Analysis of V2X Sidelink Positioning in sub-6 GHz Yu Ge Maximilian Starky Musa Furkan Keskin Frank Hofmanny Thomas Hanseny Henk Wymeersch Chalmers University of Technology SwedenyRobert Bosch GmbH Germany

2025-04-30 0 0 2.53MB 6 页 10玖币

侵权投诉

Analysis of V2X Sidelink Positioning in sub-6 GHz

Yu Ge∗, Maximilian Stark†, Musa Furkan Keskin∗, Frank Hofmann†, Thomas Hansen†, Henk Wymeersch∗

∗Chalmers University of Technology, Sweden †Robert Bosch GmbH, Germany

Abstract—Radio positioning is an important part of joint com-

munication and sensing in beyond 5G communication systems.

Existing works mainly focus on the mmWave bands and under-

utilize the sub-6 GHz bands, even though it is promising for

accurate positioning, especially when the multipath is uncom-

plicated, and meaningful in several important use cases. In this

paper, we analyze V2X sidelink positioning and propose a new

performance bound that can predict the positioning performance

in the presence of severe multipath. Simulation results using ray-

tracing data demonstrate the possibility of sidelink positioning,

and the efﬁcacy of the new performance bound and its relation

with the complexity of the multipath.

Index Terms—Sub-6 GHz, multipath, sidelink positioning,

performance bound, ray-tracing.

I. INTRODUCTION

Sensing and positioning have gained signiﬁcant attention in

the evolution of 5G mobile radio systems [1]. Much of this

attention has been devoted to the mmWave bands (both lower

mmWave around 28 GHz and upper mmWave around 140

GHz), due to larger available bandwidth and commensurate

distance resolution [2], [3]. Nevertheless, sub-6 GHz bands

hold signiﬁcant promise for accurate positioning as well,

especially in scenarios without too complicated multipath. Due

to the presence of the ITS 5850-5925 MHz band and the neigh-

boring unlicensed bands, sub-6 GHz positioning over sidelinks

has recently come into focus, complementary to enhancements

for improved integrity, accuracy, and power efﬁciency [4].

Sidelink positioning is expected to be an important enabler

for several use cases, considering positioning for the user

equipments (UEs) in in-coverage, partial coverage, and out-of-

coverage. These include vehicle-to-anything (V2X), Internet of

Things (IoT) (e.g., in private networks at 3.6–3.7 GHz), and

public safety (e.g., ﬁrst responders) [5].

Within V2X, there have been a number of studies related to

sidelink positioning at sub-6 GHz. In terms of overviews, there

are several recent papers [6]–[11]. In [6], a broad overview of

V2X positioning is provided, highlighting the limitations of

time-difference-of-arrival (TDoA) in terms of synchronization,

and proposes to use carrier phase and multipath information.

In [7], use cases and corresponding requirements are speciﬁed,

showing that beyond 5G systems need enhancements in terms

of deployments, methods, and architectures. In [8], sidelink

positioning is advocated as a positioning enabler with lower

latency, higher line-of-sight (LoS) probability, and improved

coverage, which should be able to operate both collaboratively

with the network and to operate independently when the

network is unavailable. Focusing on indoor IoT scenarios, [9]

studies combinations of different measurements together with

Fig. 1. Illustration of an urban trafﬁc situation involving vulnerable road

users, vehicles, and road-side units.

an extended Kalman ﬁlter (EKF). The relative merits of cellu-

lar vehicle-to-anything (C-V2X) and WiFi-based positioning

are discussed and evaluated in [10], based on the WINNER+

model, indicating a preference of C-V2X. In [11], different

positioning systems/architectures are described for vehicular

positioning applications, along with relevant key performance

indicators (KPIs). There have also been several studies fo-

cusing exclusively on the physical layer [12]–[15]. Among

these, [12] evaluates vehicle-to-infrastructure (V2I) ranging

and TDoA positioning for LTE under different statistical chan-

nel models. In [13], dedicated short-range communications

(DSRC) positioning is studied with ray-tracing data, com-

bining angle-of-arrival (AoA) information at road-side units

(RSUs) and vehicle-to-vehicle (V2V) cooperating ranging

links. The use of several arrays per UE is proposed in [14] and

evaluated in terms of Cram´

er-Rao bound (CRB) at 3.5 and 28

GHz, based on AoA and TDoA measurements. In [15], sensing

with 5G-V2X waveforms is considered, determining the range

and Doppler of passive targets via CRB analysis under a pure

geometric channel, subject to interference. Finally, there are

studies that focus mainly on algorithmic aspects, such as [16]–

[18]. Here, [16] proposes a method for V2X localization with a

single RSU from range measurements over time. In [17], V2V

range and angle and V2I TDoA measurements are combined

to improve positioning. In [18], a dynamic method is proposed

to switch between global navigation satellite system (GNSS)

and NR V2X TDoA measurements. While the above listed

papers adopt widely varying assumption, models, methods,

and evaluation methodologies, it is worth pointing out that

[4] has proposed a common evaluation methodology and a set

arXiv:2210.15534v1 [eess.SP] 27 Oct 2022

of common assumptions.

In this paper, we perform a realistic evaluation of sidelink

V2X round-trip-time (RTT) positioning towards 3GPP Release

18 using ray-tracing data, focusing on operation outside of the

network coverage.

•Use cases: We describe the relevant V2X use cases

towards 3GPP Release 18 and requirements that involve

sidelink positioning, as well as the related physical mod-

els, and limitations thereof.

•Novel performance bound: We propose a novel method-

ology based on Fisher information analysis to predict po-

sitioning performance in the presence of severe multipath,

by accounting for inter-path interference.

•Methods and evaluation: We verify the validity of the

new performance bound through evaluation of RTT-based

ranging using ray-tracing data in two use cases: one with

RSU and one without RSU. The simulations show that

performance is mainly limited due to multipath induced

biases.

The remainder of this paper is organized as follows. The use

cases and system model are introduced in Section II. The

basics of the Fisher information matrix (FIM) and its three

variants are described in Section III. Ranging and range-based

positioning algorithms are presented in Section IV. Simulation

results are displayed in Section V, followed by our conclusions

in Section VI.

II. USE CASES AND SYSTEM MODEL

In this section, we describe the different requirement sets

and a generic system model.

A. Use Cases and Requirements

In [5], three sets of positioning requirements are deﬁned

(both for absolute and relative positioning):

1) Set 1 (low accuracy): This set requires 10–50 m with

68%–95% conﬁdence level, mainly for information pro-

visioning use cases, such as trafﬁc jam warning.

2) Set 2 (moderate accuracy): This set requires 1–3 m with

95%–99% conﬁdence level, mainly to support so-called

day-1 use cases, including lane change warning (V2V),

intersection movement assist (V2V), and automated in-

tersection crossing (V2I) [19].

3) Set 3 (high accuracy): This set requires 0.1-0.5 m with

95%–99% conﬁdence level, to support so-called advanced

use cases, such as automated driving or tele-operated

driving [19].

More detailed requirements can be found in [19, Table 5.1-

1] and [5, Table 7.3.2.2-1], which also describe the nominal

velocity and whether the requirement pertains to absolute or

relative positioning.

To exemplify these use cases, Fig. 1 depicts a dense

trafﬁc situation in an urban environment. Many road users

are trying to cross the intersection. In this situation, V2X

communication helps to make road trafﬁc safer and more

efﬁcient. V2X communication includes the communication

between road users, namely UEs and road infrastructure, i.e.,

RSUs. In Fig. 1, lamp posts are equipped with RSUs. Within

the 3GPP framework [20], however, RSUs are assumed to be

mounted at the middle of the intersection. Originally, NR-V2X

addresses direct communication between road participants to

exchange V2X messages including warnings, information, col-

lective perception etc. Starting with Rel. 18, 3GPP studies the

possibility to perform ranging and positioning over sidelink for

V2X applications. Especially in difﬁcult outdoor environments

where classical positioning techniques, as e.g., GNSS, are

blocked or distorted, sidelink positioning arises as a valuable

complementing positioning technology. As depicted in Fig. 1,

the street canyons can block the GNSS signals, so that GNSS

is considered to be unavailable.

B. System Model

We consider a scenario with several devices, which may be

UEs or RSUs. The state components of device n, comprising

its location (in 2D or 3D) and velocity are denoted by xn

and vn, respectively. For a RSU, the state is known and the

velocity is vn=0. Devices are not synchronized. The main

functionality is the ability to estimate the time-of-arrival (ToA)

of the LoS path. Our focus is on orthogonal frequency-division

multiplexing (OFDM), where we consider a system with Ns

subcarriers with subcarrier spacing ∆f.

We drop device indices when possible, so that the received

signal at a device, based on the transmission by another device

can be expressed as a vector of length Ns:1

yt=

L−1

l=0

αl(stad(τl))e2πtvlTs/λ +nt,(1)

where t∈ {1, . . . , T }is the OFDM symbol index, Lis the

number of multipath components (which are not necessarily

resolvable), αlis the complex channel gain of the l-th path,

stis the vector of pilot signals across the subcarriers of the

t-th OFDM symbol, ad(τl)∈CNsis the delay steering vector,

as a function of the ToA τl, with

[ad(τl)]n= exp(−2πnτl∆f).(2)

In addition, vlis the radial velocity of the l-th path, λis the

wavelength, and Tsis the OFDM symbol duration (including

cyclic preﬁx (CP)). Finally, ntis the additive white Gaussian

noise (AWGN), with nt∼ CN (0, N0I). The average transmit

power is ﬁxed to Ptx, so that E{kstk2}=Ptx/∆f.

Under the assumption that the path index l= 0 correspond

to the LoS path, the parameter τ0depends on the geometry

through

τ0=kxrx −xtxk/c +B(3)

where cis the speed of light, and Bis a clock bias be-

tween the transmitter and the receiver. In contrast to TDoA-

based positioning, where the clock bias is removed by using

1After appropriate receiver-side processing, such as coarse synchronization,

cyclic preﬁx removal, and FFT [21].

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：

AnalysisofV2XSidelinkPositioninginsub-6GHzYuGe,MaximilianStarky,MusaFurkanKeskin,FrankHofmanny,ThomasHanseny,HenkWymeerschChalmersUniversityofTechnology,SwedenyRobertBoschGmbH,GermanyAbstractRadiopositioningisanimportantpartofjointcom-municationandsensinginbeyond5Gcommunicationsystems.Existingw...

展开>> 收起<<

Analysis of V2X Sidelink Positioning in sub-6 GHz Yu Ge Maximilian Starky Musa Furkan Keskin Frank Hofmanny Thomas Hanseny Henk Wymeersch Chalmers University of Technology SwedenyRobert Bosch GmbH Germany.pdf

共6页,预览2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！

Analysis of V2X Sidelink Positioning in sub-6 GHz Yu Ge Maximilian Starky Musa Furkan Keskin Frank Hofmanny Thomas Hanseny Henk Wymeersch Chalmers University of Technology SwedenyRobert Bosch GmbH Germany

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: