SUBMITTED FOR REVIEW 1 Joint Optimization of Deployment and Trajectory in UA V and IRS-Assisted IoT Data Collection System

2025-05-02 0 0 3.01MB 11 页 10玖币

侵权投诉

SUBMITTED FOR REVIEW 1

Joint Optimization of Deployment and Trajectory in

UAV and IRS-Assisted IoT Data Collection System

Li Dong, Zhibin Liu, Feibo Jiang and Kezhi Wang.

Abstract—Unmanned aerial vehicles (UAV) can be applied

in many Internet of Things (IoT) systems, e.g., smart farms,

as a data collection platform. However, the UAV-IoT wireless

channels may be occasionally blocked by trees or high-rise

buildings. Intelligent reﬂecting surface (IRS) can be applied to

improve the wireless channel quality by smartly reﬂecting the

signal via a large number of low-cost passive reﬂective elements.

This paper aims to minimize the energy consumption of the

system by jointly optimizing the deployment and trajectory of

the UAV. The problem can be formulated as a mixed-integer-

and-nonlinear-programming (MINLP), which is difﬁcult to be

addressed by the traditional solution, which may be easily fall

into the local optimal. To address this issue, we propose a Joint

Optimization framework of depLoyment and Trajectory (JOLT),

where an adaptive whale optimization algorithm (AWOA) is

applied to optimize the deployment of the UAV, and an elastic

ring self-organizing map (ERSOM) is introduced to optimize the

trajectory of the UAV. Speciﬁcally, in AWOA, a variable-length

population strategy is applied to ﬁnd the optimal number of

stop points, and a nonlinear parameter aand a partial mutation

rule are introduced to balance the exploration and exploitation.

In ERSOM, a competitive neural network is also introduced to

learn the trajectory of the UAV by competitive learning, and a

ring structure is presented to avoid the trajectory intersection.

Extensive experiments are carried out to show the effectiveness

of the proposed JOLT framework.

Index Terms—Deployment optimization; trajectory optimiza-

tion; UAV; IRS; adaptive whale optimization algorithm; elastic

ring self-organizing map

I. INTRODUCTION

Unmanned aerial vehicles (UAV) can be applied in many

Internet of Things (IoT) applications, e.g., smart farms [1],

as a data collection platform, due to its feature of ﬂexibility

and easy to be deployed. Additionally, as the UAV can move

close to the IoT devices in the real environment, it can help

reduce the energy consumption of IoT devices. However,

UAVs usually have stringent constraints of size, weight, and

energy, which may limit their ﬂight distance and time [2].

This work was supported in part by the National Natural Science Foundation

of China under Grant nos. 41904127, 41604117, 62002115. in part by the

Hunan Provincial Natural Science Foundation of China under Grant nos.

2020JJ4428, 2020JJ5105. in part by the Key Research and Development Plan

of Hunan Province under Grant no 2021NK2020. (Corresponding author:

Zhibin Liu)

Li Dong (Dlj2017@hunnu.edu.cn) is with Changsha Social Laboratory

of Artiﬁcial Intelligence, Hunan University of Technology and Business,

Changsha, China, Zhibin Liu (lzb2000@hunnu.edu.cn) is with Hunan Provin-

cial Key Laboratory of Intelligent Computing and Language Informa-

tion Processing, Hunan Normal University, Changsha, China, Feibo Jiang

(jiangfb@hunnu.edu.cn) is with Hunan Provincial Key Laboratory of In-

telligent Computing and Language Information Processing, Hunan Normal

University, Changsha, China, Kezhi Wang (kezhi.wang@northumbria.ac.uk)

is with the department of Computer and Information Sciences, Northumbria

University

Moreover, the line-of-sight (LoS) communication links may

be occasionally blocked by some obstacles, e.g., buildings or

trees. To address the above-mentioned issues and improve the

operation efﬁciency of the UAV system, intelligent reﬂecting

surfaces (IRS) can be applied as a promising solution [3] to

help reﬂect and enhance the communication signal between

UAV and the IoT devices. IRS is composed of a number of

reﬂective elements, which can reﬂect the signal by adjusting

their phase shift. IRS can be mounted on several places such as

the walls/facades of buildings, which can signiﬁcantly improve

the quality of the communication links.

Based on the above background, we aim to optimize the

UAV’s deployment and trajectory by minimizing the energy

consumption of the whole system including the UAV and IoT

devices. To achieve this goal, we propose a Joint Optimiza-

tion framework of depLoyment and Trajectory (JOLT) which

consists of an adaptive whale optimization algorithm (AWOA)

and an elastic ring self-organizing map (ERSOM). The main

contributions can be summarized as follows:

(1) The UAV and IRS-assisted IoT data collection system

is proposed, where the UAV is introduced to collect the data

and the IRS is applied to enhance the communication links

between the UAV and the IoT devices. We formulate the

optimization problem to minimize the energy consumption of

the UAV and all the IoT devices by jointly optimizing the

deployment and trajectory of the UAV.

(2) Then, the joint optimization framework named JOLT

is proposed to solve the optimization problem efﬁciently, in

which AWOA is presented to ﬁnd the optimal deployment of

the UAV, and ERSOM is applied to optimize the trajectory of

the UAV.

(3) For the deployment design of the UAV, the optimal

number of stop points is unknown and the problem is non-

convex. Hence, a variable-length population strategy in the

AWOA is presented to ﬁnd the optimal number of stop points,

and a nonlinear parameter aand a partial mutation rule are

introduced to balance the exploration and exploitation of the

AWOA for searching the locations of the stop points.

(4) For the trajectory planning of the UAV, ERSOM is

applied as a competitive neural network which can learn the

trajectory of the UAV by competitive learning between the

neurons. We also introduce a ring structure in the ERSOM to

avoid the trajectory intersection of the UAV.

The rest of our work is organized as follows. Section II

surveys the related studies. The system model and problem

formulation are introduced in Section III. Section IV describes

the proposed JOLT framework. The simulation results and dis-

cussions are given in Section V. Finally, Section VI concludes

arXiv:2210.15203v1 [cs.NE] 27 Oct 2022

SUBMITTED FOR REVIEW 2

the paper.

II. RELATED WORKS

1) UAV deployment optimization: Wang et al. [4] optimized

the location and number of UAVs through differential evolu-

tion algorithm with an elimination operator. Each individual

in the population is represented as the location of UAV, and all

population is represented as the deployment of UAV. Liu et al.

[5] designed a genetic algorithm to optimize the deployment of

UAVs. In the experiment, their algorithm was compared with

the exhaustion search and the results showed that the proposed

method could ﬁnd a better solution with less computation.

Reina et al. [6] designed an effective method to solve the

multi-target coverage problem in the deployment of UAV, in

which the optimization problem is based on the weighted

ﬁtness function.

2) UAV trajectory optimization: Yu et al. [7] presented a

new differential evolution algorithm for trajectory optimiza-

tion, in which the choice of individuals is depended on the

objective function and constraints. Li et al. [8] proposed a

new trajectory optimization algorithm called MACO, which

can decrease the probability of falling into the local optimum.

Qu et al. [9] applied a reinforcement learning based gray

wolf optimization algorithm to solve the path planning of

UAV. Shao et al. [10] proposed an improved particle swarm

optimization algorithm to optimize the trajectory of UAV,

in which the initial distribution of particles was improved

by chaotic map. Yang et al. [11] introduced an optimal

control strategy of winner-take-all model for target tracking

and cooperative competition of multi-UAVs. Furthermore, Zuo

et al. [12] summarized the ﬂight control methods and future

challenges of UAVs.

3) IRS-assisted UAV system: Jiao et al. [13] designed

an IRS and UAV assisted multiple-input NOMA downlink

network. Al-Jarrah et al. [14] analyzed the communication ca-

pacity of the IRS assisted UAV system, which was inﬂuenced

by the imperfect phase information. Pan et al. [15] applied

UAV and IRS to support terahertz (THz) communications by

optimizing trajectory of UAV, phase shift of IRS, terahertz

subband allocation and power control jointly. You et al. [16]

presented the promising application scenarios, issues, and

potential solutions of jointly applying IRS and UAV in wireless

networks.

However, the above works have not jointly optimized the

deployment and trajectory of the UAV and the phase-shift

matrix of the IRS to reduce the energy consumption of the

UAV and all the IoT devices. Moreover, the computational

complexity of the traditional solutions is high when the

number of stop points is large. Hence, here we aim to design

an efﬁcient JOLT framework to optimize the deployment and

trajectory of the UAV and the phase-shift matrix of the IRS

jointly, where the number of stop points can be reduced and

the computing time of the trajectory planning can be saved.

III. SYSTEM MODEL AND PROBLEM FORMULATION

In Fig. 1, we design an IoT data collection system involving

a UAV, an IRS and many IoT devices, where the set of IoT

Fig. 1. UAV and IRS-assisted IoT data collection system.

devices is N={1,2, . . . , N}. The UAV can collect data

through moving close to the IoT devices. If the LoS link is

blocked, the transmission signals from IoT devices can be

reﬂected and enhanced to the UAV via the IRS, which has

an uniform linear array (ULA) with Mreﬂecting elements.

Also, we assume that there are kstop points and kis a prior

unknown, and we use K={1,2, . . . , K}to represent the set

of the stop points for the UAV.

A. Data transmission model

We assume that the location of the i-th (i∈ N )IoT

device is xD

i, yD

i, which denotes the coordinate of the i-

th IoT device. The UAV ﬂies at a ﬁxed altitude Hand qj=

XU

j, Y U

j, Hrepresents the coordinates of the j-th (j∈ K)

stop point. Moreover, we consider that the coordinate of

the IRS is represented by XI, Y I, HI. Then, the distance

between the j-th stop point and the IRS can be expressed as

dU,I

j=qXU

j−XI2+YU

j−YI2+ (H−HI)2.(1)

Similarly, the distance between the i-th IoT device and the

IRS can be expressed as

dI,D

i=qXI−xD

i2+YI−yD

i2+ (HI)2.(2)

Then, the channel gain between the j-th stop point and the

IRS can be expressed as [17]

hU,I

j=v

α1

dU,I

j2h1, e−j2π

λdφU,I

j, . . . , e −j2π

λ(M−1)dφU,I

j]i

(3)

where α1is the path loss at 1m, the right of Eq. (3) is the array

response for the IRS [18], where φU,I

j=XI−XU

dU,I

represents

the cosine value of the arrival angle from the IRS to the j-th

stop point of the UAV. λis the carrier wavelength, and dis

the antenna separation distance.

Similarly, the channel gain between the IRS and the i-th

IoT device can be expressed as

hI,D

i=v

α2

dI,D

i2h1, e−j2π

λdφI,D

i, . . . , e−j2π

λ(M−1)dφI,D

iiT

(4)

SUBMITTED FOR REVIEW 3

where α2is the path loss exponent, and φI,D

i=XI−xi

dI,D

is the

cosine value of the angle of deviation from the IRS to the i-th

IoT device.

Additionally, we denote θi,m,j ∈[0,2π) (m∈ M)as

the diagonal phase shift matrix. In the matrix, imeans the

i-th IoT device, mmeans the m-th reﬂecting element, and

jmeans the j-th stop point. Hence, the total matrix is

Θi,j = diag eθi,m,j ,∀m∈ M. Through the Eq. (3) and

(4), we can get the transmission rate from the j-th stop point

to the i-th IoT device, which can be represented as

rij =ˆ

Blog2



1 +

pihU,I

jΘi,j hI,D

i

σ2



(5)

where piis the transmit power. ˆ

Bis the system bandwidth.

σ2is the noise power. aij is used to indicate whether the i-th

IoT device sends data to the j-th stop point. To reduce power

consumption, IoT devices should send data to the stop point

with the fastest transmission rate. Therefore, aij is given by

C1 : aij =(1,if j= arg max

j∈K

rij

0,otherwise

.(6)

In addition, each IoT device can only be connected to one

stop point, which can be expressed as

C2 :

j=1

aij = 1,∀i∈ N .(7)

Additionally, considering the limitation of system band-

width, each stop point is connected to no more than ˆ

MIoT

devices, which can be expressed as

C3 :

i=1

aij ≤ˆ

M, ∀j∈ K.(8)

For all data of IoT devices to be received, the condition

listed below is required:

C4 :

i=1

j=1

aij =N. (9)

If the i-th IoT device has Didata sent to the UAV, the

transmission time can be calculated as

Tij =Di

rij

,∀i∈ N , j ∈ K.(10)

Then, the energy consumption of the i-th IoT device can be

calculated by the following equation:

Eij =piTij =piDi

rij

,∀i∈ N , j ∈ K.(11)

Hence, the energy consumption of all IoT devices is given

Eiot =

i=1

j=1

aij Eij .(12)

B. UAV hovering model

The data transmission rate is limited, so the UAV needs to

hover at each stop point to collect all the IoT device data.

Thus, the hover time of the UAV at the j-th stop point can be

given by

j= max

i∈N {aij Tij },∀j∈ K.(13)

Then, the hovering energy consumption of the UAV at the

j-th stop point can be represented as

j=pHTH

j,∀j∈ K (14)

where pHdenotes the hover power of the UAV.

Finally, the hovering energy consumption of the UAV can

be given by

Ehov =

j=1

j.(15)

C. UAV trajectory model

After collecting data from the current stop point, the UAV

will choose the next stop point, which can be denoted as bi

(i, j ∈ K). Moreover, bi

jis equal to 1 if the UAV chooses the

j-th stop points as the i-th point of the trajectory. Otherwise,

jis equal to 0. Since each stop point can be reached once,

which is expressed as

i=1

j= 1,∀j∈ K.(16)

Then, we deﬁne qj(j∈ K)as the location of the j-th point

of the trajectory. Thus, the ﬂight distance is give as

j=1 

qj−qj−1

.(17)

Finally, the ﬂight energy consumption of the UAV can be

expressed as

Efly =pFL(18)

where pFis the ﬂight power of the UAV [19].

D. Objective function and constraints

The energy consumption of the UAV and IRS-assisted IoT

data collection system includes the three items mentioned

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：

SUBMITTEDFORREVIEW1JointOptimizationofDeploymentandTrajectoryinUAVandIRS-AssistedIoTDataCollectionSystemLiDong,ZhibinLiu,FeiboJiangandKezhiWang.AbstractUnmannedaerialvehicles(UAV)canbeappliedinmanyInternetofThings(IoT)systems,e.g.,smartfarms,asadatacollectionplatform.However,theUAV-IoTwirelesschann...

展开>> 收起<<

SUBMITTED FOR REVIEW 1 Joint Optimization of Deployment and Trajectory in UA V and IRS-Assisted IoT Data Collection System.pdf

共11页,预览3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！

SUBMITTED FOR REVIEW 1 Joint Optimization of Deployment and Trajectory in UA V and IRS-Assisted IoT Data Collection System

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: