Base station, communication method, and communication program

Abstract

Provided is a base station capable of reducing the possibility of wasteful power consumption by always designating a GPU as an offload destination. This base station relays communication between terminals and comprises: a first processor; a second processor which has a higher processing performance and higher power consumption than that of the first processor; and a distribution unit that distributes L1 layer processing to the first processor and the second processor, by distributing L1 layer processing with a terminal to the second processor in a communication with the terminal where artificial intelligence multi-access edge computing (AIMEC) is used, and distributing L1 layer processing with a terminal to the first processor in a communication with the terminal where AIMEC is not used. The first processor and the second processor execute the L1 layer processing distributed by the distribution unit.

Description

Base station, communication method and communication program

[0001] The present invention relates to a base station, and more particularly to a base station, a communication method, and a communication program that can realize offloading to a GPU depending on whether or not AIMEC is used.

[0002] In conventional communications, heavy High-PHY processing (physical layer, L1 layer processing) is offloaded to accelerator cards such as GPUs. There is a technology for offloading High-PHY processing in vRAN, which is composed of a CPU and accelerator cards such as GPUs, and the offload destination is specified as an accelerator card such as a GPU. There is also a technology for switching the processor responsible for processing depending on the amount of traffic.

[0003] In 5G communications, a technology has been proposed in which decoding processing is performed in parallel using multiple processor pipelines (see Patent Document 1, etc.).

[0004] In US Pat. No. 6,279,649, the information includes data that has been processed by soft demapping, and decoding the information includes layer demapping, descrambling, and de-rate matching the data in parallel.

[0005] Special Publication No. 2022-553311

[0006] However, in the technology of Patent Document 1, high-PHY processing (physical layer, L1 layer processing), which is a heavy processing load, is offloaded to the GPU, which may result in excessive power consumption because the GPU consumes a very large amount of power.

[0007] The present invention has been made in consideration of the above points, and provides a base station that reduces the possibility of unnecessary power consumption by always specifying the GPU as the offload destination, and that determines whether to allocate to the GPU depending on whether the communication uses AIMEC, so that power consumption is reduced without always allocating to the GPU.

[0008] That is, the base station of the embodiment is a base station that relays communications between terminals, and includes a first processor, a second processor that has higher processing performance and higher power consumption than the first processor, and an allocation unit that allocates L1 layer processing to the first processor and the second processor, and allocates L1 layer processing with a terminal to the second processor when AIMEC (Artificial Intelligence Multi-access Edge Computing) is used in communications with the terminal, and allocates L1 layer processing with the terminal to the first processor when AIMEC is not used in communications with the terminal, and is characterized in that the first processor and the second processor execute the L1 layer processing allocated by the allocation unit.

[0009] Furthermore, in the base station, the allocation unit may allocate L1 layer processing with the terminal to a second processor when the traffic volume in communication with the terminal is equal to or greater than a predetermined threshold, and may allocate L1 layer processing with the terminal to a first processor when the traffic volume in communication with the terminal is less than the predetermined threshold.

[0010] Furthermore, in the base station, the allocator may increase the predetermined threshold when the utilization rate of the second processor is relatively high, and may decrease the predetermined threshold when the utilization rate of the second processor is relatively low.

[0011] Furthermore, in the base station, the first processor may be a CPU, the second processor may be a GPU, and the first processor and the second processor may form a vRAN.

[0012] Furthermore, the base station may include a first communication unit corresponding to the first processor and performing communication with the terminal according to the control of the first processor, and a second communication unit corresponding to the second processor and performing communication with the terminal according to the control of the second processor.

[0013] The communication method of the embodiment includes a first processor and a second processor having higher processing performance and higher power consumption than the first processor, and a computer of a base station that relays communications between terminals allocates L1 layer processing to the first processor and the second processor, and executes an allocation step in which, when AIMEC is used in communications with a terminal, L1 layer processing with the terminal is allocated to the second processor, and when AIMEC is not used in communications with the terminal, L1 layer processing with the terminal is allocated to the first processor, and a processing step in which the first processor and the second processor execute the L1 layer processing allocated in the allocation step.

[0014] The communication program of the embodiment comprises a first processor and a second processor having higher processing performance and higher power consumption than the first processor, and allocates L1 layer processing to the first processor and the second processor in a computer of a base station that relays communications between terminals, and realizes a distribution function that allocates L1 layer processing with a terminal to the second processor when AIMEC is used in communications with the terminal, and allocates L1 layer processing with the terminal to the first processor when AIMEC is not used in communications with the terminal, and a processing function that causes the first processor and the second processor to execute the L1 layer processing allocated by the distribution function.

[0015] The base station of the present invention is a base station that relays communications between terminals, and includes a first processor, a second processor that has higher processing performance and higher power consumption than the first processor, and an allocation unit that allocates L1 layer processing to the first processor and the second processor when AIMEC (Artificial Intelligence Multi-access Edge Computing) is used in communications with a terminal, and allocates L1 layer processing with the terminal to the second processor when AIMEC is not used in communications with the terminal, and allocates L1 layer processing with the terminal to the first processor when AIMEC is not used in communications with the terminal.Since the first processor and the second processor execute the L1 layer processing allocated by the allocation unit, by always specifying the GPU as the offload destination, unnecessary power consumption is suppressed, and by determining whether to allocate to the GPU depending on whether the communication uses AIMEC, processing is not always allocated to the GPU, thereby suppressing power consumption.

[0016] Fig. 1 is a system diagram showing a configuration example of a base station according to an embodiment; Fig. 2 is a block diagram showing a CU of the base station according to an embodiment; Fig. 3 is a first flowchart showing a processing allocation operation as an operation example of the base station according to an embodiment; and Fig. 4 is a second flowchart showing a threshold change operation as an operation example of the base station according to an embodiment.

[0017] The base station of the embodiment is a base station that uses a Radio Access Network Intelligent Controller (RIC). RIC is a technology that makes the radio access network (RAN) more open and intelligent, and uses AI / ML (artificial intelligence / machine learning). This makes the operation of the network that uses RIC more efficient and automated.

[0018] The base station of the embodiment is a base station that can realize offloading to a GPU depending on whether or not AIMEC is used. The base station of the embodiment is a base station that uses AIMEC (Artificial Intelligence Multi-access Edge Computing). AIMEC executes AI using MEC (Multi-access Edge Computing), and MEC is a network computing technique for making the most of 5G (fifth generation communication standard).

[0019] Traditionally, servers for various services have often been installed on the cloud, and communication with terminals has been via the Internet, but this can result in long distances between the terminal and the server, which can cause delays. MEC, therefore, speeds up response and reduces latency by building edge servers closer to the terminal, such as on an internal network or local 5G. This improves real-time performance and enables more secure communications.

[0020] Therefore, the base station etc. (communication method) of the embodiment is positioned as a technology for realizing offloading to the GPU depending on whether or not AIMEC is used, by allocating processing to different processors depending on whether or not AIMEC is used.

[0021] The configuration of a base station 100 according to the embodiment is shown in the schematic diagram of FIG. 1 . The base station 100 is a radio base station for a 5G mobile communication system equipped with an RU / DU / CU. The base station 100 communicates with a user terminal 300 via an RU 200, which is a communication unit. The base station 100 includes a first system in which a CU 110 is connected to a DU 120, a GPU 140, and an RU (first communication unit) 200a, and a second system in which the CU 110 is connected to a DU 130, a CPU 150, and an RU (second communication unit). In the first system, the RU 200a communicates with the user terminal 300a using AIMEC. In the second system, the RU 200b communicates with the user terminal 300b without using AIMEC. The CPU 150 corresponds to the first processor, and the GPU 140 corresponds to the second processor. The CPU 150 corresponding to the first processor and the GPU 140 corresponding to the second processor form a vRAN. vRAN stands for virtual Radio Access Network, a technology that realizes a virtual radio access network, i.e., a hardware configuration in software. In wireless communication using vRAN, a general-purpose server can be used without using a dedicated hardware unit in the master station as in wireless communication without vRAN. This enables improvements and modifications. In addition, because it is a general-purpose server, other services can be placed on the server within the base station 100, enabling low-latency services to be provided.

[0022] RU / DU / CU are classifications of functions that make up a wireless base station in a 5G mobile communication system, and organize the functions and roles that make up the radio access network (RAN: Radio Access Network) that connects the base station and terminals, and are connected in the order of terminal - RU - DU - CU - core network. The "RU" (Radio Unit) controls the antenna and communicates radio waves with the terminal, and also controls MIMO (Multiple Input Multiple Output) and beamforming. The "DU" (Distributed Unit) performs signal modulation and demodulation, MAC layer communication control, etc. The "CU" (Central Unit) is responsible for processing "PDCP" (Packet Data Convergence Protocol), which controls subordinate DUs and RUs, connects to the core network, and encrypts packets, as well as "RRC" (Radio Resource Control), which manages radio resources for terminals. Base stations are divided into "child stations" with antennas and RUs, and "master stations" with CUs, and the master station aggregates nearby child stations and controls the sending and receiving of data.

[0023] 2 is a block diagram showing the functional units of the CU 110. The CU 110 includes an input unit 111, a communication unit 112, a control unit 113, a distribution unit 114, a storage unit 115, and an output unit .

[0024] The input unit 111 accepts information that the base station 100 receives from the user terminal 300 via the RU 200. This information acceptance includes both cases where the information is accepted from the user terminal 300a via the RU (first communication unit) 200a using AIMEC, and where the information is accepted from the RU (second communication unit) 200b without using AIMEC.

[0025] The communication unit 112 communicates with the user's terminal 300 via the RU 200. This communication includes both cases where communication with the user's terminal 300a is performed via the RU (first communication unit) 200a using AIMEC, and where communication with the user's terminal 300b is performed via the RU (second communication unit) 200b without using AIMEC.

[0026] The control unit 113 includes a distribution unit 114. The control unit 113 controls the distribution unit 114 to distribute whether the processing of information received from the input unit 111, the user's terminal 300 via the RU 200, is to be performed by the GPU 140 or the CPU 150.

[0027] Specifically, within the same coverage area under the jurisdiction of one server of the base station 100, there exists a first system in which the CU 110 is connected to the DU 120, the GPU 140, and the RU (first communication unit) 200a, as described above, and a second system in which the CU 110 is connected to the DU 130, the CPU 150, and the RU (second communication unit). CPU stands for Central Processing Unit and is responsible for general information processing. GPU stands for Graphic Processing Unit and is responsible for advanced information processing involving image processing. Therefore, the GPU has higher processing performance than the CPU. However, this results in higher power consumption. As a result, there is a possibility that excessive power consumption will occur in the base station 100.

[0028] Therefore, to reduce the cost of the base station 100, it is desirable to have the CPU 150 perform as much processing as possible, but because complex processing such as image processing takes time, it is preferable to have the GPU 140 perform it. However, if the GPU 140 performs many processes, power consumption will increase, and there is a possibility that excessive power consumption will occur.

[0029] Therefore, in order to execute processing quickly while avoiding high power consumption, it is necessary to offload, i.e., reduce the burden, by dividing the processing into those to be executed by the GPU 140 and those to be executed by the CPU 150. Specifically, within the same server, it is necessary to prepare a system that uses the GPU 140 and a system that uses the CPU 150 for L1 processing, and to determine which system to use. L1 refers to the physical layer, a layer of communication protocols in computer networks that handles physical connection methods. In other words, it is a layer that defines the physical connections, electrical characteristics, transmission media, etc. necessary for computers to communicate with each other. The L1 layer deals with the physical properties of communication, such as bit flow, electrical signal conversion, signal transmission distance, bandwidth, and transmission speed.

[0030] Which system to use is determined based on two criteria: whether AIMEC is used and whether the traffic volume is large. When AIMEC is used, the control unit 113 controls the allocating unit 114 so that processing is performed in the system using GPU 140 to reduce transmission delay, and when AIMEC is not used, processing is performed in the system using CPU 150. When the traffic volume is large, the control unit 113 controls the allocating unit 114 so that processing is performed in the system using GPU 140 due to the processing load, and when the traffic volume is small, processing is performed in the system using CPU 150.

[0031] The allocation unit 114 can allocate L1 layer processing with the terminal to the second processor, GPU 140, when the traffic volume (communication volume) in communication with the terminal is equal to or greater than a predetermined threshold T, and can allocate L1 layer processing with the terminal to the first processor, CPU 150, when the traffic volume (communication volume) in communication with the terminal is less than the predetermined threshold T.

[0032] The allocating unit 114 can increase the threshold T when the utilization rate of the second processor, GPU 140, is relatively high (above threshold U1), and can lower the threshold T when the utilization rate of the second processor, GPU 140, is relatively low (below threshold U2).

[0033] Furthermore, when the allocating unit 114 determines that, in a certain base station 100, a terminal 300 that is having an RU 200b connected to a CPU 150 execute L1 layer processing should be allocated to the GPU 140 midway through the process, the terminal 300 switches its connection from the RU 200b connected to the CPU 150 to the RU 200a connected to the GPU 140, that is, performs a handover. Conversely, when the allocating unit 114 determines that, in a certain base station 100, a terminal 300 that is having an RU 200a connected to a GPU 140 execute L1 layer processing should be allocated to the CPU 150 midway through the process, the terminal 300 switches its connection from the RU 200a connected to the GPU 140 to the RU 200b connected to the CPU 150, that is, performs a handover.

[0034] The storage unit 115 stores information received by the input unit 111 and information processed by the GPU 140 or the CPU 150. The storage unit 115 can hold a table linking the information received by the input unit 111 with the information processed by the GPU 140 or the CPU 150.

[0035] The output unit 116 can output information processed by the GPU 140 or the CPU 150 to the user's terminal 300 .

[0036] According to the base station 100 of the embodiment, unnecessary power consumption is suppressed by always specifying the GPU as the offload destination, and by determining whether to allocate to the GPU depending on whether the communication uses AIMEC, power consumption is suppressed without always being allocated to the GPU.

[0037] The communication method and communication program of the embodiment will now be described with reference to the flowchart of Fig. 3. Fig. 3 is a first flowchart showing an example of the operation of the base station of the embodiment, which is a processing allocation operation. The communication method of the embodiment is executed by the computer (control unit 113) of the base station 100 of the embodiment based on the communication program (see Fig. 3). The communication program of the embodiment causes the computer of the base station 100 to realize a distribution function and a processing function. Each function overlaps with the description of the base station 100 of the embodiment described above, so details will be omitted.

[0038] 3 , first, in step S201, the control unit 113 determines whether to execute AIMEC using information from the terminal 300. If the result of step S201 is YES, that is, if AIMEC is to be executed using information from the terminal 300, the allocating unit 114 allocates the processing of the L1 layer of the corresponding terminal 300a to the GPU 140 (step S203). Even if the result of step S201 is YES, the control unit 113 may make the determination of step S202 and then determine whether to branch the processing to step S203 or step S204. If the result of step S201 is NO, that is, if AIMEC is not to be executed using information from the terminal 300, the control unit 113 proceeds to step S202, where the control unit 113 determines whether the communication volume (traffic volume) with the terminal 300 exceeds a threshold T.

[0039] If the answer is YES in step S202, that is, if the communication volume (traffic volume) with the terminal 300 exceeds the threshold T, the process proceeds to step S203, where the allocator 114 allocates the L1 layer processing of the corresponding terminal 300a to the GPU 140. If the answer is NO in step S202, that is, if the communication volume (traffic volume) with the terminal 300 does not exceed the threshold T, the allocator 114 allocates the L1 layer processing of the corresponding terminal 300b to the CPU 150 (step S204).

[0040] Next, the flow of a communication method based on the utilization rate of the GPU 140 will be described with reference to FIG. 4. FIG. 4 is a second flowchart showing an example of the operation of the base station according to the embodiment, illustrating the threshold change operation. First, the control unit 113 identifies the utilization rate of the GPU 140 (step S301). Next, the control unit 113 determines whether the utilization rate of the GPU 140 is equal to or greater than a threshold U1 (step S302).

[0041] If step S302 returns YES, i.e., if the utilization rate of GPU 140 is equal to or greater than threshold U1, control unit 113 increases threshold T (step S303). If step S302 returns NO, i.e., if the utilization rate of GPU 140 is not equal to or greater than threshold U1, processing proceeds to step S304, where control unit 113 determines whether the utilization rate of GPU 140 is equal to or less than threshold U2. If the utilization rate of GPU 140 is equal to or less than threshold U2, control unit 113 decreases threshold T (step S305). In addition, the communication method may include various other necessary steps not shown.

[0042] The distribution function distributes L1 layer processing to the first processor and the second processor, and distributes L1 layer processing with the terminal to the second processor when AIMEC is used in communication with the terminal, and distributes L1 layer processing with the terminal to the first processor when AIMEC is not used in communication with the terminal (distribution step).The processing function causes the first processor and the second processor to execute the L1 layer processing distributed by the distribution function (processing step).

[0043] According to each aspect of the present disclosure described above, by always specifying the GPU as the offload destination, the possibility of unnecessary power consumption is reduced, and by determining whether to allocate to the GPU depending on whether the communication uses AIMEC, power consumption can be reduced without always being allocated to the GPU, thereby contributing to the achievement of Goal 9 of the Sustainable Development Goals (SDGs), which is to "Develop resilient infrastructure, promote sustainable industrialization and foster innovation."

[0044] The communication program of the embodiment can be implemented using, for example, a scripting language such as ActionScript, JavaScript (registered trademark), Python, or Ruby, or a compiler language such as C, C++, C#, Objective-C, Swift, or Java (registered trademark).

[0045] [Regarding Functions and Circuits] Next, the functions and circuits of the base station 100 described above will be described. Each unit of the base station 100 may be realized as a function of a computer's arithmetic processing unit or the like. That is, the input unit 111, communication unit 112, control unit 113, allocation unit 114, storage unit 115, and output unit 116 of the base station 100 may be realized as an input function, a communication function, a control function, an allocation function, a storage function, and an output function, respectively, by a computer's arithmetic processing unit or the like. A communication program can cause a computer to realize each of the above-mentioned functions. The communication program may be recorded on a computer-readable non-transitory storage medium, such as a memory, a solid-state drive, a hard disk drive, or an optical disk. The storage medium may also be referred to as a non-transitory computer-readable medium that stores the communication program. The communication program may also be transmitted online. As described above, each unit of the base station 100 may be realized by a computer's arithmetic processing unit or the like. The arithmetic processing unit or the like is configured, for example, by an integrated circuit or the like. Therefore, each unit of the base station 100 may be realized as a circuit constituting the arithmetic processing unit or the like. That is, the input unit 111, the communication unit 112, the control unit 113, the allocation unit 114, the storage unit 115, and the output unit 116 of the base station 100 may be realized as an input circuit, a communication circuit, a control circuit, an allocation circuit, a storage circuit, and an output circuit that constitute a computer's arithmetic processing unit or the like. Furthermore, the input unit 111, the communication unit 112, the control unit 113, the allocation unit 114, the storage unit 115, and the output unit 116 of the base station 100 may be realized, for example, as an input function, a communication function, a control function, an allocation function, a storage function, and an output function that include the functions of a arithmetic processing unit or the like. Furthermore, the input unit 111, the communication unit 112, the control unit 113, the allocation unit 114, the storage unit 115, and the output unit 116 of the base station 100 may be realized, for example, as an input circuit, a communication circuit, a control circuit, an allocation circuit, a storage circuit, and an output circuit by being configured using an integrated circuit or the like. Furthermore, the input unit 111, communication unit 112, control unit 113, distribution unit 114, memory unit 115 and output unit 116 of base station 100 may be configured, for example, as an input device, communication device, control device, distribution device, memory device and output device by being composed of multiple devices.

[0046] The base station 100 can combine one or any combination of the above-described units. In the present disclosure, the term "information" is used, but the term "information" can be replaced with "data," and the term "data" can be replaced with "information."

[0047] [Aspects and Effects of the Present Embodiment] Next, one aspect of the present embodiment and the effects of each aspect will be described. Note that each aspect described below is an example at the time of filing, and the present embodiment is not limited to the aspects described below. In other words, the present embodiment is not limited to each aspect described below, and may be realized by appropriately combining each of the above-mentioned parts. Furthermore, a lower aspect may in some cases cite any of the higher aspects. Furthermore, the effects of the present embodiment described below are only examples, and the effects of each aspect are not limited to those described below. Furthermore, each aspect may, for example, achieve at least one of the effects described below.

[0048] (Aspect 1) A base station of one aspect is a base station that relays communications between terminals, and includes a first processor, a second processor that has higher processing performance and higher power consumption than the first processor, and an allocating unit that allocates L1 layer processing to the first processor and the second processor, and that allocates L1 layer processing with a terminal to the second processor when AIMEC (Artificial Intelligence Multi-access Edge Computing) is used in communications with the terminal, and allocates L1 layer processing with the terminal to the first processor when AIMEC is not used in communications with the terminal, and the first processor and the second processor execute the L1 layer processing allocated by the allocating unit. This allows the base station to always designate the GPU as the offload destination, thereby reducing the possibility of unnecessary power consumption, and by determining whether to allocate power to the GPU depending on whether the communication uses AIMEC, power consumption can be reduced without always allocating power to the GPU.

[0049] (Aspect 2) In a base station according to one aspect, the allocating unit allocates L1 layer processing with a terminal to the second processor when the traffic volume in communication with the terminal is equal to or greater than a predetermined threshold, and allocates L1 layer processing with the terminal to the first processor when the traffic volume in communication with the terminal is less than the predetermined threshold. This allows the base station to determine whether to allocate L1 layer processing to the GPU based on the traffic volume, thereby avoiding always allocating to the GPU and reducing power consumption.

[0050] (Aspect 3) In the base station of one aspect, the allocating unit may increase the predetermined threshold when the utilization rate of the second processor is relatively high, and may decrease the predetermined threshold when the utilization rate of the second processor is relatively low. In this way, the base station determines whether to allocate to the GPU based on the utilization rate of the GPU, and does not always allocate to the GPU, thereby making it possible to reduce power consumption.

[0051] (Aspect 4) In a base station according to one aspect, the first processor may be a CPU, the second processor may be a GPU, and the first processor and the second processor may form a vRAN. This allows the base station to allocate processing to the GPU and the CPU in a balanced manner.

[0052] (Aspect 5) A base station according to one aspect may include a first communication unit corresponding to the first processor and executing communication with a terminal under control of the first processor, and a second communication unit corresponding to the second processor and executing communication with the terminal under control of the second processor, thereby enabling the base station to allocate processing to the GPU and the CPU in a balanced manner.

[0053] (Aspect 6) In one aspect of the communication method, a computer of a base station that relays communications between terminals includes a first processor and a second processor having higher processing performance and higher power consumption than the first processor, and distributes L1 layer processing to the first processor and the second processor, and executes the following steps: a distribution step of distributing the L1 layer processing with the terminal to the second processor when AIMEC is used in communications with the terminal, and distributing the L1 layer processing with the terminal to the first processor when AIMEC is not used in communications with the terminal; and a processing step of causing the first processor and the second processor to execute the L1 layer processing distributed in the distribution step. This allows the communication method to achieve the same effects as the base station of the above-mentioned aspect.

[0054] (Aspect 7) A communication program according to one aspect includes a first processor and a second processor having higher processing performance and higher power consumption than the first processor, and is configured to allocate L1 layer processing to the first processor and the second processor in a computer of a base station that relays communications between terminals, the program realizing a distribution function that distributes L1 layer processing with the terminal to the second processor when AIMEC is used in communications with the terminal, and distributes L1 layer processing with the terminal to the first processor when AIMEC is not used in communications with the terminal, and a processing function that causes the first processor and the second processor to execute the L1 layer processing allocated by the distribution function. This allows the communication program to achieve the same effects as the communication device of the above aspect.

[0055] REFERENCE SIGNS LIST 100 Base station 110 CU 111 Input unit 112 Communication unit 113 Control unit 114 Allocation unit 115 Storage unit 116 Output unit 120, 130 DU 140 GPU 150 CPU 200 RU 300 Terminal

Claims

1. A base station that relays communications between terminals, comprising: a first processor; a second processor that has higher processing performance and higher power consumption than the first processor; and an allocating unit that allocates L1 layer processing to the first processor and the second processor, and that allocates L1 layer processing with a terminal to the second processor when AIMEC (Artificial Intelligence Multi-access Edge Computing) is used in communications with the terminal, and allocates L1 layer processing with the terminal to the first processor when AIMEC is not used in communications with the terminal; and the first processor and the second processor execute the L1 layer processing allocated by the allocating unit.

2. The base station according to claim 1, characterized in that the allocation unit allocates L1 layer processing with the terminal to the second processor when the traffic volume in communication with the terminal is equal to or greater than a predetermined threshold, and allocates L1 layer processing with the terminal to the first processor when the traffic volume in communication with the terminal is less than the predetermined threshold.

3. The base station according to claim 2, characterized in that the allocating unit increases the predetermined threshold when the utilization rate of the second processor is relatively high, and decreases the predetermined threshold when the utilization rate of the second processor is relatively low.

4. The base station according to claim 1, wherein the first processor is a CPU and the second processor is a GPU, and the first processor and the second processor form a vRAN.

5. The base station according to claim 1, comprising: a first communication unit corresponding to the first processor and performing communication with the terminal under the control of the first processor; and a second communication unit corresponding to the second processor and performing communication with the terminal under the control of the second processor.

6. A communications method comprising: a base station computer that relays communications between terminals, the base station computer comprising a first processor and a second processor having higher processing performance and higher power consumption than the first processor, and that allocates L1 layer processing to the first processor and the second processor, the base station computer comprising: a distributing step that allocates L1 layer processing with the terminal to the second processor when AIMEC is used in communications with the terminal, and allocates L1 layer processing with the terminal to the first processor when AIMEC is not used in communications with the terminal; and a processing step that the first processor and the second processor execute the L1 layer processing allocated in the distributing step.

7. A communications program for a base station computer that relays communications between terminals, the base station computer having a first processor and a second processor that has higher processing performance and consumes more power than the first processor, and that distributes L1 layer processing to the first processor and the second processor, the distribution function distributing L1 layer processing with the terminal to the second processor when AIMEC is used in communications with the terminal, and distributing L1 layer processing with the terminal to the first processor when AIMEC is not used in communications with the terminal; and a processing function that causes the first processor and the second processor to execute the L1 layer processing distributed by the distribution function.

Images