Signal processing apparatus, method, and program

Abstract

To provide a signal processing device and the like related to forming an appropriate AP cluster by estimating the moving speed of UE.SOLUTION: A network includes a signal processing device (O-DU) and a second control device (Near-RT RIC), and a plurality of access points (APs) and user terminals (UE) in accordance with open radio access network (O-RAN) specifications. The signal processing device in the network performs a process of transmitting information on reference signal received power (RSRP) difference for each AP cluster from the signal processing device to the second control device, regarding an AP cluster to be formed for each UE, as information for enabling the second control device to form the AP cluster for each UE, after considering the moving speed of the UE.SELECTED DRAWING: Figure 10

Description

Technical Field

[0001] The present invention relates to a signal processing apparatus, a second control apparatus, a method, and a program in a network conforming to O-RAN specifications.

Background Art

[0002] In 5G (the fifth-generation mobile communication system)'s Massive MIMO (Massive Multi Input Multi Output), multiple array antennas on one RU (Raidio Unit, base station radio) form a beam, and the UE (User equipment) communicates with the RU one-to-one. As a "Beyond 5G" technology for this, in a user-centric RAN (Radio Access Network), transmission and reception with a UE (User equipment) are performed using a plurality of distributed APs (Access Points), and CF-mMIMO (Cell-free Massive MIMO) that centrally processes signals with a CPU (Central processing unit, signal processing apparatus) is used. This CF-mMIMO technology is described, for example, in Non-Patent Document 1.

[0003] FIG. 1 is a schematic diagram showing a comparison between 5G's Massive MIMO technology and "Beyond 5G"'s CF-mMIMO technology. As shown as EX10, for the former, the connection between the UE and the RU is only one-to-one, such as UE#1-RU#1, whereas, as shown as EX20, for the latter, the connection between the UE and the AP is such that the UE can be connected to a plurality of APs, such as UE#1:AP#1,2.

[0004] In addition, the Open Radio Access Network (O-RAN) Alliance is considering the opening and intelligentization of next-generation wireless access networks such as 5G (see, for example, Non-Patent Document 2). Among the O-RAN specifications formulated by the O-RAN Alliance, Non-Patent Document 2 defines an interface related to parameter information and the like used in the "Massive MIMO" technology. This Non-Patent Document 2 describes a control method and an interface for instructing parameter information of "Massive MIMO" to an O-DU (O-RAN Distributed Unit, distributed unit) using "Non-RT RIC" (Non-Real Time RAN Intelligent Controller), "Near-RT RIC" (Near-Real Time RAN Intelligent Controller), and the like.

[0005] "Non-RT RIC" is a controller that performs control at a cycle of 1 second or more. "Near-RT RIC" is a controller that performs control at a cycle of about 10 milliseconds to 1 second.

Prior Art Documents

Non-Patent Documents

[0006]

Non-Patent Document 1

Non-Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0007] Regarding CF-mMIMO, there are various considerations as follows.

[0008] (1) Regarding the formation of AP clusters Considering the wide-area deployment of CF-mMIMO, it is a problem in terms of scalability that the computational load related to signal processing increases when the UE communicates using all APs. To solve this, it is necessary to form an AP cluster, which is the minimum necessary group of APs that communicate with the UE, for each UE.

[0009] (2) Regarding the reformation of AP clusters When the UE moves, the SINR (Signal to Noise Interference Ratio) between the UE and the AP changes due to the change in the positional relationship between the UE and the AP. Therefore, it is necessary to reform the AP cluster so that the SINR of the UE in the AP cluster obtained by CF-mMIMO can be ensured even when the UE moves.

[0010] Figure 2 is a schematic diagram showing the change in SINR, etc. As can be seen in different areas A1, A2, etc., since the distribution of APs and UEs varies depending on the area, it is necessary to appropriately form an AP cluster. Also, even for the same UE, when it moves between time t1 and time t2, the area changes, so the positional relationship from the UE to each AP changes, and it is necessary to appropriately reform the AP cluster even after the AP cluster is formed.

[0011] Regarding the formation and reformation of AP clusters, there are individual problems as follows (1) to (4) in the existing technology.

[0012] (1) Difficulty in rule setting In Non-Patent Document 1, a method of forming an AP cluster based on rules using collected radio information such as RSRP (Reference Signal Received Power) has been proposed. In order to satisfy the radio quality required by the UE and to cope with all changes in SINR due to changes in the positional relationship between the UE and the AP, it is difficult to create all the rules because there are a wide variety of patterns to consider.

[0013] (2) Increase in computational complexity In Non-Patent Document 2, a method of determining an AP cluster by an optimization problem regarding radio quality using collected radio information such as RSRP and channel information has been proposed. In order to search for an AP cluster that can ensure the required SINR of the UE, it is necessary to determine for each AP whether the UE uses it. Therefore, L (where L is the number of APs), it is necessary to compare 2 combinations, so the calculation time for searching for an AP cluster on a general-purpose computer, which is generally used in RAN signal processing, becomes long, and there is a risk that the AP cluster cannot be continuously formed in response to the movement of the UE.

[0014] (3) In the prior art, appropriate AP cluster formation by estimating the moving speed of the UE has not been considered.

[0015] (4) In the prior art, keeping the size of the database referred to for forming an appropriate AP cluster within an appropriate range so as to keep the computational complexity within a reasonable range and form an AP cluster has not been considered.

[0016] In view of the above problem (3) in the prior art, an object of the present invention is to provide a signal processing apparatus (O-DU), a second control apparatus (Near-RT RIC), a method, and a program related to forming an appropriate AP cluster by estimating the moving speed of a UE.

Means for Solving the Problem

[0017] To achieve the above object, the present invention is a signal processing device in a network including a signal processing device (O-DU), a second control device (Near-RT RIC), a plurality of access points (APs), and user terminals (UEs) in accordance with Open Radio Access Network (O-RAN) specifications. In the second control device, in consideration of the moving speed of the UE, as information for enabling the formation of an AP cluster for each UE, regarding the AP cluster formed for each UE, information on the reference signal received power (RSRP) difference for each AP cluster is transmitted from the signal processing device to the second control device. Further, the present invention is a second control device in a network including a signal processing device (O-DU), a second control device (Near-RT RIC), a plurality of access points (APs), and user terminals (UEs) in accordance with Open Radio Access Network (O-RAN) specifications. Regarding the AP cluster formed for each UE, information on the reference signal received power (RSRP) difference for each AP cluster is periodically received from the signal processing device, and based on the periodically received RSRP difference information, it is managed as a plurality of databases differentiated by AP clusters. Among the differentiated plurality of databases, by referring to the database corresponding to the RSRP difference for each AP cluster of each UE received from the signal processing device, an AP cluster to be reformed for each UE is obtained. Further, the present invention is characterized by a method and a program corresponding to the signal processing device and the second control device.

Effects of the Invention

[0018] According to the present invention, regarding the AP cluster formed for each UE, by utilizing the information on the reference signal received power (RSRP) difference for each AP cluster, it is possible to estimate the moving speed of the UE and form an appropriate AP cluster.

Brief Description of the Drawings

[0019]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Embodiments for Carrying Out the Invention

[0020] FIG. 3 is a block diagram showing a configuration example of a radio access network according to an embodiment. The radio access network (RAN) 1 shown in FIG. 3 applies the O-RAN specification.

[0021] In RAN1, a plurality of access points (APs) 2 (AP#1, AP#2,...) are distributed. An AP cluster (access point group) that transmits and receives radio signals for each user equipment (UE) is formed from among the plurality of APs 2 distributed in RAN1. In the example of FIG. 1, for example, for UE#1, an AP cluster is formed from two APs 2 (AP#1, AP#2). Similarly for other UEs such as UE#2, an AP cluster corresponding to each UE is formed. (The formation is the same as the schematic example of FIG. 1 described above.)

[0022] However, the example of FIG. 3 is an example of a specific AP cluster formed for each UE at a certain specific moment (instant). As also shown in the schematic example of FIG. 2 described above, as the UE etc. move with the passage of time, the group of AP clusters configured by the RAN1 of the present embodiment will also change dynamically from moment to moment.

[0023] Each AP in the AP cluster of a certain UE transmits and receives radio signals to and from the UE via its own antenna 3. For example, the two APs 2 (AP#1, AP#2) in the access point group of UE#1 transmit and receive radio signals to and from UE#1 via their respective antennas 3.

[0024] Each AP cluster is communicatively connected to an O-DU (distributed unit) 6 that it is in charge of. The O-DU 6 executes signal processing related to the AP cluster that it is in charge of. In the example of FIG. 3, for example, O-DU#1 executes signal processing related to the AP cluster (AP#4, AP#5, AP#6) of UE#3. Similarly for other O-DU#2 etc., they execute signal processing related to the AP clusters that they are in charge of.

[0025] The signal processing related to the access point group executed by O-DU6 is signal processing such as SU-MIMO (Single User MIMO) and MU-MIMO (Multi-User MIMO).

[0026] O-DU#1 is provided at the Central site 4 and is connected to the core network CNW. O-DU#2 is provided at the Edge site #1_5 (here, an underscore is used as a delimiter to distinguish the identification number #1 and the reference numeral 5, and the same applies hereinafter), and is connected to the core network CNW via the central site 4. An MEC (Multi-access Edge Computing) server 7 is provided at the edge site #1_5. O-DU#3 is provided at the edge site #2_5. An MEC server 7 is provided at the edge site #2_5.

[0027] The communication control system 10 is communicatively connected to each site 4, 5. The communication control system 10 controls each O-DU6. The communication control system 10 provides information for forming an AP cluster corresponding to the UE to the O-DU6. The O-DU6 forms an AP cluster corresponding to the UE that it is in charge of based on the information provided from the communication control system 10.

[0028] FIG. 4 is a block diagram showing a configuration example of the communication control system 10 according to an embodiment. The communication control system 10 includes a "Non-RT RIC (Non-Real-Time RAN Intelligent Controller)" 11 and a "Near-RT RIC (Near-Real-Time RAN Intelligent Controller)" 12. The "Non-RT RIC" 11 is realized using the SMO (Service and Management Orchestration) framework.

[0029] "Non-RT RIC" 11 and "Near-RT RIC" 12 collect information such as KPIs (Key Performance Indicators) from the O-DU 6 via the O1 interface. The O1 interface is defined in the O-RAN (Non-Patent Document 2 cited above) specifications. The parameters that can be collected by the O1 interface correspond to the parameters defined in Non-Patent Document 3 below. For example, parameters such as "DL PRB usage", "UL PRB usage", "Average DL UE throughput", "Average UL UE throughput", "Number of PDU Sessions requested", etc. [Non-Patent Document 3] 3GPP TS28.552, "5G performance measurements", v16.1.0

[0030] "Non-RT RIC" 11 analyzes the information collected from the O-DU 6 and makes setting changes to "Near-RT RIC" 12 based on the analysis results. Here, "Non-RT RIC" 11 provides setting change information to "Near-RT RIC" 12 via the A1 interface. The A1 interface is defined in the O-RAN specifications.

[0031] "Near-RT RIC" 12 analyzes the information collected from the O-DU 6 and makes setting changes to the O-DU 6 based on the analysis results. Here, "Near-RT RIC" 12 provides setting change information to the O-DU 6 via the E2 interface. The E2 interface is defined in the O-RAN specifications.

[0032] FIG. 5 is a functional block diagram of a communication control system 10 according to an embodiment. As a functional configuration, "Non-RT RIC" 11 constitutes a first control unit 11 (first control device 11), "Near-RT RIC" 12 constitutes a second control unit 12 (second control device 12), and O-DU 6 constitutes a signal processing unit 12 (signal processing device 12). As shown in the figure, the first control unit 11 includes a recalculation unit 111, and the second control unit 12 includes an AP cluster candidate extraction unit 121 (abbreviated as "candidate extraction unit 121"), an AP cluster candidate DB 122 (abbreviated as "candidate DB 122"), and an AP cluster search unit 123 (abbreviated as "search unit 123").

[0033] FIG. 6 is a flowchart of the operation of the communication control system 10 according to an embodiment. The entire flow of FIG. 6 is repeated at each time t = t0, t1, t2,... at regular intervals according to a predetermined schedule or the like in advance, so as to reform (update) the AP cluster formed by each UE into one suitable for the current time, which relates to a communication control method according to an embodiment.

[0034] In the following description, it is assumed that the current time when the flow of FIG. 6 is being executed is time t = t n and the execution results of the flow of FIG. 6 at past times t = t0, t1, t2,..., t n are appropriately managed as history in the communication control system 10. (The details will be described below.)

[0035] In the following, by explaining each step of FIG. 6, mainly a formal description will be given regarding the exchange of information in the functional block configuration of FIG. 5. Regarding the specific meaning and details of these processing contents, after explaining each step of FIG. 6, it will be described later with reference to FIG. 7 and the like.

[0036] In step S1, each signal processing unit 6 (O-DU6) reports, via the E2 interface, information necessary for AP cluster reformation for each UE k belonging to the AP cluster it manages. (Hereinafter, it will be described assuming that k = 1, 2,... identifies UE#1, UE#2,... (k is the identifier of the UE). Also, the AP cluster will be described using k as the identifier, but this is separate from the identifier of the UE.)

[0037] Specifically, as also shown in FIG. 5, the following information is reported. ● Information reported from the signal processing unit 6 to the candidate extraction unit 121 · List x of SINR between UE and each AP sorted in descending order k · List AP of indices of APs corresponding to the SINR values k · Received power between UE and each AP · SINR required for the UE ● Information reported from the signal processing unit 6 to the candidate DB 122 · List of SINR of AP cluster candidates · Received power between UE and each AP · SINR achieved by that AP cluster candidate

[0038] As an option (an additional possible embodiment), in step S1, further as step S101, each signal processing unit 6 may report, via the O1 interface, information on the SINR distribution for each AP cluster for the AP cluster it manages to the recalculation unit 111.

[0039] In step S2, the candidate DB 122 outputs, for the candidate extraction unit 121 to refer to, information on the AP cluster candidates formed in the past so as to enable the processing of the candidate extraction unit 121 in the next step S3. Specifically, the following information is output to the candidate extraction unit 121. · List of SINR of AP cluster candidates · Received power between UE and each AP · The SINR achieved by the AP cluster candidate

[0040] As an option (an additional possible embodiment), in step S2, further as step S201, the recalculation unit 111 performs a recalculation process and reports the result to the candidate DB 122 via the A1 interface. In the candidate DB 122, according to the reported result (the maximum time and forgetting coefficient to be described later in FIG. 12), data determined to be unnecessary is deleted from the stored database information, and the database is updated to slim it down.

[0041] Note that when applying steps S101 and S201 as options, these two steps are applied in a pair. (Only one of steps S101 and S201 cannot be applied.)

[0042] In step S3, the candidate extraction unit 121 searches the candidate DB 122 using the information reported in steps S1 and S2, extracts an AP cluster candidate whose current UE speed, required radio quality, and achieved quality are close, and outputs the SINR distribution of each AP cluster to the search unit 123. Specifically, the following information is output to the search unit 123. · x k · AP k · A list of SINRs of an AP cluster candidate group in which the required quality and the difference in received power between the UE and each AP at the time of the previous AP cluster formation are close

[0043] In step S4, the search unit 124 that has received the above output in step S3 extracts elements from the head of x for the number of APs in each AP cluster candidate to obtain x', and selects an AP cluster with the minimum Euclidean distance between x' and the SINR list of each AP cluster candidate. For the number of APs in the selected AP cluster candidate, AP k The x' extracted from the head of the element k And select the AP cluster with the minimum Euclidean distance between the SINR list of each AP cluster candidate, and for the number of APs in the selected AP cluster candidate, AP kExtract the index of the AP from the beginning, and report this extraction result to the signal processing unit 6 (O-DU6) via the E2 interface as the set of indexes of the APs to be incorporated into the AP cluster of UE k (that is, as the result of the AP cluster to be reformed in UE k).

[0044] In this way, each UE can reform the AP cluster.

[0045] FIG. 7 is an explanatory diagram of the processing contents of each part 121, 122, 123 of the second control unit 12 ( "Near-RT RIC" 12) formally described in FIGS. 5 and 6 above. Hereinafter, the details of the processing contents of each part 121, 122, 123 will be described with reference to FIG. 7.

[0046] <Candidate DB 122…Explanation column EX1…Step S1> As shown in the explanation column EX1, in the candidate DB 122, by acquiring the information reported from the signal processing unit 6 (O-DU6) at each time t, its own database information is recorded and managed as follows.

[0047] That is, in the candidate DB 122, for the AP clusters formed before the current time t n The achieved SINR and the SINR of each AP in the AP cluster are sorted in descending order and recorded.

[0048] <Candidate DB 122 and extraction unit 121…Explanation column EX2…Steps S2, S3> Furthermore, as shown in the explanation column EX2, in the candidate DB 122 and the extraction unit 121, in steps S2 and S3, the extraction unit 121 performs the following extraction processing from the database information recorded in the candidate DB 122.

[0049] That is, assuming that it is determined that the achieved SINR satisfies the required SINR of the UE, an AP cluster candidate whose achieved SINR falls within the following range R is extracted from the candidate DB 122. (Note that the width of the range R is 1 dB, but this is an exemplary numerical value, and other widths may be set in advance.) This extraction is performed for each UE. Range R = [required SINR of UE, required SINR of UE + 1 dB]

[0050] As a schematic data example corresponding to the processing in the above description columns EX1 and EX2, in FIG. 7, as data D1, at each time t = t0, t1, t2,..., t before the current time t = t n A schematic example of the information reported from the signal processing unit 6 (O-DU6) to the candidate DB 122 is shown. n-1

[0051] Also, as data example D2, a schematic example of an "AP-SINR graph" (obtained by sorting the SINRs of each AP in the AP cluster in descending order) and its "achieved SINR", which are a total of K candidates 1, 2,..., K - 1, K in the candidate DB 122, is shown. (That is, in this schematic example D2, as an accumulation of past history, information regarding a total of K previously formed clusters is recorded.) Note that the "achieved SINR" can be calculated using an existing signal processing algorithm (for example, the partial-minimum mean square error method) disclosed in Non-Patent Document 1.

[0052] Furthermore, as an extraction processing example P3, an example is shown in which two candidates 1 and 2 (achieved SINRs are 15.1 dB and 15.2 dB, respectively) are extracted from among a total of K candidates on the condition that the achieved SINR falls within the range R = [required SINR of UE, required SINR of UE + 1 dB] from the above data D2.

[0053] <Extraction unit 121... Description columns EX3, EX4... Step S3> As shown in the description columns EX3 and EX4, in the extraction unit 121, as the processing in step S3, the following is further performed.

[0054] ​ That is, the SINR required by the UE is collected, and the SINR from the UE to each AP is collected and sorted in descending order. This sorting example is as shown in the schematic data example D4 for a certain UE #1.

[0055] Also, a list y of the SINR of the AP cluster candidates (referred to as candidate n) extracted from the candidate DB 122 n and a list x of the SINR extracted in descending order of the SINR around the UE for the number of APs included in the AP cluster candidate n n are created. This creation example is as shown in the schematic data example D5 in the form of listing the lists y1, y2 of candidates 1 and 2 (output of the extraction process P3) for the UE #1 and the list x1 of the SINR of the UE #1 in the data D4 on the same "AP-SINR graph". That is, in the data example D5, the "AP-SINR graph" x1, y1 listing the list y1 of candidate 1 and the list x1 of the SINR of the UE #1 with the number of APs combined in descending order, and the "AP-SINR graph" x2, y2 listing the list y2 of candidate 2 and the list x2 of the SINR of the UE #1 with the number of APs combined in descending order are shown.

[0056] <Effect of the extraction unit 121... Rules for fewer patterns are sufficient> According to the above processing of the search unit 123, by selecting an AP cluster based on the distribution of the radio environment around the UE instead of a specific UE-AP positional relationship as in the conventional method, the number of patterns of positional relationships to be considered can be reduced, so that an AP cluster that can satisfy the required SINR of the UE can be formed with rules for fewer patterns.

[0057] <Search unit 123... Explanation column EX5... Step S4> As shown in the explanation column EX5, in the search unit 123, as the processing in step S4, the following is performed.

[0058] That is, the Euclidean distance d(x n , y n) Select APs with high SINR by the number of APs in the AP cluster candidate n for which [[ID=]] is minimized, and use the result as the re-formed AP cluster for the UE.

[0059] This search example is schematically shown as search process P6 from two lists of data D5. That is, for candidate 1, "d(x1, y1) = 0.8", for candidate 2, "d(x2, y2) = 2.0", and since "d(x1, y1) = 0.8" of candidate 1 is the minimum value, the AP cluster of candidate 1 is used as the result of the re-formed cluster for the UE#1.

[0060] <Effect of the search unit 123... Re-formation by fast AP cluster selection> According to the above processing of the search unit 123, instead of the MIMO weight calculation whose computational complexity is proportional to the cube of the number of APs in the AP cluster by the conventional method, by using the Euclidean distance whose computational complexity is proportional to the number of APs in the AP cluster as a reference, the computational complexity is only a linear number, so that an AP cluster candidate that can ensure the required SINR of the UE can be selected at high speed, and the cluster re-formation result can be obtained.

[0061] That is, in order to search for an AP cluster from those stored in the DB (candidate DB122), the number of combinations to be searched is at most suppressed by the number of elements K stored in the DB. Furthermore, (since there is an extraction process of the candidate extraction unit 121 on the front stage side of the search unit 123,) since APs with similar radio qualities are extracted and then the AP cluster is searched, the number of combinations to be searched becomes smaller. Therefore, the AP cluster candidate can be selected at high speed rather than searching for the AP cluster from a wider range by the optimization problem.

[0062] As described above, the embodiments of FIGS. 5 to 7 provide a novel AP cluster formation algorithm that can solve the problems of the conventional method. However, in the O-RAN specification (Non-Patent Document 2), the O-RAN interface for implementing the algorithm is not defined. Therefore, below, the O-RAN interface required for the AP cluster formation algorithm according to this embodiment (including the case where additional embodiments of this embodiment are applied) will be described.

[0063] <SINR Distribution Information Collection (E2 Interface)> FIG. 8 is a diagram showing the procedure for collecting SINR distribution information by the E2 interface according to an embodiment.

[0064] In step S81, each O-DU6 transmits a "Performance measurements" message including information on the SINR distribution for each AP cluster to the "Near-RT RIC" 12 via the E2 interface. In step S82, the "Near-RT RIC" 12 recalculates the AP clusters based on the SINR information for each AP cluster to obtain the result of re-forming the AP clusters for each UE. In step S83, the "Near-RT RIC" 12 transmits an "Updated Configuration" message including information on the result of re-forming the AP clusters for each UE to each O-DU6 via the E2 interface.

[0065] Note that steps S81, S82, and S83 in FIG. 8 can be realized by step S1 in FIGS. 5 and 6, steps S2, S3, and S4, and the result transmission process in step S4, respectively.

[0066] <Effects of SINR Distribution Information Collection (E2 Interface) in FIG. 8> As a problem in the conventional method, since the SINR distribution for each AP cluster is not defined in the O-RAN standard, the RIC cannot collect the SINR distribution information for each AP cluster, and it is not possible to select an AP cluster with a similar radio environment from among the AP cluster candidates.

[0067] In response to the above problems, according to the procedure of FIG. 8, at step S81, the frequency distribution of SINR aggregated for each AP cluster is added as information collected from the base station to the interface between "O-DU6 → Near RT RIC 12" (E2). Therefore, as an effect, the SINR distribution of the AP cluster can be grasped, and an AP cluster candidate similar to the current radio environment of the UE can be selected, so that an AP cluster that can ensure the required radio quality of the UE can be formed.

[0068] FIG. 9 is a diagram showing an example of the format of SINR distribution information collection according to the present embodiment. According to such a format, each O-DU6 can aggregate the number of APs corresponding to the SINR ranges defined in Non-Patent Document 4 below (shown as columns C91, C92, and C93) for the APs within the AP cluster of each UE (shown as column C94) and report it to the "Near-RT RIC" 12. That is, for the information in columns C91, C92, and C93 defined in the existing standard, in the present embodiment, the information in column C94 as the aggregation result is newly added. [Non-Patent Document 4] 3GPP TS38.133, "Requirements for support of radio resource management", v17.4.0

[0069] <RSRP Distribution Information Collection (E2 Interface)> FIG. 10 is a diagram showing the procedure for collecting RSRP (Reference Signal Received Power) information by the E2 interface according to an embodiment.

[0070] According to the embodiment of FIG. 10, the following problems of the conventional method can be addressed. That is, in the conventional method, forming an AP cluster corresponding to the change in SINR due to the movement of the UE has not been considered in the first place. To form an AP cluster considering this, it is necessary to estimate the moving speed of the UE. However, since the distribution of the RSRP difference between the UE and the AP required for estimating the moving speed of the UE is not defined in the standard, the conventional method cannot form an AP cluster with a constant radio quality according to the moving speed of the UE.

[0071] Each procedure of FIG. 10 for solving the above problems will be described. In step S101, each O-DU 6 transmits a "Performance measurements" message including information on the RSRP difference for each AP cluster to the "Near-RT RIC" 12 via the E2 interface. In step S102, the "Near-RT RIC" 12 recalculates the AP cluster based on the SINR information for each AP cluster to obtain the result of reforming the AP cluster for each UE. In step S103, the "Near-RT RIC" 12 transmits an "Updated Configuration" message including information on the result of reforming the AP cluster for each UE to each O-DU 6 via the E2 interface.

[0072] Note that the RSRP difference is obtained as the RSRP difference between the current time t n and the previous past time t n-1 in the flow of FIG. 6 repeated at each time t = t0, t1,....

[0073] Note that steps S101, S102, and S103 in FIG. 10 can be realized by step S1 in FIGS. 5 and 6, steps S2, S3, and S4, and the result transmission process in step S4, respectively. However, if the embodiments shown in FIGS. 5 to 7 described above are referred to as the first embodiment and the embodiment shown in FIG. 10 is referred to as the second embodiment, since the first embodiment does not assume the use of RSRP difference information, the second embodiment can be realized by providing additional processing for using RSRP difference information with respect to the first embodiment.

[0074] FIG. 11 is a diagram for explaining a second embodiment realized by providing additional processing for using RSRP difference information in the first embodiment. The second embodiment can be realized by dividing the candidate DB122 in the first embodiment according to the range of RSRP differences and providing the database in the range corresponding to the RSRP difference of the corresponding UE for the extraction process from the candidate extraction unit 121. The following processes (1) to (4) etc. may be added to the first embodiment.

[0075] Process (1) "Near-RT RIC" 12 constructs the candidate DB122 by separating and storing the DB of the AP cluster for each average value of the difference in RSRP. In the example of FIG. 11, the candidate DB122 is constructed as three partial databases 122-1, 122-2, and 122-3 separated in three ranges: the range where the average of the RSRP difference is [0 to 1] dB, the range where the average is [1 to 10] dB, and the range where the average is 10 dB or more.

[0076] Process (2) Obtain the distribution of the difference in RSRP of the AP cluster formed by the UE from the E2 interface. The additional acquisition is as shown in step S91 of FIG. 10.

[0077] Process (3) Calculate the average value of the differences obtained in the above process (2), and select the DB of the corresponding AP cluster candidate from those stored in process (1).

[0078] Process (4) Obtain the SINR distribution from the interface of the first embodiment, select an AP cluster with a similar SINR distribution, and notify O-DU6.

[0079] As described above, the second embodiment is the same as the first embodiment except that the candidate DB122 is used separately based on the differential average of the RSRP.

[0080] <Control Information for DB Formation Policy Change (A1 Interface)> FIG. 12 is a diagram showing a usage procedure of control information for changing the DB formation policy by the A1 interface according to an embodiment. The embodiment of this FIG. 12 adds steps S101 and S201, which are optional in the embodiments of FIGS. 5 to 7, and solves the following problems.

[0081] That is, in the existing method, control information such as the size of the DB installed in "Near-RT RIC" and the deletion of similar elements is not defined. For this reason, as a problem, the information of the AP cluster candidates accumulates in the DB, causing the DB to become bloated, increasing the time to search for the AP cluster, and making it impossible to select an AP cluster that can ensure the radio quality required by the UE in response to the movement of the UE.

[0082] To address this problem, in the embodiment of FIG. 12, in step S121, a message is added to the interface between "Non-RT-RIC" 11 and "Near-RT RIC" 12 (A1) to notify the maximum number of elements of the AP cluster candidates in the candidate DB122 and the maximum time (forgetting factor) to hold the elements of the DB. In the candidate DB122, when the number of stored AP cluster candidates reaches this maximum number of elements or when the continuous holding time of the DB elements in the stored AP cluster candidates reaches this maximum time, the corresponding data is deleted in step S201, thereby preventing the DB from becoming bloated.

[0083] <Effect of Control Information for DB Formation Policy Change in FIG. 12 (A1 Interface)> That is, as an effect, by constructing a database of the number of elements of the DB of appropriate AP cluster candidates for "Near-RT RIC" 12, it is possible to select an AP cluster that can ensure the radio quality required by the UE in response to the movement of the UE while shortening the search time of the AP cluster.

[0084] Specifically, each step in FIG. 12 is as follows.

[0085] In step S121, each O-DU 6 transmits a "Performance measurements" message including information on the SINR distribution for each AP cluster to "Non-RT RIC" 11 via the O1 interface. In step S122, "Non-RT RIC" 11 recalculates the DB formation policy based on the interference power information for each AP cluster. In step S123, "Non-RT RIC" 11 transmits an "Updated Configuration" message including the maximum number of elements of the DB of AP cluster candidates and information on the maximum time for recording elements in the DB obtained from the result of the recalculation to "Near-RT RIC" 12 via the A1 interface. In response to the transmission result, the candidate DB 122 in "Near-RT RIC" 12 can be appropriately updated so that the size of its database information does not become excessively large.

[0086] Note that step S121 and S123, S124 in FIG. 12 can be realized by option step S101 in step S1 in FIGS. 5 and 6 and option step S201 in step S2, respectively.

[0087] FIG. 13 is a flowchart showing the details of step S123 in FIG. 12 and shows the details of the processing content of the recalculation unit 111 in FIG. 5.

[0088] In step S131, at the current time t in the candidate DB 122 nFrom the SINR elements of the newly input AP cluster candidates as new ones, select the elements with high SINR for the number of APs in each AP cluster candidate, and calculate the average of the Euclidean distances between the SINR distributions of the selected ones and the SINR distributions of each AP cluster candidate already stored in the candidate DB 122 as the past time. (Note that the process of "selecting the elements with high SINR for the number of APs in each AP cluster candidate" is a process for making it possible to calculate the Euclidean distance by matching the number of elements of each AP cluster candidate, similar to the example of the data D5 in FIG. 7.)

[0089] In step S132, the minimum value is compared with a predetermined threshold. Specifically, it is determined whether it continues to be less than the small threshold Γ_low or whether it continues to be greater than or equal to the large threshold Γ_high.

[0090] In the comparison result of step S132, when AP cluster candidates with a similarity (minimum value) less than the threshold Γ_low are continuously input (at each of the predetermined k + 1 times up to the current time t n to t n-1 , t n-2 , …, t n-k if they are continuously input), proceed to step S133, determine that there is sufficient information of AP cluster candidates stored in the candidate DB 122, and perform an update to reduce the maximum number of elements in the DB and the retention time of the elements.

[0091] In the comparison result of step S132, when AP cluster candidates with a similarity (minimum value) greater than or equal to the threshold Γ_high are continuously input (at each of the predetermined k + 1 times up to the current time t n to t n-1 , t n-2 , …, t n-k if they are continuously input), determine that there is a shortage of AP cluster candidates necessary for recognizing the radio environment, and perform an update to increase the maximum number of elements in the DB and the retention time of the elements.

[0092] Although the case division is omitted in Fig. 13, it is determined whether the condition of being less than the small threshold Γ_low continues in step S132 or whether the condition of being greater than or equal to the large threshold Γ_high continues. If neither condition is met, the maximum number of elements in the DB and the retention time of the elements are not updated and are kept as they are.

[0093] <Regarding the hardware configuration and providing this embodiment as a program> Each function of the communication control system 10 is realized by the communication control system 10 including computer hardware such as a CPU and a memory, and the CPU executing a computer program stored in the memory. Note that the communication control system 10 may be configured using a general-purpose computer device, or may be configured as a dedicated hardware device. For example, the communication control system 10 may be configured using a server computer connected to a communication network. Also, each function of the communication control system 10 may be realized by cloud computing. Further, the communication control system 10 may be realized by a single computer, or the functions of the communication control system 10 may be distributed and realized by a plurality of computers.

[0094] According to the communication control system 10 of this embodiment, it is possible to contribute to the infrastructure construction of information and communication technology. Thereby, it becomes possible to contribute to Goal 9 of the Sustainable Development Goals (SDGs) led by the United Nations, which is to "build resilient infrastructure, promote sustainable industrialization, and foster innovation."

Explanation of reference numerals

[0095] 10…Communication control system, 11…Non-RT RIC (first control unit, first control device), 12…Near-RT RIC (second control unit, second control device), 6…O-DU (signal processing unit, signal processing device)

Claims

1. A signal processing device in a network including an open radio access network (O-RAN) specification, a signal processing device (O-DU), a second control device (Near-RT RIC), a plurality of access points (APs), and user terminals (UEs), comprising: In the second control device, as information for narrowing down candidates for APs to be re-formed into an AP cluster for each UE in consideration of the moving speed of the UE, With respect to the AP cluster formed for each UE, information on the difference between the current time and the previous past time of the reference signal received power (RSRP) between the AP belonging to the AP cluster and the UE is transmitted from the signal processing device to the second control device. A signal processing device characterized by performing the process.

2. The process of transmitting the RSRP difference information from the signal processing device to the second control device is executed periodically, The signal processing device according to claim 1, wherein the RSRP difference information is obtained as a difference between adjacent times in each of the periodic transmissions.

3. In the second control device, the average value of the RSRP in the AP cluster of the difference between the current time and the previous past time of the RSRP between the AP belonging to the AP cluster formed by the UE and the UE, which is periodically received from the UE, is associated with the AP cluster, thereby managing the database of AP cluster candidates. Regarding the AP cluster formed by the UE received from the UE, an AP cluster associated with an RSRP average value determined to be close to the average value of the difference between the current time and the previous past time of the RSRP between the AP belonging to the AP cluster and the UE in the AP cluster is selected from the database. The signal processing device according to claim 1 or 2, characterized in that the moving speed of the UE is considered by narrowing down the candidate for re-forming the AP cluster.

4. A method executed by a signal processing device in a network including an open radio access network (O-RAN) specification, a signal processing device (O-DU), a second control device (Near-RT RIC), a plurality of access points (APs), and user terminals (UEs), comprising: In the second control device, as information for narrowing down candidates for APs to be re-formed into an AP cluster for each UE, taking into account the moving speed of the UE, For an AP cluster formed for each UE, a process of transmitting information on the difference between the current time and the previous past time of the reference signal received power (RSRP) between the AP belonging to the AP cluster and the UE from the signal processing device to the second control device is performed.

5. The process of transmitting the RSRP difference information from the signal processing device to the second control device is executed periodically, The method according to claim 4, wherein the RSRP difference information is obtained as a difference between adjacent times in each of the periodic transmissions.

6. In the second control device, the average RSRP value in the AP cluster of the difference between the current time and the previous past time of the RSRP between the AP belonging to the AP cluster formed by the UE and the UE, which is periodically received from the UE, is associated with the AP cluster to manage the database of AP cluster candidates, For the AP cluster formed by the UE received from the UE, the AP cluster associated with the average RSRP value determined to be close to the average value of the difference between the current time and the previous past time of the RSRP between the AP belonging to the AP cluster and the UE in the AP cluster is narrowed down as a candidate for re-forming the AP cluster from the database, thereby taking into account the moving speed of the UE. The method according to claim 4 or 5.

7. A program that causes a computer to function as a signal processing device in a network including a signal processing device (O-DU), a second control device (Near-RT RIC), a plurality of access points (APs), and user terminals (UEs) in accordance with Open Radio Access Network (O-RAN) specifications, In the second control device, as information for narrowing down candidates for APs to be re-formed into an AP cluster for each UE, taking into account the moving speed of the UE, For an AP cluster formed for each UE, a process of transmitting information on the difference between the current time and the previous past time of the reference signal received power (RSRP) between the APs belonging to the AP cluster and the UE from the signal processing device to the second control device is performed. A program characterized by that.

8. The process of transmitting the RSRP difference information from the signal processing device to the second control device is executed periodically, The program according to claim 7, wherein the RSRP difference information is obtained as a difference between adjacent times in each of the periodic transmissions.

9. In the second control device, by associating the average value of the RSRP in the AP cluster with the difference between the current time and the previous past time of the RSRP between the APs belonging to the AP cluster formed by the UE and the UE received periodically from the UE, manages a database of AP cluster candidates, Regarding the AP cluster formed by the UE received from the UE, the AP cluster associated with the average value of the RSRP in the AP cluster that is determined to be close to the average value of the difference between the current time and the previous past time of the RSRP between the APs belonging to the AP cluster and the UE is selected from the database. The program according to claim 7 or 8, characterized by considering the moving speed of the UE by narrowing down as candidates for re-forming the AP cluster.

Images