自适应模糊聚类：M-QAM信号的高效自动调制分类

57 浏览量更新于2024-08-26 收藏 368KB PDF 举报

本文主要探讨了一种新颖的基于自适应模糊聚类模型的M-ary Quadrature Amplitude Modulation (M-QAM)信号自动调制分类方法。随着现代通信系统的复杂性和高速度发展，自动调制分类(AMC)在无线通信中扮演着关键角色，特别是对于M-QAM调制，因其能够高效地传输大量数据。M-QAM信号因其多进制特性，使得传统的AMC算法面临挑战，因为不同阶数的M-QAM信号具有相似的频谱特征，这给识别带来了困难。在提出的框架中，研究者着重研究了减法聚类算法的邻域半径参数。这个参数的选择至关重要，因为它直接影响到信号的分类性能。作者创新性地设计了一个自适应机制，该机制根据M-QAM信号的幅度特性动态调整邻域半径。这种自适应性有助于更好地区分不同阶数的M-QAM信号，因为不同的M-QAM阶数会有不同的幅度分布，这为分类提供了独特的特征标识。作者利用欧几里得距离来衡量测试信号与标准信号之间的集群中心间的差异。欧氏距离是一种常用的计算两个点间几何距离的方法，在这里被用来度量信号在特征空间中的相似性。通过这种方式，算法可以根据测试信号与标准信号的距离来判断其属于哪种M-QAM调制类型，从而提高了分类的准确性。为了验证这种方法的有效性，文章采用了蒙特卡罗模拟进行大量的实验。这是一种常用的数据驱动技术，通过随机抽样和重复实验来评估算法在各种情况下的性能，包括噪声环境、信道条件等变化。通过蒙特卡洛模拟的结果，研究人员可以量化算法的稳健性和鲁棒性，确保在实际通信环境中也能达到良好的性能。这项研究提供了一种创新且实用的M-QAM信号自动调制分类策略，它结合了自适应模糊聚类模型的优势，优化了邻域半径选择，并利用了欧氏距离来增强信号特征的区分度。这对于现代无线通信系统中的信号检测和解码有着重要的理论和实践意义，为进一步提升通信系统的性能和效率奠定了基础。

A NOVEL AUTOMATIC MODULATION CLASSIFICATION FOR M-QAM

SIGNALS USING ADAPTIVE FUZZY CLUSTERING MODEL

GUO-YU ZHANG

, XIAO YAN

, SHI-HAO WANG

, QIAN WANG

School of Aeronautics and Astronautics, University of Electronic Science and Technology of China,

Chengdu 611731, China

School of Information and Communication Engineering, University of Electronic Science and Technology of China,

Chengdu 611731, China

E-MAIL: zhangguoyu@std.uestc.edu.cn, yanxiao@uestc.edu.cn

Abstract:

An automatic modulation classification (AMC) method

for M-ary Quadrature Amplitude Modulation (M-QAM)

signals using adaptive fuzzy clustering model is presented. In

the proposed framework, the neighborhood radius of

subtractive clustering algorithm is emphatically researched to

satisfy different modulation orders. An adaptive construction

mechanism of neighborhood radius is designed according to

the amplitude component of M-QAM signals. Euclidean

distance of the clustering center numbers between test signal

and standard signals are utilized to identify the modulation

types. Monte Carlo simulation results and theoretical analysis

demonstrate that the proposed AMC method can provide

promising performance.

Keywords:

Modulation classification; M-ary quadrature Amplitude

modulation (M-QAM); Subtractive clustering; Adaptive

neighborhood radius; Euclidean distance

1. Introduction

Automatic modulation classification (AMC), which

can classify the modulation schemes automatically, is an

indispensable step between signal detection and

demodulation. Since the received signals usually lack priori

knowledge, AMC plays an important role in the field of

non-cooperative communication, including military and

civilian areas, such as electronic reconnaissance,

communication confrontation, electronic warfare and

cognitive radio [1-2]. Nowadays, the increasingly complex

communication environments require a more efficient

utilization of spectrum resources which is already limited

and almost fully occupied. High-order Quadrature

Amplitude Modulation (QAM), because of its high

spectrum efficiency, is wildly used in many fields such as

intelligent communication, satellite communication and

microwave communication etc. Thus the research on AMC

for M-QAM signals is of significant value.

Classical AMC methods are generally divided into two

categories, the Likelihood-based (LB) approach and the

Feature-based (FB) approach [3]. In terms of M-QAM

signals, the FB approach is more frequently employed for

its low computational complexity and suboptimal results.

Among FB approaches, constellation shape is a commonly

used feature, existing constellation reconstruction method

in previous work including subtractive clustering analysis

[4], K-means clustering analysis [5] and density spectrum

based approach [6] etc. Among constellation reconstruction

methods mentioned above, subtractive clustering is a

simple but effective solution: (i) an initial clustering center

number is not required; (ii) the number of clustering centers

is equal to modulation order for noise-free signals in theory.

However, the neighborhood radius (hereinafter referred to

as radius) which is essential for clustering accuracy, is short

of detailed and unambiguous description of how it is valued

in previous work.

In this paper, we propose a novel AMC algorithm for

M-QAM signal via adaptive fuzzy clustering model. The

rest of this paper is organized as following. Section 2 is

background knowledge which describe the subtractive

clustering theory. In section 3, the proposed AMC algorithm

including the adaptive construction mechanism of the

clustering radius is discussed in detail. In section 4, Monte

Carlo simulation experiments are carried out to investigate

the effectiveness of our proposed work, the comparison

results are shown in the figures. And the last section is

conclusion.

2. Background

The most significant feature of M-QAM is that the

constellation shape strongly depends on modulation order.

As the modulation order increases, the number of

下载后可阅读完整内容，剩余3页未读，立即下载

weixin_38654415

粉丝: 4
资源: 1015

自适应模糊聚类：M-QAM信号的高效自动调制分类

基于K均值聚类的64-QAM相干光通信系统光纤非线性均衡技术

论文研究-基于自适应权重的模糊C-均值聚类算法.pdf

Stokes空间中基于减法聚类的调制格式识别

一种基于图星座分析的新型自动调制分类器

电子功用-基于多电平的序列检测与识别方法

Matlab正交幅度调制仿真教程及源码分享

【MATLAB信号检测与分类】：如何识别信号中的模式

yolov3 在 Open Images 数据集上预训练了 SPP 权重以及配置文件.zip

qt 5.3.2 mingw 安装包

586befcf3e78455eb3b5359d7500cc97.JPG

最新资源