KR20190131160A - Parallel Scheduler for Improving Radio Link Utilization of RAN Slicing and Method Thereof - Google Patents

Abstract

According to the present invention, provided is a parallel scheduling method of a scheduler, which comprises a step of obtaining a plurality of radio resource requests, allocating a plurality of radio resources to a plurality of radio slices based on the plurality of radio resource requests, determining a priority of at least one unprocessed radio resource request among the plurality of radio resource requests, allocating at least one unallocated radio resource among the plurality of radio resources to at least one slice which has transmitted the at least one unprocessed radio resource request among the plurality of slices according to the priority. Accordingly, a low latency service can be provided to a user even if the number of slices in a radio access network (RAN) is increased, and at the same time, the efficiency of managing a radio link can be increased.

Description

Parallel Scheduler for Improving Radio Link Efficiency of Radio Access Network (RAN) Slicing and Method Thereof

The present invention relates to a parallel scheduler for improving radio link efficiency of radio access network (RAN) slicing and a method thereof, and more particularly to a parallel scheduler for scheduling radio access control in a 5G radio access network environment. will be.

In an LTE communication system, a MAC (Media Access Control) scheduler performs scheduling based on a buffer state of a user equipment (UE). As illustrated in FIG. 1, the MAC scheduler may allocate a radio resource specified as a channel of Orthogonal Frequency Division Multiplexing (OFDM) to each terminal when performing scheduling. In the LTE communication system, since the MAC scheduler performs scheduling for all radio resources including available radio resources, the MAC scheduler can manage the entire radio link as one.

Unlike the LTE communication system described above, in the 5G communication system, radio resources are independently used in the form of multiple slices through RAN (Radio Access Network) slicing, and low latency services are provided to the user through each slice. Can support In such a 5G communication system, since several independent schedulers independently perform scheduling for each slice, scheduling efficiency for the entire radio link may be reduced. As shown in FIG. 2, when multiple independent schedulers independently process each slice independently, scheduling time can be shortened to provide a low latency service, while resource sharing between slices becomes impossible. The efficiency of the management becomes low.

On the contrary, when scheduling the entire wireless link using a single scheduler, the independent scheduling of multiple slices and multiple terminals can be performed in parallel, similarly to the scheduling scheme used in the existing LTE communication system. As a result, the time required for overall scheduling is increased, and thus it is difficult to support a low latency service to a user. As illustrated in FIG. 3, when the scheduling of all slices in the radio resource is processed in a batch, the efficiency of resource management of the radio link is increased, but the overall scheduling time is long, and thus low latency services cannot be provided.

An object of the present invention is to provide an apparatus and method for scheduling RAN slicing that can improve the efficiency of a radio link while providing a low latency service through scheduling of a plurality of RAN slices in a 5G system. To provide.

A parallel scheduling method of a scheduler according to an embodiment of the present invention includes: obtaining a plurality of radio resource requests; Allocating a plurality of radio resources to a plurality of slices based on the plurality of radio resource requests; Determining a priority of at least one unprocessed radio resource request among the plurality of radio resource requests; And assigning at least one unallocated radio resource among the plurality of radio resources to at least one slice in which the unprocessed at least one radio resource request is transmitted among the plurality of slices according to the priority. can do.

According to an embodiment of the present invention, a radio access network (RAN) consisting of several slices can be controlled by a scheduler (or device) having one parallel core, thereby increasing the number of slices of the radio access network. In spite of this, it is possible to provide low latency service to the user regardless of this, and at the same time, improve the efficiency of managing the wireless link.

1 illustrates an LTE communication system according to the prior art.
2 shows a 5G communication system according to the prior art.
3 shows a 5G communication system according to the prior art.
4 illustrates a radio access network (RAN) in accordance with an embodiment of the invention.
5 illustrates a scheduler in accordance with an embodiment of the present invention.
6 illustrates a scheduling method according to an embodiment of the present invention.
7 illustrates a scheduler according to an embodiment of the present invention, and FIG. 8 illustrates a scheduling method of a scheduler according to an embodiment of the present invention.

DETAILED DESCRIPTION Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art may easily implement the present invention. As those skilled in the art would realize, the described embodiments may be modified in various different ways, all without departing from the spirit or scope of the present invention. In the drawings, parts irrelevant to the description are omitted in order to clearly describe the present invention, and like reference numerals designate like parts throughout the specification.

Throughout the specification, a terminal may be a mobile terminal (MT), a mobile station (MS), an advanced mobile station (AMS), a high reliability mobile station (HR-MS). ), A subscriber station (SS), a portable subscriber station (PSS), an access terminal (AT), a user equipment (UE), or the like. It may include all or part of the functionality of AMS, HR-MS, SS, PSS, AT, UE, and the like.

In addition, a base station (BS) may be an advanced base station (ABS), a high reliability base station (HR-BS), a node B (node B), an advanced node B (evolved node B, eNodeB), access point (AP), radio access station (RAS), base transceiver station (base transceiver station (BTS), mobile multihop relay (BSR) -BS, relay that performs the role of a base station (relay) station, RS), a high reliability relay (HR-RS) serving as a base station, etc., and may also be referred to as BS, ABS, Node B, eNodeB, AP, RAS, BTS, MMR-BS, RS It may also include all or part of the function, such as HR-RS.

4 illustrates a radio access network (RAN) in accordance with an embodiment of the invention.

As shown in FIG. 4, the radio access network 400 includes a plurality of RAN slices 421, 422, and 423, a plurality of core slices 431, 432, and 433, and a plurality of radio access network slices. It may include a plurality of terminals (410-1, 410-2, 410-3, 410-N) and the scheduler 440 connected to the wireless access network slice (421, 422, 423).

Each of the radio access network slices 421, 422, and 423 and each of the core slices 431, 432, and 433 may be sliced and operated independently.

Each terminal 410-1, 410-2, 410-3, and 410 -N may be connected to each radio access network slice 421, 422, and 423 for each session. That is, each terminal (eg, the third terminal 410-3) may transmit a plurality of radio access network slices (eg, a first radio access network slice 421 and a second radio access network slice through a plurality of sessions). 423) at the same time.

The scheduler 440 may be located at a base station (GNB). The scheduler 440 may schedule packet exchange between the base station and the terminal through the radio access network slices 421, 422, and 423.

5 illustrates a scheduler in accordance with an embodiment of the present invention.

As shown in FIG. 5, the scheduler 540 includes a plurality of parallel schedulers 542-1 and 542-for performing scheduling for each slice (eg, the radio access network slices 421, 422, and 423 of FIG. 4). 2, 542-3, 542-4, 542-5, 542- (N-1), 542-N and each parallel scheduler 542-1, 542-2, 542-3, 542-4, 542- 5, 542- (N-1) and 542-N may include a scheduling coordinator (Scheduling Coordinator) (541).

Each parallel scheduler 542-1, 542-2, 542-3, 542-4, 542-5, 542- (N-1), 542-N may perform scheduling for radio resources allocated to each slice. Can be.

Each parallel scheduler 542-1, 542-2, 542-3, 542-4, 542-5, 542-(N-1), and 542-N can be used for each slice required to perform scheduling of radio resources for each slice. Channel state information and channel state information for each terminal may be received and stored from the physical layer (PHY) 553, and scheduling may be performed based on each received channel state information.

The scheduling coordinator 541 receives the buffer status for the downlink (DL, downlink) from the radio link control (RLC) 551, and the uplink (uplink) from the wireless control channel (560). Receive the buffer status for UL, uplink).

The scheduling coordinator 541 may determine a plurality of parallel schedulers 542-1, 542-2, 542-3, 542-4, and 542 of the downlink buffer state and the buffer state corresponding to each downlink and uplink. -5, 542- (N-1), 542-N) to allow each parallel scheduler to perform scheduling.

When scheduling is completed, the scheduling coordinator 541 transmits a scheduling result (UL) for the uplink through at least one terminal through the wireless control channel 560 (eg, the terminal 410-1, FIG. 4). 410-2, 410-3, and 410-N), and the scheduling result (DL) for the downlink may be transmitted to the data plane of the media access control 552.

6 illustrates a scheduling method according to an embodiment of the present invention.

As shown in FIG. 6, in operation S601, the scheduling coordinator 641 may receive an uplink buffer state and a downlink buffer state.

In operation S603, the scheduling coordinator 641 may check a request for a radio resource for each slice from the uplink buffer state and the downlink buffer state.

In operation S605, the scheduling coordinator may transmit a radio resource request for each slice to each of the parallel schedulers C1, C2, C3, and CN 641-1, 641-2, 641-3, and 641 -N.

In operations S607-1 to S607-N, each of the parallel schedulers 641-1, 641-2, 641-3, and 641-N allocates radio resources allocated to each slice based on radio resource requests for each slice. Parallel scheduling may be performed for each slice, and the scheduling result may be transmitted to the scheduling coordinator 641.

In operation S609, the scheduling coordinator 641 confirms the unprocessed radio resource request from the scheduling result, and performs optimization scheduling for allocating unallocated radio resources of another slice to the slice that has transmitted the unprocessed radio resource request. The above operations S601 to S607 are iterated.

The scheduling coordinator 641 may perform scheduling optimization based on the type of service of the unprocessed radio resource request. For example, the scheduling coordinator 641 determines the priority of the unprocessed radio resource request based on the type of service of the unprocessed radio resource request, and sequentially from the slice that transmitted the radio resource request having the highest priority. May allocate a radio resource for a slice that has requested a low radio resource.

For example, 5G services may be classified into broadband, ultra low latency, and massive connectivity.

First, since the low latency service is a service requiring low delay (high speed service), the scheduling coordinator 641 transmits unprocessed radio resource requests of the low latency service to the unprocessed radio resources of the broadband service and the low latency service. It can process the highest rank among the unprocessed radio resource request of the request and hyperconnection service.

Next, the superconnected service is provided in the low frequency band among the entire bandwidth, and provides data (eg, sensor data) that requires a delay higher than that required by other services. Accordingly, the scheduling coordinator 641 may process the unprocessed radio resource request of the super-connected service at a lower priority than the unprocessed radio resource request of the low latency service.

Next, the broadband service is provided in the bandwidth of the low-latency service and the bandwidth of the hyperlink service, and is lower than the low-latency service and has a higher priority than the super-connect service. Accordingly, the scheduling coordinator 641 may process the unprocessed radio resource request of the broadband service at a higher priority than the unprocessed radio resource request of the superconnected service and at a lower priority than the unprocessed radio resource request of the low latency service.

For example, there are a plurality of radio resources of unallocated N slices (S1, S2,?, SN), radio resource request of unprocessed broadband service, radio resource request of low latency service, radio resource of hyperlink service. If there is a request, the scheduling coordinator 641 first assigns the unallocated radio resource of the unallocated N slices (S1, S2,?, SN) to the highest priority by assigning the radio resource request of the unprocessed low latency service to the broadband service. The radio resource request is assigned to the unallocated radio resource of the unallocated slice remaining after allocating the radio resource request of the low latency service, and the superconnected service if there is an unallocated radio resource of the unallocated slice. May allocate a radio resource request.

In operation S611, the scheduling coordinator 641 may transmit the scheduling result to the downlink medium access control (DL-MAC) 652 and the terminal (UE) 610, respectively.

7 illustrates a scheduler according to an embodiment of the present invention, and FIG. 8 illustrates a scheduling method of a scheduler according to an embodiment of the present invention.

As shown in FIG. 7, a parallel scheduler for improving radio link efficiency of radio access network (RAN) slicing according to an embodiment of the present invention may include a plurality of parallel device core-based Graphic Processing Units (GPUs) or the like. It may be a device having a distributed structure such as a field programmable gate array (FPGA). For example, each of the plurality of parallel device cores c1, c2, c3, and cN included in the GPU may be each slice (Slice 1, Slice) in the scheduler, such as the plurality of parallel schedulers described with reference to FIGS. 4 to 6. 2, Slice 3, Slice N) may be performed for each scheduling, and the host (or Host CPU) included in the GPU may serve as the scheduling coordinator described with reference to FIGS. 4 to 6.

Information for distributed processing at the time of scheduling may be stored in a global memory or a device memory in the scheduler. In the global memory, all information related to scheduling of all slices may be stored. Information required for scheduling of each slice may be stored in the device memory and used exclusively in each device core. Information commonly used by all device cores may be stored in shared memory in the scheduler.

As shown in FIG. 8, Slice 1, Slice 2 and Slice 3 may include a plurality of Radio Resource Blocks.

Each parallel device core may read radio resource requests of R1, R2, and R3 corresponding to Slices 1, 2, and 3 from global memory. Each parallel device core may store each radio resource request of R1, R2, and R3 in device memory.

The scheduling coordinator may store radio resource blocks including unallocated radio resources in a state in which the radio resource request of R1 and the radio resource request of R2 are processed and the radio resource request of R3 is unprocessed in the global memory.

The scheduling coordinator may allocate the unallocated radio resource to the radio resource request of the unprocessed R3.

Although the embodiments of the present invention have been described in detail above, the scope of the present invention is not limited thereto, and various modifications and improvements of those skilled in the art using the basic concepts of the present invention defined in the following claims are also provided. It belongs to the scope of rights.

Claims (1)

In the parallel scheduling method of the scheduler,
Obtaining a plurality of radio resource requests;
Allocating a plurality of radio resources to a plurality of slices based on the plurality of radio resource requests;
Determining a priority of at least one unprocessed radio resource request among the plurality of radio resource requests; And
Allocating at least one unallocated radio resource among the plurality of radio resources to at least one slice having transmitted the unprocessed at least one radio resource request among the plurality of slices according to the priority;
Way.

Images