Probability-Reduction of Geolocation using Reconﬁgurable Intelligent Surface Reﬂections Anders M. Buvarp1 Daniel J. Jakubisin1 William C. Headley1and Jeffrey H. Reed2

2025-05-02 0 0 600.06KB 6 页 10玖币

侵权投诉

Probability-Reduction of Geolocation using

Reconﬁgurable Intelligent Surface Reﬂections

Anders M. Buvarp1, Daniel J. Jakubisin1, William C. Headley1and Jeffrey H. Reed2

Virginia Tech National Security Institute1and Wireless@VT2

Blacksburg, VA 24061, USA

Email: {abuv,djj,cheadley,reedjh}@vt.edu.

Abstract—With the recent introduction of electromagnetic

meta-surfaces and reconﬁgurable intelligent surfaces, a paradigm

shift is currently taking place in the world of wireless commu-

nications and related industries. These new technologies are of

great interest as we transition from the 5th generation mobile

network (5G-NR) towards the 6th generation mobile system

standard (6G). In this paper, we explore the possibility of using

a reconﬁgurable intelligent surface in order to disrupt the ability

of an unintended receiver to geolocate the source of transmitted

signals in a 5G-NR communication system. We investigate how

the performance of the Multiple Signal Classiﬁcation (MUSIC)

algorithm at the unintended receiver is degraded by correlated

reﬂected signals introduced by a reconﬁgurable intelligent surface

in the wireless channel. We analyze the impact of the direction

of arrival, delay, correlation, and strength of the reconﬁgurable

intelligent surface signal with respect to the line-of-sight path

from the transmitter to the unintended receiver. An effective

method is introduced for defeating direction-ﬁnding efforts using

dual sets of surface reﬂections. This novel method is called

Geolocation-Probability Reduction using dual Reconﬁgurable In-

telligent Surfaces (GPRIS). We also show that the efﬁciency of

this method is highly dependent on the geometry, that is, the

placement of the reconﬁgurable intelligent surface relative to the

unintended receiver and the transmitter.

Index Terms—Reconﬁgurable intelligent surfaces, MUSIC, low

probability of geolocation, direction-ﬁnding, 5G-NR, smart radio

environments.

I. INTRODUCTION

Until recently, designers of wireless communications sys-

tems have considered the radio propagation channel as outside

of the control of communication system design. However,

Reconﬁgurable Intelligent Surfaces (RIS) present the oppor-

tunity to introduce control over the propagation channel itself

[1]. Thus, RIS is a paradigm shift from previous generation

systems that has the potential to play a signiﬁcant role in the

next generation of wireless systems such as 6G. A related and

developing ﬁeld of great interest is Electromagnetic Signal and

Information Theory (ESIT) and Holographic Radios [2].

Ongoing RIS research includes redirecting signals away

from adversarial nodes or redirecting a signal in a desired

direction, and hence reducing the overall interference [1].

In this work, we explore the use of a RIS to disrupt the

direction-ﬁnding (DF) capability of a potentially adversarial

receiver that might want to jam a friendly communication

link. Disruption of DF is a key capability for Low Probability

of Geolocation (LPG) in millimeter wave systems. LPG is

important in military applications where detection, isolation,

and identiﬁcation of signals must be avoided.

In certain scenarios, a transmitter will not be able to fully

avoid radiating signal towards an adversarial node, e.g., due to

the node geometry or transmit array limitations. We propose

using a RIS in order to introduce artiﬁcial correlated multipath

into the environment to prevent geolocation. Fig. 1 shows an

adversarial node attempting to geolocate a transmitter using a

direction-ﬁnding algorithm called Multiple Signal Classiﬁca-

tion (MUSIC) and a RIS that is reﬂecting obfuscating signals

towards the Uniform Linear Array (ULA) of the adversary. The

contribution of this paper is to understand the potential for this

technique to impact DF under certain geometrical constraints

and to motivate future work on practical implementations.

Along these lines, we assume knowledge of the adversarial

node’s location and accurate control of the RIS element

phases.

Fig. 1: A signal transmitted while an adversarial node uses

the MUSIC algorithm for geolocation of the transmitter for

signal detection and jamming. A RIS is employed to reﬂect a

range of correlated multipath signals to defeat the detection.

There are several Direction-of-Arrival (DOA) estimation

methods that could be used for geolocation e.g. Doppler DF

[3], Watson-Watt [4], Correlative Interferometry [5], Time

Difference of Arrival (TDOA) [6], Maximum Likelihood Es-

timation (MLE) [7], MUSIC [8], Estimation of Signal Param-

eters via Rotational Invariance Techniques (ESPRIT) [9], and

Matrix Pencil [10]. MLE, MUSIC and ESPRIT all depend

on estimating the correlation matrix, R, using Ksnapshot

samples of the incoming signal received by an antenna array

with Nelements. The output of these algorithms is the signal

arXiv:2210.09624v2 [eess.SP] 4 Feb 2023

directions, Θi, of the Mincoming signals. The Matrix Pencil

algorithm is similar to ESPRIT, however it is a non-statistical

method.

The advantages of the MUSIC algorithm compared to

classical DOA estimation methods include general sensor con-

ﬁgurations, ultra-slow sampling, and small array dimensions.

MUSIC can be applied to both narrowband and wideband sig-

nals without prior knowledge of the signals [11]. The MUSIC

algorithm, which is closely related to Prony’s method [12], has

been evaluated as having superior high-resolution performance

at the cost of computation and storage [13]. Using peaks of a

spatial pseudo-spectrum, MUSIC is able to identify the angle-

of-arrival of signals from −90 to +90 degrees. For this reason,

we chose to evaluate our DF disruption technique against an

adversarial node equipped with the MUSIC algorithm.

This paper is organized as follows. Section II deﬁnes the

RIS wave propagation model. How to program the surface to

maximize the Signal-to-Noise Ratio (SNR) at the adversarial

node is explained in Section III. In Section IV, we evaluate

the impact of correlated multipath on MUSIC direction-ﬁnding

performance, independent on the source of the multipath. In

Section V the RIS propagation model is applied and perfor-

mance is evaluated as a function of RIS and node geometry.

We end the paper with conclusions in Section VI.

II. RIS WAVE PROPAGATION MODEL

A RIS is a type of electromagnetic metasurface [1]. Typi-

cally, it is a sheet of inexpensive and adaptive thin composite

material, which can cover walls, buildings, ceilings, etc. It

consists of individually tuned passive reﬂectors, making the

phase response of the incoming wave tunable. The individ-

ually controlled unit cell of a RIS incorporates low power

electronic circuits with components such as positive-intrinsic-

negative (PIN) diodes and varactors, where the bias voltage

of the varactor can be tuned. The RIS can be controlled and

programmed using a simple microcontroller such as Raspberry

Pi. The RIS is reconﬁgurable and can be programmed to

control and modify the incident radio waves by elementary

electromagnetic functions such as reﬂection, refraction, ab-

sorption, focusing/beamforming, polarization, splitting, analog

processing and collimation [1]. A RIS can also be used for

joint transmit and passive beamforming design [14].

For the direct Line-of-Sight (LOS) path between the RIS

and the adversarial node, the attenuation and the delay of the

received baseband signal are [15],

A0=λcqGAdv

T x GT x

Adv

4πkpAdv −pT xk, τ0=kpAdv −pT xk

,(1)

respectively, where λcis the wavelength of the carrier wave,

GAdv

T x is the antenna gain of the transmit antenna in the

direction of the adversarial node, GT x

Adv is the antenna gain of

the adversarial node in the direction of the transmitter, pAdv

is the 3-dimensional (3D) position of the adversarial node,

pT x is the 3D position of the transmitter, c0is the speed of

light through air, and k·k the l2-norm. For the RIS path, the

attenuation and delay are,

Ai,j =√µλ2

16π2qGAdv

i,j Gi,j

Adv Gi,j

T x GT x

i,j

kpAdv −pi,j kkpi,j −pT xk,(2)

τi,j =kpi,j −pT xk+kpAdv −pi,j k

,(3)

respectively, where i, j are the element index, GAdv

i,j is the

antenna gain of the RIS element in the direction of the

adversarial node, Gi,j

Adv is the antenna gain of the adversarial

node in the direction of the RIS element, Gi,j

T x is the antenna

gain of the transmitter in the direction of the RIS element,

GT x

i,j is the antenna gain of the RIS element in the direction

of the transmitter, µ∈[0,1] is the fraction of the incident

energy that is scattered, and pi,j is the RIS element location

with x, y, z-coordinates,

x= 0

y=idy−0.5dy((Q+ 1) mod 2) ∀i= [1, Q]

z=jdz−0.5dz((P+ 1) mod 2) ∀j= [1, P ],

(4)

where dyis the element spacing in the y-direction, dzis

the element spacing in the z-direction, Q is the number of

element columns of the surface and P is the number of rows

of elements. The complex baseband signal received at the

adversarial node, y(t), can be expressed as,

y(t) = A0e−j2πfcτ0x(t−τ0)

i,j

Ai,j e−j2πfcτi,j −jφi,j xt−τi,j −φi,j

2πfc

+n(t)

(5)

where τ0is the time delay from the transmitter to the adver-

sarial node, τi,j is the time delay from the transmitter to the

adversarial node via the RIS elements, φi,j

2πfcis a tunable delay

that enables the anomalous scattering of the incident radio

waves and n(t)is additive white Gaussian noise (AWGN).

III. RIS ARRAY PROCESSING

The phases φi,j are programmed by setting the bias voltage

of the varactor of each surface element. Maximizing the

received power of the baseband signal (5) requires that all

terms have the same phase. In order to maximize the SNR at

the ULA, φi,j is programmed as,

φi,j = 2πfc(τ0−τi,j ) mod 2π(6)

We assume that the location of the adversarial node is known.

IV. SIMULATION RESULTS

Without using the complex RIS model shown in Section

II, we ﬁrst analyzed the behavior of the MUSIC algorithm

using orthogonal frequency division multiplexing (OFDM)

signals generated as they arrive on the adversarial node’s ULA.

We looked at the ability of the adversarial node to detect

signals in the MUSIC pseduo-spectrum when two correlated

or two uncorrelated signals arrive at the ULA from different

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：

Probability-ReductionofGeolocationusingRecongurableIntelligentSurfaceReectionsAndersM.Buvarp1,DanielJ.Jakubisin1,WilliamC.Headley1andJeffreyH.Reed2VirginiaTechNationalSecurityInstitute1andWireless@VT2Blacksburg,VA24061,USAEmail:fabuv,djj,cheadley,reedjhg@vt.edu.AbstractWiththerecentintroductionof...

展开>> 收起<<

Probability-Reduction of Geolocation using Reconﬁgurable Intelligent Surface Reﬂections Anders M. Buvarp1 Daniel J. Jakubisin1 William C. Headley1and Jeffrey H. Reed2.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！

Probability-Reduction of Geolocation using Reconﬁgurable Intelligent Surface Reﬂections Anders M. Buvarp1 Daniel J. Jakubisin1 William C. Headley1and Jeffrey H. Reed2

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: