498 IEEE WIRELESS COMMUNICATIONS LETTERS, VOL. 6, NO. 4, AUGUST 2017
Multiple Decision Aided Successive Interference
Cancellation Receiver for NOMA Systems
Binjian Ling, Chao Dong, Jincheng Dai, and Jiaru Lin
Abstract—Non-orthogonal multiple access (NOMA) is one of
the critical technologies in fifth generation systems. Successive
interference cancellation (SIC) is a typical solution of multiple
users detection in NOMA systems. However, conventional SIC
suffers from error propagation. To alleviate this problem, in this
letter, a multiple decision SIC (MD-SIC) strategy is introduced.
Instead of choosing one codeword for each user in traditional SIC
detection, the proposed algorithm introduces multiple codewords
as the candidates to combat the error propagation. Furthermore,
a threshold is designed to control the size of candidate set for each
user. It helps the MD-SIC to achieve a better tradeoff between
performance and complexity. The simulation results indicate that
the MD-SIC algorithm obviously outperforms the conventional
SIC with a little additional computational complexity.
Index Terms—NOMA systems, successive interference cancel-
lation, multiple decision, error propagation mitigation.
I. INTRODUCTION
N
ON-ORTHOGONAL multiple access (NOMA) acts as a
promising technology to meet the requirement of increas-
ingly higher capacity in the fifth generation (5G) mobile
communication systems. Recently, sparse code multiple access
(SCMA) [1] and pattern division multiple access (PDMA) [2]
have become two typical code-domain non-orthogonal tech-
nologies. In both of the two technologies, each user’s bit
stream is directly mapped onto a sparse codeword and
multiple users share the same orthogonal physical resource
elements (PREs). This kind of overloading transmission effec-
tively supports the demand of massive connections in the
5G systems. For PDMA, the number of users superim-
posed on the same PRE may be different [3], which differs
from SCMA.
In the receiver of NOMA systems, the maximum likeli-
hood (ML) detection provides the lower bound of multiuser
detection performance by traversing all the possible signal
combinations. Nevertheless, the computational complexity of
ML algorithm increases exponentially with the number of
access users and modulation order. Therefore, it is difficult to
be applied in practical systems. Successive interference can-
cellation (SIC) is the most well-known detection technology in
Manuscript received April 14, 2017; revised May 20, 2017; accepted
May 20, 2017. Date of publication May 25, 2017; date of current version
August 21, 2017. This work was supported in part by the NSFC under Grant
61601047 and Grant 61671080, and in part by the BUPT-SICE Excellent
Graduate Students Innovation Fund. The associate editor coordinating the
review of this paper and approving it for publication was R. C. de Lamare.
(Corresponding author: Binjian Ling.)
The authors are with the Key Laboratory of Universal Wireless
Communications, Ministry of Education, Beijing University of
Posts and Telecommunications, Beijing 100876, China (e-mail:
lingbinjian@bupt.edu.cn; dongchao@bupt.edu.cn; daijincheng@bupt.edu.cn;
jrlin@bupt.edu.cn).
Digital Object Identifier 10.1109/LWC.2017.2708117
NOMA systems. Theoretically, the use of SIC provides capa-
bility to approach the boundary of MAC-RR [4]. Furthermore,
the computational complexity of SIC is much lower than that
of ML. In spite of all these advantages above, in the prac-
tical communication systems, it suffers from a performance
loss due to the error propagation [2]. Especially, when there
exist correlations among user access channels, the performance
of SIC receiver will further degrade. The mitigation of error
propagation is investigated in [5]–[10].
In this letter, to alleviate the above error propagation
problem, a multiple decision successive interference cancella-
tion (MD-SIC) algorithm is proposed. Standard SIC detection
is regarded as a process of searching the single path greed-
ily in the decision tree. Hence, there is no chance to correct
the inaccurate decisions of previous users. While in MD-SIC,
multi-branch expansion is employed at the instantaneously
detecting node. Conventional SIC detection is adopted for
the subsequent users’ signal detection. By doing so, more
points in the decision tree are considered and the error
propagation is obviously mitigated. It is worth noting that,
unlike MF-SIC in [5], the proposed strategy replaces MMSE
algorithm with a local ML (local_ML) function for better
performance and lower complexity. Furthermore, a predefined
threshold is proposed to prune the unnecessary branches,
which keeps a better balance between performance and com-
plexity. Simulation results indicate that the performance of
MD-SIC algorithm approaches the ML solution with compa-
rable complexity to the standard SIC receiver.
The organization of this letter is as follows. Section II
describes the notation conventions, and NOMA system model.
Section III is committed to introducing the MD-SIC algorithm.
Section IV performances the evaluation results. Conclusions
are given in Section V.
II. N
OTATIONS AND SYSTEM MODEL
A. Notation Conventions
In this letter, calligraphic characters such as
X is used to
denote sets. We write lowercase letters (e.g., x) to denote
scalars. Let the notation x denote a vector and x
i
to denote
the i-th element in x. The set of binary, real and complex
numbers are denoted by B, R and C, respectively. The bold
letters, such as X, denote matrices, X
i
is the i-th row of
matrix X, and X
i,j
is the element at the i-th row and the
j-th column of matrix X. For a diagonal matrix diag(x),
its i-th diagonal element is x
i
. The notations X
T
and x
T
stand for the transpose of matrix X and vector x, respectively.
Throughout this letter, the function log(·) means “logarithm
to base 2”, and the function ln(·) stands for the “logarithm
to base e”.
2162-2345
c
2017 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission.
See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
下载后可阅读完整内容，剩余3页未读，立即下载
weixin_38595019
- 粉丝: 8
- 资源: 894
我的内容管理
展开
最新资源
- IPQ4019 QSDK开源代码资源包发布
- 高频组电赛必备：掌握数字频率合成模块要点
- ThinkPHP开发的仿微博系统功能解析
- 掌握Objective-C并发编程：NSOperation与NSOperationQueue精讲
- Navicat160 Premium 安装教程与说明
- SpringBoot+Vue开发的休闲娱乐票务代理平台
- 数据库课程设计：实现与优化方法探讨
- 电赛高频模块攻略：掌握移相网络的关键技术
- PHP简易简历系统教程与源码分享
- Java聊天室程序设计：实现用户互动与服务器监控
- Bootstrap后台管理页面模板（纯前端实现）
- 校园订餐系统项目源码解析：深入Spring框架核心原理
- 探索Spring核心原理的JavaWeb校园管理系统源码
- ios苹果APP从开发到上架的完整流程指南
- 深入理解Spring核心原理与源码解析
- 掌握Python函数与模块使用技巧
资源上传下载、课程学习等过程中有任何疑问或建议，欢迎提出宝贵意见哦~我们会及时处理！
点击此处反馈
