WO2011107886A1

WO2011107886A1 - Method of and apparatus for assisting selection of a network cell of a wireless network

Info

Publication number: WO2011107886A1
Application number: PCT/IB2011/000791
Authority: WO
Inventors: Dongyan Wang; Junyuan Liu; Daqing Gu
Original assignee: France Telecom
Priority date: 2010-03-05
Filing date: 2011-03-04
Publication date: 2011-09-09

Abstract

A method of and apparatus for assisting selection, for a user equipment in an idle mode, of a network cell from a plurality of network cells of a cellular wireless telecommunication network having a plurality of network cells the network cells operating according to different network technologies, the method comprising: recording behavioural information of the user equipment over a period of time; compiling behavioural statistics data based on the recorded behavioural information; generating network cell ranking data for the plurality of the network cells according to the behavioural statistics data and the network technology of each network cell; and transmitting the network cell ranking data to the user equipment.

Description

METHOD OF AND APPARATUS FOR ASSISTING SELECTION OF A NETWORK CELL OF A WIRELESS NETWORK

Field of the Invention

The present invention relates in general to a method and apparatus assisting selection of a network cell of a wireless network for a mobile station.

Background of the Invention

When a user equipment (UE) such as a mobile telephone, PDA with mobile access capability, portable computer or the like is camped on a cell of a wireless network, the UE regularly attempts to detect, synchronize, and monitor intra- frequency, inter-frequency and inter-RAT (radio access technology) cells indicated by the serving cell. If a better cell is found, the better cell is selected and camped on according to the cell reselection criteria.

Cell reselection requires the UE to perform neighboring and serving cell measurements. The main parameter which should be measured is the RSRP

(Reference Signal Received Power) which is measured on reference symbol.

Multi-RAT(Multiple Radio Access Technology ) networks (such as GERAN and UTRAN operated by the same operator) have been widely deployed in different countries worldwide by different operators. Consequently different RATs can be provided in the same geographical area.

In particular when LTE networks are deployed in the near future, one operator could have the opportunity to operate GERAN, UTRAN and E-UTRAN in one geographical area at the same time. Hence when a multiple-mode UE or single mode multiband UE is performing cell reselection, it is important for the UE to choose an appropriate cell to camp on from multiple cells of intra/inter- frequency or inter-RAT.

In order to distribute subscribers to different RATs or frequency carriers according to the operators' policy to further improve the user experience, not only in active mode but in idle mode, a priority concept has been introduced in 3GPP

TS 36.300 release 8 for LTE as a mechanism for inter-frequency and inter-RAT cell reselection. In addition, priority has also been introduced to UTRAN and GERAN for consistency with LTE. A UE is able to carry out cell reselection according to the priority list of cells which could help to decrease unnecessary measurements being performed by the UE and save power consumption of the UE.

Fig.1 shows UE measurement rules and cell reselection procedures based on the priority list according to 3GPP TS 36.304. If a UE is camping on the cell with a frequency band of highest priority, the UE should evaluate the current serving cell in S12/S13. When the serving cell is not able to satisfy the UE requirement, the UE firstly makes measurement for intra-frequency and equal priority inter-frequency cells in step S14, if the cell fulfills cell reselection criteria in step S15 the UE chooses that cell to camp on in step S16, otherwise the UE performs the measurement for the carrier or RAT with lower priority in turn in step S17. If the UE camps on the cell with lower priority in step S20/S21 , the UE measures the frequency carriers or RATs with higher priority periodically in step S23. If the criterion is fulfilled in step S24, the UE reselects to that cell in step S25. Otherwise, the UE will evaluate the current serving cell in S26. When the serving cell is not able to satisfy the UE requirement, the UE shall make measurement for intra-frequency and equal priority inter-frequency cells in step S27, if the cell fulfills cell reselection criteria in step S28 the UE chooses that cell to camp on in step S29, otherwise the UE performs the measurement for the carrier or RAT with lower priority in turn in step S30.

In this procedure:

Srxlev value of step S12 and S22 is defined as the cell selection receiving signal level value (dB). The detailed calculation formula is specified in TS 36.304.

Snonintrasearch of step S13 and S26 is defined as the threshold (in dB) for

E-UTRAN inter-frequency and inter-RAT measurements

Sservingcell of step S13 is the Srxlev value of the serving cell. It should be calculated according to the cell selection criteria as follows

Threshserving, low of step S19 and S32 is defined as the threshold for serving frequency used in reselection evaluation towards lower priority E-UTRAN frequency or RAT.

Threshx, low of step S18 and S31 is defined as the threshold used in reselection towards frequency X priority from a higher priority frequency. Threshx, high is defined as the threshold used by the UE when reselecting towards the higher priority frequency X than current serving frequency.

Cell-ranking criterion is applied for intra-frequency and equal priority inter- frequency cell reselection, which is specified in TS 36.304.

Priority is divided into common priority and specific priority. Common priority is valid for all UEs in a cell, and is specified by system information which will indicate the same priority list to all UEs in the cell. Specific priority is applied to an individual UE according to the said UE's condition, and the specific priorities for UEs in one cell may be different from each other. The specific priority is assigned via a dedicated signaling; the UE ignores the common priorities provided in system information after receiving specific priority information.

At present, no explicit indication on how to generate the specific priority list for cell reselection is provided in 3GPP standards. Generally, cell load condition, UE radio access capability, SPID (Subscriber Profile ID for RAT/Frequency

Priority) and UE location are always seen as the drivers for inter-frequency or RAT cell reselection.

UE radio access capability includes RF band supported, measurement capabilities and RATs supported. This is the basis for specific priority list management. If the neighboring cell with certain layers or RATs indicated in the system information is not supported by UE radio capability, that cell will be excluded from the specific priority list.

In order to realize load balancing mechanism between cells, the cell load condition is always considered for the priority decision. Lower priority will be defined for the cell with overload or high load condition. Hence a UE can avoid camping on that cell when informed of such conditions. Load condition information can be exchanged between base stations through certain interfaces, whether it be intra-system or inter-system.

Load condition can be divided into: Hardware load condition, TNL load condition and Radio load condition.

Hardware Load information indicates the status of the Hardware Load experienced by the cell.

TNL Load information indicates the status of the S1 Transport Network Load experienced by the cell. Radio Resource Status information indicates the usage of the PRBs (Physical Resource Block) in Downlink and Uplink, including GBR (guaranteed bit rate) PRB and non-GBR PRB usage and the total PRB usage. In the context of multiple technologies (UMTS, GSM, LTE) network deployed in the same geographical area, operators generally define that subscribers access GSM for voice service and return to LTE or UMTS network for data services. It is therefore important to define a correct camping solution for a user equipment to camp on the right cell during the idle mode.

To improve end user experience, it is preferable to make the UE camp on the right cell according to the coming service. Otherwise, a noticeable delay will be caused in redirecting the UE from the current camping cell to the serving cell. To facilitate an operator to define priority based camp strategy, SPID(Subscriber Profile ID) has been defined in 3GPP to be used as an input to determine the priority level of the RAT (eg. GSM, UMTS or LTE) or priority level for each frequency of the RAT technology.

However, in the current concept of SPID defined in 3GPP TS 36300, the SPID parameter received by the BS from CN is an index referring to user information, such as mobility profile and service usage profile. According to 3GPP R2-080972, mobility profile refers to the information of the UE's moving speed. Three statuses are defined for the moving speed: high speed, medium speed and normal. High speed is for trains; medium speed is for handheld and normal is for fixed conditions. Service usage profile refers to the most used services or specific service subscription. There are also three statuses for service usage profile: voice centric, data centric and generic.

Subscription Policy based mobility control is indicated in TS 36300, i.e. the priority list is generated with the consideration of user's service subscription information in SPID's service usage profile. For example, if the user's specific service subscription is data centric, meaning that the user initiates or receives more PS data service each day E-UTRAN should be defined with higher priority; otherwise if the service subscription is voice centric, UTRAN or GERAN may be defined with higher priority. On the one hand, it is yet to be defined how to differentiate the voice centric and data centric terminals based on subscriptions. On the other hand, the user's service subscription information does not always reflect the real condition for user's most used services. For example, the service subscription information indicates that the user is data centric, while actually, the subscriber may use more CS voice call service than PS data service. The problem will deteriorate if the priority list is determined based on the service subscription information under that condition.

SPID may provide the most used services information in service usage profile. Unlike the service subscription information, SPID can reflect the user's real service use condition. But the problem still exists. For example one subscriber uses more CS voice service according to SPID information, and the UE will thus camp on UTRAN or GERAN based on the priority mechanism. However the real condition is that the subscriber may consume lots of CS voice service in one location (Location A), while in another location (Location B), the subscriber prefers to use data service. Although the CS voice service in Location A is more frequent than data service for the user, in Location B the user is data centric, the problem will occur if the UE still camps on UTRAN or GERAN. In reality, subscribers tend to use different kinds of services in different context. For example, some

subscribers tend to use data service (watching news) on their way to office; and the same set of subscribers may tend to use the voice centric service during working hours, and so on.

The location of a UE has been taken into account for cell reselection in US 2009/02471 17 Al and WO2004/006482 A2. According to the UE location information combined with the base station location information, it can be determined which cell provides signal with better quantity. A higher priority can then be defined for that cell. In addition, in CN101562851 A, the UE's history moving track can be recorded by a base station. Based on that information, a prediction for which cell the UE will move to can be obtained.

Existing mechanisms defined in 3GPP or other literature, such as cell load condition, UE radio access capability, SPID (Subscriber Profile ID for

RAT/Frequency Priority) and UE location are always seen as the drivers for inter- frequency or RAT cell reselection. However existing solutions cannot provide context awareness (location and time coherent) service usage prediction for the users of UEs. The service switching delay caused by camping on an inappropriate cell cannot be reduced without sufficient prediction of the probability of the pending service in advance.

Hence a more precise method is needed to predict the probability of coming service type in different locations for the concerned UE, then UE could reselect to the cell supporting that service in advance, CS fallback and cell redirection mechanism could be avoided to a great extent and user' s experience will be further improved.

Summary of the Invention

In order to better address one or more of the foregoing concerns a first aspect of the invention provides a method of assisting selection, for a user equipment in an idle mode, of a network cell from a plurality of network cells of a cellular wireless telecommunication network having a plurality of network cells the network cells operating according to different network technologies, the method comprising: recording behavioural information of the user equipment over a period of time; compiling behavioural statistics data based on the recorded behavioural information; generating network cell ranking data for the plurality of the network cells according to the behavioural statistics data and the network technology of each network cell; and transmitting the network cell ranking data to the user equipment.

Recording behavioural information of the user equipment may include recording at least one service used by the user equipment, the location where the service was used and the time interval when the service was used. The network cell ranking data may be further determined according to at least one of the group of a load condition of each network cell, subscription data relating to the user of the user equipment, the radio access capability of the user equipment. In embodiments of the invention the network technology of each cell is selected from the group of E- UTRAN, UTRAN and GERAN.

A second aspect of the invention provides a core network entity of a cellular network having a plurality of network cells the network cells operating according to different network technologies, the core network entity comprising: recording means for recording behavioural information of the user equipment over a period of time; statistics compilation means for compiling behavioural statistics data based on the recorded behavioural information; and transmission means for transmitting the behavioural statistics data to a base station of the network.

A third aspect of the present invention provides a base station of a network cell of a cellular network having a plurality of network cells the network cells operating according to different network technologies, the base station comprising: receiving means for receiving behavioural statistics data of a user equipment from a core network entity; ranking means for generating network cell ranking data for the plurality of the network cells according to the behavioural statistics data and the network technology of each network cell; and transmission means for transmitting the network cell ranking data to the user equipment.

A fourth aspect of the invention provides a user equipment on a first network cell of a cellular network having a plurality of network cells the network cells operating according to different network technologies, the user equipment comprising:

receiving means for receiving network cell ranking data ranking the plurality of network cells according to behavioural data of the user equipment and the network technology of the network cells; processing means for determining whether to change to another network cell or to remain on the first network cell, and in the case where it is determined that the user equipment should change network cell selecting a network cell to change to according to the network cell ranking data.

The processor of the user equipment is operable to select the highest ranked network cell fulfilling predetermined measurement criteria.

Aspects of the invention provide a method and apparatus for obtaining cell reselection specific priority list based on user's history statistical behavior over multiple RAT networks (E-UTRAN, UTRAN and GERAN). The context awareness service prediction for the concerned subscriber can be made based on subscribers' historical behavioural statistics in a time and location coherent way. In this way, the system is able to decide the right cell that the subscriber should camp on during idle time based on the subscriber' time and location coherent context. As a consequence the latency of the service access will be greatly reduced when the subscriber access the service and thus improve the user experience.

The methods according to the invention may be computer implemented. The methods may be implemented in software on a programmable apparatus. They may also be implemented solely in hardware or in software, or in a

combination thereof.

Since the present invention can be implemented in software, the present invention can be embodied as computer readable code for provision to a

programmable apparatus on any suitable carrier medium. A tangible carrier medium may comprise a storage medium such as a floppy disk, a CD-ROM, a hard disk drive, a magnetic tape device or a solid state memory device and the like. A transient carrier medium may include a signal such as an electrical signal, an electronic signal, an optical signal, an acoustic signal, a magnetic signal or an electromagnetic signal, e.g. a microwave or RF signal.

Brief Description of the Drawings Embodiments of the invention will now be described, by way of example only, and with reference to the following drawings in which :-

Figures 1 is a flow chart of a cell reselection method according to 3GPP TS 36.304

Figure 2 is a block diagram of a system for aiding selection of a network cell according to an embodiment of the invention

Figure 3 is a block diagram illustrating the flow of data between the network entities according to the embodiment of the invention; Figure 4 is a schematic diagram illustrating an example of location history of a user equipment; and

Figure 5 is schematic diagram comparing switching of network cells according to a method of the prior art and an embodiment of the present invention in a first example.

Figure 6 is schematic diagram comparing switching of network cells according to a method of the prior art and an embodiment of the present invention in a first example.

Figure 7 is a flow chart of a method of assisting selection of a network cell according to a first embodiment of the invention;

Detailed description

Embodiments of a method of and apparatus for assisting selection of a network cell according to the invention will be described with reference to Figures 2 to 7.

With reference to Figure 2 illustrating a general architecture of a system 100 for enabling a suitable cell to be selected for a user equipment to be camped on, the system comprises a core network 100 and a base station (BS) 200 of the core network 100.

The core network 100 comprises a record unit 101 for recording and a history of behavioural information, a Statistics unit 102, used for making statistics based on the user's history behavior information, a statistical information sending unit 103, for sending user's statistical history behavior information to the BS 200, a UE capability information sending unit 104, for sending UE radio access capability information to a BS 200, and a SPID information sending unit 105, used to send Subscriber Profile ID information to the BS 200.

The BS 200 comprises an information receiving unit 201 for receiving the user's statistical history behavior information, UE radio access capability information and neighboring cell load condition information as well as SPID information from other units, a specific priority list manager 202 for generating specific priority list for the UE based on the receiving information; a priority list sending unit 203 for sending specific priority list to the UE, and a load condition sending unit 204 for sending load condition information to other BSs.

The embodiment of the present invention aims to obtain a specific cell reselection priority list based on a UE's statistical behaviour history. Through that mechanism, a prediction of the coming service type probability for UE can be more accurately. A UE can then camp on the cell supporting that service in advance by cell reselection and the redirection or CS fallback procedure will be avoided thereby reducing the delay before service connection.

In order to obtain user's statistical behavior history, the user's behavior should be recorded firstly. That can be realized in CN of the system. The record unit in CN will record and store each service the subscriber used. In order to make the prediction more accurately, besides the record of service type (CS voice service or PS data service), the user' s behavior will be extracted in two dimensions: time and location. The location information could refer to Cell ID where the service happened. For example, one subscriber may be used to initiating or receiving more PS data services than CS voice calls during a period of time in some location each day, E-UTRAN network is preferred for that user.

However, in another place or during another period of time, more CS voice calls are initiated or received, it' s better to camp on UTRAN or GERAN.

Based on the obtained user's behavioral information, the statistic unit 102 in CN will make the user's history behavior statistics periodically. The statistical cycle could be decided by the operator, such as one week or one month. Table 1 shows an example for statistical result format:

Table 1 where P denotes the total cumulate times of initiated or received PS data service in a certain location during a certain time period each day in the last statistical cycle; C denotes the total cumulate times of initiated or received CS voice services in a certain location during a certain time period each day in last statistical cycle.

It should be noted that the deviant continuous service initiation caused by the network's failure is not considered here.

The location information can be recorded by serving cell ID.

The statistical time should cover the whole day, i.e. T1 +T2+ =24 hours.

The division of the time can be determined by operators according to the users' habit. For example, T1 =AM 8:00~PM 6:00, T2=PM 6:00~PM 1 1 :00, T3=PM 1 1 :00~AM 8:00.

The probability of voice service occurrence P_voice in Location 1 during T1 period each day in last statistical cycle is:

P_voice = (C_1 1 /P_1 1 +C_1 1 ) x 100%

The probability of data service occurrence P_data in Location 1 during T1 period each day in last statistical cycle is:

P_data = (P_1 1 /P_1 1 +C_1 1 ) x 100%

The same principle can be applied for other Locations and time periods.

P_voice can also be seen as the predicted probability that the coming service is a voice service

P_data can also be seen as the predicted probability that the coming service is data service

If P_voice> P_data, this indicates more CS service is initiated or received in that location during that period of time, and the prediction of the coming service is inclined to be a CS service. Consequently it would be better to choose UTRAN or GERAN as a cell for the UE to camp on. If P_data> P_voice, this indicates more PS services were initiated or received in that location during that period of time, and the prediction of the coming service is inclined to be data service. Consequently, E-UTRAN is preferred.

After the generation of user's statistical history behavioral information, the

CN 100 will send it to its BSs 200_1 and 200_2. The BS 200_1 and 200_2 will also receive the UE capability information and SPID information from the CN 100. At the same time, a BS 200_1 may obtain the neighboring cells' load condition from other BSs such as BS 200_2.

The user's statistical history behavior information, the UE capability information, the cell load condition information and SPID information may be seen as four factors used to make the specific priority list decision. The BS 200_1 will make use of the receiving information and obtain the specific priority list.

The following equation can be applied for each cell to calculate its priority rank F_total.

Fjotal =a%XF1 +b%XF2+c%XF3+d%XF4

where F1 ~F4 denotes the 4 priority factors on different aspects.

F1 : the value of priority factor on subscriber's statistical history behavior

F2: the value of priority factor on UE radio capability

F3: the value of priority factor on cell load condition

F4: the value of priority factor on SPID

a%~d%: the weighting factors for the different priority factors, that can be determined by the operator.

The higher the calculated F_total, the higher priority the cell has.

For the calculation of F1 (subscriber's statistical history behaviour factor), take the statistical information with T1 and Location 1 as example

If P_voice = P_data, this indicates that there is no difference between the PS service times and the CS service times. Thus an equal F1 value for different networks is feasible.

If P_data > P_voice, this indicates that more PS services were initiated or received in that location (Location 1 ) during that period of time (T1 ), hence E- UTRAN is preferred.

If P_voice > P_data, this indicates that more CS service is initiated or received in that place (Location 1 ) during that period of time (T1 ). Thus it would be better to choose UTRAN or GERAN.

Table 2 gives the mapping from user's statistical history behaviour to priority factor F1 .

Table 2

For the calculation of F2 (UE radio capability factor), if the cell with certain layer or RAT is supported by UE radio access capability: F2=1 ; if the cell with certain layer or RAT is not supported by UE radio access capability: F2=0 (cell will not be prioritized if one priority factor is 0).

For the calculation of F3 (cell load condition factor), although the load definition differs for GERAN, UTRAN and E-UTRAN, for each network, the cell load condition can be mapped to four statuses: Low Load, Medium Load, High Load and Overload.

The mapping between load status and priority factor F3 value is listed below in Table 3:

Load status Priority factor F3 on cell load

LowLoad 3

Medium Load 2

Table 3

For the calculation of F4 (SPID factor):

If the UE service is voice centric, UTRAN or GERAN should have higher priority factor F4, because they can provide CS voice service, while LTE does not support CS voice service.

If UE service is data centric, the priority factor F4 of LTE should be higher, due to LTE supporting PS data service with high data rate.

If UE service is generic, this indicates that there is no partial service between voice service and data service. Equal priority factor F4 could be configured.

Table 4 lists the mapping between user subscription policy and priority factor F4:

Table 4

Since all the networks above can support the UE with high moving speed, the UE's moving speed information in SPID is not considered here.

After the confirmation of each factor, the F_total parameter can be calculated for each cell and the corresponding specific priority list will be obtained.

Equal priority is not supported for different RATs in 3GPP specification. Hence if an equal P_total value is obtained for different RATs, the principle:

LTE>UTRAN>GERAN can be applied. In addition, a cell will not be prioritized if one priority factor is 0.

The method for specific priority ordering is not limited to the method with weighting factors as shown in the embodiments, variant algorithms taking UE behaviour information into consideration can be derived to reach the same or similar purpose. Although only LTE, GSM and UMTS are assumed in the embodiment and patent description, the scope of the patent is not limited to the three technologies, it may also include any existing cellular radio technologies, evolved cellular technologies or new technologies that apply the same or similar mechanism to determined the camp priority for the candidate cells of a UE.

Although LTE, UMTS and GSM in release 8 or further release in 3GPP are assumed as context, a similar rule may also be applied to the case where the multi-RAT network is composed of LTE and legacy UMTS and GSM networks. In that case, UE inherits the priority of the RAT can be considered.

Moreover, the scope of the invention may be applied to non-cellular radio technologies, such as WIFI and so on. In such cases, a similar mechanism is valid and should be adaptable due to potentially different network architectures.

Although only time and location information are described as input to derive the statistic behaviour of the UEs over a period of time, some other contexts, such as event contexts, holiday contexts, emergency contexts, weather contexts and etc.can also be further taken into account to reach the similar targets.

Although only UE behaviour determination methods are described in the description, the behaviour update or re-calibration mechanism can also follow similar procedure or a subset of the procedure to make the UE adaptive to the environment changes.

Although the provision of UE capability is described as being performed through the core network, other UE capability provision methods can apparently apply for this mechanism and should be within the scope of the claims.

Examples:

A possible deployment scenario for early commercial LTE networks is illustrated below as an example. In this example it is considered that a basic data services LTE network is rolled out over a former GSM and UMTS/HSPA network. In addition, due to the limitation of LTE coverage in early stage deployment, voice service should be provided with the legacy CS domain through 2G and 3G access networks, VoIP (Voice over IP) is assumed not to be available. Assuming that 5 cells are available in the same geographical area, cell 1 (LTE Layerl ), cell 2 (LTE Layer 2), cell 3 (UMTS Layerl ), cell 4 (UMTS layer 2), cell 5 (GSM).

Due to all the networks above can support the UE with high moving speed, hence UE's moving speed information in SPI D is not considered here.

After the confirmation of each factor, the F_total can be calculated for each cell and the specific priority list will be obtained.

It should be noted that equal priority is not supported for different RATs in 3GPP specification. Hence if equal P_total value is gotten for different RATs, the principle: LTE>UTRAN>GERAN can be complied. In addition, cell won't be prioritized if one priority factor is 0.

Examples:

A possible deployment scenario for early commercial LTE networks is illustrated below. It is considered that a basic data services LTE network is rollout over a legacy GSM and UMTS/HSPA network. In addition, due to the limitation of LTE coverage in early stage deployment, voice service should be provided with the legacy CS domain through 2G and 3G access networks, VoIP (Voice over IP) is assumed to be not available.

Assuming that 5 cells are available in the same geographical area, cell 1 (LTE Layerl ), cell 2 (LTE Layer 2), cell 3 (UMTS Layerl ), cell 4 (UMTS layer 2), cell 5 (GSM).

The weighting factors, a%~d% are all assumed to be 25% in this example. The information on the cells' load condition and whether the each cell is supported by UE radio access capability is listed in Table 5:

Cell UE radio capability/ Load condition /F3

F2

Cell 1 Support / 1 LowLoad / 3 (LTE Layerl )

Cell 2 Support / 1 HighLoad / 1 (LTE Layer 2)

Cell 3 Support / 1 MediumLoad 1 2

(UMTS Layerl )

Cell 4 Not support / 0 HighLoad / 1

Table 5

For SPID information, assuming the subscriber is CS voice centric, hence the priority factor can be obtained as shown in Table 6

Table 6:

According to personal habits, for users' history statistical behaviour, two periods are considered: T1 (8:00-18:00), T2 (18:00-8:00). Two locations are taken as example: Location 1 (working place), Location 2 (home). Two locations are both in Area A, which is covered by the 5 cells as illustrated in Figure 4.

The statistical period is assumed to be one month, i.e. the statistical information is based on the user's behaviour in the last month. The user's statistical behaviour information is listed below in Table 7:

Table 7 According to the above user's statistical behaviour, the priority factor F1 can be obtained as outlined in Table 8:

Cell Location & Time period Priority factor

F1 Cell 1 T1 /Location1 P voice > P data 1

(LTE Layerl ) T1 /Location2 P voice = P data 1

T2/Location1 P voice > P data 1

T2/Location2 P data > P voice 3

Cell 2 T1 /Location1 P voice > P data 1 (LTE Layer 2) T1 /Location2 P voice = P data 1

T2/Location1 P voice > P data 1

T2/Location2 P data > P voice 3

Cell 3 T1 /Location1 P voice > P data 3 (UMTS Layerl ) T1 /Location2 P voice = P data 1

T2/Location1 P voice > P data 3

T2/Location2 P data > P voice 2

Cell 4 T1 /Location1 P voice > P data 3 (UMTS layer 2) T1 /Location2 P voice = P data 1

T2/Location1 P voice > P data 3

T2/Location2 P data > P voice 2

Cell 5 T1 /Location1 P voice > P data 2 (GSM) T1 /Location2 P voice = P data 1

T2/Location1 P voice > P data 2

T2/Location2 P data > P voice 1

Table 8

Scenario A:

If a user is in Location 1 and during the period of T1

For cell 1 , P_total= a%XF1 +b%XF2+c%XF3+d%XF4=25%X (1 +1 +3+1 ) = 1 .5 For cell 2, P_total= a%XF1 +b%XF2+c%XF3+d%XF4=25%X (1 +1 +1 +1 ) = 1 For cell 3, P_total= a%XF1 +b%XF2+c%XF3+d%XF4=25%X (3+1 +2+3) = 2.25 For cell 4, due to one of the priority factor is 0, P_total needn't to be calculated. For cell 5, P_total= a%XF1 +b%XF2+c%XF3+d%XF4=25%X (2+1 +3+2) = 2

The specific priority list is:

Cell 3 (UTRAN)>Cell 5 (GERAN)>Cell 1 (E-UTRAN Layer 1 )>Cell 2 (E- UTRAN Layer 2)

Under the condition above, the UE is more inclined to receive or initiate CS voice call. With the consideration of user's statistical behaviour information, the UE will avoid camping on an E-UTRAN cell which does not support CS voice call.

As shown in Figure 5, when the UE camps on a E-UTRAN cell, in order to initiate or receive a CS voice call, a CS fall back scheme (such as redirection, cell change order or PS handover) will be performed to return to UTRAN or GERAN. However CS fall back procedures may cause an additional delay of at least 300ms for estimation according to 3GPP TS 36.133 and TS 25.133. Such a delay may put the success of the call setup at risk. With the consideration of the user's statistical behaviour information, it is possible to make the UE camp on UMTS/HSPA or GSM/GPRS/EDGE cells with good load condition for most of the time. Hence, the CS fallback probability will be minimized, and the user will have a better experience.

Scenario B:

If user is in Location 1 and during the period of T1

For cell 1 , F_total= a%XF1 +b%XF2+c%XF3+d%XF4=25%X(3+1 +3+1 )=2

For cell 2, F_total= a%XF1 +b%XF2+c%XF3+d%XF4=25%X(3+1 +1 +1 )=1 .25

For cell 3, F_total= a%XF1 +b%XF2+c%XF3+d%XF4=25%X(2+1 +2+3)=2

For cell 4, since one of the priority factor is 0, P_total does not need to be calculated.

For cell 5, F_total= a%XF1 +b%XF2+c%XF3+d%XF4=25%X(1 +1 +3+2)=1 .75

Since the LTE>UTRAN>GERAN principle is adopted if equal F_total value occurs, the specific priority list becomes:

Cell 1 (E-UTRAN Layer 1 )>Cell 3 (UTRAN)>Cell 5 (GERAN)>Cell 2 (E- UTRAN Layer 2)

Under the above conditions, a UE is more inclined to receive or initiate PS data service. With the consideration of user's statistical behavioural information, the UE will avoid camping on GERAN cell which does not support high rate PS service.

As shown in Figure 6, if the user's statistical history behavioural based priority mechanism is not used, by comparing the delay between procedures when a UE camps on an UTRAN/GERAN cell and is redirected to initiate a service versus when a UE is already camped on E-UTRAN, redirection leads to an additional delay of about 765-800 ms according to 3GPP R2-083917 for estimation.

By executing the presented method, the occurrence of redirection to E- UTRAN can be reduced. The mechanism can maintain user equipments on LTE layer for most of the time in order to provide enhanced data services to the user. Figure 7 outlines the operation of the system according to embodiments of the invention In step S1 of the method of assisting selection of a network cell the core network CN 100 records and stores the behavioural data of the user equipment 300. The behavioural data includes the service type employed by the user of the user equipment, such as a CS voice call, a low rate packet service, high rate packet service, etc as well as the time occurrence and the location (Cell ID)of the service usage. The CN 100 compiles statistical data on user's history behavior over a period. The CN may update the user's statistical history behavior information or send it to its serving BSs depending on the operator's policy. In step S2 according to the user's statistical history behavior information, cell load condition, user's subscription policy information and UE radio access capability information, the Priority List Manager 202 of the BS 200 generates a specific network cell ranking list for the UE 300. Priority List manager 202 can optionally be located in other node of network. In step S3 the UE 300 receives the specific network cell ranking or priority list through dedicated signaling and then performs a measurement for cell reselection in order to determine if the highest ranked network cell fulfils predetermined criteria. If in step s4 it is determined that the highest ranked network cell fulfils the criteria the user equipment 300 camps on that cell in step S6, otherwise the user equipment performs the measurement on the next ranked network cell to see if that cell fulfils the predetermined criteria. If a ranked network cell does not fill the predetermined criteria the user equipment continues down the ranking list performing the measurements until a network cell fulfils the predetermined criteria. The highest ranked network cell fulfilling the criteria is thus selected by the user equipment 300 for camping on.

Since the user's time and location coherent statistical behavior information history is taken into account to generate a specific cell reselection priority list in the BS, the probability of the incoming service type may be predicted with higher accuracy for the UE. UE can then reselect the most suitable network cell of the network to camp on in advance and avoid doing cell redirection or CS fallback to a great extent when initiating a service. Hence, the delay is decreased before a service is connected and the user's experience will be optimized. Embodiments of the invention hereinbefore described the invention provide a method and apparatus of obtaining cell reselection priority list based on a user's statistical behavioural history over Multiple RAT networks. According to the user's statistical history behavioural information, the probability of a coming service type for the UE can be predicted in advance. Hence a higher priority can be defined for the cell providing the predicted service type with higher probability. Consequently the UE can camp on the cell supporting that service prior to initiation of the service. This mechanism can help to further reduce the probability of the UE performing redirection or CS fallback for accessing the service and as a consequence user's experience will be improved.

Many further modifications and variations will suggest themselves to those versed in the art upon making reference to the foregoing illustrative embodiments, which are given by way of example only and which are not intended to limit the scope of the invention, that being determined solely by the appended claims. In particular the different features from different embodiments may be interchanged, where appropriate.

In the claims, the word "comprising" does not exclude other elements or steps, and the indefinite article "a" or "an" does not exclude a plurality. The mere fact that different features are recited in mutually different dependent claims does not indicate that a combination of these features cannot be advantageously used. Any reference signs in the claims should not be construed as limiting the scope of the invention.

Claims

1 . A method of assisting selection, for a user equipment in an idle mode, of a network cell from a plurality of network cells of a cellular wireless

telecommunication network having a plurality of network cells, the network cells operating according to different network technologies, the method comprising:

recording behavioural information of the user equipment over a period of time;

compiling behavioural statistics data based on the recorded behavioural information

generating network cell ranking data for the plurality of the network cells according to the behavioural statistics data and the network technology of each network cell; and

transmitting the network cell ranking data to the user equipment.

2. A method according to claim 1 , wherein recording behavioural information of the user equipment comprises recording at least one service used by the user equipment, the location where the service was used and the time interval when the service was used.

A method according to claim 1 or 2, wherein the network cell ranking data is further determined according to at least one of the group of a load condition of each network cell, subscription data relating to the user of the user equipment, the radio access capability of the user equipment.

A method according to any one of the preceding claims, wherein the network technology of each cell is selected from the group of E-UTRAN, UTRAN and GERAN.

5. A core network entity of a cellular network having a plurality of network cells the network cells operating according to different network technologies, the core network entity comprising: recording means for recording behavioural information of the user equipment over a period of time;

statistics compilation means for compiling behavioural statistics data based on the recorded behavioural information; and

transmission means for transmitting the behavioural statistics data to a base station of the network.

6. A core network entity according to claim 5 wherein the recorded

behavioural information of the user equipment comprises at least one service used by the user equipment, the location where the service was used and the time interval when the service was used.

7. A base station of a network cell of a cellular network having a plurality of network cells the network cells operating according to different network technologies, the base station comprising:

receiving means for receiving behavioural statistics data of a user equipment from a core network entity;

ranking means for generating network cell ranking data for the plurality of the network cells according to the behavioural statistics data and the network technology of each network cell; and transmission means for transmitting the network cell ranking data to the user equipment.

8. A base station according to claim 7 wherein the recorded behavioural

information of the user equipment comprises at least one service used by the user equipment, the location where the service was used and the time interval when the service was used.

9. A base station according to claim 7 or 8, wherein the network cell ranking data is further determined according to at least one of the group of a load condition of each network cell, subscription data relating to the user of the user equipment, the radio access capability of the user equipment.

10. A base station according to any one of claims 7 to 9, wherein the network technology of each cell is selected from the group of E-UTRAN, UTRAN and GERAN.

1 1 . A user equipment on a first network cell of a cellular network having a plurality of network cells the network cells operating according to different network technologies, the user equipment comprising:

receiving means for receiving network cell ranking data ranking the plurality of network cells according to behavioural data of the user equipment and the network technology of the network cells;

processing means for selecting the highest ranked network cell according to the network cell ranking data and for determining if the selected network cell fulfils predetermined operational criteria, and in the case where it is determined that the selected network cell does not fulfil the predetermined operational criteria, selecting the next highest ranked cell fulfilling the predetermined operational criteria.

12. A user equipment according to claim 1 1 wherein the recorded behavioural information of the user equipment comprises at least one service used by the user equipment, the location where the service was used and the time interval when the service was used.

13. A user equipment according to claim 1 1 or 12, wherein the network cell ranking data is further determined according to at least one of the group of a load condition of each network cell, subscription data relating to the user of the user equipment, the radio access capability of the user equipment.

14. A user equipment according to any one of claims 1 1 to 13, wherein the processor of the user equipment is operable to select the highest ranked network cell fulfilling predetermined measurement criteria.

15. A computer-readable medium having computer-executable instructions to enable a computer system to perform the method of any one of claims 1 to 4.