FIELD OF THE INVENTION

The present invention relates to a method, system, apparatus, and computer program product for balancing network load by controlling handover operations in mobile networks, such as - but not limited to - 3GPP (3 ^rd Generation Partnership Project) Long-Term Evolution (LTE & LTE-Advanced). The use of the described load balancing approach in existing technologies other than LTE is therefore not precluded. BACKGROUND OF THE INVENTION

Long Term Evolution (LTE) and Worldwide Interoperability for Microwave Access (WiMAX) systems provide simplified radio access network (RAN) structures in which the functions of centralized controllers, e.g. radio network controllers (RNCs) in Universal Mobile Communications System Terrestrial RAN (UTRAN), are partially shifted to access devices (e.g. base station type devices) in order to reduce latencies and improve the quality of service (QoS) provided to end users. The results are enhanced Nodes B (eNB) in the LTE architecture and enhanced base stations (BS) in WiMAX. Both architectures allow base station to base station communication for coordination purposes. The LTE provides this via specific interfaces (X2) between eNBs, while current WiMAX standards rely on the availability of a mediation function provided by another access serving network (ASN) gateway element (ASN-GW), since a BS to BS interface (R8 reference point) is not clearly defined yet.

Load balancing (LB) in LTE is achieved by means of handover (HO) of users from overloaded cells to cells which can accommodate additional load. The receiving cells might be adjacent in space (neighbor cells), and/or on a different frequency channel (inter-frequency LB), and/or even of a different radio access technology (RAT, inter-RAT LB). As this HO needs to be based on detailed and up-to-date knowledge of the radio environment, it is performed by the eNB in LTE. Further- more the eNB can acquire knowledge of the load situation of adjacent cells via X2 load reports.

However, the overloaded eNB might not have an understanding of the overall load situation in the larger environment, i.e. in neighbors-neighbors, or in different RATs. Such information is typically available in the operation and maintenance (OAM) system.

Yet, due to overhead and delay restriction, it is neither possible to convey the complete radio information of the eNBs to a centralized network element (e.g. self- organizing network (SON) entity), nor is it feasible to convey the OAM information on the overall load situation to the eNBs. An eNB can thus only consider the load in its direct neighbors and might therefore not distribute the load in an optimal way considering the overall load situation in the network. It is therefore desirable to provide information about the larger environment for e-NB-based LB in a way that maximizes the benefit with minimum signalling effort.

SUMMARY According to various embodiments, two interrelated methods are provided. First, a method is provided, which comprises:

• determining load balancing handover priorities for a plurality of adjacent cells surrounding a target cell of a cellular access network, said adjacent cells comprising at least one neighbor cell and at least one neighbor's neighbor cell of said target cell;

• generating based on said determined handover priorities at least one load balancing priority list with an ordered priority for load balancing purposes; and

• forwarding an information for deriving said at least one load balancing priority list from a centralized entity of said cellular access network to an access device of said target cell.

Second, a method is provided, which comprises: • receiving at an access device (e.g. base station, Node B, eNB, access point etc.) of a target cell of a cellular access network an information for deriving at least one load balancing priority list with an ordered priority for load balancing purposes, said at least one load balancing priority list indicating load balancing handover priorities for a plurality of adjacent cells surrounding said target cell of said cellular access network, said adjacent cells comprising at least one neighbor cell and at least one neighbor's neighbor cell of said target cell; and • selecting a handover target based on said at least one load balancing priority list.

Additionally, according to various embodiments, two interrelated apparatuses are provided.

First, an apparatus is provided, which comprises:

• load balancing control means (e.g. load balancing controller or load balanc- ing processor) for determining load balancing handover priorities for a plurality of adjacent cells surrounding a target cell of a cellular access network, said adjacent cells comprising at least one neighbor cell and at least one neighbor's neighbor cell of said target cell; • prioritizing means (e.g. priority controller or prioritizer or priority processor) for generating based on said determined handover priorities at least one load balancing priority list with an ordered priority for load balancing purposes; and • signalling means (e.g. signalling controller or signalling processor) for forwarding an information for deriving said at least one load balancing priority list to an access device of said target cell.

Second, an apparatus is provided, which comprises:

• receiving means (e.g. receiver or transceiver or signalling unit) for receiving an information for deriving at least one load balancing priority list with an ordered priority for load balancing purposes, said at least one load balancing priority list indicating load balancing handover priorities for a plurality of adjacent cells surrounding a target cell controlled by said apparatus, said adjacent cells comprising at least one neighbor cell and at least one neighbor's neighbor cell of said target cell; and

• selecting means (e.g. selector or handover controller or handover processor) for selecting a handover target based on said at least one load balancing priority list.

Further, a system is provided, which comprises at least of both apparatuses defined above.

Moreover, a computer program product comprising code means for producing the steps of the above interrelated methods when run on a computer device.

Accordingly, the proposed centralized LB support allows optimization of LB decisions based on overall status of the network (i.e. beyond the information available via X2 load reporting) and minimizes the involved signalling overhead (both on ltf interfaces towards the OAM system and X2 interfaces between eNBs). In addition, the proposed centralized LB support may lead to an increased acceptance of LB- based HO, since cells with less competition of other cells also trying to handover to the target cell are prioritized. As an additional advantage, the proposed centralized LB support allows to configure for each cell a list of neighbors where LB can be performed. I.e., LB can be allowed selectively on a cell-to-cell level.

According to a more specific implementation example, the at least one LB priority list may comprise at least one of a first list for intra-frequency handover priorities, a second list for inter-frequency handover priorities and a third list for inter radio access technology handover priorities. Thereby, the proposed centralization can be applied to all kinds of handovers provided by the network system. According to another more specific implementation example, the proposed at least one LB priority list may be updated and the corresponding information may be forwarded by the centralized entity at regular intervals and/or in case of detected load changes. This ensures that available priority list are always up-to-date so that handover decisions are reliable.

The priority determination may be based on an overall load distribution available at said centralized entity. The LB priority list thus reflects the overall load situation and ensures optimized handovers for LB.

According to a further more specific implementation example, the selection of handover targets at the access device may be restricted to handover targets indi- cated in the at least one LB priority list. Thereby, LB to a certain cell can be suppressed completely.

According to a still further specific implementation example, the selection of the handover target may be based on a merging of the at least one LB priority list and a load information received from neighboring cells. Hence, eNB decisions based on neighbor load reports can be enhanced by considering the received LB handover priorities.

Further modifications or developments are defined in the dependent claims.

BRIEF DESCRIPTION OF THE DRAWINGS

Examples of the present invention will now be described on the basis of embodi- ments with reference to the accompanying drawings in which:

Fig. 1 shows a schematic block diagram of a cellular system in which the present invention can be implemented; Fig. 2 shows an example of a load distribution in a cellular system;

Fig. 3 shows a schematic block diagram of an eNB according to a first embodiment; Fig. 4 shows a signaling diagram of a handover control method for load balancing at the eNB according to the first embodiment; Fig. 5 shows a schematic block diagram of an SON entity according to a second embodiment;

Fig. 6 shows a signaling diagram of a handover control method for load balancing support at the SON entity according to the second embodiment;

Fig. 7 shows a schematic block diagram of a software-based implementation according to a third embodiment.

DESCRIPTION OF THE EMBODIMENT

In the following, exemplary embodiments will be described based on an LTE system environment. However, it will be apparent from the following description and is therefore explicitly stressed that the present invention can be applied to any other mobile or cellular network system which involves network elements with a HO controlled load balancing functionality.

Fig. 1 shows schematic block diagram of a cellular LTE system in which the present invention can be implemented.

A terminal device 10 (e.g. user equipment (UE) or mobile station (MS)) is currently served by one of three adjacent or neighboring eNBs 22, 24, 26 which are configured to control radio access and load balancing in their respective cells (not shown in Fig. 1). In the scenario depicted in Fig. 1, the MS 10 is in the coverage area of a first eNB 22 and may select the cell of one of a second and third eNB 24, 26 for load balancing purposes. The neighboring second eNB 24 and third eNB 26 control adjacent cells which might be loaded. X2 interfaces provided between the eNBs 22, 24, 26 can be used for exchanging information.

According to some embodiments, HO control is performed at the network side, e.g. by the serving eNB 22 or a separate HO control unit (not shown) in a manner so that the terminal device 10 is directed to a certain cell and the load between cells can be balanced in a fashion favourable for the network capacity.

Self-optimization of intra-LTE and inter-RAT (radio access technology) mobility parameters to the current load in the cell and in the adjacent cells can improve the system capacity compared to static/non-optimized cell reselectioπ/HO parameters and can minimize human intervention in the network management and optimization tasks. The load balancing shall not affect the user quality of service (QoS) negatively in addition to what a user would experience at normal mobility without load-balancing. Conventionally, mobility load balancing optimization can be achieved by measuring load for each cell in its monioring eNB. Load information is then exchanged between eNBs 22, 24, 26 over the X2 interfaces. An algorithm is applied in a HO controller or processor of the eNBs 22, 24, 26 to identify the need to distribute the load between two adjacent cells.

In the present specification, "load" is intended to cover radio load, transport network load or even processing load. Even for radio load, it could be split between uplink load and downlink load or split among different QoS class identifiers (QCIs). The definition of load influences the algorithm to distribute the load. An algorithm needs to be defined as to when to balance the load. Due to the different possibilities of load definition, the algorithm can be based on radio load or transport network load or both of them. If the radio load is concerned as the most important factor, it could also be decided whether to differentiate among QCIs. In Fig. 1 , an OAM system 40 with an SON entity 42 and an OAM entity 44 is shown. The SON entity may be responsible for at least one of self-configuration, self-optimization and self-healing of the cellular access network. It is connected via a backbone network 30. Self-configuration may be defined as a process where newly deployed nodes (eNBs) are configured by automatic installation procedures to get the necessary basic configuration for system operation. A new eNB can be configured based on transport and radio configuration data. The new eNB connects to the OAM subsystem 40 for specific management functions. Finally, the X2 interfaces and other necessary interfaces are established. Optimization algorithms can be executed in the OAM System 40. The SON functionality or entity 42 may reside in a small number of locations, at a high level in the architecture. Furthermore, all SON functions are located in the OAM system 40, so it is easy to deploy them. To implement the centralized SON entity 42, an existing Itf-N interface needs to be extended. The 3GPP Services and System As- pects Work Group 5 (3GPP SA5) calls the interface between the network management system (i.e. managing unit) and network element management unit or network element (i.e. managed unit) as northbound Interface (Itf-N) in 3G network management. The Itf-N interface includes a series of Integration Reference Points (IRPs).

However, the SON entity 42 may also be based on a hybrid SON approach where a part of the optimization algorithms are executed in the OAM system 40 while others are executed in the eNBs 22, 24, 26.

According to some embodiments, the centralized SON entity 42 in the OAM system 40 signals respective lists of cells to the eNBs 22, 24, 26 with an ordered pri- ority for load balancing purposes. These lists are updated periodically or in case of changes for all cells that have the load balancing functionality enabled.

In specific optional implementation, the SON entity 42 may provide separate lists for intra-frequency, inter-frequency, and inter-RAT cells, as well as an information indicating the priority between intra-frequency, inter-frequency, and inter-RAT LB.

According to another optional implementation, the eNBs 22, 24, 26 may be configured so that neighbor cells not appearing on the list are not be used or selected for LB. This allows also to prevent LB to a certain cell totally. In this case the eNBs 22, 24, 26 can also suspend X2 load reporting with this cell and unnecessary X2 communication overhead can be avoided.

With the new centralized LB support, the eNBs 22, 24, 26 are now enabled to consider target cells for load balancing HO by taking the received priority list into ac- count, e.g., if multiple possibilities (with respect to user selection and target cell selection for load balancing) are of similar performance from the radio perspective.

As another option, the priority list(s) may be updated by the centralized SON entity 42 and sent to the eNBs 22, 24, 26 prior to an occurrence of overload and a result- ing need for LB. This ensures that the priority lists are always available and additional delay is avoided, e.g., due to a decrease in reaction time. Feasibility of LB is thus not limited.

Furthermore, as the priority list is conveyed to the eNBs 22, 24, 26 and not on\y load reports of neighbors' neighbors, other aspects and decision criteria than load can be included in the prioritization, specific cells can be excluded from LB, and a prioritization can be defined e.g. for inter-RAT LB. To illustrate an implementation example for inter-frequency LB consider Fig. 2 which shows an example of a load distribution for a set of cells 1 to 7 in a cellular network, where shaded cells are overloaded.

Based on the existing X2 signalling cell 1 can obtain information about its neighbor cells 2 to 7. The relative load is indicated as percentage in the respective cell of Fig. 2. Assume for simplicity that cell 1 has users suitable for LB in the vicinity of the cell borders of all its neighbor cells. Based on the X2 load information it would seem that cells 3 and 7 are best suited for LB, as they currently have lowest load (60%) of all neighboring cells. However, these cells have a large number of other adjacent cells currently in overload. Therefore the risk that cells 3 and 7 might soon get overloaded is to be judged high. It is therefore obvious that the LB decisions at the eNB of cell 1 can be improved by providing guidelines on LB priorities based on the overall load distribution available in the central SON entity. In a simple implementation the priority list could be based on the number of neighbors in overload, i.e. the central SON entity 42 would signal the following priority list:

4

5, 6

3 As can be seen in this example, multiple cell identities (IDs) can be given the same priority if there is no difference from the viewpoint of the central SON entity 42.

The concerned eNB which controls cell 1 can now merge both information and might e.g. use the following LB strategy:

• first try load balancing with cell 4

• in case further load balancing is required, perform load balancing with cell 6 (Cell 6 is preferred over cell 5, since it has lower radio load)

• next try cells 5, then 3 In this example, LB with cell 3 is significantly de-prioritized due to the additional information. LB with cell 7 is even not allowed due to reasons only known at the central SON entity 42 (e.g. many neighbor cells of cell 7 overloaded or a planned cell shut-down for maintenance or energy saving reasons). Cell 1 can also sus- pend the X2 load reporting procedures with cells 2 and 7.

It is evident for those skilled in the art, that this example can be applied also for inter-frequency and inter-RAT LB and that more refined algorithms can be used to merge the available information at the eNB of the target cell.

Fig. 3 shows a schematic block diagram of a HO control functionality or apparatus according to a first embodiment. The HO control functionality or apparatus may be implemented as an integrated circuit, a chip set, a module, a software-controlled processing or computing device, or a hardware circuit, which can be provided at the eNB 22 and/or the other eNBs 24, 26 of Fig. 1 or any other suitable network device responsible for load balancing and/or HO control.

According to Fig. 3, the HO control functionality or apparatus comprises a HO control function or control unit or control processor or controller 221 as HO selecting means configured to decide about HO of a served mobile terminal, e.g. terminal device 10 of Fig. 1 , in dependence on specific parameters and/or information 224 received via a receiver or transceiver as receiving means or derived from the network (X2 load report, priority list) and/or the served terminal device 10. Additionally, the HO controller 221 may issue queries or requests 225 towards the net- work-side to obtain load and/or priority information. The HO controller 221 is configured to decide on HOs for load balancing purposes based on the received information 224 (e.g. as explained above in connection with Fig. 2) Furthermore, the HO controller 221 controls a signaling unit 223 to generate a HO instruction or command 226 in order to control available HO targets. Additionally, a memory unit 222 (e.g. look-up table (LUT)) may be provided for storing received priority list(s) advertised by the SON entity 42 for the controlled cell.

At this point, it is noted that the functionalities of blocks 221 to 223 of Fig. 3 can be implemented as discrete hardware or signal processing units, or alternatively as software routines or programs controlling a processor or computer device to perform the processing steps of the above functionalities. Fig. 4 shows a flow diagram of a HO control procedure according the first embodiment, which may be executed by the HO controller 221 of Fig. 3. Based on this procedure it can be decided that a new eNB becomes the serving eNB. Initially, it is assumed that the terminal device 10 is served by the eNB 22 of Fig. 1. In step S101 the serving eNB 22 receives and evaluates X2 load report(s) from eNB(s) of neighboring cell(s) and the latest priority list received from the SON entity 42 and stored in the memory unit 222. The evaluation may be based on any load balancing algorithm and under consideration of the received HO priorities of the priority list(s). Based on the load and priority situation of adjacent cells, the HO controller 221 develops a HO strategy (as explained in connection with Fig. 2) in step S102 and decides on a HO target cell in step S103. The HO decision can be made according to some threshold criterion based on available loads and priorities. The decision criterion to hold the terminal device 10 in the own cell can have a hysteresis. That is, an admission threshold (which determines when the terminal device 10 is admitted to be served by a cell) is lower than a releasing threshold (which determines when the terminal device 10 is to be released by a cell). Thus, frequent HO may be avoided in cases of small load fluctuations. If HO to a neighboring cell is refused in step S103, the procedure jumps back to step S101 and waits for new load reports or priority list(s), based on which steps S101 to S103 are repeated. If however HO to another (less loaded) cell is decided in step S103, the procedure continues with step S104 and a HO command is sent to the terminal device 10.

Fig. 5 shows a schematic block diagram of a HO control functionality or apparatus according to a second embodiment related to the SON-side of the system. The HO control functionality or apparatus may be implemented as an integrated circuit, a chip set, a module, a software-controlled processing or computing device, or a hardware circuit, which can be provided at the SON entity 42 of Fig. 1 or any other suitable network device responsible for centralized load balancing in a cellular network.

According to Fig. 5, the HO control functionality or apparatus comprises a HO con- trol function or control unit or control processor or controller 421 as load balancing control means configured to allocated HO priorities to individual cells of the cellular network, in dependence on specific parameters and/or information 424 received or derived from the network and concerning overall load distribution and its implication on individual cells. Additionally, the HO controller 421 may issue queries or requests 425 towards the network to obtain load information based on which an overall load distribution can be derived or estimated. The HO controller 421 is con- figured to allocate priorities to HOs for load balancing purposes based on the received information 424. Furthermore, the HO controller 421 controls a reporting unit 423 as signalling means to advertise an information 426 indicating dedicated load-balancing priority lists to individual cells or from which dedicated priority lists can be derived at the individual cells. These priority lists may be generated at the HO controller 421 or in an optional separate priority list generator function or unit 422, both acting as prioritizing means.

At this point, it is noted that the functionalities of blocks 421 to 423 of Fig. 5 can be implemented as discrete hardware or signal processing units, or alternatively as software routines or programs controlling a processor or computer device to perform the processing steps of the above functionalities.

Fig. 6 shows a flow diagram of a load balancing support procedure according the second embodiment, which may be executed by the HO controller 421 of Fig. 5. Based on this procedure priorities for load balancing HOs can be advertised to individual cells.

According to Fig. 6, when the HO controller 421 detects in step S201 changes of cells with enabled load balancing function, e.g., based on the information 424 re- ceived from the network or terminals, it generates or updates priority list(s) of relevant cells in step S202. This can be achieved by changing the priority order of the cells or even by deleting certain cells from the priority list to suppress HOs to such deleted cells. Finally, in step S203, updated priority lists are signaled or advertised or reported to the relevant cells.

Fig. 7 shows a schematic block diagram of a software-based implementation of the proposed HO control mechanism for LB support according to the above embodiments. Here, the HO control apparatus 300 comprises a processing unit 310, which may be any processor or computer device with a control unit which performs control based on software routines of a control program stored in a memory 312. Program code instructions are fetched from the memory 312 and are loaded to the control unit of the processing unit 310 in order to perform the processing steps of the above functionalities described in connection with the respective Figs. 4 and 6. At the HO control apparatus on the eNB side, these processing steps may be performed on the basis of input data Dl and may generate output data DO, wherein the input data Dl may correspond to the load reports received from neighboring cells and the priority lists received from the SON entity 42, and the output data DO may correspond to the HO command 226.

At the HO control apparatus on the SON side, these processing steps may be performed on the basis of input data Dl and may generate output data DO, wherein the input data Dl may correspond to the information 424 received from the network or terminals, and the output data DO may correspond to the information 426 indicating the priority list.

The above embodiments thus provide better cell LB by centralized network sup- port.

To summarize, methods, computer program products, apparatuses, and a system for balancing load in a cellular network have been described. Load balancing handover priorities are determined for a plurality of adjacent cells surrounding a target cell of a cellular access network, the adjacent cells comprising at least one neighbor cell and at least one neighbor's neighbor cell of said target cell. Then, based on the determined handover priorities at least one load balancing priority list with an ordered priority for load balancing purposes is generated and an information for deriving the at least one load balancing priority list is forwarded from a cen- tralized entity of said cellular access network to an access device of the target cell, where the list is used for selecting a handover target.

It is to be noted that the present invention is not restricted to the LTE-based embodiments described above, but can be implemented in any mobile network where HOs of mobile terminals can be at least party controlled by the network. The proposed centralized Lb support can be applied in radio systems like e.g. WiMAX as currently standardized in 3GPP for WCDMA (Wideband Code Division Multiple Access), as well as 3GPP E-UTRAN (Enhanced Universal Mobile Telecommunications System (UMTS) Terrestrial Radio Access Network), such as LTE (Long Term Evolution) or 3.9G. These radio access technologies (e.g. WLAN, WiMAX, E-UTRAN or 3G LTE) may involve multiple-input multiple-output (MIMO) systems or multi-beam/multi-antenna transmitter or receiver devices (e.g. base station de- vices, access points or other access devices) capable of receiving signals via different receiving paths and/or channels. The parameters and procedures can be adapted to the respective technology, interfaces and architecture. More specifically, the invention is applicable for both frequency division duplex (FDD) and time division duplex (TDD) technologies and can be applied as mentioned above to all kind of mobile networks other than LTE. Moreover, the invention is not limited to SON-based LB support. The priority information can be generated by any network element at which an overall load distribution can be derived. Also, the different embodiments described can be combined.

As already mentioned, the embodiments can be realized in hardware, software, or a combination of hardware and software. A typical combination of hardware and software can be a processing system with an application that, when being loaded and executed, controls the processing system such that it carries out the methods described herein. The embodiments also can be embedded in an application product, which comprises all the features enabling the implementation of the methods described herein, and which when loaded in a processing system is able to carry out these methods. The terms "computer program," "software," "application," variants and/or combinations thereof, in the present context, mean any expression, in any language, code or notation, of a set of instructions intended to cause a system having an information processing capability to perform a particular function either directly or after either or both of the following: a) conversion to another language, code or notation; b) reproduction in a different material form. For example, an application can include, but is not limited to, a subroutine, a function, a procedure, an object method, an object implementation, an executable application, an applet, a servlet, a source code, an object code, a shared library/dynamic load library and/or other sequence of instructions designed for execution on a processing system.

The terms "a" and "an," as used herein, are defined as one or more than one. The term "plurality," as used herein, is defined as two or more than two. The term "another," as used herein, is defined as at least a second or more. The terms "including" and/or "having," as used herein, are defined as comprising (e.g., open lan- guage). Accordingly, the above predetermined embodiments may vary within the scope of the attached claims.