NOMA-BC融合：考虑不完美SIC的多小区频谱效率优化
需积分: 5 31 浏览量
更新于2024-08-05
收藏 183KB PDF 举报
本文档探讨了将回射通信（Backscatter Communication，BC）与多小区非正交多址（Non-Orthogonal Multiple Access，NOMA）集成的潜在优势，特别是在未来无线网络中的高谱效率（Spectral Efficiency，SE）提升和大规模连接能力。传统的NOMA技术以其能够同时服务于多个用户并实现频谱效率优化而受到关注，而BC作为一种新兴的无电池通信方式，通过利用环境中的无线射频（Radio Frequency，RF）信号进行数据传输，对环保和能源效率具有显著优势。
论文的核心贡献在于提出了一种新的优化框架，目标是在考虑不完美的 Successive Interference Cancellation（SIC）解码条件下，最大化NOMA-BC网络的SE。在这个框架中，优化问题涉及到基站的发射功率和每个回射设备的反射系数等多个变量，这使得优化问题变得复杂且难以直接求解。为了应对这个挑战，作者采用了一种分解方法，结合Karush-Kuhn-Tucker (KKT) 条件，这是一种优化理论中的重要工具，用于解决带有约束条件的非线性规划问题。
具体来说，该研究可能包括以下几个关键步骤：
1. **系统模型**：首先，构建一个包含多小区NOMA网络和回射通信模块的数学模型，定义信号传播、干扰、以及SIC解码过程中的误差模型。
2. **问题定义**：明确表示出SE作为函数，包括发射功率、反射系数和用户分配策略，同时考虑到SIC性能对系统性能的影响。
3. **优化分解**：将原问题分解成更小的子问题，以便分别针对每个子问题进行分析和优化，如分别对每个小区的发射功率和回射设备的反射系数进行优化。
4. **KKT条件应用**：利用KKT条件分析系统的局部最优解，这些条件提供了在某些约束下优化问题的必要条件，有助于找到近似最优解。
5. **算法设计**：设计一个迭代算法，如差分进化、梯度下降或基于学习的算法，来解决这些子问题，并逐步逼近全局优化。
6. **性能评估**：通过仿真和实验验证优化框架的有效性和实际性能提升，包括对比与传统NOMA和独立的BC系统的性能，以及对不同参数（如信噪比、SIC性能等）的敏感性分析。
7. **结论与未来工作**：总结优化结果，讨论其在实际网络中的实施可能性，并提出未来的研究方向，比如扩展到更大规模的网络架构或者考虑其他无线资源管理策略。
这篇论文为融合NOMA和BC技术以提升未来无线网络的能效和容量提供了一种创新的优化策略，通过优化策略克服了复杂性并有望改善现有的通信标准。
arXiv:2109.11509v1 [eess.SP] 23 Sep 2021
SUBMITTED TO IEEE 1
Integration of Backscatter Communication with Multi-cell NOMA:
A Spectra l Efficiency Optimization under Imp erfect SIC
Wali Ullah Khan, Eva Lagunas, Asad Ma hmood, Symeon Chatzinotas, and Bj¨orn Ottersten
Abstract—Non-orthogonal multiple access (NOMA) is expected
to provide h igh spectral efficiency (SE) and massive connectivity
in futu re wireless networks. On the other side, backscatter
communications (BC) is an emerging technology towards battery-
free transmission in future wireless networks by leveraging
ambient radio frequency (RF) waves that enable communications
among wireless devices. This paper proposes a new optimization
framework to maximize the SE of the NOMA-BC network. In
particular, we simultaneously optimize the transmit power of
the base station and reflection coefficient of the backscatter
device in each cell under the assumption of imp erfect decoding
of successive interference cancellation. The SE optimization
problem is coupled on multiple variables which makes it very
difficult to solve. Thus, we ap ply a decomposition method
and KKT conditions to get an efficient solution. Simulation
results demonstrate the superiority of th e proposed NOMA-BC
framework over the benchmark NOMA without BC f ramework.
Index Terms—Backscatter communication, non-orthogonal
multiple access, imperfect successive interference cancellation.
I. IN TRODUCTION
N
EXT-GENERATION wireless tec hnolog ies a re expected
to improve conn e ction density, latency, energy consump-
tion, and data reliability [1]. Non-orthogonal multiple access
(NOMA) using power multiplexing is one of the emerging
technologies to support a h uge number of wireless devices
while providing high data rates an d low transmission delay
[2]. In NOMA, multiple users can access the same spectr um
and time resources for transmission using different power
levels. In addition to th at, bac kscatter communication s (BC)
has emerged as a promising technology to use the ambient
signals of TV, WiFi, and cellular for the data transmission
between wireless devices [3]. Exploiting the ambient energy
harvesting technique , BC a llows the wireless device to transmit
data towards surroun ding users by reflecting an d modulating
radio frequen cy [4]. The integratio n of BC with NOMA can
further boost the performanc e of the system.
The coexistence of BC with NOMA networks has recently
been investigated [5]–[8]. For example, Khan et al. [5] h ave
optimized the transmit power of the base station (BS) and
reflection c oefficient (RC) of the backscatter device (BD)
to maximize the spectral e fficiency (SE) of the NOMA -BC
system under imperfect SIC detection. In a nother study, Yang
et al. [6] have optimized the time and RC of BD to improve
the minimum throughput of the NOMA-BC system. The work
of Chen et al. [7] have optimized the transmit power of the BS
and RC of BD to m aximize the ergodic capacity of the system.
Moreover, the authors of [8] have optimized the transmit power
of the BS and RC of BD to maximize the energy efficiency
Authors are with the Interdisciplinary Center for Security, Relia-
bility and Trust (SnT), University of Luxembourg, 1855 Luxembourg
City, Luxembourg (Emails: {waliullah.khan, eva.lagunas, asad.mahmood,
symeon.chatzinotas,bjorn.ottersten}@uni.lu).
of the system. Of late, Khan et al. [9] have optimized the
transmit power of the BS and roadside units to ma ximize the
SE of the veh ic ular network.
The above literature considers single-cell scenarios. Besides
that, most of the works [6]–[8] assume perfect SIC detection
which is impractical. To the best of our knowledge, the prob-
lem of resource allo cation for a multi-cell NOMA-BC network
that simultaneously optimizes the transmit power of BS and
reflection coefficient of BD under imperfect SIC detectio n has
not yet been investigated. To fill th is g ap, this paper proposes
a new optimization framework to maximize the SE of the
system while ensuring the minimum data rate of ea ch user. In
particular, we formulate the ma ximization of the SE subject to
various practical constraints. Dual theory and KKT conditions
are exploited to obtain the closed-form suboptimal solution,
where dual variables are iteratively upda ted. Numerical results
confirm the benefits of the proposed NOMA-BC framework
over th e conventional NOMA framework.
II. SYSTEM MODEL AND PROBLEM FORMULATION
We consider a downlink multi-cell NOMA-BC network,
where each BS reuse the same spectrum resources to en-
hance the spectral efficiency. We denote the set of cells as
C = {C
j
|j = 1, 2, 3, . . . , J}. We consider that C
j
serves
two cellular users (U
n
and U
m
) at a given time using NOMA
protocol [10]. If th e transmit power of C
j
is P
j
and the power
allocation coefficient of U
n
and U
m
is denoted as ϕ
n,j
and
ϕ
m,j
. The n the superimposed signa l that C
j
sends to U
n
and
U
m
can be expressed as x
j
=
p
P
j
ϕ
n,j
x
n,j
+
p
P
j
ϕ
m,j
x
m,j
.
Meanwhile, a BD (denoted as D
i
) also receives x
j
from C
j
,
modulate and r eflect a signal e(t) towards U
n
and U
m
with
a power of β
i
, wh ere E[|e(t)|
2
] = 1. Note that E[.] denotes
the expectation operation. Let model the channel from C
j
to
U
n
and U
m
as h
n,j
=
¯
h
n,j
d
−δ/2
n,j
and h
m,j
=
¯
h
m,j
d
−δ/2
m,j
.
Where
¯
h
ι,j
∼ CN(0, 1), ι ∈ {n, m} denotes the coefficient
of Rayleigh fading, d
ι,j
is the distance between C
j
and U
ι
and δ represents the pathloss exponent. This work assumes
the perfect channel state information, thus, the signals that U
n
and U
m
receive from C
j
can be written as
y
n,j
=
p
h
n,j
x
j
+
q
β
i,j
g
j
n,i
(h
i,j
x
j
)e(t)
+
J
X
j
′
=1,j
′
6=j
q
P
j
′
h
j
n,j
′
x
j
′
+
n,j
, (1)
y
m,j
=
p
h
m,j
x
j
+
q
β
i,j
g
j
m,i
(h
j
m,i
x
j
)e(t)
+
J
X
j
′
=1,j
′
6=j
q
P
j
′
h
j
m,j
′
x
j
′
+
m,j
, (2)
where the first segment in (1) and (2) is the r eceived signal
from C
j
, the second segment is the reflected signal from D
i
,
the third segment represents the in te r-cell interference, and the
下载后可阅读完整内容，剩余4页未读，立即下载
2021-03-27 上传
2021-02-07 上传
2021-05-23 上传
2021-02-08 上传
2021-02-10 上传
2021-05-06 上传
2021-02-08 上传
2021-02-08 上传
2021-02-11 上传
2021-02-05 上传
qsdftta
- 粉丝: 0
- 资源: 5
我的内容管理
展开
最新资源
- 前端协作项目：发布猜图游戏功能与待修复事项
- Spring框架REST服务开发实践指南
- ALU课设实现基础与高级运算功能
- 深入了解STK：C++音频信号处理综合工具套件
- 华中科技大学电信学院软件无线电实验资料汇总
- CGSN数据解析与集成验证工具集：Python和Shell脚本
- Java实现的远程视频会议系统开发教程
- Change-OEM: 用Java修改Windows OEM信息与Logo
- cmnd：文本到远程API的桥接平台开发
- 解决BIOS刷写错误28：PRR.exe的应用与效果
- 深度学习对抗攻击库：adversarial_robustness_toolbox 1.10.0
- Win7系统CP2102驱动下载与安装指南
- 深入理解Java中的函数式编程技巧
- GY-906 MLX90614ESF传感器模块温度采集应用资料
- Adversarial Robustness Toolbox 1.15.1 工具包安装教程
- GNU Radio的供应商中立SDR开发包：gr-sdr介绍
