Patent Application
20130316709
The present invention relates to autonomous optimization of a neighbor list wirelessly transmitted from a base station.
A neighbor list (also referred to as a neighbor cell list) is used to control handover of a mobile terminal which is performing communication. A neighbor list is able to include handover parameters such as a cell individual offset (CIO). A base station sends a neighbor list to a mobile station connected to a cell that the base station manages by wirelessly transmitting the neighbor list. Hereinafter, cells registered in a neighbor list among neighbor cells are referred to as “Listed Cells”, and cells that are not registered in a neighbor list are referred to as “Detected Cells”.
In order to achieve handover of a mobile terminal from a connected cell (referred to as a serving cell) to a neighbor cell, a handover control entity of the serving cell such as a radio resource controller (RNC) or a serving base station instructs the mobile terminal to transmit a measurement report when a predetermined event occurs. The predetermined event is, for example, degradation of radio quality of the serving cell. The measurement report generated by the mobile terminal includes results of measuring radio quality of the serving cell and its neighbor cell (neighbor cells). Upon receiving the measurement report from the mobile terminal, the handover control entity of the serving cell determines a handover destination cell (target cell) based on the measurement report and then initiates a handover procedure including signaling with the mobile terminal and the objective cell.
One of transmission events of the measurement report defined by 3GPP TS 36.331 V8.11.0 (2010-09), which is a 3GPP technical specification regarding Long Term Evolution (LTE)/Evolved UTRAN (E-UTRAN), will be introduced. The essential part of the transmission event defined as Event A3 (Neighbor becomes offset better than serving) in this technical specification is expressed by the following expression (1).
P
S
+O
S
<P
T
+O
T (1)
In the expression (1), PS denotes a result of measuring radio quality of a serving cell and PT denotes a result of measuring radio quality of a neighbor cell. In the case of LTE, PS and PT are downlink reference signal received power (RSRP) or reference signal received quality (RSRQ). RSRQ is a ratio of RSRP to total received power. Further, Os in the expression (1) denotes an offset value to the radio quality of the serving cell, and affects the radio quality of the downlink reference signal of the serving cell. Meanwhile, OT denotes an offset value to the radio quality of the neighbor cell, and affects the radio quality of the downlink reference signal of the neighbor cell. The neighbor list can hold a setting of offset OT that differs for each registered cell. The offset value OT is referred to as a cell individual offset (CIO).
When the operating condition of the expression (1) is set in the base station, the operating condition of the expression (1) is notified to the mobile terminal connected to the cell managed by the base station. When the results of measuring the radio quality of the serving cell and the radio quality of the neighbor cell satisfy the condition shown in the expression (1), the mobile terminal transmits a measurement report to the serving base station.
Further, the expression (1) may be rewritten into the following expressions (2) and (3). The parameter THHO is called a handover threshold.
P
T
−P
S
>TH
HO (2)
TH
HO
=O
S
−O
T (3)
A technique for dynamically adjusting handover parameters such as the offset value Ot (CIO) stated above and Time to Trigger (TTT) is known. TTT is lag time from establishment of the condition for transmitting the measurement report such as the expressions (1) or (2) stated above to transmission of the measurement report. For example, when a period during which the condition of the expression (2) continues to be satisfied becomes longer than TTT, a mobile terminal sends a measurement report. Thus, as TTT becomes longer, the timing to start handover is delayed, and handover becomes difficult to occur.
For example, Patent literature 1 discloses reducing handover failures (i.e., Too Late Handover, Too Early Handover) by dynamically adjusting handover parameters (e.g., TTT, CIO and the like) based on measurement reports from mobile terminals. Further, Non-patent literature 1 discloses that a base station adjusts CIO for a neighbor cell included in a neighbor list when communication load of a cell increases, thereby prompts a mobile terminal to perform handover from the cell with large load to a neighbor cell. According to the technique disclosed by Non-patent literature 1, it is possible to perform load balancing between cells.
A handover control entity that has received a measurement report from a mobile terminal typically determines a target cell from “Listed Cells” among neighbor cells included in the measurement report. In this case, a mobile terminal cannot be handed over to “Detected Cells”. Accordingly, if some neighbor cells have been omitted from the neighbor list, this may cause occurrence of handover failure or degradation of communication quality due to handover to an inappropriate cell.
Further, as stated above, the offset value Ot for each neighbor cell is notified to the mobile terminal using a neighbor list. In short, only neighbor cells registered in a neighbor list are able to notify the mobile terminal of the offset value Ot. Accordingly, if an important neighbor cell is not registered in a neighbor list, it is impossible to notify the mobile terminal of the offset value Ot regarding this important neighbor cell. In this case, since it is impossible to adjust a timing to perform handover of a mobile terminal by adjusting the offset value Ot, it is difficult to improve the handover success rate based on adjustment of the offset value Ot (e.g., Patent literature 1) or to achieve load balancing by performing handover of mobile terminals which is near a cell boundary to a neighbor cell (e.g., Non-patent literature 1). This problem also occurs when the handover control entity of the serving cell can select a target cell from Detected Cells.
A offset value for each neighbor cell as stated above may also be set, separately from the CIO, for radio quality measured by a mobile terminal at the time of initiating communication. A cell to which a mobile terminal connects at the time of initiating communication is controlled by adjusting the offset value (qoffset), and it is thereby possible to control communication load of cells. However, as is similar to the CIO, a cell which can notify mobile terminals of the offset value (qoffset) is limited to neighbor cells registered in a neighbor list. Thus, it becomes difficult to achieve load balancing if an important neighbor cell is not registered in a neighbor list.
As will be clear from the description above, neighbor list optimization is important to secure high communication quality. In particular, since the number of cells that can be registered in a neighbor list is limited, it is necessary to preferentially register, in a neighbor list, important cells that are highly likely to contribute to an improvement of communication quality.
Conventionally, neighbor list optimization has been conducted by a professional engineer based on results of a running test or the like in a service area of a cellular communication system. However, in recent years, a technique of autonomously optimizing a neighbor list during an operation of a cellular communication system has been proposed (see Non-patent literature 2).
Non-patent literature 2 discloses preferentially registering, in a neighbor list, a neighbor cell which is frequently reported by measurement reports from mobile terminals, and a neighbor cell whose number of handover attempts of mobile terminals is large.
The present inventors have found out that, in a heterogeneous network (HetNet) environment and in a situation where there are a large number of neighbor cells of more than an upper limit number that can be registered in a neighbor list, the technique disclosed in Non-patent literature 2 may not be able to sufficiently perform neighbor list optimization. The HetNet environment is an environment in which a plurality of different types of cells having different sizes in coverage (cell sizes) (e.g., a macro cell and a pico cell, a macro cell and a femto cell, or a pico cell and a femto cell) are arranged adjacent to each other. The neighbor arrangement of cells in the HetNet environment includes a hierarchized arrangement in which one cell with smaller cell size is completely included in the coverage of the other cell with large cell size. In the following description, this problem recognized by the present inventors will be described in detail.
Non-patent literature 2 only discloses selecting listed cells (i.e., cells to be registered in a neighbor list) from a plurality of neighbor cells using a common criteria for the plurality of neighbor cells (more specifically, the number of measurement reports, and the number of handover attempts). However, in the HetNet environment including, for example, macro cells and pico cells, the number of measurement reports and the number of handover attempts regarding macro cells are typically larger than those of pico cells, which means it is difficult for pico cells to be registered in a neighbor list. Accordingly, selection of listed cells according to the common criteria may not be sufficient to optimize a neighbor list.
Further, the present inventors have studied the way to reduce handover failures using the technique disclosed in Patent literature 1 and reached a technique of preferentially registering, in a neighbor list, neighbor cells with large number or rate of handover failures. However, when register is performed according to this criteria in the HetNet environment, there may be a case in which a neighbor list is dominated by neighbor cells with large number of handover failures or high failure rate (i.e., mainly pico cells), and important neighbor cells with large number of handover attempts (i.e., mainly macro cells) are not registered in a neighbor list.
Furthermore, in order to efficiently use the technique disclosed in Non-patent literature 1, it is preferable to register neighbor cells arranged for load balancing (i.e., mainly pico cells) in the neighbor list. However, the method disclosed in Non-patent literature 2 for selecting cells to be registered in a neighbor list preferentially registers in the neighbor list neighbor cells that are likely to be selected as handover destination, i.e., neighbor cells with high radio quality, using common criteria for both a macro cell and a pico cell. Accordingly, there may be a case in which neighbor cells with high radio quality (i.e., mainly macro cells) are registered in a neighbor list and neighbor cells for load balancing (i.e., mainly pico cells) are not registered in a neighbor list.
The present invention has been made based on the aforementioned findings by the present inventors, and aims to provide a neighbor list optimization apparatus, a neighbor list generation apparatus, a base station apparatus, a neighbor list optimization method, and a program capable of optimizing a neighbor list suitable for the heterogeneous network (HetNet) environment.
A first aspect according to the present invention includes a neighbor list optimization apparatus that performs optimization of a neighbor list wirelessly transmitted from a base station that manages an objective cell. The neighbor list optimization apparatus includes a cell selection unit that selects, based on cell types defined according to a difference in cell sizes, a plurality of registered cells to be registered in the neighbor list from a plurality of other cells than the objective cell.
A second aspect according to the present invention includes a neighbor list generation apparatus. The neighbor list generation apparatus is coupled to the neighbor list optimization apparatus according to the first aspect of the present invention stated above, and is configured to generate a neighbor list including information of the plurality of registered cells selected by the neighbor list optimization apparatus.
A third aspect according to the present invention includes a base station apparatus. The base station apparatus is coupled to the neighbor list generation apparatus according to the second aspect of the present invention stated above, and is configured to wirelessly transmit the neighbor list including the information of the plurality of registered cells.
A fourth aspect according to the present invention includes a neighbor list optimization method that performs optimization of a neighbor list wirelessly transmitted from a base station that manages an objective cell. The method includes a step of selecting, based on cell types defined according to a difference in cell sizes, a plurality of registered cells to be registered in the neighbor list from a plurality of other cells than the objective cell.
A fifth aspect according to the present invention includes a computer program. This program is read into a computer to be executed by the computer, thereby causing the computer to execute the neighbor list optimization method in the fourth aspect of the present invention stated above.
According to the aspects of the present invention stated above, it is possible to provide a neighbor list optimization apparatus, a neighbor list generation apparatus, a base station apparatus, a neighbor list optimization method, and a program that are capable of performing neighbor list optimization suitable for a heterogeneous network (HetNet) environment in which a plurality of different types of cells having different sizes in coverage (cell sizes) are adjacently arranged.
In the following description, specific embodiments of the present invention will be described in detail with reference to the drawings. Throughout the drawings, the same or corresponding components are denoted by the same reference symbols, and overlapping description is omitted as necessary for the sake of clarification of description.
The handover optimization system 1 includes a neighbor list optimization unit 11 and a neighbor list generation unit 12. The neighbor list optimization unit 11 selects, based on cell types defined according to a difference in cell sizes, a plurality of registered cells from a plurality of neighbor cells 62 that are adjacent to the objective cell 61. This processing of selecting the registered cells is performed by a cell selection unit 110. According to such selection of registered cells, the neighbor list optimization unit 11 performs optimization of the neighbor list regarding the objective cell 61. The neighbor list generation unit 12 generates a neighbor list in which information of the plurality of registered cells selected by the neighbor list optimization unit 11 is described, and supplies the neighbor list to the base station 2.
In the example shown in
The cell selection unit 110 uses neighbor cell information (e.g., cell IDs) regarding the plurality of neighbor cells 62 to select registered cells. The neighbor cell information may be supplied to the neighbor list optimization unit 11 through various paths. For example, the neighbor cell information can be acquired by referring to measurement reports (MR) generated by the UEs 3 or MR tabulated information of measurement reports tabulated by the base station 2 or the like. Since each measurement report includes results of measuring radio quality of the neighbor cells 62 that are detected by the UE 3, it is possible to specify the neighbor cells 62. By acquiring the neighbor cell information from the measurement reports or the tabulated information of the measurement reports, it is possible to obtain the information regarding the neighbor cells even when the objective cell 61 and the neighbor cells 62 are managed by different element management systems (EMSs) or network management systems (NMSs) that are under the control of a communication carrier and it is impossible to share information between EMSs, between an EMS and an NMS, or between NMSs.
Alternatively, the neighbor cell information may be acquired from network design information held by the base station 2 or an EMS or an NMS that manages the base station 2. The network design information includes cell layout information created by an operator of a communication carrier or the like. By acquiring the neighbor cell information from the network design information, it is possible to avoid the problem raised when measurement reports are used, i.e., information of only a part of the neighbor cells 62 can be obtained from measurement reports from the UEs 3 since the number of UEs 3 connected to the objective cell 61 is small. The neighbor cell information may be acquired by both of the network design information and the measurement reports from the UEs 3 or the tabulated information of the measurement reports. Accordingly, it is possible to obtain comprehensive information of the neighbor cells 62.
Further, as described above, the cell selection unit 110 selects registered cells based on the cell types defined according to a difference in cell sizes. Accordingly, the cell selection unit 110 uses cell type information regarding the neighbor cells 62 to select registered cells. It is only required that the cell type information is able to identify the difference of cell sizes of the neighbor cells 62. For example, the cell type information may be information that directly indicates cell size. Alternatively, the cell type information may be an identifier that indicates which one of attributes divided according to cell size, such as a macro cell, a micro cell, a pico cell, and a femto cell, the neighbor cell belongs to. Alternatively, the cell type information may be information that indicates a transmission power level of a radio signal (e.g., a pilot signal, a downlink reference signal) from a base station that manages each neighbor cell 62. Further alternatively, the cell type information may be information indicating intended purpose of a cell, e.g., information which can identify whether it is a normal cell or a cell smaller than the normal cell arranged for load balancing.
The cell type information can be supplied to the neighbor list optimization unit 11 through various paths. For example, the neighbor list optimization unit 11 may acquire the cell type information from a measurement report generated by the UE 3 (or MR tabulated information) or network design information, or the combination thereof, as is similar to the neighbor cell information stated above. In the case of acquiring the cell type information from a measurement report of the UE 3 or MR tabulated information, the cell type information may be included in a transmission signal of each base station that manages the neighbor cell 62. Alternatively, the cell type information may be acquired from a base station that manages a neighbor cell included in the measurement report of the UE 3.
Further, the cell selection unit 110 uses a cell selection index (CSI) in order to select registered cells from te neighbor cells 62. A typical specific example of the cell selection index is, as disclosed in Non-patent literature 2, “the number of detections by the UEs 3 (the number of measurement reports)” and “the number of handover (HO) attempts”. “The number of HO attempts” means the number of attempts of outbound handover from the objective cell 61 to the neighbor cell 62.
The cell selection index may include “the number of HO failures”, “a HO failure rate”, or “an occupancy rate of the number of failures”. “The number of HO failures” means the number of failures of outbound handover from the objective cell 61 to the neighbor cell 62. Further, “the HO failure rate” means a ratio of the number of failures to the number of attempts of outbound handover from the objective cell 61 to the neighbor cell 62. Further, “the occupancy rate of the number of failures” means a proportion of the number of failures of outbound handover from the objective cell 61 to the neighbor cell 62 in the total number of failures of outbound handover from the objective cell 61. The cell selection unit 110 may preferentially select, as a registered cell, the neighbor cell 62 with large number of HO failures, high HO failure rate, or high occupancy rate of the number of failures. Accordingly, it is possible to adjust handover parameters such as CIOs for these cells, and to suppress occurrence of handover to a neighbor cell in which handover failure is expected. In short, it is possible to reduce handover failures.
In the following description, two examples of selecting registered cells based on cell types will be described.
The cell selection unit 110 uses different criteria to select registered cells for different cell types. For example, for neighbor macro cells, a first criterion in which a cell with large first index (e.g., at least one of the number of detections by the UEs 3 and the number of HO attempts) is preferentially selected as a registered cell is employed. Meanwhile, for other types of neighbor cells (e.g., neighbor pico cells), a second criterion in which a cell with large second index (e.g., the number of HO failures, the HO failure rate, or the occupancy rate of the number of failures) is preferentially selected as a registered cell is employed. The first index and the second index may each include a plurality of indices. In this case, it is only required that the combination of the indices for macro cells and the indices for other types of neighbor cells are different. Alternatively, while the cell selection unit 110 uses the same index regardless of the cell type, a threshold compared to the index when registered cells are selected may be changed depending on the cell types. For example, when “the number of detections by the UE 3” is used as the same index, a high threshold may be employed for the neighbor macro cells and a threshold which is lower than that of the macro cells may be employed for other neighbor cells (e.g., neighbor pico cells).
According to the method of using a common criterion without taking into consideration the cell types as disclosed in Non-patent literature 2, only a certain type of cells may be selected as registered cells, and other types of cells may be hardly selected. However, such problem can be evaded by changing the criterion to select registered cells depending on the cell types.
With reference to
A large number of measurement reports means that a neighbor cell is frequently observed (detected) by the UEs 3 that connect to the objective cell 61, and means a neighbor cell is highly likely to be the target of handover from the objective cell 61. Typically, such a neighbor cell often has a large coverage. Accordingly, the number of measurement reports may be employed as a selection index for cells having relatively large coverage. For example, when being applied to the HetNet environment including macro cells and pico cells, the number of measurement reports may be employed as a selection index for macro cell (macro selection index I_M).
When being applied to the HetNet environment including macro cells and pico cells, the number of attempts of Outbound HO to the neighbor cell 62 may be employed as a selection index for macro cell (macro selection index I_M). This is because, a large number of attempts of Outbound HO means that a neighbor cell is highly likely to be the target of handover from the objective cell 61, and typically such a neighbor cell is often a macro cell. Meanwhile, other three indices shown in
When being applied to the HetNet environment including macro cells and pico cells, the number of reconnections of UEs to the neighbor cell 62 may be employed as a selection index for macro cell (macro selection index I_M). This is because, that large number of reconnections of UEs means that a neighbor cell is highly likely to be the target of handover from the objective cell 61, and typically such a neighbor cell is often a macro cell. Meanwhile, the average dwell time of UEs in the neighbor cell 62 may be employed as a selection index for pico cell (pico selection index I_P). Even when the UE 3 performs handover to the neighbor cell 62 in which average dwell time is short, UE 3 is highly likely to immediately return to the objective cell 61. In short, there is a high possibility that handover repeating (ping-pong phenomenon) occurs. Otherwise, the UE 3 is highly likely to immediately perform handover to a cell different from the objective cell 61 after once being handed over to the neighbor cell 62. Typically, such phenomena often occur in pico cells with small coverage. Therefore, by preferentially registering neighbor pico cells in which average dwell time of UEs is short in a neighbor list and adjusting CIOs for these neighbor pico cells to large values, handover to these neighbor pico cells can be suppressed.
A second example of selecting registered cells based on cell types will be described. The cell selection unit 110 reserves a part of the maximum registrable number NCLmax in a neighbor list as a registrable number part (reserved number) for at least one cell type (e.g., pico cell). In other words, the cell selection unit 110 sets a registration upper limit number (i.e., a value obtained by subtracting a reserved number from the maximum registrable number NCLmax in the neighbor list) to at least one cell type (e.g., macro cell). Alternatively, the cell selection unit 110 may set a registrable number for each cell type by dividing the maximum registrable number NCLmax in the neighbor list, and may select registered cells within a range of each registrable number for each cell type. Accordingly, even when registered cells are selected by the same criterion regardless of cell types, it is possible to definitely register cells of a plurality of cell types in a neighbor list.
Further, the first and second examples of selecting registered cells stated above may be used in combination. By combining them, it is possible to employ selection criteria adapted to respective cell types, and to definitely register cells of a plurality of cell types in a neighbor list.
As stated above, the neighbor list optimization unit 11 according to this embodiment selects, based on the cell types that are defined according to a difference in cell sizes, registered cells to be registered in a neighbor list from the neighbor cells 62. In this way, this embodiment is possible to perform neighbor list optimization suitable for HetNet environment.
Incidentally, the arrangement of the neighbor list optimization unit 11 and the neighbor list generation unit 12 is appropriately determined based on the design concept of the network architecture. For example, when this embodiment is applied to an evolved packet system (EPS), as shown in
Further, when this embodiment is applied to a universal mobile telecommunications system (UMTS), as shown in
Further, as stated above, the cell type information may be acquired from the measurement report of the UE 3. In this case, the configuration as shown in
In the following description, an operation of updating a neighbor cell list by the handover optimization system 1, in particular, a specific example of the operation of selecting registered cells by the neighbor list optimization unit 11 will be described in detail.
In the following description, with reference to flowcharts in
The procedure shown in
In Step S12, the cell selection unit 110 selects registered cells from a neighbor macro cell set S1 among the neighbor cells 62, according to a macro cell selection criterion, within a range of the maximum registrable number NCLmax in the neighbor list. The macro cell selection criterion may be, for example, a criterion to perform any threshold comparison of the macro selection index I_M stated above. The neighbor macro cells selected in the Step S12 are recorded in a registered cell set S3. Further, the number of neighbor macro cells selected in the Step S12 is denoted by N3.
In Step S13, the cell selection unit 110 selects registered cells from a neighbor pico cell set S2 among the neighbor cells 62, according to a pico cell selection criterion, within a range of the remaining number of cells that can be registered in the neighbor list (i.e., NCLmax-N3). The pico cell selection criterion may be, for example, a criterion to perform any threshold comparison of the pico selection index I_P stated above. The neighbor pico cells selected in the Step S13 are recorded in a registered cell set S4. Further, the number of neighbor pico cells selected in the Step S13 is denoted by N4.
In Step S14, the cell selection unit 110 determines the cells included in the registered cell sets S3 and S4 as the registered cells. When the number of cells included in the registered cell sets S3 and S4 is below the maximum registrable number NCLmax, the remaining neighbor cells 62 that have not been selected may further be selected as the registered cells. Further, registered cells may be selected from the neighbor pico cells prior to the neighbor macro cells. However, in an environment in which the number of neighbor pico cells is larger than the maximum registrable number NCLmax, the selection is preferably performed in the order illustrated in
The procedure shown in
In Step S22, the cell selection unit 110 selects registered cells from the neighbor macro cell set S1 within a range of the registration upper limit number (NCLmax-NP) of macro cells. The neighbor macro cells selected in the Step S22 are recorded in the registered cell set S3. Further, the number of neighbor macro cells selected in the Step S22 is denoted by N3.
In Step S23, the cell selection unit 110 selects registered cells from the neighbor pico cell set S2 within a range of the remaining number of cells (NCLmax-N3) that can be registered in the neighbor list. The neighbor pico cells selected in the Step S23 are recorded in the registered cell set S4. Further, the number of neighbor pico cells selected in the Step S23 is denoted by N4.
In Step S24, the cell selection unit 110 determines the cells included in the registered cell sets S3 and S4 as the registered cells. When the number of cells included in the registered cell sets S3 and S4 is below the maximum registrable number NCLmax, the cell selection unit 710 may select the remaining neighbor cells 62 that have not been selected as the registered cells. Further, the cell selection unit may select registered cells from the neighbor pico cells prior to the neighbor macro cells. In this case, the cell selection unit 110 may select registered cells from the neighbor pico cell set S2 within a range of the reserved number NP for pico cells.
The procedure shown in
In Step S32, the cell selection unit 110 selects registered cells from the neighbor macro cell set S1 according to the macro cell selection criterion within a range of the registration upper limit number (NCLmax-NP) of macro cells.
In Step S33, the cell selection unit 110 selects registered cells from the neighbor pico cell set S2 according to the pico cell selection criterion within a range of the remaining number of cells (NCLmax-N3) that can be registered in the neighbor list.
Step S34 is similar to the Step S24 shown in
The procedure shown in
In Step S41 shown in
In Step S44, it is determined whether the total number of cells (N3+N4) included in both the registered cell sets S3 and S4 is below the maximum registrable number NCLmax. When N3+N4 is smaller than NCLmax (YES in Step S44), the cell selection unit 110 selects registered cells from the remaining neighbor macro cells (S1−S3) according to the macro cell selection criterion within a range of the remaining number of cells (NCLmax-N3−N4) that can be registered in the neighbor list (Step S45). In the macro cell selection criterion in Step S45, the threshold may be changed to facilitate the selection of the registered cells compared to the criterion in the Step S42. The neighbor macro cells selected in the Step S45 are recorded in a registered cell set S5. Further, the number of neighbor macro cells selected in the Step S45 is denoted by N5.
In Step S46, it is determined whether the number of cells (N3+N4+N5) included in the registered cell sets S3 to S5 is below the maximum registrable number NCLmax. When N3+N4+N5 is smaller than NCLmax (YES in Step S46), registered cells are selected from the remaining neighbor pico cells (S2-S4) according to the pico cell selection criterion within a range of the remaining number of cells (NCLmax-N3−N4−N5) that can be registered in the neighbor list (Step S47). In the pico cell selection criterion in Step S47, the threshold may be changed to facilitate the selection of the registered cells compared to the criterion in the Step S43. The neighbor pico cells selected in the Step S47 are recorded in a registered cell set S6. Further, the number of neighbor pico cells selected in the Step S47 is denoted by N6.
In Step S48, the cell selection unit 110 determines the cells included in the registered cell sets S3 to S6 as registered cells.
In the procedure shown in
In Step S51, the cell selection unit 110 calculates the evaluation parameter by multiplying the first index (e.g., the number of measurement reports from the UEs 3) by a first weighting factor W1 for each cell included in the neighbor macro cell set S1.
In Step S52, the cell selection unit 110 calculates the evaluation parameter by multiplying the second index (e.g., the number of HO failures) by a second weighting factor W2 for each cell included in the neighbor pico cell set S2.
In Step S53, the cell selection unit 110 selects registered cells from the neighbor cell sets (S1+S2) according to the order of the magnitude of the evaluation parameter within a range of the maximum registrable number NCLmax in the neighbor list.
The cell selection indices (CSIs) stated in the first embodiment are merely an example. Further, a plurality of indices may be used in combination. For example, regarding neighbor pico cells, such a selection criterion may be employed in which cells with high HO failure rate are selected from cells of which the number of measurement reports exceeds a threshold.
Further, instead of determining all of the registered cells in the neighbor list by dynamic cell selection by the neighbor list optimization unit 11, a part of registered cells may be statically registered in the neighbor list. In this case, dynamic cell selection by the neighbor list optimization unit 11 may be performed up to a number obtained by subtracting the number of cells that are statically registered from the maximum registrable number NCLmax in the neighbor list.
Further, although the first embodiment describes the objective cell 61 and the neighbor cells 62 adjacent to the objective cell 61, the neighbor cells 62 are only necessary to be any other cells than the objective cell 61. The other cells include a cell whose coverage partially overlaps the coverage of the objective cell 62, and a cell whose coverage is completely included in the coverage of the objective cell 62. Further, although coverage-overlap relation with the objective cell 62 is not clear, a cell detected by the UE 3, which is connects to the objective cell 62, are included in the other cells stated above.
Furthermore, although the first embodiment describes the case of always executing a selection of registered cells based on the cell type, the cell selection based on the cell type may be executed only when communication load of the objective cell 61 exceeds a predetermined threshold. When load balancing of the objective cell 61 is required, it is effective that cells which are placed for load balancing, such as pico cells, are selected as the registered cells. According to the variation stated in this example, when communication load of the objective cell 61 is large, it is possible to make the cells placed for load balancing, such as pico cells, favorably selected as registered cells among the neighbor cells. Meanwhile, when communication load of the objective cell 61 is small and load balancing is not required, selection of registered cells based on the cell type is not performed. Thus, it is possible to make macro cells, with large number of measurement reports or large number of HO attempts, be favorably selected as registered cells from the neighbor cells, thereby preferentially improving HO quality of macro cells.
The processing of selecting registered cells by the cell selection unit 110 stated in the first embodiment may be achieved using a semiconductor processing device such as an application specific integrated circuit (ASIC), a digital signal processor (DSP). Further, the processing of selecting registered cells by the cell selection unit 110 may be achieved by causing a computer such as a microprocessor to execute a program. More specifically, a program including instructions to cause a computer to execute the algorithm shown in any one of
The program can be stored and provided to a computer using any type of non-transitory computer readable media. Non-transitory computer readable media include any type of tangible storage media. Examples of non-transitory computer readable media include magnetic storage media (such as flexible disks, magnetic tapes, hard disk drives, etc.), optical magnetic storage media (e.g., magneto-optical disks), CD-ROM (Read Only Memory), CD-R, CD-R/W, and semiconductor memories (such as mask ROM, PROM (Programmable ROM), EPROM (Erasable PROM), flash ROM, RAM (random access memory), etc.). The program may be provided to a computer using any type of transitory computer readable media. Examples of transitory computer readable media include electric signals, optical signals, and electromagnetic waves. Transitory computer readable media can provide the program to a computer via a wired communication line (e.g., electric wires, and optical fibers) or a wireless communication line.
Note that the present invention is not limited to the above embodiment, but can be modified in various ways without departing from the gist of the present invention already described.
Described in the first embodiment stated above is an example of using “the number of HO failures”, “the HO failure rate”, or “the occupancy rate of the number of failures” of outbound HO from the objective cell 61 to the neighbor cell 62 as the cell selection index. Non-patent literature 2 does not disclose selecting registered cells using the criterion to preferentially select cells with relatively large number or high rate of outbound handover failures from the objective cell 61 to the neighbor cell 62 as stated above. Accordingly, there is described a reference example that performs selection of registered cells using the number of failures or the failure rate of outbound handover to the neighbor cell 62 as an index without necessitating consideration of the cell type defined according to a difference in cell sizes.
The handover optimization system 7 includes a neighbor list optimization unit 71 and a neighbor list generation unit 12. The neighbor list optimization unit 71 selects, using a criterion to preferentially select cells with relatively large number or high rate of outbound handover failures from the objective cell 61 to a neighbor cell, a plurality of registered cells to be registered in the neighbor list from the plurality of neighbor cells 62. The neighbor list generation unit 12 generates a neighbor list in which information of the plurality of registered cells selected by the neighbor list optimization unit 71 is described, and supplies the neighbor list to the base station 2.
As stated above, by employing, as a registered cell selection criterion, a criterion to preferentially select cells with relatively large number or high rate of outbound handover failures from the objective cell 61 to a neighbor cell, it is possible to preferentially includes information of neighbor cells with high frequency of occurrence of handover failure to a neighbor list. Accordingly, it is possible to adjust handover parameters such as CIO to these cells and to suppress occurrence of handover to a neighbor cell in which handover failure is expected. In short, it is possible to reduce handover failures.
In the following description, an operation of updating the neighbor cell list by the handover optimization system 7, in particular, a specific example of an operation of selecting registered cells by the neighbor list optimization unit 71 will be described in detail.
In Step S201 shown in
Next, a flow shown in
In Step S302, the cell selection unit 710 selects registered cells from the neighbor cell set S1 according to the criterion that uses the number of HO failures or the HO failure rate within a range of the registration upper limit number (NCLmax-NS2). The number NS2 indicates registrable number (reserved number) reserved for cells that are selected using other criteria described later. The neighbor cells selected in the Step S302 are recorded in the registered cell set S3. Further, the number of neighbor cells selected in the Step S302 is denoted by N3.
In Step S303, the cell selection unit 710 selects registered cells from the remaining neighbor cells (S1−S3) according to a criterion (e.g., criterion that uses the number of measurement reports from the UEs 3 as an index) different from the criterion that uses the number of HO failures or the HO failure rate, within a range of the remaining number of cells (NCLmax-N3) that can be registered in the neighbor list. The neighbor cells selected in the Step S303 are recorded in the registered cell set S4. Further, the number of neighbor cells selected in the Step S303 is denoted by N4.
In Step S304, the cell selection unit 710 determines the cells included in the registered cell sets S3 and S4 as the registered cells. When the number of cells included in the registered cell sets S3 and S4 is below the maximum registrable number NCLmax, the cell selection unit 710 may further select the remaining neighbor cells 62 that have not been selected as the registered cells.
The procedure shown in
In Step S401 shown in
In Step S404, it is determined whether the number of cells (N3+N4) included in the registered cell sets S3 and S4 is below the maximum registrable number NCLmax. When N3+N4 is smaller than NCLmax (YES in Step S404), registered cells are selected from the remaining neighbor cells (S1−S3−S4) according to the criterion that uses the number of HO failures or the HO failure rate, within a range of the remaining number of cells (NCLmax-N3−N4) that can be registered in the neighbor list (Step S405). In the registered cell selection criterion in Step S405, the threshold may be changed to facilitate the selection of the registered cells compared to the criterion in the Step S402. The neighbor cells selected in the Step S405 are recorded in the registered cell set S5. Further, the number of neighbor cells selected in the Step S405 is denoted by N5.
In Step S406, it is determined whether the number of cells (N3+N4+N5) included in the registered cell sets S3 to S5 is below the maximum registrable number NCLmax. When N3+N4+N5 is smaller than NCLmax (YES in Step S406), registered cells are selected from the remaining neighbor cells (S1−S3−S4−S5) according to a criterion that is different from the criterion that uses the number of HO failures or the HO failure rate, within a range of the remaining number of cells (NCLmax-N3−N4−N5) that can be registered in the neighbor list (Step S407). In the registered cell selection criterion in step S407, the threshold may be changed to facilitate the selection of the registered cells compared to the criterion in the Step S403. The neighbor cells selected in the Step S407 are recorded in the registered cell set S6. Further, the number of neighbor cells selected in the Step S407 is denoted by N6.
In Step S408, the cell selection unit 710 determines the cells included in the registered cell sets S3 to S6 as registered cells.
The processing of selecting registered cells by the cell selection unit stated in the reference example 1 may be achieved using a semiconductor processing device such as an application specific integrated circuit (ASIC) or a digital signal processor (DSP). Further, the processing of selecting registered cells by the cell selection unit 710 may be achieved by causing a computer such as a microprocessor to execute a program. More specifically, a program including instructions to cause a computer to execute the algorithm shown in any one of
A part or all of the first embodiment, other embodiments, and the reference example 1 stated above may be described as in the following Supplementary notes. However, it is not limited to them.
A neighbor list optimization apparatus that performs optimization of a neighbor list wirelessly transmitted from a base station that manages an objective cell, the neighbor list optimization apparatus comprising:
cell selection means for selecting a plurality of registered cells to be registered in the neighbor list using a first criterion to preferentially select, from a plurality of other cells than the objective cell, a cell with relatively large number or high rate of outbound handover failures from the objective cell to one of the plurality of other cells.
The neighbor list optimization apparatus according to Supplementary note 1, wherein the failure rate is a ratio of the number of failures of outbound handover from the objective cell to the number of attempts of outbound handover from the objective cell, or a proportion of the number of failures of outbound handover from the objective cell to the other cell in the total number of failures of outbound handover from the objective cell.
The neighbor list optimization apparatus according to Supplementary note 1 or 2, wherein the cell selection means selects the plurality of registered cells based further on cell types defined according to a difference in cell sizes.
The neighbor list optimization apparatus according to any one of Supplementary notes 1 to 3, wherein:
the cell types includes a first cell type indicating a first cell and a second cell type indicating a second cell having a cell size different from that of the first cell,
the cell selection means selects registered cells from the first cell group included in the plurality of other cells according to the first criterion, and
the cell selection means selects registered cells from the second cell group included in the plurality of other cells according to a second criterion that is different from the first criterion.
The neighbor list optimization apparatus according to Supplementary note 4, wherein the cell selection means selects the plurality of registered cells so that both of the first cell and the second cell are registered in the neighbor list even when one of the number of first cells and the number of second cells included in the plurality of other cells exceeds a maximum registrable number in the neighbor list.
The neighbor list optimization apparatus according to Supplementary note 4 or 5, wherein:
the second cell is a cell having a cell size larger than that of the first cell, and
the second criterion includes preferentially selecting, as the registered cells from the second cell group, cells in which outbound handovers from the objective cell to the second cell are expected to be frequently occurred.
The neighbor list optimization apparatus according to Supplementary note 4 or 5, wherein:
the second cell is a cell having a cell size larger than that of the first cell, and
the second criterion includes preferentially selecting, as the registered cells from the second cell group, cells with relatively large number of detections by mobile terminals, relatively large number of attempts of outbound handover from the objective cell to the second cell, or relatively large number of reconnections of mobile terminals.
The neighbor list optimization apparatus according to any one of Supplementary notes 4 to 7, wherein the cell selection means preferentially selects registered cells from the second cell group, and selects registered cells from the first cell group within a range of the remaining number of cells that can be registered in the neighbor list.
The neighbor list optimization apparatus according to Supplementary note 1 or 2, wherein the cell selection means preferentially selects registered cells from the plurality of other cells according to the first criterion, and selects, according to a second criterion that is different from the first criterion, registered cells from a cell group that has not been selected among the plurality of other cells within a range of the remaining number of cells that can be registered in the neighbor list.
The neighbor list optimization apparatus according to Supplementary note 9, wherein the cell selection means performs selection of registered cells based on the first criterion within a range of a predetermined number which is smaller than a maximum registrable number in the neighbor list.
A neighbor list generation apparatus coupled to the neighbor list optimization apparatus according to any one of Supplementary notes 1 to 10, the neighbor list generation apparatus generating a neighbor list including information of the plurality of registered cells.
A base station apparatus coupled to the neighbor list generation apparatus according to Supplementary note 11, the base station apparatus being configured to wirelessly transmit the neighbor list including the information of the plurality of registered cells.
This application is based upon and claims the benefit of priority from Japanese Patent Application No. 2010-284624, filed on Dec. 21, 2010, the disclosure of which is incorporated herein in its entirety by reference
| Number | Date | Country | Kind |
|---|---|---|---|
| 2010-284624 | Dec 2010 | JP | national |
| Filing Document | Filing Date | Country | Kind | 371c Date |
|---|---|---|---|---|
| PCT/JP2011/005116 | 9/12/2011 | WO | 00 | 8/9/2013 |
