireless to the internet) as well as stationary machine-to-machine devices (gas/water counter) could benefit for the additional information about the fact that the device is neither moving nor has strict power constrains as it is typically connected to a fixed power supply: the network can take specific actions for such devices it would not take for the majority of mobile devices in the cellular network. One example could be to apply a specific timer for the LTE_ACTIVE to LTE_IDLE state transitions (which is either done to free resources or to put the device in state where less power is consumed) or it could completely disable any measurements for mobility to be applied by the device as it would never need to perform a handover to a different cell. The above listed examples are not exclusive, one can consider even more scenarios where such information about the subscriber type could be beneficial.

Claims

1. A method which allows a radio access network (RAN) to selectively influence the mobility decisions as well as the radio resource control decisions for a specific terminal based on information received from a core network (CN) entity.

2. A method as in claim 1. wherein the CN provides an indication to the RAN on registration of an UE which is subscriber specific.

3. A method as in claim 1. wherein the CN provides an indication to the RAN on transition of a mobile from IDLE to ACTIVE mode which is subscriber specific.

4. A method as in claim 1. wherein the information provided from the CN to the RAN is based on an index (“subscriber type”).

5. A method as in claim 1. wherein the CN chooses the appropriate index (“subscriber type”) from information available from the HLR.

6. A method as in claim 1. wherein the CN chooses the appropriate index (“subscriber type”) from information available from the any other instance of the CN.

7. A method as in claim 1. wherein the CN chooses the appropriate index (“subscriber type”) from information not related to subscription, but related to terminal capabilities.

8. A method as in claim 1. wherein the CN chooses the appropriate index (“subscriber type”) from information available from any other source.

9. A method as in claim 1. wherein the CN chooses the appropriate index (“subscriber type”) from information depending on the users choice.

10. A method as in claim 1. wherein the CN chooses the appropriate index (“subscriber type”) from information available from usage history of the mobile subscriber.

11. A method as in claim 1. wherein the CN send the information to the RAN in the same message as the message which triggers the transition of the mobile station from IDLE to ACTIVE.

12. A method as in claim 1. wherein the configuration of the “handling rules” is defined locally in each involved mode.

13. A method as in claim 1. wherein the configuration of the “handling rules” is defined locally in each involved mode by using an protocols which is used specifically for this purpose.

14. A method as in claim 1. wherein the configuration of the “handling rules” is defined externally by a server and distributed to the involved nodes by means of “handling labels”

15. A method as in claim 1. wherein the configuration of the “handling labels” is defined externally by a server and distributed to the involved nodes by means of e.g. XML or any other data format.

16. A radio access network with defined “handling rules” which acts during mobility and radio resource management decisions according to the indexed “handling rules” for mobiles in ACTIVE mode.

17. A radio access network with defined “handling rules” which acts during mobility and radio resource management decisions according to the indexed “handling rules” for mobiles in IDLE mode. In order to do so, the RAN will identify and group mobiles and assign “handling rules” to those mobile belonging to the same group of subscriber (same index).

18. A radio access network with defined “handling rules” which provide different “mobility rules” by allocation different cell reselection parameter sets to different groups of UEs depending on the subscriber information/index available per UE.

19. A radio access network with defined “handling rules” which acts during mobility and radio resource management decisions and signals to selective terminals an information which allows this terminal to selectively read the system information on the BCCH.

20. A procedure where the information of which part of the system information of the BCCH is relevant is used by the terminal to apply differentiated mobility decisions/procedures as if another part of the information from the BCCH would have been read.

21. A method as in claim 1. where the handling rules depend on subscription information requested from the HLR, as well as service, load and other radio related information.

22. A method as in claim 1. where the handling rules depend on subscription information requested from the PCRF , as well as service, load and other radio related information.

23. A method as in claim 1. where the handling rules depend on the preference of a user, as well as service, load and other radio related information.

24. A method as in claim 1. where the handling rules depend on the history of the users activity (e.g. if only voice calls were placed/received during the last X hours/days), as well as service, load and other radio related information.

25. A method as in claim 1. where the handling rules depend on the type of terminal the user uses (e.g. mobile, laptop, home access point (LAN router), vending machine, machine to machine device, . . . ), as well as service, load and other radio related information.