1 Extended Reality over 3GPP 5G-Advanced New Radio Link Adaptation Enhancements

2025-04-30 0 0 718.82KB 27 页 10玖币

侵权投诉

Extended Reality over 3GPP 5G-Advanced

New Radio: Link Adaptation Enhancements

Pouria Paymard, Abolfazl Amiri, Troels E. Kolding, and Klaus I. Pedersen

Abstract

One of the rapidly emerging services for ﬁfth-generation (5G)-Advanced is eXtended Reality

(XR) which combines several immersive experiences and cloud gaming services. Those services are

demanding as they call for relatively high data rates under tight latency constraints, sometimes also

referred to as dependable real-time applications. Supporting as many XR users per cell requires highly

efﬁcient radio solutions. In this paper, we propose an enhanced channel quality indicator (CQI) that

results in a better link adaptation to unleash the full performance potential of code block group (CBG)

based transmissions for XR cases. We present both an analytical analysis of the related problems and

solutions, as well as an extensive dynamic system-level performance assessment in line with the 3rd

generation partnership project (3GPP)-deﬁned advanced simulation methodologies. Our results show

an increased XR system capacity of 17% to 33% as compared to what can be supported by current

5G systems with baseline CQI schemes. We also present enhanced CQI complexity-reducing techniques

based on derived closed-form expressions that are attractive to the user equipment (UE) implementation.

I. INTRODUCTION

Standardization of the ﬁfth generation (5G) cellular, known as 5G New Radio (NR), by the

3rd generation partnership project (3GPP) was ﬁrst realized by Release-15. Currently, 3GPP

is working on Release-18 for 5G-Advanced, which introduces several key enhancements and

support for services, adopting research ﬁndings from both industry and academia. One of the

hot topics considered is extended reality (XR) [1]–[5]. XR is an umbrella term for three popular

P. Paymard and K.I. Pedersen are with the Department of Electronic Systems, Technical Faculty of IT and Design; Aalborg

University, Denmark; E-mail: pouriap@es.aau.dk

A.Amiri, T. E. Kolding and K. I. Pedersen are with Nokia, Aalborg, Denmark; E-mail: abol-

fazl.amiri@nokia.com,{klaus.pedersen, troels.kolding}@nokia-bell-labs.com

arXiv:2210.14578v1 [eess.SP] 26 Oct 2022

immersive applications i.e., virtual reality (VR), augmented reality (AR), and mixed reality

(MR) [1]. 5G-Advanced aims an offering superior XR experience in public, consumer, and

industrial sectors. XR services are characterized by strict quality of service (QoS) requirements.

As an example, typical XR services call for data rates from 30 Mbps to 60 Mbps with latency

constraints from 5 ms to 15 ms for the radio access network part and reliability targets on the

order of 99% to 99.999% [1].

In fulﬁlling these requirements for as many XR users as possible per cell, packet scheduling

and link adaptation (LA) play a crucial role in assigning the right amount of radio resources and

modulation schemes to users to fulﬁll the XR service requirements. LA assigns the modulation

and the coding rate of the error correction scheme based on the quality of the radio link. Two

fundamental components of the LA are the channel quality indicator (CQI) and outer loop LA

(OLLA) which work hand in hand to assess the quality of the radio link and assign proper

resources to satisfy a certain utility function [6]–[10].

5G NR supports code block group (CBG)-based transmissions and associated hybrid automatic

repeat request (HARQ) operation which divides a transport block (TB) into smaller groups of

code blocks (CBs) to maximize the radio resource usage efﬁciency in HARQ retransmissions.

Basically, each CB has a cyclic redundancy check (CRC). If one or more CBs in a CBG get

failed, the erroneous CBG should be retransmitted as HARQ acknowledgment (ACK) or negative-

acknowledgment (NACK) bits are provided per CBG in a TB. CBG-based transmissions are

essentially aligned with the transmission of huge payloads such as XR cases. This can effectively

reduce the retransmission payload size and, accordingly, improve the resource efﬁciency [11]–

[15].

A. State of the Art

In the recent 3GPP Release-17 study item on XR over NR [1], the basic modeling and

system-level evaluation methodologies for XR were agreed, including the deﬁnition of several

XR-speciﬁc performance indicators (KPI). This study item concluded that the current 5G network

can support XR services, but also identiﬁed several directions of possible enhancements to further

boot the XR capacity. A general overview of 3GPP XR research is also available in [2]. In 3GPP

5G-Advanced Release 18, 3GPP is now pursuing the introduction of further XR enhancements

as outlined in [3]. Among others, it includes capacity where enhanced LA and enhanced CQI

(eCQI) are under study.

The open literature is rich in LA studies, hence it would be too exhaustive to list here.

In the following, we therefore only summarize a representative subset of those that are most

relevant for this article. As an example, in [6] the authors address CQI enhancements that entail

biased interference ﬁltering of the collected channel quality measurements and so-called Worst-

M CQI reporting formats. The study in [7] proposes a machine learning-based approach to

address the CQI feedback delay problem. However, as will be discussed in greater detail in

this paper, currently known CQI designs are not designed to take full advantage of CBG-based

transmissions for XR use cases. OLLA schemes that operate with traditional HARQ with a

single bit ACK/NACK per transport block are studied in [8]–[10]. The work in [8] introduces

an algorithm to improve the link robustness to the signal-to-interference-plus-noise ratio (SINR)

variability for a 5G centimeter-wave concept. Other OLLA approaches are studied in [9]. The

solution in [10] proposes a self-optimization algorithm to adjust the OLLA initial offset.

There are also several studies of CBG-based transmissions in the literature. As an example, the

authors in [11] , and [12] propose CBG-based multi-bit feedback approaches. The study in [13]

presents new schemes which provide improvements to the current CB/CBG HARQ feedback

scheme. In [14], the authors introduced a novel multi-level CBG-based HARQ scheme, which

also offers reduced overhead of HARQ feedback. In [15], we proposed two new low-complexity

OLLA algorithms tailored to CBG-based transmissions with multi-bit HARQ feedback for the

XR downlink. Contrary to traditional OLLA algorithms for controlling the ﬁrst transport block

transmission error rate of 10%, we aimed at controlling the desired CBG error rate for the

ﬁrst and the second transmissions. Adopting such solutions was shown to help enhance the XR

capacity.

B. Contributions

In this paper, we propose an eCQI scheme for XR use cases with CBG-based transmissions.

Our main contributions are:

•We present an analytical analysis of the CBG-based transmissions to motivate its potential

advantages for the XR use cases and to determine the desired block error rate operating

point. This includes cases where errors on individual CBGs are not always independent and

identically distributed (i.i.d.), but also cases with a correlation of different error levels.

•We present a new eCQI scheme that guides the Next Generation NodeB (gNB) for which

modulation and coding scheme (MCS) to use, subject to the actual CBG-based performance

at the terminal.

•Closed-form expressions for calculating eCQI in the terminal are derived that are attractive

for a low complexity implementation.

•We assess the XR performance under realistic system-level conditions with multiple users,

multiple cells, dynamic XR trafﬁc, and accurate modeling of the major performance-determining

radio access network functionalities in line with [1]. The presented results conﬁrm that there

are promising beneﬁts to gain in terms of higher XR system capacity by adopting CBG-

based HARQ schemes with the proposed eCQI scheme.

C. Structure of the Paper

The rest of this paper is organized as follows. Section II describes the network model and

XR trafﬁc model. Section III states an analytical assessment of the CBG-based HARQ retrans-

missions. In Section IV, we propose the CBG-based. Section V discusses the computational

complexity of enhanced CQI reporting and closed-form expressions to reduce the added com-

plexity. Section VI includes the simulation setup and the performance evaluation. Finally, Section

VII concludes the paper.

II. SETTING THE SCENE

A. Deployment Scenario and Frame Structure

We adopt the agreed 3GPP XR system model and related evaluation methodologies in [1],

assuming a time division duplex (TDD) 5G network with orthogonal frequency-division multiple

access (OFDMA). The considered scenario is the indoor hotspot (InH) deployment with BgNBs.

We focus on the downlink (DL) performance where we assume 100 MHz carrier bandwidth at 4

GHz, 30 kHz sub-carrier spacing (SCS), and physical resource blocks (PRBs) consisting of 12

sub-carriers [16]. Each slot consists of 14 orthogonal frequency-division multiplexing (OFDM)

symbols. A ﬁxed TDD radio frame conﬁguration with a DDDSU slot pattern with a 2.5 ms

repetition period is assumed, where D, S, and U denote Downlink-, Uplink-, and Special-slots.

B. XR-Speciﬁc 3GPP Trafﬁc Model

The assumed downlink XR trafﬁc modeling mimics a case where the XR application server

for view-port rendering or other session techniques is performed. Then, the trafﬁc is delivered to

XR user equipment (UEs) via 5G the radio access network part in close proximity to the users.

𝜇Γ

Frame size

range

Jitter range

𝜎Γ

Frame f

Frame f+1

Time

Frame f-1

𝜆𝑓

𝜎

𝑗

𝑎𝑗𝑏𝑗

𝜇𝑗

𝑏Γ

𝑎Γ

Fig. 1. The downlink XR trafﬁc model according to Release 17 agreements in [1].

XR video frames are periodically generated at the application server based on a ﬁxed rate of λf

frames per second (fps). The arrival time of XR video frames to gNB is quasi-periodic because

of a random jitter that originates from encoding, compressing, routing, etc. The random jitter

is denoted by Jf, which follows a truncated Gaussian distribution Jf∼ T N (µj, σj, aj, bj)with

mean µj, variance σ2

j, and non-zero interval [aj, bj], aj≤bj. The arrival time of video frame f

is expressed as

Tf=f×1

λf

×1000 + Jf.(1)

The XR video frames have variable sizes due to the applied video compression algorithms. The

video frame size is modeled by a truncated Gaussian distribution Γf∼ T N (µΓ, σΓ, aΓ, bΓ)with

mean µΓ, variance σ2

Γ, and non-zero interval [aΓ, bΓ], aΓ≤bΓ. Throughout the rest of the paper,

a video frame simply is denoted as a packet. The assumed XR trafﬁc is illustrated in Fig. 1.

C. Radio Resource Scheduling

Scheduled transmissions are assumed, where each gNB dynamically schedules its users when-

ever there are pending data in its buffer to be transmitted [17]. If there are pending HARQ

retransmissions, those are always prioritized over the transmission of new data. Frequency domain

multiplexing of users on PRB resolution, using the well-known proportional-fair (PF) scheduler

with the following scheduling metric:

u∗= argmax

uψr

u,χ

ψu[t],(2)

文档加载中……请稍候！
如果长时间未打开，您也可以点击刷新试试。

下载文档到电脑，查找使用更方便

10 玖币 0人已下载

立即下载

摘要：

1ExtendedRealityover3GPP5G-AdvancedNewRadio:LinkAdaptationEnhancementsPouriaPaymard,AbolfazlAmiri,TroelsE.Kolding,andKlausI.PedersenAbstractOneoftherapidlyemergingservicesforfth-generation(5G)-AdvancediseXtendedReality(XR)whichcombinesseveralimmersiveexperiencesandcloudgamingservices.Thoseservicesa...

展开>> 收起<<

1 Extended Reality over 3GPP 5G-Advanced New Radio Link Adaptation Enhancements.pdf

共27页,预览5页

还剩页未读，继续阅读

声明：本站为文档C2C交易模式，即用户上传的文档直接被用户下载，本站只是中间服务平台，本站所有文档下载所得的收益归上传人(含作者)所有。玖贝云文库仅提供信息存储空间，仅对用户上传内容的表现方式做保护处理，对上载内容本身不做任何修改或编辑。若文档所含内容侵犯了您的版权或隐私，请立即通知玖贝云文库，我们立即给予删除！

1 Extended Reality over 3GPP 5G-Advanced New Radio Link Adaptation Enhancements

相关推荐

开通VIP享超值会员特权

作者详情

相关内容

热门标签

举报选择: