Communication control device, communication control method, and computer program

Abstract

This communication control device comprises: an acquisition unit that acquires information on computing units assigned to distributed units (DUs) in a radio access network based on O-RAN specifications; and a control unit that groups a plurality of user terminals on the basis of the information on the computing units so that signal processing related to the user terminals is collectively performed by one of the computing units, and notifies the DUs of the result of the grouping.

Description

Communication control device, communication control method, and computer program

[0001] The present invention relates to a communication control device, a communication control method, and a computer program. This application claims priority from Japanese Patent Application No. 2024-170267 filed in Japan on September 30, 2024, the content of which is incorporated herein by reference.

[0002] The Open Radio Access Network (O-RAN) Alliance is considering the opening and intelligentization of next-generation radio access networks (RANs) such as the 5th generation (5G) mobile communication system (see, for example, Non-Patent Documents 1-2). The O-RAN specifications established by the O-RAN Alliance define technologies related to, for example, fronthaul.

[0003] Non-Patent Document 3 also describes a technique related to the processing of P-MMSE (Partial Minimum Mean-Squared Error) in a Cell-Free massive MIMO (Multiple-Input Multiple-Output) system.

[0004] “O-RAN E2 Service Model (E2SM) 5.0”, O-RAN.WG3.E2SM-R003-v05.00, February 2024“O-RAN O1 Interface specification for O-DU 9.0”, O-RAN.WG5.O-DU-O1.0-R003-v09.00, February 2024Emil Bjornson, et al., “Scalable Cell-Free Massive MIMO Systems”, IEEE TRANSACTIONS ON COMMUNICATIONS, VOL. 68, NO. 7, JULY 2020

[0005] For example, in a large-scale CF-mMIMO system, it is conceivable to reduce computational load and power consumption by subdividing signal processing (Hi-PHY processing), such as MIMO matrix operations using P-MMSE with an Access Point (AP) cluster, for each User Equipment (UE). However, in the conventional O-RAN specification, the accelerator that performs signal processing for the Distributed Unit (DU) consists of a computing unit with multiple processors capable of parallel processing. However, since one computing unit is assigned to the signal processing of one UE, when the amount of signal processing for one UE is small, the processor utilization rate within the computing unit decreases, resulting in poor efficiency.

[0006] This invention was made in consideration of these circumstances, and its purpose is to improve the utilization efficiency of the computing unit that performs signal processing in the DU.

[0007] One aspect of the present invention is a communication control device comprising: an acquisition unit that acquires information on a computing unit to be assigned to a DU (Distributed Unit) in an O-RAN specification wireless access network; and a control unit that groups multiple user terminals so that signal processing for multiple user terminals is performed collectively on one computing unit based on the information on the computing unit, and notifies the DU of the grouping result. Another aspect of the present invention is a communication control device in which, in the above-mentioned communication control device, one computing unit is composed of multiple processors capable of parallel processing, and the control unit determines the number of user terminals to be assigned to one computing unit according to the number of processors capable of parallel processing in one computing unit. Another aspect of the present invention is a communication control device in which, in the above-mentioned communication control device, the control unit groups a set of user terminals that minimizes the cost based on the amount of computation when signal processing for multiple user terminals is performed collectively on one computing unit.

[0008] One aspect of the present invention is a communication control method to be executed by a communication control device in an O-RAN specification wireless access network, comprising: an acquisition step of acquiring information on a computing unit to be assigned to a DU (Distributed Unit); and a control step of grouping a plurality of user terminals so that signal processing for a plurality of user terminals is performed collectively by a single computing unit based on the information on the computing unit, and notifying the DU of the grouping result.

[0009] One aspect of the present invention is a computer program for performing communication control in an O-RAN specification wireless access network, which causes a computer to perform the following steps: an acquisition step of acquiring information of a computing unit to be assigned to a DU (Distributed Unit); and a control step of grouping multiple user terminals so that signal processing for multiple user terminals is performed collectively by one computing unit based on the information of the computing units, and notifying the DU of the grouping result.

[0010] According to the present invention, the utilization efficiency of the computing unit that performs signal processing of the DU can be improved.

[0011] This is a block diagram showing a schematic configuration example of a wireless access network (RAN) of a wireless communication system according to one embodiment. This is a diagram illustrating the signal processing (Hi-PHY processing) of a DU according to one embodiment. This is a diagram illustrating the signal processing (Hi-PHY processing) of a DU according to one embodiment. This is a diagram showing an example of a UE grouping method according to one embodiment. This is a diagram showing an example of a procedure for a communication control method according to one embodiment. This is a diagram showing an example of a procedure for a communication control method according to one embodiment. This is a diagram showing an example of a communication control method according to one embodiment. This is a diagram showing an example of a procedure for a communication control method according to one embodiment.

[0012] Embodiments of the present invention will be described below with reference to the drawings. Figure 1 is a block diagram showing a schematic configuration example of a wireless access network (RAN) of a wireless communication system according to one embodiment. The RAN1 shown in Figure 1 conforms to the O-RAN specification. However, in this embodiment, a fronthaul interface, which is not specified in the O-RAN specification, is newly added.

[0013] RAN1 comprises a RIC (RAN Intelligent Controller) 10 (communication control device), DU20 (DU#1, DU#2, ...), and an accelerator (ACC) 30. The RIC 10 and DU20 are configured to communicate with each other via a communication line.

[0014] The ACC30 includes a computing unit that performs signal processing for the DU20. The computing unit in the ACC30 has multiple processors capable of parallel processing.

[0015] The RIC 10 comprises an acquisition unit 11 and a control unit 12 as its functional units. The RIC 10 also includes a "Non-RT RIC (Non-Real Time RAN Intelligent Controller)" (not shown) and a "Near-RT RIC (Near-Real Time RAN Intelligent Controller)" (not shown), and the functions of the RIC 10 are realized by the "Non-RT RIC" and the "Near-RT RIC".

[0016] The acquisition unit 11 acquires information about the computing units assigned to the DU20 (computation unit information) using a default interface in the O-RAN specification. The computing unit information is information about the ACC 30 assigned to the DU20, and includes information such as the number of computing units 31 assigned to the DU20 and the number of processors 32 that each computing unit 31 has.

[0017] The control unit 12 groups multiple UEs based on the calculation unit information of the DU 20 so that signal processing for multiple UEs is performed together by a single calculation unit 31. The control unit 12 notifies the DU 20 of the result of this grouping.

[0018] The control unit 12 may determine the number of UEs to allocate to a single computing unit 31 according to the number of processors capable of parallel processing in that computing unit 31. The control unit 12 may also group together sets of UEs such that the cost based on the amount of computation is minimized when signal processing for multiple UEs is performed together in a single computing unit 31.

[0019] Figures 2 and 3 are diagrams illustrating the signal processing (Hi-PHY processing) of the DU according to this embodiment. Hereinafter, as an example of the Hi-PHY processing of the DU according to this embodiment, MIMO matrix operations using P-MMSE with an access point (AP) cluster will be described.

[0020] In Figure 2, access point (AP) clusters are formed for each of the three UE#1, UE#2, and UE#3. AP cluster 101 for UE#1 has 6 APs (AP cluster size = 6). AP cluster 102 for UE#2 has 4 APs (AP cluster size = 4). AP cluster 103 for UE#3 has 7 APs (AP cluster size = 7).

[0021] In Figure 3, the ACC 30 comprises a plurality of computing units 31 (31-1, 31-2, 31-3, ...). Each computing unit 31 has a predetermined number (16 in the example of Figure 3) of processors 32. In each computing unit 31, the 16 processors 32 are capable of parallel processing.

[0022] Figure 3(1) shows the processor utilization rate in conventional MIMO matrix operations. Conventionally, one computing unit 31 is assigned to each of the three UE#1, #2, and #3 shown in Figure 2, and the MIMO matrix operations for each UE#1, #2, and #3 are performed by each computing unit 31.

[0023] Specifically, the computing unit 31-1 shown in Figure 3(1) calculates a "1x6" weight matrix corresponding to AP cluster 101 (AP cluster size = 6) as the MIMO matrix operation for UE#1. As a result, in the computing unit 31-1, 6 of the 16 processors 32 are operational, and the remaining 10 processors 32 are inactive. Consequently, the conventional processor utilization rate of the computing unit 31-1 is "6 / 16".

[0024] Similarly, the computing unit 31-2 shown in Figure 3(1) calculates a "1x4" weight matrix corresponding to AP cluster 102 (AP cluster size = 4) as the MIMO matrix operation for UE#2. As a result, in the computing unit 31-2, 4 of the 16 processors 32 are operational, and the remaining 12 processors 32 are inactive. Consequently, the conventional processor utilization rate of the computing unit 31-2 is "4 / 16".

[0025] Similarly, the computing unit 31-3 shown in Figure 3(1) calculates a "1 x 7" weight matrix corresponding to AP cluster 102 (AP cluster size = 7) as the MIMO matrix operation for UE#2. As a result, in the computing unit 31-3, 7 of the 16 processors 32 are operational, and the remaining 9 processors 32 are inactive. Consequently, the conventional processor utilization rate of the computing unit 31-3 is "7 / 16".

[0026] Figure 3(2) is a diagram showing the processor utilization rate in MIMO matrix operations in this embodiment. In this embodiment, multiple UEs are grouped together so that MIMO matrix operations for multiple UEs are executed together. In the example in Figure 3(2), two of the three UEs #1, #2, and #3 shown in Figure 2, UE #1 and #3, are grouped into one group. As a result, the two UEs #1 and #3 are considered to be in an environment with an "AP cluster size = 8" consisting of eight APs belonging to AP clusters 101 and 103, and a "2 × 8" weight matrix is ​​calculated as the MIMO matrix operation for the two UEs #1 and #3.

[0027] Specifically, the calculation unit 31-1 shown in Figure 3(2) calculates a "2 x 8" weight matrix as a MIMO matrix operation for the two UE#1 and #3. As a result, all 16 processors 32 are operational in the calculation unit 31-2. Consequently, the processor utilization rate of the calculation unit 31-1 in this embodiment is "16 / 16". For the remaining UE#2, in this example, it is the same as the conventional method described above, and the processor utilization rate of the calculation unit 31-2 is "4 / 16".

[0028] In this embodiment, the processor utilization rate of the computing unit 31 is improved by grouping multiple UEs so that their MIMO matrix operations are executed together.

[0029] Figure 4 shows an example of a UE grouping method according to this embodiment. The example in Figure 4 shows the formation of a group of UEs that will be subject to MIMO matrix operations by P-MMSE with AP clustering applied.

[0030] In P-MMSE with AP clusters, a weight matrix of the form "(number of UEs) × (number of APs included in the AP cluster)" is generally calculated for each UE. However, in this embodiment, the weight matrix is ​​calculated for multiple UEs together. A specific example is shown in Figure 4 below. Note that subscripts, such as "a", may be written as "_a".

[0031] (Step S1) The surplus in the number of matrix elements resulting from combining the calculations of the weight matrices for UE#i and UE#j is defined as the grouping cost c_ij. In the example in Figure 4, when combining the calculations of the weight matrices for UE#1 and UE#3, a "2 × 8" weight matrix is ​​calculated, which is an increase of "16 - 6 - 7 = 3" matrix elements compared to calculating the weight matrices for UE#1 and UE#3 individually. As a result, the grouping cost is "c_13 = 3".

[0032] The pair of UEs with the minimum grouping cost is grouped. The matrix size and grouping cost are updated based on the grouping result. In the example in Figure 4, the pair UE#1 and #3 has the minimum grouping cost, so UE#1 and #3 are grouped, and UE#2 is left as is.

[0033] (Step S2) The grouping of the UEs with the smallest grouping cost is repeated until the grouping cost c_ij becomes equal to or greater than the computing unit size (number of processors in one computing unit) of the computing unit 31 of ACC30. In the example in Figure 4, the computing unit size is 16, and since the grouping cost of all UEs is 16, the grouping is terminated.

[0034] Figures 5, 6, and 8 show examples of the procedure for the communication control method according to this embodiment.

[0035] Figure 5 shows the case where UE grouping is performed using the "Near-RT RIC" of RIC10. The procedure for performing UE grouping using the "Near-RT RIC" of RIC10 will be explained below with reference to Figure 5.

[0036] (Step S11) The DU20 (O-DU) sends "Data collection" to the "Near-RT RIC" via the E2 interface. This "Data collection" notifies the "Near-RT RIC" of RSRP (Reference Signal Received Power), throughput (THP), etc., collected by the DU20. This is the default interface according to the O-RAN specification.

[0037] (Step S12) The Near-RT RIC forms an AP cluster for each UE housed in the DU20 based on the Data collection received from the DU20.

[0038] (Steps S13, S14, S15) Information about the ACC30 assigned to DU20 (ACC information) is notified to the "Near-RT RIC". Specifically, "Data collection" (ACC information) is sent from the "AAL implementation" to the IMS (Infrastructure Management Service) via the AALI-C-Mgmt interface, then "Data collection" (ACC information) is sent from the IMS to the "Non-RT RIC" via the O2ims interface, and then "Data collection" (ACC information) is sent from the "Non-RT RIC" to the "Near-RT RIC" via the A1 interface. These are the default interfaces according to the O-RAN specification. As a result, the "Near-RT RIC" (acquisition unit 11) acquires information about the computing unit assigned to DU20 (computation unit information).

[0039] (Step S16) The "Near-RT RIC" (control unit 12) groups the multiple UEs based on the computing unit information of the DU20 so that signal processing for multiple UEs is performed together in a single computing unit 31. For example, the "Near-RT RIC" (control unit 12) forms a group of UEs that will be subject to MIMO matrix operations by P-MMSE to which the AP cluster formed in step S12 is applied.

[0040] (Step S17) The "Near-RT RIC" (control unit 12) notifies the DU20 of the grouping result (Hardware Accelerator Assignment Control) from step S16. As an example according to this embodiment, the interface for notifying the DU20 of the grouping result "Hardware Accelerator Assignment Control" is included in the E2 interface. In this case, a new "Hardware Accelerator Assignment Control" message is added to the E2 interface.

[0041] An example of an "Information Element: IE" included in the "Hardware Accelerator Assignment Control" message added to the E2 interface is "Lists of Partial UE IDs per O-DU AAL Profile". "Lists of Partial UE IDs per O-DU AAL Profile" specifies a list of UE identifiers (UE IDs) that group Hi-PHY processing for each "AAL Profile". For example, "Lists of Partial UE IDs per O-DU AAL Profile" may be defined as a pair of "O-DU AAL Profile, Lists of Partial UE IDs". "O-DU AAL Profile" is defined in the O-RAN specification "O-RAN.WG6.AAL-GAnP-R003-v08.00 clause 7.1.3". "Lists of Partial UE IDs" is a list of "Partial UE IDs" for each group of UEs that group Hi-PHY processing. An example of the format for "Lists of Partial UE IDs" is "First UE group [Partial UE ID, Partial UE ID, …], Second UE group [Partial UE ID, Partial UE ID, …], …". "Partial UE ID" is defined in the O-RAN specification "O-RAN.WG3.E2SM-R003-v05.00 clause 6.2.2.18".

[0042] The interface that notifies the DU20 of the grouping result "Hardware Accelerator Assignment Control" corresponds to an interface that notifies the DU20 of grouping information indicating multiple UEs that have been grouped so that signal processing for multiple UEs is executed together by a single computing unit 31, with respect to the computing unit 31 assigned to the DU20.

[0043] DU20 (O-DU) receives a notification of the grouping result "Hardware Accelerator Assignment Control" via the E2 interface.

[0044] (Step S18) DU20 executes signal processing based on the grouping result "Hardware Accelerator Assignment Control".

[0045] (Step S19) DU20 notifies "Generate AAL-Profile instance for Hi-PHY processing per UE group" to "AAL implementation" via the AALI-C-App interface.

[0046] Figure 6 shows the case where the UE grouping is performed by the "Non-RT RIC" of the RIC10. Hereinafter, an example of the procedure when the UE grouping is performed by the "Non-RT RIC" of the RIC10 will be described with reference to Figure 6. In this Figure 6, the parts corresponding to each step of Figure 5 are given the same reference numerals and their descriptions are omitted.

[0047] Steps S11, S12, S13, S14 are the same as those in Figure 5. By Steps S13, S14, the "Non-RT RIC" (acquisition unit 11) acquires information on the calculation unit (calculation unit information) assigned to the DU20.

[0048] (Step S15a) The "Near-RT RIC" notifies "Data collection" (AP cluster, etc.) to the "Non-RT RIC" via the A1 interface. This is the default interface of the O-RAN specification.

[0049] (Step S16a) The "Non-RT RIC" (control unit 12) groups multiple UEs based on the computing unit information of the DU 20 so that signal processing for multiple UEs is performed together in a single computing unit 31. For example, the "Non-RT RIC" (control unit 12) forms a group of UEs that will be subject to MIMO matrix operations by P-MMSE to which the AP cluster notified by the "Near-RT RIC" in step S15a is applied.

[0050] (Step S17a) The "Non-RT RIC" (control unit 12) notifies the DU20 of the grouping result (Hardware Accelerator Assignment Control) from step S16a. As an example according to this embodiment, the O1 interface is included to notify the DU20 of the grouping result "Hardware Accelerator Assignment Control". In this case, a new "Hardware Accelerator Assignment Control" message is added to the O1 interface.

[0051] The following is an example of adding the "Hardware Accelerator Assignment Control" message to the O1 interface. The O1 interface between the SMO (Service and Management Orchestration) and the O-DU performs information queries and updates on a module-by-module basis. This is specified in the O-RAN specification "O-RAN.WG5.O-DU-O1.2-R003-v09.00". For this reason, as illustrated in Figure 7, the information "ro available-ue-list", indicated by reference numerals 201 and 202 in Figure 7, is added to the "o-ran-hardware" module.

[0052] Next, steps S18 and S19 are executed. Steps S18 and S19 are the same as in Figure 5.

[0053] Figure 8 shows the case where UE grouping is performed using the "Non-RT RIC" of RIC10. Below, with reference to Figure 8, Example 2 of the procedure for grouping UE using the "Non-RT RIC" of RIC10 will be explained. In Figure 8, the parts corresponding to each step in Figure 6 are given the same reference numerals, and their explanations are omitted.

[0054] Steps S11, S12, S13, S14, S15a, and S16a are the same as in Figure 6. Next, step S17b is executed.

[0055] (Steps S17b, 17c) The "Non-RT RIC" (control unit 12) notifies the "AAL Implementation" of the grouping result (executePerUeGroup) from step S16a via IMS. The "AAL Implementation" is a hardware infrastructure that implements DU20 in software. Specifically, "executePerUeGroup" is sent from the "Non-RT RIC" to IMS via the O2ims interface, and then "executePerUeGroup" is sent from IMS to the "AAL implementation" via the AALI-C-Mgmt interface. In this case, a new "executePerUeGroup" message is added to both the O2ims interface and the AALI-C-Mgmt interface.

[0056] Step S18 is the same as in Figure 6.

[0057] (Step S19a) DU20 notifies the AAL implementation via the AALI-C-App interface to "Generate AAL-Profile instance for Hi-PHY processing". Upon receiving the notification of "Generate AAL-Profile instance for Hi-PHY processing", the AAL implementation divides the job into groups of UEs within the OS and assigns them to processors.

[0058] In the examples shown in Figures 5, 6, and 8 above, "Near-RT RIC" or "Non-RT RIC" implemented the functions of the acquisition unit 11 and the control unit 12, but this is not limited to these. The functions of the acquisition unit 11 and the control unit 12 may also be implemented using both "Near-RT RIC" and "Non-RT RIC".

[0059] According to this embodiment, the advantage is that the utilization efficiency of the computing unit 31 that performs signal processing of DU20 can be improved.

[0060] Furthermore, this will enable improvements in overall service quality, such as in wireless communication systems, and will contribute to Goal 9 of the United Nations-led Sustainable Development Goals (SDGs): "Build resilient infrastructure, promote sustainable industrialization and foster innovation."

[0061] Although embodiments of the present invention have been described in detail above with reference to the drawings, the specific configuration is not limited to these embodiments, and design modifications and the like are also included within the scope of the gist of the present invention.

[0062] Alternatively, computer programs for realizing the functions of each of the above-mentioned devices may be recorded on a computer-readable recording medium, and the programs recorded on this recording medium may be loaded into a computer system and executed. The term "computer system" here may include hardware such as an operating system and peripheral devices. Furthermore, "computer-readable recording medium" refers to writable non-volatile memory such as flexible disks, magneto-optical disks, ROMs, and flash memory, portable media such as DVDs (Digital Versatile Discs), and storage devices such as hard disks built into a computer system.

[0063] Furthermore, "computer-readable recording media" includes volatile memory (e.g., DRAM (Dynamic Random Access Memory)) within a computer system that acts as a server or client when a program is transmitted via a network such as the Internet or a communication line such as a telephone line, which retains the program for a certain period of time. In addition, the above program may be transmitted from the computer system that stores the program in a storage device, etc., to another computer system via a transmission medium or by transmission waves within the transmission medium. Here, the "transmission medium" for transmitting the program refers to a medium that has the function of transmitting information, such as a network such as the Internet or a communication line such as a telephone line. Furthermore, the above program may be for the purpose of realizing a part of the above-mentioned functions. In addition, it may be a so-called differential file (differential program) that can realize the above-mentioned functions in combination with a program already recorded in the computer system.

[0064] According to the present invention, it is possible to improve the utilization efficiency of the computing unit that performs signal processing of the DU.

[0065] 1...Wireless access network, 10...RIC, 20...DU, 30...Accelerator (ACC)

Claims

1. A communication control device comprising: an acquisition unit that acquires information on computing units assigned to a DU (Distributed Unit) in an O-RAN specification wireless access network; and a control unit that groups multiple user terminals so that signal processing for multiple user terminals is performed collectively by one computing unit based on the information on the computing units, and notifies the DU of the grouping result.

2. The communication control device according to claim 1, wherein one computing unit is composed of a plurality of processors capable of parallel processing, and the control unit determines the number of user terminals to be assigned to one computing unit according to the number of processors capable of parallel processing in one computing unit.

3. The communication control device according to claim 2, wherein the control unit groups a set of user terminals such that the cost based on the amount of computation is minimized when signal processing for multiple user terminals is performed together by a single computing unit.

4. A communication control method to be executed by a communication control device in an O-RAN specification wireless access network, comprising: an acquisition step of acquiring information on a computing unit to be assigned to a DU (Distributed Unit); and a control step of grouping multiple user terminals so that signal processing for multiple user terminals is performed collectively by one computing unit based on the information on the computing unit, and notifying the DU of the grouping result.

5. A computer program for performing communication control in an O-RAN specification wireless access network, comprising: an acquisition step of acquiring information of a computing unit to be assigned to a DU (Distributed Unit); a control step of grouping multiple user terminals so that signal processing for multiple user terminals is performed collectively by one computing unit based on the information of the computing units, and notifying the DU of the grouping result; and a computer program for causing a computer to perform these steps.

Images