基于改进LDA和自编码器的调制识别算法  

Modulation Recognition Algorithm Based on Improved LDA and Autoencoders

作　　者：郭业才[1,2] 张浩然 Guo Yecai;Zhang Haoran(School of Electronic and Information Engineering,Nanjing University of Information Science&Technology,Nanjing 210044,China;Jiangsu Collaborative Innovation Center on Atmospheric Environment and Equipment,Nanjing University of Information Science&Technology,Nanjing 210044,China)

机构地区：[1]南京信息工程大学电子与信息工程学院,江苏南京210044 [2]南京信息工程大学,江苏省大气环境与装备技术协同创新中心,江苏南京210044

出　　处：《系统仿真学报》2021年第2期494-500,共7页Journal of System Simulation

基　　金：国家自然科学基金(61673222,61371131);江苏省高校自然科学研究重大项目(13KJA510001);江苏高校品牌专业建设项目(PPZY2015B134)。

摘　　要：传统调制识别算法是基于高斯白噪声信道的,在复杂信道条件下识别性能明显下降。针对此问题,提出基于抗混淆线性判别分析A-ALDA (Anti-alias Linear Discriminant Analysis)和堆叠稀疏降噪自编码器SSDAE (Stacked Sparse Denoising Autoencoders)的调制识别算法。该算法中,A-ALDA算法将信号累积量特征重构为新的特征,这些特征具有更优的分离性能;将原始特征与新特征输入SSDAE进行分类,SSDAE具有提取关键信息和抗噪声的能力。结果表明,本文算法的识别准确率高于已有的算法;并且在有限信号长度条件下和相位、频率误差干扰情况下,识别准确率均有提高。The traditional modulation recognition algorithms are based on the Gaussian white noise channel, which significantly degrade recognition performance in complex channel conditions. Aiming at this problem, a modulation recognition algorithm based on A-ALDA(Anti-alias Linear Discriminant Analysis) and SSDAE(Stacked Sparse Denoising Autoencoders) is proposed. In this algorithm, A-ALDA algorithm reconstructs signal cumulants feature into new features, which has better separability. The combination of original features and new features is input into SSDAE for classification, and SSADE has the ability to extract key information and resist noise. Simulation results show that recognition accuracy of the proposed algorithm is higher than that of the existing algorithms, and recognition accuracy is improved under the condition of limited signal length and phase and frequency offset interference.

关 键 词：复杂信道 抗混淆线形判别分析 稀疏降噪自动编码器 高阶累积量 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]